Handbook of Ugaritic Studies (Handbook of Oriental Studies)

+ y W +

?

1

0

H H

O

Ί

w

À

YY
? φ

y Φ

w

T

ν κ £D

K

1

?

/1 H ή

o n

-o -

w

^

—

χ

X

X

_

_ H-X 4

I

t f Ο Φ * -M e

* à* o

ι

oo

Λ7 V ï ? O 31

Λ« Ä : e m r H*

W

4

λ

X W f«

-> 1

w

r"

1

A 3

n ? Il 96 I

Ψ

o-

1

î

<7

II

ĪT

*

6

=

t=

-

d n

«D

-» ri

à 1 r t s

π

T

(-Π)

W ri

π

r

- —> O

Ό

i

\ X

?

a . . . Ill- -»w V - n>rt ¥

û

ν u

III -

^

Saf.

ή

ζ -»a cr) r r É

H*

OSA Iih.

•4 X

«

*>-

1 C

*

? *

-

ν / γ<

~~

3—

Λ h an

-

va —

N

S

®

i k

-> π

I

^«J " F

Â

ï

U

3

V *

2

Southern alphabets

LONG ALPHABET

* }

1

>;

>

X IX

1 D3

t χ- -t

? I C

(P) fill)

Fig. 5

The Ugaritic script in relation to the Western and Southern scripts ( D i e t r i c h - L o r e t z 1988a, 102)

Fig. 6

Fig. 7

The Alphabet tablet from Beth Shemesh ( D I E T R I C H - L O R E T Z 1988a, 285)

Tablet with the South Arabic Alphabet (BORDREUIL

-

PARDEE

1995,

856)

Beth Šemeš Ugarit {J. Ryckmans

H L H M Q. w H L H M Q. w H L H M Q. w

Beth Šemeš

'

Ugarit

' '

{J. Ryckmans

'

Fig. 8

'

'

Š R Τ R Β

TS Κ N U Β [Š] F T D Š K N U S S Ρ

Š

T

S

Κ

T

Š

T

S/Z}

R

B(G)

Ζ

G

D

G

Τ

Ζ

D Τ

D

G

D

G

Τ

Ζ

Y

D

G

D

G(B)

Τ

Ζ

D Y

Ν

H

S

F}

Deciphered Alphabets of the South-Eastern A-/-^-m-Tradition

with the traditions of Beth Shemesh and Ugarit gives the following results (Fig. 8)—in the third line the South Semitic alphabet is given in brackets for convenience, as established by R y c k m a n s ( 1 9 8 5 ) from texts dating to the first millennium bce. This comparison shows clearly that the sequence of letters of the cuneiform alphabets attested in Beth Shemesh and Ugarit as well as the South Arabic-South Semitic alphabet reconstructed by Ryckmans show very few divergences. T h e fact that the signs, as shown by hand-copies and photographs, vary slightly as well, leads to the conclusion that the traditions, certainly not least in view of their geographical separation, must to some extent have developed independently. These divergences are basically so slight that their common origin never completely vanished from view. This report on the discovery of the two cuneiform alphabets as transmitted in Ugarit as well as their summary comparison should show that the commercial centre Ugarit in the third quarter of the second millennium was a turning point in the early history of the alphabet in two ways: 1. With the earliest recorded alphabet so f a r — w h e t h e r in the '-b-g sequence of the north west tradition or the h-l-h-m sequence of the south eastern tradition—we are quite unexpectedly—or rather, as was to be expected—at a point in its development which already has the appearance of the conclusion to a forerunner. Just when we thought we were close to the origin of the evolution of the alphabet and finally were able to propose a date for the 'discovery of the alphabet', we have to envisage a world with two alphabets (Fig. 9).

2. T h e origin of the cuneiform alphabet from Ugarit also reflects a history of two alphabets: a Levantine tradition with 22 letters was shall we say overlaid by an Arabic tradition so that, with some additions, a 30-letter alphabet was derived. T h u s we find ourselves in the midst of a historical process on which it is worthwhile to reflect a little: the inhabitants of a city who towards the middle of the second millennium had migrated from south eastern Palestine to the northern Levant, had developed a commercial ruling dynasty in Ugarit which in the history of literature and the art of the scribe had thus created a lasting monument, so that on arrival its native tongue could be inserted into an existing scribal school and writing tradition and thus the alphabet already discovered could be expanded. T h e authentic Ugaritic alphabet which arose in this way is ultimately to be perceived as a witness of inter-cultural activity with branches as far afield as Cyprus (Hala Sultan Teke). With the examination of the cuneiform alphabets transmitted in Ugarit we find that in terms of the history of the alphabet we are not, as many would argue, at its inception. Instead, we are once again at the conclusion of separate developments which nevertheless in terms of their significance represent an important step in universal history. Although we find ourselves in the mid-second millennium bce, we have still not yet reached a point at which we can speculate on the origin of the alphabet. (Translation:

W.G.E.

Watson)

.ArstaiTai© •Hamm W Bars®© ,8f*S© -Halet>/Ateppo •SfTre« • Äf is ö • Ris Šamrj/UqaritÍ

/larnaka/Kition· Hata Sultan Tekke •

Uj CQ • Palmyra/Tadncr

Neti Mwd/CLadeS.

Α· Sarafand/Sar»pta, #Ruweisah±

"•TK3md ei-LÛz/Kumidi •Damaskus

Sean©

Mountains of s G ILE AD

• Oer ΘΤ.0»$ί

A

•«HWj Heön»i J .Balu'an •

.QakSBarnea θ

Ugaritic cuneiform

ψ

Proto-Canaanlte Inscriptions

•

Ancient Canaanite Inscriptions

φ

Ancient Phoenician Inscriptions (up to the end of the 8th cent. BCE)

©

Ancient Aramaic Inscriptions (up to the end of the 8th cent. BCE)

β

Ancient Hebrew Inscriptions (up to the end of the 8th cent. BCE)

0

is

ill

7S

100kw

drawing: G. Neuber

Fig. 9

The spread of cuneiform alphabets in the Eastern Mediterranean ( D I E T R I C H - L O R E T Z 1988a, 359: Map)

3

Ugaritic

Josef

3.1 3.1.1

Grammar

Tropper

Introduction

The classification of Ugaritic

Ugaritic, the local language of the city state of Ugaritic, is one of the Semitic languages. T h e classification of Ugaritic within Semitic is still a matter of dispute. T h e Semitic languages can be broadly divided into an East Semitic branch (Akkadian) and a West Semitic branch. West Semitic can be further divided into a South Semitic and a central Semitic branch (Canaanite, Aramaic and North Arabic). Of the central Semitic languages, Canaanite and Aramaic can be traced back to a common former stage which is called 'Northwest Semitic'. T h e r e is hardly any doubt nowadays that Ugaritic is more closely related to Canaanite and Aramaic than to North Arabic. 1 T h u s Ugaritic is shown to be a Northwest Semitic language. Since Ugaritic is closer to the later Canaanite languages than to Aramaic due to more important linguistic isoglosses, either it belongs to Canaanite or it is the (only) representative of a separate language branch of Northwest Semitic closely related to Canaanite.3.1.2

Current research in Ugaritic grammar

3.1.2.1 T h e four most important overall summaries of Ugaritic g r a m m a r published so far are: 1. C.H. G o r d o n , Ugaritic Grammar (Rome 1940) with two further editions: Ugaritic Manual (Rome 1955) und Ugaritic Textbook (Rome 1965). 2 . S . S e g e r t , A Basic Grammar of the Ugaritic Language (Berkeley, Los Angeles, London 1984).

' The opinion maintained in the early period of research that Ugaritic is particularly close to North Arabic can no longer be held today. The two most important arguments for this theory, the extensive phoneme inventory of Ugaritic and Ugaritic-Arabic isoglosses have been substantially modified by the results of recent research. 2 For most recent discussion see especially ISAKSSON 1989 and T R O P P E R 1994b.

D. S i v a n , Ugaritic Grammar (Encyclopaedia Miqra'it 9 ; Jerusalem 1993) [in Hebrew], 4. D. S i v a n , A Grammar of the Ugaritic Language ( H d O 1/28 Leiden, New York, Cologne 1997). 3.

By now, Gordon's summaries have been largely superseded. T h e later grammars listed merely present a concise grammatical outline of Ugaritic. As yet there is no complete, modern reference grammar of Ugaritic. In his Habilitation submission presented in 1997 at the Freie Universität Berlin with the dtie Untersuchungen zur ugaritischen Grammatik. Schrift- Laut- und Formenlehre3 the author has laid the foundation stone for such a grammar. 3.1.2.2 In spite of research spanning several decades, a large number of problems of Ugaritic g r a m m a r remain unsolved, due to the following factors: (a) Ugaritic is a dead language attested only in writing. (b) Closely cognate contemporary languages are known only sketchily. (c) T h e corpus of Ugaritic texts is recorded in a writing system without vowels, except for a few words in syllabic spellings. (d) T h e range of grammatically important Ugaritic texts is restricted. (e) M a n y texts are of uncertain interpretation. (f ) T h e corpus of Ugaritic texts contains a set of very different genres, each with its own grammatical features. 3.1.3

Aim of this grammatical outline

Here follows a brief grammatical outline of spelling, phonology, morphology and morpho-syntax of the Ugaritic language. It contains only the central topics of grammar. Facts on which scholars are widely in agreement are described without reference to secondary literature. However, areas where there are problems are provided with a short commentary. In this way the outline of grammar will describe the present state of research on Ugaritic grammar.

3

Research on Ugaritic Grammar. Script, Phonology and Morphology.

3.2 3.2.1

Orthography

The principles of consonantal orthography

T h e basic principle of the Ugaritic alphabetic script is that each consonantal phoneme of the language is equivalent to one sign in the script. By the introduction of the three additional signs of the Ugaritic alphabet, < i > , < u > and < s > , the principle is violated in two ways: the phoneme / V is represented in Ugaritic—depending on the following vowel—by < a > , < i > or < u > (cf. § 3.2); the phoneme / s / is represented by <s> or < s > . 4 Lengthened (doubled) consonants were not differentiated from single consonants. They can be determined only by comparative philology or on the basis of syllabic spellings. 3.2.2

The aleph signs

a. T h e most remarkable feature of the Ugaritic alphabet is that it has three different signs for writing the glottal stop / V , i.e. an a-aleph = < a > , an z'-aleph = < i > and a w-aleph = < u > . b. It is generally agreed that the three Ugaritic aleph signs are used in syllable-initial position: < a > represents the syllables Γa/ and /'ā/, < i > represents / V , Γ ΐ / , Γ ΐ Ι and probably also / W (shcwa) and lasdy < u > represents I'ul, Γū/ and /'0/. c. T h e way in which syllable-final, i.e. vowelless aleph is written is a matter of dispute. T h e r e are quite different theories. T h e extreme views are as follows: 1.) Any syllable-closing aleph is represented by . 2.) T h e choice of sign for syllable-closing aleph depends on the quality of the preceding vowel. 3.) Syllable-closing aleph is no longer expressed in Ugaritic: / û V becomes / â / , / f / becomes I I I and / u } / becomes / « / . T h e aleph signs < a > , < i > and < u > act as vowel letters (matres lectionis) for the resulting contracted vowel: / á / , lîl and IÛI?

4 For the phonetic value of this grapheme, which gained entry into the Ugaritic alphabet only at a later date, see S E G E R T 1 9 8 3 (<s> = [su]) and recently T R O P P E R 1995. Thus the grapheme <s> has the value ['s]. It was inserted after the Ugaritic phoneme / s / written with <s>, which originally (also) had the phonetic value ['s], had been deaffricated to [s] in certain phonetic contexts. 5 For the proponents of these theories see V E R R E E T 1983a, 2 2 3 - 6 .

d. T h e theory of the position of aleph defended here is more complex than the proposals mentioned above. 6 O n the one hand it envisages the possibility that the syllable-closing glottal stop in Ugaritic was not always strongly articulated (= quiescent aleph). O n the other hand it follows that in Ugaritic / a V after the loss of a syllable-closing glottal stop—most probably irrespective of stress—became either / â / or / 0 / (cf. Heb. näsä'tä < *naša>tā as distinct from Heb. roš < *ra>š ). O n this basis the following 'rules' can be formulated: the (articulated) syllable-closing glottal stop is written with an < i > , irrespective of the preceding vowel. If the glottal stop is not articulated, i.e. the aleph is quiescent, then the following applies: < a > stands for / â / < *a}; < i > stands for III < *i'; < u > stands either for /û/ < *u} or for loi < V . e. T h e graphic notation of a syllable-closing glottal stop is non-homogeneous in the corpus of Ugaritic texts, as some alephs evidendy represent a strong aleph, others a quiescent aleph. T h e former are phonemic spellings, the latter phonetic spellings: e.g. yihd lya'fyud-l 'he takes/took' ( K T U 1.6 = RS 2. [009]+5.155 ν 1, etc.) againstyahd lyâhud-I <*ya}(}ud- ( K T U 4 . 4 4 = RS 9.453:28) or yuhd(m) lyôffVd-l <*ya'hud- ( K T U 1.4 = RS 2.[008]+3.341+3.347 iv 16; K T U 1.22 = R S 2.[024] ii 17*; K T U 1.103+ = R S 24.247+: 17) and tuhd/tôfj. Vdl<*ta'hud- ( K T U 1.2 = RS 3.367 i 40).7 Within a word there are more spellings with strong aleph, but at the end of words there are more with quiescent aleph. 3.2.3

Vowel notation

In principle, in alphabetic spelling vowels are ignored, apart from the aleph signs which in syllable-initial position express the vowel following an / V (as well) (§ 3.2.2b). Exceptions are rare. T h e sporadic use of < y > as a vowel letter (mater lectionis) for / í / is worth emphasizing, especially at word-close, e.g. ily ugrt /}ilī 3UgaritV/ 'the gods of Ugarit' ( K T U 2.26 = R S 16.264:4-5). 8

6

On this see T R O P P E R 1990b. Cf. Heb. yôhèz (18x; alongside 3 X ye"höz). 8 On vowel letters in Ugaritic see BI.AU - L O E W E N S T A M M 1970 and L O R E T Z 1973c. 7

DIETRICH

-

3.3 3.3.1

Phonology9

The consonantal phoneme system

According to the Ugaritic alphabet, which comprises twenty-seven 'ordinary signs' and three 'supplementary signs', there were at least 27 consonantal phonemes in Ugaritic. This relatively wide range of consonants can be subdivided into obstruents (plosives and fricatives), nasals ( / m / , / « / ) , liquids ( / / / , /r/) and semi-vowels (/w/, /y/). Depending on how they were articulated (unvoiced, emphatic, voiced) most of the obstruents can be arranged in rows of three columns, as follows: unvoiced labials dentals interdentals (alveolar) sibilants (palato-alveolar) sibilants velars uvulars pharyngeals laryngeal

3.3.2

/p/ /t/ /t/ /s/ /š/ /k/ / y /h/ /h/,

emphatic -

/t/ /z/ /s/ /q/ —

voiced /b/ /d/ /d/ /z/ 'g' igi /7

/7

Comments on uncertain consonantal phonemes

a. T h e Semitic phoneme / d ! is written consistently with < d > only in texts K T U 1.12 = RS 2.[018] and K T U 1.24 = R S 5.194. In the other texts it is written predominantly with the sign < d > — a p a r t from specific phonetic environments 1 0 —which indicates a conditioned coalescence of / d / and / d / in Ugaritic. b. Etymological / ζ / usually occurs in Ugaritic spelling as . However, there are a few words in which < g > stands for etymological / £ / . " Incontrovertible examples are ΛIngr, 'to guard', Vgm\ 'to be thirsty' and gr, 'mountain'.

9

On this topic see FRONZAROLI 1955b. Often in the vicinity of Irl (cf. UT § 5.3), /m/ and /n/. 11 See FRONZAROLI 1955b, 33-5, Ί / Τ § 5.7 and D I E T R I C H - L O R E T Z , 1967, 312-4. RÖSSLER 1961 has a different view. 10

c. Ugaritic has four sibilants, i.e. / s / , / s / , / z/ and / š / . T h e first three phonemes stand out by their distinctive feature of affrication and should have been realised as ['s - 's5 - d z]. In contrast, /Š/ is a fricative palato-alveolar sibilant, i.e. [š]. d. T h e lateral phonemes /s2/ (= / š / ) and / / / attested in other Semitic languages are regularly represented in Ugaritic spelling by the graphemes <š> and <s>. Rarely—for instance in K T U 1.12 = RS 2.[012]—etymological / / / appears in Ugaritic as < ? > . 3.3.3

PS

Equivalence table of selected Semitic consonants12

OSA

Arab.

Ugar.

Heb.

Aram.

Eth.

Akk.

h

s s ζ s s Š d h

š s ζ s Š Š s h

c

<

early t ζ d s3 s1 s2 Š X y

t ζ d s 3 /s s'/s S2/Š d h g

3.3.4

t ζ d s s Š d h g

t ?» g* d, d s š Š s, z* b g

š s ζ s Š s s h c

late t t d

š s ζ s š Š q

s c

The vocalic phoneme system

Proto-Semitic has a) three short vowels, / á / , /Γ/, / « / , b) three long vowels, / ā / , / í / , / ū / , and two dipththongs, / a y / and /aw/. Ugaritic has preserved the primary vowels / a / , /i/ and /u/ as both long and short. T h e PS diphthongs *ay and *aw, however, have been contracted to monophthongs: *ay > /ê/\ *aw > /ô/ (see § 3.3.5.3a). 3.3.5

Sound changes

A large number of sound changes are documented in Ugaritic spelling following phonological rules. T h e most fundamental are set out here.

12

OSA — Old South Arabic; Arab. = Classical Arabic; Aram. = Aramaic; Eth. = Ethiopie; Akk. = Akkadian; PS = Proto-Semitic.

3.3.5.1

Consonantal sound changes

(a) Sound change *w > lyl at word-initial, e.g. \ybl < *wbl, 'to carry, bring'. 13 (b) Voicing—devoicing, 14 especially with labials, e.g. ybcl /yib'alu/ 'he makes' ( K T U 1.17 = RS 2. [004] vi 24; / t t / , e.g. aht /'ah(h)att-/ < *'ahadt'one' ( K T U 1.48 = RS 1.019:13 etc.); *nC > /CC, e.g. ap = syllabic spelling ap-pu /'appu/ < * 'anpu 'nose' ( K T U 1.2 = R S 3.367 i 13 etc.); *IC > /CC/, only in Ίqh 'to take' (see § 3.4.4.4c). 3.3.5.2

Vocalic sound changes 1 ' 1

(a) Vowel harmony: *qattv\l > /qv\ttv\l/, e.g. ibr /'ibbīr-/ < *'abbīr 'bull' ( K T U 1.10 = R S 3 . 3 6 2 + iii 35 etc.); *>ViC.v2C(C) > ^V2C.V2C(C) ( V = short vowel), e.g. urbt /'urubbat-/ < *'arubbat'opening, hatch' ( K T U 1.4 = R S 2. [008]+ ν 61, etc.), iršt Γirišt-/ < *>arišt- 'wish' ( K T U 1.104 = RS 24.248:1, etc.). (b) Vowel syncope: pretonic: (Cv)CvCvC.v > (Cv)CvCC.v (V = short vowel), e.g. rišt /ra'sat-/ < *ra'ašāt- 'heads' ( K T U 1.2 = RS 3.367 i 23.24, etc.); post-tonic: C.vCvCvCv > C.vCCvCv or C.vCCvCvCv > CvCCvCCv (V = short vowel). 3.3.5.3

Sound changes in diphthongs and triphthongs 16

(a) Contraction of diphthongs: *aw > / / /î/\ >

*uw

/Û/.

(b) Preservation of some types of triphthong ( / u w a / , /iyv/, /aw/yā/, /vw/yv/) as opposed to the contraction of other types of triphthong (*awu> /û/\ *awï> /î/*ayu > /«/; *ayÍ > / f / ; *awa> /â/\ *aya > /á/; *uwu > /«/).

13

However, word-initial /w/ is preserved (a) in the conjunction w, 'and', (b) before the vowel lui (D-stem infinitives of the I-y < I-w roots, e.g. wld lwullad-1 'to bear (a child)' [KTU 1.14 = RS 2.[003]+ iii "48 etc.] and wpt-m lwuppat-1 'to insult' [KTU 1.4 = RS 2.[008]+ vi 13]). 14 On this topic see G A R R 1 9 8 6 and V O I G T 1 9 9 1 . 15 For syllabic spellings see H U E H N E R G A R D 1987b, 268-83. 16 For syllabic spellings see H U E H N E R G A R D 1987b, 288-92.

3.3.5.4

Sound changes within syllables

(a) Prothesis (to avoid a word-initial consonant cluster): e.g. usbc/(')usbac-/ < *sibac- 'finger (pi.)' ( K T U 1.2 = R S 3.367 iv 14, etc.). (b) Specific pausal forms: words at the end of a sentence occasionally have a phonetically altered form (e.g. reduction of the ending or special lengthening of the stressed syllable).

3.4 3.4.1 3.4.1.1

Morphology and morphosyntax

The pronoun T h e personal pronoun

a. Nominative forms: l.c.sg. ank — syll. a-na-ku Γanāku/ (longer form) or an /3anā/ (shorter form); 2.m.sg. at = syll. at-ta /}atta/ < *}anta; 2.f.sg. at /3atR/ < *'anti; 3.m.sg. hw = syll. ú-wa /huwa/ < *hu}a; 3.f.sg. hy /hiya/ < ^hi'a; 2.m.pl. atm /'attumü/ < *'antumü; 2.c.du. atm Γattumā/ < *'antumā; 3.c.du. hm /humā/.17 b. Oblique forms (gen./acc.): 3.m.sg. hwt /huwati/ ( K T U 1.3 = R S 2. [014] vi 20 etc.); 3.f.sg. hyt /hiyati/ ( K T U 1.3 = R S 2. [014] iii 10 etc.); 3.m.p1. hmt /humūti/ ( K T U 1.19 = R S 3.322+ iii 9 etc.); 3.c.du. hmt /humāti/ ( K T U 1.17 = R S 2. [004] ν 20.30; K T U 1.19 = RS 3.322+ iii 44). 3.4.1.2

Pronominal suffixes

l.c.sg. possessive suffix -0 or -y / - í / , -y /-ya/;ÌR l.c.sg. object suffix -η /-nī/; 2.m.sg. -k /-ka/; 2.f.sg. -k /-kï/; 3.m.sg. -h = syll. -û /-hü/; 3.f.sg. -h /-ha/; l.c.pl. -n /-na/(?); 2.m.pl. -km /-kurnu/; 2.f.p1. -kn /-kun(rì)a/; 3.m.pl. -hm /-humū/; 3.f.p1. -hn /-hun[n)a/; I.e.du. -ny /-nayā/; 2.c.du. -km /-kumā/; 3.c.du. -hm /-humā/. 3.4.1.3

Demonstrative pronouns

- 'this' (adjectival): hnd (many); hndn ( K T U 2.71 = RS 29.095:10). - 'this' (nominal): hndt ( K T U 1.19 = R S 3.322+ iv 62; K T U 2.38 = RS 18.031:12; K T U 2.45 = R S 18.140:7).

17

l.c.pl., 2-f.pl., 3 m./f.pl. and I.e.du. forms are not attested. After a short H-/ vowel and after various long vowels it is / -ya/, otherwise it is / I / . 18

- 'that' (adjectival): hnhmt ( K T U 3.3 = RS 15.128:8; perhaps also K T U 4.659 = RS 19.166:6). - 'that' (nominal): hnk ( K T U 2.33 = R S 16.402:23); hnkt ( K T U 2.41 = RS 18.147:13; K T U 2.21 = RS 15.174:10). 3.4.1.4

T h e determinative pronoun (relative pronoun)

T h e forms of the Ugaritic determinative pronoun, which also functions as a relative pronoun, are: m.sg. d — /dû/ (Nom.), / d ī / (gen.), /dā/ (acc.) and d = /dī/ (only K T U 1.24 = R S 5.194:45 [gen.]); f.sg. dt = /dātu/, /dāti/, /data/; c(?).p1. dt /dūtV/. T h e r e is also an indeclinable variant d = /da/(?).19 3.4.1.5

Interrogative pronouns

- 'who?': my /mĪya/(?) (several occurrences); mn ( K T U 1.3 = RS 2.[014] iii 37; K T U 1.3 iv 4; perhaps K T U 1.5 = RS 2. [022] + iv 23. - 'what?': mh /mah(a)/ (several occurrences); mhy ( K T U 2.14 = RS [Varia 4]:9 only); mat ( K T U 1.14 = R S 2. [003]+ i 38 only); mn (uncertain occurrences: K T U 1.5 = RS 2. [022] + iv 23; K T U 2.45 = R S 18.140:25; K T U 2.72 = RS 34.124:22; in K T U 1.16 = R S 3.325+ ii 19.20 it means 'how many?'). 3.4.1.6

Indefinite pronouns

- 'anyone': mnk ( K T U 3.2 = R S 15.111:12 [mnk mnkm]); mnkrn ( K T U 2.19 = RS 15.125:12; K T U 3.2 = RS 15.111:13); mnmn ( K T U 1.123 = R S 24.271:22 [mr mnmn]);20 mnn ( K T U 5.9 = RS 16.265 i 2). - 'anything': mhk ( K T U 2.38 = RS 18.031:26); mhkm ( K T U 2.30 = RS 16.379:22; K T U 2.71 = RS 29.095:14 [mhkm]); mnm (many occurrences). - 'whatever' (adjectival): ay ( K T U 1.23 = RS 2.002:6; K T U 1.24 = R S 5.194:44).

19 It occurs only as a determinative pronoun before a noun clause or as a relative pronoun before a nominal relative clause. 20 Cf. Akk. mammon < *man-man 'anybody' as well as the expression mar marnmana(ma), 'anybody's son' (CAD M / l , 200-1).

3.4.2 3.4.2.1

The noun Noun formation

Several different patterns are used for the formation of nouns (nouns and adjectives) in Semitic: monosyllabic forms; polysyllabic forms; forms with lengthened components; forms with prefixes, infixes and suffixes. Most patterns can be assigned to specific classes of meaning. T h e r e are severe limitations on identifying nominal patterns in Ugaritic as the alphabetic spelling often permits no conclusions regarding formation. 21 T h e Ugaritic noun forms attested in syllabic spelling are rich in information. 2 2 3.4.2.2

Gender

T h e r e are two grammatical genders: masculine (masc.) and feminine (fem.). Masc. nouns are basically unmarked, whereas as a rule fem. nouns have a special ending. By far the commonest feminine morpheme is -(a)t. T h e choice of the m o r p h e m e variant -at instead of -t is largely dependent on syllable structure. T h e -(amending also denotes nomina unitatis (singular nouns), e.g. mnht '(single) gift' ( K T U 4.709 = R S [Varia 13]:6) in relation with the generic name mnh 'gift(s)' ( K T U 1.2 = RS 3.367 i 38 etc.). Besides the feminine ending -t = /-{a)/ probably a rare feminine ending y — /-ayV/ is also attested: ncmy 'the (exceedingly) lovely' ( K T U 1.5 = RS 2. [022]+ vi 6.28; K T U 1.17 = RS 2. [004] ii 42). T h e r e are also grammatically feminine nouns without a feminine ending, e.g. um 'mother'. 3.4.2.3

Number

T h e r e are three numbers in Ugaritic: singular (sg.), dual (du.) and plural (pl.). T h e sg. is unmarked. T h e du. and pi. are denoted by special morphemes. In Ugaritic the du. is productive and is marked by the m o r p h e m e - 0 = / ā / which always coalesces with the case ending: nominative -a, oblique (gen./acc.) -e. In the absolute state the ending is lengthened by mimation: nom. -m — /-āmi/, obi. -m — /-êma/ (alternatively: /-êmi/). As a rule, the dual ending is added onto the singular 21

C f . SEGERT

22

On this topic see

1 9 8 4 § 4 3 ; SIVAN

1997,

HUEHNERGARD

60.

1987b, 302 17.

form. It comes after the gender m o r p h e m e -t — /-(a)t/ of (marked) fem. nouns. T h e pi. is marked by a m o r p h e m e which causes vowel lengthening. With fem. nouns the plural marker comes before the gender marker and the case endings: nominative /-ātu/, oblique /-āti/; in the masc. noun it merges with the case endings: nominative / - « / , oblique /-Ī/ (absolute state: /-ūma/, /-īma/). T h e nominal base of the pi. is mosdy the same as the sg. In certain nominal patterns and certain weak root classes, the plural basis differs from the base of the singular. It should be emphasized that -forms generally have a bisyllabic base in the plural (e.g. pi. rašm/1 /ra'as-/ of sg. ris / ra's-/). 3.4.2.4

Case

a. Ugaritic has three main cases: nominative (nom.), genitive (gen.) and accusative (acc.). They are marked by vocalic morphemes which follow the gender marker in fem. nouns. In the (masc. a n d fem.) sg. the three main cases are mostly differentiated by three different vowel endings: nom. -u, gen. -i, acc. -a (triptotic endings). Nouns of certain patterns (including certain personal names) have only two different case-endings in the sg.: nom. -u, gen./acc. = oblique (obi.) -a (diptotic). In the du. and masc. pi. the inflection is exclusively diptotic: d u . n o m . -ā, du.obi.-ê; m.pl. nom. -ū; m.pl. obi. -f.23 b. Besides the three main cases, Ugaritic has two further cases with primary adverbial function, i.e. the terminative and the locative. Both are comparatively little used. T h e terminative functions as an independent adverbial case, primarily for denoting direction. It is marked by the -Α-ending which, in connection with the so-called 'he locale (locative h) of Hebrew g r a m m a r is probably to be vocalised as /-ah/. T h e terminative ending is probably added on to the uninflected noun stem: arsh = /',arsah/ 'towards the earth' ( K T U 1.14 = RS 2. [003]+ i 29). It only occurs for certain in the abs. state. T h e locative is marked by the ending / - « / , e.g. sbu špš /sabVu/

23 Occasionally the oblique ending seems (already) to have assumed the function of the nominative ending, e.g. ily ugrt (= /ill UgaritV/) tgrk tšlmk 'may the gods of Ugaritic guard you (and) grant you well-being' (KTU 2.16 = RS 15.008:4-6).

'at sunset' ( K T U 1.41 = RS 1.003+:47.53). Examples are difficult to identify as the locative ending is only evident from spellings in forms of III-' roots. T h e r e seem to be several different functions of the locative. It denotes place (locative), time, the ablative, the instrumental, measure and quantity, final nuances (with infinitives or verbal nouns) and the paronomastic infinitive (e.g. bt krt bu tbu, 'she did enter Kit's house' [ K T U 1.16 = RS 3.325+ vi 3]). c. T h e form of the noun in the imperative and in direct address— the vocadve—is expressed by various syntagmata: (a) by an unintroduced noun, (b) by the noun introduced by the particle /, (c) by a noun introduced by the particle y or (d) by a noun with a 1st pers. noun suffix. There is no explicit information on the case-endings of the vocative in Ugaritic in spite of a few occurrences of III-' radical nouns. It is uncertain whether a noun in all the constructions just mentioned (a-d) has the same morphological form. It is also uncertain whether the vocative has the same form as one of the three main cases. T h e r e are indications that in the singular the vocative can be expressed without any case-endings and that the accusative case serves as a vocative. d. In the corpus of Ugaritic texts there are occasionally m.sg. nouns without any inflection. 24 This could be a relic of what is known as the 'absolute case', 25 comparable with the 'absolute state' of Akkadian g r a m m a r (cf. GAG § 62 c-j). 3.4.2.5

State

T h e noun has two states which depend on the syntactic position of a noun: 1. the absolute state (abs. st.), 2. the construct state (cstr. st.). T h e abs. st. is unmarked in the singular and so is the same as the cstr. st. In the dual and plural it is sometimes marked by a final -m, known as nominal mimation. T h e cstr. st. is unmarked for all numbers and differs from the abs. st. in the dual and plural by the lack of mimation. Both states in Ugaritic have (the same) case-endings.

24 25

f o r syllabic spellings see H U E H N E R G A R D 1987b, 300-1. On the absolute case in Hamito-Semitic see SASSE 1984.

3.4.2.6

Determination/Indetermination

Ugaritic has no morphological marker for determination or indétermination. T h e r e is neither a definite article nor a specific determined case, and mimation on nouns has neither a determinative nor an indeterminative function. 26 3.4.3

Cardinal numbers

a. T h e cardinal numbers 1 - 1 0 are as follows: one' two' three' four' five' six' seven' eight' nine' ten'

ahd /'ah(h)ad-/;

aht/'ahhatt-/

in /tinā/ (nom.), /tinê/ (obi.); tt /tittā/ (nom.), /tittê/ (obi.) tit / talāf, tilt /talātat-/ arb'/'arba'-/; arb't /'arba'at-/ hmš /hamiš-/\ hmšt /ham(i)šat-/ It / M - / ; ttt //tittat-/ šb' /sab'-/; šb' t /sab'at-/ tmn / tamānû/ < *tamāniy-; tmnt /tamānît-/ ts" /tiše-/\ tš' t /lis'at-/ 'šr /'aš(a)r-/ < *Cašar-; 'šrt /Cas\a)rat-/

T h e uninflected (masc.) forms of the cardinal numerals 3 - 1 0 can be coupled with nouns of either gender. In the Baal Cycle, the Aqhat Epic and a few other poetic texts, however, fem. numerals are generally used with masculine countables (syntax with 'polarity of gender'). In prose, fem. numerals are used exclusively with the ellipsis of tql 'shekel' and ym 'day'. b. T h e numerals from 11-19 are made up of the units 1 - 9 and the expression for 'ten' (Cšr / Cšrt / esrh). T h e sequence is mosdy 'unit— ten', e.g. (a) hmš Cšr, (b) hmš Cšrh and (c) hmšt Cšrt '15'. Type (a) is only used together with masc. nouns. Type (b) occurs with fem. and masc. nouns. Type (c) is used only with the ellipsis of tql 'shekel' or ym 'day'—here as an ordinal numeral. Numerals 12-19 can also be constructed in the reverse sequence ('ten—unit'). In these cases the unit is always followed by the word kbd which can be rendered 'plus', e.g. csr arbc kbd '14'. c. T h e numeral 20 is formed from the dual or plural form of Cšr ' 10', the tens from 30 to 90 from the plural forms of the numerals 3 to 9: Cšrm, tltm, hmsm, ttm, sb'm, tmnym, ts'm. T h e cardinal numerals 26

Against

SEGERT

1 9 8 4 , §§ 5 2 . 6 ,

62.6,

73.21.

2 1 - 9 9 comprise two or at most three words: the ten, the unit and usually a word linking the ten and the unit, e.g. tt I ttm '66' ( K T U 1.4 = R S 2. [008]+ vii 9) or tmnym tmn kbd '88' ( K T U 4.179 = RS 15.103:14). d. '100' is mit /mi't-/, '200' mitm (dual of mit). T h e hundreds from 300 are formed by connecting a unit and mat /ma'at-/ (pi. of mit), e.g. tit mat. e. '1,000' is alp /calp-/, '2,000' alpm (dual of alp). T h e thousands from 3,000 are formed from a unit and alpm (pi. of alp), e.g. hms alpm '5,000' ( K T U 4.181 = RS 15.106:2). f. T h e word for '10,000' (or 'myriad') is rbt or rbbt. 3.4.4

The verb

3.4.4.1

Introduction

T h e inflected verb differentiates gender, number, person, aspect/tense (imperfective or perfective; antecedent, contemporaneous, subsequent), mood (indicative or volitive [imperative, jussive]), diathesis (active, reflexive, passive) and aspect (e.g. factitive, causative). Gender, number and person are differentiated by various prefixes a n d / o r suffixes. Various verb stems differentiate diathesis and aspect (see § 3.4.4.3); aspect/tense and mood are differentiated by a) subtypes of the prefix conjugation, b) the imperative and c) the suffix conjugation. Alongside the genuine (finite) verbal forms two nominal (infinitive) categories occur in connection with the verb system, i.e. participles and infinitives. T h e y are morphologically and semantically directly related to verbal categories. Besides gender and number their inflection also differentiates diathesis and aspect. 3.4.4.2 3.4.4.2.1

Morphological classes of the basic stem T h e imperative

a. T h e imperative (impv.) is the mood of c o m m a n d in the 2nd pers. Morphologically, it is identical with the short form of the prefix conjugation without the prefix and phonemically monosyllabic, i.e. (V = / a / , / i / or / « / ; the same thematic vowel as in the prefix conjugation). As a word-initial consonant cluster is not tolerated, the impv. becomes bisyllabic, usually by insertion of an auxiliary vowel

after the first radical (— anaptyxis), generally / 2 / , m o r e rarely / « / : e.g. isp /}isfn/ < *'Vsupī{f.sg.) 'collect!' ( K T U 1.107 = R S 24.251+:33 etc.); uhd /"hud/ < *>hud 'seize!' ( K T U 1.82 = R S 15.134:6). b. T h e inflected endings of the impv. are the same as the endings of the 2nd person of the short form of the prefix conjugation: qHVl·, f.sg. qHVlî·, m.pl. q'tVlû (f.pl. not attested); c.du. q'tVlâ. c. Besides the uninflected form of the m.sg. impv.—in line with H e b r e w — t h e r e is probably a lengthened ('emphatic') form qUVla, marked by the suffixed m o r p h e m e /-a/,'21 e.g. sa /sa'a/ < *sa'a (^Ins3) 'raise/lift up!' ( K T U 1.5 = R S 2.[022]+ ν 13; K T U 1.14 = R S 2. [003]+ ii 22). 3.4.4.2.2 T h e prefix conjugation a. 'Prefix conjugation' (PC) is the generic term for various different morphological subtypes which have differing verbal meanings. Inflection is by means of prefixes and suffixes. In morphological terms and functions the following subtypes of the PC can be distinguished (cf. § 3.4.4.2.5):

Form

short form extended short form long form

Function

a) perfective aspect, preterite b) 'jussive' mood jussive/cohortative mood imperfective aspect, present

Abbreviation

PC s p PC s j PC s e PC1

b. T h e prefix c o n s o n a n t s of the P C are: 3.m.sg. y-\ 3.f.sg. a n d 2.m./f.sg. t-; l.c.sg. 3.m.p1. t-/*y-;26 3.f.p1. and 2.m./f.p1. t-\ l.c.pl. 29 3.f.du. a n d 2.c.du. /-. «-; 3.m.du. y-/1-;

27

It is the same morpheme as occurs in the PCfe (cohortative mood); see § 3.4.4.2.2f. 28 Normally a /-prefix (see D O B R U S I N 1981). There was also a variant with a jy-prefix—probably attested only in two cases (KTU 1.4 = RS 2-1008]+ ν 17: yblk 'they should bring' || tblh, K T U 1.4 = RS 2.[008]+ ν 38.40:yblnn 'they brought'). This variant is no (longer) productive and only attested in grammatical parallelism with the 'normal' /-prefix form. 29 Forms with the y- and /-prefix both occur (with almost the same frequency).

c. T h e personal suffixes of the PC are the endings of the short form of the prefix conjugation (PC S ): 3.m./f.sg., 2.m.sg., l.c.sg./pl. -0; 2.f.sg. /-Ī/; 3./2.m.pl. / - « / ; 3./2.f.pl. /-«a/(?); 30 3./2.du. /-ā/. d. T h e PC in the basic stem of the underlying 'strong' roots has the following structure: (paradigm root Vqtl; C = any prefix consonant; V | = prefix vowel; V 2 = thematic vowel). T h e thematic vowel (TV) is / a / , / z / or / « / , the prefix vowel (PV) either / a / (before T V /u/ or til) or lit (before T V /a/). T h e following PCpatterns occur: , und .3] T h e choice of T V is essentially dependent a) on the semantic class of the root (roots with a fundamentally stative meaning usually have the T V / a / , roots with a fientic basic meaning have either t u l or t i t as a T V , b) and on the phonetic quality of the third and second root consonant (roots with gutturals in second or third position often have / a t as TV). e. T h e paradigm of the PC S is therefore (paradigm root ^lqtl, T V / u l ) :

Singular 3.m 3.f. 2.m. 2.f. I.e.

yaqtul-0 taqtul-0 taqtul-0 taqtul-ī c aqtul-0

Plural taqtul-ū taqtul-na(?) taqtul-ū taqtul-na(?) naqtul-0

Dual y/taqtul-ā taqtul-ā taqtul-ā

taqtul-ā (as masc.) - ( ? )

f. T h e forms of the PC S without endings have a morphological variant with the suffixed m o r p h e m e I-at instead of - 0 (= PC s e). 32 It is only attested in connection with jussive forms and so can be termed a lengthened or 'emphatic' jussive. T h e lengthened jussive is attested in an unequivocal spelling a large n u m b e r of times only in the 1st p. sg. In analogy with Hebrew this form can be called 'cohortative'. As the lengthened jussive is not attested for every person and is not in functional opposition to the ordinary jussive (= PC s j), it is not an

30

VERREET

1984, S

3 1 7 - 9 has a different view and postulates a P C L tqltn = /taq-

tulūna/ and a PC Iqtl = /laqtulū/ for the 3.f.p1. However, a PC-form of the 3.f.p1. without the -n ending does not exist. 31 For these vocalic sequences see already B A R T H 1894, 4-5. For the validity of 'Barth's Law' in Ugaritic see especially V E R R E E T 1983b. 32 It is therefore the same morpheme as probably also occurs in Ugaritic on the lengthened impv. (m.sg.); see § 3.4.4.2.1c.

autonomous grammatical category. In other words, unlike Arabic, Ugaritic does not have a specific 'yaqtula-mood'. g. T h e long form of the prefix conjugation (PC 1 ) differs from the PC S by an additional suffixed morpheme. Forms without an ending in the PC S have the ending / - u / in the PC1'; forms with a vocalic ending in the PC'S (except for the 3./2. f.pl.) have the additional ending -n = /-na/ or /-ni/ in the PC 1 :

Singular 3.m. 3.f. 2.M. 2.f. 1 .c.

yaqtul-u taqtul-u taqtul-u taqtul-īna 'aqtul-u

Plural

Dual

taqtul-ūna taqtul-na (?) taqtul-ūna *taqtul-naa (?) naqtul-u

h. In earlier research the question was hotly debated whether in Ugaritic there was also a long form of the prefix conjugation with the pattern , comparable to Akkadian iparrVs, Ethiopie y^qathl or similar formations in modern South Arabic languages. 33 F e n t o n (1970) and M a r c u s (1975, 75-104, esp. 97ff.) demonstrated independently, however, that this category does not exist in Ugaritic. T h e Ugaritic texts published over the last twenty years provide absolute proof of this. 3.4.4.2.3 T h e suffix conjugation a. T h e Suffix conjugation (SC) is not a homogeneous category in Ugaritic, as is also the case in other West Semitic languages. T h e r e is a fundamental difference between SC-forms with stative meaning and those with a fientic (perfective, mostly preterite) meaning. T h e former can be called 'statives', the latter 'perfects' (abbreviations: 'SCs' and 'SCp'). T h e subtypes mentioned also differ from each other morphologically by different thematic vowels (see § c). b. T h e paradigm of the suffix conjugation is as follows:

33

See

esp. GOETZE

1938,

296-309.

Singular 3.m. 3.f. 2.m. 2.f. I.e.

qatVl-a qatVl-at qatVl-ta qatVl-ti qatVl-tu35

Plural qat VI- ū qatVl-ā/ ū(?) qatVl-tum(Vf4 *qat Vl-tun(n)a *qat Vl-na/ ü

Dual qatVl-ā qatVl-tā qatVl-tumā

qatVl-tumā (also masc.) qatVl-na/iyā

c. T h e thematic vowels of the SC are / a / , / i / and / u / . / a / is reserved exclusively for the fientic-perfective subtype of the SC (SCp), / u / for the stative subtype of the SC (SCs). I i i occurs in both subtypes. In the fientic subtype, however, it is limited to roots with I I / III guttural, where the thematic vowel of the PC is / a / . T h e system of thematic vowels in the SC and their equivalents in the PC can be summarized as follows: SC qati/ ula - PC Ciqtal (stative); SC qatala PC Caqtu/il (fientic); SC qatila - PC Ciqtal (fientic II/III-guttural). 3.4.4.2.4 Finite Verb Forms with Energie Ending a. Besides the inflectional endings, finite verb forms—especially in poetry and in object suffixes—often exhibit a so-called energic ending,36 spelled either -n or -nn. At least two perhaps even three different alloforms of the energic m o r p h e m e can be distinguished (energic types I / I I / [ I I I ] ) . As the energic endings may occur, basically, on all finite verb forms, whether they are indicative or volitive, the energic is not a mood in the strict sense. b. By far the the most commonly attested allomorph of the energic is -n — /-(á)nnV/ (= energic type I). It may stand alone and or before the 3rd pers. sg. pronominal suffixes. In combination with 3rd pers. sg. suffixes, the ending is -nh, to be vocalised as I-anna-hüI or / -annaha/ respectively. c. T h e r e is also an energic allomorph: -nn = l-ninl(?) (= energic type II). It occurs exclusively in combination with 3rd pers. sg. suffixes. T h e initial consonant, I hi, of the pronominal suffix is thus

34 Alternatively /i/- vowel, i.e. / qatVl-tim(V)/. Similarly for 2.f.p1. (/qatVl-tin(n)ā/) and 2.c.du. (/qatVl-timā/). 35 Alternatively: /-til (as in Canaanite). In favour of /-tū/ however is that the independent personal pronoun l.c.sg. in Ugaritic also ended in / - « / (/'anāku/). 36 On this topic see especially V E R R E E T 1 9 8 8 , 7 9 - 9 8 and K R E B E R N I K 1 9 9 3 .

always assimilated to the second /n/ of the energic ending: -nn — /-nVnnü/ < *-nin-hu or /-nVnna/ < *-nin-ha respectively. d. Perhaps Ugaritic also had a third allomorph of the energic, i.e. -n = /-an/ (= energic type III). T h e orthographical proof for this allomorph has not yet been produced. 3.4.4.2.5 Aspect and tense Verbal aspects and (relative) tenses are mainly differentiated by the subtypes of the PC and the fientic variant of the SC. T h e functions of these categories can be set out in the following table (paradigm root ^lqtl, 3.m.sg.):

anteriority contemporaneous posteriority

yaqtul (PCsp) *yaqtul(a) (PC s j/e)

perfective

imperfective

qatala (SCp) *qatala (SCp) qatala (SCp)

yaqtulu (PC L ) yaqtulu (PC L ) yaqtulu (PC L )

All the fields on the right = imperfective column of the table are filled by the long form of the prefix conjugation (PC 1 ). T h e left = perfective column includes the short form of the prefix conjugation (PC S ) and the perfective-fientic suffix conjugation (SCp). In the field 'perfective anteriority', the P C s p and the SCp have practically the same function. T h e field 'perfective-contemporaneous' is empty because facts which occur simultaneously are essentially imperfective. Only a special function of the SCp, i.e. the function of the so-called 'performative perfect', 37 can be placed in this field. T h e field 'perfective-posteriority'—with reference to indicative statements—is only covered by the SCp. T h e function of the variants of the PC S also in that slot is exclusively volitive (jussive). T h e table shows clearly that the PC1· is always imperfective and the PC S is always perfective. As the PC L is used for simultaneous situations, this category is conventionally labelled the 'present'. As, on the other hand, the indicative P C s p generally expresses previous events, this category is conventionally called the 'preterite'. These labels, which suggest an opposition of tense between PC 1 ' and PC S , are not in fact correct, as the PC1- can also denote previous events, 37

E.g. I rgmt Ik Ί hereby surely tell you . . .' (KTU 1.2 = RS 3.367 iv 7).

provided that they are imperfective. An imperfective presentation is demanded especially for situations which are marked by the features of plurality or repetition. 38 T h e P C s p occurs for certain only in narrative verse and is used there as the usual narrative form for single and instantaneous actions of the past. 39 In other genres perfective actions of the past are obviously always expressed by the SCp. However, the SCp also occurs in narrative poetry, sometimes as a free variant of PC s p. 4 0 3.4.4.2.6 Moods In the Ugaritic verbal system the 'indicative' (= declarative mode) and 'volitive' (= wish and c o m m a n d mode) moods are differentiated. T h e categories PC s p, PC1- and SC (SCp and SCs) are used for indicative statements. 41 T h e following have volitive functions: a) the imperative, b) the PC S ] (jussive) and the PC s e (cohortative) as well as—relatively rarely—c) both subtypes of the suffix conjugation, i.e. SCp and SCs. T h e r e is no specific use of mood in subordinate clauses. Volitive moods, i.e. P C s j and PC s e, occur only in subordinate clauses with volitive (final) meaning. In Ugaritic there is no specific 'subordinating mood', comparable to the Akkadian 'subjunctive', which only occurs in dependent clauses. 42 3.4.4.2.7 Participles T h e pattern for the formation of the active participle of the basic stem (G-ptc. act.) is . For the passive participle of the basic stem (G-ptc. pass.) probably the pattern is generally used, as in Canaanite. 4 3 38 E.g. p'rih I tmgyn hdm / rišh I ymgy apsh 'His feet did not reach the footstool, his head did not reach as far as its (upper) end' (KTU 1.6 = RS 2.[009]+ i 59-61 [general-continuing event, expressed by the PCL]). 39 See for instance 1šu ilm rašthm 'the gods lifted up their heads' (KTU 1.2 = RS 3.367+ i 29). The morpho-syntactic autonomy of the category PC s p is, however, questioned by some scholars; for discussion see M . S . S M I T H 1994, 39-41. According to Smith, PO s p and PC 1 —although in morphological contrast—are free variants. 40 On the parallelism between PC s p and SCp see M.S. S M I T H 1994, 49-51 and 1995, 797-9. On other functions of the Ugaritic S C see M . S . S M I T H 1994, 45-57 and 1995. 41 For the use of these categories see § 3.4.4.2.5. 42 For a different view cf. V E R R E E T 1 9 8 8 , esp. 8 - 1 0 . 43 Over the years scholars have repeatedly stated that in Ugaritic there could also or only be other patterns for the passive participle of the basic stem. The discussion centres on the patterns (G-Ptc. pass, in Aramaic) and <maqtūl> (G-Ptc. pass, in Arabic). So far, however, no proofs have been provided.

3.4.4.2.8 Infinitives/Verbal nouns As in Hebrew, an absolute infinitive and a construct infinitive may be differentiated. T h e former corresponds syntactically to a noun in the absolute state, the latter to a noun in the construct state or a noun after a preposition. T h e infinitive (inf.) of the basic stem generally has the pattern . Besides this there is in Ugaritic a series of differently constructed verbal nouns of the basic stem. T w o patterns are noteworthy. O n e is , which occurs several times in syllabic spellings, e.g. ni-ib-rù /nigru/ 'guard' (Ug 5, 137 = RS 20.123+ i 5').44 T h e other is , which occurs only in I - w / y roots and ^1hlk, e.g. sat /si'at-/ 'going out, expression' ( K T U 1.4 = RS 2. [008]+ vii 30.32); d't /dacat-/ 'knowlege' ( K T U 1.2 = RS 3.367+ i 16.32); Ikt /likt-/ 'going' ( K T U 1.10 = RS 3.362+ ii 28, 29). Whereas the pattern is used for both the absolute and the construct infinitives, other patterns can only denote the construct infinitive. 3.4.4.3 T h e system of verbal stems: basic and derived stems a. T h e Semitic languages use a large number of different patterns to express aspect and diathesis, called 'verbal stems'. T h e simple basic stem of Semitic is morphologically unmarked. T h e 'derived' verbal stems are, instead, indicated by specific morphological markers. b. Ugaritic has the following ten verbal stems, which can be arranged as follows: symbol

stem

functions

G Gp Gt D Dp

basic stem passive basic stem reflexive basic stem intensive passive intensive

basic lexical function passive of G45 reflexive of G etc.46 intensive, factitive etc. passive of D

For further examples see H U E H N E R G A R D 1987b, 305-6. On occurrences of the Ugaritic Gp-stem see M A R C U S 1971. 4
see 47

KREBERNIK

1991.

In stative verbs the function is chiefly factitive/causative (e.g. 'be good': 'to make [someone/somethingl good'). In intransitive-fientic verbs the D-stem chiefly provides transitivity. In transitive-fientic verbs it strengthens or pluralizes the verbal

(table cont.)

symbol

stem

functions

tD Š

reflexive intensive Š or causative passive Š reflexive Š Ν or passive

reflexive of D etc. causative passive of Š reflexive of Š etc.48 reflexive, passive49 etc.50

p

l St Ν

Four of the ten patterns can be called 'cardinal stems': G, D, Š and N. G is unmarked (= Heb. qal); D is marked by gemination of the middle radical (ξ Heb. piel); Š is marked by the prefix /- (= Heb. hiphil in function); 51 Ν is marked by the prefix n- (= Heb. niphal). All the 'cardinal stems'—except for Ν—have both a passive and a reflexive variant. T h e reflexive forms have an additional element t\ in the Gt it comes after the first radical, in tD (= Heb. hithpael) before the first radical and in Št directly after the causative marker Š-. T h e passive forms, i.e. Gp, D p (= Heb. puai) and Šp (Ξ Heb. hophal in function) are differentiated from the corresponding active forms only by different vowels. As this characteristic is usually not expressed in the alphabetic spelling, in most cases examples of the passive stems can be determined only from syntax or context. Some authors 52 doubt 5 3 their existence in Ugaritic. c. T h e paradigm of the verb stems (forms are vocalized; finite forms are always 3.m.sg.; ptc. and inf. u n i n f e c t e d ; n.o. = no [certain] occurrences): meaning (plurality of agents or objects; repetition of an action). The D-stem is also used for denominative meanings. 48 On the Š, Šp and Št in Ugaritic see T R O P P E R 1990a, 2 1 - 1 1 1 . 49 Chiefly or only in prose texts. 50 E.g. reciprocal, ingressive and inchoative. 51 As non-sibilant causative morphemes occur in other Northwest Semitic languages (e.g. the causative marker h- in the Heb. hiphil), repeated attempts have been made by sholars over the years to prove these types of causative also occur in Ugaritic. The attempts in question have not been convincing, however (see T R O P P E R 1990a, 113-82). 52 The most uncompromising opponent of the existence of the passive stem in Ugaritic is V E R R E E T 1 9 8 5 , 3 2 4 - 3 0 . It should be noted, however, that all the central Semitic languages have passive stems. 53 There are no other verbal stems in Ugaritic. On the so-called 'lengthened stems' (L) see under § 3.4.4.4f and § 3.4.4.4h, on the so-called 'reduplicated stems' see under § 3.4.4.4i. In Ugaritic there are no stems corresponding to IX, XI or XII-XV of Arabic.

PC s j G Gp Gt D Dp tD Š St Ν

3.4.4.4

yaqtu/il yiqtal yuqlal yiqtati/ al54 yuqattil36 yuqattal ? yVtqattVl yušaqtil39 yušaqtal y Vštaqtil yiqqatilM

impv.

se

q1 tu/il q'tal

qatala qati/ula quti/ ala ('î)qtat(a)la qattila qutti/ala ? ('i)tqatti/ ala57 šaqtila šuqta/ ila

n.o. ('i)qtatil qattil

n.o. n.o. šaqtil

n.o. n.o. naqtVl ?

ptc. qātil (act.) qatül (pass.) —

muqtatil ? muqattil muqattal ?

inf. qatāl

ρ tVqtatil55 quttal

n.o.

n.o. šVqtVl

n.o.

musaqtil mušaqtal muštaqtil

naqtala

n.o.

naqtal

n.o. n.o.

Morphological peculiarities of the 'weak' verbs

a. Five I - ' verbs have irregular G-PC-forms of the type yuC2C3 — /yôCi VC3/ instead of or as well as yiC2C3 - /yaCC2 VC:ì/ (cf. § 3.2.2e): ^bd 'to perish', ^hb 'to love', ^hd 'to seize', Ψkl 'to eat', yl'sp 'to gather'. 6 1 b. I-h verbs usually have strong forms. Exceptions are verbs with / / / as the second radical, i.e. ^lhlk 'to go' and ^ìhlm 'to strike, hit'. Both verbs have G-PC-forms without /h/, e.g. ylk- /yalik-/ o r y l m - /yalum-/. Whereas the remaining forms from VA/m are strong (e.g. G-impv. him- /húum-/), ^Jhlk is weak in other ways, i.e. it produces forms without /h/: G-Impv. Ik- /lik-/; G-verbal noun Ikt /likt-/; Gt-PC ytlk /yitalik/. Š-PC-forms from ^lhlk are instead strong: ashlk Γašahlik/ ( K T U 1.3 = R S 2.[Ol4]+ ν 2, etc.).62 54

In forms with endings there was probably syncope of the corresponding vowel: /yiqlatlū/ < *yiqtatVlū (cf. § 3.3.5.2b). The same applies to other forms of the paradigm with similar syllabic structure. 55 Occurrences: Imthsh 'her fighting' (KTU 1.3 = RS 2. [014]+ ii 19); trntbs 'fighting' (KTU 1.3 = RS 2.[014]+ ii 29); tljtsb 'quarrel' (KTU 1.3 ii 20.30). 56 /y/t/nuqattil/·, l.c.sg. but /'aqattil/ < *'uqattil (vowel harmony; see § 3.3.5.2a). 57 Alternatively: /taqatti/ala/. The only certain example: w Ikms /wa-t(a)kamm Vsa/ 'he fell to his knees' (KTU 1.12 = RS 2. [012] ii 54). Another possible example: tmz' (KTU 1.19 = RS 3.322+ i 36.46). 58 Possibly /tuqattil/·, cf. the uncertain syllabic spelling: tu-a-pí-[ku(?)] /tuhappiku/ (ιUg 5 137 = RS 20.123+ ii 23). 59 /y/t/nušaqtil/\ l.c.sg. however /'ašaqtil/ < *'ušaqtil (vowel harmony). 60 < *yinqatil (chiefly undifferentiated by spelling). 61 The same verbs also have similar forms in Hebrew (verbs with 'weak aleph' as the first radical). On the topic see esp. T R O P P E R 1990b, 367. 62 On the morphology of I-h verbs see T R O P P E R 1990d.

c. In I-n verbs and Mqh 'to take', the first radical—when vowelless—is assimilated to the following consonant, e.g.ygr /yaggur-/ < *yangur- (Vngr 'to guard' G-PC) or yqh /yiqqah/ < *yilqah (Mqh G-PC). 6 3 In most I-n verbs the G-impv is formed without the first root, e.g. la/sa'a/ < *ša3a (^lnš' 'to raise', lengthened impv. m.sg. [ K T U 1.4 = R S 2.[008]+ viii 5]). šu /ša'ū/ (VraT, m. pl. [ K T U 1.2 = R S 3.367 i 27 etc.]), sk /safā/ (<nsk 'to pour', f. sg. [ K T U 1.3 = R S 2. [014]+ iii 16 etc.]) as well as qh/qah/ Çilqh, m. sg. [ K T U 1.4 = R S 2. [008] + ii 32, etc.]). T h e verb ^lngr 'to protect' constitutes an exception: impv. m. sg. ngr /mgurā/ 'look out!' ( K T U 1.4 viii 14 [alternatively: N-impv. /naggVrā/ < *nangVra\). d. III- J verbs are—besides \\\-w/y verbs—of central importance for understanding the Ugaritic verbal system as in principle in such verbs the spelling allows verbal aspects and moods to be differentiated clearly. PC1· forms occurs with the spellingyC\C 2 u (3.m.sg); for forms of the PC S , however, the spellingyC\C 2 i (3.m.sg) is expected. In fact, though, the situation is more complex as in Ugaritic word-final aleph was no longer reliably pronounced. 6 4 O n this basis only verbs with the PC thematic vowel H / permit an unequivocal differentiation of the underlying classes: e.g. P Q· ysu /yasi'u/ 'he goes out' i^yf) versus PC S ysi /'yasiV (or /yasV/ < *yasï) 'he should go out / he went out'. In PC S forms with the thematic vowels / « / and / a / this differentiation is not given with certainty: spellings such as ybu (ynš') and yšu (ynš1) can be understood as PC 1, (/yabu'u/ or /yišša?u/), but possibly they could also be PC S forms (/'yabû/ < *yabu'·, /yisšá/ < *yinša').65 e. Verbs with /w/ or /y/ as the first, second or third radical present several forms which are irregular with respect to the forms of the paradigm of the 'strong verb', as the semi-vowels /w/ or / y / , depending on position in the syllable, can occur either as consonants or as vowels. f. T h e paradigm of I - w / y verbs, which in Ugaritic includes ^lytn 'to give', is characterised by the occurrence of forms both with and without a (consonantal) first radical. If the word begins with a semi-

63 However, there is no assimilation of / / / in the N-stem: SC nlqht /nalqahat/ (KTU 4.659 = RS 19.166:1). 64 See § 3.2.2d (quiescent aleph). 65 O n the grapheme < u > for / 0 / < *aJ see § 3.2.2d. There are no attested forms of the spelling ysi. On the topic see T R O P P E R 1990c.

vowel it is normally retained, as long as there no aphaeresis of the whole initial syllable occurs as in impv. G. As word-initial *w in Ugaritic generally becomes /y/ (see § 3.3.5.1a.), except before a / u / vowel (inf. D), it is not possible to distinguish I-w roots from I-y roots by spelling in most cases. In non-initial position the semi-vowel (Iwl) is retained only after a / « / - v o w e l (D-PC); in the other cases either it causes a lengthening of the preceding vowel (e.g. in forms of the Š stem) or it disappears unreplaced (e.g. G - P C forms). T h e function of the inf. cstr. is generally filled by verbal nouns of the pattern (see § 3.4.4.2.8). Significant forms of the paradigm for l-w verbs i^yrd < *wrd 'to descend'): 66 PC S G Gp Gt D S Ν

yarìd yurad yittarid yuwanid yusôridbl

impv.

SC

ptc.

inf.

nd

yarada

yārid

yarād

šoúd

sonda nôrada

musôrid

wurrad

g. In forms of II-w/y verbs, the second (weak) radical never occurs as a c o n s o n a n t but always causes a lengthening of the original preceding or following vowel, G"PC' 3.m.sg. /yaqûmu/ < *yaqwumu, G - S C /qâma/ < *qawama (yqurni).68 T h e thematic vowel of the Ο ΡΟ, is usually / « / in II-«; verbs (e.g. /yaqûmu/), in II-y verbs usually H/ (e.g. lyašîtuI). Instead of the ('normal') intensive stems (D, D p and tD), I I - w l y verbs form stems which are marked by lengthening the vowel between the first and second radical and reduplication of the third radical (= Heb. polel, polal, hithpolel). T h e y are conditioned variants of the 'normal' intensive stems (D, D p , tD) and have the same functions. H e r e they are denoted by the symbols D*, Dp* and tD*. 69

66

The paradigm for I-y verbs is still blank at many points due to the lack of significant forms. 6 ' However, l.c.sg. /'aš(md1. <>8 Conventionally vocalized as /yaqūmu/ or /qāma/. 69 These stems are conventionally called 'lengthened stems' and denoted by 'L', 'Lp' and 'tL'. The term 'lengthened stem' has been taken from Arabic but is

Significant forms of the paradigms for II-«; and II-y verbs i^qwm 'to rise' and ^Ìšyt 'to place'): 70 PC1

impv.(m.sg./pl.)

yaqûmun yašîtu yuqo/âmimu yuqô/ âmamu yitqô/ âmimu yuš(a)qîmu

G D* Dp* tD* Š

SC

ptc.

qum/ qûmū šit/šîtū qo/âmim

qâma12 šâta

qâm-(?)73 sat-(?) muqo/âmim-

šaqim/šaqîmū

šaqîma

muš(a)qîm-

inf. abs. qûm-(?)

JSJK?)

h. T h e paradigm of I I I - w / y verbs is marked by the occurrence of forms both with and without a (consonantal) third radical. In originally syllable-closing position (also when final) the weak third radical is always vocalic (contraction of diphthongs *iy > / î / , *ay > /ê/, *uw > /û/, *aw > /0/ [see § 3.3.5.3a]). In intervocalic position it is partly preserved as a consonant (e.g. ybky = /yabkiyu/ 'he weeps' [G-PC 1 · 3.m.sg.]), partly the relevant triphthong is contracted. Which triphthongs in particular remain and which are contracted is still not clearly explained 74 in spite of extensive research. 70 It seems that occasionally paradigmatically identical forms occur both with and without contraction. Forms III-«; and III-jy verbs are usually orthographically identical. Some indications of different paradigms of these two classes are provided however by forms such as atwt Γatawat/ (y'tw) 'she came' ( K T U 1.4 = R S 2.[008]+ iv 32) versus mgyt /magayat/ ^mgy) 'she arrived' ( K T U 1.4 = R S 2. [008]+ ii 23, etc.).

misleading. Ugaritic 'L'-stems correspond to the III and VI verbal stems of Arabic neither in form nor in function. 70 The II·? forms in the derived stems are the same as II-w forms. 71 Shortening of the long vowel between the first and third radical when the syllable in question is closed, e.g. PC * /yaqum/ and PC L/S 3.f.p1. /laqumnā/·, analogically, impv. m.sg. /qum/, PC S II-j /tašit/, impv. m.sg. U-y /šit/. 72 Before personal endings beginning with a consonant, either Iqam-I < *qâmor /qâ/ûmâ/ (cf. Heb. bīnātī [byn 'to understand', G-SC l.c.sg.]). The same applies analogously to the SC-forms of the derived stems. 73 Uncontracted forms of the type *qāyim or *qā'im are unattested. 74

75

See

SIVAN

1984

and

VERREET

1985,

330-41.

The rules concerning triphthongs in respect of noun forms (cf. § 3.3.5.3b) cannot be transferred to verbs without being modified.

O n the basis of comparative linguistics, one can conclude that in P C as well as in SC different thematic vowels existed (III-w: / u / and /a/; III^: /i/ and /a/). In III-y the commonest type was definitely *yaqtiy- (PC) versus *qataya (SC), in III-w *yaqtuw- (PC) versus *qatawa (SC). Significant forms of the lll-y paradigm, basic stem (ΛIbny 'to build'): PC yabmyu (PC1·); yabnf6 (PCS). impv. b'nî (m.sg.); ÏÏniyî or bvnî (f.sg.); bVniyū or bVnû (m.pl.).77 SC

ptc. inf.

banaya or (rarely) banâ (3.m.sg.); banayat or banât (3.f.sg.); banêta

(2.m.sg.); banayū (3.m.pl.). bāniyu/a (m.sg. nom./acc.); bāniyi (m.sg. gen.); bānît- (f.sg.).78 banāyu/i/a

or

banû/î/â.

i. T h e paradigm of weak geminate verbs (II-gem.) still presents m a n y uncertainties. In the basic stem they are both 'strongly' constructed forms, i.e. forms with reduplicated 2nd = 3rd radical, and 'weak' forms, i.e. forms with doubled 2nd = 3rd radical. Doubled consonants, when final, are reduced to a single consonant. T h e distribution of strong and weak forms is not as in classical Arabic. However, there are parallels with the H e b r e w paradigm for II-gem. Significant forms of the II-gem. paradigm, basic stem iysbb 'to go round'): PC

with endings: yasubb- (e.g. PC1· 3.m.sg. yasubbu).™ without endings: yasbub or yasub < *yasubb.

impv. SC ptc.

sub < *subb (m.sg.); subbū (m.pl.). sabba (3.m.sg.); sabbata (2.m.sg.);80 sabbanVyā (I.e.du), sābib-.

Forms of the derived stems—as far as can be ascertained—are strong throughout. In agreement with H e b r e w (pölel = pö c e c ) it is possible that instead of or alongside 'usual' D-stems (D, Dp, tD) the Ugaritic II-gem. verbs form so-called 'lengthened stems' (D*, Dp*, tD*), 8i e.g. 76 Analogously III-w: PC s j 3.f.sg. tdu /tad'û/ < *tad'uw 'she flew' (KTU 1.16 = RS 3.325+ vi 6.7). 77 Analogously III-w: du /ds'û/ (m.sg.) (KTU 1.19 = RS 3.322+ iii 28); di /diÎ/ < *ds'iyi (f.sg.) (KTU 1.16 = RS 3.325+ ν 48); du /d$'û/ < *ds'uyū (m.pl.) (KTU 1.19 = RS 3.322+ iii 14). 78 Analogously III-«;: dit /dâ'ît-/ (KTU 1.108 = RS 24.252:8). 79 II -gem. verbs which are also I-« provide an exception: all the forms have a reduplicated 2nd = 3rd radical, e.g. tlt/latutā/ < *lanlutā 'they (3.f.du.) trembled' (KTU 1.3 = RS 2. [014]+ iii 33, etc.). 80 Cf. Heb. sabbotā < *sabbātā. 81 For the term 'lengthened stem' and the sigla see § g.

D*-PC *yusabib- instead of D-PC *yusabbib- (vowel lengthening instead of consonant lengthening). T h e orthography is ambiguous. j . Only a few Ugaritic verbs have four radicals. T h e three most important formations are Crr/l-C3-C4,82 b) Ci-C 2 -C r C 2 83 u n d c) Cr C2-C3-C3.84 T h e forms of types (b) and (c) are conventionally understood as 'reduplicated stems' (R-stems) of roots with three or two radicals. 85 3.4.5 3.4.5.1

Particles86 Adverbs

a. Adverbs of place: tm, tmt, tmn, tmny. 'there'; 7, cln 'above'; I pnm 'before'; b'dn 'behind'; atr '(directly) after'; pnm 'within/inside'. b. Adverbs of time: ht, htm, cnt 'now'; idk, ap(.)hn, apnk, b km 'then; thereupon; a!}r 'after(wards), later', atr '(directly) after'; ahrm 'in succession'(P); clm 'on the following/next (day)' (alternatively: 'further'). c. Modal adverbs: k, kd, kmt 'thus, in this way'; Ibdm 'alone'. d. Interrogative and indefinite adverbs: iy, i, 'where?'; an 'whither?'; ik, ikm, iky, 'how? why?'; Im 'what for? why?'. 3.4.5.2

Prepositions

T h e Ugaritic prepositions mostly denote an adverbial position but— in connection with certain verbs—can also be used directionally. T h e y can then fundamentally express both directions, terminative and ablative. 87 82
a. monoconsonantal prepositions: b (also: by), syll. bi-i /bĪ/ 'in, at, on, with, from'; / (also: ly), syll. le-e /ÏÏ/ (or / l e / ) 'towards, for, against; from; away from; to'; k /ka/ with the (lengthened) variant kmm 'as, like'. b. prepositions formed from bi- or triconsonantal roots: yd /yada/ 'next to, together (with)'; 'm /'imma/ '(together) with; towards'; bn /bêna/ 'between'; cd /cadê/('?) 'until'; 7 /Calê/â/ 'upon; towards; down from; away from'; tht /tahta/ 'under, beneath'; qdm /qudāma/(?) 'before; in front of'; b'd /ba'da/ 'behind'; atr /'at(a)ra/ 'in the wake of; (directly/immediately) after'; tk /tôkâ/(?) '(right) in the midst of; into'; qrb /qarba/ 'in the interior; into'. c. Composite prepositions: (preposition b or / + noun): bd /bâdi/ < *bi yadi ' i n / f r o m the hand of; from'; b tk /bi tôki/ 'in the midst of'; b qrb /bi qarbi/ 'in the midst of'. I p /li pî/ 'according to, in the m a n n e r of'; I pn /li panî/ 'to the front of; before; before (temporal); away from (spatial)'; I p'n /li paenê/ 'at (both) feet of; (low) before'; I ir /li iân/ 'on top of; on; onto (movement); from o f f / o n (movement)'; I bl /li balî/ 'without'. d. Prepositions can be lengthened by the enclitic particles -m or -η90 with no essential change in meaning. T h e forms b-m, k-m, l-m, cm-m, l-n und cm-n are attested. T h e y are especially favoured in poetry. 3.4.5.3

Conjunctions

a. coordinating: - w /wa/ 'and; but': copulative conjunction. p /pa/ 'and t h e n / t h e r e u p o n / c o n s e q u e n d y ' : copulative conjunction; it marks a temporal or logical sequence. 90 - ap (extended variant: apn) 'thus, just as; even'. 91 - u /'0/ < *}aw 'or': disjunctive conjunction. b. subordinating: - ahr 'after': to introduce a temporal clause. - id 'when; as soon as': to introduce a temporal clause. c d 'while; as long as; until': to introduce a temporal clause. 88 89 90 91

See § d. See § 3.4.5.9. On Ugaritic p see W A T S O N 1990e, 1994e. Also functions as an asseverative particle.

- hm /him/ with the phonetic variant im /'im/ 'if, in case': to introduce a conditional clause. k /kī/ (variant spelling: ky) 1. 'because': to introduce a causal clause; 2. 'that': to introduce an object clause; 3. 'as, when i f : to introduce a temporal or conditional clause. - km /kīma/ 'as; as soon as'. - him Ίο!; as; as soon as'. 3.4.5.4

Interjections

- Presentation particles ('lo!'): hn (extended variants: hnn\ hnny); hi (extended variants: him, hin, hlk); mk.92 - vocative particles: y /yā/\ l (cf. § 3.4.4.2.4c). i /'ī/ 'truly!' (only in oaths); an 'oh!' (exclamation). 3.4.5.5

Asseverative particles

- k, al, dm, I, mc: 'truly!; certainly!'. 3.4.5.6

Optative particles

- I /lū/ (alternatively: /la/): sive (PC s j/e). - ahl: 'alas!; if only!'. 3.4.5.7

proclitic optative particle before a jus-

Negatives

- / / l a / : for negating words and verbal clauses. 93 - al /'al/: for negating volitive verbal clauses (only before P C s j / e ) . bl /bal/ (extended variant: bit): for negating words and noun clauses; in verbal clauses only in an interrogative sense. 3.4.5.8

Existential particles

- it /'itê/ < *'itay: 'there is/are'. - in /'êna/ < *'ayna (extended variants: inm; inn): 'there is/are not'.

92

Alternatively: 'then, after that' (adverb of time).

93

WATSON

1991C.

3.4.5.9

Enclitic particles

T h e most important enclitic particles in Ugaritic are -m,94 -n, -y, -k and -t. T h e enclitics -m and -n, which are by far the most frequent, serve generally to emphasise certain constituents of a sentence. T h e enclitic -y obviously acts as a marker of direct speech; 95 -k and -t occur chiefly in connection with pronouns and adverbs. 96 (Translation: W.G.E.

94

Watson)

O n which see W A T S O N 1992c, 1 9 9 4 F . See T R O P P E R 1994a. 96 Note the special abbreviations: C = any consonant; c. = common gender; obi. = oblique case; PC = prefix conjugation; PC S = short form of the prefix conjugation; PC s e = lengthened short form of the prefix conjugation (cohortative); PC s j = short form of the prefix conjugation with jussive function; PC s p = short form of the prefix conjugation with perfective-preterite function; PC1' = long form of the prefix conjugation; PV = vowel of prefix; SC = suffix conjugation; SCp = fientic sub-type of the suffix conjugation; SCs = stative sub-type of the suffix conjugation; syll. = syllabic; T V = thematic vowel; V - any vowel. 95

4

U g a r i t i c

W i l f r e d

4.1

L e x i c o g r a p h y

G . E .

W a t s o n

Previous work

Not unexpectedly, the meanings of Ugaritic words have been a matter for discussion and analysis right from the earliest days of Ugaritic studies and it is due to the difficulties inherent in these texts that many words remain unexplained to this day. Several surveys of previous studies are available 1 and there is no need to repeat all this material here. As yet, the only complete dictionary is Aistleitner's Wörterbuch der ugarìtischen Sprache (1963, etc.). T h e only other comparable 'dictionary' is the glossary which formed part of Gordon's series of handbooks to Ugaritic: with each new edition, the glossary was revised as new texts were discovered and different solutions were proposed. 2 Like the Wörterbuch, Gordon's glossary listed all the words found in the Ugaritic texts discovered at the time, including personal and place names. Partial glossaries are to be found in Segert's g r a m m a r ( S e g e r t 1984, 175-205) and in the various translations of the Ugaritic texts now available. 3 Specialised glossaries are included in studies on the following: the hippiatric texts, 4 the ritual texts. 5 Also important are the studies of prepositions 6 and the particles 7 and in spite of its age, d e M o o r 1971 remains a mine of lexical information. Of particular interest are the study of terms connected with textiles used in the Ugaritic texts ( R i b i c h i n i - X e l l a 1 9 8 5 ) and the discussion of terms for sheep and goats ( d e l O l m o L e t e 1 9 9 3 , 1 8 3 - 9 7 ) and of terms for sacrifice ( d e l O l m o L e t e 1 9 9 5 ) . Useful, too, are Pardee's listing of lexical items with bibliography ( P a r d e e 1987) and the studies of syllabic spellings. 9 T h e personal names of 8

1

DE M O O R

2

GORDON

3

DRIVER

not 4

TO

1 9 7 3 ; DEL O L M O L E T E 1965

1955;

(1967);

1940;

GIBSON

1978;

1947

SANMARTIN and

1988; WATSON

1995a.

1955.

DEL O L M O

LETE

1981;

TO

II

and

C A R T U

but

I.

COHEN

-

SIVAN

1983;

PARDEE

1985.

See

also

SANMARTIN

1988b

and

COHEN

1996. 5

DEL O L M O

6

PARDEE

LETE

1975,

7

AARTUN

1974,

8

See

review

9

HUEHNERGARD

the

1992a,

1976,

20-1

=

1999,

00-0;

XELLA

1981.

1979.

1978. b y DURAND 1987b;

SIVAN

1990. 1984a;

see review by

HUEHNERGARD

1987a.

Ugarit are also a source of lexical items even though their meaning may not have been noticed overtly either by those who gave them or by those who bore them. T h e classic collection by G r o n d a h l ( G r o n d a h l 1967) is a useful if somewhat dated reference work in this respect. Some recent studies have provided additional material. 10 T o p o n y m s also contribute lexical items but their origins are more difficult to ascertain." Several series of articles on Ugaritic lexical problems have been written by various authors, some of which are to be continued, 1 2 as well as sets of studies on Ugaritic semantics. 13 T h e r e are many notes and articles on individual words or groups of words which cannot be listed here. 14 It is very helpful when a study is devoted to words belonging to a particular semantic field: sociology ( R a i n e y 1963), fabrics and dyes ( v a n S o l d t 1990), sacrifice ( d e l O l m o L e t e 1995), crafts ( S a n m a r t i n 1995) and the army ( V i t a 1995a). For various reasons, some words receive more attention than others, for example, words which occur in the mythological texts. 15 A reverse glossary (English-Ugaritic) is provided in UT, 530-7. 1 6 In recent years actual dictionaries are starting to be published. O n e is the Diccionario de la lengua ugarítica (DLU) by del O l m o Lete and Sanmartin, a two-volume work of which the first volume has appeared and the second is at an advanced stage of preparation. Another is Cohen's Comprehensive Dictionary of the Ugaritic Language, as yet unpublished. 1 7 T h e third such lexicon (UHw), ] R which was initiated in Münster at Ugarit-Forschungen several years ago, will soon be ready for publication but is available in the form of a word-list 19 ( D i e t r i c h - L o r e t z 1996b). 1990a, 1990b, 1993, 1995b, 1996a.

10

WATSON

11

S e e ASTOUR

1 9 8 7 ; cf. §

12.2.

and D I E T R I C H - L O R E T Z SANMARTIN; bibliography in SEL 5 1988, 2-12. Many of their studies are concerned with differentiating homonyms. 12

DIETRICH -

AARTUN

1968,

LORETZ

1984,

1985,

1 9 9 1 ; BERGER

1978, 1979, 1980, 1988;

XELLA

GREENFIELD 1 9 6 7 , BADRE

BORDREUIL

1984, WATSON 13

14

See

now

1978,

1 9 7 0 ; MARGAI.IT

1978b, 1980. See also MUDARRES

1 9 8 2 ; SANMARTIN DE

MOOR

1977,

1965, 1979,

AJJAN - V I T A L E 1 9 7 6 ,

RENFROE

etc.

DEL O L M O L E T E

1 9 8 4 ; SANMARTIN

1973.

See, for example, W A T S O N 1996c. 15 E.g. šlmm, which denotes a type of sacrifice, has 25 entries in P A R D E E 1987, 410. 16 A semantic glossary is provided in C A R T U , 177-92. 17 As mentioned in SIVAN 1997, xix. 18 Ugaútisches Handwörterbuch. 19 Supplemented by the extremely helpful 'glossary' in D I E T R I C H L O R E T Z 1996a, 543—926. Card indices of lexical material are also held in research institutes (see, for example, R I B I C H I N I - X E L L A 1985, 1 1 ) .

4.2

The texts

T h e texts under discussion are, of course, those in Ugaridc found at Ras Shamra, Ras Ibn Hani and elsewhere (conveniently collected in KTU'2), supplemented by more recent discoveries. T h e Akkadian texts are relevant chiefly for the lexical material they provide, either directly (as in the lexical texts) or indirectly (see K ü h n e 1974, 1975). O f particular interest is the treaty 20 ( K T U 3.1 = R S 11.722+) of which large sections in Ugaritic correspond to its Akkadian exemplars (RS 1 1.732, 17.227, 17.382; K n o p p e r s 1993 with previous bibliography). It can also be noted that some Ugaritic letters may in fact be translations from Akkadian, Egyptian and Hittite.

4.3

Problems

Aside from the large n u m b e r of words which are known from comm o n Semitic (um, 'mother', klb, 'dog', etc.)21 it is difficult to determine the m e a n i n g of m a n y lexical items in Ugaritic for several reasons. For one thing, the corpus is small and the range of significant contexts is accordingly quite limited. Also, vowels (aside from the use of the three aleph signs, 'a, 'i, 'u) are not indicated, and it is therefore not always easy to distinguish homographs. While prose and verse texts share much of the vocabulary (e.g. thm, 'message'), certain words are found only in non-literary contexts (e.g. gzl 'spinner' [ K T U 4.358 = RS 18.048:9]) whereas others occur only in verse (e.g klat 'both' [ K T U 1.1 = RS 3.361 iv 10 etc.]; phi 'stallion' [ K T U 1.4 = R S 2.[008]+ iv 5.9.15, etc.]; rt 'dirt' [ K T U 1.16 = R S 3.325+ ν 29, etc.]). 22 M a n y words are difficult or obscure, or have uncertain etymologies, e.g. ilqsm, bnn, b^r, gml, gpr, ddym, dnt II, dqr, dnt, kb, kbm, kdr, kmlt, knh, kpsln, krln, cbk, cprt, crgz, ctrb, gb, gbt, gprt, šdmt,23 tkt, etc., and the meanings of others (aktn, agzr, aqhr, askrr, idm, idrp, idt, udbr, unk, Cšd, etc.) cannot as yet be determined. Some words occur in broken or difficult contexts, e.g. hkm, hnn, hsm, htn, kdt, gbz, gbt, gdm (see DLU for details). With the discovery of new texts, previously unknown words continue to be added to the lexicon but the meanings of these

20

O r letter accompanying a treaty ( K N O P P F . R S 1 9 9 3 ) . Even here there can be false assumptions, as SANMARTIN ( 1 9 9 6 ) has shown in respect of ahl which means 'town' (Akkadian alum) rather than 'tent' (as in Hebrew). 22 For the vocabulary of the Ugaritic letters see C U N C H I L L O S § 8.1.5. 21

23

But see WYATT

1992C.

can also be uncertain, e.g. udn, 'to give ear', ghr, 'to sound forth(?)', zb, 'to foam(?)', qnn, perhaps 'to stand up', in RS 92.2014 ( P a r d e e 1997a, 327-8)

4.4

Principles

In view of the vast literature on Ugaritic lexicography and the often conflicting or at least divergent conclusions reached by scholars, there have been several attempts to set out solid methodological principles for the determination of meanings. 24 These are discussed here briefly. O n c e the conect reading of the text has been established, the context is of crucial importance. In fact, all agree that context is the most important single element for ascertaining what a word may mean. Syllabic spellings must also be taken into account, 25 and finally, comparative philology can be used. For this approach to be valid a set of rules must be applied: context is more significant than etymology; without context, etymology can only uphold a hypothetical proposal; phonological rules should only be flouted with supporting evidence; words in another language may not necessarily have the same meaning in Ugaritic; homographs and homonyms should be assumed only as a last resort; syntagmata and idioms as well as words need to be compared; a distinction must be made between the (archaic) poetic texts and the language of the letters, rituals and administrative texts; in the poetic texts it is important to determine stichometry and parallelism; in general, the rules of g r a m m a r and syntax should be applied. Finally, non-linguistic evidence should not be neglected. Some illustration of these principles is provided below.

4.5

Use of cognate languages

Comparison with other Semitic languages can provide a significant contribution to determining the meanings of words, but a degree of caution is required. Healey has surveyed the contributions available from Hebrew, Phoenician, Arabic, Akkadian, South Arabian 2 6 and Ethiopie ( L e s l a u 1 9 6 8 ) , particularly Aramaic and Syriac 1 9 8 8 ) . 2 7

24

HELD

25

See S I V A N 1984a; H U E H N E R G A R D 1987b. See especially R E N D S B U R G 1 9 8 7 .

26

1959,

1 6 9 ; DE M O O R

1973,

98;

PARKER

1979-80.

His conclusion, though, is that context is 'the ultimate arbiter'. 2 8 Arabic has been much used (and misused) as a resource for determining the meaning of Ugaritic words. This approach has been examined in detail by Renfroe 2 9 who has shown that there are many genuine Arabic-Ugaritic isoglosses but an equal if not greater number of spurious ones. In many cases we may simply have to say that there is insufficient evidence for any firm conclusions. It always has to be remembered that the meaning of a word in a cognate language cannot simply be transferred to Ugaritic and at times is no more than a guide. T h e same applies to the contribution from Eblaite (Sanmartin

1991).

4.6 Methodolog)) T h e first task necessary before resolving the meaning of a Ugaritic word is to survey all previous attempts, which is often very timeconsuming, with no guarantee of complete coverage. T h e scholar must then establish the correct reading on the tablet, determine the context, perhaps use etymology based on established language laws, refer to a wide range of Semitic languages, if necessary, use other languages (including Egyptian, Hittite, Hurrian, and even Sanskrit and Sumerian) and avoid the multiplication of homonyms and homographs. These rules, however, are an over-simplification. In practice, several other factors need to be taken into account, as the following examples show.

4.7

Selected examples

Some examples can help to illustrate the above. Evidence from cognate (Semitic) languages can come from Phoenician ( d e l O l m o L e t e 1 9 8 6 ) , Hebrew, Akkadian, Aramaic and Arabic as well as from such languages as Ethiopie 30 and even Syriac. For example, the verb nsr, parallel to bky 'to weep' ( K T U 1.16 = R S 3.325+ vi 4-5), can be explained from Syriac n'sar/nasar, 'to sigh, groan, m u r m u r , howl, shriek, lament'. 3 1 Choice of the correct cognate is important; for 27

HEALEY

1988.

28

HEALEY

1988,

29

RENFROE

30 31

68.

1985, 1986a, 1986b, 1989, 1992. See D I E T R I C H - L O R E T Z 199Id on 'šr. H E A L E Y 1 9 7 6 ; SANMARTIN 1 9 7 8 , 4 5 1 . However, cf.

WYATT 1998C, 237,

n.

280.

instance in the expression b bz'zm ( K T U 1 . 8 0 = RS 1 5 . 0 7 2 : 4 ) which could mean 'from the booty of goats', in view of Heb. baz, 'plunder, spoil' (HALOT, 117). However, in the context—a list of sacrifices— it is more likely to mean 'a young animal from the udder of the goats', i.e. an unweaned kid, where Ugaritic bz is an isogloss of Arabic buzz, Jewish-Aramaic bīzzā3, etc. all denoting 'teat' ( S a n m a r t i n 1 9 7 9 , 7 2 3 - 4 ) . Extra-linguistic evidence can also help determine meanings, for example, glp may denote murex used as a body-dye, since this type of shellfish was common near Ras Shamra ( d e M o o r 1 9 6 8 ) . Correct syntactic analysis is important for determining the meanings of words as shown by Husser ( H u s s e r 1995) in respect of atr in I'pr dmr atrh K T U 1.17 = RS 2.[004] i 2 7 - 8 (and par.). This expression has been understood in various ways but because of the parallelism with the previous line (lars mssu qtrh, 'who makes his spirit come out towards the earth'), which refers to correct burial, the preposition I also means 'to(wards)'. It is probably to be translated 'who protects his step towards the dust', and therefore atr cannot mean 'place', 'shrine' or even 'sanctuary'. T h e meaning of a word can depend on several factors including the structure of a text and recognition of the correct meaning of another word in the same passage. For instance, in K T U 4.392 = RS 18.130, prs means neither 'steering pole (of a chariot)' nor 'horse' but more probably a type of 'ration': Ihms mrkbt hmš csrh prs bt mrkbt, 'For the five chariots of the five divisions: ten />rí-rations from the chariot-house', since hmš C šrh does not mean 'fifteen' but 'five divisions (of the army)' followed by the numeral 'ten' ( V i t a 1996).32 It is also important to compare not just isolated words but syntagmata in Ugaritic with those in another (Semitic) language. Del O l m o Lete has provided a list of syntagmata common to Ugaritic and Phoenician. 33 O f interest, too, is the term hrs which occurs in the economic texts in connection with chariots (e.g. K T U 4.145 = RS 15.034:8-9) as a syntagm in the form w.hrs and means 'precisely, exactly', a usage borrowed from Akkadian. 34 T h e personal names provide a wealth of vocabulary, with many items not otherwise attested, e.g. rgln ( K T U 4.619 = RS 19.047:7),

32

T h e text remains difficult because the term šant (line 2) is not yet understood. 1986b, 46~7 = 1996a, 32-3. For comparison with a syntagm from Aramaic cf. W A T S O N 1992d. 34 DEL O L M O L E T E 1979; cf. V I T A 1995a, 57. 33

DEL O L M O L E T E

which is formed from the word rgl, 'leg'. T h e same applies to placenames such as bir, 'well' ( K T U 1.91 = RS 19.015:29, etc.). For both types of names syllabic spellings can be of use in determining meanings.

4.8

Lexical tablets

Of considerable importance are the polyglot vocabularies which have been found in Ugarit. These list the equivalents of words in four languages (Sumerian, Akkadian, Hurrian and Ugaritic) and in five of the eight tablets discovered so far, the Ugaritic column has been preserved (details in H u e h n e r g a r d 1987b, 21-3). According to Huehnergard, about 114 Ugaritic words have been vocalized in syllabic cuneiform spellings. For example: Sumerian

Akkadian

Hurrian

Ugaritic

Meaning

ŠUL EZEN

et-lu za-am-ma-rù35

uš-ta-an-ni hal-mi

ba-ah-hu-rù si-i-ru

'youth' 'song' 36

T h e sources for the vocalization of Ugaritic are (1) the three aleph signs, (2) syllabic spellings of Ugaritic words and to a lesser extent (3) comparative Semitics. T h e aleph signs give some indication of the associated vowels (or the absence of a vowel; see § 4.3.2.2). T h e syllabically written Ugaritic words and names are particularly helpful. T h e list of such items in v a n S o l d t (1991a, 301-8) has 156 entries. H u e h n e r g a r d provides a glossary with approximately 280 entries (Huehnergard 1987b, 103-94). Similarly, S i v a n 1984, 185-295, although his sources are not confined to texts found at Ras Shamra (see § 4.5). 3 ' Reference to other Semitic languages can only provide an indicadon of possible spellings and has to be used with caution.

4.9

Non-Semitic words in Ugantic

T h e city of Ras Shamra was a melting pot of several nationalities speaking different languages and both court and administration dealt with documents in several languages. In ritual, particularly, sections of text were written in Hurrian and Hurrian words occur liberally

As H U E H N E R G A R D 1987b, 97 notes, za-am-ma-rù stands for zamāru (the double -mm- is incorrect) and all the forms are nouns rather than infinitives. 36 VAN S O L D T 1991a, 7 4 7 - 5 3 : 'Appendix C : The lexical texts at Ugarit'. 37 See VAN S O L D T 1989d for review. 35

in the Ugaritic texts. It is not suprising, then, that m a n y words in the Ugaritic lexicon are in fact b o r r o w e d f r o m H u r r i a n , occasionally f r o m Hittite a n d m o r e rarely f r o m Egyptian or f r o m other nonSemitic languages. O v e r the years m o r e a n d m o r e such words have been identified. 3 8 T h u s , although most Ugaritic lexical items have a Semitic etymology, several are (or m a y be) of non-Semitic origin. S o m e of these words are listed here u n d e r the following headings: (1) H u r r i a n words, (2) Hittite words, (3) Egyptian words, (4) Sumerian words, (5) I n d o - E u r o p e a n words, (6) words from other languages. 4.9.1 H u r r i a n words include alhn, 'steward' ( K T U 4.392 = R S 18.130:4; cf. K T U 4 . 1 0 2 = R S 11.857:25; K T U 4 . 3 3 7 = R S 18.024:11) b o r r o w e d f r o m H u r r o - U r a r t i a n allae-hhi-nn, 'housekeeper', all, '(festive) g a r m e n t ' ( K T U 1.12 = R S 2.[012] ii 47, etc.), H u r r i a n alālu (Neu 1996, 314, η. 22); itnn, 'gift' ( K T U 1.100 = R S 24.244:74), Hurrian uatnannuf grbz, 'helmet' ( K T U 4.363 = R S 18.055:2), Hurrian gurpisi; hdm, 'footstool' ( K T U 1.3 = R S 2.[014] + ii 22, etc.), H u r r i a n atmû ( W a t s o n 1996b); hbrt, 'vessel, container' ( K T U 1.4 = R S 2. [008] + ii 9), H u r r i a n hubrushi; hptr, 'pot, c a u l d r o n ' (K TU 1.4 ii 8), H u r r i a n huppataru; hrd, 'warrior', Hurrian huradi/e ( S t i e g l i t z 1981); kht, 'throne' ( K T U 1.2 = R S 3.367 i 23, K T U 1.4 vi 51 etc.), 40 probably H u r r o U r a r t i a n ; kkrdn, ' c h e f ' ( K T U 4.126 = R S 14.084:27); kmn, '(a surface measure)' ( K T U 1.3 = R S 2. [014]+ iv 38 etc.), H u r r i a n kumānw, llh ( K T U 4.363 = R S 18.055:5) denotes part of trappings or harness, H u r r i a n lulahhi, (DLU, 245); gr, 'total', H u r r i a n heyarì; pg[n)dr, 'a type of fabric' ( K T U 4.270 = R S 17.111:10), H u r r i a n pahandam-f tbl, 'smith' ( K T U 4.790 = R S 86.2235:15), Hurrian tabid- ( D i e t r i c h L o r e t z 1990); tgpt, 'fe1t(?)' ( K T U 4.183 = R S 15.116 ii 10, etc.), H u r r i a n tahapše ( W a t s o n 1995c, 540); tkt, 'chariot' ( K T U 1.4 = R S 2. [008]+ ν 7, etc.), H u r r i a n šukītu ( L o r e t z 1996). Note that some words are Semitic with H u r r i a n endings, e.g. hdgl, 'arrowsmith' ( K T U 4.138 = R S 15.016:2, etc.) which is a H u r r i a n form of U g . hz, 'arrow' with the Hurrian -(hii)li ending ( S a n m a r t i n 1995, 179). Others

38

See the list provided by DE M O O R 1973, 98. Not all are correct, of course. For additional material the indices of Ugarit-Forschungen and other periodicals may be consulted. For a survey see W A T S O N 1995c, 1996c. See also P A R D E E 1996. 30 Borrowed through Middle Assyrian utnannu: cf. VON S O D E N 1988. 40 However, cf. DEL O L M O L E T E SANMARTIN 1 9 9 5 . 41

DIETRICH -

LORETZ

1977;

RIBICHINI

XELLA

1985,

61.

are Semitic words in H u r r i a n guise, e.g. kid, 'bow' ( K T U 4.277 = R S 17.141:1) is a form of qaštu, 'bow' ( D i e t r i c h - L o r e t z 1978b). 4.9.2 Hittite words: ans, 'small of the back' ( K T U 1.3 = R S 2. [014] + iii 3 5 ) , Hittite anašša ( d e M o o r 1 9 8 0 ) ; uiyn, '(an official)' ( K T U 6 . 2 9 = R S 1 7 . 3 6 4 : 3 ) , Hittite ur(/i]yanni; dgt, 'incense' ( K T U 1 . 1 9 = RS 3 . 3 2 2 + iv 2 3 , etc.), Hittite tuhhui-/tuhf}uwai-; htt, silver' ( K T U 1 . 1 4 = R S 2 . [ 0 0 3 ] + iv 1; K T U 1 . 1 4 ii 1 7 ) , Hittite (or Haitian) (fottu(š)-; hndlt, '(coloured wool)' ( K T U 4 . 1 8 2 = R S 1 5 . 1 1 5 : 1 7 ) , Hittite siGhandala; hsn, 'domestic' ( K T U 4 . 1 3 7 = R S 1 5 . 0 1 5 + : 1 . 1 0 ) , Hittite baššann-; mtyn, '(garment)' ( K T U 4 . 1 4 6 = R S 1 5 . 0 3 5 : 5 ) , Hittite maššiya—\a garment)' ( R i b i c h i n i - X e l l a 1 9 8 5 , 5 2 ) ; spsg, 'glass' ( K T U 1 . 1 7 = R S 2. [004] vi 3 6 - 7 etc.), Hittite zapzagi-, which denotes precious stones or a mineral ( N e u 1 9 9 5 ) ; tpnr, 'chief scribe' ( K T U 3 . 1 = R S 1 1 . 7 7 2 + : 3 2 ; K T U 4 . 4 4 = R S 9 . 4 5 3 : 2 8 ) , Hittite tuppanuri, etc. 4.9.3 Egyptian words: 4 2 br, ' b o a t , (war)ship' ( K T U 4.81 = R S 11.779:2-3, etc.); ht, ' b r e a d ' ( K T U 1.41 = R S 1.003+:22), Egyptian ht3 ( W a t s o n 1995a, 223-4); htt, 'silver' (see above), Egyptian hd; kw, 'drinking vessel' ( K T U 4.691 = R S 20.010:6), Egyptian kb; krk, ' p i c k a x e ' ( K T U 4 . 3 9 0 = R S 18.119:8 etc.), E g y p t i a n grg 'pick' ( S a n m a r t i n 1987b, 151); ktp, '(weapon)' ( K T U 1.6 = R S 2. [009] + ν 2); rrí, Ί p r a y ' (enclitic of entreaty; K T U 1.4 = R S 2. [008]+ i 20, etc.), tkt, 'ship' ( K T U 4.81 = R S 11.779; K T U 4.366 = R S 18.074); etc. 4.9.4 Sumerian words: ad, 'father' ( K T U 1.23 = R S 2.002:32 etc.); ilg, 'stone' ( K T U 4.751 = R S 29.096:11); ksu, 'seat, throne', ( K T U 1.3 vi 15 etc.), krs/su, '(a type of forage or fodder)' ( K T U 4.225 = R S 16.198 [a]+:16); plk, 'spindle' ( K T U 1.4 ii 3.4), etc. 4.9.5 I n d o - E u r o p e a n / I n d o - A r y a n words: agn, 'cauldron' ( K T U 1.23 = R S 2 . 0 0 2 : 1 5 . 3 1 . 3 6 ) m a y be cognate with Sanskrit agni 'fire'; mryn probably Indo-Aryan, e.g. Sanskrit marya, 'hero'; sm, 'king', ( K T U 1 . 2 2 = R S 2 . [ 0 2 4 ] i 1 8 ) ; ssw/ssw, 'horse' ( K T U 1 . 7 1 = R S 5 . 3 0 0 : 7 etc.), Sanskrit asva ( D i e t r i c h - L o r e t z 1 9 8 3 ) ; tnn, 'archer' ( K T U 4 . 3 5 = R S 8 . 1 8 3 + ii 1 1 etc.) a n d p e r h a p s others such as smrgt

42

WARD

1961

now needs updating.

'emerald', 43 Sanskrit marakata and Greek smaragdos (also found as maragdos), both meaning 'emerald' (WYATT 1 9 9 8 C , 9 1 , n. 9 0 ) . 4.9.6 Words from other languages: adr, 'door (?)' ( K T U 4.195 = R S 15.184:5), explained by anduru of uncertain origin; 44 irp, 'vase, container' ( K T U 4.123 = RS 13.014:20), is perhaps Hurro-Hittite, unless to be explained by Egyptian irp, 'wine' and therefore, possibly, 'wine-container'; utiyn ( K T U 3.1 = R S 11.772+:30 has the syllabic spelling uš-r[i-ia]-ni ( P R U 3 203 = R S 16.257+ iv 21) and may derive from Hurrian, Hittite or some other language (cf. DLU, 62). Generally speaking, in the case of some loanwords it is difficult to know whether they have been loaned directly, or indirectly through another language such as Akkadian, or even whether they are in fact Kulturwörter or Wanderwörter. Some words may even have been borrowed back from the language which initially borrowed them, e.g. kht, 'throne', from Hurrian keshi, itself a loan from Semitic ksu (DEL OLMO LETE - SANMARTIN 1995) and the same may apply to mgn, 'gift' and mryn, 'warrior' ( O ' C O N N O R 1989). T h e r e were also inner-Semitic borrowings, 45 and a distinction must be made between cognates and actual loans, such as nmrt from Akk. namurratu, 'splendour' (PARDEE 1988b, 115).46

4.10

Homonyms

H o m o n y m s can be distinguished by context, comparative philology and occasionally from syllabic spellings. Simple examples of homonyms are bt 'house' and bt ' d a u g h t e r ' , both nouns; from comparative Semitics and (where attested) syllabic spellings, it is possible to determine that the first word corresponds to / b ē t u / and the second to / b i t t u / (SIVAN 1984, 210.212). In the case of weak verbs it is also difficult to determine the correct form of the root (e.g. does gl derive from gll, gly or gyl?). Since the Ugaritic corpus is so small, it is quite possible that a 'word' which occurs only a few times may have as

43

K T U 1.4 = RS 2.[008]+ i 32, which is read mrfrt in KTU2. Listed as = daltu, 'door' in C A D A / 2 , 117; AHw, 51a ( C E C C H I N I 1984, 47). 45 Including loans from Ugaritic to Akkadian, e.g. Ug. mit, 'oar', which was borrowed by Ugaritic Akkadian ( V I T A 1995b). 46 In K T U 1.108 = RS 24.252:21.24; nouns with preformative n- are Akkadian, not Ugaritic, as P A R D E E notes (ibid.). 44

many different meanings. 47 Examples include b'r I, 'to burn', b'r II, 'to a b a n d o n ' (only in the D stem); 48 gl I, 'shout of joy', gl II, 'cup' and gl III, '(type of field)'; ptt I, 'linen' and ptt II, '(make-up) case' ( K T U 4.247 = R S 16.399:22; SANMARTIN 1987a, 54, n. 7). A clear example of the importance of distinguishing homonyms (and incidentally of correct word division) is provided by yrk tcl bgr She climbed the mountain by the flank, mslmt bgr tliyt by the incline, the immense mountain. wfl bkm ban She climbed bkm, Araru, bm an wbspn Araru, Sapūnu, bn'm bgr tliyt the fair, the immense mountain (KTU 1.10 = RS 3.362+ iii 27-31) Although the sequence bkm looks like the particle bkm, 'thereupon', this is impossible here as such particles are never postpositive: they always come first in the clause (RENFROE 1 9 9 2 , 58). Instead, here km means 'hill, m o u n d ' (as proposed by AARTUN 1 9 6 8 , 2 9 1 ) and it is preceded by the preposition b (as part of the syntagm cly + b, 'to climb'). 49 Hence the third line should be translated 'She climbed the m o u n d , Araru'. 5 0

4.11

Ghost words

Non-existent words are due to scribal error, false readings, incorrect analysis or incorrect word division. Examples of words written incorrecdy are any ( K T U 1.16 = R S 3.325+ i 7-8) which is to be read liny, a place-name and tdrs ( K T U 1.45 = RS 1.008+:5), to be read tdrq, 'tread'. T h e word ski, 'vizier' in K T U 3.1 = R S 1 1.772+:38 2 (KNOPPERS 1993) may have to be read skn (so KTU ). A classic example is ulp, taken by some scholars to mean 'noble, chief or the like', though it is really to be understood as u, 'and' + Ip, 'like' 51 or as u + / + p, 'whether from the mouth o f (see § 13.6.2). In some cases, the word division is uncertain, e.g. the sequence grbtil ( K T U 1.19 =

47

See especially the studies by

DIETRICH -

LORETZ

and

DIETRICH

-

LORETZ

-

SANMARTIN. 4,1

DIETRICH

LORETZ

49

Cf.

1975,

PARDEE

SANMARTIN

1975.

362.

50

However, cf. DLU, 107. Where tp is the preposition I + p, 'mouth'; cf. 2 9 1 - 2 , with bibliography. 51

DE M O O R

-

SANDERS

1991,

R S 3.322+ iii 47) could be analyzed as gr bt il 'resident in the house of Ilu', as grbt il, 'leprosy of Ilu' 52 or even as grb til 'may you seek asylum as a leper'; 53 kgmn may = k + gmn, 'like a funeral offering(?)' or kgmn = Hurrian 'three-year old'. 54 O n the other hand, in K T U 1.96 = RS 22.225:1, most scholars corrected cnn to cnt[ 'Anat', thus eliminating a previously unnoticed word which may mean 'evil eye'. 55

4.12

Future research

Although the core vocabulary of the Ugaritic texts is now understood to a large extent, there still remain many lexical items which either need to be determined or require further clarification. For example, in the Keret epic, msb'thn bslh ttpl, ' T h e seventh of them fell by (the) sW ( K T U 1.14 = R S 2. [003]+ i 20-1), it is uncertain whether the deaths described refer to his wives or to his children or indeed to the way the last victim died. T h e word slh could mean 'a throwing weapon', 'a sword', 'war', 'lightning', the god 'Salhu', 'parapet' or a disease which affects babies (Babylonian šulhu). If the last meaning applies, then this death must have affected Kirta's children, which in turn is significant for the meaning of the epic (WATSON 1997c). However, the task of determining the meaning and or etymology of individual words is not simply a luxury for students of Ugaritic. Scholars in other branches of Semitic studies or in other disciplines frequently need to consult reference works on Ugaritic for their own purposes. Examples include the compilers of DNWSI and HALOT or of encyclopaedias of various kinds or of comparative studies (e.g. H O C H 1994). It is important, therefore, to establish as accurately as possible what Ugaritic words mean. O u r main difficulties in understanding correctly many a difficult passage are principally that there is no similar passage in Ugaritic or that the context is uncertain. In other words, the Ugaritic corpus is simply too small owing to lack of texts. Future discoveries and continuing research are our only hopes in this exercise.

52

HILLERS

53

RENFROE

1985.

1986, correcting amd, the first word of the line, to tmd. See

WATSON

1989a, 47~8. 54

DIETRICH -

55

DEL

§ 6.5.3.

OLMO

LORETZ LETE

SANMARTIN

1992b;

LEWIS

1976.

1996,

WYATT

1998c, 375 n. 1. See

SPRONK

5

U G A R I T I C W O R D S IN SYLLABIC

JOHN

TEXTS

HUEHNERGARD

E m b e d d e d within the syllabic cuneiform texts written by scribes at Ugarit are over three hundred Ugaritic lexical items. These Ugaritic forms appear in all genres of Akkadian texts. In one group of syllabic cuneiform texts, those of the polyglot Syllabary A Vocabulary (Sa Voc.), the Ugaritic words were intentionally recorded by the scribes. T h e S a Voc. was a Mesopotamian lexical series in which columns of individual cuneiform signs, in a fixed order, were equated with one or more Akkadian words in a second column (LANDSBERGER - H A L L O C K 1 9 5 5 ) . This lexical series was imported to scribal centres in the west, including Hattuša, Emar, and Ugarit. T h e Ugarit exemplars of the S a Voc. are unusual in that they do not have merely the two columns of the exemplars found elsewhere; instead, they add either one additional column giving lexical equivalents in Hurrian 1 or, more often, two additional columns with equivalents in both Hurrian and Ugaritic. Six exemplars of this quadrilingual type are known (VAN S O L D T 1 9 9 0 , 7 2 8 - 3 0 ) , on which more than one hundred Ugaritic words are wholly or partly preserved. Thanks to the presence of Akkadian equivalents (and, when those are broken away, the fixed order of the cuneiform signs), the meanings of the Ugaritic words in the S a Voc. exemplars can be established with more precision and certainty than is the case with the Ugaritic lexical items attested in other text genres. Nearly all parts of speech are attested a m o n g these forms as the examples on p. 135 illustrate. Apart from the S a Voc. exemplars, Ugaritic words appear in Akkadian texts either (a) as parts of the names of local geographical features or plots of land or, much more often, (b) by chance, essentially lapsus calami in which the scribe either forgot the appropriate Akkadian word and substituted a Ugaritic form or thought that the (Ugaritic) form he was writing was proper Akkadian. T h e former group, which by their nature are substantives and adjectives, occur in legal and

1

R S 21.062 (Ug 5 no. 135); also R S 94.2939, discussed by M. S A L V I N I and B. A N D R É - S A L V I N I at the 45th Rencontre Assyriologique Internationale, Cambridge, Mass., July 6 1998.

ω υ Ζ —

Ά

in ιΟ ιη IT) in LH LO ιΟ m 3 S» rc en á?.Ô S CO . -50 S? á? à? ûo áf cb (Μ Il C II :CM M II CM ^ II II CMà? CO &) II = II ; ~ Il n " sa "Φ + :3 + :S + :s + + := + a ïh σIl> 13 Oï C D C O ^ C O C O co C O ι-» I-03 o O O TtCMco CMco Tt- co Ζ CO § 2 - CO^ Ë - Ε - Ε - Β - Ë ^ CO co CO ο . Ο ~' Ο O o oM o oCM o öCM Ο o o o Ο ο CM C0 CMC οCM 0 CM QCM Q CM Q ο CM co CM C O CM C o CMC C o G C C C/} C/3 C/3
Ο ζ «J ja ζ o P < 0N 1 ΐ

Ì

3 Ζ

Λ ; 3 ja

π! j3 jC y 'Ja 5

ι3 T3

ο Ë

J3

»3 C 3 j3

3 C« IC Ί3

Λ -a

3 ε ΙΟ

aj > Λ

ja CL : χ: ο α- S

a 'S c

3 C _aM Q

2

3 'S.

Λ =3 3 α

g

3 'ci. D. jaaΛ

CL

CL

Ο-

•3

a •o.

-S

£

CL

ά

'3

ct •S I "3

•S», c

α α

£ aI a α £

a £

>s

Ο -D —! J

D Ζ

Q D

α •«a.

Ζ D

Q D

13

_3

ζoo a. χ

S -Si

Q.

H U ω

O

c — c ω >

aν

'Ço α ε aέ

a

Ζ ο c/3

^ λ js. o

=e jag

•—

% ο D

c -S

α 3 υ

α κ

α -S

J <

CQ

3 Ζ

Q

D

S3

Ζ ο οχ

1α.

economic texts; many of them correspond to designations attested in alphabetic texts, such that alphabetic gt X corresponds to syllabic A.ŠÀ (me5/1,i ' a) * or ( é ) AN.ZA.GÀR X (see H U E H N E R G A R D 1987a, 11 n. 51), as in gt gwl = A.ŠÀmeŠ (/guw(w)â1i/ 'circuit'); gt dpm = A.ŠA-,ia dì-ip-ra-ni-ma (/diprānrma/ 'junipers'); gt m'br = "AN.ZA.GÀR ma-ba-n (/ma'bari/ 'ford'); gt 'mq = [AN.ZA].GÀR: am-qa (/'amqa/ 'stronghold?'); gt gl = A.ŠÀ* 3 : hu-li (/gôli/ 'low ground?'). T h e largest n u m b e r of Ugaritic words in syllabic texts, however, are those that appear, seemingly at random, for an expected Akkadian form. About a fourth of the Akkadian texts contain one or more such Ugaritic words. T h e y are found in all genres, although they are, understandably, relatively u n c o m m o n in texts that are copies of Mesopotamian originals, i.e. lexical texts (other than the S a Voc.) and literary texts; note, however, the following examples: lexical Ú EME.UR.GI 7 = la-ša-nu UR.GI 7 meŠ 'hound's-tongue' (a plant name), with Ugaritic /1ašānu/ for Akkadian lišān(u) (RS 22.034 + 349 = MSL 10 107ff. A, 110); literary lip-hu-dú-ma 'may they fear' RS 17.155 = Ug 5, no. 17a r. 7', in which the root is Northwest-Semitic p-h/h-d 'to fear' (VON S O D E N 1969) but the form, as is usually the case in the few examples found in literary texts, has been made to conform to an Akkadian paradigm (here, precative). Ugaritic vocabulary is much c o m m o n e r in the many legal and economic texts written in syllabic cuneiform. Some economic texts contain several Ugaritic words, or even a preponderance of them (e.g. RS 19.071= PRU 6, no. 114). An extreme instance is R S 17.240 = PRU 6, no. 136, a list of men of different professions who are owed a shekel (of silver), in which it is likely that every syllabically-written word is Ugaritic; this text may be compared with the very similar alphabetic text K T U 4.99 = R S 11.845, in which most of the same terms occur, albeit in the plural and not in the same order: RS 17.240 = PRU 6, no. 136 r 1Ί GÍN UG[U IÚ 1 GÍN MIN 1ύ[ 1 GÍN MIN 1ύ[ 1 MIN '"Š[À.TAM] 5

1 MIN

pa-[si-lu]

iù

1 MIN '"SAN [GA]

KTU 4.99 = RS 11.845

pslm (line 17)

khnm (line 9)

1 MIN

^ma-h[i-sú]

1 MIN ^karbi-s[Ú] 1 MIN

10

ÌÙ

na-ffi-ru

mf}sm (line 15) kbšm (line 7) (cf. ngr krm in KTU 4.609 = RS 19.016.12)

1 MIN '> 5 (QA)-/a-%] 1 M I N xHa-si-[ru] 1 MIN

iÙ

ia-qí-š [u]

ysrm (line 11) or

iu

ia-si- [hu] — y shm

(line 19) yqšm (line 6)

1 MIN '"UGULA ma-š[i 1 M I N ^mur-ú [

15

1 MIN lùna-s[i-ku]

mru ibr^n* (line 12) and mru skn (line 13) nsk ksp (line 14)

As the examples cited thus far suggest, most of the Ugaritic forms that occur in Akkadian texts (except for the S a Voc. quadrilinguals) are nouns. A few finite verbs are also found, however, such as the following suffix-conjugation forms: G D

3ms ta-ba-'a /taba'a/ 'he departed' RS 19.032 = PRU 6, no. 77:1; 3mp sa-ma-tù /samatū/ 'they devolved' RS 16.147 = PRU 3, 90b: 13; 3ms šal/ša-li-ma /ša11ima/ 'it delivered' RS 20.012 = Ug 5, no. 96 passim·,

?N 3mp na-ap-ta-ru /naptarū/ 'they exchanged?' RS 15.123 + 16.152 = PRU 3, 89a:5. T h e syllabically-written Ugaritic words are usually not identified as such by any graphic device; they simply occur within an otherwise Akkadian context, as in ul-ma-tu GALmeä sa S'SMÁ '4 large ship's hammers' (/hu1mātu/)' RS 19.112 = PRU 6, no. 141:4; 'AN.ZA.GÀR TN η (kVìÌR<meŠ) (la) Šal/ša-li-ma 'the manor of TN has (not) delivered (/sallima/) η slaves' RS 20.012 — Ug 5 no. 96, passim·, ù ma-ša-ra sa TN 'and the tithe (/ma'šara/, acc.) of TN' RS 16.244 = PRU 3, 93b:7. 4

urudu

In many instances, however, the Ugaritic words are preceded by a special sign that is written with two small angled wedges. This sign, usually termed a 'gloss mark' ÇGlossenkàr), has several functions in the syllabic texts from Ugarit (see H u e h n e r g a r d 1987a, 204-8), but its most c o m m o n use is to mark the word that follows it as nonAkkadian (i.e. in all but a few examples, as Ugaritic). T h e gloss mark is indicated by a colon in transliteration: ŠEmcs-/w KAŠmeâ-Â sa : ma-a'-ša-ri-ša 'the grain and beer of its (sc. a TN) tithe (/ma'sari/)' RS 16.153 = PRU 3, 146-7:10-1 (compare the last example cited above);

i-na A.ŠÀ : ad-ma-ni 'in "redland (/'admāni/) field'" RS 15.145 = PRU 3, 122-3:8, 12; u É-tu4 PN a-na 'MÎJLUGAL'-ft : sa-ma-ta 'and PN's field devolved (/samata/) upon the queen' RS 15.086.15-16 = PRU 3, 51-2. Most of the syllabically-written Ugaritic words are also attested in alphabetic texts. O v e r one-fifth of the forms, however, are thus far unknown in alphabetic form. In the case of some presumably comm o n words, such as the first example cited below, the absence of an alphabetic attestation m a y be due to the poetic nature of m u c h of the Ugaritic corpus. ri-i\g]-lu /riglu/ 'foot' RS 20.123+ = Ug 5, no. 137 i 10'; b[i]-f}i-ru UTUu-g[a-ri-it] 'the elite troops (/bihirū/) of Ug[arit]' RS 17.432 = PRU 6, no. 71:5'; [k]a-ma-'a-\lu) /kama'ätu/ 'truffles' RS 19.035B + Ε = PRU 6, no. 159:3'; ti-ib-nu /tibnu/ 'straw' RS 20.149 = Ug 5, no. 130 iii 17'.

kimc5

Several Ugaritic consonantal p h o n e m e s do not occur in Akkadian. T h e s e were generally represented in syllabic writings by signs whose consonantal c o m p o n e n t approximated that of the Ugaritic sound: / 0 / appears only rarely, as in i-zi-ir-\tu^\ /'iöirtu/ 'help' RS 20.149 = Ug 5, no. 130 iii 7'; / 0 / is written with Š-signs: mu-ša-bu /mo6abu/ 'seat' RS 20.123+ = Ug 5, no. 137 iii 32"; ^šu-uq-du(-)ma /9uqdu/ 'almond' RS 19.035B + Ε = PRU 6, no. 159:4'; ma-áš-I}a-tu-ma /ma0hatūma/ '(cloths)' RS 19.028 = PRU 6, no. 126:1; / z / probably appears in: zu-ur-PI /zurwu/ '(aromatic) resin' EA 48:8; / h / and / g / are written with Η-signs: ha-ra-^šu} /harrašu/ 'artisan' RS 20. 189a + β ( L a r o c h e 1979b, 479) 7; fré-qu / h ê q u / 'lap' R S 20.123+ = Ug 5, no. 137) i 9'; ^ha-ma-ru-ú / g a m a r u - h u / 'his apprentice' RS 19.042 = PRU 6, no. 79, 11; /}u-ul-ma-tu4 /gu1matu/ 'darkness' RS 20.123+ = Ug 5, no. 137: üi 15'; / V , / h / , and / V are sometimes written with the Akkadian '-sign, as in m-1 a-tu /ni J ātu/ '(implements)' RS 19.135 = PRU 6, no. 142:2; &meisa-'a-tu /sā'ātu/ '(wooden) bowls' RS 19.064 = PRU 6, no. 163: r. 4'; ma-a'-sa-H /ma'sari/ 'tithe' RS 16.153 = PRU 3, 146-7:11; sometimes indicated by 'broken writings', as in ma-aš-a-li /maš 5 a1i/ •orac1e(?)' RS 15.092 = PRU 3, 54ff:25; tu-a-pÍ-[ku] /tuhappiku/ 'to be upset' RS 20.123+ = Ug 5, no. 137: ii 23'; si-il-a /sil'a/ 'c1iff(?)' RS 16.249 = PRU 3, 96ff.:5; and sometimes, especially word-initially, not represented, as in a-na-ku /'anāku/ Τ RS 20.149 = Ug 5, no. 130: iii 12'; á-[P]I-[/]u 4 /huwātu/ 'word' RS 20. 189A + Β ( L a r o c h e 1979b, 479) 12; ab-du / ' a b d u / 'slave' R S 20.123+ = Ug 5, no. 137: iii 4.

T h e greatest linguistic benefit of the syllabically-written words is the evidence they provide for the vocalization of Ugaritic. T h e y show, for example, that the patterns of some Ugaritic words differed from those of their Northwest Semitic and Arabic cognates, as in da-ab-hu / d a b h u / 'sacrifice' RS 20.123+ = Ug 5, no. 137: iii 6, versus Aramaic, Hebrew and Arabic *dibh\ : ma-ad-da-tù /maddatu/ 'measurement' RS 17.022 + 087 = Ug 5, no. 5:9, versus Hebrew middâ; a-du-rù /'aduru/ 'mighty' RS 20.123+ = Ug 5, no. 137: ii 34', versus Hebrew 'addîr. A n u m b e r of phonological processes are also exposed by the vocalized syllabic forms. A m o n g these are vowel assimilation around gutturals: tu-ú-ru /tuhūru/ < *tahūru 'pure' RS 20.123+ = Ug 5, no. 137: ii 1; [ u ' r u d u ' m e J W ^ ' - [ j ] ^ W m e ä /mihrsūma/ < *mahīsūma '(implements)' RS 19.135 = PRU 6, no. 142:4; ta-a-ma-tu4 /tahāmatu/ < *tihāmatu 'sea' RS 20.123+ = Ug 5, no. 137: iii 34"; raising of a and ā before w and y: [h]u-V\-tu^ /huwwatu/ < *hawwatu 'land' RS 20.123+ = Ug 5, no. 137: ii 10'; hé-V\-ma /hiyyūma/ < *hayyüma 'life' RS 20.426c + 20Ιο = Ug 5, no. 131:6'; optional syncope of short vowels in open syllables: na-ba-ki-ma and : na-ab-ki-ma /nab(a)kīma/ 'springs' RS 16.150 = PRU 3, 47a: 16 and RS 16.263 = PRU 3, 49b:5;' [b]a-ma-ru-m[a] and 1ύ·meif}a-am-rumamei/gam(a)rūma/ 'apprentices' RS 15.042 + 110 = PRU 3, 196: i 1 and RS 25.428:6 (see PRU 6, 150 n. 3); Ä-W-[<]« 4 /šantu/ < *šanatu 'year' RS 20. 189a + b ( L a r o c h e 1979b, 479) 11 (see v a n S o l d t 1990b). T h e Ugaritic vocabulary attested in Akkadian texts has been studied in detail in B o y d 1975, S i v a n 1984a (see the reviews of H u e h n e r g a r d 1987b, v a n S o l d t 1989d), H u e h n e r g a r d 1987a (see the important review of v a n S o l d t 1990b), and v a n S o l d t 1991a. Several studies of individual lexical items have also appeared, including, recendy, Lambert 1988; S a n m a r t i n 1987b, 1992; v a n S o l d t 1989a; V i t a 1995b, 1996a; W e g n e r 1995; W i l h e l m 1992; X e l l a 1990.

UGARITIC STYLISTICS

1

Ugaritic

Meindert

1.1

Prose

Dijkstra

Introduction

It is rather difficult to give a precise definition of Ugaritic prose texts. In general, they include all those texts that are assumed not to be poetic texts, or at any rate do not reveal clear marks or criteria of poetry as found in the major Ugaritic myths and legends. However, the distinction between poetry and prose is rather clear, where poetic sections of myths and legends are interrupted by prose sentences containing ritual prescriptions, instructions for performance and recitation, or colophons ( K T U 1.4 = R S 2. [008]+ le.edge, K T U 1.6 = R S 2.[009]+ vi 5 4 - 8 ; K T U 1.16 = R S 3.325+ le.edge, K T U 1.17 = R S 2. [004] le.edge). For instance K T U 1.4 ν 4 2 - 3 : wtb Imspr. . ktlakn glmm 'and repeat the recitation that the lads were sent' and K T U 1.19 = R S 3.322+ vi. le.edge whndt.ytb.Imspr 'and this (passage) should be recited once more' (referring to the legend from iv 23 onwards). W e shall see that hndt is a typical prose word. A similar line of instruction is included in the myth of K T U 1.23 = R S 2.002.56 ytbn yspr 1hmš 1slmm* wyšr pf}r klat, ' O n e shall repeat the recitation five times before the images and the congregation together shall sing. . .'. K T U 1.23 is a good example of how prose ritual prescriptions are interspersed in a poetic text, particularly in the opening sections ( K T U 1.23.12, 14-5, 18-22), but in the expiation ritual K T U 1.40 = RS 1.002+.35 the instruction w.tb.lmpsr 'and start to recite again . . .' appears in a prose discourse. T h e majority of documents published in KTUX and 2 are prose texts. T h e largest group, the economic or administrative texts, usually contain lists of persons and cities, or villages, often introduced by a label or heading identifying the nature and purpose of the list and some-

times also preceded by the general marker spr. O f t e n only these headings permit some grammatical and syntactic analysis. Together with the letters, they may help to assess the criteria and character of the Ugaritic vernacular used during the years of Ugarit's final flourishing, basically the last fifty years following the reign of Ammittamru III, though some older documents survived ( K T U 3.1, 3.4, 7.65 = R S 11.772+, 16.191+ and 16.402[B]) mentioning such kings as N i q m a d u II and his son Niqmepa. It has been assumed that these latest texts reflect the m o r e developed language of everyday use (SEGERT 1984, § 13.1).

T h e purpose of this chapter is to review the different types of prose style and syntax found in distinctive prose genres such as letters, contracts, or rituals. Distinct use of a given verb form may occur in different types of discourse. Prose discourse is a constellation of functionally used verbal or noun clauses pertaining to a given type of prose. Discourse types may, for instance, be narrative, precative, persuasive, prescripdve or performative and each function implies the use of certain modes of verbs and noun clauses. This review starts from the assumption that each type of prose is, in effect, such a cluster of functionally and semantically used verbal or noun clause types ( L O N G A C R E 1 9 9 2 , 1 7 7 - 8 ) . It implies that reports are basically narradve, ritual and medical instructions prescripdve, and prose incantantions and letters persuasive. T h e borderlines between the different types of discourse are not always well defined; performative elements may also occur in rituals and incantations. Letters may contain narrative parts in so far as they function as reports. Such a functional approach related to context and genre for the study of verb and syntax in Ugaritic prose is more appropriate than the generic aspectual a n d temporal distinction made, for instance, by S E G E R T 1984 (particularly § 64.2, but see R A I N E Y 1987, 397; T R O P P E R 1993a, 389ff.). W e cannot deal extensively here with the function of perfect and imperfect in poetry in comparison with its function in prose, but there is more overlap between prose and poetry than Segert suggests. H e states that the perfect a n d imperfect acquired temporal character in the late Ugaritic vernacular (about 1200 BCE; S E G E R T 1984, § 64.21). However the perfect is used in poetry as a narrative m o d e describing a completed action in the past, whereas it still may a p p e a r in its constative and performative function in late Ugaritic prose depending on the context, for instance in contracts and rituals. Segert's assumption may in general apply to Ugaritic

correspondence and administradve texts, but the modes of use m a y be different for other types of discourse.

1.2 1.2.1

Classification of the prose texts

Administrative texts

KTU2 lists 792 texts as economic or administrative. T h e y are by far the largest group of prose texts. Not all of t h e m are administrative texts (e.g. K T U 4 . 6 6 9 + = R S 1 9 . 1 7 4 A B is possibly H u r r i a n [ D I J K S T R A 1 9 9 4 , 1 2 5 - 6 ] and 4 . 6 5 9 = R S 1 9 . 1 6 6 a sales contract for a female slave) a n d m a n y fragments are chips a n d bits that m a y in time be joined to other documents (e.g. K T U 4.412 + 545 + 518 + 512 = R S 1 8 . 2 5 1 + 1 8 . [ 4 7 1 ] + 1 8 . [ 4 3 5 ] + 1 8 . [ 4 2 6 ] ) . However, some texts listed as religious texts, letters or juridical documents would better be assessed as administrative documents (e.g. K T U 1 . 9 1 = R S 1 9 . 0 1 5 , l 2 . 2 7 = R S 1 6 . 3 7 8 A , 2 . 6 9 = R S 2 4 . 6 6 0 C a n d spr mnh bd mnny K T U 3 . 1 0 = R I H 8 4 / 3 3 , c o m p a r e also K T U 4 . 9 1 = R S 1 1 . 7 9 5 ) . 1.2.2

Letters

T h e second largest corpus of Ugaritic prose texts that are susceptible of basic linguistic analysis are letters ( K T U 2.1-83). T h e r e is some doubt as to the epistolary nature of some of the texts ( K T U 2.2 = R S 3.334, 2.5 = R S 1.020, 2.7 = R S 1.026+, 2.19 = R S 15.125 [manumission of a royal slave], 2.27 = R S 16.378A, 2.31 = R S 16.394, 2.60 = R S 18.[528], 2.62 = R S 19.022 and 2.69 = R S 24.660G). Quite a few letters are purely formal epistles, or contain only short messages apart from the usual airs and graces ( K T U 2.4 = R S 1.018, 2.10 = R S 4.475, 2 . 1 1 - 7 = R S 8.315, 9.479a, 11.872, [Varia 4], 15.007, 15.008, 15.098, 2.24 = R S 16.137 [bü\+, 2.26 = R S 16.264, 2 . 3 0 - 3 1 = R S 16.379, 16.394, 2.40 = R S 18.040, 2 . 6 3 64 = R S 19.029, 19.102, 2.68 = R S 20.199 a n d 2.71 = R S 29.095), or they are too broken for coherent translation ( K T U 2.1 = R S 3.427, 2.3 = R S 1.013+, 2.6 = R S 1.021, 2 . 8 - 9 = R S 1.032, 2.[026], 2.18 = R S 15.107, 2.20 = R S 15.158, 2.22 = R S 15.191 [a], 2.25 = R S 16.196, 2.35 = R S 17.327, 2 . 4 8 - 5 9 = R S 18.285[a], 18.286[ab], 18.287, 18.[312, 364, 380, 386, 387, 400, 443, 482, 500], 2 . 6 5 - 6 7 = R S 19.158B, 19.181 AB, 2 . 7 7 - 8 0 = R I H 7 7 / 0 1 , 77/21A, 7 7 / 2 5 , 7 8 / 2 1 and 2.83 = R I H 78/25). Only a few offer larger portions of prose to give an impression of the 13th century West Semitic per-

suasive m o d e of discourse used in diplomatic and business letters (KTU 2.33

2.10

=

RS

4.475,

=

RS

16.402,

17.434

and

17.434ba,

=

RS

18.113a,

2.45

2.36

2.23 +

2.38 =

RS

37 =

=

RS

+

73

16.078+, +

74

RS

18.031,

18.140,

2.70

=

RS

2.39 =

2.31

RS

=

=

RS

16.394,

17.435+,

17.438,

RS

29.093,

18.038, 2.72

2.42 =

RS

a n d 2 . 8 1 = R I H 7 8 / 0 3 + 7 8 / 3 0 ) . Twenty or so more letters were found in 1 9 9 4 in the house of U r t e n u ( M a l b r a n - L a b a t 1 9 9 6 ; D i e t r i c h - L o r e t z 1 9 9 7 ) , but are not yet available for analysis. 34.124

1.2.3

Ritual texts

Quite a large group are about fifty Ugaritic rituals and five lists of gods. With this group should also be mentioned about 26 completely or partially Hurrian ritual texts. T h e Ugaritic rituals include a series of monthly rituals as a kind of service book through the cultic year. T h e y contain prescriptions for daily sacrifices, seasonal festivals a n d prayers. T h e H u r r i a n texts include sacrificial lists, sacrificial agr hid. hymns a n d perhaps incantations: 1.2.3.1

Monthly rituals through the year

yrh rìšyn K T U 1.411| 1.87 = R S 1.003+, 18.056 and partial duplicates 1.39 = R S 1.001a.2-10|| 1.41 = R S 1.003+.11-9, 1.126 = R S 24.276.18ff.il 1.41.44-9; yrh sm[et] K T U 1.87.54ff.|| 1.46+ = R S 1.009+ [ D i j k s t r a 1984, 6 9 f f ] and partial duplicates 1.109 = R S 24.2531| 1.46+. 10-32, 1.130 = R S 24.284111.46+. 11-21, 1.58? = R S 1.047, 1.134? = R S 24.294; yrh n[ql] K T U 1.138 = R S 24.298; yrh ib'lt K T U 1.119 = R S 24.266; yrh hyr K T U 1.105 = R S 24.249, 1.112 = 24.256, 1.132 = R S 24.291 (partially Hurrian, continuation of 1.112? = R S 24.256), 1.148 = R S 24.643 rev?; yrh gn? K T U 1.106 = R S 24.250+, partial duplicates 1.134 = R S 24.294 obv., 1.171 = R I H 7 8 / 1 6 . 1.2.3.2

Related texts with daily rituals a n d lists of sacrifices

K T U 1.48 = R S 1.019, 1.49 = R S 1.022, 1.50 = R S 1.023, 1.53 = R S 1.033, 1.57 = R S 1.046, 1.58 = R S 1.047, 1.76 = R S 6.215, 1.81 = R S 15.130, 1.91 = R S 19.015, 1.104 + 7 . 1 3 3 = R S 24.248+24.305 ( D i j k s t r a 1998, 280-2), 1.110 = R S 24.254 (Hurrian

with Ugaritic gloss bW pamt), 1.111 = R S 24.255 (obv. Hurrian); 1.134 = R S 24.294, 1.136 + 1.137 = R S 24.296ab, 1.146 = R S 24.253, 1.156 = 24.656, 1.159 + 1.160 = R S 28.059AB, 1.162 = R S [Varia 20], 1.165 = R I H 7 7 / 0 4 + 7 7 / 1 1 , 1.170 = R I H 7 8 / 1 1 , 1.171 = R I H 7 8 / 1 6 , 1.173 = R I H 7 8 / 0 4 , 7.46 = R S 1.042, 7.177 = R S 24.653B; id yph/ydbh mlk: K T U 1.41 = R S 1 . 0 0 3 + . 5 0 - 5 , 1.90 = R S 19.01311 1.168 = R I H 7 7 / 1 0 b + 7 7 / 2 2 , 1.115 = R S 24.260, 1.164 = R I H 7 7 / 0 2 B + , 1.139? = R S 24.300. 1.2.3.3

Procession

rituals

km t'rb GM,Ν) bt mlk: K T U 1.43 = R S 1.005, 1.148 = R S 2 4 . 6 4 3 . 1 8 22, 1.139? = R S 24.300. 1.2.3.4

Occasional sacrificial festivals

K T U 1.91 = R S 19.015 obv. a catalogue of royal festivals; spr dbh φι K T U 1.161 = R S 34.126; dbh il bldn K T U 1.162 = 1.91.6 (= R S [Varia 20], R S 19.015.6); dbh spn K T U 1.148 = R S 2 4 . 6 4 3 . 1 - 1 2 , 1.91.3; dbh cttrt qrat.bgrn (partially H u r r i a n ) K T U 1.116 = R S 24.261; a lung model with ritual a n d sacrificial instruction K T U 1.127 = R S 24.277. 1.2.3.5

G o d lists

K T U 1.47 = R S 1.017, 1.74(?) = R S 6.138, 1.102 = R S 24.246, 1.113 = R S 24.257 (deified kings), 1.118 = R S 24.264+. 1.2.3.6

Expiation rituals

K T U 1.4011 1.84111.121 + 1.122 + 1.153 + 1.154 + 7.162? = R S 1.002 II 17.100[A]+ || 24.270a[b] + 24.650b + 24.652G+ + 24.652b?. 1.2.3.7

H u r r i a n rituals

K T U 1.26 = R S 1 - 1 1 . [ 0 4 8 ] , 1.30 = R S 1 - 1 1 . [046], 1.32 = R S 1.[066], 1.33 = R S 1.[067], 1.34 = R S 1.[076], 1.35 + 1.36 + 1.37 = R S 1. [069 + 070 + 071], 1.42 = R S 1.004, 1.44 = R S 1.007, 1.51 + 52 = R S 1.027 + 1.028+, 1.54 = R S 1.034+, 1.59 = R S 1.[049a], 1.60 = R S 2.[006], 1.64 = R S 3.372, 1.66 = R S 5.182, 1.68 = R S 5.200, 1.110 = R S 24.254, 1.111 = R S 24.255 (rev.

Ugaritic), 1.116 = RS 24.261, 1.120 = R S 24.269+, 1.125 = R S 24.274, 1.128 = R S 24.278, 1.131 = RS 24.285, 1.132 = R S 24.291, 1.135 = R S 24.295, 1.148 = R S 24.643.13-7, 1.149 + 150 = R S 24.644 + 24.644[a], 4.669+ = R S 19.174A. Administrative texts quite often also contain information about rituals, in particular when they deal with the distribution and allocation of wine, food and other commodities for the cult, e.g. hmšyn.bdbh mlkt bmdr' 'five (kd) of wine for the sacrifice of the Q u e e n in the sown land' ( K T U 4.149 = R S 15.039.14 6, see further K T U 1.91 = R S 19.015, 4.168 = R S 15.082, 4.182 = R S 15.115, 4.213 = R S 16.127.24, 4.219 = R S 16.179.2-3). Most ritual texts stem from the High Priest's house and the house of the Hurrian Priest (PH rooms 10-11) and just a few from Ras Ibn Hani. 1.2.4

Religious texts in literary prose

T h e r e is quite a large group of smaller texts and fragments which may be classified as literary religious texts, if not myths. Some of them were clearly composed as poetry (e.g. K T U 1.10 = RS 3.362+, 1.12 = R S 2.[012], 1.92 = RS 19.039+, 1.96 = R S 22.225 and 1.100 = RS 24.244), but some may be fragments of myths and incantations in prose or a kind of poetic prose ( K T U 1.9 = R S 5.229, 1.24 = RS 5.194, 1.25 = R S 5.259, 1.45 = RS 1.008+, 1.65 = R S 4.474, 1.82 = RS 15.134, 1.83 = R S 16.266, 1.96 = RS 22.225, 1.107 = R S 24.251+, 1.151 (?) = R S 24.647 and 1.169 = R I H 78/20). T h e r e are also occasionally prayers, one in poetic form ( K T U 1.119 = R S 24.266.26-36), but also in prose ( K T U 1.65 = R S 4.474 and 1.123? = R S 24.271), a blessing or dedication ( K T U 1.77 = R S 6.411), a small collection of fables ( K T U 1.93 = R S 19.054) and, perhaps, a wisdom text ( K T U 2.21| 2.5? = R S 3.334, 1.020). With this group, we may also mention the few Akkadian .n/?íM-prayers in Ugaritic alphabetic script ( K T U 1.67 (+) 1.69 = R S 5.199 + 5.213, 1.70 = R S 5.156+, 1.73 = R S 5.303fc, 7.50 = R S 5.157, 7.52 = RS 5.196 and 7.55 = R S 5.218). 1.2.5

Other miscellaneous prose texts

Minor groups of prose texts are the juridical texts, medical prescriptions and omens. T h e juridical texts are mainly found in K T U Section 3, but see also K T U 2.19 = R S 15.125 (manumission of a slave) and K T U 4.659 = R S 19.166 (sales record of a female slave?)

a n d the custom licences (or h a r b o u r dues?) a n d caravan licences (KTU

4.172 =

RS

15.093, 4.266

=

RS

17.074, 4.336

=

RS

18.023

and 4 . 3 8 8 = R S 1 8 . 1 1 3 ) . As sealed documents, the licences can be considered a kind of juridical contract. T h e distinction between legal documents and administrative records is not always clear. K T U 3.7 = R S 1 8 . 1 1 8 and 3 . 1 0 = R I H 8 4 / 3 3 are rather administrative lists, though they deal with legal charges such as ^ - o b l i g a t i o n s and debts, as do K T U 4 . 3 3 8 = R S 1 8 . 0 2 5 and 4 . 3 4 7 = R S 1 8 . 0 3 5 + . Medical prescriptions are found in K T U 1 . 1 7 5 = R I H 7 7 / 1 8 . T h e y are sometimes included in other texts K T U 1 . 1 1 4 = R S 2 4 . 2 5 8 . 2 9 - 3 2 (an incantation for medical treatment of delirium) a n d 1.124 = R S 2 4 . 2 7 2 (an oracular report), a n d also perhaps the fragment K T U 1 . 8 8 = R S 1 8 . 1 0 7 . T h e s e prose texts and sections are related to the hippiatric medical text of which four copies have been discovered (KTU

1.71 =

RS

5.300,

1.72 =

RS

5.285+,

1.85 =

RS

17.120

and

T h e large corpus of ancient N e a r Eastern o m e n literature was also represented in Ugarit by a d r e a m book (spr hlmm K T U 1 . 8 6 = R S 1 8 . 0 4 1 ) , a collection of astronomical omens ( K T U 1 . 1 6 3 = R I H 7 8 / 1 4 ) and birth omens of the summa izbu type ( K T U 1.103+ = R S 2 4 . 2 4 7 + a n d 1.140 = R S 24.302: D i e t r i c h - L o r e t z 1990a); also omens inscribed on fields of lung and liver models ( K T U 1 . 1 2 7 = R S 2 4 . 2 7 7 , 1.141-4 = R S 24.312, 2 4 . 3 2 3 , 2 4 . 3 2 6 , 2 4 . 3 2 7 and 1 . 1 5 5 = R S 2 4 . 6 5 4 ) and a report of an astronomical o m e n ( K T U 1.78 = R S 12.061). T h e r e is, perhaps, also a protocol of n e c r o m a n c y with some ritual prescriptions ( K T U 1.124 = R S 2 4 . 2 7 2 : D i e t r i c h - L o r e t z 1990a). 1.97 =

RS

23.484; C o h e n

1.3

1996).

Administrative prose

By far the largest group of prose texts are the administrative texts, which include census lists of persons, guilds and cities, p a y m e n t rolls, receipts and records of received or distributed commodities. T h e y are a main source for private n a m e s a n d also a lexicographic goldmine, though m a n y words are still poorly understood. For the structure of the language they are less informative, since their syntactical structure and style is often very simple. M a n y texts only have a simple label as heading mentioning a guild (hrtm 'ploughmen', K T U 4.65 = R S 11.602, 4.122 = R S 13.012; tnnm a kind of soldier, K T U 4.66 = R S 11.656; mrynm 'knights', K T U 4.623 = R S 19.049[b]; nqdm, 'sheep breeders', K T U 4.681 = R S 19.180; mdrglm 'guards ? ', K T U 4.751 = R S 29.096; khnm, 'priests', K T U 4.761 = R S 34.123)

or a village/city/gentilic. T h e s e guild markers themselves are listed as such too ( K T U 4.29 = R S 3.320, 4.38 = R S 8.272, 4.47 = R S 10.043, 4.68 = R S 11.716.60ff., 4.99 = R S 11.845, etc.); likewise geographical markers in topographical lists (K TU 4.63 = R S 10.052, 4.232 = R S 16.355, etc.). Both serve as headings in texts which contain persons grouped by trade, profession or provenance ( K T U 4.35 = R S 8 . 1 8 3 + , 4.69 = R S 11.715+, 4.71 (+) 72 = R S 11.721, 11.722, 4.103 = R S 11.858, 4.183 = R S 15.116, 4 . 4 1 2 + = R S 18.251 a n d 4.633 = R S 19.086A). T h e different p a r t s of these simple syntactical structures are: (1) heading (with or without introductory spr); (2) lists of persons, towns, etc. (together with n u m b e r , commodity, etc.); (3) s u m m a r y or total (with or without tgmr). T h e s e sections are often extended by descriptive, or restrictive remarks in relative clauses. T h e style is usually concise in the extreme, leaving out self-evident terms a n d phrases (e.g. tql, kbd, dd, tgmr, etc.). F r o m such texts, only a few prose sections can be gleaned, in particular f r o m texts such as K T U 4.145 = R S 15.034, which is a small report revealing the p o o r condition of the king's chariotry: (1) tmn.mrkbt.dt. (2) 'rb.bt.mlk (3) yd.apnthn (4) yd.hzhn (5) yd trhn/ (6) w.l.tt.mrkbtm (7) inn. utpt/ (8) w.tlt.smdm.w.hrs (9) apnt.bd.rb.hršm (10) d.šsa.hwyh Eight chariots, which entered the royal palace with their wheels, their ax1es(?), their bearings(?), but two chariots have no quiver; and of three two-horse carriages(?), the wheels are in the hands of the chief smith, who took (them) out for repair. It is a good example of the descriptive style found in administrative texts (also K T U 4.136 = R S 15.013). T h e nature of the d o c u m e n t s is often indicated by the w o r d spr, while the subject m a t t e r m a y be persons, trades a n d professions, or commodities, tribute, rations a n d fields u n d e r these headings extended with different types of relative sentences: spr npš d.crb bt.mlk w.b spr.l.st, 'List of people w h o entered the royal palace, but who were not put into the list. . . ' ( K T U 4.338 = R S 18.025.1-3). Usually, clusters with construct nouns do not exceed three nouns or names, such as spr argmn špš ( K T U 4.610 = R S 19.017); spr ksp mnny ( K T U 4.791 = R I H 8 4 / 0 4 ) ; but cf. spr hpr.bns.mlk ( K T U 4.609 = R S 19.016), spr hr's qst iptl ( K T U 4.215 = R S 16.130). D o c u m e n t s often have no h e a d i n g and start in médias res. C o m modities listed and other entries are occasionally extended by descriptive relative n o u n clauses or participles: ktn.d.sr.phm.bh, '. . . a robe

that has a string(?) of carbuncles on it' ( K T U 4.132 = R S 15.004.4); w.lpš. d sgr.b/ι, '. . . a garment that has a fibula' ( K T U 4.166 = R S 15.078); tit mrkbt spyt.bhrs [.] f ir[.] smdm.trm.d[.l.s]py/w.trm. ahdm. spym/tit mrkbt d.l.spy, 'Three chariots covered with gold, ten pairs of tr which are [not co]vered and a doub1e(?) set of tr covered, three chariots that are not c o v e r e d . . . ' ( K T U 4.167 = R S 15.079.1-7), but also verbal clauses: yn d.ykl.bd.k[hnm] (2) b.dbh.mlk, 'Wine that is delivered into the hands of the pr[iests] for the sacrifice of the king' ( K T U 1.91 = RS 19.015.1-2; T R O P P E R 1991b, 355); qmh. d.kly.ksh.illdrm bd.zlb[n], 'Flour that was completely spent according to the order(?) of Illdrm into the hand of %b[n] . . . ' ( K T U 4.362 = RS 18.052.1-2), ksp.d.slmyrmn.'l.bt, 'Silver that Yrmn paid for (the mortgage on?) the house . . . ' ( K T U 4.755 = R S 31.080; also K T U 4.95 = R S 11.836+, 4.166 = R S 15.078, 4.213 = RS 16.127, 4.290 = R S 17.297 and 4.348 = R S 18.036) and . . . prs qmh d nšlm, '. . . a /w-measure of flour that has been paid for' ( K T U 4.328 = R S 18.008.1). In relative noun clauses the expression for existence it is often added ( K T U 4.235 = RS 16.369, 4.422 = R S 18.293, 4.617 = RS 19.044, 4.752 = RS 29.097 [ S E G E R T 1982, § 55.7]), and the usual negation in such relative noun clauses is in(n) ( K T U 4.53 = RS 10.090, 4.180 = R S 15.105, 4.214 = R S 16.128 and 4.379 = R S 18.098). Ugaritic scribes seem to have used two conventions to sum up the totals of their administrative documents, either by writing the totals (ŠU.NÍGIN = napharu/gabbu) in cuneiform Sumero-Babylonian shorth a n d ( K T U 4.48 = RS 10.045, 4.63 = R S 10.052, 4.68 = R S 11.716, 4.69 = RS 11.715+, 4.71, 4.72 = RS 11.721, 11.722, 4.90 = R S 11.797, 4.93 = R S 11.776+, 4.100 = R S 11.850, 4.102 = R S 11.857, 4.165 = R S 15.076, 4.219 = RS 16.179, 4.232 = R S 16.355, 4.299 = R S 17.345, 4.308 = RS 17.386, 4.340 = R S 18.027, 4.435 = R S 18.[306], 4.610 = R S 19.017, 4.704 = R S 21.002, 4.745 = R S 25.417, 4.754 = R S 31.043, and 4.784 = R S [Varia 38],2; VAN S O L D T , 1995, 485-6) or in Ugaritic tgmr, and sometimes even both ways (e.g. sb'.mat ttm kbd/7 me-at 60 H I . M E Š , K T U 4.340 = RS 18. [027]). As with the headings, the pattern of such totals is not consistent. Usually the tgmr of the commodity received or distributed, or the group or city is mentioned first (e.g. K T U 1.91 = R S 19.015, 4.67 = R S 11.714, 4.156 = R S 15.053, 4.269 = R S 17.106 and K T U 4.151 = R S 15.044, 4.179 = RS 15.103, 4.777 = R I H 8 3 / 0 7 + respectively), but it may also follow the total a m o u n t ( K T U 4.230 = R S 16.341, 4.764 = R S 34.176 and 4.137 = R S 15.015+, 4.141

= R S 15.022+ and 4.173 = R S 15.094 respectively), but the word tgmr is also often left out ( K T U 4.164 = R S 15.075, 4.344 = RS 18.030, 4.427 = RS 18.299, 4.163 = R S 15.073.15ff, and 4.174 = RS 15.095 respectively). T h e style of these texts is basically descriptive and strongly paratactic. Complex syntactic structures with subordinate clauses are almost absent.

1.4

Literary prose of incantations, stones and reports

Everybody knows, or rather thinks he knows the difference between prose and poetry ( W a t s o n 1984d [1995] 44). T h e problem is to establish sound criteria. We cannot deal here with this question in extenso (see P a r d e e 1993a). T h e criteria often used to distinguish prose from poetry in Hebrew literature, namely the absence or rarity of prose elements such as the relative marker }"ser (less often the relative pronoun), the definite article, the object marker and the narrative waw, are not very helpful for Ugaritic prose. T h e existence of a narrative waw discourse in Ugaritic is still disputed and indeed, if it is not to be found in the context of the incantation K T U 1.100 = RS 24.244.67-8: mgy.hrn.l bth w (68)ystqilh^rh, ' H o r o n reached his house, and he entered his court', it occurs, perhaps, in a few prose texts such as, for instance, the report about a necromantic inquiry ( K T U 1.124 = R S 24.272): ky mgy.adn (2) ilm rbm.'m dtn (3) wysal.mtpt.yld (4) w/ny.nn.dtn (5) t'ny. . . uymg (11) mlakk.'m dtn (12) Iqh mtpt (13) uy'ny.nn dtn . . .' W h e n the Lord of the Great Gods came to Ditanu and asked for the boy's (oracular) decision, Ditanu answered him: "You will a n s w e r . . . and your messenger to Ditanu arrived after he received the (oracular) decision." T h e n Ditanu answered:. . .' As long as no vocalized narrative texts are available, the question will remain undecided. O n the other hand, the use of parallelism or parallelizing style in Ugaritic texts is not confined to poetry. As in the Hebrew Bible, it is also found in many prose texts. T h e greater use of relative pronouns and particularly, a set of demonstratives (hnd-hnk/hndt-hnkt/hnhmt, including the definite article hn-, R a i n e y 1971, 160; C u n c h i l l o s 1983b) and interrogative pronouns (mn(m)-mnk(m); mh-mhkm, etc.) in prose texts such as economic texts, letters, contracts, etc. ( W a t s o n 1984d [1995], 62) is a fairly clear criterion, but not the only one. Several texts show a mixed style of prose and poetry. It is hard to say whether these literary texts are prose containing poetic elements, or a kind of poetry in which the rules of poetic parallelism are weakly

used. T h e r e can be no doubt that m a j o r works from Ugarit such as the Ba'al Cycle, the legends of Aqhat and Keret, the astral myth of Shahar and Shalim were composed as oral poetry, in which inserted prose elements clearly stand out in their context. Some minor myths and incantations were also composed in the concise prosody of the major works ( K T U 1 . 1 0 = R S 3 . 3 6 2 + , 1 . 1 2 = R S 2 . [ 0 1 2 ] , 1 . 8 3 = R S 1 6 . 2 6 6 and 1 . 9 2 = R S 1 9 . 0 3 9 + ) , but a few are in prose, or have prose sections alternating with mythical passages in poetry. For instance, K T U 1 . 1 0 7 = R S 2 4 . 2 5 1 + clearly opens with a mythical poem ( K T U 1 . 1 0 7 . 1 - 1 4 ) , but the poison-expelling charms on the reverse are in prose ( K T U 1 . 1 0 7 . 3 2 - 4 5 ) . T h e incantation text K T U 1 . 8 2 = R S 1 5 . 1 3 4 , so far as it is readable and understandable, is also couched in prose. Its sections are marked by quick changes of subject and by subordinate clauses in persuasive style, but no clear parallelism can be traced: \y\mhs.bcl [xxjy.tnn.wygl.wynsk.'d [x] [x]xy.lars[.id\y.alt.I ahš.idy.alt.in ly [x\b/dt.b'l.hz.rsp.bn.km.yr.klyth.wlbh [xx]x.pk.b gr.tn.pk.bhlb.k tgwln.šntk [xx]wšptk.l tsšy.hm.tgrm.l mt.brtk [xx]hp.an.arnn.qlJpš.hw.btnm.uhd.bClm [wa]tm.prtl.l rišh.hmt.tmt. Let Ba'al smite the breed(?) of Tunan and reveal and pour out the . . . (2). . . on the earth. Then I shall not feel the curse, then the curse will not for me be (3) harmfu1(?). The archer Reshef (is) between the two of you. He will shoot at his kidneys and his heart. Let your mouth . . . in the lowland, let your mouth resound in the woods, when you grind(?) your teeth (5) [against him]. And your lips will surely rejoice, if you keep until death your covenant. (6) [I shall ] myself, I shall ring out with the Sun's voice: the life of the serpents I shall take away, Ο Ba'al, and nail down the iron pole on his head (KTU 1.82.1-7) Likewise the minor myths of Yarikh and Nikkal ( K T U 1.24 - R S 5 . 1 9 4 ) , Astarte the Huntress ( K T U 1 . 9 2 = R S 1 9 . 0 3 9 ) a n d El's drunkeness ( K T U 1 . 1 1 4 = R S 2 4 . 2 5 8 ) show a mixed style of prose and poetry. Of the last text the medical prescription is entirely in prose ( K T U 1 . 1 1 4 . 2 9 - 3 1 ) . T h e first text has a narrative framework in prose, whereas some of the speeches are couched in verse. This may also be true of the fable K T U 1 . 9 3 = R S 1 9 . 0 5 4 ( D I J K S T R A 1994,

125):

(1) arh td.rgm.b gr (2) b py.t'lgt.b 1šn[y] (3) gr.tyb.b npšy.rg[m] (4) hzvt.b'l.ištm['.y gr'] (5) M ly.y pš. i[k(?)] (6) hkr(.b]Šry[ } . . .

The Cow let out a cry against the Mountain: 'In my mouth is stammering, on [my] tongue is agitation, in my throat rolls "thun[der]". Heed the word of Ba'al [O Mountain!] Listen to me, you braggart! Why this distress of my [bo]dy . . . ? . . . ' K T U 1.92 and 1.114 contain prosodie phrases and epic formulae borrowed from the major myths ( D i j k s t r a 1 9 9 4 , 1 1 6 ) , but the narratives are basically prose compositions despite this poetic flavour. For K T U 1.92, one has to assume that many verses have been shortened to monocola, if it was a piece of narrative poetry. In K T U 1 . 1 1 4 , the sequences of perfect (with inversion of the subject) and imperfect forms suggest the transformation of poetically structured verses into prose sentences (so also K T U 1.96 = R S 22.225.Iff.). Consider il dbh . . . sh . . . tlhmn.ilm.wtstn . . . After El slaughtered . . . called . . . the gods ate and drank . . . After he rebuked El his father, b il.abh.gcr. ytb il.. . El sat down . . . After El sat down at his il.ytb.b mrzhh yšt. . marzeah, he drank . . . After El wanted to go home, il.hlk I bth . . .y'msn.nn . . . they carried him . . . A similar mixed style is also present in the ritual K T U 1.161 - R S 34.126, the incantation K T U 1.169 = R I H 7 8 / 2 0 and the prayers K T U 1.65 = R S 4.474, 1.108 = R S 24.252 and perhaps 1.123 = R S 24.271. W e observe in these texts a certain repetitive style, even chains of adverbial clauses and comparisons: k qtr.urbtm.k btn. cmdm (3) kyim.zrh.k Ibim.skh, 'like smoke from a chimney; like a snake from a pillar; like a mountain-goat to the hill-top; like a lion to a lair' ( K T U 1.169.3-4); il hš il add (10) bcd spn bcd[ (11) ugrt (12) b mrh il (13) b nit il (14) bsmd il (15) b dtn il, etc., Ό El, hurry! Ο El, stand up, on behalf of Saphon, on behalf of Ugarit, with the javelin of El, with the spade(?) of El, with the span of El, with the threshings1edge(?) of El, etc.' ( K T U 1.65.9-15); b mrmt (8) b miyt.bzlm.b qdš, '. . . on the heights, in the lowland, in darkness and in the sanctuary' ( K T U 1.169.7-8); b cz (22) [rpi.] mlk.'lm. b dmrh.bI (23) [anh].bhtkh.b nmrth . . ., 'in the safety [of the Healer], the eternal king, in his protection, in his strength, in his dominion, in his benevolent power . . .'

( K T U 1.108.21-3; also 1.108.4-5). C o m p a r e also the chain of adverbial p h r a s e s a n d //^-greetings K T U 1 . 1 6 1 . 2 2 - 6 (with tht), 3 1 - 4 and the chain of epithets in K T U 1.100.1: um phi ph.lt bt.'n.bt.abn. bt.šmm.wth. . ., ' T h e m o t h e r of the he-ass, the she-ass, daughter of the spring, daughter of the stone, daughter of heaven and flood . . .'. Similarly, the chains of epithets and participial predicates in the liturgical prayer K T U 1.108.Iff. T h e y are all instances of poetic prose with repetition and even occasional parallelism within a prose context. Despite these poetic elements, such passages as K T U 1.65.9ff. and K T U 1.108.Iff. form one extended prose sentence, bursting the bounds of an originally poetic structure. Even the small fragment K T U 1.83 = R S 16.266, p e r h a p s part of an incantation (DE M O O R 1987, 181-2), shows this mixed style: . . . [ts]un. b ars (4) mhnm. trp ym (5) 1šnm.tlhk (6) šmm. ttrp (7) ym dnbtm. (8) tnrí. I šbtn (9) ist. trks (10) I miym Ibnm* (11) pi. tbtn. yymm* (12) hmlt.

ht. ynh*[r] (13) itph. mk* [ ] (14) thmr. [ ] . . . (text newly collated by PITARD

1998,

263

. . . [She we]nt out into the land of Mahanayim to vanquish(?) Yam with the forked tongue that licks the sky. She vanquished Yam with the forked tail, Tunan she muzzled. She bound him onto the heights of Lebanon (saying): You swill no longer humiliate me, Ο Yam. Mankind may be scared, Ο River (but) you will not see [me] collapse. You may foam [. . .] Related to the persuasive prose style of incantation is the prose of letters and an occasional wisdom text such as K T U 2.2 = R S 3.334 (II 2.5? = R S 1.020), perhaps dictated as a scribal exercise: (1) (2) (3) (4)

[/] rHšSyyš[al] [š]lm.brtš.yšl[m.rn\ \s\gr.l šlmt.šl[m] b*th.p šlmt.p* šlm

(5) b*t.lbnš.trgm*[?] (6) / stmt.I šlm.b[tk] (7) by.šnt.mlit.t[mla] (8) (9) (10) (11)

ymgyk.bnm.ta[rš] bnm.wbnt.ytnk* [bYLbny.šht.w\x.x.] [wš]h*t.msgr.bnk[ ]

(12) [wh]n.thmM[ ]

First of all, my friend, one should a[sk] for [p]eace. A man should keep the bolt(?) secure. (If) it is not safe, will its house be at peace? When it is safe, then the house will have peace. To a man you should say: 'It is not safe, so [your] house will not be at peace!' Believe me, after a full year has surely [passed] children you asked for will come to you. Sons and daughters will Ba'al give you. My son, spoil (it) and . . . you will have [spoi]1ed the bolt(?) of your children. [And 1o]ok, the word of Ba'al. . .

1.5

The prose of letters

Letters form the most interesting group in which to study the syntax of Ugaritic prose in the latter days of its floruit. W e shall not deal here with the well known formulae of sender and addressee (thm X rgm l Y), the airs and graccs (yslm Ik, ilm tgrk tšlmk, with its variants), the prostration formula (lpen PN[šbCd šbCd/tnid] mrhqtm qlt/qlny), enquiries about health (šal šlm formula), requests for an answer (wrgm [t]ttb l/cm-PN), to reply or to pay attention to the message (wbcly. . .ydc [rgmh]), or even the closing remark to reassure somebody (wap mhkm b Ibk al tst, etc.). These have all been thoroughly and properly studied in the past ( A h l 1 9 7 3 ; K r i s t e n s e n 1 9 7 7 ; P a r d e e 1 9 8 4 ; P a r d e e Whiting 1987; C u n c h i l l o s 1983a, 1989a). Here attention will be paid to the larger passages of prose only in so far as they are preserved and help to give a glimpse of late colloquial Ugaritic. Characteristic of this Ugaritic 12th-century written vernacular is the increased use of plene writing with -y in prepositions by, ly, ky, iky, construct state: ily ugrt and verbal forms tmgyy, etc., but also an increase in enclitic - j as a marker of direct speech, in particular in letters ( T r o p p e r 1994d, 474-5). T h e difference between plene written -y and enclitic -y is not always easy to detect ( T r o p p e r 1994d, 480-1). O t h e r changes in vocalization and phonology include the quiescent aleph, sbcd < šbC,id, yrš < yarš, etc., though also incidentally found in poetry, cbdnn < acbdnn; cdbk < a'dbk ( K T U 1.6 = RS 2. [009] + ii 21; 1.18 = R S 3.340 iv 22);ytmr < yitmr ( K T U 1.3 = RS 2. [014] + i 22), the shift of ' < h, for instance in im < hm ( K T U 2.15 = RS 15.007.8; 2.72 = RS 34.124.9, 10, 17; 3.9 = RS [Varia 14].6; T r o p p e r 1989b, 421-3); vowel harmony ulp (*ullupi) < alp (*allupi); ihy and uhy < ahy, ibr (Hbbiru) < abr (*abbiru). See further below § 8.1. T h e g r a m m a r and syntax of this late Ugaritic prose are enriched by the use of the article and demonstrative element hn, rarely independent in hn Ws, K T U 1.40 = R S 1.002 (perhaps also 1.114 = RS 24.258.28 and in assimilated form in K T U 2.70 = RS 29.093.15-6, w.hwt (16) hbt, 'and I repaired the house'), but frequently as part of a set of demonstrative pronouns: masc. hnd—hnk; fem. hndt—hnkt pi. hnhmt. Morphological developments may include the loss of diptotic plural and the occurrence of imperatives with prothetic aleph i: ibky, išhn ( K T U 1.161 = RS 34.126; T r o p p e r 1993a, 391-2); igr ( K T U 2.33 = RS 16.402.1); or aleph a: add ( K T U 1.65 = RS 4.474.9). Furthermore, we may note the wider use of the absolute infinitive

with separate personal p r o n o u n to continue a finite verb or imperative, wtbc ank ( K T U 2 . 1 7 = R S 1 5 . 0 9 8 . 6 ) ; w.ttb.ank ( K T U 2 . 3 8 = R S 1 8 . 0 3 1 . 2 3 ) ; wrgm hw/ank ( K T U 2 . 4 2 = R S 1 8 . 1 1 3A. 1 9 , 2 5 ) ; hbt hw (6) hrd w.sl hw (7) qrt, '. . . it eliminated the guard a n d pillaged the c i t y . . . ' ( K T U 2 . 6 1 = R S 1 9 . 0 1 1 . 5 - 7 ) ; w.ybl.hw ( K T U 2 . 7 2 = R S 3 4 . 1 2 4 . 2 7 a n d passim), though this is incidentally also attested in poetry ( S E G E R T 1 9 8 4 , § 6 4 . 4 2 ) . T h e variation in the use of the conjunction p(m)- is significant, a n d greater than in poetry (DE M O O R 1969, 2 0 1 - 2 ; K T U 2.2 = R S 3.334.4, 2.3 = R S 1.013+.19, 2.10 = R S 4.475.12, 2.14 = R S [Varia 4].12, 2.15 = R S 15.007.7, 2.23 = R S 16.078+.17, 2.26 = R S 16.264.7, 2.33 = R S 16.402.28, 2.70 = R S 29.093.27, 2.71 = R S 29.095.11 [pm], 2.72 = R S 34.124.1 1, 22, 42, 2.73 = R S 17.434.14; W A T S O N 1990e, 1994e) a n d the occurrence of pi and pn 'lest, you may not' ( K T U 1.83 = R S 16.266.11, 1.114 = R S 24.258.12). Also new, specific verbs such as dhl, 'to be afraid' ( K T U 2.16 = R S 15.008.12, 2.31 = R S 16.394.21); hbt, 'to knock down, eliminate' ( K T U 2.4 = R S 1.018.19, 2.47 = R S 18.148.16, 2.61 = R S 19.011.5), or verb forms like štn cm/l 'put something at the disposal of P N ' ( K T U 2.36+ = R S 17.435+.6, 13, 2.45 = R S 18.140.19, 2.50 = R S 18.287.16, perhaps also 2.32 = R S 16.401.7, 10, 2.39 = R S 18.038.35, 2.79 = R I H 77/25.3), presumably a Š-stem ofy/ntn; Š-stem 'hr 'to withhold, keep back (things)' ( K T U 2.42 = R S 18.113A.11, 2.79 = R I H 7 7 / 2 5 . 4 ) a n d the Gt-stem sal 'to make a request, e n q u i r e ' (2.17 = R S 15.098.15, 2.42 = R S 18.113A.23, 2.70 = R S 29.093.12, 2.71 = R S 29.095.10). Further, we may note the continued use of bl in compounds like blym, 'never' ( K T U 2.45 = R S 18.140.23; 4.272 = R S 17.118.7), bl bns, 'nobody' ( K T U 2.45 = R S 18.140.25), l.bl.hrb || Ibl ks, 'without a knife, or cup ( K T U 1.96 = R S 22.225.4-5) a n d bl sml ( K T U 1.169 = R I H 78/20.7), though also used in poetry: bl spr/hg ( K T U 1.14 = R S 2. [003]+ ii 37-8); blmt, 'immortality' ( K T U 1.17 = R S 2. [004] vi 27, etc.), the increased use of the perfect or participle Nstem ntkp ( K T U 2.10 = R S 4.475.14); nšUi (2.34 = R S 17.139.13); nskh (2.38 = R S 18.031.15); nmkr ( K T U 2.48 = R S 18.285[A].5); nplt (2.82 = R I H 7 8 / 1 2 . 4 , 11); also in administrative and legal texts: nkly ( K T U 4.213 = R S 16.127.24, 4.230 = R S 16.341.15; 4.280 = R S 17.236.6, etc.); night ( K T U 4.659 = R S 19.166.1) a n d nšlm ( K T U 4.328 = R S 18.008), but also a new preposition like ml(y), 'opposite (me)' ( K T U 2.50 = R S 18.287.12; 2.75 = R S 34.148.11). T h e r e is no

clear evidence that a special subjunctive m o o d was maintained in prose texts ( T r o p p e r 1991b, 3 5 3 - 5 , pace V e r r e e t 1988). T h e deictic or a n a p h o r i c use of -n, in p a r t i c u l a r in the apodosis of o m e n s ( D i e t r i c h - L o r e t z 1990a, 104; T r o p p e r 1994c, 466-7), but also elsewhere ( K T U 1.124 = R S 24.272.14, 2.37 = R S 17.438.10, 2.39 = R S 18.038.21, 2.42 = R S 18.113a.6, 10, 26) may suggest influence of the H u r r i a n 'article' -ni (but see T r o p p e r 1993b, 468). It is inherent in the nature of letters that we find narrative parts (reports using the perfect, e.g. K T U 2.38 = R S 18.031) and prescriptive sentences (instructions); but the f u n d a m e n t a l convention is that of an oral message exchanged between parties. This means that the m o d e of discourse in letters is usually persuasive, a m o d e of discourse couched in a kind of virtual verbal exchange between sender and addressee, but from the temporal perspective of the writer. A m o d e of discourse that varies with the field (diplomacy, international commerce, royal bureaucracy) and tenor (grade of social relationships between parties, see also P a r d e e - B o r d r e u i l 1992, 711). Letters are essentially a verbatim account of verbal exchanges between parties w h o argue their case. As part of the discourse the words of the other party are quite often quoted or referred to ( D i j k s t r a 1987). This style of persuasion is marked by statements accentuated with particles or adverbs such as ap/ p/ hn, emphatic use of the separate personal pronouns, interrogative particles and pronouns, such as ik, 'how', Im, 'why', my/mn(m), 'who(ever)', mndc, 'whoever knows, perhaps', rhetorical questions and other turns of speech introduced by ht, 'now', hm . . . p/w/zero, ' i f . . . then', etc. If we are aware of the sequence of inferences and thrust for persuasion m a d e coherent by a set of characteristic phrases and other cohesive devices in gramm a r and style, the letters are essentially a genre of persuasive discourse, even if stories are told, oaths are sworn and instructions are given within such a persuasive m o d e of discourse. Unfortunately only a few of the m o r e elaborate letters are complete or have a passage that allows for coherent translation and rhetorical analysis. Any translation given below admits that other translations are possible in some instances, but I a m concerned here only with the mode of discourse and the general thrust of a given passage. An example of such a dialogical discourse of persuasion including an oath is K T U 2.10 = R S 4.475.5-15, . . . trgds (6) w.l.klby (7) šrrít.hti (8) nhtu.ht (9) hm.inmm (10) nhtu.w.lak 'my.wyd (12) ilm.pkmtm (13) 'z.mid (14) hrn.nlkp (15) m'nk

(11)

. . . from Trgds and Klby I have heard that we suffered a heavy defeat. However, if we did not suffer a defeat, send me a message and, as for the hand of the gods, it will indeed be as strong as Death (against you) if your answer be negative(?). Letters often include reports. A good example is the letter from the king of Tyre to the king of Ugarit about the shipwreck of a Ugaritian fleet sent to Egypt, but there are also short reports such as K T U 2.17 = RS 15.098, 2.30 = RS 16.379, 2.33 = RS 16.402.4ff., 2.40 = RS 18.040, 2.61 = RS 19.011 and 2.75 = RS 34.148. We often find here as a typical report marker the formula 'the king my lord may know it/his word!' ( K T U 2.17, 2.33, 2.40, 2.75; perhaps also 2.35 = RS 17.327), or advice not to fear or to worry too much about the reported developments ( K T U 2.16 = RS 15.008, 2.30, 2.38 = RS 18.031 and 2.71 = RS 29.095). (10) anykn.dt (11) likt.msrm (12) hndt.b.sr (12) mtt.by (14) gšm.adr (15) nškh.wìb.tmtt (17) lqh.kl.dr'\ 18) bdnhm.w.ank (19) kl.npš (20) kMm.bd.rb.tmtt.lqht (21) w.Ub.anUhm (22) wanyk.lt (25) by.'/cy.'ryt (26) w.ahy.mhk (27) b.lbh.al.yšt This fleet of yours, which you sent to Egypt, was shipwrecked at Tyre. It was hit by a heavy storm. And the master of shipwrecks took all the cargo from their holds. However, I in turn took all their cargo, all the livestock for their provision from the hand of the master of shipwrecks and returned it to them. And your second fleet is in for repair(?) at Acre, but my brother should worry about nothing. (KTU 2.38.10-27). In particular, the mode and tenor of the discourse in international diplomatic letters becomes very persuasive, if not suggestive. In many instances we are here perhaps dealing with drafts and translations of official letters, of which the originals were sent and received by the royal chancelleries in official Akkadian. They negotiate about tribute, settle border disputes and deal with conflicting interests and loyalties. Consider the following anthology: ky.hkt. bt. mlk. thmk. hin [y] (6) \lj\rs.a rgmny [ .\'m. špš. štn ,\ank\ (7) [w]at.m[h]r.k[x] t.d.štt.b.ms[mt] (8) \ht\.Iqdm.udh.mgt.wmlkn.[χχ] (9) [m]hrt[.]nib.'mnkm.I.qrb.[xx] (10) [xj.i[x]t.w.at/my.l.mgt.[xY.] (11) [w.]mÌa[k]tk/my.l.likt ([12) [x] χ.km.šknt.ly.ht.hln.hrs.[xx] (13) | xxx] Jtnt.Cmy.'m.špšJtn[t] When you sent your message to the royal palace (saying), 'Herewith I put the gold of my tribute at the disposal of the Sun', [as for] you, the equivalent of the . . . as was settled in the trea[ty], you should present it now! Its payment(?) is due. And the king [said?], 'Tomorrow we shall return to you to deliver the . . ., but you have not come to me [. . . nor] did you send your embassy to me. [Now was this] as

arranged with me? So, the gold [of your tribute that was] put here at my disposal, I shall put (it) at the disposal of the Sun. (KTU 2.36+ = RS 17.435+.5-13) Li[th]dn.špš (6) ad[nk.'] bdk.ukJkn (7) k.'[bdm.]sglth.hw (8) w.b[nh].uk.ngr (9) rg[mh.l]adny.l.yfysr (10) w.[ap.y]d'.l.ydet (11) ht[.hm].l.špš.belk (12) 'b[dm.]sglth.at (13) ht[.hm].špš.b'lk (14) yd'm.l.yd't (15) 'myJps.b'lk (16) šnt.šntm.lm.<.>l.tlk Did not the Sun his father and his servant make an agreement(?) either that he would record that he and his sons would be servants of his own property, or that he would keep his promise (saying): 'My father will not lack anything' and [also]: Ί acknowledge you fully'. Now, if to the Sun your lord, you are servant of his own property, so then, if you recognize the Sun your lord fully, why did not you come to the Sun your lord for one, for two years? (KTU 2.39 = RS 18.038.5-16) Likewise K T U 2.23 = RS 16.078+. T h e tenor of such letters is often haughty, if not aloof on the part of the Great King, his queen and his officials. T h e Great King speaks about himself in the third person, the greetings are curt and there is no love lost, whereas the attitude of the vassal king is submissive and the airs and graces are elaborate, if not exhaustive. More than half the letter from Ammittamru to the Egyptian Pharaoh is filled with the repeated string of royal tides: špš.mlkJb.mlλ.mpm.mlkMcm.mlk.sdq.mlk.mlL·LbΊMI^wt msrm . . . , ' . . . the Sun, the great king, the king of Egypt, the benevolent king, the righteous king, the king of kings, lord of the whole country of Egypt . . .' ( K T U 2.81 = R I H 7 8 / 0 3 + ; see somewhat less tediously K T U 2.23 = R S 16.078+, 2.76 = RS 34.356.1-2, 9 - 1 0 [a draft?]). All along, the writer repeatedly praises his overlord as benevolent king, trying to negotiate a lower tribute: [mtn.ì]gm.bCly.nCm.hn.ksp.d.šsen (25) ['bdk.bš] nt.qdm.alpm.mznh (26) [ht. cbdk.] yirš.snp.ln.dym.hw, '. . . Another matter, Ο my benevolent lord. Look, the silver which [your servant] has paid for many years, two thousand (shekels) is its weight. (Now your servant] asks, will two thirds be sufficient for us? . . .'. Another flower of such submissive speech is K T U 2.23 = RS 16.078+. 15-24: w.an[k.cbdk.]d (16) arš[.hym.lšp]š (17) mlk.r[b.b'l}y.p.l. (18) hy.np[šh. «]/í (19) /./w.//[/.] spn.b'ly (20) w. urk.ym. b'ly (21) l.pn.amn.w.l.pn (22) il.msrm. dt.tgm (23) npš.špš [.]mlk.rbMy . . . And I am [your servant] who begs [for life to] the Sun, the great king, my lord. Then do I not pray for the life of his soul before Ba'al Saphon my lord, and length of days for my lord before Amun and

before the gods of Egypt who protect the soul of the Sun, the Great King, my lord? In the exchange of messages between the king a n d officials, we sometimes find such elaborate phrases in addition to the usual formulae of submission, ankn.rgmt.l.b'l.spn. (7) Ups.'lm.l.'ttrt (8) l.'nt.l.kl.il.alty (9) nmiy.mlk.'lm, ' . . . I pray to Ba'al Saphon, to the eternal Sun, to Astarte, to Anat, to all the gods of A1ašia for the splendour of an everlasting kingship!' ( K T U 2.42 = R S 18.1 13a.4-9). N o t only are developments reported, but problems are also discussed a n d instructions given in the same persuasive sort of style. T h e U g a n d a n king reports violations of his territory by Egyptian caravans, a n d the Hittite q u e e n , probably in consultation with the Hittite deputy-king of C a r c h e m i s h , instructs him to direct t h e m past Q a d e s h through the valley of the O r o n t e s ( K T U 2.36+ = R S 17.435+ .16ffi; D I J K S T R A 1989, 142-4). An interesting instance is the letter f r o m G e n e r a l Iwri-tarruma ( K T U 2.33 = R S 16.402), reporting an attack by the kings of Mugišhe (Alalakh) a n d , p e r h a p s , Nuhašši against N i q m a d d u II. After some explanations a b o u t the course of strategy taken, he c o m e s with a m a z i n g e l o q u e n c e to the subject of reinforcements: w.mlk.b'ly (23) ImJkn.hnk (24) l.'bdh.alpm.sswm (25) rgmt.Hy.lh.lm (26) l.ytn.hm.mlk.'ly (27) w.hn. ibm.šsq.ly (28) p.l.ašt.atty (29) riry.lh.l pn.ib (30) ht.hm.yrgm.mlk (31) b'ly.tmgyy.hn (32) alpm.ššwm. hnd (33) w.mlk.b'ly.bnl (34) bnny.'mn. (35) mlakty.hnd (36) ylak.'my (37) w(l.lh.hn (38) [a]lpm[.]ššumi (39) [x].l.[yx]xs/l.w.ib And the king my lord, why did he assign such a thing to his servant? Two thousand horses, you said, would come soon! Why has the king, my lord, not provided them yet? Look, the enemies are pressing me hard, but I cannot put my womenfolk and children just in front of the enemy! Now, if the king my lord orders it, they will arrive here, those two thousand horses. And the king my lord may also send to me mediators(?) with this my embassy. And let them come up soon hither, the two thousand horses [and] let him not [. . .] and withdraw! (KTU 2.33 = RS 16.402, 22-39) Fragments of such eloquent pieces of prose, in which someone is pleading his case, are also found, for instance, in K T U 2.41 = R S 18.075, 2.42 = R S 18.113A, 2.45 = R S 18.140, etc., but unfortunately they are too broken for their lines of reasoning to be followed in detail. In K T U 2.70 = R S 29.093, we find a complaint a n d a request. Obviously one of the senders of this letter is a w o m a n (as also K T U 2.11 = R S 8.315):

hlny.bn.'yn ( 12) ystal.'m.amtk (13) lak.lh.wkhdnn (14) w.ank.hrš ( 15) Iqht.w.hwt (16) hbt.wlm.tb (17) bn.'yn (18) w.lqh.tqlm (19) ksp.bd.amtk tn.'bdk (21) tmt.'mnk (22) k I On.akl.Uun (23) w.k tšal (24) bt.'bdk (25) w.kymgy (26) ebdk.l ihn (27) 'mk.p.l.ysb' I (28) hpn.l b'ly (29) mnm.it.l 'bdk

Here, the son of Ύη came to request (silver) from your handmaid. It was sent to him, but he hid it. I enlisted a contractor and I repaired the house. Why has the son of Ύη returned to take the two shekels of silver from the hand of your handmaid? And two of your servants are there with you in order that you may give food to them. Now if your servants ask for accommodation and if they arrive in peace to you, will not then a handful satisfy them? T o my lord belongs everything that your servants own. (KTU 2.70 = RS 29.093.11-29) Letters evidently often react to messages received about information, allegations of disloyalty, requests for help or neglect to pay outstanding debts a n d tardiness in fulfilling obligations. T h e r e is nothing new u n d e r the sun! M a n y of the letters refer to messages received and even quote from them (examples in D i j k s t r a 1987a, 3 8 - 9 to which K T U 2.36+ = R S 17.435+ passim, m a y be added). lm.tlikn.hpt.hndn (11) p.mšrrít.mlk (12) inn.im.bn.qln* (13) im.bn.alyy.im (14) msm't.mlk (15) wtlkn.tn. tnm (16) 'my.wttbm.lby (17) wlht.bt.amr (18) ky.tdbr.umy (19) l.pn.qrt (20) im.ht.l.b (21) msqt.yt_bt (22) qrt.p.mn (23) likt.ank.lht (24) bt.mlk.amr (25)ybnn.hlk (26) 'm mlk.amr (26) w.ybl.hw.mit (28) hrs.w. mrdt.l (29) mlk.amr.w.lqh.hw (30) šmn.b.qrnh (31) w.ysq.hw.l.riš (32) bt.mlk.amr (33) mnm.ht[at.brt]

Why did they release these two, as if they were not subordinate to the king? Either being the son of Qln, or the Son of Alyy, or a subordinate of the king, these two should have come to me together and should have broken my heart. As for the letter about the daughter of the king of Amurru—when my mother speaks on behalf of the city: 'If now the city will not go on to live in anxiety, who then must I send with the letter about the daughter of the king of Amurru?' Let Yabninu go to the king of Amurru and let him bring a hundred (shekels) of gold and the repudiated woman(?) to the king of Amurru and let him take oil in his horn and pour it on the head of the daughter of the king of Amurru. Whatever sin [she] committed [she will be free?] . . . ' (KTU 2.72 = RS 34.124.10-33) It is characteristic of this m o d e of prose to construct complicated sentences in an elaborate rhetorical, often conditionally phrased style. For instance, the passage quoted from K T U 2.39 = R S 18.038.1 1 - 6 is probably one long sentence. Sentences introduced with conditional

hm, temporal k(y)- (preceded or followed by the conjunction p(m)-, but often also simply the copula w-, K T U 2.10 = R S 4 . 4 7 5 . 9 - 1 0 , 2.31 = R S 16.394.16ff., etc. or without connection, K T U 2.33 = R S 16.402.30-1) are numerous, as are statements a n d conditional sentences emphasized with hn, ht and ap (e.g. ht hm . . . K TU 2.10.8-9, 2.33.30, 2.39.1 I f f ; wap ht... 2.3 = R S 1.013+.20; wap.ank.. . 2 . 1 1 = R S 8.315.13, 2.33.15, 2.41 = R S 18.075.19), and rhetorical a n d real questions with or without ik(y), mh(y), e.g. w.k tšal bt cbdk w k ymgy c bdk l šlm cmk pi ysbc Ihpn, 'And if your servants ask for a c c o m m o d a tion and if they reach you safely, would not a handful be sufficient?' ( K T U 2.70 = R S 29.093.23-8, also 2.23 = R S 16.078+. 17ff. and perhaps, 2.39.5ff). Quite often in the prose of letters and elsewhere (e.g the narrative K T U 1.114 = R S 24.258, the ritual K T U 1.112 = R S 24.256.6-7) the preferred word order V S is changed to SV: wum tsmf} mad. . ., 'and my mother should rejoice g r e a d y . . . ' ( K T U 2.16 = R S 15.008.10); whn.ibm. šsq.ly . . ., 'And look, the enemies are pressing m e hard . . .' ( K T U 2.33 = R S 16.402.27), perhaps for emphasis (other examples T R O P P E R 1994c, 467-70). A related interesting p h e n o m e n o n in this m o d e of prose is casus pendens or nominative absolute ( W A L T K E O ' C O N N O R 1990, § 4.7), e.g. in the oath-sentence: wyd ilm p kmtm cz mid hm ntkp m'nk, 'as for the h a n d of the gods, it will indeed be as strong as Death, if your answer is negative(?)' ( K T U 2.10 = R S 4.475.11-5); note the incongruity of yd (f.) and cz (m.)5 but this seems to be the idea (pace P A R D E E 1987; W A T S O N 1990, 8 1 - 2 , 1994, 495); [w].b.ym.k.ybt.mlk (15) [t]ydr.w.ap.ank (16) [i]hd.lgr. amn, '[And] as for today, if the king stays in [T]ydr, I for my part took hold of M o u n t A m a n u s . . . ' ( K T U 2.33 = R S 16.402+. 15-6; for the geographical n a m e ty[n\dr, see D I E T R I C H - L O R E T Z 1994, 65-7); wmlk bcly Im skn hnk rbdh 'and as for the king my lord, why did he assign such a thing to his servant?' ( K T U 2.33.22-4); hn.mrt d štt ašsu b Idtk, 'Look, the patrimony which was (legally) settled, I shall release after you have given birth' ( K T U 2.34 = R S 17.139.32-3); cmy špš b'Ik snt sntm Im Itlk, 'to the king your lord, why did you not come for one or two years? ( K T U 2.39 = R S 18.038.15-6); wlht akl Iy likt cm špš bclk ky akl bhwtk. inn. špšn tubd, 'As for the tablet about food, when you sent to the Sun the message that there is no food in your country, the Sun was indeed d i s h e a r t e n e d . . . ' ( K T U 2.39.17ff). T h e r e are m a n y other examples, for instance, in administrative texts: mlbš.trmnm (6) k.ytn.w.b.bt (7) mlk.mlbš (8) ytn.lhm, '. . . As for the clothing of the

trrnnm-gods when it is old, then in the house of the king clothing should be given to them' ( K T U 4.168 = RS 15.082.5-9).

1.6

Performative and prescriptive prose

In this group I would include the medical texts, omens, rituals and contracts, though a distinction from other prose texts is not always easy. For instance, the protocol of necromancy through the medium Dtn, one of the royal deified ancestors, contains as a report an amalgam of narrative discourse, ritual and medical prescriptions ( D i e t r i c h L o r e t z 1990a, 212, 216). T h e Ugaritic liturgical prayer K T U 1.108 = RS 24.252 is an interesting amalgam of performative ritual language and descriptive hymnic prose with occasional parallelism. Though the style of contracts is basically performative and very formal, lym hnd RN mlk ugrt ytn bt/šd. . ., ' O n this day, R N the King of Ugarit gave the house, the field . . ( K T U 3.2 = RS 15.111; 3.5 = RS 16.382); lym hnd iwrkl pdy . . ., ' O n this day, Iwrikalli redeemed P N . . . ' ( K T U 3.4 = RS 16.191+), we occasionally find instances of persuasive style complete with metaphor: . . . l.yihd stqšlm (2) b unt.km.špš (3) d brt.kmt. (4) br stqšlm (5) b unt cd clm, 'No one shall take Stqšlm in corvée-service. As the Sun who is free, so Stqšlm is free from corvée-service for e v e r . . .' ( K T U 2.19 = RS 15.125). T h e largest category in this section comprises rituals. They are written with an exasperating concision ( P a r d e e - B o r d r e u i l 1 9 9 2 , 709). Many of the ritual texts look like administrative texts: a list of gods and the sacrifices administered to them. Some of them are even simple onomastica of gods ( K T U 1 . 4 7 || 1 . 1 1 8 ; Akkadian RS 2 0 . 2 4 , N o u g a y r o l , 1 9 6 8 , 4 2 6 4 ) , perhaps a kind of canonical list of gods. Excerpts from this list and others, are extended in the rituals by sacrifices administered to them. This may happen by simple juxtaposition of name and sacrifice, e.g. b'l i, atrt s tkmn wšnm š 'nt š, ršp /, etc. ( K T U 1 . 4 1 = RS 1 . 0 0 3 + . 15-6, with parallels), but also with a dative I (wtn sm lbclt bhtm, csrm lins ihn, K T U 1 . 4 1 . 5 , see also 1 . 8 1 = RS 1 5 . 1 3 0 ; in H u m a n texts dative -d/-da, plural -tt/-asta). These lists can be preceded by date formulae of months and days: byrh ris yn. bymhdt. . . btltt esrt, 'In the month Risyn (First Wine) on the day of the new moon . . . on the 13th day . . .' Occasionally, and often interspersed between this lists of sacrifices, we find references to processions and other cultic rites to be performed. A basic question remains as to whether these rituals were meant to be 'prescriptive'

or 'descriptive' ( L e v i n e 1963; P a r d e e - B o r d r e u i l 1992, 709). M y own preference is to see the references to ritual activities as habitual. This is consonant with the frequent use of imperfects indicating an incomplete action, if not an action to be performed, i.e. a jussive mode, or a prescriptive imperfect ( P a r d e e - B o r d r e u i l 1992). This prescriptive nature of the rituals is also consonant with the fact that some rituals have complete or partial duplicates. T h e ritual prescriptions usually mention or imply the king and the priest as officiants: btltt 'srt. yrtfis. mlk.brr, 'on the 13th day the king shall wash himself clean' ( K T U 1.41 = R S 1.003+.3 a n d passim)·, wynt qrt/db I cnt walp wš l il wb urbt y tk gdlt ilhm, ' a n d he shall p r e p a r e a city-pigeon before Anat, a cow a n d a sheep for El a n d in the chimney he shall p o u r out (the blood o f ) the cow for the ancestor-gods(?)' ( K T U 1.41.10 a n d passim)·, wtlhm. att. . . kl lylhm bh . . . , ' . . . a w o m a n m a y eat (from i t ) . . . nobody should eat f r o m i t . . . ( K T U 1.115 = R S 24.260.8, 10).;. . .yšt.rpu.mlk.Clm. wyšt (2) [/] gtr.wyqr. . wtst.'nt.gtr. . ., '. . . let the Healer, the eternal king, drink a n d let [the god] G a š r u - a n d - Y a q a r u d r i n k . . . a n d let A n a t of G a š r u d r i n k . . .' ( K T U 1.108 = R S 24.252.1-2, 6). This prescriptive prose style is particularly clear in the ritual where the king a n d the officiant priests celebrate together: id[.yd]bh.mlk l.prgl.srqn.b.gg (51) ar[b]'.arb'.mtbt.azmr.bhJJr[p] (52) al[p.]w.š.šlmm.pamt.šb'.klbh (53) yr[gm.]mlk.sbu.špš.w.hl.mlk (54) w.l[bí]n.spm.w.mh[y pn]h.t[t]tbn (55) b.b[t].w.km.it yš [u.l.]smm. yd[h] Then the king shall sacrifice to PRGL SRQN on the roof, on which there are four by four dwellings of foliage: a sheep as a burnt offering and a cow and a sheep as a peace-offering. Seven times the king shall say whole-heartedly (the prayer): 'Host of the Sun and army of Maliku'. Clothe him with covers(?) and wipe his face, and let him return to the temple; and when he is present, he shall raise his hands to heaven. (KTU 1.41 = RS 1.003+.50-5) . . . [ s]l[m.uš]hiy ylbš (23) mlk.ylk.lqli.ilm (24) atr.ilm.ylk.p'nm (25) mlk.p'nm.yl[k.] (26) šb' pamt.lklhm . . . [the] sta[tue of Uš]hari one shall dress. The king shall proceed to accompany the gods, people shall go behind the gods barefoot. The king shall also go barefoot, seven times for both(?) of them. (KTU 1.43 = RS 1.005.22-7) A n o t h e r instance of performative style in ritual, but in the second person plural, is the sacrificial ritual of the ancestor cult ( K T U 1.161 = R S 34.126). T h e perfective forms have been u n d e r s t o o d as prescriptive narrative (as also in the H e b r e w Bible Lev. 8 - 9 ; d e M o o r

1976, 335; H E A L E Y 1978b, 85), but performative or precative perfects:

I

would suggest that they are

spr dbh ζΐτη (2) qritm.rpì.a[rs] (3) qbitm.qbs.d[dn] (4) qra.ulkn.rp[a] . . . tqdm C sr (31) šlm.šlm 'mr[pi] (32) wšlm.bn/ìJlm.tiyl (33) šlm.bth.šlm.ugrt (34) šlm.lgrh Book of the Sacrifice of the 'Shades': Y o u shall n o w invoke the Healers of the E a r t h , you shall n o w s u m m o n the Assembly of D D N : 'Invited be U L K N , the H e a l e r . . .' Y o u shall a p p r o a c h the festive assembly by saying the peace: Peace to A m m u r a p i ; p e a c e to his sons; p e a c e to Tryl\ p e a c e to his house; p e a c e to Ugarit; p e a c e to its gates.

Performative imperatives and jussive imperfects 2nd person sing, and plur. also occur in rituals: b tšC Cšrh (2) trbd.'rs []pd-{?>)-iy.b št.mlk.. . (25) I pn ll.trír (26) V/. . ., ' O n the 19th day, you shall prepare the nuptial bed of Pidriya in the suite of the king. . . before the night, you shall shake up the b e d . . . ' ( K T U 1.132 = RS 24.291.1-3, 25-6). Note also the poetic prose incipit of the prayer in K T U 1.119 = RS 24.266: k gr 'z.tgrkm.qrd (27) hmytkm.'nkm.lb'l.tsun, 'When a strong one attacks your gates, a warrior your walls, you shall raise your eyes to Ba'al (saying:)...' (R.P. G O R D O N 1991, 161-3). Such performative style is characteristic of Ugaritic rituals, just as for ancient Near Eastern rituals in general. A last good example is the expiatory ritual K T U 1.40 = RS 1.002 found in several mutilated copies: A n d present a y o u n g he-ass to obtain the re1ease(?) of the Ugaritians a n d the expiation of the sojourners within the walls of Ugarit, the expiation of Tman, the expiation of crmt, the expiation of Ugarit a n d the expiation of N i q m a d d u . W h e t h e r y o u r faithfulness d e p a r t e d f r o m the Q a t i a n clans, the D a d m i a n clans, the H u m a n clans, the Hattic clans, the A1ašian clans, the clans of Gbr, the clans w h o r o b b e d you, the clans of y o u r faithfu1(?), the clans of Qrzbl—whether y o u r faithfulness d e p a r t e d either because of y o u r a n g e r or because of y o u r i m p a tience, or because of the quarrels you h a d , w h e t h e r your faithfulness d e p a r t e d for sacrifices a n d oblation, o u r sacrifice we should sacrifice. T h i s is the oblation we oblate, this the offering we offer. Let it rise to the F a t h e r of the gods, let it rise to the family of the gods, to T u k m a n a n d Š u n a m , this he-ass. ( K T U 1.40.26-34)

T h e omen texts show a fixed pattern of prose sentences and syntax, like their Mesopotamian counterparts. Obviously, this type of literature derived from Babylonian tradition. Fragments of a dreambook ( K T U 1.86 — RS 18.041) and a collection of astronomical omens ( K T U 1.163 = R I H 78/14) have been preserved. T h e birth omens are represented by two main groups: the summa izbu 'If an abortion'

( K T U 1.103+ = RS 24.247+) and the šumma sinništu (.. .) ulid 'If a woman gives birth to a . . .' omens ( K T U 1.140 = RS 24.302). T h e former is the better preserved collection and shows the well-known protasis-apodosis k tld X . . . Y(-rì) structure. Characteristic of omens is the subject (+«) verb order in the apodosis ( T r o p p e r 1 9 9 4 c , 469-70): O m e n s of small livestock. [If] a ewe(?) bears a stone, then the m a j o r ity in the land will fall victim; (if) a snake follows after it, the y o u n g of its cattle will be weak, (if) also a . . .; famine will be in the land, (if) it has no . . ., the country wall be destroyed; a n d (if) [its belly] is open, a famine will be in the land . . .

Certain omens were checked by a second opinion of the haruspex ( K T U 1.78 = RS 12.061), but inspection of the omina also drew forth ritual activity to eliminate the effects of bad omens. T h e lung model K T U 1.127 = RS 24.277, for instance, reveals an interesting instance of relationship between omen interpretation and ritual. It may have been an instruction model, but this is far from certain, since parallels are still lacking ( M e y e r in D i e t r i c h - L o r e t z 1990a, 270-1). Nevertheless certain parts arc marked by 'borders' containing small texts with ritual instructions, seemingly derived from omina observations. T h e most interesting instance is the instruction of a scape-goat rite to eliminate the danger of a city taken or a plague (Dietrich L o r e t z 1990a, 32-38, 270-1): hm qrt tuhd.hm mt y'l bnš (31) bt bn bus yqh

c

z (32) wyhdy mrhqm

If a city is besieged (and) if plague attacks a m a n , the citizen's household will take a goat a n d banish it to the remotest parts.

(KTU 1.127.30-2) However not many other examples of such characteristic prose can be gleaned from the rituals. T h e style is often extremely concise and many phrases are still poorly understood, for instance Crb.špš. whl.mlk ( K T U 1.87 = RS 18.056.56-7; 1.46+ = RS 1.009.9-10, 44, 1.112 = RS 24.256.9, 1.119 = RS 24.266.4, 2 3 - 4 , 1.126 = R S 24.276.23?, 1.132 = RS 24.291.27-8), variant sbu špš (whl ym crb špš) whl mlk ( K T U 1.41 = RS 1.003+.47-8, 5 3 - 4 , 1.112.14-5), also ttb rgm (bgn) whl mlk ( K T U 1.106 = RS 24.250+.23-4, 33). T h e context suggests a kind of morning or evening prayer spoken by the king.

2

Ugaritic

Wilfred

2.1 2.1.1

G.Ε.

Poetry

Watson

Introduction

Scope

Almost from the very beginning of Ugaritic studies, account was taken of the verse component, 1 but as yet there has been no fullscale description of Ugaritic poetry nor has there been an exhaustive examination of the principles involved, although several partial surveys are available. 2 T h e best and fullest account so far is P a r k e r (1989, 7-98) 3 while a very detailed analysis of parallelism including phonological features in only one short passage ( K T U 1.3 = RS 2.[014]+ i 2~25) has also been completed ( P a r d e e 1988c, 1-67). In view of this state of affairs, the presentation here is not systematic; instead, it sets out a number of related topics under several headings. First, though, some account is required of the nature of the material under consideration and the problems it entails. 2.1.2

The texts

T h e corpus of Ugaritic poetry is fairly easy to define: it includes all the mythological and epic texts, which are mostly in narrative verse, and excludes letters, legal and economic texts as well as most of the ritual texts as non-poetic. 4 Some texts, however, are borderline, e.g. K T U 1.119 = RS 24.266:28-36); K T U 1.161 = RS 34.126 ( P a r d e e 1993a) and some verse texts contain prose elements ( C r o s s 1974) e.g. K T U 1.113 = R S 24.247:1-11 = verse; 12ff. = prose list. Some

1

DUSSAUD

1935,

1 9 3 6 (19412); GASTER

1933,

GINSBERG

1936,

OBERMANN

1936,

as well as C O P P E N S 1 9 4 6 , Ρ Α Ί Τ Ο Ν 1 9 4 4 , Y O U N G 1 9 4 8 , 1 9 4 9 , etc., though inevitably the relationship to Hebrew poetry was to the fore. 2 T / T § 1 3 , K O S M A L A 1 9 6 6 , 1 7 2 - 6 ; G R A Y 1 9 6 5 passim·, S E G E R T 1 9 7 9 ; 1 9 8 3 ; 1 9 8 4 , 109-10;

DEI. O I . M O LF.TF. 1 9 8 1 ,

3

31-62;

AVISHUR

1994,

13

25.

For a critique cf. DEL O I . M O L E T E 1990, esp. 190-4. 4 'Features known as parallelistic may also occur in prose texts. Some Ugaritic epistolary formulae exhibit parallelism, and even some of the expressions in the body of the letters are arranged in parallelistic structures. This criterion alone cannot determine whether the text is poetry or prose. The other distinguishing feature of a poetic text is its division into prosodie units of approximately the same length' (SEGERT

1979,

730).

speech introductions are extra-colonic but others comprise an integral part of the verse (see § 5.2.4.1). T h u s , the corpus includes the Baal Cycle (KTU 1.1-6 = RS 3.361, 3.367, 3.346, 2. [014]+, 2. [008]+, 2.[022]+, 2.[009]+); the 'Stories' of Keret and Aqhat ( K T U 1.14-16 = R S 2.[003]+, 3.343+, 3.325+; K T U 1.17-9 = RS 2. [004], 3.340, 3.322+); the Rapi'uma texts ( K T U 1.20-1.22 = RS 3.348, 2.[019], 2.[024]); a wedding poem ( K T U 1.24 = R S 5.194); incantations ( K T U 1.82 = R S 15.134; K T U 1.96 = R S 22.225; K T U 1.100 = RS 24.244; K T U 1.107 = R S 24.251+); a prayer ( K T U 1.119 = RS 24.266:28-36; see d e l O l m o L e t e 1987, W a t s o n 1996); a mythic marriage ritual ( K T U 1.23 = RS 2.002; cf. W a t s o n 1994a); a funeral ritual ( K T U 1.161 = RS 34.126; see P a r d e e 1993a) and various fragments. 5 A broad distinction can therefore be made between longer texts (mosdy narrative) and relatively short texts (the remainder), with K T U 1.100 = RS 24.244 occupying a mid-posidon. O n the whole problem see § 5.1.4. 2.1.3

Problems in studying Ugaritic verse

Apart from the poor condition of some tablets which readings uncertain as well as leaving large gaps in the and the fact that the corpus is relatively small, certain culties combine to make the analysis of Ugaritic verse

makes many poetic texts, specific diffiproblematic.

• Principally, for most of the tablets the stichometry is uncertain or at least not made clear.6 Exceptions include K T U 1.10 = RS 3.362+ (and to some extent K T U 1.23 = R S 2.002), where the verse-line corresponds to tablet line (cf. d e l O l m o L e t e 1991a, 463 and n. 3, W a n s b r o u g h 1983 and W a t s o n 1982). Occasionally, stretches of text are written with correct stichometry (e.g K T U 1.15 = RS 3.343+ iii 1-23, with 23 consecutive verse-lines which match the lines on the tablet) and some of the worst sections are in Aqht. • At times, the vocabulary poses difficulties (e.g. gmn in K T U 1.6 = RS 2. [009]+ i 19ff.). Generally, this is not particularly an obstacle to determining poetic structure. However, when whole passages which are repeated are not entirely understood, (e.g. K T U 1.3 = RS 2. [014]+ iii 14-7 and par.) problems do arise.

5

E.g.

MARGALIT 6

See

1.8 = R S 3.364; K T U 1.83 = R S 16.266—see P I T A R D 1998, DEL 1996; K T U 1.92 = R S 19.039+—see DE M O O R 1985, D I J K S T R A 1994, 1989b; K T U 1.93 = R S 19.054—see D I J K S T R A 1986.

K T U

OLMO LETE

LORETZ

1976,

1986,

KOTTSIEPER -

LORETZ

1987.

• T h e lack of vocalization and in general the absence of a tradition of pronunciation (although syllabic spellings are of some help where available) mean that we do not know how this poetry was recited or sung and there is no indication of metre (see below), especially as it is not known for certain where the stress lay. • T h e r e are several scribal mistakes; in addition, very often lines appear to be omitted (as is apparent from comparison of nearparallel passages) 7 but it is not always clear when this was intentional (see § 5.7 on expansion/contraction) and when not (see d e M o o r 1978a, 130-1). • Almost all Ugaritic narrative is in verse, with no strictly comparable material in prose. Exceptions are the letters and to some extent the ritual texts, but these have their own special styles and to some extent Akkadian influence is evident in the letters, some of which were translations (see, e.g. M a r q u e z R o w e 1 9 9 2 ) . It is, for instance, difficult to describe 'normal' syntax and then compare it with the syntax of poetry, due to the lack of material ( S i v a n 1 9 9 7 , 2 1 0 ) (however see § 5 . 2 . 3 . 3 and § 5 . 2 . 3 . 5 below on verb forms and ellipsis). 2.1.4

Approaches

O n the positive side, some assistance is provided by the tablets and several factors make study of Ugaritic verse easier. Quite often, as has been noted, whole passages of verse are repeated. Sometimes the correct stichometry is used and as has been said, the corpus is reasonably well-defined. Enough of the texts is understood for a degree of certainty in describing their poetic aspect and information from similar traditions (Akkadian, Hebrew, Phoenician) can be of some help. Also, text editions with translations now tend to include comments on poetic structure. 8 M.S. Smith, who provides vocalization, semantic parallelism, word-count, syllable count, comments: Some remarks bearing on various sorts of parallelism—syntactic, morphological and sonant. . .—often follow the presentation of cola, although syntactic parallelism is not treated according to any specific system. Rather, it has been my interest to indicate how these sorts of

7

See, e.g.

8

DEL O L M O L E T E

MARGALIT

ROSENTHAL

1939.

1991, P A R D E E 1988a, S M I T H 1994a, W Y A T T 1998c, etc.; see also 1980b (with the critique by DEL O I . M O L E T E 1983) and M A R G A L I T 1989a.

parallelism may bind and contrast cola, especially in the absence of apparent semantic parallelism (M.S. S m i t h 1994a, xxxiv).9 2.1.5

Metre

It is now generally accepted that Ugaritic verse is not metrical. Young discussed the possible existence of metre in Ugaritic verse and concluded: 'If there is any metric [i.e. metrical] system in Ugaritic, it should show itself in some regular manifestation observable in the texts themselves without our trying to fit any system into t h e m ' (YOUNG 1950, 124). T w o factors led him to this conclusion: (1) 'there is no consistency in the sequence of similar stich combinations within a poem or within sections of a poem, much less a consistency of an accent-per-word pattern for the successive stichs themselves' (YOUNG 1950, 124) (2) 'if an accentual meter existed at Ugarit, it might be seen in consistency between parallel passages within single poems, if meter were important to the composer, singer, or reciter of the poetry. N o such consistency is found' (YOUNG 1950, 128). He added: Ά poetry in which the outstanding feature is parallelism of thought; a poetry written in a language in which the majority of words are of one, two, or three syllables, and in a language in which almost any clause can be couched in from two to four words, is a poetry which naturally lends itself to the creation of lines of uniform metric length' (YOUNG 1950, 132). His conclusions have been accepted by Pardee who states that since there is no 'regular, predictable or at least observable recurrence' of rhythmic units in Ugaritic poetry it has no metre (PARDEE 1981, 116) and PARKER (1989, 9-10) provides additional arguments. 1 0 It has been suggested by de M o o r that Ugaritic verse is written in free rhythm to match its musical accompaniment. Such music was probably led by a soloist and would fit the short stichoi of ancient West Semitic poetry. T h e fact that consecutive lines had a degree of regularity can be explained by the poets' 'pursuit of symmetry' (DE M O O R 1978a, 132). Ultimately, 'parallelism was the primary structural principle of Ugaritic poetry and . . . length

9

T h e aspect of literary translations of these texts cannot be discussed here; cf.

SEGERT 10

1979,

MARGALIT

259-72).

738,

HORWITZ

cf. SEGERT

PARKER

1975,

1979,

1973, 730

1990 and

LEWIS

1996.

endorsed by Z E V I T 1 9 8 3 (but cf. P A R D E E 1 9 8 1 - 2 , argues that the word-divider had a metrical function, but

1995, 215; (on

K T U

1.24

=

RS

5.194)

and

WANSBROUGH

1983.

of line was only prescriptive in the general principle of approximation' ( P a r d e e 1 9 8 1 , 1 2 6 ) .

2.2

Language

In general, the language of Ugaritic verse is archaic ( P a r d e e 1 9 8 1 - 2 , 267) and to some extent it also has its own vocabulary (see above). In respect of verb forms, the rules followed are unlike those for prose. T h e use of the qtl (or qatala) form in Ugaritic verse is distinctive: 'While *qatala constitutes the characteristic form for past narration in the prose texts, its poetic usage is more restricted; the prefix forms [yqtl, etc.] arc the norm for poetic narration' ( M . S . S m i t h 1 9 9 5 , 7 8 9 , following F e n t o n 1 9 7 3 ) . " Besides being used for the stative, reporting the past, continuing other perfects, the pluperfect, subordination and the performative, there arc three usages specific to Ugaritic verse. These are 'contrast with prefix forms', 'report of action c o m m a n d e d in the imperative' and 'delimitation of a section' (M.S. S m i t h 1 9 9 5 , 7 9 0 , with further references). Different or identical verb forms can occur in parallel lines (see § 5 . 2 . 3 . 3 ) . Little research has been carried out on syntax, but for the Story 12 of Keret, it has been shown 'that word and sentence order remains completely unaffected by the type of verbal form present' ( W i l s o n 1 9 8 2 , 3 1 ) . 1 3

2.3 2.3.1

Parallelism and the verse-line

General

T h e basic component of Ugaritic verse is the verse-line which can be divided into two (parallel) half-lines or provided with a parallel line to form a bicolon. It is generally accepted, then, that parallelism is a fundamental component of Ugaritic verse, and it differs from prose precisely because parallelism is so prevalent." T h e r e are several 11

However, cf. § 4.4.2.5. According to G I B S O N 1975, Keret is a myth and Ai/liai a folktale. On literary forms in Ugaritic cf. DEL O I . M O L E T E 1984b. I;! W I L S O N ( 1 9 8 2 , 3 1 ) concludes: 'Such interchangeabilitv of verb forms with no appreciable effect on word order or sentence structure may well indicate a "frozen" state for some poetic passages, in which word order is fixed and immovable. Such passages could be inserted at any point of a narrative with only the necessary change of verbal form in the new context'. See also SIVAN 1 9 9 7 , 2 1 0 - 4 . 14 'There is little disagreement that the most obvious and pervasive convention of the Ugaritic poems is parallelism' ( P A R K E R 1989, 7, cf. 10).

types of parallelism, depending on meaning (i.e. semantic parallelism which can be synonymous—including numerical parallelism, antithetic or contrasting, alternating), syntax (grammatical; nominal and verbal; chiastic) and the lines (or parts of lines) comprising parallelism can have various degrees of separation (standard or near, internal, and distant) and can be grouped into bicola, tricola, etc. These types may or may not overlap. 2.3.2

Semantic parallelism

T h e r e are various sub-types of semantic parallelism. 2.3.2.1

Synonymous parallelism

This is the standard form of parallelism, 15 where line A and line Β say virtually the same thing: (A) They ply with gifts Lady Athirat of the Sea, tmgnn rbt atrt ym tg zyn qnyt ilm (B) they implore the Progenitrix of the gods (KTU 1.4 = RS 2. [008]+ iii 25-6) 2.3.2.2

Numerical parallelism

Since numbers have no synonyms, synonymous parallelism involving numbers takes the form of η || η + 1 (where η is an integer), as in hm tn dbhm ska b'l lit rkb crpt dbh bit wdbh dnt wdbh tdmm amht (KTU 1.4 iii 17-21)

For two sacrifices Baal hates, three, the Cloud-rider: a sacrifice of shame and a sacrifice of prostitution and a sacrifice of handmaidens' lechery

T h e r e are several other examples. 16 2.3.2.3

Antithetic parallelism

Although relatively rare, contrasting or antithetic parallelism does occur, 17 e.g.

P A R K E R 1974, P A R D E E 1988b, S E G E R T 1984, 109. See the bibliography in W A T S O N 1991b, 241, n. 2 and 242, n. 3, esp. 1973 and 1981 and L E E 1973. 17 W A T S O N 1986 = 1994b, 468-77. 15

16

AVISHUR

št špt lars špt 1šmm set a lip to the earth, a(nother) lip to the sky (KTU 1.23 = RS 2.002:61-2)18 and, in spite of the missing text: [A1]oud did Anat laugh, [g]/n tshq cnt wblb tqny [. . .] but in her heart she was hatching [a plot?] (KTU 1.17 = RS 2. [004] vi 41-2) 2.3.2.4

Alternating parallelism

More problematic is the existence of what appear to be non-parallel bicola (SEGERT 1983, 300). Such non-parallel lines can be explained in several ways. For example, a line such as hm it Ihm wtn wnlhm, 'If there is food, then give so that we may eat' appears to be prose, but in view of its counterpart, the lines are evidently in alternating parallelism (indicated by A / A ' and B/B'): hm [it 1 ]hm (A) If there is food, wtn wnlhm (B) then give so that we may eat hm it [yn] (A') If there is wine, \yi\tn. wnšt (B') then give, so that we may drink (KTU 1.23 = RS 2.002:70-2). 19 Elsewhere the overriding pattern of parallel couplets tended to generate non-parallel couplets. 2.3.3

Grammatical and syntactical parallelism

A complete match in grammatical terms is evident in šlhm ššqy ilm Give food, drink to the gods, sad kbd hmt wait on, honour them (KTU 1.17 = RS 2. [004] ν 19-20).21 T h e verbs in the parallel giving rise to types such etc. 22 O n qtl in Ugaritic Note also that a verbal

WO* WO

lines of a couplet can be in various 'tenses', qtl || qtl, qtl || yqtl, yqtl || qtl and yqtl || yqtl verse see above. clause can be in parallel with a noun clause:

18

See MERLO

19

adduces K T U 1 . 1 9 = R S 3 . 3 2 2 + iv 4 6 5 0 . V = 'verb'; Ο = 'object'. Parallel to K T U 1 . 1 7 = R S 2. [ 0 0 4 ] ν 2 9 - 3 0 , perhaps with hendiadys here.

20 21 22

DE M O O R

CASSUTO

WATSON

1996.

1993, 2 0 4

1971,

FENTON

1969,

1989 = 1994c, 240-9.

1973,

HELD

1962,

1965,

M.S.

SMITH

1994,

al tšrgn ybtltm Do not delude me Ο Virgin, dm Igzr srgk hhm for your delusion to a hero is sheer rubbish. (KTU 1.17 = RS 2. [004]+ vi 34-5) Perhaps chiastic parallelism belongs here, as in sb ksp Irqm turned had the silver into sheets hrs nsb llbnt the gold had turned into bricks (KTU 1.4 = RS 2. [008]+ vi 34-5) where the components of line 1 are switched around in line 2. 23 O n chiasmus see W e l c h ( 1 9 7 4 ) . 2 4 2.3.4

Parallelism based on degree of separation

Internal or half-line parallelism occurs when only one line is involved, for example: rgm cs 1hšt abn Word of tree and whisper of stone (KTU i . 3 = RS 2.[014]+ iii 22-3 and par.). Here, the two halves of the line are in parallel. 2 ' 1 Standard or near parallelism is none other than the couplet formed by adjacent lines (see above). In distant parallelism, which serves to bind together longer sections of text, there is a gap between the lines involved ( P a r d e e 1988c, 193-201, esp. 199-200). In general, the standard verse pattern comprises two sequential lines, as if parallelism were the accepted norm, although it is not always present. Quite often couplets can be formed from two (formulaic) monocola ( P a r k e r 1989, 23). 2.3.5

Ellipsis and ballast variant

Ellipsis (or gapping) is the absence of one or more elements (e.g. a verb) from a line which would be expected but is or are understood to be present. 26 For example, the verb tbl (+ suffix) is taken to be present in the second line of tblk grm mid ksp May the mountains bring you plenty of silver, «tblk» gb'm mhmd hrs the hills «bring you» choicest gold27 (KTU 1.4 (= RS "2. [008]+) ν 31-3) 23

S e e WATSON

24

Note also 'gender-matched parallelism' (on which cf. W A T S O N 1981a). See W A T S O N 1984b, 1985, 1988b = 1994b, 104 44. K O R P E L - DE M O O R 1998, 11. See S I V A N 1997, 215-6. The 'missing' element is indicated by « » (DE M O O R 1993, 200).

25 26 27

1 9 8 3 C , DE M O O R

1993,

193.

A ballast variant is the use of a longer expression (usually in the seco n d line) for its corresponding a n d evidently shorter equivalent (usually in the first line) a n d is related to ellipsis. ' T h e principle of ellipsis in poetry is the converse of (and goes h a n d in h a n d with) the principle of ballast variants.' (UT § 13.105). For example: wykn bnh bbt Sired be a son for him in the house, «wykn» 1rs bqrb hklh «Sired be» a scion within his palace (KTU 1.17 = RS 2. [004] i 25-6 || 42-3) H e r e , the ballast variant of bbt in the first colon is bqrb hklh in the second colon, which compensates for the the ellipsis of wykn in the second colon Similary, zbl b'l ars is the longer equivalent of aliyn bcl (matching the ellipsis of widc) in: wid' khy aliyn b'l kit zbl b'l ars

Then I shall know that Mightiest Baal lives, «I shall know» that the Prince, Lord of the Earth, exists. (KTU 1.6 iii 8-9; also K T U 1.17 i 36-7; K T U 1.18 iv 17-8). Ellipsis in the first colon is rare, occurring almost exclusively in 'staircase parallelism' (see § 5.2.4.2): ht ibk b'lm «tmhs» Now, your enemy, Ο Baal, «do strike», ht ibk tmhs Now, your enemy strike etc. (KTU 1.2 = RS 3.367 iv 8-9) As A l t e r (1985, 23-6). has explained, such ellipsis frees space in the second line for some elaboration of a parallel item in the first line a n d also isolates the e x p a n d e d topic for attention. 2 8

2.4

Verse

paragraphs

Lines of verse can occur in relative isolation (monocola) or clustered in sets varying f r o m two (couplets) to seven or more. T h e s e combinations of cola 29 are described here.

28

See also G R E E N S T E I N single word in K T U 1 . 2 = 1.14

=

2. [ 0 2 4 ] 1.3 = 29

RS

i RS

2. [ 0 0 3 ] +

21-4,

RS

iv 4 2 - 3 ;

SIVAN 3.367

K T U

i

YONA 20-1;

1.16 =

1983,

iv 302;

39-40. 1984,

108 §

71.32.

(1998,

K T U RS

and ellipsis of an expression in

2. [ 0 1 4 ] +

SEGERT

1983.

404-5)

1.3 =

3.325+ K T U

vi 1.2

RS

discuss ellipsis of a iv 5 ; K T U

2.[014]+

11-2; =

RS

K T U

1.22

3.367

iv

6;

=

RS K T U

2.4.1

Monocola

Although the standard strophic form is the couplet, single lines or monocola occur very frequently. T h e y appear as introductory monocola drawing attention to speaker and in this form are prose, e.g. šrrí Imtt hry 'Listen, Ο Lady Hurriya' ( K T U 1.16 = R S 3.325+ vi 16-7). W h e n expanded to a bicolon or tricolon (see section on expansion/contraction), they are verse. Several are speech-openers. It is sometimes difficult to determine whether a line of verse forms part of a longer unit (bicolon, tricolon, etc.) or is entirely separate, e.g. e

d Ihm sty ilm While they ate the gods drank. (KTU 1.4 = RS 2. [008]+ vi 55) Is this line a monocolon or an introduction to the next two lines? T h e function of the monocolon is to open and close sections of verse. T h e y can also mark a climax, as in sgrthn. abkrn The youngest of them I shall make the firstborn (KTU 1.15 = RS 3.343+ iii 16). Some monocola exhibit inner parallelism 30 and so conform to the prevailing feature of Ugaritic verse. An example is the standard formula balp Id rbt kmn

by the t h o u s a n d šiddu, the m y r i a d kumānu

(KTU 1.3 = RS 2. [014]+ iv 38 and many times elsewhere). 2.4.2

The tricolon

T r u e tricola, where all three lines are in parallelism (AAA"), are rare, e.g. I ys' alt tbtk Surely he will I yhpk ksa mlkk Surely he will I ytbr ht mtptk Surely he will ( Κ Τ υ Ί . 6 = RS 2. [009]+ vi 27-9

remove the prop of your seat, upset the throne of your kingship, smash the sceptre of your rule. (and par.).

Some comprise a monocolon followed by a bicolon (ABB'): She prepared a lamb in flour tcdb imr bphd hps ktr whss for the appetite of Kothar-and-Hasis 1brlt hyn dhršydm for the craving of Hayin, skilful with both hands. (KTU 1.17 = RS 2. [004] ν 22-5)

30

See

WATSON

1984b, 1985a, 1988b = 1994b, 104-44.

or, have the reverse formation (AA'B): ydd wqlsn He stood and insulted me, yqm wywptn He got up and spat on me, btk phr bn ilm right in the gathering of El's sons (KTU 1.4 = RS 2. [008]+ iii 12-4). A special type of tricolon, called 'staircase parallelism', where the initial line is interrupted by a vocative and then repeated in complete form, occurs several times, 31 e.g. irš hym laqht gzr Ask for life, Ο Hero Aqhat, irs hym watnk Ask for life I and I shall give it to you, blmt waslhk for non-death, and I shall grant (it) to you. (KTU 1.17 = RS 2. [004] vi 26-8) O n e function of tricola is to slow the pace of the narrative which then requires more attention ( P a r k e r 1 9 8 9 , 1 6 ) . 2.4.3

The quatrain or tetracolon

T h e r e are various forms of the four-line strophe, including AAA" A' " Her father set the beams of the scales, her mother, the trays of the scales. Her brothers settled the pointer, her sisters, the stones of the scales. (KTU 1.24 = RS 5.194:33-7) adnh yšt msb mznm umh kp mznm ihh yt'r mšrrm ahth labn mznm

AA'BB' tšmh ht atrt wbnh Let Athirat and her sons now rejoice, ilt wsbrt aryh the goddess and the throng of her kin, kmt aliyn b'l for dead is Mighty Baal, khlq zbl b'l ars expired has the Prince, the earthlord (KTU 1.6 = RS 2. [009]+ i 39-43) Some are formed when one couplet is inserted into another, as in dnilysb palth Daniel went round his blighted land, bsql yph bpalt he saw a stalk in the blighted land, bsql yph byglm he saw a stalk in the weeds, bsql yhbq wynsq he hugged and kissed the stalk. (KTU 1.19 = RS 3.322+ ii 12-5).32 31

See

281-309.

AVISHUR

1972,

GREENSTEIN

1974,

1977,

LOEWENSTAMM

1969

=

1980,

'Staircase parallelism' is more correctly analysed as apostrophe plus ep-

analepsis ( W A N S B R O U G H 1 9 8 2 ) . 32 As identified by D I E T R I C H

-

LORETZ

1973a (but cf. 1973b) and

DEL O I . M O

LETE

2.4.4

The pentacolon

An example is: bh p'nm ttt On her, her feet quake, e behind her, her loins burst, b dn ksl ttbr 'In pnh td' above, her face perspires, tgs pnt kslh the joints of her loins quiver, ans dt irh the muscles of her back. (KTU 1.4 = RS 2. [008]+ ii 16-20; || KTU 1.3 = RS 2. [014]+ iii 32ff. Κ KTU 1.19 = RS 3.322+ ii 44ff.) Again, much like the quatrain (see above) pentacola arise when a tricolon is inserted into a bicolon, as in Who sings and plays dyšr vuydmr bknr wtlb with lyre and flute, btp wmsltm with drum and cymbals, bmrqdm dsn with ivory castanets bhbr ktr tbm in the company of sweet(-voiced) Kothar. (KTU 1.108 = RS 24.252:3-5)33 2.4.5

Longer sequences

Hexacola are relatively rare in Ugaritic (cf. L o r e t z 1 9 8 9 ) . T h e following comprises an introductory line and a five-line speech: tm ydr krt t' itt There Keret the votary vowed a gift:34 atrt srm wlilt sdynm 'Oh, Athirat of Tyre, and goddess of Sidon, hm hiy bty iqh if I take Hurriy to my house, asr'b glmt hgry and bring the damsel into my dwelling, tnh kspm atn twice her mass in silver shall I give, wtltth hrsm and three times her mass in gold!' ( K T U Ί . 14 = RS 3.343+ iv 36-43) O t h e r hexacola may be K T U 1.3 = RS 2.[014]+ iii 3 - 8 a ; K T U 1.3 iv 4 8 - 5 3 (Ι) ν 3 9 - 4 4 || K T U 1.4 = R S 2. [008]+ i 12-8; iv

1981a, 34 and DE M O O R 1978a, 137, n. 41. Other examples in W A T S O N 1997a, 30-5 and 1997b. 33 For another example cf. D I E T R I C H - L O R E T Z 1 9 8 2 . 34 The meaning of iitt (= i + lit?) is disputed (cf. DLU, 1 and 60); here I follow W Y A T T 1998C, 2 0 0 - 1 ; see esp. nn. 115 and 117. Another possible translation is 'The gift of 'Atiratu of Tyre, the goddess of Sidon (is this):' ( P A R D E E 1997a, 336). For yet another translation cf. M A R G A L I T 1997.

50-7); K T U 1.12 = R S 2. [012] ii 5 8 - 6 1 ; K T U 1.17 = R S 2.[004] vi 43-5. 3 3 O t h e r sets are heptacola, K T U 1.6 = R S 2. [009]+ ν 11 - 9 and its near-parallel K T U 1.6 ii 31 -5,36 the eight-line sequence K T U 1.5 = R S 2.[022]+ i 14-22 and the nine-colon set K T U 1.107 = R S 24.251+ 3 8 - 4 4 .

2.5

Strophe and

stanza

A stanza is 'a fixed . . . or variable . . . grouping of lines that is organized into thematic, metrical, rhetorical, musical, or narrative sections' ( M e y e r s - S i m m s 1989, 288), though whether this definition would be recognised by the poets of Ugarit is a moot point. T h e only p o e m with an a p p a r e n t sequence of strophes or stanzas 37 is K T U 1.100 = R S 24.244 which is divided into 14 sections by ruled lines. O f these, sections 2 - 1 1 have the same n u m b e r of verse-lines (i.e. 10) due principally to almost wholesale repetition. T h e first section ( K T U 1.100:1—7) has 14 lines because the initial couplet has been expanded to a quatrain (see W a t s o n 1997a, 3 5 ) . 3 I ! T h e last two sections differ completely from all the others. T h u s , although these sections are actually marked off on the tablet, as P a r d e e (1978, 104) comments: 'this is unquestionably o w i n g . . . to the extra-poetic structure of the text and the comparable length of the sections is owing to the repetition within this structure' and 'any attempt to find strophes in Ugaritic poetry as a prosodie or poetic e l e m e n t . . . is d o o m e d to failure'. 3 9 Analysis is limited to identifying shorter or longer sets of verse-lines (couplets, tricola, quatrains, etc., as above), with no regularity or predictability. 40 Even so, this remains a useful exercise, and Ugaritic poetry can be segmented into sections based on content, sometimes marked off by features such as certain particles (apnk, etc.)

35

KORPEL -

DE M O O R

1986,

190-1

36

=

1988,

30-1.

See previous note. Cf. VVYATT 1998c, 1 3 5 n. 8 3 and 1 4 1 n. 108. 37 The terms 'strophe' and 'stanza' as denoting lines of verse grouped into sets are used almost interchangeably (cf. C U D D O N 1 9 9 2 , 9 1 5 - 6 , 9 2 1 ) , although stanza is more correct. On strophe, M E Y E R S - SIMMS 1 9 8 9 , 2 9 1 note: 'In contemporary usage, the term usually refers to any stanzaic unit containing irregular lines'. 38 Each 'stanza' has 10 lines (= 1 + 9); the first has 14 (4 + 1 + 2 + 7). 39 See also P A R D E E 1993a, 208, n. 2 (ruled lines do not mark off'strophes'). 40 A very interesting attempt at dividing the Story of Keret into (three) chapters, each further subdivided into 6 cantos of 5 canticles or strophes, has been made by S P R O N K 1 9 8 8 , although the incomplete form of the original text precludes cast iron conclusions. See also LICHTENSTEIN 1 9 6 6 . On a smaller scale cf. D I E T R I C H - L O R E T Z 1 9 7 8 , HUSSER

1995,

KLEVEN

1988.

a tricolon, peculiar syntax ( d e M o o r 1993, 197-200), speech introductions and the like, though these generally reinforce what has already been indicated by the meaning of the passage concerned (see, for example, the headings in d e l O l m o L e t e 1991a, 158-235, etc.).41

2.6

Repetition

Repetition takes on various forms. Repetition of sound takes the form of end-rhyme, alliteration, assonance and wordplay and is discussed below (§ 5.2.10). Repetition of words can occur at the beginning of a line (anaphoric), at line end (cataphoric), as immediate repetition of words (epizeuxis), and in the form of identical word pairs 42 (or repeated over several cola, as keywords). Sometimes whole lines (or sets of lines) are simply repeated, which in terms of strophe and stanza, results in envelope figure and refrains. Lasdy, complete passages recur, sometimes unchanged, sometimes in altered form (see § 5.2.7; P a r k e r 1989, 26-52). A selection of these types of repetition is considered here. 43 2.6.1

Repetition of words

For example, repetition of one word at line-initial, as in idk al ttn pnm Then you shall set off towards Mount Trgzz, 'm gr trgzz c m gr trmg towards Mount Thrmg, towards the two hillocks at the edge of the earth 'm tlm gsr ars (KTU 1.4 = RS 2. [008]+ viii 1-4) T h e function here is apparently to depict a long journey. Repetition of a word consecutively occurs rarely and most examples come from a single text ( K T U 1.23 = R S 2.002). O n e function may be to d e m a n d attention or convey urgency: J n k r n S r pÛ Hey! Watchman! Watchman! Open! (KTU 1.23 = RS 2.002:69-70) Similarly, y mt mt, O h , husband, husband!' ( K T U 1.23:40 and 46); y ad ad and ad ad, 'Father, father!'; urn urn, 'Mother, mother!'. M o r e

41

For an attempt along these lines cf. S A U R E N - K E S T E M O N T 1 9 7 1 , although their scheme was much too rigid. 42 See especially B O R N E M A N N 1 9 7 0 . 4:í On repetition in Ugaritic see Z U R R O 1 9 8 7 and H E N S - P I A Z Z A 1 9 9 2 .

striking is the repetition of six consecutive lines beginning tld pgt. . . 'She shall bear a girl. . ( K T U 1.15 = RS 3.343+ iii 7-12). 44 2.6.2

Repetition of lines

ytlk llbnn ufsh They went to Lebanon and its trees, 1šryn mhmd arzh to Siryon (and) its choice cedars; hn llbnn w'sh Yes, Lebanon and its trees, 1šîyn mhmd arzh to Siryon (and) its choice cedars. (KTU 1.4 = RS 2. [008]+ vi 18-21) Again, the purpose of the repetition, perhaps, is to depict a long journey. 2.6.3

Repetition of passages

Ugaritic narrative poems are constructed using passages which are repeated, sometimes verbatim, sometimes with slight variations. 43 Sets that recur several times include the list setting out the six duties of the model son ( K T U 1.17 = R S 2. [004] i 2 6 - 3 3 and par.; H u s s e r 1995); the 'gift-list' in the Keret Story ( K T U 1.14 = RS 2.[003] + iii 2 2 - 5 and par.; H e n s - P i a z z a 1992) and the 'peace-offering ritual' ( K T U 1.3 = R S 2.[014]+ iii 14-7). Typically, a passage comprising a ' c o m m a n d ' is then repeated for the 'performance' or fulfilment of that c o m m a n d , e.g. K T U 1.14 ii 6—iii 19 and K T U 1.14 iii 52-iv 31, or else an invitation ( K T U 1.22 = R S 2. [024] iii 2~4) and its acceptance (lines 5 - 8 ; see d e l O l m o L e t e 1981a, 58-60) It is significant, perhaps, that segments of verse (couplets, etc.) could occur in different sequences, for example, in K T U 1.14 ν 4 0 - 2 (restored) || vi 10-2, the couplets A

wng mlk Ibty rhq krt Ihzry al tsr udm rbt wudm trrt udm ytnt il wusn ab adm

B C

And depart, Ο king, from my house, go far, Keret, from my dwelling! Do not besiege Greater Udum, or Lesser Udum; Udum is a present of El and a gift of the father of Man

come in the sequences ABC ( K T U 1.14 iii 27-32), BCA ( K T U 1.14 ν 40-5) and ACB ( K T U 1.14 vi 10-5) and in addition, the word 44

According to P A R D E E 1997a, 338, n. 51 this may be a list of the daughters' names, now lost. 45 Cf. DEL O L M O L E T E 1991a, 58-62, P A R K E R 1989, 26-52.

pair ng II rhq is inverted the last two texts. Unless due to dictation error, this indicates that verse was composed in formulaic passages, perhaps orally.

2.7

Expansion and contraction

In Ugaritic the poets were free to expand single lines to bicola and in turn form tricola from bicola. T h e process could also be reversed, with longer strophes becoming shorter. It is certainly the case that a line can be expanded to a couplet, as in: šrrí rrí latiyn b'l Listen, please, Ο Mightiest Baal! (KTU 1.4 = RS 2. [008]+ vi 4) which becomes šm' laliyn b' I Listen, Ο Mightiest Baal, bn Irkb 'rpt understand, Ο Cloud-rider! (KTU 1.4 ν 59-60)46 A single line can also be extended to a tricolon. For instance, Pughatu's core epithet is tkmt mym '(she who) shoulders water', and it can occur alone ( K T U 1.19 = R S 3.322+ iv 28); it can also be expanded by two further epithets: hspt Is'r tl, 'scooping dew from the barley' and yd't hlk kbkbm, 'knowing the course of the stars' (cf. M a r g a l i t 1989a, 364-5). T h e prose formula wrgm Ikrt t' thm pbl mlk,47 'And say to Krt the Noble, "Message of King P b l " ' ( K T U 1.14 = RS 2.[003]+ ν 32-3), which derives from everyday usage, is expanded by applying parallelism to each half to form two bicola: wrgm Ibn ilm mt And say to divine Mot, tny lydd il gzr Repeat to El's beloved, the hero thm aliyn b'l "Message of Mightiest Baal, hwt aliy qrdm The word of the Mightiest warrior". (KTU 1.4 = RS 2. [014]+ viii 29-35)48 However, it is sometimes difficult to determine whether a poet has intentionally added or omitted a line (or lines) or whether these are accepted variants. For example, in

46 47 48

See L O E W E N S T A M M 1980a = 1992, 230-9, Restored from similar formulae. L O E W E N S T A M M 1980b, 256-61.

DE M O O R

1978b, 1980.

sh hrn bbhtk Call a caravan into your house, c dbt bqrb hklk merchandise into the midst of your palace. tblk grm mid ksp The rocks will bring you much silver, gb'm mhmd hrs the hills attractive gold. yblk udr ilqsm let the quarries bring you choice gems. (KTU 1.4 = RS 2. [008]+ ν 15-6 and KTU 1.4 ν 38-40) T h e last line is present in the two parallel passages but has been omitted in K T U 1.4 = RS 2.[008]+ ν 31-3. Is this a mistake or a deliberate contraction? Since it is of little significance, it was probably left out unintentionally. However, in the two club-naming passages, it is only when the c o m m a n d line yprsh ym wyql lars May Yam crumple and fall to earth (KTU 1.2 = RS 3.367 iv 22-3) is given to the weapon that it is effective and achieves the intended result (line 26); this line does not appear either in the previous n a m ing or in the unsuccessful previous attempt, which indicates its omission there to be intentional. Each case has to be judged carefully because the copyists 49 did occasionally leave out lines by mistake but generally speaking the poets could expand or contract as they saw fit.

2.8

Word pairs

It has been noticed that there seems to be a large set of word pairs which recur in Ugaritic verse and that many have equivalents in other verse traditions of the ancient N e a r East. 50 T h e problem then arises: Did the poets have a sort of 'dictionary of word pairs' on which they drew to compose their verse, which was probably originally oral, 51 or were these pairs the side-effects of their use of parallelism and of (oral) formulaic language?'' 2 As is evident from the following, the matter is unresolved. In such word pairs, the Ά - w o r d ' is usually c o m m o n e r than the 'B-word', e.g. klb || inr, 'dog' || 'puppy' ( K T U 1.16 = RS 3.325+ i 12; K T U 1.114 = RS 24.258:13) and any particular Α-word may

49

50

See

HORWITZ

1974,

1977,

1979.

Cf. AVISHUR 1984, D A H O O D 1972, 1975, 1981, DEL O I . M O L E T E 1984a, W A I T E R S 1976. However, see the cautionary remarks of VAN DER L U G T DE M O O R 1974. 51 See P A R D E E 1988a, 160, D I E T R I C H - L O R E T Z 1980b. 52 Another possibility is that they arose through word association (BERLIN 1 9 8 3 ) , but see below.

be paired with several different B-words (e.g. ib, 'enemy' || srt, 'adversary' or qm 'one who rises against' or šnu, 'hater'). 5 3 Often a word pair is related to a theme, e.g. ars || šd, 'earth' || 'field', and is connected with fertility as in n'm lars mtr b'l Pleasant to the earth is Baal's rain, wild mtr 'ly and to the field the rain of the Most High. (KTU 1.16 = RS 3.325+ iii 5-6, 7-8) as also in K T U 1.3 = R S 2.[014]+ iii 16-7 and par., K T U R S 2. [022]+ ν 18-9; K T U 1.6 = R S 2. [009]+ ii 16-7, 19-20. word pairs are bound to a formula or formulaic expression fewer that are non-formulaic and m a n y of these are repeated as Aitken has shown for the Aqhat Narrative. H e comments:

1.5 = Most with pairs,

This calls into question the notion of the word pair as a compositional device, functioning independently of the formulas and formulaic expressions or of a thematic or/and formulaic context within the tradition. While there are indications that a narrator could 'learn' a word pair as a word pair in one context and use it quite independendy in another, this was the exception rather than the rule. In the overwhelming majority of examples, the repository of word pairs is the formulas and themes of the narrative tradition, and their appearance in the narrative is a spontaneous reflex of the formulary and thematic habits of that tradition, and not of the production and manipulation of word pairs. Neither spontaneous word association, nor the 'learning' and subsequent deployment of 'generally useful' word pairs has played a significant part in the generation of recurrent word pairs within the narrative 54 ( A i t k e n 1989b, 38). Very rarely, word pairs are reversed, generally to denote some form of reversal of events. C o m p a r e al tšt urbt bbhtm You shall not install a window in the mansion, hin bqrb hklm an aperture within the palace. (KTU 1.4 = RS 2.[008]+ ν 64-5 || 61-2; vi 5-6, 8-9) with the reversal of urbt || hin, which matches the sense, in ypth hin bbhtm Let an aperture be opened in my mansion, urbt bqrb hklm a window within the palace. (KTU 1.4 vii 17-9 II 25-7).55

53 54 55

Cf. AVISHUR 1984, 344-9 and S E G E R T 1984, 108. See also K U G E L 1981, 27-40. See W A T S O N 1981b = 1994b, 262-6.

W o r d pairs, then, were an integral part of the poet's composing technique and the very traditional character of versification in Ugaritic resulted in most pairs remaining 'fixed'.

2.9 2.9.1

Formulae and formulaic patterns

Formulae

T h e Ugaritic poems were composed using traditional formulae, sometimes with modifications or complete transformations. 5 6 A c o m m o n type of formula is the one which introduces speech and it can take many forms. Typical examples are wy'n Itpn il dpid Answer did Lutpan, kindly god (KTU 1.1 = RS 3.361 iv 13) where the slot Itpn il dpid could be filled by the appropriate name or epithet, and yšu gh wysh

He raised his voice and exclaimed

which can be altered to suit gender and n u m b e r where necessary. T h e r e is a whole range of such formulaic introductions. 5 . It was also accepted convention that such introductions could occasionally be omitted, either because they were implicit or for dramatic effect.'' 8 2.9.2

Formulaic patterns

T h e formulaic patterns to portray the passing of time are of two types. T h e first has as its core two expressions, one denoting time and the other an activity, as in the single line: hn ym ysq yn Behold, for a day they pour wine (KTU 1.22 = RS 2.[024] i 17) and either or both expressions can be extended to cover more time (up to a sequence of seven days) or further activity (cf. K T U 1.16 = R S 3.325+ vi 21-4; K T U 1.17 = R S 2. [004] ii 30-40). In the second pattern, the time expression denotes a period of days, months or years and activity is continuous, e.g. 56

See

57

WATSON

1 9 8 1 , also M E I E R 1 9 8 6 . 1983a = 1994b, 414- 24; 1992a. Some of them had elaborate preambles; cf. W A T S O N 1994d. 5B W A T S O N 1990 = 1994b, 425-30. WHITAKER

ym ymm yt'qn A day, two days elapse, lymm lyrhm from days to months, rhm 'nt tngth Maid Anat (still) looks for him. (KTU 1.6 = RS 5.180+ ii 26-7). O n c e again these patterns can be extended. They are used for themes such as making a journey ( K T U 1.14 iv 31-5), preparing a banquet ( K T U 1.22 = R S 2.[024] i 21-6) or to depict a ruling monarch ( K T U 1.6 ν 5 - 1 0 ; K T U 1.16 vi 21-4). T h e y also have other functions within the wider framework of the narratives. 59

2.10

Sound patterns

In spite of the lack of vocalization, some idea of the patterns of sound exploited by the poets can be gained from the texts and a few examples are provided here. 2.10.1

Alliteration

Particular words and forms were often chosen for reasons of alliteration although this feature should not be exaggerated. 6 0 First comes the simple word-initial type, as in ap ank ahwy aqht gzr I too shall revive Hero Aqhat (KTU 1.17 = RS 2. [004] vi 32-3). Quite often consecutive lines of verse began with the same letter which may indicate they were intended to be read as well as performed. An example, with line-initial i- twice, is in b'l bbhth Baal is not in his mansion, il hd bqrb hklh the god Hadd, in his palace. (KTU 1.10 = RS 3.362+ ii 4-5) 1980 = 1994c, 431-4). In addition, multiple alliteration seems to have played a part in longer sections of verse as has been shown by P a r d e e ( 1 9 8 8 c ) .

(Watson

5N A I T K E N 1987. See also LOEWENSTAMM 1965 = 1980, 192-209, P A R K E R 1989,46-52 ('Repetition with a numerical framework'), T R O P P E R 1995 and YVHITAKER 1969. On theme cf. A I T K E N 1990, 1991a, H I L L E R S 1973, L L O Y D 1990 and on motif, W A T S O N 1984a. 60 M A R G A L I T 1975, 310-3, 1979, 1980a.

2.10.2

Wordplay

Undoubtedly, plays on words formed part of Ugaritic verse but because our knowledge of the language is limited, many puns escape us.1'1 A few examples can be given, however. ytt nhšm mhrk I hereby give you a snake as your bridal gift bn bin itnnk a serpent's son as your present. (KTU 1.100 = RS 24.244:75-6) Here, the use of the rare word itnn is apposite as it evokes tnn, the mythical serpent monster, and the wordplay between itnn ('present') and nhš || bn btn, both denoting 'serpent' is transparent. Another example is 'ζ ym lymk Yam was strong, he did not sink. (KTU 1.2 = RS 3.367 iv 17) T h e play between Ym (probably pronounced yammu) and the verb form ymk (yamukku) is self-evident. Another verb such as yql, 'he fell', would have been less effective. 62

2.11

Figurative language

Ugaritic poetry is relatively rich in figurative language and includes similes, metaphors and metaphorical expressions, personification and apostrophe 6 5 as well as imagery. Occasionally it is difficult to know where to draw the line between mythological language (as in 'the rain of the Cloud-Rider', which refers to Baal) and extravagant expression (e.g. 'rain with which the stars anointed her'). Here, examples are provided under appropriate headings, though it is not always easy to make clear-cut distinctions. 64 2.11.1

Simile

In similes, the particle k (or km), 'like', is used, as in tit kyn udm't She drinks tears like wine (KTU 1.6 = RS 2. [009]+ i 10)

61 62 63 64

For a complete survey see W A T S O N 1999. 'Janus parallelism' is another form of wordplay; cf. See, e.g. W A T S O N 1984c = 1994b, 460-4. See, in general, K O R P E L 1 9 9 0 .

NOEGEI.

1995.

though it can be omitted through ellipsis. Similes rarely come singly, as they are mostly in sets of two, for example: klbs km 1pš dm ahh

He was clothed like a mantle in his brothers' blood, km all dm aryh like a cape, in his kinsmen's blood. (KTU 1.12 = RS 2.[012] ii 46-7) 65 or in sets of three: thth kkdrt rc[/] Beneath her like balls were hea[ds], lh kirbym kp above her like locusts were palms, kqsm grmn kp mhr like grasshoppers in a swarm, warrior palms. (KTU 1.3 = RS 2. [014]+ ii 9-11) c

and once, a set of four: K T U 1.169 = R I H 7 8 / 2 0 : 3 - 4 . Similes can be drawn out at length (as extended similes), indicative, perhaps of an oral, improvising style, e.g. km tdd cnt sd Just as Anat hurries to the chase, (and) sets the birds of the skies soaring away, tštr cpt smm tbl} alpm ap sin (so) they butchered oxen as well as sheep, šql trm wmri ilm felled bulls and the fattest of rams, etc. (KTU 1.22 = RS 2. [024] i 10-4).66 Only one cumulative simile has been identified: klb arh I'glh Like a cow's heart for its calf, klb tat limrh like a ewe's heart for its lamb, so is Anat's heart after Baal. km lb cnt atr b'l (KTU 1.6 = RS 2.[009] ii 6-9) 2.11.2

Metaphor

M e t a p h o r is little used in Ugaritic verse. For example, fertility is expressed metaphorically as šmm šmn tmtm Let the heavens rain oil, nhlm tlk nbtm the wadis run with honey. (KTU 1.6 = RS 2. [009]+ iii 6-7; M a n y metaphorical expressions are used such as dm csm 'blood of trees', for grape juice and tl šmm 'dew of heaven' for rain. In some metaphors the mythological meaning may be muted, e.g. wytn qlh b'rpt, 'and may he (Baal) give his voice in the clouds' ( K T U 1.4 = RS 2. [008]+ ν 8). 65 66

For another example cf. W A T S O N 1991a. See also I R W I N 1983, but cf. P A R D E E 1988a, 127-9.

However, 'the lack of comparable contexts in Ugaritic prose makes the recognition and evaluation of these and other poetic figures difficult and uncertain' ( S e g e r t 1 9 7 9 , 7 3 3 ) . 2.11.3

Imagery

Besides expressions such qr cnk 'the well of your eyes' ( K T U 1.16 = R S 3.325+ i 26) to denote tears, weapons are said to fly off and strike like birds of prey ( K T U 1.2 = RS 3.367 iv 13-7), a tree stands for descendants ( K T U 1.19 = R S 3.322+ iii 53-4; cf. P a r d e e 1997a, 354, n. 121), Mot's domain is described as being a town (qrt) called 'Miry' {hmiy), in a land called 'Filth or M u d ' {fih: K T U 1.4 = R S 2. [008]+ viii 12-4 || K T U 1.5 = R S 2. [022]+ ii 15-6), a mountain weeps (KTU 1.16 = RS 3.325+ i 7) and so on. Occasionally the imagery is obscure due to our lack of knowledge (e.g. K T U 1.16 = R S 3.325+ vi 5 7 - 8 ; K T U 1.17 = R S 2.[004] vi 36-7). Hyperbole also occurs ( W a t s o n 1979 = 1994b, 452-60). T h e use of abstract for concrete is extremely rare, perhaps only srt, 'adversary' (see above) and t'dt 'legation' 67 ( K T U 1.2 = R S 3.367 i 22; cf. G r a y 1965, 22, n. 6).68

2.12

An example

In order to illustrate some of the poetic structures, rhetorical techniques and expressive language described above, a passage is set out here in tabular form, with analysis, comments and discussion. 2.12.1

A king is chosen ( K T U 1.6 = R S 2.[009]+ i 43-65)

description

text

line translation

introd. to sp.

"gm ysh il 44 Irbt alrt ym

01 02

Aloud shouted El to Lady A.Y.

non-par. bicolon

šrrí ''Irbt alrt ym

03

"Listen, Ο Lady A.Y.

monocolon

Give (me) one of your mlk sons so I can make him king!"

monocolon

speech: El

In 46ahd bbnk amlkn 04

67 68

As aptly translated by M . S . S M I T H 1994a, 2 6 6 . For an example of imagery cf. W A T S O N 1992b.

keyword ||m

(table cont.) description

text

line translation

introd. to sp.

[7

wt'n rbt atrt ym

05

And Lady A.Y. replied

keyword ||m monocolon

speech: Athiratu

w

bl nmlk yd' yIhn

06

'Shall we not make a person of intellect king?'

introd. to sp.

n

uy'n Itpn il dpid

07

And El, merciful god, replied

monocolon

dq anm lyrz

08

tricolon

h[

09

'One feeble of strength cannot run (or) like B. release the javelin like the son of Dagan '

11

And Lady A.Y. replied

monocolon

'Shall we not make Awesome Athtar king? Let Awesome Athtar be king!'

speech: El

'm b'l ly'db mrh

b2

'm bn dgn ktmsrn 10

introd. to sp.

M

speech: Athiratu

51

bit nmlk 'ttr 'τζ

12

5i

ymlk 'ttr 'rz

13

narrative

w'n rbt atrt ym

mlk

mlk

monocolon

bicolon

mlk

56

apnk 'ttr 'τζ j7 bsrrt spn

14 Then Awesome Athtar 15 went up to the heights of Sapūnu, ™ylb Ikht aliyn 59b'l 16 he sat on the throne of Mighty Baal;

tricolon

p'nh Itmgyn a)hdm

his feet did not reach the footstool, his head did not reach its extremity.

bicolon

monocolon

57

rilh lymgy

61

aps/i

17 18

introd. to sp.

uy'n 'ttr 'τζ

19

And Awesome Athtar said:

speech: Athtar

,a

20

Ί shall not be king in the heights of Sapūnu'

narrative

lamlk bsrrt spn

m

yrd 'ttr 'rz

yrd "'Mt aliyn b'l K

'uymlk bars il klh

Awesome Athtar came down, 22 came down from the throne of Mighty Baal, 23 and was king in all the vast earth.

mlk

21

monocolon tricolon

mlk

2.12.2

Comments

T h e following selective remarks are set out in the sequence of topics adopted above, and there is a brief overall evaluation (on the whole passage see now X e l l a 1996a). General T h e passage is a combination of speech and narrative, linked by the formulaic introductions to speech which are all monocola (except for 0 1 - 0 2 which is a non-parallel bicolon). Stichometiy In 13 of the 23 lines, the verse-line corresponds to the line on the tablet (i.e. 05-16, 0 9 - 1 5 and 20-23). Lines 14-18 could be analysed in several other ways (e.g. monocolon + two bicola) or one could argue that they form a pentacolon. fiy O n the other hand, although amlkn (04) could be a single-word separate line, it would seem to belong to tn ahd bahk amlkn, like the corresponding expression tn ahd bahk ispa (see below). Similarly, some scholars argue that ktmsm (10) comprises a separate line, 70 although this view has not been accepted here. Others argue that 0 8 - 1 0 comprises a couplet.' 1 Language Difficulties are caused by the lack of a clear translation, especially of 0 8 - 1 0 (survey: d e l O l m o L e t e 1984, 77). T h e expression bars il klh has been translated in various ways. 72 T h e epithet ydc ylhn, perhaps 'he knows, he understands', may be a hendiadys and occurs only here. Also unique is aps, 'edge' and the form nmlk is found only here (twice);73 on bit, see below. T h e verb form amlk is used elsewhere only in K T U 1.16 = R S 3.325+ vi 37, 53, also in the context of a failed would-be usurper to the throne. T h e formula tn ahd bahk, 'Give (me) one of your brothers (so that I may . . .)' occurs later in this text ( K T U 1.6 = R S 2. [009]+ ν 19-20, where Mot is speaking to Baal), but nowhere else.

69 70 71

According to M A R G A L I T (1980b, 143), lines 01 13 are all 'monostichs'. See e.g. G O O D 1994, P A R D E E 1997a, 269, n. 246. K O R P F . I . - DE M O O R 1 9 8 6 , 1 8 0 = 1 9 8 8 , 12. For a different approach see

EMERTON 72

1965, 441

2.

E.g. 'la terre dont il est maître', by X E L L A 1 9 9 6 , 3 8 7 , discussion 3 9 0 - 1 . 73 T h e first occurrence (06) could be an N-stem, but this verb form is rare and here it is causative, probably D (SIVAN 1 9 9 7 , 1 1 6 . 1 3 5 ) .

Parallelism T h e best examples are 12-13 and 17-18. In many 'couplets' it is absent (eg. 01-02); 'staircase parallelism' is present in lines 21-23. Antithetic, distant parallelism is effected by lines 16 and 22 (_ytb Ikht aliyn b'l—yrd Ikht aliyn b'l) and lines 20 and 23 (lamlk bsrrt spn ymlk bars klh). Strophes and stanzas T h e strophic sequence of monocola, bicola and tricola is unpredictable, 74 and inasmuch as there is any demarcation into stanzas or sections, these seem to be 0 1 - 0 6 , 0 7 - 1 3 , 14-18 and 19-23. T h e only clear division is signalled by the combination of apnk and a tricolon (14-17). Repetition T h e title rbt atrt ym, 'Lady Athirat Yam', occurs 4 times in 23 lines, c ttr crz, 'Awesome Athtar', occurs 5 times, but mlk occurs 6 times. In view of the content, which primarily concerns kingship, the keyword here, then, is mlk, 'to rule, be king'. Crucially, it is missing from the first narrative section (14-18), where Athtar attempts to sit on Baal's throne and fails miserably (mlk - 0). However, in the second narrative section, where he actually rules over the earth, it does occur, only once but to effect. It is interesting that when bl is repeated it takes on a form with enclitic -t which is found only here ( A a r t u n 1974, 27).75 Sound patterns Consecutive lines beginning with the same letter are 0 9 - 1 0 ( f ), 15-16 (y) and 2 1 - 2 2 (y again). Alliteration also occurs in the n a m e + epithet sequences, notably 'ttr 'rz, and perhaps in the obscure yd' ylhn; see also srrt spn. T h e r e is probably wordplay between 'rz and yrzWord pairs These include p'n || ris (found only here) and the repeated pairs / || I, mlk Κ mlk, mgy || mgy and 'm || 'm. Figurative language Apart from the comparisons in the difficult tricolon where the wouldbe king is compared to Baal, there is virtually no figurative language, but Athtar does perform a symbolic act (descent from the throne). ' 4 It is possible that lines 2 0 - 2 3 form a split couplet with inserted bicolon; cf. 1997a, 3 1 . 75 For a similar usage in respect of enclitic -m see W A T S O N 1992c, 238-9.

WATSON

2.12.3

Discussion

T h e passage demonstrates the difficulties caused by uncertain stichometry and obscure language, especially in 06 and 08-10. Also, it is not always clear how lines were grouped together to form couplets or higher units. However, overall the structure is quite evident and there are no textual problems to complicate matters. T h e mix of speech and narrative is fairly typical and the use of repetition and stock formulae is offset by the presence of rare vocabulary and grammatical forms. T h e interpretation of the passage is quite clear (no suitable successor to Baal has been found) 76 and is reinforced in particular by the keyword (mlk), while of course the passage forms part of a larger whole (the Baal Cycle).

2.13

The character of Ugaritic poetry

M u c h of the above is concerned with techniques and rhetorical devices, 77 but this does not mean that the aesthetic dimension is absent. Even though we do not know who the poets of Ugarit were nor for w h o m they composed their verse, it is evident that they were masters of the language and well able to manipulate it in a variety of ways, attracting and holding the attention of their listeners or readers. 7 " T o do justice to such aspects would require detailed study of each composition for which there is no space here. 7u Enough has been provided, one hopes, to whet the appetite for closer reading. As in some other ancient N e a r Eastern verse traditions, perhaps the most salient feature of Ugaritic poetry is its unpredictability, a feature which runs right across the board from prosodie structure to complete compositions. This means in effect that, with a few exceptions, length of line, whether two lines will be parallel or not and if so, the type of parallelism adopted, whether a speech will or will not have an introduction and whether the introduction will comprise one line or several, sequences of bicola, tricola and so on, how

76

For a nuanccd approach cf. W Y A T T 1 9 9 8 C , 1 3 2 , n. 7 5 . Though the survey has not been exhaustive, e.g. rhetorical questions have not been considered (cf. H E L D 1969). 78 See, for instance, W A T S O N 1988a = 1994b, 434-45 on delaying devices in Ugaritic verse. 7 '' For examples of close analysis see H E T T E M A 1989 90, M A R G A L I T 1989a, T S E V A T 77

1986,

VERREET

1987.

many lines there are in a 'strophe' or 'stanza', etc., are all completely variable. Generally speaking, studies of Ugaritic poetry pay considerable attention to similarities with other verse traditions, particularly ancient Hebrew (e.g. A v i s h u r 1 9 9 4 , P a r k e r 1 9 8 9 ) . While this is useful, especially when features from other traditions can throw light on Ugaritic, it is also of interest to determine in what respects Ugaritic differs from such verse traditions. T h e question to be asked is: what is unique or special to Ugaritic verse? According to S e g e r t ( 1 9 7 9 , 7 3 1 ) , 'The most prominent feature of Ugaritic poetry is its parallelistic structure. It can be said that no other literature of the ancient N e a r East, Semitic or non-Semitic, exhibits such consistent application of this structure'. O t h e r features which could be mentioned include the use of verse for narrative (which though rare or virtually unknown in Hebrew or Phoenician, is common in Mesopotamian tradition) and a general lack of hymns (though this could be due to chance). In addition, there is a tendency to alter repeated (parallel) passages slightly. Special verse patterns such as 'staircase parallelism', and its combination with anadiplosis (notably in K T U 1.10 = RS 3.362+), the use of word pairs in fixed sequences, with variation and inversion rare, the use of chiasmus to show two or more individuals acting as one (e.g. K T U 1.17 = R S 2. [004] ν 10-1) and the split couplet 80 all seem to be peculiar to Ugaritic. Also, unlike Hebrew and Akkadian verse, there is no clustering of similes or of lines with inner parallelism. 81 However, descriptions of actions as preludes to speech and lengthy introductions to speech are c o m m o n e r in Ugaritic than in Hebrew verse, whereas abrupt or unmarked speech is rare. 82 'If there is a specific contribution made by Ugaritic to a poetic tradition . . . [it] seems to lie in injecting originality into a well-worn, stereotyped body of versification . . . T h e single copies of the Ugaritic tablets that have been found are, perhaps, the work of a single school which re-worked stock and static verse and made it sparkle with new life'. 83

80

WATSON

81

WATSON

82

WATSON

83

WATSON

1997a, 1985b 1994d, 1983b

1997b. = 1994b, 157. cf. W A T S O N 1990 = 1994b, 425-30. = 1994b, 68.

ΓΗ Ε U G A R I T I C L I T E R A R Y T E X T S

1

T h e

Mythological

John

1.1 1.1.1

C.L.

Texts

Gibson

The Baal cycle

The tablets

T h e Baal mythological cycle is the largest text from ancicnt Ugarit, taking up six tablets ( K T U 1.1 6: 1.1 = RS 3.361, 1.2 iii = R S 3.346, 1.2 i—iv = R S 3.367, 1.3 = RS 2.(014]+, 1.4 = RS 2.[008]+, 1.5 = RS 2. [022]+, 1.6 = RS 2.[009]+) and comprising in its surviving portions around fifty per cent of the contents. Each tablet is written on both sides and usually has six columns, although 1.2 has only four and 1.4 has eight. K T U 1.6 has the heading 'of Baal' and, although the other tablets have their tops missing and it remains a conjectural point, this was probably written on them all and is the title of the whole composition. K T U 1.4 has a note on the edge, ' T h e scribe is Ilimilku, the master, Niqmad being king of Ugarit', while K T U 1.6 has at the close a full colophon, giving the information that Ilimilku was not only the scribe but a student or assistant of a high religious officer of King Niqmad, probably the second of that name. So the composition was officially approved. At a banquet scene in K T U 1.3 i there is talk of chanting and singing and of a minstrel (n'm). This suggests a possible Sitz im I^ben in the royal palace, though equally possible is some festival in the temple of Baal, in the library of which the tablets were inscribed. No doubt the king himself was often present. For a translation and full bibliography consult W y a t t , 1998c, 33 146. 1.1.2

The contents

T h e plot or story-line centres around a contest (under the overall supervision of El) for the kingship over gods and men, and divides itself usefully into three parts, cach consisting roughly of two tablets.

1.1.2.1

Baal and Y a m - N a h a r ( K T U 1.1-2)

T h e first part leads up to Baal's defeat of his rival Prince Yam (sea), also called J u d g e N a h a r (river), with the help of two maces constructed for him by the divine craftsman, Kothar-and-Hasis, a story told in the last column of 1.2 (iv). T h e evidence of what happened prior to the battle is, however, not at all so clear, since 1.1 is very imperfectly preserved. But near the beginning (1.1 ii) a message is sent from the supreme god, El, to Anat, Baal's sister, calling on her to perform what seems to be some kind of ritual, which involved setting an offering of war in the earth (perhaps the burying of weapons) and, following this, offerings of love and peace. This ritual may originally have had to do with ceremonies for the cessation of hostilities; but it was not, as far as we can tell, performed by Anat, so it is more likely that the passage uses ritual language to express a wish on El's part that the notorious goddess of war and love should a b a n d o n her more savage ways and, in particular, not employ them in her brother's support. It is an important indication of the way El desires things to work out. Thereafter in 1.1 iii Kotharand-Hasis is summoned to El's distant abode, perhaps as an ally of Baal, to be given a similar warning; for clearly Y a m - N a h a r is at this point being favoured by El, since in 1.1 iv he accords him royal status in a kind of ceremony of coronation. By 1.2 i, however, Y a m - N a h a r is worried; for he sends an embassy to the divine assembly, complaining that Baal has been reviling him and demanding his surrender. El appears to sympathize but Baal, who is present, objects strongly and sends an angry reply back to Yam-Nahar. In 1.2 iii, a large fragment (perhaps out of place in its present position), El instructs Kothar-and-Hasis to build a palace for Y a m - N a h a r , and the claims of a minor rival, Athtar, are dismissed. W h e n , after a sizeable gap, the text resumes, the battle between Baal and Y a m - N a h a r is joined, Baal with the encouragement of Kotharand-Hasis wins and, though it is not according to his plans, the supreme god has presumably to accept that Baal is now king. 1.1.2.2

T h e Palace of Baal ( K T U 1.3-4)

These tablets concern the building of a palace for Baal, from which he may exercise his newly achieved kingly power. After a victory banquet (1.3 i), the goddess Anat resumes her warlike ways and

mercilessly slaughters the inhabitants of two u n n a m e d towns, thereafter repeating the process with a n u m b e r of soldiers and guests in her own mansion (1.3 ii). At the beginning of the next column (1.3 iii) she sings of her affection for her brother, but Baal, perhaps like El before him, perturbed by her behaviour, sends messengers to her requesting her to perform the same 'ritual' for peace as El had previously asked for, but also tells her to visit him and help him search for the secret of the lightning. She has to be reassured that YamN a h a r and his cohorts, w h o m she had worsted in the past, had been finally dealt with and were no longer a threat to Baal (1.3 iv); but she decides to call on him to find out for herself what is really worrying him. It transpires that what Baal has set his mind upon is to have a palace like other gods, and she herself goes to El's abode to press Baal's suit, which she does in too threatening a m a n n e r and apparendy has it turned down (1.3 v). A new plan has to be concocted, the working out of which takes up the last column of 1.3, all of 1.4 i-iv and half of 1.4 v. This involves the preparation of gifts for El's consort, Athirat, and the enlisting of her intercession with the supreme deity. T h o u g h we know from elsewhere that she is no friend of Baal's, she persuades him to let Baal have his palace. Anat takes the good news to Baal, and he immediately summons Kothar-and-Hasis to build a palace for him on his sacred mountain Saphon (1.4 ν 27ff.). At the end of of colu m n ν the divine craftsman suggests that it should have windows in it, but at the beginning of column vi Baal refuses to entertain the idea lest, it seems, his old enemy Y a m - N a h a r may gain entrance and again wreak havoc on earth. The house is soon finished and a celebratory feast held (1.4 vi). Column vii tells how Baal then marches through the surrounding territory, annexing a large n u m b e r of cities and towns and thereby forming an empire for himself. Returning home flushed with success, he puts away his former fears and resolves that after all he will have windows in his palace. H e thunders out of them; the earth reels and people far and near are terrified, his enemies cling to the rocks in dismay, and he openly taunts them: would anyone, prince or commoner, now dare to resist his royal power? Column 1.4 viii nicely leads into the third main division of the cycle as Baal sends messengers to the underworld abode of Mot inviting him to a feast to acknowledge his sovereignty.

1.1.2.3

Baal and Mot ( K T U

1.5-6)

Mot's embassy back to Baal dismisses this invitation contemptuously (1.5 i). Rather Baal, just as he once transfixed Yam-Nahar, will soon be descending to Mot's subterranean domains where he will be swallowed down and torn limb from limb by him. Baal, now in dread of Mot, sends an abject reply (1.5 ii). In 1.5 ν (after two very poorly preserved columns) Mot's messenger is advising Baal to take his wind and cloud and his other attendants down to the underworld and assume the condition of the strengthless dead. Baal obeys, but on his way he has connexion with a heifer, who is delivered of a boy w h o m Baal clothes in his own robe. 1.5 vi relates how the substitute's dead body was found at the edges of the earth and El, on hearing the news in his mountainous home, descends from his throne and wallows in sackcloth and ashes, wondering what will now become of Baal's followers. In K T U 1.6 i Anat and the sun-goddess Shapsh together bury the surrogate corpse, and Anat goes on to the abode of El to inform Athirat and her brood that they can now rejoice because Baal is dead. A short interlude follows in which Athirat proposes Athtar for the vacant kingship, but on proving unfit to take Baal's place he resigns. In K T U 1.6 ii Anat, seemingly now aware of what has happened, goes to the underworld to search for the real Baal. She confronts Mot and summarily slays him, and then (1.6 iii) returns to El to inform him that Mot is no more. She invites El to have a dream, in which he sees the heavens raining oil and the valleys running with honey; this shows that Baal still lives. In 1.6 iv Anat is invited by El to speak to Shapsh, and they make plans for Baal's return. K T U 1.6 ν and vi tell of Baal's restoration to full vigour in the world above, and also of Mot's recovery after seven years. T h e y argue threateningly with each other, at least some of the time on Mt Saphon, and finally fall to fighting. T h e y butt and gore like wild animals, and both fall exhausted to the ground. At this juncture the goddess Shapsh arrives to warn Mot that fighting with Baal is useless, for El is now on Baal's side. Mot, at last afraid, picks himself up and declares that Baal is rightfully king. 1.1.3

Interpretation

Since the early days of Ugaritic research the Baal cycle has nearly always been interpreted as a ritual and seasonal text, either enacted

dramatically at Ugarit's New Year festival in the autumn or, more broadly, as reflecting ritual events that took place at certain points of the Syrian agricultural year (see e.g. H v i d b e r g 1962 [1938], G a s t e r 1950, d e M o o r 1971, 1972). Such views, in various shapes, are still influential in ancient Near Eastern, including Old Testament, studies, but over the years they have been increasingly and quite severely criticised ( F o n t e n r o s e 1966, G r a b b e 1976, M . S . S m i t h 1986, W y a t t 1996a, 1998a). Few would argue today that where we have a myth, a ritual basis must be assumed or that the proper ordering of the seasons and their role in natural, or even h u m a n fertility were the only or even the chief interests of Ugaritic mythology. A good example of an agrarian meaning being read into a passage is the description of Anat's destruction of Mot in K T U 1.6 ii 3Iff., which speaks of her threshing him with a 'blade', winnowing him with a sieve, burning him with fire, grinding him with millstones, and then throwing his remains into the open fields for the birds to eat. Commentators have seen in this a mythological counterpart of a ritual ceremony held each year at the time of the grain harvest. H o w Mot's discomfiture at this time is to be equated with his normal role later in bringing about the summer dryness, or with his ongoing role in swallowing h u m a n bodies, is not said. M u c h more likely is the view that the whole scene is metaphorical ( L o e w e n s t a m m 1972); cf. the not dissimilar language used of the destruction of the golden calf in Exodus 32:20. In other words, Anat destroys Mot thoroughly; there is nothing more to it than that. O f late it has been argued (e.g. by W y a t t 1996a, 1998a) that some of the themes in the Baal cycle, notably the Chaoskampf, are much more archetypal than merely agricultural, and that they may really be referring to issues of kingship and the exercise of power in the ancient world. T h e way the gods behave is a mirror image of the way rulers in this world behave or should behave. Approaches along these lines could be much more fruitful. T h e part played in the story of the cycle by the supreme god El is particularly revealing here (see further G i b s o n 1 9 8 4 ) . T h e three great deities (and one minor candidate), whose exploits fill the cycle, are battling essentially for supremacy over the earth. T h e kingship to which they aspire is, as El's viceregent, to control the earth, YamN a h a r through the waters which surround it and are the source of its streams and rivers, Baal through his rains and thunder and the air which people breath, Mot through the droughts and dryness he

can cause. Baal is the victor in the contest, becoming the 'prince, lord of earth' ( K T U 1.3 i 3 - 4 ; 1.5 vi 10; 1.6 iii 9; 1.6 iv 29), holding at bay the unruly waters from outside and bringing the dry season to an end by his rains. But there is much more than these naturalistic roles to the three of them. Y a m - N a h a r also engenders moral chaos or evil in the lives of mankind, Mot eventually finishes everyone off, while Baal by his control of the atmosphere can be regarded as the life-force in the world of Ugarit. H e did not only secure year by year the ordered succession of the seasons, but every day of every year he had to wage a constant battle against evil and death in the lives of humanity, so that the one did not cause too m a n y depredations or the other gain too m a n y victims too soon. T h a t was really why Baal became the favourite god of the people of Ugarit. But he became this in the last resort u n d e r the supervision, indeed by the connivance of El reigning from his distant abode beyond the earth. From the standpoint of people on earth, looking around them in fear and trembling, the encounters between the negative and positive forces in their environment were tense and awesome affairs, and Baal's victory was always in the balance and never certain. But the people of Ugarit could also take comfort from their belief that beyond the squabbling powers that impinged so insistendy on their everyday lives stood a remoter but by no means disinterested figure, the god El, who had fathered the gods, the nice and nasty both, and had created the universe, contrary powers and all, who must therefore have planned it that way and built both good and evil into its very fabric, and who could for that very reason be trusted to uphold its equilibrium. For all his mistakes, his choosing the wrong side, his blustering, his pleading, his changes of mind, his putting up with impertinence from his underlings, his—if you will—lack of power, he was the one ultimately in charge, whom the other gods had to visit for approval whenever they had an enterprise planned, and whose was the final decision, which, however reluctandy, they had to accept. And perhaps most remarkably of all, he masterminded the balance he sought, not by calling on openly superior force but by relying upon an engaging mixture of diplomacy and conciliation, sharpness and persuasion. H e arranged it so that usually good and life triumphed, but even evil and death were his 'darling' and 'beloved' children ( K T U 1.1 iv 20; 1.4 viii 23~24) and had, as it were, their rights too. This is the view of reality espoused by the people of

Ugarit, their explanation of the divine ways with the universe and with h u m a n beings, their estimate of power and the m a n n e r it is exercised; it is, for all the fancifulness with which ancient peoples in their myths expressed themselves, mature and not lacking either faith or irony. It may have involved naturalistic reasoning, but it involved a great deal more besides.

1.2

Other mythological texts

T h e r e are in KTU quite a few other mythological texts or at any rate partly mythological texts, that is, passages embedded in legends, hymns or rituals. T h e larger of these, like Keret ( K T U 1.14-16 = R S 2.[003]+, 3.343+ and 3.325+) and Aqhat ( K T U 1.17-19 = R S 2.[004], 3.340 and 3.322+) which some may prefer to call legends or sagas, or the Rpum texts ( K T U 1.20-22 = RS 3.348, 2.[019] and 2.[024]), are given separate treatment later in this chapter (§§ 6 . 2 - 4 below). M a n y of the rest are mere fragments, e.g. extracts from the Baal cycle used probably for scribal practice ( K T U 1.7 = RS 5.180+, K T U 1.8 = R S 3.364, and K T U 1.133 = R S 24.293). I concentrate here on the most intriguing and (relatively) well preserved texts. They are all commented on, with up-to-date bibliography, in W y a t t , 1998c. 1.2.1

Baal and Anat ( K T U 1.10 = R S 3.362+)

This tablet has three columns of text on one side of the tablet only. O f the first column little readable text survives, but at the beginning of the second Anat calls on Baal (also called Hadd) in his palace. O n being told that he is out hunting in the Shamak marsh, she follows him there, and is warmly welcomed. She sees a cow giving birth and is apparently seized with passion, as is Baal who mounts her before returning to his sacred mountain. As a consequence she gives birth to a bull and, on her taking the news to Baal, he rejoices. Some scholars attach the tablet to the Baal cycle, but it is more likely that it belongs to a series recounting his dalliances with his sister (e.g. K T U 1.11 = R S 3.319 and K T U 1.13 = RS 1.006). It is not obviously théogonie, ritualistic or seasonal. 1.2.2

The Devourers ( K T U 1.12 = R S 2.[012])

This difficult text survives in two columns. T h e first tells of the conception of monstrous creatures by the handmaids of the deities Yarih

(the moon-god) and Athirat (the wife of El), who complain to El that they are being caused distress by carrying them. T h e head of the gods, doubtless their begetter, is amused by this, and instructs the handmaids to go into the desert to bear their offspring. In his naming of them they are likened to bulls and steers. Baal is present, and he expresses a great interest in them, perhaps for hunting purposes. In the second column, after a long gap, the offspring, called the 'devourers', set upon and destroy Baal, who falls into a swamp, after which the earth suffers a drought for seven or eight years. H e is eventually found by his brothers and and restored. At the end of the text a few lines tell the king to perform a water ritual, presumably to guard against the disaster caused by the 'devourers'. This disaster cannot be a seasonal disaster but is a long-lasting one. T h e real point does not seem to be about Baal but about El fathering such dangerous creatures. 1.2.3

Hymn to Anat ( K T U 1.13 = RS 1.006)

A hymnic text, interesting mythologically for its portrayal of the complex character of Anat, called the 'virgin' in the Baal cycle, and in the final lines here described as voracious to bring forth, although her w o m b had not known conception. T h e r e is no need to connect the hymn with any specific ritual, e.g. an incantation against infertility. 1.2.4

The Gracious Gods ( K T U 1.23 = R S 2.002)

A quite substantial text, written on both sides of a single column tablet and nearly complete. It is clearly a cultic tablet, most of the obverse consisting of little hymns, blessings on the king and queen and the ministerial personnel, instructions about repeating certain lines, about niches for the gods, about incense offerings, and so on. T h e r e are also citations from a few mythological texts, in some cases merely a heading, but in two cases rather fuller; in 11. 8 - 1 1 , there is a short excerpt about Mot-and-Shar ('death and the prince', a byname of the god of death), and in 11. 3 0 - 7 6 a longer story about El's seduction of two w o m e n (perhaps the goddesses Athirat and Anat), who give birth to Shahar and Shalim and then to the gracious gods as a whole. Mot-and-Shar holds in his hands the sceptres of bereavement and widowhood, and is felled by the 'pruners of the vine'. His removal from the scene makes it possible for El,

the progenitor of the gods, to father offspring on the two women. It is interesting that the first deities to be born are S h a h a r and Shalim, whose names mean 'dawn' and 'dusk'; we may compare the beginning of Genesis where the division of day and night is the first act of creation. But it is the behaviour of the gracious gods after birth that is worth remarking on. T h e y open their mouths greedily to swallow the birds of the air and the fish of the sea, and are sent off by El with their mothers into the desert where for seven or eight years they hunt for food. T h e y eventually come upon someone callcd the 'watchman of the sown land' who invites them in to continue their eating and drinking. Is the meaning of this myth that the gods are not satisfied with the natural provision of the open country, but require in addition the offerings of the cultivated land which h u m a n beings bring them? It is unlikely that such a profound observation, in effect that though mankind are clearly dependent on the gods, they in their turn are dependent (or at least partially so) on mankind, would be confined to one particular ceremony, agricultural, fertility-angled or otherwise. Doubdess this myth found expression on numerous liturgical occasions at Ugarit. It is a not untransparent theogony or explanation of how and why the gods came into existence. 1.2.5

Nikkal and the Kotharat ( K T U 1.24 = RS 5.194)

T h e mythological portion of this text, written like the last one on two sides of a single tablet, relates how a deity called Hirhib, king of summer, who behaves like a typical eastern marriage-broker, arranged the betrothal of a lunar goddess Nikkal to the moon god Yarih. It is probably, like the second narrative in the previous text, an extract from a fuller théogonie myth. This tale is preceded and followed by hymns of praise and invocation to Nikkal, Hirhib and the Kotharat, who are the sages-femmes of the Ugaritic pantheon. In the first hymn the Kotharat are summoned to oversee the birth of a son to the two moon deities. T h e last lines of the second hymn with their allusions to incantations to the Kotharat, betray the purpose of the whole poem, which is to secure for a h u m a n girl Prbht the same blessing and protection in her forthcoming marriage as had been enjoyed by the goddess Nikkal in hers. Probably, with the necessary change of the girl's name, the text was regularly recited at ceremonies of engagement and courtship.

1.2.6

Shapsh and the Mare ( K T U 1.100 = RS 24.244)

This is a long and excellently preserved but difficult text containing in the opinion of most commentators a charm against snake-bite. T h e daughter of the sun-goddess Shapsh (or perhaps simply a mare, as her n a m e may be translated) calls on her to carry a message to El, Baal and various other deities in order to obtain help from them in curing the malady. Only when the god H o r o n (apparently a chthonic deity) is a p p r o a c h e d is a positive response forthcoming. According to others, however, the text is chiefly a mythical narrative, not a charm and the serpent mentioned represents some cosmic disaster which is removed by H o r o n . 1.2.7

El's Banquet ( K T U 1.114 = R S 24.258)

This is description of a banquet to which El invites the other gods and at which he falls outrageously drunk. T h e last lines on the reverse contain an incantation for the cure of a disease or perhaps, as has been suggested, a hangover. T h e texts assembled here, some of recent discovery, give us a glimpse of the diversity of Ugaritic mythology but, apart from the Baal cycle and other larger texts like Keret and Aqhat, their extent is not very great. W e have some way to go before a comprehensive account is possible. Perhaps the only things we can say is that myths are not always, if much at all, connected with ritual, and especially that Near Eastern, including Ugarit, mythology is not always, if much at all, obsessed with matters of seasonal agriculture or fertility.

2

The

Legend

Baruch

2.1

of

Keret

Margalit

The history of the text: discovery, publication, editions

2.1.1 T h e poem of Keret is one of the three major literary works which gifted Canaanite poets of the Late Bronze Age (ca. 1500-1200 bce) bequeathed serendipitously to 20th century civilization. Excavated at modern Ras Shamra on the northern Mediterranean coast by a French archaeological team in the early thirties (1930-31), the poem was published in three u n e q u a l instalments by the Assyriologist V i r o l l e a u d , beginning with a monographic study in 1936 and concluding several years later with a series of articles in the periodical Syna (vols. 22-23) published during the war. Accordingly, only the monograph was widely known and accessible before 1945, although a great deal of interpretation, much of it fanciful, was current in the interim. 2.1.2 After the war, the text was republished or re-edited several times. G o r d o n reproduced Virolleaud's text in the successive editions of his Ugaritic G r a m m a r . A new and critical edition was published by H e r d n e r in 1 9 6 3 . This much acclaimed two-volume work consists of all the alphabetic texts, literary and other discovered in the thirties, together with photographic plates and autograph facsimiles. Generally abbreviated as CT[C)A, this edition established the by-now standard numeration of the Keret text as 1 4 - 1 5 - 1 6 , corresponding to Virolleaud's IK-IIIK-IIK respectively. 2.1.3 CT[C)A was followed by a new edition published in 1976 by a team of Ugaritic specialists affiliated with the University of Münster (Westphalia) headed by Oswald Loretz. Entitled Die Keilalphabetischen Texte aus Ugarìt (KTU), it contains (in transliteration only) all the alphabetic texts discovered up to 1970. In this edition the Keret poem appears as K T U 1 . 1 4 - 1 5 - 1 6 = RS 2.[003]+ 3.343+ 3.325+, the initial cipher indicative of its classification as a literary-poetic text. A revised edition appeared in 1995. It too has the transliterated text only. 2.1.4 T h e text of the poem, labelled krt by the ancient scribe after the royal hero of the story, is distributed over three rectangular clay

tablets of similar size (21x17 c m ; 15x17 cm; 2 3 x 1 7 . 5 cm). Each tablet comprised originally six double-ruled columns, three on each side, a n d altogether contained a p p r o x i m a t e l y a t h o u s a n d lines of compactly written text. O f the three, only K T U 1.14 = R S 2. [003] + is relatively well-preserved, a n d with the aid of the m a n y duplicate passages in this portion of the p o e m can be restored nearly to its pristine state. Most Ugaritic specialists hold that the extant tablets never comprised the entire p o e m a n d assume that one or m o r e tablets have been lost, especially at the conclusion ( K T U 1.16 = R S 3 . 3 2 5 + vi); this despite the colophon at the end of K T U 1.16 = R S 3 . 3 2 5 + vi n a m i n g the scribe Ilumilki w h o committed the p o e m to writing (spr). 2.1.5 Since its discovery, Keret has been translated m a n y times a n d into several languages, either as part of an anthology of Ugaritic literary texts or m o r e broadly of ancient N e a r Eastern texts. T h e m a j o r translations, a n d the most widely cited, are the ones by (a) G i n s b e r g , first in his A S O R m o n o g r a p h ( 1 9 4 6 ) a n d subsequently as part of P r i t c h a r d ' s anthology (AJVET); (b) G . R . D r i v e r ( 1 9 5 6 ) later substantially revised by G i b s o n ( 1 9 7 8 ) ; (c) H e r d n e r , in a work jointly p r o d u c e d with C a q u o t a n d S z n y c e r (TOi) in 1 9 7 4 , a n d (d) d e l O l m o L e t e in 1 9 8 1 , where the Spanish translation is a c c o m p a n i e d by extensive analytical discussion a n d a unique synoptic presentation of alternative translations. 1 2.1.5.1

2.2

The history of (misinterpretation

2.2.1 T h e history of the p o e m ' s interpretation during the past sixty years can be divided into three phases. Initially it was the subject of extravagant claims of historicity a n d 'biblicization'. 2 T h e hero was thought to have been a Phoenician king whose legions, including m e m b e r s of the Israelite tribes of Asher a n d Zebulon, waged w a r in the N e g e b region of Palestine a n d in E d o m . Progeny or devotees of the biblical T e r a h , the father of A b r a h a m , were also thought to be involved. However, rapid progress in Ugaritic philology p u t a 1

Also noteworthy are G O R D O N 1977, 34-59; DE M O O R 1987, G R E E N S T E I N 1997, 9 - 4 8 ; L O R E T Z 1997 and P A R D E E 1997a, 333-43. See also W Y A T T 1998C, 175-243. 2 Uncannily reminiscent of the Eblaite euphoria some 40 years later; plus ça change. . . .

quick and merciful end to this euphoric era and to the fata M o r g a n a of biblical persons and tribal entities. O f the alleged geographical allusions, only the shrine of Asherah in the environs of Tyre and Sidon would survive the debacle; and even this determination was destined not to go uncontested (Cf. A s t o u r 1973, 29-39). 2.2.1.1 Still, there is wide if not consensual agreement today that two fundamental insights of this early era in Ugaritic studies retain their validity, viz., (a) Late Bronze Age Keret, like Aqhat, reflects a literary genre qualitatively different from the mythological tales of Baal, Anat, and the members of the Ugaritic pantheon generally—this notwithstanding the prominent roles of divine actors in both poems; (b) the main works of Ugaritic literature—Baal-Mot, Keret and Aqhat— are 'classics' of Late Bronze Age Canaanite civilization and culture and as such were known in Iron Age C a n a a n generally and in ancient Israel specifically. 2.2.2 A second and similarly fleeting phase in the interpretation of Keret was introduced by the Scandinavian, secondarily British, Myth and Ritual School, represented—at its most extreme—by the publications of Engnell and Mowinckel. T h e former considered the poem 'a ritual for the Ugaritic sukkot festival' ( E n g n e l l 1967, 149) T h e wedding party for Keret and his bride described in K T U 1.15 = RS 3.343+ is characterized as 'originally the [ιερός γάμος] of the god and goddess, celebrated annually and co-experienced by the participants as they watched the cult-drama and also when indulging in sacral prostitution' ( E n g n e l l 1967, 148). For Mowinckel, on the other hand, the poem exemplified myth attenuated as legend; behind the portrait of the hero as a Phoenician king stands the figure of Adonis: '[in Keret] the god is . . . strongly anthropomorphized; the original god has become the dynastic founder, the mythic first ancestor of the royal family . . . the poem is no longer a real myth, but a mythic hero-legend.' 3 2.2.2.1 If the shortcomings and misconceptions of the French historical school were the result of inadequate philology, those of the myth-ritualists were the product of faulty methodology. By means of

3

MOWINCKEL

MOWINCKEL

1954,

1941, 142-3, as translated by 52-5.

ENGNELL

1967, 148. See further

careful selection and tendentious interpretation of certain model texts, 'evidence' is created proving the existence of 'an organic [ANEastern] culture . . . whose special feature is the domination throughout by the divine kingship idea' ( E n g n e l l 1967, 2). This 'pattern' is then applied to other texts assumed a priori to reflect this 'pattern'. T h e explanatory value of this theory is commensurate with its (non-)falsifiability. Basic to this approach, which breathed its last in Ugaritic studies with G r a y ' s monograph in the mid-fifties, 4 is the axiom that any ancient N e a r Eastern literary text, be it myth or legend, is necessarily 'functional', and almost invariably so in the cultic sphere where the ancients are presumed to have spent all their leisure time. In Gray's words, 'the t e x t . . . was not an aesthetic exercise'—presumably the author's understanding of literary creativity—'but served a practical purpose in the community where it was current to achieve some desired end or to conserve . . . all the social conventions and the social order'. 5 2.2.2.2 It is also typical of this approach that its advocates do not feel constrained to demonstrate precisely how this 'conservation of values' is actually implemented in the poem or how an audience might infer such a conclusion. It never occurs either to Engnell or to Gray to query whether the story might not be understood by at least some readers or listeners in a quite different, even opposed, manner, e.g., as underlining the perilousness of a social order predicated on the health of an individual, mortal king supported in turn by a bunch of rather inept gods; or that the author of Keret, far from preaching the doctrine of divine kingship, might in fact be condemning it by means of a lethal dose of parody. In the final analysis, the failure of the Myth and Ritual school lies in its denial of the literary ontology of the text. 6 revised and attenuated in G R A Y 1 9 6 5 . 4 - 5 . In all fairness it should be noted that he concedes that the poem 'was not deliberately [so] designed' (ibid., 5). But nowhere are we informed what this 'original' design may have been, since it surely was not a mere 'aesthetic exercise'. This concession, however, contradicts the Myth-Ritual postulate of a cultic 'Sitz-im-Leben' for all texts with divine characters and in fact heralds the school's demise. 6 Contrast the astute observation of DE L A N G H E 1 9 5 8 , 1 3 1 , citing B A U M G A R T N E R ( 1 9 4 1 , 8 9 - 9 1 ) that 'aesthetic interests stand side by side with religious interests' in the Ugaritic literary texts. However, very few specialists in Ugaritic have taken this admonition to heart, either before or after. This is true even of B E R N H A R D T ( 1 9 5 6 ) , who is at pains to criticize—very successfully—the Myth-Ritualists but whose own 4

GRAY

1954,

5

GRAY

1954,

2.2.3

O n e cannot take leave of the 'pre-historic' era of Ugaritic studies without taking note of the important study by the Danish Semitist P e d e r s e n . Published in 1 9 4 1 with only V i r o l l e a u d ' s monograph at hand ( 1 9 3 6 ) , Die Krt Legende is probably the only study of this era whose influence abides to the present. This influence has unfortunately perpetuated a basic misconception in the interpretation of the poem. 2.2.3.1

to his credit, rejects outright the Myth-Ritual interpretation of M o w i n c k e l — E n g n e l l ' s study ( 1 9 6 7 ) had yet to app e a r — w h e n he states categorically that Keret 'ist nicht der Ausdruck kultischer Vorgänge und ist kein Mythus' ( P e d e r s e n 1 9 4 1 , 6 4 ) . Methodologically he stands close to the French School. He assumes that we are dealing with a historiographie work whose historical kernel is heavily overlaid with legend—'wie der israelitischen Passahlegende' ( P e d e r s e n 1 9 4 1 , 6 4 ) . T h e hero 'ist ein Urkönig, Gründer einer Dynastie' ( P e d e r s e n 1 9 4 1 , 6 5 ) — i n reality, not just in the plot; and the principal theme is 'die Sicherung der Dynastie durch N a c h k o m m e n schaft' ( P e d e r s e n 1 9 4 1 , 6 4 ) . In other words, Keret is at heart a work of propaganda commissioned by a royal house and executed by a poet with the soul of a priest. 2.2.3.2

Pedersen,

2.2.3.2.1 T h e theme of divinely sanctioned dynastic kingship, moreover, is deemed to reflect a society 'deren Königsgeschlecht schon als eingewurzelt betrachtet werden kann', analogous to the Israelite society which spawned the Davidic royal ideology: 'hier wie dort handelt es sich u m Legenden welchen den festen Bestand der herrschenden Dynastie bestätigen und begründen' ( P e d e r s e n 1 9 4 1 , 1 0 4 ) . T h e author of Keret is so to speak a 'kept w o m a n ' of the political authorities. His hand is free to write but his soul is in bondage. O n e does not normally take the work of such writers seriously, whatever their technical virtuosity.

2.2.3.3 This view of the Keret poem as a work of royal propag a n d a — b y implication if not explicitly, by the ruling house of the

position is merely a re-statement of PEDERSEN'S: 'Richtiger wird man jedoch von einer Bestätigung der Erwählung der Keret-Dynastie sprechen' {ibid., 119). This is the raison d'être of a political manifesto, not a work of art.

kingdom of Ugarit whose dynasty Keret is presumed to have founded— is very widespread in contemporary Ugaritic scholarship; and it is hugely mistaken. 7 2.2.3.3.1 T h e curious omission, on these assumptions, of any reference to K e r e t — o r his son and heir Yassib—as king(s) of Ugarit has long been noted. It is reinforced by a similar omission in the so-called 'Ugaritic king list' ( K T U 1.113 = R S 24.257). However, the real shortcoming of this view is that it fails to distinguish the substance of the plot from the authorial intention, the creation from the creator. T h e statements placed in the mouths of the characters are naively taken as the author's own point of view. 8 2.2.3.4 In fact, a close, methodologically unbiased scrutiny of how the author of Keret depicts his characters must surely lead to the conclusion that far from endorsing sacral dynastic kingship the poet actually ridicules it. T h e opening scene, for example, portrays the king as a hapless soul who has gone through seven wives—the first of w h o m simply 'walked out' (tbc) on him!—and who can think of no better expedient than, like a baby, to cry himself to sleep (only the soothing lullaby is missing. . .). Subsequently he will conscript all the men of his kingdom—including the disabled and the newly-wed— for a 'historic' military campaign to the Bashan for the grand purpose o f . . . obtaining a wife! It is inconceivable that this entire scenario should have evoked from a contemporary audience anything but gales of laughter. 9

7

Cf. e.g., B E R N H A R D T ( 1 9 5 6 , 1 2 0 ) : 'der text [steht] in enger Verbindung mit der durch Keret begründeten Herrscherdynastie . . . und [hat] als Tendenzdichtung die Aufgabe . . . die besondere göttliche Erwählung gerade dieser Dynastie ganz augenfällig darzustellen.' Cf. also above, η. 6. 8 Thus, even if it be true 'dass wir in Keret einen typischen Vertreter des altorientalischen Sakralkönigtums vor uns haben' ( B E R N H A R D T 1 9 5 6 , 1 1 6 ) , it does not follow that this is an ideology which the author either espouses or wishes to propagate. Can one legitimately infer from the detailed description of the Persian monarchy in Esther that the author is desirous of propagating an ideology of oriental despotism? Indeed, but for the strong nationalistic motives attributed to its (supposedly) Jewish author, the book of Esther might well have been understood as political satire. 9 T h e fact that the latter stratagem is concocted by Keret's divine patron—with Baal nowhere in sight!—does not make it more 'respectable'; it merely adds to the scope of the ridicule. El in Late Bronze Age Ugarit is a museum piece and a soulbrother of Shakespeare's Falstaff.

2.2.3.5 But the most telling refutation of the dynastic interpretation comes from the final scene of the poem (a scene unknown to Pedersen at the time of his essay) in its portrayal of Yassib, the king's eldest son and divinely-ordained (as well as politically confirmed) heir. It is difficult to imagine a less favourable comment on dynastic kingship or a more incongruous endorsement of a royal line supposedly founded by Keret. 2.2.3.5.1 In short, there is altogether too much comedy ody in Keret for it ever to have served as propaganda for but the joy of living. For the author of Keret, not even the sacred, much less the political institution of kingship. His devotion are given unconditionally only to his art." 1

and paranything gods are love and

2.2.4 2.2.4.1 G i n s b e r g ' s short monograph published in 1946, inaugurated a new era in the poem's interpretation—or rather, explication. O n e of his severest critics, Gaster, hailed it as 'a marked and revolutionary advance in our understanding [of the text]' ( G a s t e r 1947, 385). Ginsberg's was the first study to have addressed the material in its (extant) entirety: K T U 1 . 1 4 - 1 5 - 1 6 = RS 2. [003]+ 3.343+ 3.325+, are fully at his disposal and will be so henceforth for the scholarly world to study and analyse. It is Ginsberg's contribution to have been the first to establish the narrative coherence of the text. However, Gaster—a dues-paying m e m b e r of the Myth and Ritual school—took Ginsberg to task for 'his obvious lack of acquaintance with c o m m o n facts and methods of comparative religion, anthropology, and folklore . . . This leads . . . to an egregious disregard for the cultural context and background of the narrative.' ( G a s t e r 1947, 286-7). 2.2.4.1.1 But Ginsberg consciously and deliberately eschewed 'metaphysical' interpretation; he was a devout positivist. His strength lay lu

In a recent interview to a Montreal newspaper on the occasion of his 85th birthday (The Gazette, Mar. 7/97), Irving Layton—Canada's (unofficial) poet-laureate— offered the following assessment of his life in the service of his art: 'Poetry never let me down. My worry is, have I ever let poetry down? I should like to think that I've never dishonoured poetry or turned my back on it. . . . A world without poetry would be just intolerable. Unbearable.' O n e cannot mistake the (unintentional) piety of this inveterate God-baiter and iconoclast. T h e ancient Canaanite bard would surely have given this credo his unqualified assent.

in his philological dexterity at the level of grammatical analysis. H e was primarily interested in words and how they combine to form grammatical structures. T h e ideational content is secondary and the literary craftsmanship incidental. Nevertheless, Ginsberg took a definite stand on several 'metaphysical' issues (without however making them a part of his discussion or interpretation). H e considered it 'probable' that the story 'contains a certain core of history'; he also deemed it 'probable' that text K T U 1.14 = R S 2. [003]+ was preceded 'by one or more lost tablets'. Probability becomes certainty on the question of the poem's alleged non-conclusion in K T U 1.16 = RS 3.325+." 2.2.4.2 For all his skill in explicating the text, Ginsberg either misconstrued or overlooked several key elements in the plot, beginning with the mistaken notion 12 of Keret as the victim of the catastrophic loss of countless children deemed to have perished in bunches: a third, a fourth, a fifth, etc. 13 H e is completely unaware that the real reason for convening the nobility of Bêt-Hbr ( K T U 1.15 = RS 3.343+ iv-vi) is to confirm Yassib as Keret's successor. In fact, Ginsberg's translation of K T U 1.15 = R S 3.343+ v-vi lacks the thread of narrative coherence. This in turn leads him to wonder whether K T U 1.16 = R S 3.325+ i is the direct continuation of K T U 1.15 = R S 3.343+. 2.2.4.2.1 T h e r e are also some incongruities in Ginsberg's rendering of K T U 1.16 = R S 3.325+. T h e phrase pnh. tgr. ysu ( K T U 1.16 = R S 3.325+ i 52-3), referring to the king's daughter 'Octavia' as she emerges from the gateway of her (convent) residence, is under" The assumption of a story with large gaps before, within, and after the extant tablets is very useful for the philologist operating exclusively with the microscope of comparative grammar. If we possess only a half of the original work, then we obviously are severely handicapped, if not outright precluded, from interpreting it macroscopically on the metaphysical level. It is also something to fall back on when one's interpretation of a given section, at the philological level, seems literarily vapid or even inconsistent with that of a preceding or following section, whether in terms of characterization or plot. It is not without irony that one notes the common ground shared by Ginsberg and the myth-ritualist: neither takes seriously the poem of Keret as literature or its author as artist. Both approach the text as so much grist for the grinding. 12 Corrected early on by Cassuto but ignored by Ginsberg, even in his later work. 13 Here loo one senses how the lack of esteem for the pagan as poet enables the Western scholar to attribute to him such a literary inanity, not to say moral obtuseness, in so quantifying human beings. Surely the biblical prejudice of the idolatrous Canaanite and his 'debased' culture is here subverting the scholarly enterprise.

stood to mean 'Its sheen (i.e., of brother Ilhu's lance \mrh\) lights up the gateway'. H e makes no effort to translate K T U 1.16 = RS 3.325+ ii 2 4 - 3 4 , although the text is quite well preserved; and he passes over in silence the sudden a p p e a r a n c e of Octavia in her father's bed-chamber in lines 5 0 - 1 . In col. iii, 8 - 9 , the language tnnth . . . tltth does not evoke in Ginsberg's mind the association with Keret's vow in K T U 1.14 = R S 2.[003]+, and his understanding of the Ša C tiqat episode (v 28-vi 14) is both faulty and incomplete; especially curious in his failure to render yqrs, while citing the analogy with Gilg. I ii 34 and its reference to tīta iqtaris. Finally, the curse which concludes K T U 1.16 — R S 3.325+ (vi 54-8) is deemed by Ginsberg to be 'unintelligible', in which case one wonders at the certitude which informs his opinion as to the non-conclusion of the poem at this point. 2.2.4.3 T h e foregoing critique, be it noted, is based not on Ginsberg's early translation ( G i n s b e r g 1946) but rather on his contribution to P r i t c h a r d ' s anthology, first published in 1950 and subsequently (unrevised!) in 1955 and 1969. T h e authority of Ginsberg's name— he was widely considered to be the 'doyen of Ugaritic studies' in the fifties and sixties—and the popularity of Pritchard's anthology, which soon became a standard reference work for biblical and ancient Near Eastern studies, go a long way towards explaining the rather limited progress made subsequently in the elucidation of the poem at the most basic level of narrative explication. T h e unspoken if not also unconscious assumption is that short of a windfall discovery of additional copies, Ginsberg's translations of the major Ugaritic poetic texts ( K T U 1.1-6 = R S 3.361, 3.367, 3.346, 2.[014]+, 2.[008]+, 2.[022]+, 2.[009]+; K T U 1.14-16 = R S 2. [003]+, 3.343+, 3.325+; K T U 1.17-19 = RS 2.[004], 3.340, 3.322+) have defined the limits of what scholars can ever hope to know of them. 2.2.5 2.2.5.1 A new phase in the study of Keret is introduced by M e r r i l l ' s short essay ( 1 9 6 8 ) , marking the first serious attempt to deal with the poem as a literary œuvre and providing the inspiration for an important essay by P a r k e r ( 1 9 7 7 ) nearly a decade later. 2.2.5.2 ' T h e hypothesis of this paper', writes Merrill, 'is that the poem . . . points to the 'house of Keret' as the basic issue. Every part

of the narrative finds its focus and delineation in this motif.' ( M e r r i l l 1968, 7). T h e story, it is supposed, 'begins with the ruined and impoverished house of Keret. T h e king stands alone, without heir, wife, or progeny.' ( M e r r i l l 1968, 9). By the end of K T U 1.15 = R S 3.343+, 'the narrative of the king who has lost his 'house' and regains it appears to be complete in itself. It has a beginning, a middle, and an end . . .'; and with mild surprise, 'yet the story continues' ( M e r r i l l 1968, 9-10). T h e 'fact' that the story continues beyond its 'logical' conclusion leads Merrill to the conclusion—taken up and elaborated subsequently by Parker—that the unfulfilled vow to Asherah, and her ensuing wrath, 'become the basis for the addition of the other 'narratives' which are woven around the central concern for the 'house of Keret' and find their sub-themes in the three areas of fertility, salubrity, and sovereignty.' Keret, on this hypothesis, is a composite work, although Merrill stops short of assuming multiple authorship. 2.2.5.3 It is the merit of Merrill's essay to have dealt with the story in its own terms and with a vocabulary drawn from the field of literary criticism rather than comparative religion or Semitic linguistics. Implicit at least is the assumption of an author who has something interesting, perhaps even important to say, and who commands the necessary tools of the trade which he employs with the skill and imagination worthy of an artist. This approach also implies an audience who can appreciate such a work, not as a cultic libretto or a catechism of theological-political indoctrination, but as an artistic endeavour, which, like good wine, is to be savoured and enjoyed. 2.2.5.3.1 But for all the freshness and originality of its approach, Merrill's essay, like Parker's subsequently, goes astray in its effort to determine what the author is trying to say, as well as the specific techniques which he has chosen for this purpose. T h e hypothesis of a composite work, and a fortiori of multiple authorship, is symptomatic of a basic misconception, or rather, misperception. 2.2.5.4 T h e view of an ancient work of Semitic literature as composite comes easily to scholars trained primarily in Old Testament exegesis, as their partiality to myth-ritualism and cultic solutions generally tends to reflect their roles as (practising) theologians in the Judaeo-Christian tradition. But it is nonetheless a view quite unfounded

here in Keret and in Ugaritic literature generally. 14 T h e r e is no evidence for a 'history' of any of the m a j o r Ugaritic poems, although such is not to be precluded a limine.^ 2.2.5.4.1 T h e vow-to-Asherah episode, it must be insisted, is absolutely central to the plot of the story for the simple reason that it alone supplies the story with its dramatic quality. Without the vowepisode the story is a tale not worth the telling, much less the price of admission to its performance. T h e absence of a corresponding instruction in the dream-episode does not prove the vow to be secondary: if someone were intent on tampering with the original by 'grafting' on the vow episode, he would have had little difficulty making the necessary emendation in K T U 1.14 = R S 2.[003]+. 1 6 2.2.5.4.2 T h e omission, on the other hand, speaks volumes for the authorial intention. T h e r e is nothing more characteristic of the (male) dramatis personae in Keret than their personal shortcomings and imperfections—including most definitely the head of the pantheon who (like Y H W H in the Garden-of-Eden story) fails to anticipate his clientservant's initiative. 1 ' 2.2.5.4.3 However, there is a second and more basic problem in Merrill's theory, viz., his initial assumption that the well-being of 14

It would be inappropriate in this connection to cite in rebuttal the complicated history of the Gilgamesh epic for obvious reasons related to the chronological spans of the respective works. 15 One should also not wish to deny the existence of 'parallel traditions' in Ugaritic literature, notably the stories dealing with the construction of Baal's palace (KTU 1.3 II 1.4). However it has yet to be demonstrated (though often assumed) that 1.3 and 1.4 belong to a single literary work or that they constitute a consecutive narrative. 16 T o be noted in this connection are the ill-preserved conversations of the Udumite king, first with his wife Na'amat (KTU 1.14 = RS 2.[003]+ ν 14-23) and subsequently with his messengers, commissioned to scale Mt Inbb and offer sacrifice to the gods {ibid., 24-9; cf. Margalit, 224 31), both of which are unforeseen in Keret's dream. Since no authorial design can be discerned in their omission from the dream, and since nothing in the sequel would seem to presuppose these conversations, the theoretical possibility of a 'second hand' can be entertained here. However, as presently constituted the scene has the positive effect of 'humanizing the enemy', a sentiment very close to the (original) author's heart, as is evident from the emotional departure-scene which follows shortly at the beginning of K T U 1.15 = RS 3.343+. 17 It should not be overlooked that once Keret awakes, El 'disappears' from the story. He will return as a guest at the wedding reception, but he cannot be supposed to have monitored his client's actions in the interim.

KereCs dynasty stands at the centre of the poet's concern and creation. It is simply not true that the king is portrayed at the beginning of the poem as impoverished note how easily Keret dismisses El's offer of 'silver and gold' in the dream, and the king of U d m ' s bribe subsequently during the siege. H e lacks progeny, but not for having been bereaved; like Dan'el, he lacks a male heir for not having sired one! N o w just as the birth of a son in Aqhat does not signal the completion of the story but more nearly its commencement, so too does the birth of Keret's offspring provide the impetus for moving the story to its climax. T h e truly important developments in the story come after the birth: in the case of Aqhat, the lad's treacherous m u r d e r by the goddess Anat and her Sutean mercenary, followed by the homicidal act of blood-redemption by the hero's sister. In the case of Keret, the 'meat' of the story is the king's illness and the behaviour of his offspring in response. T w o of them, without aspirations to the throne, are devoted, loving, and obedient. T h e third, predesdned by birth as heir-apparent, is the spoiled-brat antithesis. T h e attempted putsch by Yassib and the thunderous curse called down on his head by his enraged father ( K T U 1 . 1 6 = R S 3 . 3 2 5 + vi) bring the story full circle as it drives home the principal message: Keret is miserable at the beginning of the story for want of a son and heir; he is equally miserable at its conclusion precisely because of his son and heir. If the curse were not so funny—Yassib examining his teeth in the cup of his h a n d — t h e ending would indeed be sad. This is the essence of the poem as tragi-comedy, mixing the tears of laughter with those of pain. 2.2.5.4.4 T h e fate of the 'house of Keret' is thus of no particular interest either to the poet or his audience. T h e real 'star' of Keret is neither the king nor the gods but the invisible Moira who like the poet delights in irony and makes the h u m a n life-experience at once fascinating and unpredictable—the very qualities required of a good story! 2.2.6 For P a r k e r ( 1 9 7 7 , 1 6 7 ) , the poem of Keret is a conflation of three originally independent stories executed by different poets at different times and with variable degrees of editorial skill. ' O u r conclusion . . . is that the first section [= A] of Keret originally stood on 2.2.6.1

its own, and the material dealing with Keret's sickness [= B] was attached to it by the insertion of the promise [= vow] passage into the journey to U d m . . Λ 18 U p to this point Parker is echoing Merrill. But he goes further in positing multiple authorship and in his understanding of the Yassib episode ( K T U 1.16 = R S 3.325+ vi) as an 'originally independent story . . . tacked [iic] onto section B' ( P a r k e r 1977, 169). 2.2.6.1.1 T h e alleged 'third story' (C) is of course quite incomplete, and is assumed to have continued on (a) no longer extant tablet(s). It follows accordingly that 'we are scarcely in a position to speak of the theme or function of the whole work' ( P a r k e r 1 9 7 7 , 1 7 4 ) . Here too (cf. critique of Ginsberg above) the assumption of incompleteness serves as a safeguard against criticism based on literary considerations; the 'answers' to difficult questions can be conveniently assumed to lie in the unattested and empirically unverifiable 'hereafter'. 2.2.6.1.2 E.g. it would be most surprising if a literary work which had evolved in this 'tacky' way could be shown to have a unifying theme or structure. Yet according to Parker, the combination of story A and the 'neatly grafted' story Β yields, remarkably, a unified theme described by him as 'the vulnerability and helplessness of the king on the one hand, but also the benevolent power and wisdom of El on the other." 9 2.2.6.2 We have discussed earlier some of the weaknesses in Merrill's argument for the secondary nature of the vow episode. In his monograph P a r k e r ( 1 9 8 9 ) tries to meet one of these objections, but in so doing actually reinforces it. 2.2.6.2.1 Parker acknowledges that (a) the reason for suspecting the vow is its absence from the list of detailed instructions in the dreamtheophany of K T U 1.14 = RS 2. [003]+ which the king subsequently

IB

'The poets [TTD] who thus extended the poem . . (PARKER 1 9 7 7 , 167). 1977, 174. In P A R K E R 1989 he claims to have discovered significant structural differences between A and B. But at best these differences do not necessitate a distinction of authors. They are certainly consistent with the stylistic versatility and literary virtuosity of a single writer. 19

PARKER

carries out to the letter and which include, incidentally, a sacrifice to the god Baal, presumably to enlist his support for the venture; (b) a good 'grafter' would accordingly have encountered little difficulty in making the necessary adjustment. Therefore (c) the omission was motivated ideologically, viz., by reverence for El's reputation. T h e grafter did not want to make El responsible for the subsequent debacle. 2.2.6.2.2 Setting aside the conjectural and suspiciously ad hoc nature of this latter supposition, it is surely clear that it effectively undermines the case for multiple authorship. T h e same pious concern for El's reputation could as easily have motivated the original author of the poem! In other words, the vow-episode loses through this 'explanadon' its entire value as an empirical indication of multiple authorship. 2 0 2.2.6.3 T h e case for viewing the Yassib episode as secondary— Parker's own contribution to the hypothesis of a composite work— is devoid of even the prima facie evidence supporting the secondariness of the vow episode. O n e suspects that the very idea owes its birth to the widely held view of the poem as lacking, in its extant form, a conclusion, and hence needs have been continued elsewhere. It seems improbable that Yassib would make his one and only appearance at the end of the story. T h e missing conclusion, it is supposed, will have described how Yassib was punished for his insolence by forfeiting his claim to the throne in favour of his younger sister Octavia, the favourite of El and the gods (.sgrthn. abkm. etc.). 2.2.6.3.1 Were such a denouement actually attested, it might well be taken to support a theory of compositeness and multiple authorship, for it would totally contradict, in substance and spirit, much of what has transpired in the poem up to this point. 2.2.6.3.2 T h e fear is however unfounded. In point of fact, the role of Yassib is much more firmly rooted in the story than is readily apparent from his single appearance in K T U 1.16 = RS 3.325+ vi. Parker's hypothesis appears to antedate the realization that the baro-

20 Here too the influence of O.T. scholarship is readily apparent. Bible scholars commonly assume 'pious glosses' in the text originating with 'pious Jews' of the post-exilic era.

niai council in K T U 1.15 = RS 3.343+ v-vi was only ostensibly convened to 'weep for Keret', and tbat its agenda was secretly political: to confirm Yassib as successor in the event of Keret's demise— a decision ultimately taken, albeit after stormy debate. 2 ' With this 'certificate' in hand, Yassib has no reason to challenge his father during the latter's illness (as Parker suggests he ought to have done if his role were original); he need but bide his time until the king's imminent demise. It is only when his expectations are frustrated by Keret's miraculous recovery that he makes a pathetic attempt to unseat him. Yassib, for whose confirmation so m u c h energy had been expended but whose true character the author has skillfully concealed up to this point in the story, is now revealed at the conclusion for the 'wimp' that he is! 2.2.6.4 T h e r e are two points to be emphasized in connection with Parker's hypothesis: (a) that story Β the king's illness—is securely tied to the figure of Yassib and his succession and can never have existed independently thereof; (b) that the case for the Yassib episode as an independent story C hangs entirely on the assumption that the poem is not concluded at the end of K T U 1.16 = R S 3.325+ vi. 2.2.6.5 T h e final point to be m a d e against Parker's case is the catastrophic consequences which the deletion of the Yassib episode has on the literary structure and message of the poem. 2.2.6.5.1 If Keret were a typical 'happy ending' narrative, the story (Parker's story Β in particular) ought to have ended with the king's recovery (much as Merrill's 'original story' ought to have ended with the wedding reception in K T U 1.15 = RS 3.343+!). For if at the beginning ( K T U 1.14 = R S 2.[003]+) the king is alone and in tears, and then subsequently, facing death, he is tearfully embracing—possibly for the last time—his beloved 'blossom' Octavia ( K T U 1.16 = R S 3.325+ ii 50ff.), he is surely smiling from ear to ear, surrounded by his faithful wife and adoring children, at the feast described (laconically) in K T U 1.16 = R S 3.325+ vi 15-21; and his resumption of

21

The language of the council's decision can be presumed identical with the resolution presented by its president, the šrk-il (KTU 1.15 = RS 3.343+ ν 18-21): 'rb. špš. lymg19 krt/sbia. špš'10 b'lny/uymlkn [ j ] ^ ( . ) 'In 'When Keret arrives at the western horizon (i.e., dies)|Our lord, at the setting s u n | T h e n will [Ya]ssib rule over us'. Cf. M A R G A L I T 1982, 425; 1995, 252-2.

work as king of Bêt-Hbr (ibid., 11. 22-4) ought to have been greeted with much fanfare and public celebration. 2.2.6.5.2 T h e 'addition' of the episode of filial infidelity—a sin punishable by death in the Bible and which the very name 'Aqhat' (lit., 'the-obedient-one') attests to as heinous in ancient Canaanite society—turns this would-be happy ending on its head at the same time as it brings the story full-circle to tragi-comic conclusion. In K T U 1.14 = R S 2. [003]+ Keret is miserable for want of a son and heir; at the end of K T U 1.16 = R S 3.325+ vi he is miserable for having obtained a son and heir. Nothing more can or need be said. 2.2.7 2.2.7.1 But a word should be said, in conclusion of our critique, on vestiges of the historical approach to the poem still current. 2.2.7.1.1

Parker writes:

I would see the origin of the poem in a story about a king who undertook a campaign against another king to claim the king's daughter as his bride. Negotiations between the king resulted in the ceding of the woman in question, and hence in the marriage of the two and the birth of children. . . . It is this much that forms the most solid basis for those who claim that die poem reflects historical events . . .' ( P A R K E R 1 9 8 9 , 39) 2.2.7.2 This statement, I submit, is as true (or false) of Keret as it is (mutatis mutandis) of Hamlet, Julius Caesar, or Antony and Cleopatra, none of which can be considered 'historical' works reflecting historical events. T h e y are works of the imagination, pure fiction, in which historically attested personal a n d geographical names, scraps of history, social and religious customs are expertly utilized as trappings for the plot and its characters by craftsmen minutely knowledgeable in historical arcana and—all importandy—masters of disingenuity in the service of artistic integrity. 2.2.7.2.1 But even if the poet be inspired by a 'real event'—which in the case of Keret one is entitled to doubt—this determination is no more consequential for understanding the poem and its author, than is the Danish chronicle which inspired Shakespeare's Hamlet. At most, such knowledge can produce some learned footnotes to the text, enhancing its appreciation by cognoscenti but irrelevant and boring for poet and audience alike.

2.3

The story in outline, the message in detail

2.3.1 Although there is no hard evidence to indicate that the poem of Keret was ever the subject of dramatic presentation in a theatre or like setting, it is useful, and certainly not misleading to summarize its contents as if it were. T h e material is most amenable. 2.3.1.1 T h e 'prologue' in the opening lines of K T U 1.14 = RS 2. [003]+, now largely defective, introduced the hero, Keret, as king of Bêt-Hbr, situated '[by the se]a'. T h e king is a man of valour ([gbr. hyl]) and a devotee of El (glm. il) who is his 'patron' (ab); but he is wretched for want of wife and children to fill his 'naked' (crwt) palace. Seven dmes was Keret wed, but each marriage ended abrupdy, for the most part tragically with the death of the spouse; in one case, in childbirth. T h e absence of a (male) heir apparent causes his seven brothers to cast greedy eyes on his throne. Comment: (1) 'Bêt-Hbr by-the-sea' ([gblt. y]rrì) is a pseudonymic riddle to be solved by the audience in the course of the p o e m / p l a y . (2) By presenting the king as a devotee of El—in contrast to the Baalworshipping poet and his audience—the author conveys the message that the story is about a historical figure of long-ago, the era of the ancestors. 22 Keret is thus a 'patriarchal narrative'. T h e members of the hero's clan (lim || umt) are to be found roaming the steppelands between the (Phoenician) coast and the Euphrates ([ f ]
22

For the author of Keret, this 'patriarchal era' began with Ditanu (Ug. dtn, var. ddri) cited obliquely in K T U 1.15 = RS 3.343+ iii 2 - 3 || 13-5) thought to have lived in the early MBA (ca. 2100 BCE; cf. K I T C H E N 1977, 131-42; H F . L T Z E R 1981, 1 10) and developing into an eponymous ancestor. It needs be emphasized however that Keret's Ditanu-ancestry does not make him a direct ancestor of the Ugaritic kings Niqmaddu and Ammittamru ( K T U 1.161 = RS 34.126). Like Abraham, Ditanu is 'the father of many [Amorite] nations'. This much however can be said: the city-state kingdoms which speckled the Phoenician and north-Syrian coastline in the early 2nd millennium BCE were all ruled by classes of Amorite stock. The dynastic houses of Byblos and nearby Ugarit in particular could accordingly have been related by ties of blood a n d / o r marriage, and both of them to clans residing in Bashan (cf. K T U 1.108 = RS 24.252: 23-4). T h e phrase qbs. dtn, roughly 'union of Ditanu' (KTU 1.15 = RS 3.343+ iii 2 - 3 || 13-5), like its parallel rpi. ars (ibid.·, K T U 1.108) denotes the transnational aspect of this consanguinity, the word ars 'land' contrasting with socio-political terms like qrt 'city' and mlk 'kingdom'. (This usage of ars [Heb. 'eres] survives in the Hebrew Bible in the phrase 'am-hā-'āres, denoting an institution of landed gentry who can 'make or break' a king. Cf. provisionally M A R G A L I T 1995, 255-6.)

the El-worshipping, Harran-based patriarchs of the Bible, one of whom complains bitterly that he is without (legitimate) heir and who subsequently sends his trusted servant to obtain a wife for his son in the 'old country'. 2.3.2

Act 1, Scene 1

2.3.2.1 T h e curtain rises on the king about to retire for the night to his sleeping c h a m b e r which he enters shedding tears of self-pity. Curled up in bed in a foetal position, he falls asleep. His patron deity now appears in his dream, having heard the heart-rending sobs of his valiant servant. Wise but not omniscient, El inquires as to the cause of the king's distress. Ever the jester, he speculates whether Keret, dissatisfied with the modest extent of his kingdom, has designs on his own; or perhaps, he wonders aloud, the king is short of money to cover his regal expenses. In reply, the king assures his patron that he wants for nothing material, and that his only wish, and the panacea to his pain, is to sire a family, sons in particular. 2.3.2.2 El is sympathetic; and the remainder of the dream (and scene) is devoted to divine monologue wherein the deity issues a series of detailed instructions as part of an elaborate plan for the hero to realize his ambition. At the centre of the plan is the full mobilization of the kingdom for a military expedition to the (Bashanite) kingdom of Udum(u), to be followed by a siege and ultimatum to its king: surrender your eldest daughter, the fair Hry, to be Keret's wife (or face the consequences). 2.3.3

Scene 2

2.3.3.1 T h e King awakens with a start, but with total recall of the dream which he immediately begins to implement to the letter. He attends first to his person—washing (for cleanliness) and rouging (for war) then to the gods to w h o m he offers sacrifice, and then to the business of war. Comment: (1) El's oneiric thcophany is part and parcel of n o m a d i c / Amurritic religion centering on El and his consort Asherah. El resides in the subterranean fresh-water deep which feeds the palm-trees of the oasis where the n o m a d pitches camp. W h e n the latter retires for the night and sets his sleepy head down to rest, he is lulled to sleep

by the gurgling stream nearby. It is both logical and natural that El, residing close by, should pay him an occasional visit in his dream, especially when the nomad is troubled. (2) Again, the correspondence with the biblical tradition-complex (in its Ε and Ρ versions particularly) is uncanny. T h e El(-Shaddai)-worshipping patriarchs receive their divine visitations in nocturnal dreams. T h e same is true of the Elworshipping Aramean clairvoyant Balaam of the D A P T . Y H W H , by contrast, never appears in a dream to his servant Moses, 23 and there are no dream-theophanies in the Hexateuch outside Genesis and N u m . 2 2 - 4 . (3) T h e characterization of both El and his protégé is parodical. T h e valiant warrior of the prologue is totally deconstructed by the pathetic king crying himself to sleep like a baby. Crying is womanish (2 Sam. 1:24; Lam. 1:2, etc.) and a sign of weakness in men except in well-defined special circumstances. It never occurs to the biblical author to depict A b r a h a m as weeping in Gen. 15; nor does the son-less Dan'el weep when petitioning for a son (1.17). Both are in contrast with the similarly situated H a n n a h (1 Sam. 1:10). T h e proverbially wise El (by dint of age and experience) is parodied by means of the patently ridiculous plan which he concocts: the total mobilization of the kingdom, including the sick, the blind, and the newly-wed normally exempt from military draft, and a strenuous and very expensive seven-days march to the hinterland region of Bashan for no better reason or exigency than to obtain (yet) a(nother) wife for the king. Were this not enough, the poet will subsequently inform us that Pbl, the king of Udum(u), is himself a devotee of El, and his kingdom a 'gift' (ušn) from this same deity ( K T U 1.14 = R S 2. [003]+ vi 12-4). All El need have done was to send an oneiric message to Pbl and the fair princess would have been on her way to Bêt-Hbr. 2 4 This contrasting of exaggerated means utilized for trivial ends is of course a staple of comedy and burlesque (cf. the M a r x

23 Cf. Num. 12:6 8. A later tradition, no longer familiar with the religio-historical presuppositions of the patriarchal faith, attributed this fact to the uniqueness of Moses' prophetic status. 24 H e could also have spared U d u m the pains of siege, and its monarch the pangs of uncertainty, by revealing himself in a dream to Pbl and thereby confirm Keret's ultimatum as indeed inspired and supported by divine degree. But then El would be seen to be truly wise and compassionate rather than the comical dotard intended by the poet.

brothers), a contrast further accentuated here by the disparity between the normally peacable and compassionate El (Itpn. dpid) advising, and devizing, a strategem of war. 25 (4) T h e use of parody at this early stage in the story must be understood as setting the tone for all that ensues. It is the dramaturgic equivalent of Shylock's 'poundof-flesh' bond contracted (ostensibly) 'in a merry sport', and to the over-reaction of foolish king Ahasuerus (Est. 1) to the queen's refusal of a royal summons (itself a parody of a king 'ruling from India to Ethiopia'). It serves notice that the poem of Keret is a species of 'mock epic', perhaps the oldest of its kind in recorded history. 2.3.4

Scene 3

2.3.4.1 T h e a r m y of Bêt-Hbr marches in battle array to U d m (= U d u m u in the land of G a < s h u - > r u [EA 256]). T h e march is broken up into two more or less equal segments: 3 days from BêtH b r to Tyre, where Keret pays an unscheduled (or at the least unanticipated in the dream) visit to the shrine of Asherah, El's wife, where he takes a vow (cf. Gen. 28) that if his mission be successful (one senses clearly the insecurity of this valiant warrior) he will pay to Asherah's shrine 'twice [his bride's] weight in silver, thrice in gold'. 2.3.4.2 T h r e e days later, on the seventh day of the campaign, Keret's army arrives at U d u m and camps outside its walls after having cleared the countryside. T h e r e follow the futile efforts of Pbl, king of U d u m , and his queen N a ' a m a t to relieve the siege, first by offering Keret a bribe of silver, gold, three horses and chariot (with attendant squire), and, simultaneously, sending messengers to offer sacrifice atop nearby Mt. Inbb, the mythological abode of the (war-)goddess Anat. T o no avail; Keret is a d a m a n t (and the gods, by implication, unresponsive): only the surrender of beautiful H r y — in the description of whose (as yet unseen) beauty (he has only El's word for it) the king waxes poetic—will suffice to remove the siege (cf. mutatis mutandis 2 Sam. 20:14-22). T h e scene concludes with Hry taking tearful leave of her family and friends as she sets out for Keret's c a m p and her new life as queen of Bêt-Hbr.

25

O n e may note the uncanny if fortuitous resemblance of El's plan with that of Portia's 'virtuous father' (The Merchant of Venice), mocking the 'holy men [who] at their death have good inspirations'. El's plan is similarly 'inspired'.

Comment: (1) O n the identifications of U d m and Mt. Inbb respectively, cf. M a r g a l i t 1995, 2 2 5 - 4 3 . (2) Although formally a married couple, El and Asherah do not live together (cf. K T U 1.4 = R S 2. [008]+ iv).26 (3) T h e fact that the king's initiative, for all its good intentions, eventually lands him in hot water could—if Keret were a 'serious' piece of literature—be taken as implying the futility of human endeavour and the advisability of resignation to divine will. But if, as I maintain, Keret is tragi-comical, then the crisis precipitated by the king's ill-fated initiative (the result, be it recalled of his absentmindedness) can and should be seen as contributing to his portrayal as a pathetic figure, a 'Schlemiel' or 'Sad-Sack' who can do no right, a master bungler. Keret, like Dan'el, is a 'talker', not a 'doer'. (4) This characterization of the king is underscored by Pbl and Na'amat's reluctance to become Keret's in-laws. After all, such an attitude is not self-evident given Keret's credentials. A king of the backwater kingdom of U d u m would normally have given his eye-teeth for a liaison with the royal house of Bêt-Hbr, alias Byblos. However, Keret's reputation as a matrimonial 'jinx' has preceded him to Udum. 2.3.5

Act II

2.3.5.1 Scene 1. T h e reception celebrating the marriage of Keret and Hry is attended (i.a.) by the gods, including El and Baal. Asherah is conspicuous by her absence. During dinner, Baal prompts El to toast the newly-wed couple. El is glad to oblige: raising his wineglass, his blessing consists of a promise that Keret's wife will bear him multiple offspring (cf. Gen. 15:5, etc.): seven || eight boys and a like n u m b e r of girls. T h e eldest of the boys, to be named Yassib, will be Keret's heir (poetically, he will be nursed by goddesses); the youngest of the girls, 'Octavia', will be El's favourite (bkr, literally, 'first'). Comment: (1) T h e senior gods arrive at the party in pairs; the 'assembly', consisting of the minor (younger) and anonymous gods,

2I

' The separation of El and Asherah on the mythopoeic level is surely a reflection of the transformation of their originally pastoral-nomadic cult following the sedentarization of their worshippers. El is put out to pasture in the Upper Jordan Valley, his domain extending from the foot of Mt Hermon near Dan as far as the Sea of Galilee. But his consort starts up a new career among Tyrians as rbl. alrt. ym 'Lady Asherah-of-the-Sea' where she is presumably worshipped as the patroness of fishermen (cf. her attendants qdl. wamrr described (KTU 1.3 = RS 2.[014]+ vi 10-1; K T U 1.4 = RS 2. [008]+ iv 2-4) as dgy. rbl. alrt. y m 'fishermen of Lady Asherahof-the-Sea'.

arrive in threes. T h e absence of Asherah is accentuated by pairing El with Baal. Anat (here labelled 'Rhmf [Heb. rehem]) is accompanied by the similarly bellicose Reshef. Kathir-and-Hasis, sporting a binomial n a m e , escorts himself (our poet is a 'kibitzer'). (2) T h e choice of R h m y as an alias for Anat is motivated by two considerations: (a) the alliteration with Rip; (b) the synonymity with ( c )Anat, both referring to the female genitalia. 27 2.3.5.2 Scene 2. T h e scene shifts to the (unspecified) domicile of Asherah. Seven years, and several birth-days, have elapsed and Keret's pledge is still unpaid. With the king now in default, Asherah takes her own vow: to make Keret pay . . . with his life! Comment: 'Heaven has no rage . . . nor hell no fury, like a woman scorned'. Cf. also Ps. 50:14, Eccl. 5:3. 2.3.5.3 Scene 3. A party in Keret's h o m e — m o r e precisely, his atr— for the nobility of Bêt-Hbr, its 'Bulls' and 'Stags' in the poet's saucy language. In preparation, the king instructs his queen to 'dress-up like a maiden' (km[.n]crt) by hiding her bulges, doing up her hair, and manicuring her fingernails. T h e ostensible purpose of the party is to 'weep' (ritually) for the ailing Keret. But the secret agenda is political, viz., to confirm the juvenile Yassib as heir and successor to the throne of Bêt-Hbr upon the king's supposedly imminent demise. O n c e this political purpose is made known to the guests at the party, a furious debate erupts, accompanied by shouting and clenched fists, in the course of which the 'president' (irk. il, lit., 'chief member') stands up to speak and pledges the support of the assembly for the young prince. T h e ailing king replies. Invoking the private parts of the president's wife, he blesses him for his support. H e then informs the council that he expects to die within the month, blaming his misfortune on Athirat's abiding hatred for his kingdom. His personal fault he passes over in silence. However, the king's remarks, far from stilling debate, add fuel to its fire, in the course of which both the king and his queen are forced to intervene to restore order. T h e king accuses his opponents of 'drinking his blood', while the queen reprimands her guests for their indecorum as well as for their insin-

27

Cf. D E E M 1978; M A R G A L I T 1995, 241 2. T h e basic meaning of 'nh is 'open up', normally of speech. Its use with sexual activity (cf. Ex. 32:18b) reflects a perceived symmetry between oral and vaginal anatomy (cf. Prov. 30:20).

uation that the king might be feigning illness in order to obtain an endorsement of the crown-prince as successor to the throne. Hry assures the noblemen that Keret's illness is, unfortunately, neither dream nor fantasy; and he has the body sores and fever to prove it! T h e conclusion of the scene is lost, but a political victory for the royal family is a necessary inference. 28 From this moment on, Yassib is heir-apparent in fact as well as in theory, and his enthronement evidently a matter of days. Comment (1) T h e location of the banquet in a tent (hmt) set up in the family atr or burial-ground (cf. K T U 1.17 = RS 2.[004] i-ii)— points to a kispum or mrzfi, i.e., a feast associated with the cult of the ancestral dead (dbh. ilm). This would furnish a convenient pretext for convening the nobles and a suitable occasion for 'beweeping' the sick king. It also is consistent with the all-male guest list as well as the king's instructions to his wife to dress appropriately (cf. K T U 1.4 = R S 2. [008]+ iii 10-22). This banquet is definitely a 'stag affair'. (2) T h e scene implies the existence of a political group whose authorization the king requires to transfer power to his son and thus establish a dynasty. This council of barons is a fcudalistic body whose duties, and prerogatives, are to 'advise and consent'. Neither servile nor rubberstamping, it can make or break a king. A residue of tribal confederation, this group of grandees may be seen as the sociopolitical equivalent and 2nd-millennium precursor of the biblical 'am-hä'äres. (3) It is further implied in this scene that while the legitimacy of the dynastic principle is acknowledged, its cultural roots are shallow indeed. T h e tribal tradition of charismatic leadership, understood in terms of military prowess, is still very much alive, and it goes far towards explaining the fierce opposition to the blankcheque endorsement of the king's son, still very young and untried in battle and leadership. (4) Keret is in all likelihood (portrayed as)

28

T h e spirit and circumstances of Keret's speech are strongly reminiscent of the speech delivered by Hattusi1iš before the pankuš-assembly: 'Behold I have fallen sick. . . . Behold, Mursilis is now my son . . .' It may not be too venturesome to suggest accordingly that the missing portion of Keret's speech may have been formulated in a vein similar to the continuation of Hattusilis' address: 'In the hour when a call to arms goes forth . . . you . . . must be [at hand to help my son]. When three years have elapsed he shall go on a campaign. . . . If you take him (while still a child) with you on a campaign, bring [him| back [safely].' (Translation apud GURNEY

1990,

171).

the first m e m b e r of his family to have occupied a throne, which he may well have seized by overthrowing an incumbent ruler. His rise to p o w e r — a n d this is probably the extent of the story's historicity (which in any case is only presupposed by the narrative)—would have been a model Idrimi who, with a b a n d of ruffians and outlaws, conquered Alalakh and set himself up as king. Like Keret's El-religion and his affiliation with a clan dispersed in the Syrian steppeland, the present scene reflects the political ethos of an earlier epoch, viz., the formative stages of Amorite settlement in Phoenicia and N. Syria and the struggle to establish the legitimacy of dynastic kingship in a society barely weaned from non-hereditary charismatic leadership and tribal organization. 2.3.6

Act III

2.3.6.1 Scene 1. As the scene opens, preparations are underway for Keret's funeral (although the king is still quite alive). T h e sound of caterwauling w o m e n — t h e poet prefers the comparison with howling dogs and coyotes—fills the royal mansion. Overcome emotionally by these depressing sounds and by the realization which they spur of his father's imminent demise, the loving and devoted son Ilhu approaches the king's bedside. With tears rolling down his pubescent cheeks, he queries his father in disbelief (in the process giving expression to the current ideology of divine kingship in Canaan): 'Is Keret, the divine offspring of El and Athirat, not immortal?! D o gods die?!' 2.3.6.1.1 T h e compassionate Keret responds with words of comfort to his distraught son; and by way of occupational therapy counsels him to set out on a mission to sister Octavia, residing elsewhere, and to bring her home. T o spare her sensitive feelings, the pretext is to be an invitation to a family feast rather than a funeral. (But since Keret's funeral will doubtless be followed by a lavish wake— cf. K T U 1.6 = R S 2. [009]+ i 18-31—the lie is truly lily-white.) Ilhu obediently complies and takes his leave. Comment: T h e text at this point is in disarray, resulting in the dism e m b e r m e n t and dislocation of Ilhu's speech. T h e awareness of this disturbance by a subsequent copyist led to its rewriting; but the corrected version unfortunately found its way into the second column of K T U 1.16 = R S 3.325+, causing yet another disturbance. (2) W e are not informed here either as to the reason for Octavia's residence away from home or its location. If this information was not forth-

coming in a no longer extant part of the preceding text (e.g., at the end of K T U 1.15 = RS 3.343+ iii), the poet left it to the audience to fathom the answers from the continuation. 2.3.6.2 Scene 2. After climbing a mountain and praying there (to Baal!) for a safe journey, Ilhu sets out. Arriving at Octavia's convent-residence (hmh<m>), he squats on a nearby hillock, with the gateway-entrance in view, to await his sister. As she emerges from the gate, on her way to fetch water, Octavia espies her brother. In her excitement, she drops (more likely, hurls) the encumbering vessel in her hand and makes a dash to embrace Ilhu, her head now nestling tenderly in his shoulder. But it does not take long for Octavia to collect h e r s e l f a n d to sense that brother Ilhu has not just droppedby for a chat. Her female intuition tells her that something is awry, possibly relating to her father. T o her query, 'is father ill?' Ilhu replies that, of course, all is well with the king, and that he has come to invite her to a family party. Sensing her incredulity, Ilhu unfolds his cover-story with lavish improvisation: it will be a sumptuous affair, attended by the gods and the who's-who of Bêt-Hbr; musical entertainment will be provided by nubile lasses singing songs 'to set one on fire'. He himself, Ilhu continues in his prevaricating best, was asked by the king to go out and trap birds as delicacies for mother Hry and brother Yassib; 'and since father knows that my hunting would take me near where you live, he thought it might be nice to extend you an invitation. So here I am!'. 2.3.6.2.1 Octavia's response leaves no doubt as to her incredulity. First she asks her brother to pour a cup of wine from his portable jug. After removing the plug, Ilhu obliges. She now turns to her brother and, in a tone mixing hurt pride with barely concealed anxiety, she asks: 'Why do you make a fool of me? H o w long has father been ill?' 2.3.6.2.2 Faced with such precocity, Ilhu breaks down and proceeds to tell his sister the sorry truth. Upon hearing this, the heartbroken Octavia cries out and shudders (V hi); she commences a funereal song-and-dance around her brother. She then repairs with him to the parental home. 2.3.6.2.3 Upon arrival, Octavia enters, silently and abashcdly, her father's bed-chamber. Approaching his bed, she kisses him affectionately

on his feverish forehead. She is his little 'blossom' (ib). She leaves shortly thereafter, heeding her father's request to climb a mountain and pray there for his recovery. 29 Comment: (1) T h o u g h not expressly stated, it is a reasonable inference from this portion of the narrative that Octavia has become a nadītu-priestess, or nun, residing in a cloister and in the service of the sun-goddess, Shapsh. A m o n g other things, this hypothesis will explain (a) why Octavia is residing away from her parental home; (b) why Ilhu does not even consider entering the gateway to notify his sister of his arrival. As a female retreat, it is presumably off-limits to men. (2) Octavia's funereal song-and-dance, encircling her brother, is described in language similar to that used by Ilhu to describe the wailing-women in Keret's house. This may help to explain the intrusion here of extraneous material originating as a (corrected) version of Ilhu's plaintive speech to his father. (3) It is typical of Ugaritic epic literature to portray w o m e n as superior in intellect a n d / o r courage to men; 30 and the present encounter of brother and sister is certainly no exception. El compares unfavourably with Athirat, Dan'el with his daughter Pughat, and Ilhu with Octavia. 31 2.3.7 2.3.7.1 Scene 3. T h e text of this scene is very fragmentary, and its contents consequently are obscure. T h e king's illness, like Aqhat's murder, has resulted in drought, and the stocks of grain, wine, and oil are depleted. A set of obscure ritual acts, intended presumably to induce rainfall, is followed by a delegation of farmers to the king, presumably (since the continuation is lost) to apprise him of the situation and to ask for help.

29

A considerable part of the text summarized above is missing, and the summation at certain points presupposes the correctness of the restorations. Cf. the discussion in M A R G A L I T 1 9 9 5 , 2 6 4 - 8 9 for this and other matters relating to this passage. 30 It is also not uncommon in O.T. literature: Adam is clearly inferior in intellect to wife Eve (which is why the 'wily' snake takes her on first). The same holds true for Isaac and Rebekkah, Barak and Deborah, Sisera and Jael, Haman and Esther, etc. A notable exception is David and Michal. The latter is possibly the most 'trag(ed)ic' figure in the entire Bible: bright, beautiful, and courageous—and an habitual 'loser'. 31 Anat is only seemingly an exception: for while nominally female, she acts and dresses like a (violent) man, and is therefore the villain of Aqhat. She contrasts both with her virtuous brother Baal and the heroine Pughat; and the poet does not stop short of ridiculing the penis-envy of this self-hating goddess by depriving her of the coveted bow once acquired (KTU 1 . 1 9 = RS 3 . 3 2 2 + i).

Comment: W e have here another expression of the 'ideology of divine kingship': the illness of the king induces a paralysis of Nature. However, there is no more reason here than in the previous instance to assume that this ideological stance reflects the authorial point-ofview. Like the portrait of the sick hero, and (shortly) the inept gods, this ideology is also subject to satirization, as if what is true of the great Baal ( K T U 1.5 = RS 2.[022]+ ii 5-7) is true of the pathetic king of Bêt-Hbr. 2.3.7.2 Scene 4. T h e desperate situation created by the king's illness sets the stage for a curious development: the artisan god, Kathirwa-Hasis, whose wisdom (say the gods) is second only to El's, is approached by a delegation of the divine assembly—'El's sons' ((dt. bn. il)—and asked to take an urgent message to a hitherto and otherwise unknown character named lis and his (characteristically unnamed) wife, bearing the title ngr(t) of the H o u s e / T e m p l e of El, (var. Baal). In the message promptly delivered by the hobbler Kathir-waHasis—described by the poet as running with the grace of an ass— lis is instructed to go up to the tower and to shout at the top of his lungs—'like a waterfall || like a bull' to the inhabitants of the city. T h e sequel is lost, and with it presumably the statement of purpose, viz., a call to prayer and supplication on behalf of the dying king and the drought-imperiled kingdom. Comment T h e present scene, if correctly interpreted, brings the satirical tone of the poem strongly to the fore; indeed, the satire comes close to becoming farce. El is in deep trouble: his plan for his protégé has miscarried, and the protégé himself and his famished kingdom are teetering on the brink of disaster. El's distress signal (which one may presume to have been lost in the lacuna at the beginning of K T U 1.16 = RS 3.325+ iv) has been picked up by the assembly of his sons which now goes into (pathetically ineffectual) action. In their infinite divine wisdom they turn to the wise but crippled Kathir to deliver an urgent message to the temple crier (ngr) or mu'addin32—and for good measure, to his wife—to summon the 32

The semantic correspondence of Ug. ngr and the Arabic mu'addin (< 'dn (II) 'cause-to-hear, announce') is very close indeed. T h e Ugaritic term is cognate with Akk. nagaru 'Ausrufer, Herold' (.ΑΗιν, 711). The translation 'herald' in the present context is, however, misleading in that it implies a palace functionary charged with making public pronouncements. Ils is rather a temple functionary; and since his j o b is to summon the faithful to prayer, he is necessarily (a) mortal, and (b) a BronzeAge Canaanite precursor of the Islamic mu'addin.

faithful to prayer and supplication, without which the gods can do nothing to save Keret and his kingdom. From here it is surely but a small step to the conclusion that the 'sons of EP (notably excluding heroic and virtuous Baal) can do nothing because they are good for nothing. 2.3.7.3 Scene 5. T h e race is on to save Keret's life, for which purpose El has convened the divine assembly in emergency session. El arrives accompanied by wife Athirat. H e opens the session with a plea to his consort to spare Keret's life. Asherah replies that the king's life 'is in the hands of his wife Hry': she (now that the king is incapacitated) has to pay Athirat 'twice her weight in silver, thrice in gold' (with the interest waived) if Keret is to be cured. 2.3.7.3.1 Even El cannot apparently raise such a sum, much less Keret, whose kingdom is down at the heels. T h e 'father of m a n ' turns accordingly to his sons for help; but they remain deafeningly silent, either because they are cowed by the presence of their mother or they are simply at a loss for ideas. El is now left with no choice but to deal personally with the problem. 2.3.7.3.2 His solution: to create a female exorcist named Š'tqt (< etq '(cause to) remove'), the details of which are obscured by the tablet's poor state of preservation. But they are not beyond recovery. El creates his creature, in primordial fashion, from (red) clay (cf. J o b 33:6), inserts snake-poison into her vagina (to thwart reproduction?), christens her over a cup of sanctified wine, and then brings her to life by 'pouring' into her mouth '[the soul] of a god and the blood of a [human]'. Fortified by El's blessing and directed by his detailed instructions, Sctqt flies off to Bêt-Hbr. Comment: (1) T h e divine assembly, as we learn from K T U 1.15 = R S 3.343+ ii, is m a d e up of the minor deities consisting of the sons—not including daughters—of El. It corresponds, one may assume, to the 'council of princes' reflected in 1 Kgs. 12:8ff. T h e presence of Athirat is thus a breach of protocol warranted by the emergency as well as by the divine mother's direct responsibility for creating it. (2) Athirat's offer at first sight seems to hold the key to a satisfactory resolution of the crisis. However, it must be remembered that after fifteen births and nursing periods, the once streamlined Hry is now bulging at both waist and bosom (cf. above, § 2.3.5.3 [= K T U

1.15 = R S 3.343+ iv 10-3]). (3) T h e echoes of cosmogony in this scene are part of the parody. Like the exaggerated mobilization in K T U 1.14 = RS 2.[003]+, the creation of S'tqt is a case of mockheroic 'overkill'. W h a t S'tqt does for Keret could, and would, have been done by any run-of-the-mill exorcist in Bêt-Hbr. El, says the poet, may be very wise (who else knows how to create life?); but he is sorely lacking in c o m m o n sense. 2.3.7.4 Scene 6 describes how S'tqt saved Keret's life. U p o n arrival, she immediately gets down to business. She fastens a string to the patient's navel and places a wreath of medicinal leaves on his feverish head, while repeatedly wiping his brow of sweat. She then opens his mouth and forces him to eat—we are not told w h a t — a n d presto! the king is well. A rejuvenated Keret promptly orders wife H r y to slaughter a fading lamb for dinner to celebrate his recovery. Comment T h e poet's feminist bias finds expression here once again. Just as the w o m a n Athirat foils the male El's plan, so the female S'tqt saves the day (while foiling the foiler). But woman's superior intelligence, courage, and enterprise are, alas, not enough to alter the existential 'female condition': S'tqt, like Athirat, is subordinate and subject to El. T h e 'tree' of Athirat cannot exist without El's fertilizing water, and S'tqt will die, without fruit, once her mission is accomplished, as will Pughat, whose heroism is recounted in a tale n a m e d for her b r o t h e r (in marked contrast to the 'book of Esther'). 2.3.7.5 Scene 7. T h e crown-prince and heir-apparent, whose birth, heralded by the gods, fulfilled his father's most fervent wish, makes his first—and last—appearance in this scene, as if to say: if you meet him once, it is enough for a lifetime. Obviously displeased and disappointed by his father's dramatic recovery—Yassib, be it recalled, is still an adolescent!—he decides to act. Encouraged by 'the fiend at his elbow', he enters the throne-room where his father is seated and orders him to step down, on grounds of incompetence in the discharge of his royal duties. Keret's reaction is a mixture of rage and anguish; and as the curtain falls there resounds a curse, at once fearsome and funny, called down by the king on the head of his perfidious son. Comment: (1) Keret's recovery obviated a tragic end to the story; but the finale is its tragi-comic equivalent. For upon hearing his son's

words, the king might well have wished he were dead. (2) Yassib is 'instructed' (V wsr) by his jinn (Ug. ggrì)—he is a school-boy listening to the wrong teacher. (3) T h e charges laid by Yassib against his father are instructive for the insight they provide to the Canaanite view of kingship. T h e most important task of the king is not to lead in battle (which is precisely what Keret does in K T U 1.14 = RS 2.[003]+) but to administer justice fairly and compassionately (something he is never seen to do). 33 T h e Canaanite king is first and foremost a 'judge', in contrast to his Amorite counterpart whose claim is based on personal charisma as a warrior proven in batde (gibbôr hayyil). In this sociological sense, Aqhat is older than Keret, in that the former describes its (male) heroes, young and old, in terms derived from the military lexicon. Keret is a mlk,H Dan'el a gzr. (4) In addition to ferocity and hilarity, the curse also contains the most important clues, suitably and cleverly embroidered into the finale, to the identity of pseudonymic Bêt-Hbr, lit., 'House-of-Union'. 3 5 T h e king calls on 'Astarte-name-of-Baal'—i.e., Baalat—and Yassib's dislodged teeth are to fall out 'altogether', for which the poet chooses the rare (b)gbl (Palmyrene-Aramaic gbl — 'community', M H e b . gbl 'to mixtogether (as porridge)', which plays on the original form of 'Byblos', i.e. G B L / G u b l a (Heb. Gëbaï).

2.4

The moral of the story (in sum)

2.4.1 T h e moral of the story is clear, a proud and praiseworthy testimonial to the venerable Stoic tradition c o m m a n d i n g the allegiance of the wise throughout the ages:

33 Contrast the description of the (non-royal!) judge Dan'el! For all their impudence, Yassib's words thus contain a germ of truth. This motif of 'truth from the mouth of babes' is especially prominent in Aqhat (cf. M A R G A L I T 1989, passim). Noteworthy too is Absalom, like Yassib motivated by a desire to depose his father, who sets himself up as a judge in the gateway, intercepting his father's 'clients', in order to establish his credentials for kingship. 34 As well as £ (Heb. sô'a) a title which he (presumably) shares with the other members of the Bêt-Hbr nobility. 35 Cf. Akk. /)ibru(m), a Canaanite loanword denoting 'clan' (OB) and 'in gathering (of fruit)' (LB)—cf. AHw, 344. In the 11th cent. Egyptian Wen-Amon story, f}-b-r denotes a joint commercial venture (AJVET 27, n. 17). ^l(fbr and ΛIgbl are thus fully synonymous terms.

2.4.1.1 G r a n d the plans of gods and man, But when the day is done— Bones broadly scattered dry in the sun, For ironic Moira the fray hath won. And nought remains for Apollo's progeny, But to sing her praise In comic agony. 2.4.2 'Life', not 'kingship', stands at the centre of Keret as it does in Aqhat and Baal-Mot. But whereas Baal-Mot focuses on the uncanny dialectic of Life and Death on the (awesome) cosmic plane, Aqhat and Keret focus on the (absurd) human-life condition, the former on its tragic aspect, the latter on the tragi-comic ('if it weren't so funny, it would be sad . . .'). If Aqhat is a C a n a a n i t e Hamlet, Keret is a Canaanite Merchant of Venice. Like his great English counterpart, the Canaanite bard is a master at manipulating emotion; but to mistake him for a 'politician' (or a 'preacher'), and his art for propaganda (or a sermon), is at once an insult and a betrayal. 2.4.2.1 O u r poet- indeed any poet (of integrity)—writes (or sings) for an audience which is 'free', not 'captive'. He is by nature the foe of tyranny, be it of the body or of the mind. He is anathema equally in Plato's Republic and in Augustine's 'City of God'. He kneels (only) in the Temple of Moira, at the feet of Apollo. 36

36 What J. H U I Z I N G A (Homo loudens) has said of 'play' is equally true of writing poetry: 'all play is a voluntary activity. Play to order is no longer play; it could at best be a forcible imitation of it. By this quality of freedom alone, play marks itself off from the course of the natural process. It is something added thereto and spread out over it like a flowering, an o r n a m e n t , a g a r m e n t . ' (Beacon ed., 1955, 7). Elsewhere (ibid., 132) he rightly observes that poetry as such is a form of play.

3

The

Story

of

Aqhat

Nicolas

3.1

(KTU

1.17-19)

W y a t t

Introduction

Tablets R S 2.[004], 3.340 and 3.322+349+366, discovered in the 'High Priest's House' on the acropolis at Ras Shamra-Ugarit in 1930 and 1931, 1 were quickly established as constituting the same literary work.- T h e mention of Danel by name in another third-season find from the same location, R S 3.348 (IV D = 1 R p = U T 121 = C T A 20 = K T U 1.20), led to the initial incorporation of this tablet in the series, but its successor Rpum tablets were never thus regarded, and for practical purposes it too was eventually discarded from the sequence. 3 N o authorship is mentioned on any of the tablets. However, the lower edge below K T U 1.17 vi reads [ ]prln, (KTU2 prln) and is generally restored on the basis of K T U 1.6 = RS 2. [009]+ vi 5 4 - 5 as [spr.ilmlk.šbny.lmd.atn.]prln, thus restoring the n a m e of Ilimilku, the scribe to w h o m K T U 1.1-6 = R S 3.361, 3.367, 3.346, 2.[014]+, 2. [008]+, 2. [022]+, 2. [009]+ and K T U 1.14-16 = R S 2. [003]+, 3.343+, 3.325+ are attributed (with colopha at 1.4 viii lower edge, 1.6 vi 5 4 - 8 and 1.16 vi 59 lower edge). RS 92.20 1 6 4 (as yet unpublished) also apparently bears the name of Ilimilku. In the case of the published tablets, the script is similar in all the tablets at-

1 See B O R D R E U I L - P A R D E E ( 1 9 8 9 , systems for the texts are as follows:

26, 30-32).

The most widely used numbering DIETRICH -

RS

VIROLLEAUD

GORDON

HERDNER

2. [004]

2D

3.340

3 D

3.322+

1

2 Aqht 3 Aqht 1 Aqht

C T A 17 C T A 18 C T A 19

D

LORETZ -

SANMARTIN

K T U 1.17 K T U 1.18 K T U 1.19

T h e tablets are located as follows: K T U 1.17, 19 in the Louvre (AO 17. 324 and A O 17.323 respectively), K T U 1.18 in the British Museum (AO 17.325 = BM L84). 2 Published by V I R O L L E A U D 1936a (Editio Princeps). 3 Typical expressions are 'clearly at least one further tablet must have followed': G I B S O N 1 9 7 5 , 6 6 . 'At least four tablets': DE M O O R 1 9 8 7 , 2 2 4 . P A R K E R 1 9 9 2 , 9 9 , 1 3 4 - 5 , still evidently had a fourth tablet in mind, but refrained from identifying it with K T U 1 . 2 0 = RS 3 . 3 4 8 . See also P A R K E R 1 9 9 7 , 4 9 . 4 Provisionally K T U 9.432. See C A Ç U O T 1992, B O R D R E U I L 1995a, 2.

tributed to Ilimilku, so that the identification is reasonable. 5 As will be clear from the synopsis below, considerable portions of Aqhat are missing. K T U 1.17 is a tablet originally of six columns, of which two are entirely missing, while the beginning and end of the four surviving columns are missing (an estimated twelve or so lines in all on the recto, and on the verso, with a shearing break down the upper surface, perhaps twenty or so), with the wedgeshaped breaks on columns i and vi resulting in even further loss. K T U 1.18 is a tablet of four columns originally, of which two are entirely missing. Again, the beginning a n d end of the surviving columns are missing (some twenty-five lines or so), and wedge-shaped breaks further reduce the surviving text, with no complete lines in col. i, and only nine complete in col. iv. K T U 1.19 is the best preserved of the three; the only substantial losses here result from friable edges along the three sections into which the tablet has broken, with the additional loss of a central section in col. i, the bottom right hand corner of col. ii, the top corner of col. iii, and a small vertical section in the lower part of col. iv. T h e surface is also eroded at the beginning of col. i and in the upper central section of col. iv. At a rough estimate, from these three tablets slightly over half the lines are missing, say approximately 430 from an original 840 (fourteen columns of ca 60 lines each), or fifty-one percent. While these figures are not set in stone (or clay), any adequate interpretation of the text must remain provisional, having to take into account the fact that it can never tell more than half the story.

3.2

Synopsis of the story

Substantial portions of the text are missing, as we have noted. T h e following narrative sequence can however be understood. 17

5

i

Danel has no son and so performs devotions to obtain divine assistance. For six days he sacrifices to the gods, apparently spending every night in the temple. O n the seventh day Baal intercedes for him, asking El to provide a son who will perform all the filial duties necessary for a father to be blessed. El blesses Danel and promises . . .

For recent discussion on the programme of Ilimilku see W Y A T T 1997, 1998a and below in this chapter. I also enlarge on his significance at § 13.4.2 below.

ii

iii iv ν vi

18

i

ii iii iv

19

i

ii

iii

a son. T h e son is probably born in the missing section between the columns, and news is brought to Danel, who rejoices and holds a feast in his palace for the goddesses of childbirth. missing missing Kothar arrives, bringing a composite bow as a gift. He is feasted/' and the bow is given to A q h a t . . . A feast is apparently taking place. Anat asks Aqhat to give her the bow. H e tells her to take the raw materials to Kothar: he will make her one. She persists in her demand, and offers him immortality. Aqhat tells her to stop lying, and says that a bow is in any case a man's weapon. She departs to El in a rage, accusing Aqhat of impiety. Anat threatens El that she will be violent if she does not get her way, and he gives her a free hand. She approaches Aqhat, seemingly mollified, inviting him to go hunting with her. missing missing After a successful hunt, Anat summons Yatipan, instructing him to assume the form of a falcon, and pounce on Aqhat, killing him. H e does so . . . and the bow falls into the river and is shattered. Anat mingles regret for the bow with a savage assault on Aqhat's corpse, which she tears in pieces. Back in his capital, unaware of what has happened, Danel sits to dispense justice; Pughat sees the withering of the plants, and sensing a disaster, rips Danel's cloak. H e utters a curse. Danel, still evidently not appreciating the situation, wishes that his son would harvest the now shrivelling grain. T w o messengers arrive and tell of Anat's complicity. D a n e l . . . then curses the falcons. As each falls from the sky in turn he examines its entrails. Those falcons found to be empty are healed. Finally he finds Aqhat's remains in Sumul's stomach, and buries him. H e then goes round the country,

6 This is the folk-tale idiom for worship with sacrifices, but uses the figure of face to face, person to person, communication between deity and devotee.

iv

cursing all the villages in the vicinity of the murder. Aqhat is mourned for seven years, Danel concluding the rites with a sacrifice. Pughat then dons a r m o u r beneath her woman's clothes, and sets off to find Yatipan. Already half drunk, and thinking that she is Anat, he demands wine, and while she plies him with it boasts of his exploit. . J

As can be seen from this synopsis, there are tantalizing gaps in the narrative. Particularly u n c l e a r — a n d consequently open to variations in reconstructive guesswork—is the sequence of events in K TU 1.18. T h e synopsis above represents this author's reading of the story. Again, the last column of K T U 1.19 breaks off at the most inopportune moment. It is commonly supposed that Pughat went on to kill Yatipan, which would provide a relatively satisfying dénouement, but would leave Anat, the true villain of the piece, unscathed. O n the other hand, as pointed out below, 8 this is to apply our moral expectations to a divine power, and may misread the author's intention. W h e t h e r or not K T U 1.20-22 = R S 3.348, 2.[019], 2.[024] have any close relationship with Aqhat must remain an open question. But while they are narrative in form (and they appear to be three versions of substantially the same narrative, though Pitard opines below—§ 6.4—that tablets K T U 1.21 and 1.22 = RS 2. [019], 2. [024] may be two parts of one whole), we cannot assume that the mention of Danel proves a link, since a n u m b e r of stories may have been attached to the same figure.

7

T h e following translations have been published: V I R O L L E A U D 1936a, G A S T E R 1936, 1937, 1938. 1950, 257-313, 1961, 316-76, C . H . G O R D O N 1949, 84 103, 1977, 9-29, FRONZAROLI 1955a, D R I V E R 1956, 48-67, J I R K U 1962, 115-36, AISTLEITNER 1964, 65-82, G I N S B E R G 1969, 149-55, C A Ç U O T - S Z N Y C E R 1974, 401 58, C L E A R 1976, 5 0 - 6 9 . X E L L A 1976, 1982, 193-216, G I B S O N 1978, 103-22, C O O G A N 1978, 27-47, M A R G A I . I T 1989a, DEL O L M O I J Î T E 1981a, 327-401, DE M O O R 1987, 224-66, A I T K E N 1990, BALDACCI 1996, 333-65, P A R D E E 1997a, 343-56, PARKER 1997, 49-80, W Y A T T 1998c, 245-312. Other studies include CASSUTO 1938, BARTON 1940, STOCKS 1943, GINSBERG 1945a, 1945b, O B E R M A N N 1946, H E R D N E R 1949b, G R A Y 1957, 73-91 = 1965, 106-26, EISSFELDT 1966, K O C H 1967, K A P E L R U D 1969, 70-82, H I L L E R S 1973, DIJKSTRA DE M O O R

1976,

1975,

DRESSLER

1975,

1979,

1983,

GIBSON

1 9 7 5 , WATSON

1976,

ΧΕΙ.Ι,Α

1979, M A R G A I . I T 1981a, 1983a, 1983b, 1984a, 1984b, 1989a, DEL O L M O I J Î T E 1984a, 115-42, CAÇHJOT 1985, 1987, 1990, PARKER 1987, 1989, 99-144, C O O P E R 1988, DE M O O R 1988a, A I T K E N 1989a, 1989b, 1990, HUSSER 1995, 1996. For a fuller bibliography down to 1988 see M A R G A L I T 1989a, 503-6. R See § 13.3.2.1. DIJKSTRA

3.3

History of interpretation*

published the editio púnceps of the Aqhat tablets. H e accepted Ilimilku's authorship of the present narrative, whatever their antecedents, 1 0 and dated the tablets to the mid-fourteenth century ( 1 9 3 6 , 8 2 ) . O n genre he was imprecise, referring to 'legends' ( 1 9 3 6 , 8 3 ) and to 'mythological texts' ( 1 9 3 6 , 8 5 ) , or even both together ( 1 9 3 6 , 1 0 9 ) without demur. By ordering them as he did (n. 1) he inevitably missed the logical progression which was subsequently recognized. M a n y of his explanations of the vocabulary have had to be revised, but his study is not to be underestimated as important pioneering work, however m u c h may now be of primarily historical interest. An interesting instance is his discussion ( 1 9 3 6 , 8 7 - 9 6 ) of the names of the characters. H e concluded (p. 96) that Aqhat is 'a member of the family of the god of death; he is, in effect, one of the gods who dies . . .'; he resumed this argument later (p. 110), claiming that Mot 'personifies the ripe ear of wheat', with the result that Aqhat is also supposed to be 'the harvest-genius'. T h e r e was an unfortunate tendency to draw conclusions of this kind in early Ugaritic scholarship, which it took decades to escape. Everything was allegorical! T h e result was the wholesale application of various permutations of the seasonal interpretation (a variant on the myth-and-ritual theme) to all the larger compositions, with a consequent delay in the recognition of more balanced assessments. Virolleaud also understood the terms qst ('bow') and qs't ('arrows') to mean 'chalice' and 'vases' respectively," which rather destroyed the symbolic centrepiece of the whole story. Virolleaud

(1936)

in a brief note on the text, placed the tablets in what is now regarded as the correct order, and recognized Aqhat's h u m a n nature. B a r t o n 1 9 4 0 drew attention to the apparent links Cassuto

9

1938,

There is insufficient space to offer a complete survey here, and much of the earlier discussion is in any case now outmoded. I shall therefore merely oudine one or two salient features of early discussion, and concentrate on later work. For literature before their respective publication dates see also the surveys in CAQUOT SZNYCER - HERDNER 1974, 401-15, DEL OLMO LETE 1981a, 327-401, and MARGALIT 1989a, 3-92. 1(1 He writes of them being 'redacted in the fourteenth century' and of a considerable lapse of time between their original formation and reduction to writing by Ilimilku (VIROLLEAUD 1936a, 83). This dating is now in course of modification. See PARDEE 1997c, 376 n. 2, and below, § 13, nn. 284, 289, 311. 11 VIROLLEAUD 1936a, 117, 203-5.

of the narrative with the Galilee region, 12 while G i n s b e r g 1945a, 1945b, recognized that Danel was a king, and in his detailed treatment of a n u m b e r of key passages broadly set Aqhat studies in their present mode. Gaster developed his views through a n u m b e r of articles and two editions of Thespis ( G a s t e r 1936, 1937, 1938, 1950, 257-313, 1961 [1966 printing], 316-76). We may take his final account as his considered view. H e treated Aqhat as myth, and classified it as 'the disappearing god type'. It is a purely literary work as it stands, but with its roots in ritual drama: 'it was, au fond, nothing but an artistic transformation of the time-honored seasonal drama'. 1 3 After offering a synopsis of the narrative, in which the reader may feel uncomfortably that he is being led more by rhetoric than by hard facts, he launched into his interpretation (1961 [1966], 320-7). 'If our basic approach is correct,' he averred, 'this story will go back to a primitive seasonal myth relating how a mortal huntsman challenged the supremacy of the goddess of the chase and how his subsequent execution for this impiety caused infertility upon earth.' H e went on to invoke T a m m u z , Osiris, Adonis 'and the like'. We can see the patterning process at work. Only the prior assumption of some seasonal theory (probably also misrepresenting fundamental elements in these traditions too) could justify a parallel treatment. More substantial, however, was his invocation of the Orion myth. This is indeed a widespread tale (in my view possibly quite independent of T a m m u z and company), of the hunter who confronts and insults a goddess or is in some way brought to her attention. 14 Unfortunately, Orion is the subject of a large n u m b e r of myths, and not one of them corresponds very closely to the plot of Aqhat. It requires a synoptic approach to discern any extensive parallels between what are essentially variations on a theme. Furthermore, Orion is inseparable from the constellation of the same name, while Aqhat

12 O n e of the grounds for considering that the rulers of Ugarit were originally from the Hauran-Galilee region lies in the familiarity of the tradition with the toponyms of the region. See discussion, with further references, in M A R G A L I T 1989a,

14-7. 13

GASTER

1961 (1966), 316.

See also G R U P P E 1 9 0 6 , i 6 9 - 7 0 (cited Gaster), i 1 5 1 - 4 ( 4 1 ) , and A S T O U R 1 9 6 7 , 1 6 3 - 7 5 (discussed below). Graves' explanation of Aqhat ( 1 9 6 0 , i 1 5 3 - 4 ) , apart from calling it a Hittite (!) myth, interprets it astronomically. 14

GASTER

1961

(1966)

320-6.

FONTENROSE 1 9 8 1 , G R A V E S 1 9 6 0 ,

has no obvious links with the stars; 15 and while it would be nice to find them, we should beware of assuming them on the basis of 'parallels' some centuries younger. But Gaster's work on this aspect deserves more extensive re-evaluation as our knowledge of the stellar dimension to U g a n d a n religion, now known only fragmentarily, develops further. But even establishing a stellar basis does nothing for the explication of a text from which any such putative elements are now clearly missing. 16 D r i v e r (1956, 8) gave only a slight treatment of the significance of the story. H e stated that the theme of Aqhat 'is a righteous king's need of a son', but a couple of paragraphs later wrote that 'the main theme of the myth is clearly the death and resurrection of Aqhat', thus introducing a new perception of what concerned the narrator, before concluding that in view of the damaged condition of the material 'no satisfactory interpretation of the myth is possible'! This final assessment is certainly the most cautious. But it should be noted that Driver raised two interesting issues: the problem of whether Danel was a king, and the death and resurrection motif.

(1967, 163-75) referred to Gaster's treatment of Aqhat in relation to Orion, but, without discounting it and noting its Mesopotamian antecedents, argued that a much closer figure for fruitful comparison is the Greek Actaeon. H e noted that Actaeon's mother Autonoë was daughter of C a d m u s and H a r m o n i a , thus evincing a Semitic pedigree, since C a d m u s and the whole Boeotian tradition reflect West Semitic influence. H e argued (p. 165) that the two names, Aqhat and Actaeon, are related, and that not only are both torn asunder, one by 'eagles' (rather falcons), 17 the other by dogs (falcons and dogs are animals in the service of hunters), but (p. 167) that in both stories there is a seasonal element, the fifty hounds of Actaeon representing the cycle of the year, while Aqhat's death provokes a severe drought. Like Gaster, he went on to suggest a stellar element behind the Orion parallels (p. 168), and noted that the latter's name, too, is susceptible of a Semitic etymology (y'Sr, 'ûr) and may even appear in the form aryn as an Ugaritic personal name. 1 8 Astour

|R> G A S T E R ' S (1961 [1966], 322) linking of the bow with the constellation of Canis major is certainly intriguing! "' T h e only hint at a richer background is Pughat's epithetal yd't hlk kbkbm, 'who know(s) the courses of the stars', K T U 1.19 = RS 3.322+ ii 2 - 3 etc.

17

WYATT 1998C, 2 8 4 a n d n .

18

Cf.

GRONDAHI.

151.

1967, 27, 220,

365.

'Hurrian?': DLU,

54.

He further argued for a link between Sumul {ml), 'mother of the eagles (falcons)' and Greek Semele. 19 C a q u o t - S z n y c e r ( 1 9 7 4 , 4 0 9 ) drew attention to the incomplete state of the text and advised caution in interpretation. Critical of mythological and seasonal interpretations, they insisted (p. 413) that the story was neither history, not historical epic, nor myth, and was not the subject of seasonal, national or exceptional recital. T h e y found a definition of its genre elusive, but compared it with Gilgamesh, Adapa and Etana, ragarding it as an Ugaritic 'classic' (conveniently vague!). They did not use the term, but to judge from their treatment (p. 414) appear to have assessed it as wisdom literature. 20 G i b s o n ( 1 9 7 5 ) set out to clarify thinking on the nature of myth and other genres. Noting that one of K i r k ' s ( 1 9 7 1 , 2 6 8 ) features of myth was the fantastic dimension, he noted such features in Aqhat (and Keret), but added that 'a speculative or perhaps better, an ideological bias' (p. 62) should be present for a narrative to qualify as myth. But he denied any link between the present narrative and Ugaritian royal ideology. T h e scenes such as the confrontation between Anat and Aqhat, which he considered to have an ideological dimension, he regarded (p. 67) as 'secondary, supplying for all their vigour only the backcloth against which Daniel's piety is put to the test'. H e further opined that the bow too was a secondary feature, and that in a putative fourth tablet Aqhat was finally restored to life.

dealt briefly with Aqhat in the broader context of ancient Near Eastern tales. Her treatment was too cursory to contribute much to the discussion, but she served the useful purpose of highlighting the conventional folklore motifs to be found in the story.- 1 In his edition of the texts, d e l O l m o L e t e 1 9 8 1 offered an extensive analysis of Aqhat. H e classified it as 'epic', along with Keret, and ran through the scenes, analyzing the literary sub-type of each episode, drawing on parallels in biblical and other ancient Near Eastern literatures. In reverting to a general statement of the text's 'sense and function' (pp. 3 5 4 - 6 4 ) , he reiterated its epic nature, judging it however Irvin

19

(1978,

ASTOUR

76-78)

read the

D N

sml at

K T U

1.39

=

RS

1.001.14,

but this is now dis-

counted. 20 M A R G A L I T 1989a, 58, characteristically summed up their exposition as an account of 'a pastor or a priest. . . [who] teaches platitudes'! 21 She identifies them according to T H O M P S O N 1 9 5 5 - 8 .

more 'mythical' than Keret, since the deities are more involved as dramatis persona, and not merely invoked in conventional religious terms. ' T h e gods avenge themselves' in response to Aqhat's insolence, he stated (p. 355), discerning a general theological argument here, and the supreme god 'has to yield to the caprice of an inferior deity'. 22 This theological quality makes it difficult to estimate a historical basis for the story, as though that were desirable. Del O l m o Lete attempted to give a serious theological account, but his assessment of 'the caprice of the gods, their amoral conduct' (p. 356) seems to me to misconstrue the significance of mythological 23 action. H e later (p. 358) drew attention to another theological point, the contrast between Danel who is obedient to the gods and receives a positive response, and Aqhat who confronts them and is accordingly rebuffed. But while this is true in terms of narrative device, and of psychology and pastoral theology, it perhaps disguises the real problem on a purely metaphysical level, which is that the different deities encountered in the story are quite differently motivated in their relationships with humans. T h a t is, the deities as reifications of certain metaphysical principles are credited with their own motivation, which operates independently of immediate h u m a n motivation. Anat is after all, as goddess of war and hunting, by nature vicious, pitiless and scheming. T h a t is the role she is constructed to play. We are left feeling that no a m o u n t of diplomacy on Aqhat's part would have saved him. Certainly no a m o u n t of piety on Danel's part does him any good. Margalit has written a n u m b e r of studies on Aqhat, culminating in his large-scale commentary ( M a r g a l i t 1981a, 1983b, 1984c, 1989), the first study on a single narrative from Ugarit on this scale. 24 This is extremely thorough, but is a very difficult volume to work with, in view of its division into separate blocks dealing with the same

22 For my slightly different assessment of the general theological principles at work see below, § 13.3.2.6. 23 I am writing here of the mental disposition rather than the literary genre, though the two naturally overlap. As though grappling with this issue, del Olmo Lete (p. 356 and n. 90) writes that Aqhat is 'nearer "myth" than the "epic of Kirta"', and (n.) '[Aqhat's] intermediate position between saga and myth is recognized'. 'Caprice' is also a term appearing in de Moor's assessment (below). For further observations on the adequate assessment of myth see § 13.4 below. 24 The studies on the Baal cycle by DE M O O R (1971) and VAN Z I J L (1972a) are not formal commentaries in the same sense, and in any case the Baal material is altogether more heterogeneous than Aqhat or Keret.

materials from different perspectives and no adequate cross-referencing aids. Trying to achieve this during use is a taxing occupation. It begins (pp. 3-92) with a useful survey of previous work, though this is perhaps excessively negative, not to say waspish, in its assessment of others' efforts. 25 T h e r e follows a prosodie analysis (pp. 93-105), separated by nearly four hundred pages from the appendix on the principles of Ugaritic prosody (pp. 495-502). It is fair to say that Margalit has ploughed a lonely furrow on this topic, for few have expressed support for his approach, or are as optimistic as he that he has solved the considerable problems the topic raises. Following the initial position-statement on prosody, he offered a textual analysis (pp. 107-14), followed by a text layout (unvocalized, pp. 117-41); this is followed in turn by a translation (pp. 143-66), then by textual and epigraphic notes (pp. 167-246), and finally by a literary commentary, prefaced by short units of the unvocalized text (again!) and punctuated by excursi on various topics (pp. 2 4 7 469), before an exegetical overview, a brief statement on Ugaritic literature and the Hebrew Bible and the final appendix. T w o theoretical positions dominate the work, the non-royal nature of Danel (on which see further below), and the so-called 'Kinneret hypothesis', according to which the narrative is at home in the Galilee region. He even considered the Kinneret to be an actor in the drama (p. 411: 'the personified Kinnereth, "unwilling" to disclose the identity of the assailant, must be punished as "accessory after the fact"'). This seems a trifle excessive. Parker has written two studies ( P a r k e r 1987, 26 1989, 99-144) on Aqhat. In the former, he deplored the atomistic nature of previous philological approaches to the poem, and the patternistic bias of religious approaches. T h e time had come for a literary approach. While caution must be urged in view of the fragmentary nature of the text, a useful approach, on essentially form-critical terms, was the establishment of the types of traditional material employed. 27 'Hypothetically

25

The treatment of K A P E L R U D 1969 encapsulates this rather well: 'Author (sic) states at the outset (p. 70): "The Aqht text is still an enigma, and so far no satisfactory solution of its problems has been found". Had he limited himself to this statement, the net balance of author's contribution would have been more positive than it is in fact.' Some put-down! 2li Published in the M.H. Pope Festschrift ( M A R K S - G O O D 1987). Given as an SBL paper in 1980, and discussed briefly in M A R G A L I T 1989a, 7 1 6 . 27 Cf. discussion of del Olmo Lete above.

any consistent thrust uniting those peculiarities may be treated as the theme of the whole' (p. 71). H e isolated five main sections, and treated each in turn. These are as follows, in his treatment. A), the birth of Aqhat, deals with the familiar theme of the childless hero who appeals to the god for help; the god responds, and the child is born. T h e Egyptian tale of the Doomed Púnce and the Hurrian Appu story, the story of H a n n a h and Samuel in I Samuel, and the Ugaritic Keret story, are cited as comparable examples of the type. B), the bow of Aqhat, describes the making and delivery of the bow. T h e account of Kothar's visit has analogues in Genesis 18:1—16 and 19:1-16. Sharing certain traits is 2 Kgs 4:8-17, and a modified version of the form appears in 1 Kgs 17:9—16. C), the death of Aqhat, describes Anat's coveting of the bow, her overtures to the hero, and confrontation with El when rebuffed, culminating in her being given a free hand in accomplishing Aqhat's death with Yatipan's help. Comparison is made with the hero with Ishtar in Gilgamesh, both episodes deriving from an older Vorlage, and also with Anat's dealing with El in K T U 1.3 ν = R S 2.[014]+. 2 8 D), the consequences of Aqhat's death. Parker noted that the narrative movement almost comes to a standstill in this section, apart from describing a n u m b e r of ritual activities which accompany the inevitable environmental consequences of Aqhat's murder. Awareness of Aqhat's death (as the cause of drought) dawns only slowly, and then Danel's curses are directed towards the birds who have devoured his son, 29 and to the cities held responsible for unresolved homicides in their neighbourhoods. N o similarly extended parallels from ancient near eastern literature are cited. E), Pughat's mission of vengeance. Comparisons scholars have made with the stories of J u d i t h and Jael and Sisera are noted. While the latter connection is discounted, extensive similarities with the former are discussed.

28 P A R K K R 1 9 8 7 , 7 7 , notes that while the language in Baal and Aqhat is remarkably similar, the theme of the goddess' insubordination before the high god is far less apposite in the former. He adduces a closer relationship between Aqhat and Gilgamesh. 29 P A R K E R 1 9 8 7 , 7 9 , appears to hold the father of the raptors responsible. In fact it is their mother, Sumul, who is so described, in K T U 1 . 1 9 iii 2 8 - 3 9 . The birds are identified as vultures, p. 78. For the present author they are rather falcons.

Parker concluded that Aqhat would have originally ended with an account of the fulfilment of Pughat's vengeance, and a return of fertility to the land, 30 but with no reference to Aqhat's restoration to life. H e ended with an assessment in which the mythological emphasis drawn by previous scholarship was played down, while the social dimension was highlighted as the main theme of the author's intention. H e made some interesting observations on gender roles, contrasting Anat's 'innate and blatant masculinity' with Pughat's 'assumed and concealed masculinity' (p. 82). his conclusion raises a n u m b e r of questions (expressed rhetorically) rather than providing answers for them. H e saw the possibility of a critique of the values of the contemporary monarchy and administration. Useful as this analysis of the structures and congeners of the Aqhat story are, it does not actually tell us much about the moral or ideological dimensions which may lie behind it (that is, the author's intention). If his final questions had been answered, we might have some clear idea where Parker stood. Margalit, though too harsh in his judgment, 3 1 is perhaps justified in complaining that this tells us more about comparative literature than about Aqhat. It certainly shows the relatively sterile nature of analysis which gives no account of why an author works in this or that way, beyond the fact that it all boils down to 'traditional themes'. Parker returned to the topic in a further study ( P a r k e r 1989). Here he set out the broad characteristics of Ugaritic narrative verse, as it was evidenced in particular in the Keret and Aqhat stories. H e then turned to Aqhat itself (pp. 99-144), and outlined much the same discussion as above. His conclusion was extended to a demonstration of how, while drawing on common mythic and legendary themes, the author(s) ('composers') have, 'by adopting, transforming and combining several different traditional narratives, produced a larger work of striking unity' (p. 142). Again he emphasized the familial values promoted in the story, as distinct from conventional mythological themes, raised the question of authorial motive, and now suggested that as a piece of 'classical' literature in Ugarit, Aqhat may

30 Does he mean an element of the 'fertility cult' here? He does not say. At most, what can be said is that fertility represents divine blessing, while sterility is the outcome of a curse, and in broad terms fertility also has to do with royal power and its effective implementation. The loss of a prince is a threat to a kingdom. However, P A R K E R notes ( 1 9 8 7 , 8 3 ) the lack of overt emphasis on royal issues. 31 M A R G A L I T 1989a, 72.

have afforded its readers and hearers the opportunity of seeing themselves mirrored in the world of the story, 'a satisfying portrayal of life in an idealized past era, a life with its own tragedies, but also with its own orderly and beautiful institutions that in the end prevailed' (p. 143). In his translation of the texts, d e M o o r 1 9 8 7 , 2 2 4 - 6 6 , made a n u m b e r of comments on the literary features of the story. In keeping with his broader assessment of Ugaritian theology, 32 he saw Aqhat as dealing 'with life and death, and with the fate of m a n who all too often appears to be the victim of divine caprice'. Read in the light of his earlier and later treatments of Ugaritic theology, this is not a perspective to be taken seriously, since he appears to have envisaged a culture incapable of the moral insight to question its own bankrupt theology. T h e authors are thus as benighted as their literary characters. At best Ilimilku reflects a disillusioned and pessimistic oudook supposedly typical of the Late Bronze Mediterranean world. 33 Into this scenario d e M o o r (1990, 97 = 1997, 99) wove an argument developed some years earlier ( d e M o o r 1988a), discerning in Aqhat a further outworking of the seasonal pattern he had previously argued to be the foundation of the Baal Cycle ( d e M o o r 1971). In the 1988 article he expressed the principle thus: 'Ilimilku . . . deliberately wove a seasonal pattern into the Legend of Aqhatu out of his conviction that life on earth revolves according to a circular pattern that had been laid down in the pristine age of myth' (p. 61). H e then proceeded to fix episodes in the narrative in sequence through the calendrical year in the same m a n n e r as had been done for the Baal cycle. T h e substantial objections raised by some scholars to the seasonal interpretation were dismissed as of no consequence ( d e M o o r 1988a, 75 n. 6). (1990) 34 offered a very thorough analysis of the narrative from a folk-literature perspective, drawing on the work of Propp, Dundes and Dolezel. H e saw the narrative structure in terms of a series of different thematic levels, and of alternating patterns; 'lacks' Aitken

32 D E M O O R 1986b, 1990, 42-100 (= 1997, 41-102). For my views on this issue see § 13.3 below. 33 See DE M O O R 1990, 99 (= 1997, 101). In my view the observations made here result from a mistranslation of K T U 1.19 ii 34-36. For my translation see W Y A T T 1998c, 301. 34 This is the published form of an Edinburgh PhD dissertation from 1978. The latest entries in the bibliography are from 1984.

being 'liquidated' (not the most apposite term in view of Aqhat's liquidation!) as desires were met or situations reversed (e.g. a son for the hero, a bow for the hero, the bow for Anat, and so on) or a status quo maintained. A m o n g the oppositions a set of equivalences (called 'the synonymous sequence') is also developed, and periodic repetition (e.g. searching the falcons' gizzards for the remains of Aqhat maintain tension and development to a climax. Aitken was able to achieve this, quite legitimately, in spite of the considerable gaps in the narrative, and showed the tight construction of the surviving text, and, as he put it (p. 206), 'of the skill and artistry of its narrator, the Ugaritic teller of tales'.

3.4

Some recurrent and unresolved issues in Aqhat

A n u m b e r of individual episodes and themes in the story have been the subject of particular discussion. 3.4.1

The Incubation theory

Like O b e r m a n n (1946), Gaster interpreted the temple episode ( K T U 1.17 i) as an incubation scene, 35 as did Gray, 3 6 del O l m o Lete 37 and Parker. 3 8 This view has however been persuasively challenged by M a r g a l i t 1989a, 2 6 0 - 6 , and by H u s s e r 1992, 2 9 - 6 2 , 1996, 9 3 - 5 , who marshall substantial arguments against the incubation interpretation. Margalit, citing H a m i l t o n ' s (1906) study, observes that none of the conditions required is fulfilled: chthonian gods are not involved, no illness is involved, there is no reason to think that Danel's sleeping is part of the ritual, there is no direct theophany, and no cultic personnel are involved. Furthermore, no other ancient Near Eastern candidate fulfils the conditions either, and we are left with a late hellenistic institution with no obvious points of contact. Husser's original discussion was complex and extended, taking Oberm a n n on at every juncture. His later paper summarized his main findings, broadly in accord with Margalit. H e noted that it was gaps in the text, filled out in academic imagination, which appeared to justify the incubation interpretation. Furthermore, it was not to Danel

35 36 37 38

GASTER 1 9 6 1 , 3 1 6 . GRAY 1 9 6 9 , 2 9 6 . D E L OLMO LETE 1981a, 3 3 2 - 3 , PARKER 1 9 8 7 , 7 2 ; 1 9 9 2 , 1 0 0 .

1984a, 119 2 0 , 1984b.

that Baal drew near (in a theophany), but to El, to whom he speaks about Danel in the third person. If 'incubation' is to be used to describe the scene, it requires a considerable extension of the classical meaning of the term. 3.4.2

The occasion of Aqhat's birth

W h e n was Aqhat born? T h e conventional interpretation of K T U 1.17 ii has been that it is part of the build-up to the account of the birth of Aqhat, which must have been narrated in the gap following. This is explicitly stated, for instance, by M a r g a l i t 1989a, 147, and in some other discussions appears to be assumed, though not spelt out (e.g. P a r k e r 1987, 73). But a convincing case has been m a d e by Caquot - Sznycer, and developed further by Husser, that the birth must have taken place in the gap between the end of col. i and the beginning of col. ii.39 T h e later presence of the Kotharat is to be understood, no doubt, as for that purpose, but they evidently delay for some days after, perhaps to confirm a safe birth and the healthy state of the child. O n this alternative interpretation, the arrival of the Kotharat and the counting of days and months in K T U 1.17 = R S 2. [004] ii refer not to conception and pregnancy, but to the immediate post-natal period and the child's infancy. 40 3.4.3

Aqhat's encounter with Anat

T h e encounter between Anat and Aqhat in K T U 1.18 i (in particular 1. 24) has also given rise to much discussion. T h e key part of the text reads at.ah.wan.x[ ]. It has been frequently restored, to read at.ah.wan.a[htk].41 Was there a sexual encounter between Aqhat and Anat? This was asserted by A l b r i g h t (1944, 33-4), but rejected by G i n s b e r g (1945b, 19). T h e best argument against the c o m m o n view, i.e. that there was, is that of D r e s s l e r 1979, followed by X e l l a 1984b. I have accepted this view. 42

39

N.

CAÇ>UOT -

SZNYCER

1974,

405,

425

n.

f.,

HUSSER

1996,

WYATT

1998C,

262

50. 40

Note the pluperfect sense of 'rb bbth ktrt, 'the skilful goddesses had entered his house', in K T U 1 . 1 7 ii 2 6 : H U S S E R 1 9 9 6 , 9 1 - 2 , W Y A T T 1 9 9 8 C , 2 6 4 and n. 6 1 . 41 Thus KTU1. 42 Instead of 'You are my brother and I am your sis[ter]', the passage is to be read 'Come, brother, and I shall [ ]'. The putative reading a[htk\ is impossible. See WYATT

1998C, 2 7 9

and

n.

132.

3.4.4

Danel's social status

There has been some debate as to whether Danel is a king. G i n s b e r g (1945a, 4 n. 6) noted that Danel sits at both threshing-floor and city gate to dispense justice, citing 1 Kgs 22:10, 2 Chr. 18:9 as parallels, and also noting the widespread ancient Near Eastern use of the royal theme of caring for widows and orphans. G i b s o n (1975, 66) remarked that 'Danel is only once called a king', as though apologizing for the author's loose use of language! He preferred to see in Danel a patriarch of the J o b a n or Abrahamic kind. T h e formal royal view has not gone unchallenged, as noted above. M a r g a l i t (1989a, 2 5 3 - 4 , 278, 292-3, 309, 361 2, 410, 424-7) in particular has been most persistent in arguing that Danel is not to be seen as a king. His arguments (1989a) were as follows, though the case was m a d e more by assertion than by demonstration. A premonarchical society was envisaged by Margalit (p. 309), in which 'notables' dispensed justice. Danel's 'political status is that of unus (doubtfully primus) inter pares. H e is one of the "city elders" . . . ' (p. 361). Finally, he did not recognize the sense 'king' of mlk in K T U 1.19 = RS 3.322+ iii 46. He construed it (1989a, 163, 410) as 'your down-course' (Ar. may I). In view of Gibson's comment, it is worth noting that with substantial portions of the text missing, not too much should be made of the apparent hapax appearance of mlk here. Margalit's case can hardly be said to be very strong, and much rhetoric and repetition can hardly substitute for reasoned discussion. T h e counter-arguments are as follows. Firstly, the title mt rpi, however it is taken, has a royal significance. In my translation (1998c, 250 n. 5) I have taken it in the sense ' m a n (i.e. ruler) of R a p h a ' . T h e alternative sense is to take it as a promise of Danel's later (postmortem) incorporation into the rpum (deified dead kings of high rank). T o this cf. the element in the blessing of Keret which foresees his inclusion a m o n g the rpum, K T U 1.15 = R S 3.343+ iii 2 - 4 , 13-5). Secondly the blessing formula restored at K T U 1.17 i 3 4 - 6 , on the strength of K T U 1.15 = R S 3.343+ ii 16-20, is to be construed as explicitly royal in its object. As Jackson and Dressier have noted, there is a close affinity between the scene described and the Ugaritian royal seals. 43 Within this formula, the form cbd reappears (pace Margalit), with reference to Danel, at K T U 1.17 i 36 (previously

43

JACKSON -

DRESSLF.R

1975. See also WYATT 1997, 787-9.

misread as .bdh, with preceding word-divider), and this too is an explicitly royal title.44 Thirdly, Danel also performs an action reminiscent of El himself at K T U 1.17 i 10-11, in placing his feet on his footstool. This is to be understood as an accoutrement of royal rank (cf. Ps. 110:1). Fourthly, the description in K T U 1.17 i of Danel 'enrobed' (uzr) for the performance of his devotions may be an allusion to the ritual garment in which Ugaritian kings are shown robed in various representations. 4 5 Fifthly, as already observed by Ginsberg (above), Danel's sitting at the city-gate to judge, at K T U 1.17 ν 4 - 8 , 19 i 19-25, and particularly the reference to widows and oiphans as the beneficiaries of his dispensation of justice (a cliché for royal justice throughout ancient Near Eastern literature; contrast Keret at K T U 1.16 vi 3 3 - 4 , 45-50!) is the fulfilment of a specifically royal duty. Margalit's attempt to democratize this is without substance. Sixthly, the use of hkl at K T U 1.17 i 26, 43, ii 25, is most reasonably to be construed as denoting a royal palace. T o accept it as less is to require that the term is used as a sustained hyperbole. T h e former alternadve is altogether the more natural. While it is of course the pair-word to bt, it is used elsewhere in the Ugaritic corpus only of kings' and gods' 'houses', that is, palaces in the first instance and temples in the second. Finally, the form mlk occurring at K T U 1.19 iii 46 is most reasonably to be taken to m e a n 'king', with reference to Danel himself, as the subject of the verb in the sentence. T h e syntax raises no problems, and an appeal to Arabic cognates is unnecessary. These pieces of evidence represent a prima facie case for Danel's kingship, and it can only be the prior assumption that he is not a king which can challenge their cumulative force. O n the other hand, the anti-monarchical case requires the demolition of the whole argument, point by point. T h e royal status of Danel is not necessarily an essential element in the broad interpretation of the story, for the heroes of tales are often kings or the sons of kings, and it merely adds an aristocratic gloss to the narrative; but it also lends some support to the view expressed below that the story has been given

44 A king is 'bd DN, 'servant (or "gardener"—WYATT 1990b) of DN'. Thus with Judahite kings 'bdyhuoh, 'servant of Yahweh', or 'gardener of Yahweh'. 45

See

WYATT

1998C,

251

n.

6.

an ideological twist by Ilimilku. Quite apart from its interpretative significance, the fact remains that Danel's kingship appears to be taken for granted. 3.4.5

The bow

Dressier and Hillers have both drawn attention to the significance of the bow in the story of Aqhat. It is evidently of some considerable importance, since the desire for it motivates Anat to murder the hero. In the broadest sense it is a symbol of power. But what else is it? H i l l e r s (1973) argued that it was in effect a phallic symbol. Its theft by Anat would constitute an act of emasculation. He read into this a n u m b e r of modern psycho-literary insights. D r e s s l e r ( 1 9 7 5 ) countered that much of Hillers' argument centred on damaged text (and its restoration), hardly a sound basis for far-reaching hermeneutical claims. T h e bow was indeed a masculine symbol, but not a phallic one. T h a t is, it pointed to manly virtues, but without explicidy sexual innuendo. He drew attention to H o f f n e r ' s ( 1 9 6 6 ) discussion of gender-symbols in defence of his own restrained view, and rebutted each instance of an ancient literary allusion to bows, arrows and quivers in which Hillers had discerned a sexual innuendo. 3.4.6

What becomes of dead men?

T h e retort offered by Aqhat to Anat's offer of eternal life in exchange for the bow has long puzzled scholars. M a r g a l i t ( 1 9 8 9 , 3 0 7 - 1 0 ) , while having nothing to say on the specific matter of post-mortem existence, presents Aqhat's observations in K T U 1 . 1 7 vi 3 6 - 3 8 as an allusion to the Neolithic liming of skulls attested at Jericho. T h e idea that a LBA text would be preserving ritual details of a very specific kind (and long discontinued, to judge from mortuary evidence throughout the intervening period) seems most unlikely to me, and may be discounted. The further idea that LBA people might discover such ancient skulls (lying in shallow deposits) and contemplate their meaning (p. 309), while a charming idea, is scarcely a sufficient basis for postulating a systematic anthropology of death. T h e more general issue of the nature of beliefs regarding the afterlife has been dealt with at some length by S p r o n k ( 1 9 8 6 ) , who takes a maximalist line, much in the tradition of D a h o o d ( 1 9 7 0 ) , who cited the present passage in

his argument for a positive view of the afterlife, and even attributed a firm belief in a beatific vision to the psalmists. T h e following translation of K T U 1 . 1 7 vi 3 6 ~ 3 8 4 6 represents my rather less fulsome view of the matter. T h e key word in the problem of translation has been hrs occurring in 1. 37. I have explained it as representing perhaps a misspelling of the more c o m m o n hrs, 'gold'. D i j k s t r a - d e M o o r ( 1 9 7 5 , 1 9 0 ) offered the same explanation, considering hrs however as a parallel form, d e M o o r ( 1 9 8 7 , 239) later abandoned this view in favour of 'potash', but I consider that his initial insight was to be preferred. Man, (at his) end, what will he receive? What will he receive, a man (as his destiny? Silver will be poured on his head, gold on top of his skull, [and] the death of all I shall die, and I shall surely die. As I read this, Anat does indeed voice the possibility of a life after death, probably based on the beliefs held regarding dead kings, all of w h o m are 'divine' in some sense (cf. K T U 1.113 = RS 24.257. 13-26, where each R N is prefixed by il). Some dead kings (though to judge from the evidence, legendary rather than historical figures, since no members of the king-list apparently qualify) are even given the accolade rpum (cf. § 6.4), which I understand to denote dead and deified kings, comparable perhaps to the heroes of the Greek cult. However, while this idea is here broached, and Ilimilku may indeed be floating the idea through the agency of Anat's speech, he also offers through the medium of Aqhat's wholly negative answer a sound critique of such unrealistic views. 46 Cf. W Y A T T 1998C, 274 and n. 115. I have modified the third colon here. T h e colon begins spsg, translated in 1998c as 'a precious substance ? ', and variously translated as 'glaze' (|| 'quicklime': G I B S O N 1978, 109), 'enamel' (|| 'whitewash': DEL O L M O L E T E 1981a, 378), 'glaze' (|| 'potash': DE M O O R 1987, 239), 'a coating' (|| 'limeplaster': M A R G A I . I T 1989, 151). 1 noted (1998c, 274 n. 115) that this is reminiscent of Prov. 26:23, where the expression kesep siggim, was read by G I N S B E R G (1945b, p. 2In. 55) as k-spsg\ym\, 'like glaze' on the strength of the Ugaritic. But this argument may be regarded as circular, in which one unknown is used to interpret another. Perhaps the original kesep is to be retained in Proverbs, and restored to the Ug. text as spsg || hrs/*/j'rs. T h e -sg(ym) remains unexplained in both cases, but that is another issue. But Ugaritic is relatively stereotypical in its use of regular word-pairs, and ksp || for s is fairly commonplace, occurring many times in the repertoire. Furthermore, the Heb. term is pi. in form, while the Ug. is sg., thus reducing the appearance of a seemingly identical terminology.

3.5

New angles on Aqhat

T h e present author has attempted 4 7 to inteipret the work of Ilimilku overall, as motivated in all probability by the concerns of his office as priest and sacrificer of the king. Such a substantial amount of material ( K T U 1.1-6, 1.14-6, 1.17-9, perhaps 1.10, and now 9.432 = R S 92.2016, still unpublished as this goes to press) by one named person, however we apportion his editorial and compositional input, is quite remarkable in the Late Bronze, and we may at least ask whether he did not have some broad ideological concern to express. Working on the basis of the view, now under challenge, that the N i q m a d d u of the colopha is N i q m a d d u II, it is proposed that we have in this substantial opus a legitimization of the new king, whose reign may have begun in inauspicious circumstances (death of a prior claimant? usurpation?), whose claim required every kind of support available in a hearts and minds campaign. T h e place of Aqhat in this hypothetical p r o g r a m m e was almost circumstantial, since the weighty arguments were already spelt out in the Baal myths and in Keret. But it would tend to reinforce some features of the Keret story: thus the restitution of the blessing episode to the text would tend to confirm the author's view that Danel is indeed a king. This in turn would highlight the significance of the magnificent bow, fit for a god, and of course of its function as a royal weapon (the king as hunter is an analogue of the king as warrior), and thus attach a peculiar importance to Aqhat, the hapless recipient of this wonder. While not in the same form as the search for Baal and the allusion to it in searching for help for Keret, Danel's search for Aqhat's remains is perhaps intended to invoke the same theological associations, as though Baal's death is a type of a king's, here a future king, with an aspiration to some happy hereafter. This of course is what Anat offers the prince, but which he sharply rejects. Are we to see in the raising of such questions a grappling with issues which tradition has indeed sanctified by habit, but whose answers are no longer regarded as good enough? Ilimilku might thus be seen as one of O'Shaunessy's poets, the movers and shakers of the world. In attempting to identify the poet's motivation (as well as to quantify his personal input into a representation of essentially traditional

W Y A T T 1997, 1998a. See also notes to the text in 1998c, 3 4 - 6 , 1 7 6 - 8 , 2 4 6 - 8 (written before 1998a), where I initially explored some of the ideas in question. 47

material), I am of course entering into a mode as subjective as that which I have implicitly or explicitly criticized above in other scholars' work. But there seems to me to be a world of difference—however difficult it may be to achieve it—between attempting to foist on an ancient author one's contemporary prejudices and discovering what were indeed his own concerns in the matter. In the matter of the 'caprice' of the gods, mentioned by two authors cited, I think we have an example of the all too c o m m o n tendency to allow m o d e r n theological values, already c o m m o n l y imported into the Hebrew Bible, and declared to be resident there, to affect the agenda. Not content with an invented 'biblical world view' which is more often that of the post-reformation period, there seems to be an attempt to make Ilimilku himself an early reformer! Yet I see no tension at all in broad matters of metaphysics between Ilimilku and his world. So far as the internal and traditional theology of Aqhat is concerned, and which there is no reason to believe is under attack by Ilimilku, the kind of perspective outlined below (§ 13.4.2) is to be discerned: polytheistic theory at the same time offers a coherent overall structure at the macrocosmic level for the maintenance of meaning and value in the world, and in the interaction of its particular deities accommodates the microcosmic realities of individual problems, individual decisions and their consequences, and the tensions which are bound to exist between the real and the ideal world. Anat's behaviour is predictable, and in no way a reflection of inadequate or immoral theology. As the embodiment of precisely the dysfunctional aspect of the world, represented by all forms of killing (hunting and war), she is a terrible power to encounter. Aqhat's brisk rebuttal of her overtures, while c o m m a n d i n g our respect, inevitably brings on his own head the whole weight of the traditional sanction on those who blaspheme (sc. question the divinely ordained order of things). T h e r e is a degree of tragedy and of awful inevitability, as the sequence of automatic cursing and automatic revenge is set in train, but this does no more than express in graphic terms the principle of accountability. O n the matter of genre, there has been much discussion on that of the Keret and Aqhat stories. Are they myths, legends or sagas? W h e n myth is defined as 'stories about the gods' (e.g. by Gunkel and Eissfeldt), then the presence of h u m a n characters in the story compromises any attempt to categorize the story as myth. Since no such inhibitions seem to affect discussion of mythology in any arena

other than the biblical one (where also it is arguably an inappropriate basis for classification, since it really belongs to a polemic rather than a detached analysis), it may be regarded as irrelevant to serious discussion of the Ugaritic texts. More to the point is perhaps the matter of authorial intention. If myth be defined, as it is likely to be in social scientific terms, as stories bearing an ideological (which may include a religious) or paradigmatic message to their public, then the issue of the nature of the characters, divine, h u m a n , or even animal, is secondary. These elements are important. Fables, folk and fairy tales, though they contain characters drawn from myth, have lost the absolute imperative of the ideological norm, which demands obedience to a conventional set of values, duties and taboos, and effectively authorizes sanctions against non-conformists. But an element of freedom in the development of a tradition exists. T h u s the myth (e.g. the Chaoskampf, which is integral to the maintenance of royal ideology) may slowly evolve into other forms, such as 'St George and the dragon', which retains vestiges of ideology, and the 'Celtic dragon myth', which does not. T h u s genre is not absolute, as a given narrative may be developing from one genre to another. We noted above del O l m o Lete's and Parker's remarks on the matter of genre. These highlighted, to my mind, the inadequacy of the 'genre' approach, in so far as it seeks to establish literary types, each with its own distinctive mental disposition, attitude to history, to religion, and so on. These divisions are part of our need to structure our thoughts on such issues, and all too frequently bring a sledgeh a m m e r to crack a nut. Absolute categorizations, of the 'saga', 'legend', 'myth', 'epic' kind, are always subjective, if only because no one agrees on definitions, and merely cut off various avenues of retreat from the absolute judgments which each genre is felt to entail. I prefer like many scholars to speak more neutrally of the 'story' of Aqhat (and of Keret), noting different tendencies on various issues. This avoids hostages to fortune, and still leaves narrators free to indulge in a degree of eclecticism, perhaps drawing on different, or even hybrid genres, and us free to estimate the broad mental stance of the tradition, unencumbered by the demands imposed by arbitrary classification. T h e most obvious starting-point for an assessment should be the global one of the LBA Weltanschauung of Ugaritian culture, as established through broad studies of its cosmology, theology, mythology and ritual forms. I am entirely happy to call this 'mythological', in

the sense in which a modern religious response in devotion, scripturereading and cult remains mythological, since it operates on a different level from the purely empirical. 'Mythological' is the more useful in that while it relates to myth, it may also denote quasi-mythic features in other genres, such as deities featuring as characters, the suspension of empirical laws for narrative effect, and so on. 48 Aqhat is to be seen as a story, built up as Parker showed around a n u m b e r of motifs, and as Aitken showed around a n u m b e r of themes, motifs, formulae and word-pairs. T h e stages of its literary prehistory are no longer recoverable, partly on account of the considerable skills of the tradents, partly because no one motivation seems evident in its construction; but in the hands of Ilimilku there is a case to answer that the poet pressed it into service in the interests of royal propaganda. T o this extent it has become an ideological text. And in so far as Ilimilku has brought an ideological element into traditional material, he has blurred the distinction between genres, and produced composite works.

3.6

Some observations on style

A n u m b e r of commentators have remarked on the 'patriarchal' characterization of Danel, undoubtedly with an eye to similarities in the presentation of the patriarchs of the Genesis narratives. T h e r e too a domestic, almost bucolic gloss is given to narrative themes which address the most urgent needs of h u m a n societies, their very physical survival, expressed most typically in the yearning of a m a n for a son, who will support him in his declining years and perform his obsequies. T h e r e is a surprising tautness to the text (well illustrated in Aitken's treatment), with no word too m a n y and an action that proceeds deliberately, its pace tailored to the various levels of meaning requiring weaving into the whole. T h e fourfold repetition of the duties of the pious son, for instance, is no mechanical overkill by a poetaster, but a skilful development of one of the main themes of the story. While it is unprovable, it is worth suggesting that this is one of Ilimilku's own insertions into the traditional Vorlage, since it is so germane to his own concerns, if my analysis is correct. In K T U 1.17 = R S 2. [004] i 2 6 - 3 3 the for-

48

Much biblical historiography is 'mythological' in this sense.

mula is voiccd by Baal, interceding on Danel's behalf with El. In 1.17 i 42~7 (and the missing conclusion) it is repeated by El in his response to Baal. These two narrations therefore take place in heaven, in the divine abode. As 1.17 ii becomes legible in 11. 1 - 8 the formula is already being repeated in a message by an unnamed deity to Danel, confirming the promise of the birth that has now taken place, and finally, the fourth instance, in 1 . 1 7 ii 1 6 - 2 3 , Danel himself repeats the formula as guarantee of divine blessing. Not only is this a classic instance of the 'semantic rectangle' ( J A M E S O N 1 9 7 2 , 1 6 3 - 7 ) in use as a literary tool, but it shows a clear progress from an idea deliberated a m o n g the gods and only finally, via a revelation, appropriated by the h u m a n recipient. T h e theological weight this confers on the text goes far beyond the merely folkloric. A similar technique is used in the episode of Danel searching the gizzards of the raptors. In the three successive scenes in K T U 1.19 = RS 3.322+ iii (edge and) l - 1 4 a , 14b-28a and 2 8 b - 4 0 a we see Danel progressively home in on the place where his son's remains lie. Firstly a rather unfocusscd look a m o n g falcon-gizzards in general; then an examination of Hargab's gizzard, and finally the discovery in Sumul's gizzard of fragments of the hero. This time there is only a threefold repetition; but this too is masterly, for with the semantic rectangle incomplete it leaves the reader with a sense of tension, of a search not really brought to a final conclusion. This is further heightened by the consequent threefold treatment of the cursing of localities held responsible. Again, the sense of tension. The reason for this is surely that the real culprit is still at large. W e may see the problem resolved on one level with the probable vengeance wrought on Yatipan by Pughat, appropriately disguised as Anat, as the final column breaks off. O n another level Anat is of course to blame, and yet as a goddess she must remain above formal, or at any rate articulated suspicion. It is on account of no weakness, and certainly of no theological bankruptcy, that the goddess appears to emerge unscathed. While her cruelty is not perhaps to be compared in too facile a m a n n e r with Yahweh-El Shaddai's inscrutability in J o b , we are faced with the same fruitless search for the fulfilment of h u m a n expectations in the face of the divine nature. 4 9

49

For further comments on Anat's theological significance see below, § 13.4.

In both contexts too, in Genesis and Aqhat, we discern ideological concerns either overtly expressed, or simmering beneath the surface. T h e patriarchs are royal and priesdy ancestors, and Danel is a king, whose fecundity determines the future of his kingdom. T h e domestic flavour is deceptive: in their final form both traditions are pregnant with ideological power.

4

The

Wayne

4.1

RPUM

T.

Texts

Pitard

Introduction: the tablets

T h e so-called "Rpurrí or 'Rephanrí texts, K T U 1.20 = RS 3.348 (CTA 20, U T 121), K T U 1.21 = R S 2.[019] (CTA 21, U T 122) and K T U 1.22 = R S 2. [024] (CTA 22, U T 124, 123), are three small fragments of at least two large, multi-columned tablets.' T h e y preserve in very broken form a portion of a narrative concerning a group of beings called rpum.'1 While extremely problematic and ambiguous, these texts have played an important role in the discussion of Ugaritic and Canaanite concepts of death and the afterlife, as well as in the reconstruction of the Ugaritic political and social order. T w o of the fragments ( K T U 1.21 = R S 2. [019] and 1.22 = R S 2. [024]) were found during the second season of excavations (1930), either inside the House of the High Priest on the acropolis or in the rubble directly outside the southern wall of the house. 3 T h e third 1

K T U 1.20 = R S 3.348 was first published in V I R O L L E A U D 1936a, 228-30, in his editio pnnceps of the Aqhat Epic. T h e other two, K T U 1.21 + RS 2. [019] and K T U 1.22 = RS 2. [024], along with a réédition of K T U 1.20, first appeared in V I R O L L E A U D 1941a. Besides the transcriptions in the standard collections (CTA and KTU, KTU2 = CAT), a recent edition of the texts, with extensive photographs, may be found in P I T A R D 1992. Major translations include V I R O L L E A U D 1941a, 1-30; D R I V E R 1956, 66-71; A I S T L E I T N E R 1964, 83-86; C A Q U O T etat. 1974, 461-80; G O R D O N 1977, 29-31; C O O G A N 1978, 48-51; L ' H E U R E U X 1979, 129-59; DEL O L M O L E T E 1981a, 405-24; S P R O N K 1986, 163 77; DE M O O R 1987, 266-73; D I J K S T R A 1988, 35-51; L E W I S 1996b, 128-31; L E W I S 1997, 196-205; W Y A T T 1998C, 316-23. 2 T h e meaning of the word is usually related to the root rp', which means, 'to heal'. T h e correct pronunciation of rpum in Ugarit remains controversial. T h e two most popular proposals are (1) räpi'üma, vocalizing it as a participle, which would mean, 'healers', and (2) rapi'ûma, a stative form that can be rendered, 'the healthy, hale ones'. See the discussion of the possibilities in L ' H E U R E U X 1 9 7 9 , 2 1 5 - 2 1 ; P A R D E E 1981-82, 3

266;

and

NACCACHE

1995.

Unfortunately the inventory lists for the first two seasons were lost, probably during World War II (cf. B O R D R E U I L P A R D E E 1 9 8 9 , 1 3 ) . It is known that these tablets were registered with topographic points that were marked on ground-plans that still exist. T h e tablets of the second season were numbered with points 2 1 0 - 6 4 . Unfortunately, excavations that year took place all along the western side of the house, in the street along its southern boundary, in the room to the west of the southern entry, and in a room to the east of the larger interior courtyard. So these topographic points are scattered throughout the house (see the plan in B O R D R E U I L P A R D E E 1 9 8 9 , 2 5 ) . Because the third fragment and most of the other literary texts were found in the entry room or just outside the doorway, it seems probable that

was discovered the following year in or near the house's southern doorway, in the same area where the Keret and Aqhat epics emerged, along with some of the Baal tablets. Of the three texts only K T U 1.22 = R S 2. [024] preserves some complete lines. K T U 1.20 = R S 3.348 contains parts of two columns on one face. T h e left column is m a d e up of the right halves of eleven lines of text, while the right column preserves somewhat over half of twelve lines. K T U 1.21 = RS 2.[019] preserves part of a single column on one face (the right two-thirds of thirteen lines), but only the last five letters of one line on the other side. T h e largest of the three fragments, K T U 1.22 = RS 2. [024] preserves a left column of twenty-eight lines, twenty-two of which are complete, and a badly broken right column of twentysix lines, with only about one-third of each line preserved. O n the other side of the latter tablet are two identifiable letters, plus fragments of two others, each the first letter of a line. We thus have an aggregate of 95 attested lines, only 22 of which are complete, while an additional 24 are somewhat more than half-preserved. K T U 1.21 = RS 2. [019] and 1.22 = R S 2. [024] may come from a single tablet. T h e scripts of both appear to have been written by the same scribe, probably Ilimilku, who also produced the other major narrative texts in the archive. O n the other hand, K T U 1.20 = RS 3 . 3 4 8 seems to belong to the work of a different scribe (cf. P i t a r d 1992, 75, n. 8), and thus presumably to a separate tablet. Since the literary tablets found in this archive range in size from about 200 to about 500 lines, we may conclude that the 95 lines attested on the rpum texts are probably no more than a quarter and perhaps only one-tenth of the lines that would have existed on the two tablets when intact. In the entire collection of narrative poetic texts from Ugarit few are as obscure and difficult to interpret as the rpum fragments. Rarely have the ambiguities of a vowelless script and the piecemeal preservation of so many damaged lines conspired so powerfully to frustrate attempts at drawing decisive conclusions about a text. T h e broad context within which the action depicted in these fragments occurs is quite obscure, a n d only a few things about the events described in the preserved lines can be discerned with certainty. It is clear that the characters called the rpum (also designated with the parallel terms ilnym, 'godly ones', and ilm, 'gods') are invited to a banquet. At least one invitation seems to be given by the king of the gods, Ilu ( K T U 1.21 = R S 2. [019] ii 8), but virtually identical

invitations are offered five other times in the three fragments, each in a broken context that does not allow us to confirm whether these come from the same deity. Each invitation seems to be followed by a description of the journey to the banquet.' Both K T U 1.22 = R S 2.[024] and K T U 1.20 = R S 3.348 seem to describe the arrival of the group at a threshing floor, where the banquet is apparently given, and K T U 1.22 = R S 2. [024] i 10-25 describes the feast as lasting for a week. O n the seventh day, it appears that the god Baal arrives. But his function in the story, and indeed the purpose of the gathering of the rpum for the banquet remains unclear, and the text breaks off at this point. Beyond these few elements of the story, little certainty about the plot of the narrative and its meaning is possible. Not only is the larger context of the fragments lost, but the broken nature of the narrative makes it impossible to ascertain even some of the basic elements of the preserved story line. For example, not only is it uncertain whether all six invitations are offered by the same character, but it is also unclear whether the invitation is addressed to the same set of rpum, or whether different groups of rpum are being invited to the feast. Further, the fragments contain almost no information about the characteristics and identity of the rpum. In fact, about all we can say with certainty about them is that they travel by chariot and that they eat heartily at the feast. O n e passage in the more complete column of K T U 1.22 = RS 2.[024] (col. i, lines 4-10), seems to describe a company of rpum who arrive at the banquet, but the text, though completely preserved, is frus-

K T U 1.21 = RS 2.[019] and K T U 1.22 = RS 2. [024] were discovered either in the street or in the room west of the entry. 4 Four of the five versions of these lines in K T U 1.21 = RS 2. [019] and K T U 1.22 = RS 2.[024] appear to read (with the possibility of some variation) atrh. rpum. lldd I atrh.ltdd.ilnym, while the last version of these lines, which directly precedes the description of their arrival at the banquet, drops the I before the one preserved Idd. This is also the case in K T U 1.20 = RS 3.348, where the lines appear in col. ii 1. The ambiguity of the I has encouraged several interpretations of the lines. Some take it as an emphatic particle, and thus render these lines with the understanding that the rpum are travelling in each case; thus, for example, 'After him the shades verily proceed' ( G O R D O N 1977, 30-31). Others propose that the I is a precative particle and that the lines should be translated with jussive force, either as part of the invitation (e.g., S P R O N K 1986, 169 72: 'May the rpum flutter to the holy place'), or as an expostulation of the poem's narrator (e.g., L E W I S 1996b, 129: 'To his shrine, Ο shades, hasten'). DIJKSTRA 1988, 4 1 - 4 3 and others have argued that the / is best understood as a negative particle. They suggest that the rpum refuse to hasten until they are finally convinced by the fifth invitation (thus, 'After him the shades did not move').

tratingly ambiguous. O n e finds critical words here that may be interpreted either as proper names of the rpum, or as verbs describing actions taken by some of the story's characters. 5 In other cases, one cannot be sure whether certain construct nouns are to be construed as singular or plural. T h u s the mhr b'l, mhr ent, and rpu b'l, in lines 8 - 9 , may be titles of individual characters (i.e. 'the warrior of Baal', 'the warrior of Anat', and 'the rpu of Baal'), 6 or they may designate large numbers of persons (i.e. 'the warriors of Baal', 'the warriors of Anat', and 'the rpum of Baal'). 7 T h e ambiguity here makes it impossible to use this section to help define the nature of the rpum. O t h e r problems arise. H o w is one to understand the relationship between the three fragments? Since K T U 1.22 = R S 2. [024] and K T U 1.20 = R S 3.348 both describe the arrival of the rpum at the threshing floor where a banquet is served, is it best to assume that the two fragments are separate versions of the same scene, fortuitously overlapping, 8 or are they describing two different banquets? T h e latter interpretation is possible since K T U 1.20 = R S 3.348 depicts the banquet as being hosted by a h u m a n , Danel, a character also known from the Aqhat epic, while the banquet of K T U 1.22 = R S 2.[024] may be hosted by Ilu. Since the former text almost certainly comes from a different tablet, it may actually be part of a completely different narrative that just happened to have a similar convocation of the rpum as a story element. In addition, what is the relationship between K T U 1.22 = R S 2.[024] and K T U 1.21 = R S 2.[019]? 9 D o they belong to a single

Beyond this problem of interpretation, the translations just quoted also point up another major ambiguity in these lines. T h e word alrh may be analyzed as a preposition with a suffix, 'after him', or it can be understood as a noun, atr, 'place, shrine' with a possessive pronoun, 'to his shrine', or simply a directional marker, 'to the place/shrine'. 5 For example, is tmq in line 8 a verb (as translated by D R I V E R 1956, 69; L ' H E U R E U X 1979, 152-53; C A Q U O T 1974, 474-75; S P R O N K 1986, 171; DE M O O R 1987, 272; L E W I S 1997, 203) or the proper name of a character (as translated by AISTLEITNER 1964, 85; DEL O L M O L E T E 1981a, 423; DIJKSTRA 1988, 47; W Y A T T 1998c, 321 n. 38)? The same question arises concerning the word yhpn in line 9. Some commentators take it as a proper name (AISTLEITNER 1964, 85; DEL O L M O L E T E 1981a, 423; DIJKSTRA 1988, 47; W Y A T T 1998c, 321); others as a verb ( G O R D O N 1966, 141; D R I V E R 1956, 69; L ' H E U R E U X 1979, 152-53; SPRONK 1986, 171; DE M O O R 1987, 272; L E W I S 1997, 203). 6 Rendered thus by D R I V E R 1956, 69; C A Q U O T et al. 1974, 474-75; DEL O L M O L E T E 1981a, 423; DIJKSTRA 1988, 47; W Y A T T 1998C, 321. 7 Translated thus or similarly in G O R D O N 1 9 6 6 , 1 4 1 ; A I S T L E I T N E R 1 9 6 4 , 8 5 ; L'HEUREUX 8 9

E.g. See

1 9 7 9 , 1 5 2 - 5 3 ; SPRONK 1 9 8 6 , 1 7 1 ; DE M O O R

DE M O O R DIJKSTRA

1987b, 267; DIJKSTRA 1988, 35-39. 1987b for a discussion of this issue.

1 9 8 7 , 2 7 2 ; LEWIS 1 9 9 7 ,

203.

tablet, or may they belong to two separate ones, once again overlapping in the part of the story recounted? T h e latter possibility cannot be ruled out, since the vast majority of the lines in K T U 1.21 = RS 2. [019] i are repeated in K T U 1.22 = RS 2. [024] ii. In particular, the former preserves two invitations to the rpum, followed by two descriptions of them making the journey toward the shrine or palace, while the latter has a threefold appearance of the same basic lines. Would such an extraordinary amount of repetition in two small fragments of a single tablet be plausible, or is it better to identify them as duplicates? Again, there is no definitive argument for either interpretation. T h e a m o u n t of repetition is indeed surprising in such a small n u m b e r of preserved lines. O n the other hand, multiple reiterations are well attested in the Ugaritic poems, 1 0 and it is possible to develop scenarios of the story that would allow for so m a n y recurrences of the invitation (e.g., that different groups of rpum are being invited to the feast). And finally, is there a relationship between these fragments and the Aqhat epic? T h e appearance of Danel in K T U 1.20 = RS 3.348 has led a n u m b e r of scholars to argue that the rpum texts are the remains of a fourth tablet of the Aqhat narrative." T h e y interpret the banquet as a preliminary to the restoration of Aqhat to life or some similar dénouement which would have brought the story to its conclusion. T h e problem with this proposal is that, apart from the presence of Danel in this text, there is nothing in the rpum texts that suggests that the action described in these fragments relates in any way to such a proposed scenario. Others have pointed out that Danel may have been the subject of more than one narrative at Ugarit, and that his presence here does not require that the fragments be related to the other known narrative in which he plays a role. 12 At this point it seems best not to insist on a relationship between them and the Aqhat story.

4.2

The Identity of the R p u m

Undoubtedly the biggest hindrance to gaining a proper understanding of these texts is the uncertainty about the nature of the rpum, 10 See for example the four-fold repetition of the duties of an ideal son in the Aqhat epic (KTU 1.17 = RS 2. [004] i 25-33; 42-7; ii 1-9; 14-23). " For example, S P R O N K 1986, 160-1; C A Q U O T et al. 1974, 463; DE M O O R 1976, 332; M A R G A L I T 1989a, 464-5; G R A Y 1965, 126-9. 12 E.g. P I T A R D 1992b, 73; D I J K S T R A 1988, 36; and L E W I S 1996b, 119.

the primary characters in the narrative. In spite of their appearance not only in these texts, but also in a few others, 13 scholars have been unable to reach a solid consensus on their identity. T h e primary proposals include the following: (1) They are ghosts of the dead, most likely the spirits of deceased kings, and perhaps of the nobility. Several scholars argue that these spirits were thought of as deified. (2) T h e y are a group of deities who join Ilu in special gatherings and who are called upon to protect the king and his city. (3) T h e y are living m e m b e r s of the aristocracy, perhaps an elite group of chariot warriors, or perhaps a group of priests involved in rituals of fertility. In this context the term may be viewed as a tribal name, probably related to another tribal designation, Ditanu, which appears in parallel with rpum in the Keret epic and in K T U 1.161 = R S 34.126. (4) A n u m b e r of scholars argue that the term may in fact be used to designate more than one of these three groups. 4.2.1

The r p u m as spirits of the dead

T h e most commonly-accepted proposal is that the rpum are spirits of the dead. 14 There are several good reasons to support this identification. (1) T h e cognate of rpum in biblical Hebrew, r'pâ'îm, has as its primary meaning, 'spirit of the dead, ghost'. T h e same meaning attaches to the word in Phoenician. (2) K T U 1.161 = R S 34.126, in which the rpu ars, 'the rpum of the earth', and the rpim qdmym, 'the ancient rpum\ appear in the context of a funerary ritual, suggests that they have a relationship to the dead. In this context they are summoned to take part in the funeral of N i q m a d d u III of Ugarit (late 13th century) and perhaps to bless the new king, 'Ammurapi. In the ritual, the summons of the 'ancient rpum' is followed immediately by the invocation of the spirits of two identifiable, deceased kings of Ugarit (not explicitly referred to as rpum, however). Scholars have argued that the 'rpum of the earth' 13

They are also mentioned in K T U 1.161 = RS 34.126, a funerary ritual text; in the Keret epic (KTU 1.15 = RS 3.343+ iii 3 and 14); at the conclusion of the Baal cycle (KTU 1.6 = RS 2. [009]+ vi 46), and perhaps in K T U 1.108 = RS 24.252.23-4 and K T U 1.82 = RS 15.134.32, both of which are damaged. 14 See for example, C A Q U O T 1960; D I E T R I C H - L O R E T Z - S A N M A R T I N 1976c; P O P E 1977; S P R O N K 1986, 161-96; F O R D 1992; W Y A T T 1998C, 3 1 5 n. 1.

and the 'ancient rpunÎ are royal ancestors of 'Ammurapi from the distant past. Since the word, ars, 'earth', was sometimes used to designate the netherworld, one can render rpu ars, 'the rpum of the netherworld'. (3) A passage at the end of the Baal epic K T U 1.6 = R S 2. [009]+ vi 45-9) contains two bicola in which the four words rpim, ilnym, ilm, and mtm may be understood as synonymous with one another: špš rpim.th.tk \ špš thtk.ilnym | 'dk.ilm \ hn.mtm.'dk. This may be translated as 'Shapshu (the sun goddess), you rule over the rpum, | Shapshu, you rule over the godlike ones. | T h e gods are your company, | See, the dead are your company'. It should be noted, however, that these interpretations of K T U 1.161 = RS 34.126 and 1.6 = RS 2. [009]+ are by no means certain (see below). T h e y cannot be considered decisive for favouring this identification of the rpum. 4.2.2

The r p u m as deities

M u c h of the same evidence can be used to argue that the rpum are better understood as deities, rather than spirits of the dead. 1 5 Some scholars would identify them specifically as minor netherworld deities, closely associated with Baal. Others have proposed that the term may designate any deity, major or minor, who is called upon by Ilu to perform a special function. Arguments for identifying the rpum as deities include the fact that the term rpum is several times paralleled by the term ilnym, 'godlike or godly ones', quite plausibly a divine title, and perhaps by the term ilm, 'gods' as well."' Arguments against identifying them as the dead (as described above) and for seeing them as deities instead also include the following: (1) Later meanings of words do not always parallel earlier meanings of the cognates in other languages. T h u s the Hebrew and Phoenician

15 See S C H M I D T 1 9 9 4 , 8 3 - 9 2 and L ' H E U R E U X 1 9 7 9 , 1 1 6 - 9 for a history of this view. L ' H E U R E U X also argues that the rpum of our texts are deities, not spirits of the dead, though in other passages the word may designate such spirits. See ibid. 2 0 5 - 6 , 2 1 5 - 3 0 . The idea that the rpum are deities should be distinguished from the view of those who argue that the dead are deified at Ugarit and thus can be identified as both spirits of the dead and gods. 16 The occurrences are: K T U 1 . 2 0 = RS 3 . 3 4 8 i 1 - 2 ; K T U 1 . 2 1 = RS 2 . [ 0 1 9 ] i 3 - 4 , 11-2; K T U 1.22 = RS 2. [024] ii 5 - 6 ; all relatively certain, but all in broken contexts, and in K T U 1 . 6 = RS 2 . [ 0 0 9 ] + vi 4 5 - 9 . Parallels with the word dm, 'gods', probably occur in K T U 1 . 2 0 = RS 3 . 3 4 8 ii 1 2 and 8 - 9 , both, however, in very broken contexts.

cognates are far from decisive. T h e context of a word like rpum within the local literature is more significant for its interpretation. However, nothing in our three rpum texts specifically points to a ghostly identity for the rpum. In fact, there is no hint in any of the fragments that the regular habitation of the rpum is the netherworld. T h e only clear action tells of the rpum preparing their chariots, travelling to a threshing floor and having a feast, none of which particularly hint that the beings are spirits of the dead. Near Eastern evidence for spirits of the dead travelling from the netherworld in horse-drawn chariots is nil, while such transportation for deities, whether from the netherworld or elsewhere, seems quite reasonable. T h e lack of netherworld indications in the rpum texts might suggest that, while K T U 1.161 = R S 34.126 indicates a connection between the rpum and funerary activities (see below), their sphere of activity may not necessarily be restricted to that realm. (2) Although the rpu ars and rpim qdmym of K T U 1.161 = R S 34.126 are involved in a funerary context, this does not mean that they are necessarily spirits of the dead. T h e y may also be identified plausibly as deities. T h e names of the beings specifically identified in K T U 1.161.4-7 as rpum are not attested as royal names of Ugarit (or elsewhere), but in fact resemble divine names more than h u m a n ones (see particularly the composite name, sdn-w-rdn, (lines 6, 23) and tr 'limn (lines 7, 23-4). W h e n the two recognizable, deceased kings of Ugarit are named (lines 11-2, 25-6), each is designated as mlk, 'king', rather than rpu. T h u s those who are obviously spirits of the dead appear to be given a different designation in the text. In sum, the evidence from this tablet does not compel one to identify the rpum as ghosts. (3) T h e reference to the rpum in the Baal Epic ( K T U 1.6 = RS 2. [009] vi 45-9), which can be interpreted as equating the rpum with the dead, is in fact ambiguous, because the word, mtm, often translated 'the dead' in this passage, could actually be a h o m o n y m well attested at Ugarit that means 'humans'. In this case, the passage may be translated, 'Shapshu, you rule over the rpum, | Shapshu, you rule over the godlike ones. | Gods are your company, | See, humans are your company'. 1 7 If this is the correct rendering, there would be no direct parallel connection between the first bicolon and the second, as assumed by those who suggest that all four words refer 17

See

SCHMIDT

1994,

84-8.

to the dead. Rather, the two extol the importance of Shapshu in her relations with various elements of the world order—the rpum/ilnym, minor gods, perhaps related to the netherworld; then ilm/mtm, the divine and h u m a n spheres as a whole. Although none of the three points made here disproves the identification of the rpum with the dead, they each emphasize the fact that none of the arguments for such identification is fully compelling. O n the other hand, while the texts are compatible with the view that the rpum are deities, none of them require that meaning to make sense. 4.2.3

The r p u m as living persons

Some scholars have argued that at least certain of the Ugaritic references to the rpum are best understood as referring to living persons. 18 In this interpretation, they are usually identified as an elite group of chariot warriors who had strong connections with the king. Those who support this idea argue primarily from the passage in the Keret epic, K T U 1.15 = R S 3.343+ iii 2 - 4 and 13-5 in which Keret is praised by Ilu himself: 'Greatly exalted is Keret in the midst of the rpum of the earth (or land), | in the gathering of the assembly of Ditanu'. Supporters argue that 'rpum of the earth' in this passage can hardly refer to the spirits of the dead, since it would be inappropriate in the context of the exaltation of Keret to proclaim his preeminence a m o n g the dead. 1 9 T h e 'assembly of Ditanu' (qbs dtn), the phrase that is parallel to rpu ars in the passage, can plausibly be identified as a designation for the leaders of the Ditanu clan. This suggests an earthly, political and social context for the rpum. T h e military imagery of our three rpum texts (especially the use of chariotry and the appearance of what seems to be a military host in K T U 1.22 = R S 2. [024] ii 4 - 1 0 ) fits reasonably into this reading as well. Most scholars who identify the rpum as living humans

18 See L ' H E U R E U X 1 9 7 9 , 1 2 0 — 7 for a history of this type of proposal. More recently see S C H M I D T 1 9 9 1 , 7 1 - 1 2 1 for a detailed defence of the argument, including a number of new elements. G R A Y 1 9 4 9 argued that the rpum were elite leaders, perhaps heads of clans, or perhaps priests, who accompanied the king in rituals designed to insure fertility of crops. This view has generally not been followed. 19 Those who believe that the rpum are spirits of the dead or deities dispute this, arguing that since the rpum play an important role in the preservation and support of the king, Keret's exaltation in the midst the ancestors is a reasonable image to use in the story. See P O P E 1 9 7 7 , 1 6 6 - 7 ; F O R D 1 9 9 2 , 7 3 - 6 .

also assume that even after these chariot warriors died, they continued to be called rpum, so that there were both living and dead rpum. This would explain the origin of the usage of the term for the dead. Note is often made of the fact that the Bible refers to an ancicnt tribe of giants, located in northern C a n a a n and Transjordan before the establishment of Israel, as rephaim (Gen. 14:5, 15:20; Deut. 2:11, 20; 3:11, 13; Jos. 12:4, 13:12; 17:15). K n o w n for their military prowess, these rephaim are thought to be a dim memory of the elite warrior class of that name in the Late Bronze Age. So again we find some plausible arguments, but again none of the texts used to support the interpretation provide irrefutable evidence. Plausible, alternative interpretations of each of these passages have been made. T h u s in the final analysis, no decisive conclusions about the identity of the rpum can yet be drawn. It is quite possible, as several scholars have argued, that the word had more than one meaning in the Ugaritic texts and that different contexts require different meanings. O n the other hand, other scholars have made plausible cases for seeing all the occurrences of the word as referring to a single group of beings, either spirits, gods, or humans. Only further discoveries of texts relating to the rpum are likely to improve the present situation. From the preceding discussion, it is clear that these texts are exceedingly ambiguous and that great caution should be used in drawing upon them to reconstruct aspects of Ugaritic or Syro-Palestinian culture. In many cases such caution has not been employed, so that the rpum texts have been used extensively as the basis for elaborate descriptions of Ugaritic concepts concerning afterlife by scholars who identify the rpum as spirits of the dead. 20 A similar situation has also occurred in some of the reconstructions of the Ugaritic military and

20

For example, DE M O O R 1 9 7 6 , 3 2 9 - 3 3 and S P R O N K ( 1 9 8 6 , 1 5 5 - 6 , 1 7 0 - 4 ) , using K T U 1 . 2 2 = R S 2 . [ 0 2 4 ] ii 5 - 7 and K T U 1 . 2 1 = R S 2 . [ 0 1 9 ] i 5 - 6 as their primary sources, reconstruct an event at the Ugaritic New Year Festival in which the deceased nobility, the rpum, are brought back to life, along with Baal, their patron. But such a reconstruction relies for its foundation upon the identification of the rpum as spirits, which remains uncertain, alongside problematic readings of both key passages. Their interpretation of the first passage is based on questionable renderings of two problematic verbs, the first of which (his ytbš, which occurs twice here) is based on a misreading of the text (the verb is y'bs, instead), and the second, qym, more plausibly rendered as a noun than a verb (see VAN DER T O O R N 1 9 9 1 , 5 2 ) . In the second passage, they must reconstruct both of the critical verbs needed for their interpretation.

its relation to the royal house by those who view the rpum as living elite warriors. 2 ' It is important not to place too much interpretational weight on ambiguous and problematic texts such as these. Before they can be used as sources for dealing with the wider issues of Canaanite religion and society, a clearer understanding of the texts themselves is necessary.

21

See, for example,

GRAY

1952,

39-41.

5

THE

INCANTATIONS

KLAAS

5.1

SPRONK

Introduction

T h e incantation can be defined as 'rhythmic or formulaic words of power to accomplish a desired goal by binding spiritual powers'.' A n u m b e r of Ugaritic texts written in alphabetic script unearthed in Ras S h a m r a and Ras Ibn H a n i fit, completely or in part, under this heading. In none of these texts, however, do we find a word specifically denoting the incantation, like Akkadian siptu.2 This word is attested at the end of a syllabic Akkadian incantation found in Ras S h a m r a 'against fire' (RS 17.155). 3 We do find the more general indication mnt? This has an equivalent in Akkadian minûtu. In Akkadian incantations it is used next to siptu. It seems to refer in the first place to the act of reciting and repeating the incantation, because it is derived from the verb manû,b 'to count' (cf. Hebrew mnh). In K T U 1.24 = RS 5.194:46-7, mnt denotes the enumeration of goddesses listed in the following lines. It is paralleled by spr, 'list'. In K T U 1.82 = R S 15.134:20 and K T U 1.100 = R S 24.244 it is used in texts that as a whole can be labelled as incantations. In other 'pure' incantations, like K T U 1.96 = R S 22.225 and K T U 1.169 = R I H 7 8 / 2 0 , this or another general term for the genre is missing. It appears to be impossible to distinguish the 'Ugaritic Incantations' sharply. 6 Also, the calling of divine beings by their names at the end of K T U 1.24 = R S 5.194 may have had some kind of magical purpose. T h e same can be said of other parts of some of the mythical or ritual texts. Add to this, firstly, that is often difficult to distinguish

1

2

LUDWIG

1987,

151.

For the much larger corpus of Akkadian incantations see the surveys by F Ä R B E R 1981, 1984 and 1987. 3 Cf. A R N A U D 1995a. Within the context of the incantations' 'fire' can be regarded as a reference to demons; cf. K T U 1.2 = RS 3.367 i 3, where the demoniacal helpers of Yam are described as 'one, two fires'. 4 See on mnt P A R D E E 1 9 8 8 , 2 0 6 - 8 . According to DE M O O R 1 9 8 7 , 2 4 8 the masculine mn is attested in K T U 1 . 1 9 = RS 3 . 3 2 2 + i l l . 5 Cf. the expression manû šiptu, 'to recite an incantation', in Akkadian, cf. CAD Š, III, 89. 6 Cf. J E F F E R S 1 9 9 6 , 1 8 , facing the same problem with regard to the more general theme of magic and divination.

a prayer from an incantation, in the second place, that as a rule magical and related texts are difficult to interpret, and that, finally, many of the tablets concerned are damaged, then it comes as no surprise that in the editions of and commentaries on Ugaritic texts we find different proposals for classification: • According to Avishur the only texts which 'can clearly be classified as incantations' are K T U 1.100 = R S 24.244; K T U 1.107 = R S 24.251+; and K T U 1.169 = R I H 78/20. 7 • Xella lists under 'preghieri ed incantesimi': K T U 1.65 = RS 4.474; K T U 1.123 = R S 24.271; K T U 1.100; and K T U 1.107. 8 • DE M O O R comes to five incantations as 'more or less independent prayers without ritual prescriptions', next to incantations 'embedded in rituals': K T U 1.82 = R S 15.134; 1.83 = RS 16.266; K T U 1.169; K T U 1.93 = RS 19.054; and K T U 1.108 = RS 24.252. 9 • Dietrich - Loretz come to eight alphabetic texts that in their view are representative of this 'Gattung', leaving out two thematically related but heavily damaged texts. T h e y subdivide these eight texts into four categories: (1) 'Evokationen königlicher Ahnen': K T U 1.124 = RS 24.272 and K T U 1.161 = R S 34.126; (2) 'Beschwörungen gegen Krankheit, Unfruchtbarkeit, D ä m o n e n , Folgen von Trunkenheit und Totengeistern': K T U 1.13 = R S 1.006; K T U 1.82; K T U 1.114; K T U 1.169; (3) 'Beschwörung gegen Schlangengift': K T U 1.100; (4) 'Beschwörung gegen die schädliche Naturkräfte': K T U 1.23 = R S 2.002.'° • C a q u o t lists K T U 1.82; K T U 1.114 = RS 24.258; K T U 1.100; K T U 1.107; and K T U 1.169 under the heading 'tablettes mythicomagiques'."

7

B ,J

AVISHUR XELLA

1981,

13.

1981, 207-50.

DE MOOR 1987,

175-90.

1988b, 328-7. In KTU1 the following texts are also marked as possible incantations: 1.20-22 = RS 2. [024], 2. [019], 2. [024]; 1.65 = RS 4.474; 1.86 = RS 18.041; 1.96 = RS 22.225; 1.107 = RS 24.251+; and 1.123 = RS 24.271. Apparently, these belong to the 'damaged' or 'related' texts referred to by Dietrich - Loretz. " C A Q U O T 1989, 51-123. It is remarkable that he pays no attention to these texts in his survey in C A Q U O T 1979b. 10

DIETRICH -

LORETZ

• In his description of Ugaritic religious practices in daily life del O l m o Lete mentions as 'conjuras': K T U 1.100; K T U 1.107; K T U 1.82; K T U 1.96; and K T U 1.169; 12 and as 'recelas mágicas': K T U 1.124 = R S 24.272:13-5 and K T U 1.114:29-31. 1 3 • A recently published survey of documents from the biblical world contains as examples of Ugaritic incantations: K T U 1.100; K T U 1.169; K T U 1.114; and R S 92.2014.' 4 Apparently the old Ugaritic texts on these matters cannot be clearly classified. It is better in this situation not to put too much weight on our modern definitions and choose a more general approach to the Ugaritic texts about h u m a n efforts to have an influence upon the supernatural, from raising one's hands in prayer to binding hostile spiritual powers by magic. This means that the boundaries between 'literary' and 'cultic', and those separating 'myth', 'incantation', 'ritual', and 'god lists' are not always as clear as editors of a handbook might want them to be.

5.2

Speaking to the gods in hymns and prayer

T h e genre of prayer appears to be rare in the texts of ancient Ugarit. 15 O n e should not, however, conclude from this that the people of Ugarit did not have deep religious feelings or that they were reluctant to address their gods direcdy. T h e lack of separate hymns or prayers is simply due to the fact that praising the gods or seeking their favours is usually set in a larger context. Recitation of the great myths can be seen as a means of expressing respect for the gods and their glorious deeds. For instance, telling each other about Baal's victory over Yam and Mot with the words of the myth of Baal expresses one's confidence in the power of the supreme god over chaos and death. T h e hymn addressed to the sun-goddess Shapash at the end of the myth ( K T U 1.6 = R S 2.[009]+ vi 45-53) is put in the mouth of Anat, but it is also a way in which the people of Ugarit thankfully praise the sun-goddess for watching over the boundaries between night and day, the world of the living and the world of the dead. Hymnic elements can also be found in the second part

12 13 14

15

1992a, 241-60 = 1999, 359-87. 1992a, 261 = 1999, 388. H A L L O 1997, 295-8; 301-5; and 327-8. Cf. W A T S O N 1984a, 360 and M I L L E R 1988. D E L OLMO LETE

DEL OLMO

LETE

of the much debated text K T U 1.108 = RS 24.252, , e which is discussed below together with texts related to necromancy. Hymns also seem to have formed a standard element in Hurrian prayers accompanying incense offerings ( K T U 1.44 = RS 1.007; K T U 1.51 = R S 1.027; K T U 1.54 = RS 1.034+; K T U 1.128 = RS 24.278; K T U 1.131 = RS 24.285). 17 In the legend of Aqhat we hear of his father Daniel praying (Ug. sly) for rain ( K T U 1.19 = RS 3.322+ i 38 46). He calls on the name of Baal, as 'rider of the clouds', and on his 'delightful voice', that is of the thunder heralding coming showers. In a subsequent scene, Daniel beseeches (Ug. sly)™ the gods that the small stalks in the dry land may shoot up ( K T U 1.19 ii 15-25). It is more c o m m o n for prayer to be part of ritual actions, as we can see in the legend of Keret. T h e c o m m a n d to raise the hands (in prayer) is paralleled by a reference to a sacrifice to El ( K T U 1.14 = R S 2.[003]+ ii 22-3). W e can also find this combination in the ritual text K T U 1.41 = RS 1.003+, with prescriptions about the annual celebration of the grape harvest in the month 'First of the Wine'. T h e text ends with the same call for prayer as in the legend of Keret. In K T U 1.87 = R S 18.056, a copy of K T U 1.41, these last lines containing the reference to the king's prayer are missing. T h e action described in K T U 1.41:50-5 is situated in another place: not in the temple, but on its roof; and it refers to a sacrifice offered to an unknown deity (prgl sqrn). This is probably a deity of Hurrian origin. So the expansion of the text can be explained as due to later Hurrian influence upon an older Ugaritic ritual. T h e king is said (or prescribed) to offer a recitation (yrgm mlk), but we hear nothing of its contents. Perhaps building on the assumption of Hurrian influence one should think here of something like the Hurrian incense prayers mentioned above. These texts all follow a similar pattern: after the heading we read the names of the gods to w h o m the prayer is addressed, together with a short hymn. T h e gods are asked to come and receive the offerings and then to do something on behalf of the suppliant. T h e texts end with mention of the messenger and in some

16 A V I S H U R 1994, 297-8 even speaks of the entire text as a 'hymn in honor of EF, because of the striking similarity with Hebrew psalms and its vocabulary being reminiscent of that found in hymns. 17

18

Cf.

DIETRICH -

MAYER

1994.

Cf. DE M O O R 1987, 252, n. 190 and interpretation of 'ahl see DEL O L M O L E T E

MARGALIT -

1984b, 140-1; for a different 1996, 16.

SANMARTIN

cases with promises of new offerings and a final doxology. T h e words spoken by the king according to K T U 1.41:53, on the roof of the temple, could have been something like this calling up the gods and asking their favours. Because the tablet is damaged here, it is not clear whether this invocation is accompanied by the king wiping his face (mh pnh)i9 or by clapping his hands (mh ydh).20 T h e reference to prayer in line 55, back in the temple, could be related to the closing hymn in the Hurrian incense prayers. In the older secondary literature K T U 1.65 = R S 4.47 4 21 has been interpreted as a prayer to El and the assembly of the gods. 22 More recently commentators of the text appear to be reluctant to classify it. Xella points to the resemblance of the first lines (naming El, the sons of El, the family of the sons of El, the assembly of the sons of El, and tkmn-w-snm) with the repeated address of the ritual text K T U 1.40 = R S 1.003+. H e assumes as a working hypothesis that K T U 1.65 is some sort of prayer. 2 3 Dietrich - Loretz take this text as a scribal exercise. 24 In his elaborate study of this text Avishur also concludes that the old view (of H.L. Ginsberg) that this text is a prayer, seems to be closest to the truth. Comparison with the Q p m r a n W a r Scroll (chapters 4 - 6 and 9, about names with Ί as a second element to be written on banners and weapons) leads him to classify it as a list of war banners. 2 5 Because these banners are 'battle cries intended to arouse the deity to assist the warriors', this text resembles a prayer. Although m u c h remains uncertain, one should not rule out the possibility that this is indeed the text of a prayer, related to sacrifices as mentioned in K T U 1.40, calling up the gods (lines 1-5), appealing to the consideration of the supreme

19

C f . DE M O O R

1987,

159,

20

165, a n d

DE T A R R A G O N

1989,

154,

159.

Cf. L E V I N E DE T A R R A G O N - R O B E R T S O N 1 9 9 7 , 2 9 9 , 3 0 1 . On the clapping of hands serving 'to intensify the accompanying words and perhaps even to effectuate the action' see Fox 1995. 21 Cf., for instance, B E R N H A R D T 1975, 239-40: 'Bittgesang an El und die Versammlung der Götter'. See on this text A V I S H U R 1994, 308-9, who also lists and discusses previous research. 22 Cf., for instance, B E R N H A R D T 1975, 239-40: 'Bittgesang an El und die Versammlung der Götter'. 23 X E L L A 1 9 8 1 , 2 1 3 : 'almeno come ipotesi di lavoro, una sorta di "preghiera"'. 24 D I E T R I C H - L O R E T Z 1 9 8 1 , 6 4 - 7 ; in their translation of this text in 1 9 8 8 they call it 'Opfer für El und seine Repräsentanten'; note also the problems with classifying this text in ATI/ 2 , 91: 'scribal exercise?, invocation of II?, prayer?, incantation?'. 25

AVISHUR

1994,

326,

525,

gods by referring to their noble character (6-9) and naming places and divine attributes in and with which these words have to be recited (lines 1 Off.). Del O l m o Lete sees it as a 'cultic invocation of the divine panoply and to its apparent presence in the sanctuary'. 2 6 In his opinion this primarily concerns Baal's weapons, celebrated in mythology. A more generally accepted example of a prayer in alphabetic Ugaritic is the end of the ritual text K T U 1.119 = R S 24.266. 27 This text starts as a common prosaic ritual prescription about the right time, place, and sort of sacrifice to the right god. In line 26 there is a transition to a direct address to Baal by referring to the problem of a strong foe attacking the gates of the city. T h e style changes here from prose to poetry. 28 T h e prayer (sit, line 34; cf. the verb in K T U 1.19 = RS 3.322+ i 39) is introduced by the command: 'raise your eyes to Baal' (line 27). T h e request to drive away the enemy is accompanied by a n u m b e r of vows and sacrifices by the suppliant, in exchange for Baal's help. T h e text ends with the statement, repeating the words at the beginning, that Baal will hear the prayer. K T U 1.123 = R S 24.271 29 is probably best described as a benediction,'" because of the repeated šlm in the opening lines, followed by a n u m b e r of divine names. T h e text seems to end in a similar way, the last word being again slm. Lines 14ff. mention righteousness and mercy. This is reminiscent of K T U 1.65 and can be interpreted in the same way as expressing the hope for and confidence in a positive attitude of the gods towards the one saying these words. However, any interpretation of this text in its present severely damaged state must remain uncertain.

26

D E L OLMO LETE

341-3

and

WATSON -

1992C, 255; cf. also WYATT

DEL O L M O L E T E

1992a, 228-9 = 1999,

1997.

27 See on this text X E L L A 1981, 25-34; W A T S O N 1984d, 360-2; DE M O O R 1987, 171-4; M I L L E R 1988; DEL O I . M O L E T E 1989; DEL O L M O L E T E 1992a, 197-205 = 1999, 292-306; P A R D E E 1993, 213-7; AVISHUR 1994, 253-6; W A T S O N 1996b. 28 O n this phenomenon see especially P A R D E E 1993a. 29 Cf. X E L L A 1 9 8 1 , 2 1 6 - 3 , with references to the older literature. 30

DE MOOR

1 9 7 0 3 1 2 ; RAINEY

1974,

191. CAQUOT

1979,

1404 and

suggest that it is a scribal exercise. KTU1, possible genre: 'prayer, liturgy?, scribal exercise'. LORETZ

1981,

74-5,

135,

DIETRICH

-

gives as the

5.3

Binding hostile supernatural forces by incantations

T h e texts discussed under this heading are the ones that best fit the narrow definition of an incantation given at the beginning. It concerns independent texts with words of power used against evil forces from the realm of gods and demons. T h e interpretation of these texts is very difficult, not only because of the state of conservation of most tablets, but also because of the genre of the texts, with unknown vocabulary and often without a clear structure or line of thought. T h e best example of an Ugaritic incantation is K T U 1.169 = R I H 78/20. 3 1 Although there is much difference of opinion a m o n g the interpreters about many details, it is generally accepted that we are dealing here with a spell to drive off evil powers causing sickness, with the help of Baal, H o r o n and Ashera. It is not clear which disease is meant here, 32 nor which power is causing it; according to some it is indicated by dbbm in lines 1 and 9, although it is translated in different ways: 'flying demons' (de Moor), 'tormenters' (Pardee), or 'accusers' (Fleming). Others (Dietrich - Loretz, Caquot) relate it to Akkadian dabābu, 'word', and interpret it as a reference to the words spoken to expel the (unnamed) demon. This difference of opinion returns in the interpretation of kspm (line 9) as 'sorcerers' indicating the black magic of demons, but according to others the magic with which one can expel the forces of evil.33 T h e r e is more consensus about the verbs used in connection with the expelling magic : ydy (line 1), 'to drive off', and gr/(line 9), 'chase away'. Both are used in the legend of Keret in the repeated question 'who a m o n g the gods is able to cast out {ydy) the disease, to expel (grs) the illness?' ( K T U 1.16 = R S 3.325+ vi 10-28). In line 10, in close connection with ksp and grs, we find the root hbr. This is reminiscent of the use of hbr in the H e b r e w Bible and Akkadian abāru, 'to bind', in Mesopotamian incantations. Avishur points to Deut. 18:10-11 and to Isa. 47:9 with the word pair hbrym || kšpym, 'enchantments 11 spells', and to a similar pair in the Akkadian Maqlû-

31

Cf.

A V I S H U R 1981; DE M O O R , 255-7; DE M O O R 1987, 183-6; D I E T R I C H 1988, 333-6; C A Q U O T 1989, 53-60; FLEMING 1991; DEL O L M O L E T E 1992a, 2 5 9 - 6 0 = 1999, 385-6; P A R D E E 1993, 211-3; FLEMING 1997; W Y A T T 1998C, 442~9. 32 D E M O O R 1980b, 257, and 1987, 184, n. 13, assumes that the patient suffered from a 'cataleptic or epileptic seizure', Pardee thinks of 'male sexual disfunction'. 33 According to J E F F E R S 1 9 9 6 , 6 7 - 8 , one can leave open both possibilities. LORETZ

incantation. 3 4 So here hbr would denote the negative influence of evil spells. Avishur translates: 'Horon will expel the binders and the Youth soothsayers', relating the last word (dctm) to Hebrew yd'ny. Dijkstra interprets these terms in a similar way, but he assumes a positive meaning: ' H o r o n be the enchanter, and the Young M a n the one who provides knowledge. 35 In the hymn at the end of the Baal myth this word pair hbr || dct ( K T U 1.6 = R S 2. [009]+ vi 49-50) would have the same meaning. Most commentators, however, prefer the more c o m m o n meaning of hbr, 'friend', and d't, 'intimate'. 3 6 In the text we hear of the one who has to recite the incantation: 'the /^-priest' (line 3). This title is also used in the colophon of the Baal myth ( K T U 1.6 vi 57) and seems to refer to a high-ranking teacher. 37 In K T U 1.40 = RS 1.002:32 we find the related verb parallel to dbh, 'to sacrifice'. According to some interpreters this officiant used a staff as a magic device, 38 but the meaning of the word ht denoting it (line 5) is, again, disputed. 39 This person executing the incantation by word and probably also by gestures and other ritual activities can be compared to the Mesopotamian incantation priest called āšipu. It is interesting to note that this exorcist is often mentioned in the colophon of the incantation texts as a scholar who wrote and checked the tablet. 40 O t h e r correspondences with Mesopotamian incantations are the use of similar metaphors, especially the spirits being said to leave 'like smoke' 41 and the naming of gods acting on behalf of the oppressed against the evil spirits. In some Mesopotamian rituals the incantation priest even says that it is not he himself who speaks, but that it is an incantation of Ea 42 or Ninkilil, 'lord of the incantation. 4 3 This

1981a, 22-3.

34

AVISHUR

35

DIJKSTRA

36

C f . DEL O L M O L E T E -

1985,

150. SANMARTIN

1996,

126-7

and

172. DIETRICH -

LORETZ

1988b, 335 translate 'Genossen || Komplizen'. According to J E F F E R S 1 9 9 6 , 33 both suggested meanings of hbr are related: 'comrades can be linked together by sworn words, oaths and the like'. 37

C f . VAN S O L D T

38

1988 and

FLEMING

1991,

146.

Cf. F L E M I N G 1991, 148-50. 39 Cf. D E L O L M O L E T E - S A N M A R T I N 1996, 202-3. 40 See the texts mentioned in CAD A, II, 434, s.v. āšipu a. 41 See, for instance, Maqlû V: 166-169 (translation by F Ä R B E R 1987, 265). Cf. A V I S H U R 1981, 18; F L E M I N G 1991, 146; and W A T S O N 1994b, 405-6. 42 Cf. CAD Α, II, 431-2, s.v. āšipu a.2; CAD Š III, 90, s.v. siptu e.2'. 43 Cf., for instance, the recently discovered incantation against Lamaštu, discussed by

MICHEL

1997.

can be compared to K T U 1.169 = R I H 7 8 / 2 0 beginning with the statement that it is the breath of Baal which drives out the evil spirits. Unfortunately, the text is broken here. Next to Baal a special function seems to be reserved for H o r o n (lines 9 - 1 0 ) and Ashera (line 16). A n u m b e r of these basic elements of K T U 1.169 are also found in another clear example of an incantation in alphabetic Ugaritic script: K T U 1.82 = R S 15.134. 44 This seems to be a collection of six different incantations, to be recited on different occasions, but also sharing c o m m o n elements (such as the reference to the snake in lines 6 and 35). T h e fourth part is explicitly introduced as an incantation with the technical term mnt (see p. 270) in its first line (= line 20). Like K T U 1.169, this text is difficult to interpret, but it gives us more information about the gods invoked to help and especially about the demons to fight. T h e benign gods are Baal (lines 1 and 6), his consort Anat (line 11 and twice in line 39), and the sun-goddess Shapash (line 6). T h e evil forces they have to destroy are: • T u n n a n (line 1), known from the myth of Baal ( K T U 1.3 = R S 2. [014]+ iii 40) as a monstrous helper (dragon) of Yam, the god of the sea, one of Baal's prime opponents. According to the myth T u n n a n is slain by Anat. H e also seems to have been mentioned in the small fragment K T U 1.83 = R S 16.266, next to Yam, 'bound (by c Anak?) on the heights of Lebanon'. 4 5 This reference to T u n n a n , however, is uncertain, not only because of the poor state of conservation of the tablet, but also because it is based on a correction of the text in line 8, reading tnn instead of t'an.*6 • Reshep (line 3), the god of pestilence, who is mentioned next to the 'lads of Y a m ' in the legend of Keret as the god who caused the death of one of the king's wives ( K T U 1.14 = R S 2.[003]+ i 19). • Mot (line 5), the god of death. Next to Yam he is the other powerful opponent of Baal ( K T U 1.5 = R S 2.[022]+).

44

Cf.

VAN

ZIJL

1972,

1974,

1975;

DE

MOOR

-

SPRONK

1984;

CAQUOT

1988

1988, 336-9; C A Q U O T 1989, 61-70; DEL O L M O L E T E 1992a, 251-5 = 1999, 373-9. 45 Cf. P I T A R D 1 9 9 8 ; DE M O O R 1 9 8 7 , 1 8 1 - 2 assumes that this text was part of an incantation. 46 C A Q U O T 1 9 8 9 , 2 8 - 3 0 states that tnn is the key term of the text, but he ignores the fact that for this interpretation the text has to be changed. DIETRICH

-

LORETZ

• Serpents (lines 6 and 35), who are the prime object of another Ugaritic incantation ( K T U 1.100 = R S 24.244 and K T U 1.107 = R S 24.251+; see below). • Creatures of H o r o n (line 13). This reading is uncertain, but the name of Horon, who is the lord of the demons, returns in lines 27 and 41. Horon is mentioned in the legend of Keret as a threatening power in a curse: 'may H o r o n break your head' ( K T U 1.16 = 3.325+ vi 56-7; this phrase can also be restored in the broken text K T U 1.2 = R S 3.367 i 7-8). In the Ugaritic incantations H o r o n plays an ambivalent role: on the one hand he is a fearful threat, on the other h a n d he can be called u p o n to take the demonic threat away (cf. K T U 1.100 and K T U 1.169). • In the second part of the text the evil forces are indicated more 'poetically' as 'creatures of agitation' (lines 18 and 41), 'creatures of insanity' (line 18), 'sons of disease (or: terror)' (line 23), '1egions(P)' (line 26; cf. Mark 5:9), 'flies (or: accusers)' (line 26), 'those of the flood(?)' (line 27), 'stupor(?)' (line 28), 'the fugitive' (line 38; cf. K T U 1.5 = R S 2. [022]+ i 1, where it is used as epithet of a seamonster related to Yam). It is not clear what is precisely the nature of the distress caused by these evil forces. In the first lines there seems to be reference to problems of a girl with her menstruation, that is with her fertility. T h e names of the demons in the second part of the text point in more general terms to disease and insanity. W h a t is clear is that these evil forces have to be driven out (grs, lines 12 and 40; see also K T U 1.169:9) or have to be b o u n d (rky, lines 10 and 38). Both verbs are c o m m o n terms in this genre. For the place of this and other incantations within the religion of Ugarit it is important to note the close relation with the myth of Baal. T h e batde described there of Baal and Anat, supported by Shapash, against Yam, Mot and their helpers does not appear to be something from a distant past. It has its repercussions on daily h u m a n life. T h e victory over the forces of evil has to be gained time and again. As was remarked above, Horon takes a central place in K T U 1.100 = R S 24.244. 47 Fortunately, this text is well-preserved. It is in itself not

47

See on this text especially the elaborate study of references to previous studies. Cf. also DE M O O R 1 9 8 7 ,

PARDEE

1988,

193-226,

1 4 6 - 5 6 ; DIETRICH -

with

LORETZ

a pure incantation, but can be classified as a ritual in mythological form. It contains, however, a n u m b e r of brief incantations indicated as mnt. T h e text tells of a mare, 'the mother of the stallion'. She is probably a goddess acting on behalf of her worshippers. W e are told that she seeks support from the great deities of the Ugaritic pantheon against venomous serpents. T h e sun-goddess Shapash is indicated as her mother and acts as an intermediary. Each request ends with the same incantation in order to charm (lf}š; cf. the use of Akkadian lahāšu, 'to whisper', together with siptu, 'incantation', 4 8 and the use of Hebrew 1hš, specifically related to the charming of snakes i n j e r . 8:17; Ps. 58:5-6; Q o h . 10:11; and Sir. 12:13), expel (ydy) and bind (ytq) the snake and its poison. Apparently, this incantation was repeated eleven times, each time on the basis of another authority. T h e twelfth, Horon, responds in a different way. With a magical rite, using a m o n g other things a tamarisk and 'the tree of death', he succeeds in letting the poison 'become weak and flow away'. This climax of the text is, as is appropriate in magic texts, described with a n u m b e r of puns. 49 T h e text ends with a dialogue between a groom and a bride; apparently these are H o r o n and the 'mother of the stallion'. T h e y speak about marriage with the serpents (a phallic symbol?) as bride-price. A clearly related text is K T U 1.107 = R S 24.251+. 5 0 Here the snake is called 'devourer', a c o m m o n designation of demons (lines 10 and 20).51 In the more elaborate mythological part of the text Shapash plays a more active role. 32 T h e r e can be no doubt about H o r o n being viewed here in a positive way, be it that he is clearly not the first choice. T h e eleven incantations preceding the final invocation seem to be meant to show that no other choice was left than to go to Horon's 'fortress', probably an indication of his hardly accessible residence in the nether-

1988, 345-0; L E V I N E - DE T A R R A G O N 1988; C A Q U O T 1989, 79-94; P A R D E E 1997a and W Y A T T 1998C, 378-87. 48 Cf. CAD Š, III, 90, s.v. siptu A.e.3'. 49 Cf. G R E A V E S 1994 and P A R D E E 1997a, 298, n. 26. See on this phenomenon in Mesopotamian incantations F Ä R B E R 1986. 50 Cf. X E L L A 1981, 241-0; P A R D E E 1988b, 227-56; and C A Q U O T 1989, 95-100. 51 Cf. DE M O O R 1 9 8 1 - 2 , I 1 0 . He refers, among other things, to the clear representation on a Phoenician amulet from Arslan Tash (seventh century BCE) of a demoniacal man-eater. 52 O n this part of the text and a number of resemblances with the story of the Garden of Eden according to the Hebrew Bible, see DE M O O R 1988b.

world. O n e demons by It is also and ritual. liturgy.

can compare this to Jesus being accused of casting out Beelzebul, the prince of demons (Mark 3.22). important to note the combination of myth, incantation, This appears to be a c o m m o n feature of the Ugaritic

O n e of the new Ugaritic texts discovered in 1992 is the incantation R S 92.2014, 5 3 which is in many respects similar to K T U 1.100, K T U 1.107, and K T U 1.169. As in K T U 1.169 the offending evil force remains unnamed: dy lyd\ 'the one not known'. H e is presented as a foaming snake and as a scorpion and is exorcised in a magic rite with 'bits of sacred wood'. This is reminiscent of the action taken by Horon against snakebite according to K T U 1.100 and K T U 1.107. In this way he prevents the serpent from coming up (ely) and the scorpion from standing up (a new Ugaritic root qnrì). T h e second part of the text (lines 8-13) is an incantation against dbbm and kšpm. These words are also used in K T U 1.169. In R S 92.2014 they are mentioned next to rs\ 'the evil m a n ' and bn nšm, 'son of m a n ' , which seems to be an indication of all possible men. T h e incantation is directed against any evil word spoken: 'may they pour it to the earth'. T h e incantation is dedicated to Urtenu, the holder of the archive to which this tablet belongs: 'for his body (gb), for his members (tmntf. These two words also occur in K T U 1.169:5-6. T h e incantation has a function in securing the physical well-being of Urtenu. O n e final text to be mentioned within this framework of independent incantations is K T U 1.96 = RS 22.225. This is usually interpreted as a short mythological text about Anat literally or, what seems to be more likely, figuratively devouring Baal. 54 T h e reference to Anat was found in the first word of the tablet. New collations, however, show that the first letters are cnn, not W.55 In the first edition of K T U it was suggested that cnn is a scribal error for cnt, but in the second edition this 'rectification' was left out. A n u m b e r of scholars now suggest that cnn is related to the repeated reference to 53 It is briefly described by B O R D R E U I L - P A R D E E 1995, 2 8 and 3 1 ; a first translation was offered by P A R D E E 1997a. 54 Cf. DE M O O R 1 9 8 7 , 1 0 9 - 1 0 ; and A S T O U R 1 9 8 8 , with a survey of previous research. 55 See now L E W I S 1996a, with excellent photographs and drawings.

c

n, 'eye', in lines 5 - 1 3 and explain the text as an incantation against the evil eye. 56 This is a well-known object of incantations in Mesopotamia. 5 7 A clear example is also found in a later Phoenician incantation against 'the coming of the big eye' and with m a n y other descripdons of the evil eye, just as in the second part of K T U 1.96. 58 Even more interesting within this comparison is that on the tablet of the Phoenician incantation we see a drawing of a demon devouring the one he attacks. In the heading the d e m o n is called mzh, 'sucker', namely of blood. This has a counterpart in K T U 1.96:4-5 which states that the demoniac power eats the flesh and drinks the blood of his victim.

5.4

Conjuring up the spirits of the dead

T h e demons to be driven out with the help of incantations are associated with death and the netherworld. As we have seen above, some of them are represented as helpers or satellites of Yam and Mot (cf. K T U 1.82 = R S 15.134:1, 5, 27, 38). It is very likely that as in Mesopotamia the people of Ugarit feared the influence of malign spirits of the dead. 5 9 From Mesopotamia we know many incantations with the object of expelling them. In Ugarit we hear more of their positive counterpart: invoking the dead to ask their advice and help. 60 This was also an act of veneration. By offering their sacrifices and honouring them by calling their names, they hoped to prevent hostilities from the dead towards the living. T h e interpretation of the texts concerned is a matter of much dispute. According to some scholars there is not enough evidence to speak of a cult of the dead. In their view there was probably no more than a funerary cult intended to offer a good burial for the deceased, helping them on their way to the netherworld; which is to be clearly distinguished from any belief in supernatural power of

56 Cf. DEL O L M O L E T E 1992a, 255-9 = 1999, 379-84, and 1992b; his interpretation is accepted by W A T S O N 1992b, 367, n. 5; W A T S O N 1994b, 237; L E W I S 1996a, 118 and W Y A T T 1998C, 375-7. T h e suffixed -n is usually explained as a deictic element, comparable to Hurrian -ne\ cf. D I E T R I C H - L O R E T Z 1990a, 104. 57

C f . THOMSEN

58

O n this seventh century incantation from Arslan Tash see

1992.

111. 59

C f . BOTTÉRO

60

Cf.

SPRONK

1983 and

1993.

1986, 145-206 and

LORETZ

1993, 287-30.

DE M O O R

1981-2,

the dead. 61 This is not the place to enter that discussion. M a n y of these texts are already discussed elsewhere in this handbook. Within the present context the survey can be confined to the elements related to the incantation texts. K T U 1.161 = R S 34.126 62 is a ritual text associated with the burial of a king of Ugarit. It reports the invocation of all possible ancestors, with many names of former kings, but also with more general indications such as 'rephaim of the earth' and 'rephaim of old'. Apparently, one was anxious not to forget any of the important deceased ancestors. This can be compared to a similar p h e n o m e n o n in the 'Genealogy of the H a m m u r a p i Dynasty' 6 3 listing all spirits of the dead that are considered important to the well-being of the living king: the royal dead, the heroes and also the spirits who might become hostile, namely 'any dynasty which is not recorded on this tablet, and any soldier who fell while on his lord's service, princes, princesses, all humanity, from the east to the west, who have no one to care for them or to call their names'. 6 4 T h e idea behind this was that the dead who remained u n n a m e d and uncared for could become a threat to the living. So especially the unknown spirits of the dead (cf. dy lydc, 'the unknown one' in R S 92.2014) were feared. Also the fact that in K T U 1.161 of the great gods only the sun-goddess Shapash is mentioned, is reminiscent of the incantations. She appears to be the most important intermediary between the living and the dead. T h e Rephaim texts ( K T U 1.20-22 = R S 3.348, 2. [019], 2.[024]), only partly preserved, seem to describe a similar invocation and actual gathering of the spirits of the dead. T h e state of the tablets hardly allows any conclusion on their function. 6 5 T h e relation with the legend of Aqhat suggests that the rephaim may have been called up by the father of Aqhat on the occasion of the burial of his son. T h e fact that the rephaim come together on the threshing floor may indicate that they could be of help in restoring the fertility which was lost at the death of Aqhat.

61

Cf. S C H M I D T 1994, 47 1 3 1 and P A R D E E 1996a. See on this text especially B O R D R E U I L P A R D E E 1 9 9 1 , 1 5 1 - 6 3 , and also the recent study of T S U M U R A 1 9 9 3 . A survey of recent research can be found with 62

LORETZ

1993,

296-300.

63

C f . FINKELSTEIN

64

Cf. on this part of the text also L A M B E R T 1 9 6 8 . Note the call for a 'minimalist' approach by P I T A R D 1992a and

65

1966. LEWIS

1996b.

K T U 1.108 = R S 24.252 can be regarded as an example of the belief in ancient Ugarit that the spirits of the dead could be invoked to help and bless the living. 66 Baal seems to be presented here as the first of the rephaim. Together with Baal and other gods these spirits of the dead enjoy a banquet presented to them in order to propitiate them. In K T U 1.124 = R S 24.272 we may find another way in which the spirits of the dead could support the living. 67 T h r o u g h a mediator they give precise advice on how to cure a sick child. If this interpretation is correct, it would offer a good illustration of the spirits of the dead acting according to the probable meaning of their name: rp'um, 'healers'.

5.5

Elements of incantations in other texts

We have already come across the phenomenon of incantations embedded in other texts. Some of these also deserve our attention. K T U 1.13 = R S 1.006 is interpreted by Dietrich - Loretz as an incantation against infertility: 68 a hymn to Anat is followed by a prayer for fertility, which is answered by a mythological fragment about Anat and Baal solving a similar problem. In particular, the urgent call for supernatural assistance (lines 23-29) can be regarded as an incantation. Note also the special role in this context of 'messengers from heaven' (ml'ak smm, lines 21-22). T h e y belong to the class of divine beings who, standing in between humans and the great gods, often play a prominent role in incantations, either as helpers or as offenders. A combination of myth and ritual can also be found in K T U 1.23 = RS 2.002. Although there is no consensus about the interpretation of this text, there can be no doubt about the relation to the question of fertility. 69 T h e text is associated by Dietrich - Loretz with the incantations because of its beginning: }iqr'a, Ί invoke (the

66

See for a survey of previous research on this 'Zankapfel der Ugaritologen' 1993, 293-5; cf. also P A R D E E 1988b, 75-118, and A V I S H U R 1994, 277-307. 67 See on this text D I E T R I C H - L O R E T Z 1990a, 205-40; L O R E T Z 1993, 289-93; and, for a different interpretation, P A R D E E 1988b, 179-92. 68 D I E T R I C H - L O R E T Z 1988b, 339-42 with a reconstruction of the ritual; cf. also DE M O O R 1980a. For a different interpretation see DEL O L M O L E T E 1981b and M A R G A L I T 1995, 231-8. LORETZ

69

C f . SEGERT

1985

and

DE M O O R

1987,

117-8.

gracious gods)'. In their opinion it is an incantation against malign forces of nature. 7 0 Also the part of the text dealing with mt w sr, 'death and evil' (lines 8-11) resembles the incantations; in particular the reference to the binding (smd, line 10) of the demon-like god points in this direction. K T U 1.114 = RS 24.258 is labelled by Dietrich - Loretz as an incantation for the medical treatment of drunkenness. 7 1 T h e text describes El drinking himself nearly to death and two goddesses finding a remedy to cure his sickness. T h e remedy seems to be described in the last lines as a recipe for humans with the same problem. 7 2 This and similar texts (cf. K T U 1.23; K T U 1.100; K T U 1.107) can teach us something about the use of myths in the daily life of ancient Ugarit. Apparently, it was believed that reciting the right story on the right occasion, combining it with the right prayer and cultic acts, had magical power. Interpreted in this way, K T U 1.114 is related to 'pure' incantations. T o this can be added that in lines 19-20 we hear of a demon-like figure threatening El in his drunkenness (lines 19-20). This hby is described as 'the one with two horns and a tail'. T h e name itself can be translated as 'crawler' and seems to refer to a scorpion." In Ugaritic incantations this is a common indication of the evil force to be expelled. In the myth about the moon-god Yarikh obtaining his bride Nikkal ( K T U 1.24 = R S 5.194) we find some elements related to the incantations in the second section of the text (lines 40-50), which is separated from the rest of the text by a horizontal line. It concerns a hymn to goddesses called the Kathirat, daughters of the new moon Hilal. T h e y are described as birds settling down on the trees. T h e singer calls them by their names, stating that 'their list' (mnthn] cf. the use of mnt in K T U 1.82:20 a n d K T U 1.100) is on his lips. According to this text the Kathirat can be regarded as lower goddesses who are especially related to marital affairs. Their status is comparable to that of demons and (deified) spirits of the dead, who just like the Kathirat, are often compared to birds. 74 Calling their

1988b, 3 5 0 - 7 . 1988b, 342-5; cf. also P A R D E E 1988b, 13-74; 1997a; and C A T H C A R T 1996; W Y A T T 1998c, 4 0 4 - 1 3 . 72 Cf. W A T S O N 1990a. 70

DIETRICH -

71

DIETRICH

-

LORETZ

LORETZ

73

C f . CATHCART

74

C f . SPRONK

1996,

1986,

5.

100.

PARDEE

names at the end of the myth about a divine wedding probably functioned as a way to invoke their blessings on the occasion of a h u m a n wedding. For the sake of completeness another two texts deserve some attention. In the second edition of K T U the genre of tablet 1.86 = R S 18.041 is indicated as 'myth?, ritual?, incantation?'. 7 5 Recently, it has been demonstrated that it is likely that we are dealing here with a hippiatric text about breeding. 7 6 K T U 1.93 = RS 19.054 is listed by de M o o r a m o n g the incantations, interpreting it as a prayer to Anat for help in reciting his incantation properly, that is, without stammering. 7 7 If this interpretation of the short and damaged text is correct, 78 then it would illustrate the importance of incantations in the religious life of the people of Ugarit.

75

KTU\

76

C f . DEL O L M O L E T E -

77

DE

78

106. MARQUEZ

1987, 186-7. For other interpretations see

ROWE

1995.

MOOR

CAQUOT 1989, 3 7 - 9

and the literature listed there.

C H A P T E R SEVEN

THE UGARITIC CULTIC TEXTS

1

Paolo

1.1

T h e

M e r l o

Rituals

-

Paolo

X e l l a

The problem of the ritual documents

T h e Ugaritic texts which can be classed as ritual texts or have a ritual background did not at first attract the attention of scholars to the same extent as the mythological texts, on which an enormous bibliography has emerged. T h e very formal characteristics of these documents have contributed to discouraging any approach to them. T h e y are mostly schematic texts, written in a concise and technical language which proceeds by allusions, using an accurate and precise liturgical vocabulary which cannot be studied with the help of, for example, the parallelism characteristic of poetry. In addition, there is the frequent use of stereotyped formulae, the understanding of which depends on actually identifying the rites to which they allude. Lastly, the focus of the syntax is extremely difficult because 'prescriptive' rubrics and 'descriptive' sections alternate with no obvious criteria and are accompanied by long lists of gods followed (often asyndetically) by the victims or offerings intended for them. A good stimulus to the study of ritual texts, which have been neglected for so m a n y years (apart perhaps from a couple of specific contributions) 1 came from the article by Levine 2 on the possibility of identifying prescriptive and descriptive rituals, even if this distinction now seems applicable only to a limited extent. This is either because 1 9 5 5 . Cf. also U R I E 1 9 5 9 . 1963. The distinction he proposes between 'prescriptive' rituals and 'descriptive' rituals is only acceptable in broad outline since it is clear that even a description is significantly prescriptive in nature (libretto for ceremonies; cf. the use of the imperfect/future). It is, thus, a spurious problem. Cf. also L E V I N E 1965; 1974, 8ff. and 1983. 1

DE

2

LEVINE

GUGLIELMO

several texts do not, strictly speaking, belong to either of the two categories or else because the descriptive texts are also standard and are also often devised and written down as m e m o r a n d a . However, the situation has improved markedly over the last twenty years, which have seen the first monographs on these documents as well as a series of minor studies devoted to analysing the lexicon, structure and function of the various texts, their implications for the history of religions, the divinities involved in the rites, the typology of the rites, etc. If we limit ourselves here only to wide-ranging works, first of all comes the publication in 1979 of a long and excellent comprehensive t r e a t m e n t of the ritual texts in the Supplément au Dictionnaire de la Bible written by Caquot. 3 T h e n , at the beginning of the eighties, there appeared the first monographs on the topic by de Tarragon 4 (which is more discursive) and by Xella 5 (which is more systematic). Besides a large n u m b e r of other minor studies (which appeared chiefly in Ugant-Forschungen), a new wide-ranging contribution appeared in the section written by de T a r r a g o n in the book edited by Caquot, de T a r r a g o n himself and Cunchillos (TO 2) which was published in 1989 (although written a few years earlier) even though it did not provide any remarkable new interpretations (in a few rare instances even marking a regression). With del O l m o Lete's monograph, 6 substantial progress has been achieved in spite of the way the author has chosen to present the material (in practice it is a comprehensive study of the religion of Ugarit). In this survey can also be noted the translations m a d e by Dietrich and Loretz of a certain n u m b e r of ritual texts in the collection TUAT and elsewhere 7 as well as the inclusion of several cultic documents in the anthology of de Moor 8 and in the study by Pardee of texts which he curiously called 'para-mythological'. 9

1.2

Classifications of form and content

T h e ritual texts supply direct and extremely valuable information about religious practices (chiefly but not exclusively royal and pub3

CAQUOT

4

D E TARRAGON

5

XELLA

6

DEL OLMO

7

DIETRICH -

8

DE

9

PARDEE

1979. 1980.

1981.

MOOR

LETE

1992a

LORETZ 1987.

1988b.

=

1988b,

1999. 1990a.

lie) of ancient Ugarit, even if we reiterate that the material is difficult, to be approached with caution, using clear methodological principles. Apart from attempts to consider these documents as a specific 'literary form', the main fixed points of departure for analysis are as follows. In terms of the history of religions, they have an undoubtedly cultic character, while in terms of linguistics, they belong to a form of linguistic expression that can be defined as 'chancellery language', 10 which places them on the same level as the economic and administrative texts and the hippiatric texts. As for their being documents with a religious purpose, the ritual texts have to be studied against the background of all that we know about the beliefs and cults of Ugarit during the Late Bronze Age, with the proviso of also using, to the greatest extent possible, the archaeological data from Ras S h a m r a and Ras Ibn Hani in order to provide a better framework as to how the rites functioned, and who were their participants and the recipients of the various ceremonies. O f course, local mythology also has to be a constant and fundamental reference point for understanding the rites and their underlying ideology. T h e objection (which is often raised) that the 'literary' texts reflect a stage of Ugaritic religion which is different and older (than the ritual texts) in fact has very little foundation, for it is based only on the formal opposition between poetic language and the nonpoetic or non-literary language" in which the ritual texts are cast. In terms of the history of religions it is true that in some cases there is a certain divergence, for example between the rank and personality of the divine figures described in the myths and those venerated in the cult. However, the history of religions teaches us that the mythic dimension has its own laws, times and coordinates, such that the actions or features of a god in the mythological narratives need not necessary have an exact equivalent in the ritual universe regulated by man in terms of his own needs (even such banal and practical needs as the eating and [re-]distribution of meat by means of ritualizing immolations and celebrating religious feasts). T o cite only one striking case, it is precisely in the 'later' ritual texts that the god El—incorrectly considered by some to be in decline with the passage of time—occupies a position of undoubted pre-eminence over all the other deities. 12

10

FRONZAROLI

11

LIVERANI

12

Cf.

XEI.LA

1975.

1964.

(forthcoming).

If, instead, we consider the content of the texts rather than their outward form, mythology and culdc documentation comprise two complementary and fairly organic aspects of Ugaritic religion which, as in all the cultures of the ancient world, is expressed differently at different levels. Lastly, as regards the matter of the relationships between myth and ritual, which are inextricably connected, it is sufficient to r e m e m b e r the existence of texts such as K T U 1.23 = R S 2.002, correctly defined as 'cultic myth'; 1 3 or, the fact that recitations, prayers, exorcisms and incantations are religious acts which can all be projected into the mythic dimension (while the recitation of a myth is itself a rite!). Besides these considerations, another important fact to keep in mind in the study of the Ugaritic rituals is the comparative dimension within the religious traditions of Syria-Palestine and Northern Mesopotamia. With the increase in our knowledge of the religion and sacrificial system of centres such as Ebla, 14 Emar 1 5 as well as Mari (with the requisite changes), 16 it is necessary to get away from the usual and repeated references to the Bible in order to reconstruct the religious tradition peculiar to Syria, the consistency and essential continuity of which are perceptible. It should also be r e m e m b e r e d that additional and often direct information on the cult and on various rites, both sacrificial and non-sacrificial, can be gained from other types of text, i.e. the economic and administrative texts, 17 the letters and the mythological texts. T h e most famous example in the last category comprises the rites carried out by Keret ( K T U 1.14 = R S 2. [003]+ ii 50ff. and parallels) which, though to some extent awaiting proof, are undeniably connected with cultic practice. However, it is understandable how, faced with a mass of documents sharing a definite (and more or less direct) connection with the cult, but without uniform formal characteristics, specialists have attempted to sort this material, proposing classifications and subclassifications of various kinds. In a monograph written in 1 9 8 1 ( X e l l a 1 9 8 1 ) one of the authors set out a subdivision of the texts as follows: 1) monthly liturgies and 13

14 15 16 17

D E L O L M O L E T E 1992a, 15 = 1999, 15. Cf. in general F R O N Z A R O L I 1993 and P O M P O N I O - X E L L A 1997. Cf. especially F L E M I N G 1 9 9 2 and, for example, D I E T R I C H 1 9 9 0 . An excellent synthesis in D U R A N D 1 9 9 5 . DEL OLMO LETE S A N M A R T I N 1998. On these aspects see below.

lists of offerings; 2) divination texts and oracles; 3) prayers; 4) incantations; 5) atonement sacrifices; 6) liturgies for kings; 7) votive texts; 8) lists of gods. 18 This proposed classification has largely been followed by G . del O l m o Lete ( d e l O l m o L e t e 1992a = 1999) with a series of additional subdivisions which refine the original grid still f u r t h e r , even for example distinguishing prescriptive rituals into sacrificial or non-sacrificial, pure or mixed, single or multiple, etc. 19 T h e table proposed by the Spanish scholar is undoubtedly a good theoretical grid for classification, provided that it is not taken rigidly as a formal reference point. In this Handbook, which has a more general approach, we shall retain the distinction between prescriptive and recited rituals. Whether the sacrificial action does or does not involve blood is a further classification which, in some sense, cuts across the others and can connote a n u m b e r of other ceremonies. Likewise, whether the ritual actions are more or less complex, the length of time they take and where they are performed, are factors which do not alter the basic typology. Ultimately it has to be said that our as yet imperfect knowledge of the sacrificial vocabulary of Ugarit and our continuing ignorance of the deep structures of Syrian religion in the Bronze Age should put us on guard against claims of elaborate classification which are too detailed and too systematic.

1.3

Rite and sacrificial material: terminology and

typology

As indicated above, to describe or record various actions of the cult, the Ugaritic ritual texts use a series of technical terms which sometimes have a more generic meaning and sometimes are very technical. Even if we are still very far from having resolved all the problems of interpretation, the meanings of some terms have now been determined with enough certainty. 20 T h e word dbh is a prime

X E L L A 1 9 8 1 ; in fact there is also a section on Hurro-Ugaritic rituals, due only to the bilingualism marking these documents, recently studied in a systematic way particularly by M. Dietrich, W. Mayer and M. Dijkstra in various publications (chiefly UF and A O A T / A L A S P ) . 19 A different approach has been adopted by D I E T R I C H - L O R E T Z 1988b, 300ff., which however is also reductive because it is determined by the anthological nature of the series (TUAT) in which it appears. 20 X E L L A 1989; D E L O L M O L E T E 1992a, 19-22 = 1999, 20 1; 1995. Nevertheless there are still quite a few disputed or even completely enigmatic cases, for example, 18

example of a term to denote the sacrificial ceremony which is both specific and generic at the same time. O f as yet unknown etymology, 21 dbh (generally attested as a noun although it also occurs in verbal form) is a key term in the Ugaritic sacrificial lexicon. It has been suggested that a whole series of terms (for example,ytn, kbd, nkt, np, nrr, nša, ntk, ql, qrb, šlm, št (?), t'y) should be considered essentially as its synonyms, 22 but each of t h e m — a p a r t from proven cases of functional synonymity—is of course distinct and must be investigated separately as a specific element of a highly technical lexicon. In fact, dbh is a polysemantic term, as noted already by Aistleitner, 23 who correctly distinguished between (a) the meaning 'schlachten/Schlachtopfer' and (b) the meaning 'Gastmahl essen/Mahl'. T h e Ugaritic term is not semantically related to Hebrew zbh (verb and noun) but instead to Akk. zību, 'food offering', which in turn is synonymous with naptānu (originally an accounting term used for food rations, used in the cult but also in non-cultic texts). T h e so-called polyglot vocabulary of Ug 5 (137 iii 6)24 gives the equivalents EZEN = i-si-nu = e-\l]i = da-ab-hu, from which can be deduced the clear meaning 'feast', 'soirée'. This meaning is abundantly confirmed by the use of the term in the ritual texts. It is a sacrificial meal, i.e. a sacred banquet, as has been proved by several terms which are parallel or actually synonymous to it—dgt, msd, trm, esrt. However, the meaning of the word sometimes seems to be more generic (precisely 'feast' in general), 23 since dbh sometimes includes not only foodstuffs (for example, cf. oil and honey in K T U 1.41 = RS 1.003+:20-l) but also objects, clothes or metals. T h u s it is an offering in a generic sense even if, as seems to be the case, the gift par excellence to the gods is still food (cf. the implicit equivalence dbh — kispum in K T U 1.142 = R S 24.323). 26 An excellent example of a grandiose and complex dbh is provided by K T U 1.148 = R S 24.643, 27 the dbh of Sapunu, which seems to

cf. terms such as iyn, m'rb (in spite of its apparently clear etymology), sin, trmt and others as well. 21

FRONZAROLI

1965.

1992a, 2 0 - 1 = 1999, 21-3, although he notes that the synonymy refers only to the basic meaning of 'to offer'. 22

DEL OLMO

23

AISTLEITNER

24

Cf. H U E H N E R G A R D 1987b, 117; VAN S O L D T 1991a, 303. Cf. § 4.5. Cf. for example the opening line of K T U 1.161 = RS 34.126: spr. dbh. zlm. Cf. DLU, 128a for the references. D E L O L M O L E T E 1992a, 88-91 = 1999, 129-33. (with previous bibliography).

25 26 27

LETE

1974, §

722.

include a whole series of different rites within the ceremony. In this case, however, it seems that the iZmm-sacrifice does not form part of the general typology of the dbh, but it cannot be excluded completely that in the first section of this text the offerings are to be considered as performed as a /r/?-sacrifice (which was not mentioned because it was implicit). A text such as K T U 1.170 = R I H 7 8 / 1 1 shows that the dbh included sacrifices such as the srp, as is also indicated by the semantic field of the Hurrian term athl, sometimes used as equivalent to dbh and sometimes as equivalent to srp.2* If, out of several possible cases, we consider the so-called atonement ritual K T U 1.40 + 84 + 121 + 122 + 154 = RS 1.002,29 significandy, the terms dbh and tc denote the inner cultic sphere in which the Ugaritic homo religiosus has committed sin. Above we have referred to attempts at formal classification of the ritual texts. Without going into over-precise subdivisions, we note that the ritual actions can be simply classed into the following categories: 30 ritual blood sacrifices, bloodless sacrificial offerings, processions, enthronements-investitures, cultic meals. T o these can be added texts or parts of texts comprising recitation (including prayers). T h e first category is thus represented by ritual actions involving blood. A m o n g these, the designation attested most often in the rituals of Ugarit is provided by the brace of terms srp and šlmm, which almost always occur together in close succession. These are two specific terms for sacrifice, the first of which is evidently connected with the action of 'burning' the victim (as the root *srp shows) as is the case in the other Semitic languages (although this does not mean that it was always a holocaust); the second term, šlmm, is possibly to be translated 'communion sacrifice' or 'peace sacrifice', based on etymology and also on comparison with H e b r e w šelamîm.U Explicit confirmation that (with the requisite modifications) in Ugarit also it is a communal meal of the offerers, comes from K T U 1.115 = R S 24.260:9-10: š l il bt šlmm kl l y Ihm bh, 'a ram to the god of the (royal) house as a //mm-offering; all eat it'. 32

2R

DIETRICH -

MAYER

1995,

12-3.

1981, 251-76; DEL O L M O L E T E , 1992a, 9 9 - 1 0 9 = 1999, 1 4 4 - 6 0 ; DE M O O R - SANDERS 1991; W Y A T T 1998C, 342-7. 30 D E T A R R A G O N 1980, 55-129; DEL O L M O L E T E 1992a, 28-32 = 1999, 3 4 - 4 0 ; cf. also DEL O L M O L E T E 1995, passim. 31 Current views and bibliography in DEL O L M O L E T E 1992a, 3 0 - 1 and n. 81 = 29

XELLA

1999, 32

37-40

XELLA

and 1981,

n.

82.

1 0 8 ; DEL O L M O L E T E

1992a,

177

81

= 1999, 257-64.

Also part of the other actions of the blood ritual is the act of 'immolation' (nkt) or 'slaughter' (qll) of the victim. Nevertheless, the ritual texts provide evidence of further types of sacrificial offerings of which the meaning is not always precise, as in the cases of šnpt (related perhaps to Hebrew tenûpâ),33 tzg (which may denote either the victim or the rite)34 and also t'y, mentioned already, a polysemantic term which also seems to denote an offering connected with the royal cult. 35 Besides rites involving blood, the Ugaritic texts also mention bloodless ritual actions. A m o n g the ceremonies which do not involve the sprinkling of blood can be mentioned here the fairly widespread practice of performing 'libations' (the verb *ntk or the derived noun mtk), especially of water and wine, documented in several texts (cf. for example, K T U 1.41 = RS 1.003+:12; K T U 1.107 = RS 24.251:46; K T U 1.119 = R S 24.266:25), although they provide no information about the details of the ceremonies. T h e bloodless offerings also include, of course, offerings of vegetable food, and of various types of object, metal or cloth, which are amply documented in the ritual texts as well as in the economic and administrative texts. 36 A further category of ritual actions comprises the processions, identifiable by a series of terms (for example, yrd, hlk, Iqh, cly, crb, etc.) and correcdy included by G. del O l m o Lete among the 'non-sacrificial liturgies'. 37 T h e processions mentioned in the Ugaritic rituals have in fact as a central point a series of ceremonial actions in honour of divine statues, 38 in which, as usual, the role carried out by the king and his family is completely in the foreground. In this connection, the ritual K T U 1.43 = R S 1.005 can be mentioned which is focussed completely on the procession of divine statues. 39 It begins with k t'rb* 'ttrt. hr. gb/bt mlk 'When 'Attart of Hurri 4 0 enters the gb of the royal palace', and has the important conclusion (lines 24-26): mlk. ylk. Iqh. ilm || atr. ilm. ylk. penm./mlk. p*c*nm. yl[k]/sbc pamt. Iklhm, 'the king will go to take (the statues o f ) the gods || behind (the stat33

34

MILGROM

1972.

1981, 39-40. 35 D E L O L M O L E T E 1988a; cf. also F R E I L I C H 1992. 36 D E L O I . M O L E T E - SANMARTIN 1998; cf. in general 1999, passim. 37 D E L O L M O L E T E 1992a, 96 = 1999, 1 4 0 - 1 . 38 D E T A R R A G O N 1 9 8 0 , 9 8 - 1 1 2 ; in general D I E T R I C H 39 D E L O L M O L E T E 1992a, 189-94 = 1999, 282-91. 40

XELLA

C f . BONNET -

XELLA

1996.

DEL O L M O L E T E

-

LORETZ

1992.

1992a =

ues o f ) the gods he will go on f o o t / t h e king will go on foot/seven times with all of them'. Further references to ceremonies of transporting statues are also given in K T U 1.91 = R S 19.015, a key text (classed as administrative) which lists concisely various rites which can be identified in more detail in other documents. 4 1 Line 10: k t'rb. cttrt. šd. bt m[lk] 'When 'Attart of the steppe enters the royal palace' (cf. K T U 1.148 = R S 24.643:18-22) and in line 11: k t'rbn. ršpm. bt. mlk, 'When the Rašaps enter the royal palace' (cf. K T U 1.106 = R S 24.250+, as well as K T U 1.43 = RS 1.005:9ff.). K T U 1.112 = R S 24.256: 42 6 - 9 , where images of the deities (genuine gods and divinized ancestors) are moved around in various cultic installations as part of the dynastic cult, is not exactly of the same type. Another category of ritual actions is that of investiture-enthronement.4:i Here we are referring, for example, to actions described by verbs such as ytb 'to sit (down)' or lbs 'to dress/be dressed' (for the semantic field of clothes or ritual dressing cf. especially azr and nps, as well as other terms) but which, in a religious context, refer to specific ceremonies the details of which are unknown to us. A clear case of royal investiture can be identified, for example, in K T U 1.41 = R S 1.003+:53-5, sbu. špš whl mlk/w l*[bš]n. spm. w mh[pn]h* t*[t]tbn/b. b*\t\ w km. ity[šu. L] šmmyd[h] 'the sun sets and the king is desacralized and, robed splendidly and with his face cleansed, they shall enthrone him (lit. they shall make him return) in the palace and once there, he shall raise his hands towards the sky'. 44 Although it mentions the royal throne, the expression 'and at night the throne is prepared' alludes to a ritual action which is not completely clear in K T U 1.106 = RS 24.250+:26~8 (w III tcr[k] ksu). Important moments in the liturgies, apparently without involving blood, can be recognized in the invocations, recitations and prayers (cf. for example dn (?), nsa + yd, pth. + yd (?), sh, qra, šr), in acts of divination and oracles (phy (?), rgm, šr, ttb + rgm, etc.) and in expiation rites (cf. K T U 1.40 + 84 + 121 + 122 + 154 = RS 1.002, cited already and marked by its peculiar structure and lexicon).

1979; DEL O L M O L E T E 1992a, 173-7 = 1999, 257-64. Rites to be celebrated in the month of tjyr (likewise K T U 1 . 1 0 5 = RS On the passage quoted cf. X E L L A 1 9 8 8 . 43 D E L O I . M O L E T E 1992a, 96 = 1999, 1 4 1 . 44 D E L O L M O L E T E 1992a, 83-5 = 1999, 122-5. 41

42

XELLA

24.249).

Finally, a m o n g the ritual actions, communal meals can be mentioned (cf. what has been said above concerning dbh and šlmrrì) which frequendy acquire religious significance, and include the consumption of drinks (specified by terms such as kly, Ihm, nsl (?), (db, Cšr, sty), either as a convivial occasion which joins together men, the gods and the spirits of the ancestors, or else as a ritualized and regulated occasion where food (especially meat) is eaten. As for the sacrificial material, 45 in rites involving blood, the offerings of animals were substantially similar to those known in the other religious traditions of the ancient N e a r East. T h e animals offered most frequently are bovidés (the ox, alp, the bull, tr, the cow, gdlt, lit. 'female head of cattle'), ovines, denoted genetically as sin (the ram, i, the sheep, dqt, lit. 'head of small catde' or tat, she-goat, cz), birds, called generically csr, with the dove, ytnt or the turtle-dove, tr specified. However other kinds of animals also occur such as donkeys, V and also fish, dg. Besides complete animals, the various parts of the victim were offered, limbs and entrails (ap, nps, lb, kbd, mtnt are the easiest to identify) as studied in detail by G. del O l m o Lete. 46 Animal offerings are certainly not the only ones attested in the Ugaritic rituals and several times gifts were dedicated to the gods either of vegetables such as wine (yn), oil (šmrì), e m m e r (ksm), flour (qmh) honey (nbt), or else of precious metals such as silver (ksp) and gold (}}rs) or even objects in c o m m o n use including articles of clothing and crockery.

1.4

The role of the king

By now it is well known that the king had a role of particular importance within the Ugaritic cult and was by far the principal officiant. 47 T h e texts of Ugarit, in fact, show not only how the palace has control over the personnel appointed to the cult but how the king himself is often the main celebrant within a liturgy and how frequently the ceremonies take place in locations and internal buildings actually within the royal palace. Moreover, the importance of the dynas-

45 Cf., for example, 1999, 40-2.

1980, 31-54;

DEL O L M O L E T E

1992a, 32~3 =

L E T E 1989b. 1984; Y O N 1985; D E L O L M O L E T E 1992a, 115-95 = 1999, 166-291; L E T E 1993b; A B O U D 1994, 123-92.

4,1

DEL OLMO

47

HEALEY

DEL O L M O

DE T A R R A G O N

tic cult together with the veneration of the divinized royal ancestors has emerged as one of the most typical aspects which characterize the religion of Ugarit in a peculiar symbiosis between the living a n d the d e a d for the c o m m o n good. 4 8 T h e high n u m b e r of rituals f o u n d in Ugarit testify to a series of liturgies where the ritual actions clearly refer to the king, expressed frequently by the use of fixed 'ritual formulae' which indicate his state of (ritual) purity a n d refer to special m o m e n t s in the day. 49 T h e best known and best attested form of ritual action where the king is seen as the protagonist is the one concerning his ritual purification. This must certainly have been achieved by means of special ablutions, expressed by the f o r m u l a yrths mlk bn 'the king washes < a n d is> purified'. This formula often introduces a series of ceremonies in which the king takes part, at the close of which there occurs, connected with nightfall or sunset (sba/u špš, crb špš) the further formula whl mlk 'and the king is desacralized'. T h u s the purification of the king seems to be a prerequisite for him to be able to p e r f o r m , assist at or take part in the ritual (cf., for example, K T U 1.119 = R S 2 4 . 2 6 6 : 4 - 6 ; K T U 1.112 = R S 2 4 . 2 5 6 : 1 0 - 5 , etc.). A n o t h e r series of 'formulae' which are quite similar to each other even if not exactly identical, allude to the king's role in oracle practice (mlk brr rgm y ttb/rgm y ttb mlk bn, ttb rgm whl mlk) probably acting as mediator for the replies a n d always in conditions of ritual purity. 0 " T h e r e are m a n y other examples of the sovereign being involved in the cult. A m o n g the most i m p o r t a n t a r e K T U 1.119 = R S 2 4 . 2 6 6 : 1 3 - 4 in which 'the king sacrifices in the temple of ΕΓ, or K T U 1.164 = R I H 7 7 / 2 b + ' W h e n the king sacrifices in the hmn (i.e. the palace chapel)'; the king takes part in processions, as in K T U 1.43 = R S 1.005:23-5, cited already; a cantor has to sing in front of the king, w h o has his h a n d s spread out: K T U 1.106 = R S 24.250+: 15-7; the king's throne is p r e p a r e d at night, probably to allow him to be seated d u r i n g the p e r f o r m a n c e of nocturnal rites: K T U 1.106 = R S 2 4 . 2 5 0 + : 2 7 - 8 , a n d so on.

48

Restated by D E L O L M O L E T E 1992a, 130 4 = 1999, 192-8 (in respect of K T U 1.161 = RS 34.126) and passim, where five whole chapters (3- 6) are devoted to the religion and cult of the king. 4 " X E L L A 1984c. 50 D E L O L M O L E T E 1992a, 22 = 1999, 24.

T h e ceremony described at the end of K T U 1.41 = R S 1.003:505 is particularly interesting; here the cultic role of the king is very obvious: the rite, which is still difficult to identify exactly, is performed in the month of rišyn and takes place on the terrace of the royal palace (it is less likely that it was a temple); it exhibits remarkable similarities with the H e b r e w ritual of the New Year which was also celebrated on the day of the full moon in the first month of the year, at the season of the wine harvest, with the construction of huts. 52 51

Also worthy of mention is the possible sacred marriage ceremony mentioned in K T U 1.132 = R S 24.291, studied again recently (with new proposals) by Dietrich and Mayer, 5 3 in which it seems that the union of the king with the goddess Pidrayu was celebrated. She was the daughter of Baal and evidently must have embodied the ideal wife at the highest level. 54 Besides the rites in which the king took part, the locations where the liturgy took place were often included within the palace complex and the chapels attached to it. O n e of the places in which ritual actions involving the king's presence were often performed is the hmn, undoubtedly a chapel reserved for the royal and dynastic cult, a raised and covered structure, a sort of earthly projection of the heavenly dwelling, which forms part of the royal palace and was used chiefly for the royal dynastic cult. 55 T h e r e are other places in which the rituals were performed and probably must have belonged to the king's residence such as the 'terrace' (gg) mentioned above or the 'garden' (gn). This second term, which is also a name for a m o n t h in the Ugaritic calendar when an impressive n u m b e r of ceremonies are performed, is already attested in the texts from Ebla' 6 and denotes a cultic area inside the royal palace which was a sort of cemetery, the location for funerary rites intended for the ancestors and the chthonian deities. 37

51

Lines 50-55 of this tablet are separated from lines 1-49. Only the latter have a duplicate in K T U 1.87 = RS 18.056. 52 D E L O L M O L E T E 1992a, 83-5 = 1999, 122-5. 53 D I E T R I C H - M A Y E R 1996a; cf. the new proposal to read bb*t mlk instead of bit mlk in line 3. 54 D E L O L M O L E T E 1992a, 1 4 3 = 1999, 212. 55 D E L O L M O L E T E 1984d; X E L L A 1991, I69ff. and passim. 5,I X E L L A 1995c. 57 Cf. for example S P R O N K 1986; L E W I S 1989; D E L O L M O L E T E 1992a, 149-56 = 1999, 219-32. For a more reserved critical attitude cf. VAN DER T O O R N 1991.

T h e reference to gn brings us to the ceremonial role of the king in his capacity as principal celebrant, a role exercised even more within the funerary cult in honour of the dead and deified kings of the dynasty. This fundamental aspect of Ugaritic religion is attested in a whole series of indications and texts (three, in particular K T U 1.108 = RS 24.252, K T U 1.113 = RS 24.257 and K T U 1.161 = RS 34.126 which, although their general meaning and many details are still disputed, throw considerable light on the phenomenon which seems to be peculiar to the Syrian region). T h e first of these documents ( K T U 1.108 = RS 24.252) 58 seems in fact to attribute to the dead king the title of rpu mlk clm 'Rapiu, eternal king' (lines 1 and 21-22). Unfortunately, the break and the theoretical possibility of attributing the title rpu to a god (in this case Baal) rather than to the king makes the interpretation of the whole text very uncertain. K T U 1.113 = RS 24.257 (see § 7.2) is a clear testimony that the dead kings were considered as divine. Lastly, K T U 1.161 = RS 34.126 59 is introduced as is known as the 'liturgy of sacrifice of the shades [= protective spirits]' (spr dbh zlm), i.e. a 'libretto' of the funeral celebration which was performed to accompany the descent of the dead king into the underworld. T h e liturgy begins with the invocation of the Rapiuma, the ancestors belonging to the dynasty, after which we are present at a lament with a ritual meal followed by the c o m m a n d given by the goddess Šapaš to the dead king (here probably Niqmaddu III, line 26) to descend into the underworld. T h e text then continues with the sacrifice of a bird (probably as an offering for the dead king) performed seven times and it ends with a propitiatory blessing of well-being towards the king, the ruling house and the whole city of Ugarit. It is precisely this final blessing which shows us the purpose of the cult of the ancestors which the Ugaritic kings carried out: it expresses in fact the concern for obtaining prosperity, protection and oracular responses 1 '" from the Rapiuma, fittingly honoured for that purpose. 61 Finally, it can be noted that the role of the royal house in Ugaritic liturgy does not seem to have been strictly confined to the king. 1988b, 75-118; ( C A Q U O T - ) DE T A R R A G O N (CUNCHILLOS) 1989, 1 1 1 - 8 ; 1992a, 126-30 = 1999, 184-92, W Y A T T 1998C, 395-8. 59 D E L O L M O L E T E 1992a, 130-4 = 1999, 192-8, with previous bibliography. See also W Y A T T 1998C, 4 3 0 - 4 1 . 60 T R O P P E R 1989a; L O R E T Z 1993. 61 T h e 'minimalist' position recently adopted by P A R D E E 1996a, 281 does not seem completely justified. 58

PARDEE

DEL O L M O L E T E

Already K T U 1.112 = RS 24.256 studied above shows that the whole royal family took part in ritual actions. In addition, K T U 1.170 = R I H 7 8 / 1 1 , even in a broken context, seems to show the existence of a sacrifice carried out by the queen, which is confirmed by the administrative text K T U 4.149 = R S 15.039:14-6 which records quantities of wine for the sacrifice by the queen on sown field.62

1.5

Other cult personnel

Besides the foreground role performed by the king as officiant, we also know of other personnel appointed to the cult even if, surprisingly enough, neither khnm nor rb khnm ever feature in the ritual texts, apart from a the doubtful reference to a 'throng of priests', tltt khn[m], in the second broken incantation against serpents' venom, K T U 1.107 = R S 24.251 + : 18. Otherwise note can be taken of a 'cantor', sr, in K T U 1.106 = R S 24.250+: 15, a qdl χη K T U 1.112 = R S 24.246:21, a category which can perhaps be defined as 'purifiers', mhllm in K T U 1.119 = RS 24.266:23, whereas a certain exorcist termed mihi is the h u m a n protagonist in the great incantation against serpents, K T U 1.100 = R S 24.244. All this personnel belongs to the category of dependents of the royal house (bnl mlkf3 and therefore, strictly speaking, a class of priests independent of the king is not attested in Ugarit. 64 T h e rare mention of personnel with religious functions in the cultic texts is partly completed by references which can be derived from the administrative and economic texts, 65 although there is no mention of their cultic duties. In this category of documents we find several references to 'priests', khnm (in the Akkadian texts: l u . m e s s a n g a ) ; individually or collectively, in 'community', dr khnm ( K T U 4.357 = R S 18.046:24), with their 'chief' rb khnm (in the Akkadian texts: l ù . u g u l a 66 s a n g a ) , a title which occurs also on the famous hatchets and in the colophon to the 6th tablet of the Baal cycle mentioned above.

62

Already noted in X E L L A 1 9 8 1 , 1 4 9 . Cf. for example H E L T Z E R 1982, 131-9; LIPINSKI 1988. 64 According to the colophon of the tablet K T U 1.6 = RS 2.[009]+ vi 55-7, even the high priest Attanu-prln was a dependent of king Niqmaddu, albeit at the highest level. T h e title nqd can be related to the cult only hypothetically. 63

65

DEL

66

See most recently

OLMO

LETE -

SANMARTIN BORDREUIL

1998, 1998.

176-84.

T h e r e are also the 'consecrated ones', qdšm, for w h o m a divinatory function has also been proposed, hypothetically, 67 and then the 'singers', šrrn, personnel connected in various ways with music in the cult and perhaps also outside the cult and lastly the 'water carriers of the sanctuary', šib mqdšt, a function which has parallels in the Hebrew world, in Hcllenistic-Roman Syria and elsewhere.

1.6

The times of the celebrations

As is almost the general rule in the ancient Near East, the calendar 6 8 of Ugaritic liturgical celebrations is based on the lunar cycle and therefore the days in which the cult takes on greater importance are evidently at the beginning or middle of the month, corresponding to the new moon (ym hdt, i.e. the first day of the month) or the full moon (ym mlat, i.e. the fifteenth day of the month). O n these days great sacrifices are carried out, accompanied by the usual rite of purification of the king, as shown for example by K T U 1.46 = RS 1.009, K T U 1.109 = R S 24.253 and K T U 1.41 = R S 1.003+. However, there are also other liturgies for other days of the month, often in 'weekly' cycles (i.e. a quarter of a month), but also on several other days of the month. And then some rites took place in the evening or at night, as can be deduced from the indications I II and Ipn II. With regard to the annual cycle of the cult, however, unfortunately we are not yet completely certain of the exact sequence of the months of the Ugaritic calendar since the local names are considerably different from the standard names of Mesopotamian tradition. And comparison with what we know of the months of various Eblaite calendars from over a millennium earlier does not help much either. Some scholars have tried to reconstruct a particular seasonal liturgical sequence on the basis of the mythological texts, 69 but this attempt has not been accepted by scholars. As far as the strictly ritual texts is concerncd, we know of liturgical texts related to certain

S A N M A R T I N 1 9 9 8 , 1 8 0 - 1 . On the root cf. X E L L A 1 9 8 2 . 1980, 17-30; D E L O I . M O L E T E 1992a, 2 2 - 4 = 1999, 2 4 7 ; C O H E N 1993, 3 7 7 8 3 ; A R N A U D 1993b. For chronological indications to the cult which can be obtained from the economic and administrative texts cf. DEL O L M O L E T E - S A N M A R T I N 1998, 186-8. 6fl Especially DE M O O R 1971. Cf. also DE M O O R 1972 and 1988a. 67

DEL OLMO

68

DE

LETE -

TARRAGON

specific m o n t h s such as the m o n t h s of ib'lt ( K T U 1.119 = R S 24.266:1-17), nql ( K T U 1.138 = R S 24.298) and hyr, the second m o n t h of spring corresponding to A p r i l / M a y ( K T U 1.105 = R S 24.249 and K T U 1.112 = R S 24.256). In addition we are acquainted with the ritual of the m o n t h of ris yn ( K T U 1.41 = R S 1 . 0 0 3 / K T U 1.87 = R S 18.056). This last m o n t h ('first of the wine') fell in the time of the wine harvest and corresponded to the month S e p t e m b e r / October; it was probably the beginning of the cultic year in Ugarit. Its ritual, with an extremely long list of sacrifices, is marked by an initial and highly symbolic offering of a bunch of grapes, undoubtedly to be understood as firstfruits, to the god El.

1.7

The places for the celebrations

T h e Ugaritic liturgy certainly did not only take place in the temples. In fact the ritual texts mention several sacred places, some of which have been confirmed from archaeology whereas others still remain unidentified. 70 For reasons of completeness it is necessary to note, however, that certain 'cult places', identified on the basis of archaeological evidence, actually have no equivalents in the specific terms of the ritual texts. 71 T h e temple of Baal, identified with one of the two great temples on the acropolis, takes on a foreground role in the ritual texts ( K T U 1.119 = R S 24.266; K T U 1.105 = R S 24.249:19; K T U 1.104 = R S 24.248:13; K T U 1.109 = RS 24.253:11), a n d also in the mythological texts (especially K T U 1.4 = R S 2. [008]+; but see also K T U 1.17 = R S 2. [004] i 31). It is built on a platform which is approached by means of a monumental stairway; it comprises a vestibule which comes before the naos (or inner cella) and, in the southwest corner, is preceded by a structure which can be explained as the holy of holies. In front of the entrance, within the court surrounding the sanctuary to the south, there is a structure measuring 2 x 2 metres, usually explained as an altar (it is thought that it may be the altar

70

The first of these is the so-called sanctuary of the Hurrian gods close to the royal palace, the nature and cultic function of which seem to be certain; the second is the so-called sanctuary of the rhyta, located in the residential area of the city; cf. Y O N 1 9 9 6 . 71 On cult places in Ugarit cf. generally Y O N 1984; D E L O L M O L E T E 1992a, 2 4 - 8 = 1999, 27-34; DE T A R R A G O N 1996; DEL O L M O L E T E - S A N M A R T I N 1998, 184-6.

mentioned in K T U 1.41 = RS 1.003+:41 mdbh b'l). It is interesting to note that the raised structures mentioned as the 'tower of Baal of Ugarit' ( K T U 1.119 = RS 24.266:12) and the 'terrace' (gg in K T U 1.41 = RS 1.003+:50), where sacrifices were offered, are probably confirmed from archaeology by the enormous width of the foundation walls of the temple of Baal and in the purported existence of a stairway on the eastern side of the temple. Archaeology has brought to light a second great temple on the acropolis. Due to the finding of the two inscribed stelae K T U 6.13 = RS 6.021 and K T U 6.14 = R S 6.028 in its vicinity, it has been considered as dedicated to the god Dagan. However, unfortunately the Ugaritic ritual texts never mention a temple of Dagan, 7 2 so that it seems more reasonable to attribute this temple to the god El in agreement with the evidence from the ritual texts K T U 1.87 = RS 18.056:42 and K T U 1.119 = R S 24.266:13-4 and with the mythological texts K T U 1.17 = RS 2. [004]+ i 3 1 - 2 and K T U 1.5 = R S 2. [022]+ iv 21. This temple also, like the other one dedicated to Baal, is built on a platform, has massive foundations which suggest the existence of a very high upper storey and is subdivided into a vestibule and a cella. Besides these temples which have been identified archaeologically, the ritual texts also mention the temple of Ilat ( K T U 1.41 = R S 1.003+:24 mdbh bt ilt 'altar of the temple of Ilat') and the 'temple of the lady of the high t e m p l e s ' (ibid. 37 [b]t b'lt bt[m rmm]). Unfortunately, nothing worthwhile can be said about these two temples given that the context provides nothing useful regarding their description. From the texts in prose we know of the existence of a temple dedicated to El (cf. above on the hypothesis regarding Dagan), a temple to 'Attart ( K T U 4.216 = R S 16.165:2), a temple of Rašapgn (ibid, line 3) and other sacred structures dedicated to Iiis ( K T U 4.781 = R I H 8 3 / 2 8 + 31 + 8 4 / 1 5 + 26:2) and to the unknown deity dml (the letter K T U 2.26 = RS 16.264:6). Besides the temple structures, a whole series of names of other specific places of cult are known. T h e most important of these is the hmn, which has already been mentioned, i.e. the palace chapel dedicated chiefly to the dynastic cult (and comprising perhaps qdš, c ly and kbm, i.e. a small sacred area, steps and other unidentified 72 Cf. N I E H R ing in KTU1.

1994,

and on

K T U

1.104

=

RS

24.248:13

cf. also the new read-

structures). But there are also other places about which nothing certain is known, such as the 'garden' (gn) already mentioned, cited for example in K T U 1.106 = R S 24.250+:22~3, or else the gb ('sacrificial pit'?, 'cistern' ? or 'platform' ?, cf. K T U 1.43 = R S 1.005:1-2), the urbt, the cgml, the gb and others as well (such as the 'tower' mgdl, the 'staircase' (?) m'lt, etc.). In spite of the lacunae and the uncertainties, the combined use of textual, archaeological and comparative data makes the ritual system of Ugarit certainly the best known within west Semitic religions with the prospects of further knowledge in connection with excavations which continue uninterruptedly.

2

T h e

Offering

Lists

and

Gregorio

del

O l m o

2.1

the

G o d

Lists

Lete

Introduction

T h e 'list' as a literary form is well known in the scribal practice of the ancient N e a r East 1 and is well represented in its archives. T h e s e provide us with canonical series which comprise a sort of universal encyclopaedic vocabulary arranged by theme, together with other lists of similar origin. These lists, which come from administrative archives, record different products a n d items that ' c o m e into' or 'leave' the public warehouses, evidently for the purposes of accounting. In principle, their fate is irrelevant as far as accounting goes, which m e a n s that there is no need for a special category for recording products intended to be consumed in the cult as sacrificial victims. In fact, in the archives of Ugarit, such account records occur mixed together with others which have all kinds of destinations ( d e l O l m o L e t e - S a n m a r t i n 1998). However, we are able to identify t h e m and given their syntactic structure, consider them to be cultic texts. This structure is set out as the attribution of an offering to a divine recipient following various grammatical models ( d e l O i . m o L e t e 1992a, 1 3 20 = 1999, 11—21 ); such an attribution implies a cultic action even though its model does not need to be m a d e explicit. As a result, in their simplest form these texts a p p e a r as a series of two juxtaposed lists, one of offerings a n d the other of gods. T h e y can be analysed separately or in relation to each other, in both cases providing some information on the development of cultic practice in this community. It is clear that in such a case the 'list of gods' determines the origin and a r r a n g e m e n t of the ritual, either implying the existence of cultic p a n theons or using other p a n t h e o n s which originated outside the cult. In terms of administration, t h o u g h , the i m p o r t a n t element is the record of offerings. However, this concise form of cultic list is not the only one; there are others which supply information a b o u t the m o m e n t , place or

1

OPPENHEIM

1 9 7 7 , 244FF.; L A M B E R T 1 9 5 7 - 7 1 , 4 7 3 - 9 ; C A V I G N E A U X 1 9 8 0

3,

609-41.

type of cultic act, even providing descriptions of it, which suggests that at least some of these texts are rather more than mere account 'records' ( d e T a r r a g o n 1995a, 104). Even so, in principle the references to the recipient or sacred m o m e n t of offering, even to the rite, seem to have the purpose of merely certifying or justifying delivery, issue or expense in the accounts, just like their civil counterparts concerning the moment or method of a business transaction. Thus, they do not seem to be accounts which strictly tabulate items but rather they all seem to be tinged with contextual references. This is possibly connected with the formation and function of the scribe, Quite often the development of these facts takes on an informative value which transcends mere accounts and becomes a more or less schematic description, in space and time, of a complex cultic action. Given the importance of the cult in ancient cultures, it is legitimate therefore to suppose that these texts were produced as meta-economic texts, with their own sacral meaning, and have been analysed as such. 2 Keeping this in mind, it is often quite difficult to determine when the text is merely a list for accounts (list) and when it is descriptiveprescriptive (ritual). It all depends on what is implied by the information as a whole. Here we shall discuss basically texts which seem to be pure lists or largely lists, i.e. in which the other information can be considered (merely) circumstantial. Texts which lie outside this frame of reference are left to be analysed as a whole as rituals, i.e. examples of the liturgy of Ugarit, and from them we shall extract here only the series of offerings and divine names they provide. O n the other hand, these elements (offering and recipient) which are supplied together in the lists mentioned, could be recorded quite separately. This happens especially in the pantheons or n a m e lists of gods, whereas the possible lists of products or offerings can easily be camouflaged a m o n g texts which we consider to be strictly administrative if there is no indication of their use or purpose (del Olmo

2

Lete

-

Sanmartin

1998,

192-4).

This has been the approach adopted by us (DEL and earlier by DE T A R R A G O N 1980 and X E L L A 1981.

OLMO

LETE

1992a = 1999),

2.2

God Lists

We shall begin with the lists of gods, given the importance this literary form acquired in ancient near Eastern religion: from the beginning, the principal entities to be listed were the gods. These lists could have functioned merely as templates with which to fill the records of delivery or lists of offerings, but they also undoubtedly had a meaning transcending their practical use. This is suggested by the n u m b e r of copies that were made and their translation into Akkadian, as we shall see below in connection with List A of the gods. This is a process of systematization which combines profession of faith in the divine person with the invocation of his name, elements basic to all ancient Near Eastern religions. Apart from the 'List Αηιΐ, which is arranged in the Mesopotamian style and so is foreign, there were at least two indigenous lists, originating in a different period and a different ideology, both translated into Akkadian and occurring together in the cult in the lengthy text K T U 1.148 = RS 24.643 ( P a r d e e 1997b, 67-71). T h e palace shared in this religious process of systematic theology in respect of the divine universe providing its own 'pantheon', also for cultic use. Besides these and other name-lists or exempt lists we shall provide next those to be found in ritual contexts, sacrificial and non-sacrificial. 2.2.1 2.2.1.1

Lists of divine names T h e List Ann

T h e Ugaritic fragments of this classic list of Mesopotamian gods copy their model faithfully and are a witness to how venerable and widespread they were throughout the ancient N e a r East, no doubt together with the school tradition of the literature of lists. However, as such, they provide nothing new or important about the religion of Ugarit. For that topic we refer to the bibliography. 3 More important as part of the same scholarly tradition is the information provided by the polyglot vocabularies since they allow us to trace the equivalences made in Ugarit between the Sumerian-Akkadian and

3

On the Mesopotamian list Anu in general cf. the fundamental article by W E I D N E R 1929 and the bibliography provided later by B O R G E R 1975, I I I 64-5; also L I T R E 1958. For this list in Ugarit cf. N O U G A Y R O L 1968a, 2 1 0 - 3 0 , 246 9, 324. Another new fragment has been published by A R N A U D 1982a.

Hurrian pantheons and their own; however little of the text has been preserved ( N o u g a y r o l 1968a, 246-9). 2.2.1.2 List A ( K T U 1.118 = RS 24.268+, K T U 1.47 = RS 1.017, K T U 1.148 = RS 24.643:1-9, R S 20.24) T h e basic text of the principal or canonical list (A) is K T U 1.118 = R S 24.268+, whereas K T U 1.47 = RS 1.017 which has the addition of il spn at its beginning, is very damaged, and can be reconstructed from the other text. O n the other hand, the good condition of the Akkadian text RS 20.24, a version of the previous text, is of inestimable value for determining what the Ugaritic names/epithets mean and their relation to the Mesopotamian pantheon ( d e l O l m o L e t e 1986a, 293-9; H e a l e y 1985, 115-23; 1988a, 103-12). In turn, K T U 1.148 = R S 24.643:1-9 confirms the fixed character of this canonical sequence in the cultic domain: its first nine lines give us a simple listing of the gods of List A, followed by the victim allotted them in the festival of Sapunu (cf. below). This first section is repeated in lines 10-12 as a sort of summary, in accordance with the two sets of offerings required by the šrp wšlmm ritual. RS

20.24

DINGIR a-bi Hum*"m ci da-gan A adad be-el huršān ha-zi A adad II A adad III A adad IV d adad V A adad V I A adad V I I D

IDIM

A

Ù IDIM

sa-sú-ra-tum A sin A l}uršan ha-zi A é-a %é-bat A aš-ta-bi d huršanumrì u a-mu-tu[m] aš-ra-tum A a-na-tum A šamaš

K T U

1.118

ilib il dgn b'l spn b'lm b'lm b'lm b'lm b'lm b'lm ars wšmm kt[r]t \y)rh [s]pn kir pdry 'ttr grm w\'mqt] WW 'nt špš

1.47

1.148:1-9

il spn [il spn] ilib [ilib] i[l] il dgn [dgn] b'l spn b'l spn] b'lm b'lm b'lm [b'lm] b'lm b'lm b'lm [b]'l[m] [b]'lm [b'lm] [b'l\m [ars] wšmm ars wšmm [ktr]t klr[t] yrb Μ] spn [spn] klr m pdry [pdry] m [grm w'mqt] grm w'mqt atrt [alrt] 'nt M špš [špš]

d

al-la-tum iš-ha-ra d is'tar"lm HlāmT* til-la-ad Aadad d nergal d dá-ad-mi-iš d pu-lfur ilānimes d tāmtum d

dDU(

dgis d

'BUR.ZI.NÍG.NA

ki-na-rum

MA.LIK.MEŠ

d

sa-li-mu

arsy ušhry 'ttrt il t'dr b'l r[Š]p ddmš phr ilm ym utht km mlkm šlm

[a]rsy [u]šhry ['] ttrt il t'dr b'l ršp ddmš phr ilm ym knr mlkm Ilm

arsy 'ttrt ušhry il 't'dr b'l ršp ddmš phr ilm ym utht (k]nr

Leaving out the addition of the reinterpretative title (il spn), the structure of this list provides the following elements: a) It begins with a 'triad' which to some extent defines the personality of the s u p r e m e god u n d e r three cultically distinct epithets: ilib, il, dgn (see also K T U 1.123 = R S 24.271:1-3: ab wilm. . . il. . . il šr dgn . . .). Very probably they suppose a development (il as 'father' of the divine family) of his personality a n d an attempt at combining p a n t h e o n s , as a reflection of myth, portraying Baelu as 'son of Dagānu' a n d considering Ilu to be his 'father' (DEL O L M O L E T E 1992a, 56 n. 77 [1999, 74 n. 78] with bibliography). b) A similar process is assumed in the case of Ba'lu a n d (H)ad(a)du, 4 H e is the great C a n a a n i t e god of the second millennium, the protector of Ugarit (b'l ugrt), defined in principle by the attribute of his residence spn, specified by the Akkadian version as '(H)Adad, lord of M o u n t H a z ï . In his sevenfold epithet are concentrated all the m a n y (local) e p i p h a n i e s (b'l ugrt, hlb . . .)} In K T U 1.118 = R S 24.264+, a line separates this group f r o m the following heading. c) T h e descriptive series invoking Ba'lu is followed by a g r o u p of seven deities, h e a d e d by the dual divinity 'Earth a n d H e a v e n ' , a p p a r endy 'chthonian-astral' or cosmic in nature, in chiastic relation (stellar/

4 Akkadian makes a clear distinction between the proper noun Adad (IM) and the common noun be-el, 'lord', which are combined in Ug. b'l (cf. line 4: ''adad be-el t}uršan fra-zi = b'l spn). The Ug. epithet hd of myth is not used here nor is add of ritual (cf. K T U 1.65 = RS 4.474:9), in spite of the Akkadian translation. But it is possible that d IM was not read as AAdad in Ugarit; the Akkadian version is a Ugaritic interpretation (interpretatio) for 'foreigners' (DIETRICH - LORETZ 1981, 67-8). 5 In this context K T U 4.15 = RS 9.469 must be considered, which is possibly a list of local or family epithets (stelae?) of Ba'lu.—On divine 'heptads' in Mesopotamia cf. the bibliography given by BORGER, above, n. 2.

terrestrial deities) with this polar pair which heads the group (ktrt,6 yrh, 'ttr || spn, ktr, pdry). d) T h e next group, again of seven deities, is also headed by the dual name/divinity 'Mountains and Valleys'. It is combined with six goddesses (1 + 6 ) , apparently arranged hierarchically and related to the male gods of the previous groups. This separation of the sexes confirms the absence of 'pairing' in the Ugaritic pantheon. e) T h e last group is different in each of the various texts. T h e most complete list is provided by K T U 1.118 = R S 24.264+ and the Akkadian version, whereas the others omit one or other element. Here too the series is introduced by a composite divine name, 'the helper gods of Ba'lu'. T o it corresponds another composite epithet, 'the assembly of the gods', equivalent to 'the family of Ilu'J At all events, this last group is somewhat of an appendix and possibly contains later additions. In it is developed the process of 'divinizing' objects (utf}t, km) and persons (mlkm).8 T h e final result is a pantheon of 33 divine invocations, excluding the title added in K T U 1.118 = R S 24.264:1 (and K T U 1.148 = R S 24.643:1). T h e hierarchical principle governing the composition of this 'god list' is not clear; they do not seem to be arranged according to 'personal' importance. Even so, it is possible to say that the 'canonical list' (A) does in fact include the group of principal gods of Ugarit, exacdy as they appear in myth and the official cult ( K T U 1.148 = RS 24.643:1-9]. It represents a mythologizing expansion which tends to make organic distinctions between the gods and at the same time to assimilate other foreign pantheons within its own religious framework, from the multiple cultural influence which affected Ugarit: Amorite, Hurrian, Hittite, Sumerian and Akkadian.

6 This is an overall term for a group of 'seven' deities, as we know from K T U 1.24 = RS 5.194:47-50 (DEL O I . M O LETE 1991, 74-5). 7 Cf. K T U 1.123 = RS 24.271:32 which refers to drm ilm, 'the two divine families' (?); K T U 1.40 = RS 1.002:25 and par.; and K T U 1.65 = RS 4.474:2-3, with its invocation of dr/mpl}rt bn it. 8 In the cult a small reduction of the standard list as it has come down to us in its final form, is evident: from a pantheon of 33/34 epithets it has become a group of 28/29, a number which is 'cultically' determined in K T U 1.148 = RS 24.643:1-9, in all likelihood, by the clays of the month, as will be apparent in the fifth section (lines 23-45). One b'lm as well as 'ttr, uhtt, mlkm and šlm are omitted and the sequence of ušf)iy and 'ttrt is inverted.

2.2.1.3

List Β ( K T U 1.148 = R S 24.643:23-45; R S 26.142)

K T U 1.148 = R S 24.643:23-45, instead, provides us with a new list of the il hyr, 'gods of the (month) Hiyaru. W e have no separate Ugaritic version of this text, but it is evidently a fixed and canonical list as is shown by the coincidence of the cultic use with the independent Akkadian version (RS 26.142). As both texts are d a m aged, we lack a complete list of gods. 9 W e now set out both texts arranged in order (the Akkadian text follows the recent reconstruction by A R N A U D 1994, 107-9) a n d their Ugaritic equivalents: RS 26.142 1. dingir [a-bi\ d Ki [à clidim] dingir-[/M4] d Nin.mah x? [ d 5. Da-gan d U Hal-bi d U d hur.sag Ha-zi d 'Iškur' tur d [XX]X 10. [ d hur.]sag Ha-zi [ d É]-a ΓáAš-ta-bv. dBe-el]'šul-ba)-ad-da

K T U 1.148 23. ilib ars wšmm 25. il a/ktrt dgn b'l hlb b'l spn trty yrh spn 30. ktr 'ttr atrt šgr witm ršp idrp ]gšr gl]mt 35

15. 'dingir.meš giš.sag.kul' dingir.meš da-ad-me-na

9

km ] mšr 40. il m[ [...] w thmt [...] xmr [z7] sk[r] \il.d\dm

Interpretation ilib Earth and Heavens Ilu Nin.mah [ ] Dagānu Ba'lu of Halba. Ba'lu of Mount Hazi. Ba'lu the second (?) [£]» (?). [Moun]t Hazi. Ea 'Attaru Atiratu Sagru and Itmu ρ Rašpu of idrp ? The Damsel ? ρ ρ Kināru Mišāru The god of [...] and Tihāmātu Gods of the bolt Gods of Dadmena

As yet unpublished texts may perhaps help complete and identify them (BORP A R D E E 1995, 31). P A R D E E (1992, 167) suggests that K T U 1.148 = RS 24.643 is an incomplete tablet which does not reproduce the full god list RS 26.142.

DREUIL -

DINGIR.MEŠ la-ab-a-na dug bur.zi.ni.naga.t! d giš"ZA.MÍM 20. d U d U d U d d

d

u

Ma-[lik. mes] 25. d D[I

il.lb[n]n

[...] b'lm 45. [....]

Gods of Labana Incense burner (?) Lyre Ba'lu Ba'lu Ba'lu Ba'lu Mali[kuma] ?

In our view this is yet another pantheon used in the cult. As in the case of K T U 1.148 = RS 24.643:1-9, it is very probable that this is also a modification of an earlier pantheon, reduced to 28 deities, in line with the monthly nature of the ritual. 10 Everything indicates that K T U 1.148 = RS 24.643:23ff. is to be considered a ritual connected with the 'funerary' cult of the month of Hiyaru, which is older than the 'festival of Sapunu' (lines 1-9). 2.2.1.4 List C ( K T U 1.102 = R S 24.246; K T U 1.139 = R S 1.001:13-9) W e have to consider the list of gods provided by K T U 1.102 = R S 24.246:1-14 along the same lines (exempt list and cultic use). It is a list (C) used in the cult in K T U 1.139 = R S 1.001:13-9, i.e. this is a standard list not a casual one. Thus, the group of 14/16 deities mentioned there presupposes a selection which in n a m e and number largely agrees with the one provided by the group of texts which we call 'dynastic' (cf. below). 1. il bt ušhry ym.b'l yrf} 5. ktr pdry dqt trt 10. r'sp 'nt f}bly 10

T h e 'month' would only be a pattern, without necessarily implying, on the other hand, a daily celebration in honour of each one of the gods. Also, comparison with RS 26.142:16 suggests correcting ktrt to atrt (nin.mah) in K T U 1.148 = RS 24.643:25, so written probably because of the sequence šmm w ars ktrt in list A (but cf. atrt in line 31; Pardee reads [-]-rt).

špš pgr iltm hnqlm yrh kty

This 'list of gods' contains a group of deities exclusive to the quoted texts (trmn, dqt, trt and the epithets cnt hbly, špš pgr, iltm hnqtm; yrh(m) kty also occur in K T U 1.123 = R S 24.271:7), and also includes the major deities of Ugarit [b'l, ym, yrh, ktr, cnt, špš). 2.2.2 2.2.2.1

Lists of names of divine kings List D ( K T U 1.102 = RS 24.246:15-28)

T h e dynastic nature mentioned above explains why the list of K T U 1.102 = RS 24.246:1-14 is continued on the reverse of the tablet by another (D) of theophoric names, to which in K T U 1.39 = RS 1.001:19" correspond the references to glmt, bclt bhtm, ins ilm. This suggests that these theophoric names are also divine names, actually of the kings of Ugarit, of their ins ilm, 'divine peoples', of which K T U 1.106 = R S 24.250:3-5 already provides proof (three: ydbil, yaršil, cmtr) as recipients of offerings and the colophon K T U 1.6 = R S 2.[009] vi 58 assigns one (yrgb.bcl) to king Niqmaddu of Ugarit. 12 This interpretation, however, has incurred some opposition, although alternative suggestions are not convincing. 13 Consequently, we consider K T U 1.102 = RS 24.246 to be one list or single pantheon which belongs to the dynasty with its gods and divinized dead, which as such (ins ilm, gtrm, rpum) are only listed globally in the c o m m o n standard patheon under the official title of mlkm. These divine names are:

11 Understood in this way, K T U 1.39 = RS 1.001 is a 'generic' royal ritual which closes by invoking the ins ilm, specified in K T U 1.102 = RS 24.246:15-28, and as such could be added to it together with the sacrificial element thus supposed, e.g. in K T U 1.106 = RS 24.250+:2-5 (gdlt). V1 On K T U 1.102 = R S 24.246 in general cf. V I R O L L E A U D 1968, 594; DE M O O R 1970b, 326-7 (see 1990, 241); H E R D N E R 1978, 3-7; D I E T R I C H - L O R E T Z SANMARTIN 1975b, 545-6; C A Q U O T 1979b, 1404; S T A M M 1979, 753-8; X E L L A 1981, 328-31; D E L O L M O L E T E 1986a, 282-5; 1987:43-6; 1992a, 117-20 = 1999, 168-75. 13 Cf. P A R D E E 1996b, 273-87; S C H M I D T 1994, 71; 1996, 300; the reply by DEL O I . M O L E T E 1996b, and the counter-reply by P A R D E E 1998b.

15. y(r)gbhd yrgbb'l ydbil yaršil yrgmil 20. 'mir ydbil yrgblim 'mtr yaršil 25. ydbb'l yrgmb'l 'zb'l ydbhd 2.2.2.2

List Ε ( K T U 1.113 = R S 24.257)

A supplementary text, possibly with a similar pattern to the one cited, is K T U 1.113 = R S 24.257 on the reverse of which (lines 13-26) we have the list (Ε) of the proper names of the (dead) kings of the dynasty of Ugarit. 14 It reflects the same royal ideology as supposed in the previous text, providing us with the list of 'personal' names of the kings of Ugarit preceded by the determinative il. It is, therefore, a classification of them as 'divine beings' or divinized beings, 15 as already dead of course, and reigning in the 'Underworld', for we have no indications that the kings of Ugarit were deified in their lifetime, in spite of the possible proleptic use of these names while a king was still alive. T h e reverse seems to be a hymn or evocation, whereas the obverse is a text divided into two columns, as is evident from the traces of a double vertical line before the last lines on the right. T h e text has been studied in detail by Kitchen and interpreted as a 'rising' or

14

Cf. V I R O L L E A U D 1962, 94~5; 1968. 561-2; K I T C H E N 1977, 131-42; X E L L A 1981, 288-91; D E L O L M O L E T E 1987a, 47-9, 68; 1992a, 121-3 = 1999, 176-80; D I E T R I C H - L O R E T Z 1983a, 496f.; P A R D E E 1988b, 165-78; L E W I S 1989, 47-52; Y O U N G E R 1997, 356-7; W Y A T T 1998C, 399-403. 15 An interpretation opposed to the divinization of kings, following an earlier proposal by Liverani, is defended by S C H M I D T 1994, 67fT.; 1996, 289-304; however cf. P A R D E E 1996, 276. L E W I S 1989, 49ff., instead, suggests a weakened 'divinization' of the kings of Ugarit which in our opinion does not do justice to all the ritual texts which give abundant evidence of the cult of dead kings and their entry into the 'pantheon', though this does not necessarily mean they were worshipped as the 'great' gods were; nor were the other 'minor' gods of the pantheon worshipped in that way either.

'retrograde' list of the kings of the Ugaritic dynasty, starting with the predecessor of 'Ammurāpī, the last king of Ugarit, under whose c o m m a n d this list of divinized ancestors was drawn up. Inevitably, the various possible reconstructions of the text leave a wide margin for hypotheses and there is no use trying to insist on any particular one. T h e text is as follows: 12. [ [ [ 15. [ [

[ [il [il 20. [il [il [il [il [

25. [ [

] ] ]

'mttm]r nqm\ d arhi]b[ nq]mpC[ 'mt]tmr nq]md ] ]

¥

]xx[ ] il 'm]ttmr il n]qmp' il Cmpi il ibm il/drd il tiqmp' il ibm i]l 'mrpi il] nqmpC il ibm il nqmp' il ibm il nqmd il yqr

Elsewhere we have suggested reading this list in descending order. T h e left-hand column, which ended with the historical sequence Ar-H alba/Niqmepac/cAmmittamru/Niqmaddu (lines 2 4 - 2 6 are written on the edge) continued with the last three names on the right which corresponded to the last three kings of Ugarit Ibiranu/Niqmaddu/ Taqaru, with the last name applied to the ruling king, CAmmurāpi, bearer of the dynastic seal inscribed with that name. 1 6 Each king is the 'new' Yaqaru,]1 in this case the last sovereign of Ugarit. Even so, we do not know the structure of the names in the left-hand column and their possible relationship to those on the right. 16

Cf. DEL O L M O L E T E 1992a, 123 = 1999, 179. See also the new combined reading of both columns and their comparison with the names of K T U 1.102 = RS 24.246:15-28, proposed by DE M O O R 1990, 2 4 0 - 1 . For a general discussion of these matters cf. A B O U D 1994, 3-11. 17 Cf. N O U G A Y R O L 1955, X L X L I I I ; L I V E R A N I 1962, 137; C A Q U O T 1978a, 574f.; X E L I A 1983, 404; R I B I C H I N I - X E L L A 1979, 155-6.; D E L O L M O L E T E 1992a, 123 = 1999, 179. This in turn would explain why the determinative il is placed before yqr and not before the personal name of the ruling king, which would imply some 'divinization' in his lifetime. T h e king would undoubtedly be supposed as having divine 'character' in his lifetime, at least for reasons of protocol and in anticipation as is apparent from K T U 1.16 = RS 3.325+ i 9 1 1 , 20-3, and from his epithets adn (ilm rbm), b'l (?) and trmn (cf. K T U 1.6 = RS 2. [009] vi 58).

2.2.3

Lists which are litanies of gods' names

A couple of texts, which seem to be prayers in the form of litanies, each provide lists of divine names which are completely strange in respect of their formal characteristics and the sequence of epithets, which also seem unrelated to each other. Whereas the first list keeps to the list of known deities, the other list has a series of divine names otherwise unknown in Ugaritic literature. Perhaps this is an expression of social religion which lies outside the 'official' systematic arrangement. K T U

1.65

il bn il dr bn il mphrt bn il tkmn wšnm il watrt il hš il add b'l spn b'l ugrt il ( 7 )

1.40

ab bn il dr bn il mphrt bn il tkmn wšnm

1.123

il wilm il il šr dgn wb'l tt wkmt yrh wksa yrhm kty tkmn wšnm ktr whss thr wbd 'ttrt 'ttpr šhr wšlm ngh wšrr 'd wšr sdq mšr hn bn il dn kbd wnr

ilqds mlk kbd dil gd[ mr mnmm bmn aryn a^hn tlyn atdb wtr qdš wamn ktr hss il bt il hst ršp ins ilm drm ilm

These texts which are prayers in the form of litanies have a set of characteristics in c o m m o n which are extremely interesting: a) First of all, they all have a clear preference for double invocations, either of the 'construct' type (bel spn) or of the 'copulative' type (il watrt). This predominant use of a particular linguistic construction possibly has cultic reasons and is perhaps connected with the correct use of the ' n a m e ' of the god and his proper match as an expression of his particular personality and therefore of his invocation. This match follows criteria which are not 'matrimonial', apart from a few exceptional cases (il watrt). Criteria of ethnic and functional origin are more prevalent (bcl-dgn, 'nt-'ttrt, yrh-ršp . . .). (DE M O O R 1970a, 227).18

18

On the possible merging of pantheons implicit in such double names cf. DEL L E T E 1992a, 40 = 1999, 52.

OLMO

b) In this type of cultic text there is also a marked persistence of fixed sequences. A text such as K T U 1.65 = R S 4.474, possibly also a 'litany', begins with the same sequence which forms the concluding invocation, repeated 5 or 6 times, of the long expiation text K T U 1.40 = RS 1.002+. It comprises, undoubtedly, a cultic sequence of prayer or general invocation of the supreme god and of all 'his family', the pantheon, in a nutshell, which ends in the first text by the invocation of il watrt, as a suitable literary and genealogical inclusio. More surprisingly, instead, at first is the mention in this series, as a single specific deity (or as separate deities), of tkmn wšnm, about whom we have no exact information, 1 9 although, as is clear from this series, they are connected specifically to the supreme god (cf. K T U 1.114 = R S 24.258:18-9.) and to the pantheon in general. Accordingly, the hypothesis can be put forward that they are 7/m's 'messengers' or 'dioscuroi', in parallel with qdš amn (cf. the reference to both pairs in K T U 1.123 = R S 24.271:26), the messengers of Atirat according to mythology. 2.2.4 2.2.4.1

Cultic lists of gods Lists of gods in sacrificial texts

Most of the cult texts of Ugarit are provided with rubrics which prescribe or describe the sacrificial ritual to accompany the various festivals or days of the month. T h e literary plan (cf. d e T a r r a g o n 1980, 55ff. who analyses the cult vocabulary; for the syntax of the texts and their formulae cf. d e l O l m o L e t e 1992a, 19ff. = 1999, l l f f . ) is extremely simple and it invariably specifies, as said above, the victim of the offering or sacrifice and the divinity to whom it is offered, together with directions for the ritual to be performed on it, but without spelling out the details. These texts, therefore, comprise the best evidence of the cultic and official pantheon of Ugarit and of the catalogue of gods of liturgical practice in its most generic and popular form: sacrifice. T h e other texts belong to rarer and more specific events (divination, atonement, royal cult). T h e texts considered here are as follows, omitting a long series of fragments which are inconclusive in this respect and add nothing new: K T U 1.39 =

19 They occur in other ritual texts, but are unknown in mythology and in the Canaanite pantheon; cf. P A R D E E 1988b, 195 9.

R S 1.001, K T U 1.41 = R S 1.003+, K T U 1.43 = R S 1.005, K T U 1.46 = R S 1.009, K T U 1.48 = R S 1.019, K T U 1.49 = R S 1.022, K T U 1.50 = RS 1.023, K T U 1.81 = R S 15.130, K T U 1.87 = R S 18.056, K T U 1.105 = R S 24.249, K T U 1.106 = R S 24.250+, K T U 1.109 = R S 24.253, K T U 1.112 = R S 24.256, K T U 1.115 = R S 24.260, K T U 1.119 = R S 24.266, K T U 1.126 = R S 24.276, K T U 1.130 = R S 24.284, K T U 1.139 = R S 24.300, K T U 1.148 = R S 24.643, K T U 1.162 = R S [Varia 20], K T U 1.163 = R I H 7 8 / 1 4 , K T U 1.164 = R I H 7 7 / 0 2 B + , K T U 1.165 = R I H 7 7 / 0 4 + , K T U 1.168 = R I H 7 7 / 1 0 B + , K T U 1.170 = R I H 7 8 / 1 1 , K T U 1.171 = R I H 7 8 / 1 6 , K T U 1.173 = R I H 7 8 / 0 4 (for other fragments cf. X e l l a 1981, 128-68; d e T a r r a g o n 1989, 222-3). Besides the separate lists reproduced above, these cultic texts of Ugarit have preserved for us stable and repeated sequences which presuppose the existence and persistence, at least in the cult, of fixed series of divine names, transmitted in these 'lists of offerings' (cf. below) a n d are thus indirectly 'lists of gods'. However, most, of KTU 1.41/1.87 KTU 1.39 il b'lt bhtm ins ilm t' ils t'm ilhm t'm2i ym ilh ilh ilhm ilhm ilhm ilhm b'l b'l atrt atrt ikmn wšnm tkmn wšnm 'nt 'nt ršp ršp dr il wphr b'l dr il wphr b'l šlm šlm ilhm b'lm ilhm b'lm spn b'l spn glmt ilt mgdl yrh ilt asrm nkl b'lt bhtm ins ilm ilhm špš

KTU 1.41/1.87 'nt il ilhm tkmn wšnm ršp ršp ilh ilhm il tkmn wšnm ilt bt b'l spn spn b'l ugrt ilib [atrt] ridn atrt ins ilm b'l spn b'l ugrt 'ttrt

KTU 1.39 il ilhm tkmn wsnm ršp

KTU 1.39: b'l spn spn b'l ugrt ilib

b'l ugrt 'nt spn

these sequences of recipients of offerings are not fixed a n d seem to be somewhat h a p h a z a r d , although in fact they were not a n d this impression is due to lack of documentation. 2 0 In any case, it is irelevant n o w to call it a 'list' of gods. O n l y in a few cases does the repetition of the sequence in different texts or in different sections within the same text prove it to be normative. In fact, several of t h e m must be split up in different series, as they represent different cultic actions or the repetition of the same action on different days. T h u s we have in K T U 1.41/1.87 = R S 1 . 0 0 3 + / 1 8 . 0 5 6 , K T U 1.39 = R S 1.001 a n d K T U 1.109 = R S 24.253 a significant a n d important list of divine n a m e s (or rather, a set of sub-sets) which we can consider standard in the cult. Apparently, this sequence is related to the ' c o m m u n i o n ' sacrifice (šlmm), which is expressly mentioned. Within this sequence can be noted several repeated sub-sections (e.g. il/h, ilhm, (il), tkmn wsnm || (il), bclt bhtm, ins ilm, (ils), ilhrrì). It is possible that each g r o u p or section of the list belongs to a specific ritual. T h e d a m a g e d condition of the texts makes proof difficult. Smaller sections also occur in other sacrificial texts such as: b'l spn, rmŠ, šlm, b'l spn, spn (cf. K T U 1.46 = R S 1.009:12-5; K T U 1.109 = R S 24.253:5-10; b'l ugrt, ilib, il, b'l, 'nt, spn, pdry (cf. K T U 1.46:16-7; K T U 1 . 1 0 9 : 1 1 - 4 ; cf. DE M O O R 1970a, 221); b'l ugrt, b'l hlb, yrh, 'nt spn, pdr(y) (cf. K T U 1.130 = R S 2 4 . 2 8 4 : 1 0 - 5 ; K T U 1.109:16-8). Yet other sections or even couples are very stereotyped a n d not very relevant, e.g. the beginning of K T U 1.46 or the end of K T U 1.130. Lasdy, some texts such as K T U 1.50 = R S 1.023, K T U 1.81 = R S 15.130 a n d K T U 1.119 = R S 24.266, have completely a n o m alous sequences both f r o m the canonical a n d the cultic points of view. O t h e r s again are too broken for a n y conclusions to be drawn, e.g. K T U 1.48 = R S 1.019, K T U 1.49 = R S 1.022, K T U 1.90 = R S 19.013, etc. As an example of the incongruity of the lists of gods in the ritual texts see some of the sequences provided by the subgroup of texts which, however, as their tenor is royal or of the palace, are to some extent pertinent ( K T U 1.43 = R S 1.005, K T U 1.46 = R S

20

Cf. P A R D E E 1997b, 68: 'les listes divines que l'on peut tirer des rites sacrificiels . . . sont, potentiellement au moins, innombrables', DE M O O R 1970a, 2 0 4 1 6 , provides a table arranging the gods in 38 sets; it is hardly relevant. 21 Other scholars consider the term to denote a generic type of 'offering'.

1.009, K T U 1.105 = R S 24.249, K T U 1.106 = R S 24.250+, K T U 1.109 = R S 24.253, K T U 1.112 = R S 24.256, K T U 1.115 = R S 24.260, K T U 1.162 = R S [Varia 20], K T U 1.165 = R I H 7 7 / 0 4 + , K T U 1.168 = R I H 7 7 / 1 0 B + , K T U 1.170 = R I H 7 8 / 1 1 , K T U 1.171 = R I H 7 8 / 1 6 , K T U 1.173 = R I H 78/04). KTU 1.112 b'lt bhtm ins ilm 'ttrt hr gtrm il b'l spn b'l ugrt atrt btbt il msd(?) btbt

1.106 1.105 b'lt bhtm rsp b'l 'rkm ins ilm btbt ydb il b'l yarš il 'mtr spn ršp bbt ršp ins ilm ins ilm ršp mlk pdry b'lt bhtm ilm ars ktr arsy

T h e r e is no fixed group seen above, ushr, bbt, hlmi, qlh ( K T U 1.43 = R S ( K T U 1.105 = RS R I H 78/16); ršp, 19.015:14-5).

1.43 ('ttrt) il(m) ktr (gtrrrì) špš yrh gtr 'nt špš yrh gtr ilt bt gtrm 'ntm 'ntm šlm(?) ušfyiy(?)

1.115 ušhr hlmz bbt il bt hlmz qlh il bt att il bt bbt ushr hlmi qlh

1.139 ilib b'l il t'tr b'l 'ttrt glmt pdry(?)

1.162 ilib z(!)/ b'l dgn yrh ym il t'dr b'l 'nt hbly a$.)rt22 dr il wphr b'l spn

order in these sequences of deities, but as in the within the same text, some sections are repeated: ( K T U 1.115 = R S 24.260); špš, yrh, gtr, 'nt/ilt bt 1.005); and some of two names only: ršp, ins ilm 24.249, K T U 1.106 = R S 24.250+, K T U 1.171 = b'lt bhtm(?) ( K T U 1.105; cf. K T U 1.91 = R S

T h e only conclusion in this case is the strong dynastic or palatine nature of these series of deities (DEL O L M O L E T E 1992a, 4 4 - 5 ; 1999, 58-9.). Here also there is the shortened group of the great gods of the official pantheon, not always in dominant position. Arranged hierarchically, we can set them out followed by the index of their frequency: ilib (4), il (5), dgn (2), b'l (5), b'l spn (3), b'l ugrt (1), atrt (3), spn (3), ktr (2), s'ps (2), yrh (3), 'nt (4), 'ttrt (1?), pdiy (2).

22

Cf. DEL O L M O L E T E 1 9 9 7 , 164ff. Note that this list agrees both with List A (beginning: ilib, il, b'l, dgn) and the dynastic List C (where dgn, ym, 'nt f}bly are specifically mentioned).

From all this it can be deduced, as was already apparent in the separate lists, that within standard orthodoxy the palace, the royal family, has a special cult of particular or selected deities. Some are dynastic in type and almost all of them are foreign to the original Semitic pantheon of the city; these dynastic gods correspond, perhaps, to what the letters call 'gods of the king/queen'. Others are 'tutelary' deities of the palace and as generic epithets can occur in other rituals as protectors of the 'house' of each of the faithful. Lastly, others still are 'ancestral' and correspond to the cult ancestors. T u r n i n g to the general group of sacrificial texts, their series of divine names have the following features: a) T h e r e is consistency in the heading to the texts, in the main occupied by Ilu, in one or other epithet (il, ilib), followed by Ba'lu in one of his epithets (b'l, b'l spn, spn). b) Starting with the initial position, in any case, the sequence does not always follow the hierarchy expected from mythology (most often it does in the three or four first positions). T h e order is now determined by different criteria from importance and function in the pantheon; perhaps the criteria are patronage and cult specialization. c) T h e absence of a vertical order is also corroborated by the lack of a horizontal one: in these lists there are no sequences of divine 'pairs', a lack already noted in myth. d) Some local epithets retain their meaning in these texts (b'l ugrt, b'l hlb, 'nt hlb, 'nt hlš[?]), as well as b'l spn, 'nt spn from mythology. e) Some deities, who frequently feature in the dynastic and divinatory pantheons in special forms, also appear here (gtr[rri\, 'ttr, 'ttpr, bbt[m], b'lt bhtm, ins ilm), clear testimony to the introduction of the palace cult into the town cult. f ) T h e absence of Môtu as a recipient of sacrifices may be noted, although he is known as a harmful deity against whom one has to take precautions (cf. K T U 1.127 = RS 24.277:29). Statistically, in terms of their occurrence in various literary genres (DE M O O R 1970a, 217; above, n. 20) the principal gods of the culticsacrificial pantheon of Ugarit, those receiving the most offerings are: 1. 2. 3. 4.

ill dgn b'l/hdd 'nt ršp

5. yrh. 6. ktr 1. 'ttr 8. tkmn wsnm

9. 10. 11. 12.

špš šlm ins ilm b'lt bhtm

13. 'ttrt 14. ušhry 15. alrt 16. pdry 17. arsy

2.2.4.2

Lists of gods in non-sacrificial texts

Another series of texts of ritual practice refers to non-sacrificial cultic actions, in which the worshipper addresses a divinity with the word or goes in search of it. These actions may be accompanied by sacrifices, although they remain marginal and these texts do not have the form of stricdy sacrificial offerings to a deity. As they comprise a uniform category, to a certain extent, due to the language or life-setting which they share, it is worth considering the divine recipient of such rituals, although we shall not analyse them in detail. These texts are: K T U 1.40 = R S 1.002+, K T U 1.65 = R S 4.474, K T U 1.78 = R S 12.061, K T U 1.82 = R S 15.134, K T U 1.90 = R S 19.013, K T U 1.96 = R S 22.225, K T U 1.100 = R S 24.244, K T U 1.103+ = R S 24.247+, K T U 1.104 = R S 24.248, K T U 1.107 = RS 24.251+, K T U 1.123 = R S 24.271, K T U 1.124 = R S 24.272, K T U 1.127 = R S 24.277, K T U 1.140 = R S 24.302, K T U 1.141 = R S 24.312, K T U 1.142 = R S 24.323, K T U 1.143 = R S 24.326, K T U 1.144 = R S 24.327, K T U 1.155 = R S 24.654, K T U 1.169 = R I H 78/20. Others in this category (divinatory and oracular) do not mention gods, an extremely important fact, which emphasizes the specific nature of magical liturgy in contrast to sacrificial liturgy. T h e result is as follows, divided into two categories: b) incantation-magic

a) divination-oracle KTU

Ipl ršp

1.78

KTU

1.124

adn ilm(?) dtn dtn

KTU

1.100

(.pi) (phlt) (šps') il b'l dgn 'nt w'ttrt yrh rip 'ttrt ZZ wkmt mlk ktr whss Ihr wslm hm

KTU

1.107

hm( 3) IpI(3)

il whm b'l wdgn 'nt w'ttrt yrh wrsp 'ttr 'ttpr ZZ wkmt mlk b'ttrt ktr whss Ihr wslm

In the divination and oracle texts, as noted above, the scant reference to deities and then basically only in descriptive terms (cf. K T U

1.124 = RS 24.272:2), as granting omens, not as recipients of offerings, is evident, a d a t u m which highlights the importance of 'magical liturgy' in contrast to sacrifice. O n the other hand, texts of magic and incantation, like the litanies discussed above, have a long series of double deities in almost the same sequence. T h e standard double series is given in K T U 1.107 = R S 24.251+. This text and K T U 1.100 = RS 24.244 also invoke hrn and špš in a special way, as the principal deities in incantation. Again, the repetitive sequence (partly reflected in K T U 1.123 = R S 24.271; cf. above) exhibits a fixed cultic structure of curse-prayer with possibly a magical and thus unalterable meaning. Several of these couples are already known in mythological literature (šhr-šlm, ktr-f}ss and to some extent, cnt-cttrt, bcl-dgn); others occur determined by cultic-magical use (il-hrn, yrh-rsp); and others still need to be studied more closely to explain the meaning of the relationship ('ttrt-'ttpr, ZZ-kmt; for mlk-bettrt cf. K T U 1.108 = RS 24.252:1-2). Otherwise, apart from these last deities or invocations/epithets, the remainder belong to the mythological pantheon. As a whole, then, the pantheon of these texts is evidently very fixed and sequential, with a preference for 'double' and 'paired' names, as expressed in K T U 1.100 = R S 24.244 and K T U 1.107 = R S 24.251+.

Appendix:

T h e

Lists

of

Hurrian

Gods

T h e influence of Hurrian religious, mythological and cultic tradition in Ugarit is strong and is attested by the many texts written in that language found in various archives ( K T U 1.26 = RS 1-11.[048], K T U 1.30 = R S 1-11.[046]·, K T U 1.32 = R S 1.[066], K T U 1.33 = R S 1.[067], K T U 1.34 = R S 1.[076], K T U 1.35 = R S 1.[069], K T U 1.36 = RS 1 .[070], K T U 1.42 = R S 1.004, K T U 1.44 = RS 1.007, K T U 1.51 = RS 1.027, K T U 1.52 = R S 1.028+, K T U 1.54 = RS 1.034+, K T U 1.59 = RS 1. [049 a], K T U 1.60 = RS 2.[006], K T U 1.64 = R S 3.372, K T U 1.66 = RS 4.474, K T U 1.68 = RS 5.200, K T U 1.110 = RS 24.254, K T U 1.116 = RS 24.261, K T U 1.120 = R S 24.269+, K T U 1.125 = R S 24.274, K T U 1.128 = RS 24.278, K T U 1.131 = R S 24.285, K T U 1.135 = RS 24.295, K T U 1.149 = RS 24.644) as well as some bilinguals, for

example K T U 1.111 = R S 24.255 and K T U 1.132 = RS 24.291, in which the Semitic and Hurrian epithets are mixed ( X e l l a 1 9 8 1 , 303-21; P a r d e e 1997b, 66ff.). Besides this evidence from liturgical practice, the Hurrian pantheon in Ugarit was also known through canonical lists (?), as shown by texts such as K T U 1.26 = R S 1 - 1 1 . [048] (broken; note the attributive m o r p h e m e -d) and is included in others of a ritual nature such as K T U 1 . 1 1 0 = RS 2 4 . 2 5 4 and K T U 1 . 1 1 6 = RS 2 4 . 2 6 1 (cf. K T U 1 . 1 4 8 = R S 2 4 . 6 4 3 : 1 3 - 7 ) . By combining the various series it is possible to sketch out the following list of the Hurrian pantheon exacdy as it was known and used in Ugarit. Little is known about its actual meaning, for which reference may be made to discussion of Hurrian religion. 2 ' It is given here together with a normalized transcription (in Akkadian) ( L a r o c h e 1 9 8 0 ; W i l h e l m 1 9 8 2 = 1 9 8 9 ; for the Hurrian normalisation cf. D i e t r i c h - M a y e r 1 9 9 4 , 1 9 9 7 ) . KTU 1.110

KTU 1.116 KTU 1.132 KTU 1.111

in tin/Eni Salanni in atn!Eni attanni il/Ilu ttb/ Teššub/p kdg/ Kušuh irw prz/(?) kmrw/Kumarw/bi

'ttrt tutk kzg in hmn nnt kit nbdg bbt/Bibita tutk/ Sawuška in Ilm in atn il ttb tutk kmrb kzg

iy/E(y)a attb/Aštabi 'nt/'Anatu tmg/ Simike

attb in ard in hmn

in tin hbt šbdr dqt hdn Ijdlr hnng nbdg

in atn il ttb kmrb kdg in prz nkl yrh

tgn kid

in sin il ttb

Ugaritic Pantheon ilib il dgn b'l yrh mlk nkl yrh

kmrb kdg iy iy attb in ard tmg

'ttrt špš

23 Cf. V I E Y R A 1970, 5 3 8 - 6 6 ; V O N S C H U L E R 1965a, 141-215; W I L H E L M 1982, 6 9 - 1 0 5 ('Götter, Mythen, Kulte und Magie') = 1994, 49-76; D I E T R I C H - M A Y E R 1994, 7 3 - 1 1 2 ; (1995, 7 42); M A Y E R 1996, 2 0 5 - 1 1 ; P A R D E E 1997b, 63-80; D I E T R I C H M A Y E R 1997, 161-81. T h e r e is a great deal of material in the first two volumes in the series Studies on the Civilization and Culture of Nuzi and the Hurrians, Winona Lake, IN 1981/1987.

nkl/Nikkalu in ard/(?) nbdg/Nub adig

nbdg 'nt tmg pddph/Pišašaphi hbt/Heb/pat dqt/D/ Taqitu hdn/Hud/ tena hdlr/Hud/ tellura išhr/ Išhara aln/Allani nkl/Nikkalu nnt/Ninatta klt/ Kulitta adm/Adamma kbb/Kubaba

T h e first series of the Hurrian pantheon of Ugarit (in tin, in atn, il, ttb, kmrb, kdg, iwr prz, nkl, tutk) corresponds to the beginning of the Ugaritic canon and in fact represents the group of the great Hurrian gods, repeated with remarkable consistency in K T U 1.110 = RS 24.254:1-5; K T U 1.111 = RS 24.255:3-5, 8 - 1 0 ; K T U 1.116 = RS 24.261:11-4, suggesting a certain established sequence in which the position of certain deities can fluctuate and Sawuška is not always present. T h e mother goddess Hebat is missing. This series, in which Teššub does not occupy a privileged position, is in fact an adaptation of the classical Hurrian pantheon to the canon of Ugarit. A second series (cf. K T U 1.110 = R S 24.254:6-10; K T U 1.111 = R S 24.255:10-2; K T U 1.116 = R S 24.261:14-7) includes the following deities: iy / Eya attb / Aštabi 'nt / 'Anatu in ard / (?) in hmn / (?) tmg / Simegi nbdg / Nubadig

Kotaru(?) 'Attar(t)u 'Anatu (?) (?)

Sapšu

(?)

These two series are completed by the list of gods provided by texts such as K T U 116 = R S 2 4 . 2 6 1 : 1 8 - 2 3 a n d K T U 1.132 = R S 24.291:4-12, already difficult to square with the Semitic pantheon. This is, in principle, a cultic pantheon known from the sacrificial texts. However, the syntax used for listing some of them makes them equivalent to a 'god list'. W e are unable, however, to decide whether

the lists are canonical or simply functional. T h e same problem is presented by K T U 1.42 = R S 1.004, a cultic text in which seventeen repeated records list as many Hurrian deities in this sequence: 1. 2. 3. 4.

in atn il kmrb ttb lb iy kdg

5. 6. 7. 8.

ddmš Íu[k] ]nd attb

9. hdn f}dlr 10. tgrbn pddph 11. tmg 12. irxxi/hdrp

13. cnt 14. ibnkl pdgl 15. nbdg 16. in k[]l/d in trhn in atthn ttb 17. bbt

These are the same gods just quoted but in a different order.

2.3

Lists of offerings

As was noted above, most of the texts labelled as 'cultic' or 'ritual' are really only 'lists of offerings' where the divine recipient is noted. T h e y can thus be considered as administrative accounts which entail descriptive ritual elements. Their classification will be along similar lines to those used before: first we shall consider lists which are independent and then those which form part of more or less complex cultic contexts. We shall only consider data that are certain; disputed or purely hypothetical elements are dealt with in a description of the Ugaritic cult as a whole. 24 2.3.1

Lists without offerings

2.3.1.1 Record of sacrificial material ( K T U 1.91 = R S 19.015, K T U 1.48 = RS 1.019, K T U 1.87 = R S 18.056) a) Perhaps the clearest and purest example of a 'list of offerings', in terms of administration, is provided by K T U 1.91 = R S 19.015. 23 T h e introduction (lines 1-2), specifies the product (yrì) and the occasion of its generic use in the cult (dbh mlk). Next there follows the only example we have of a 'record of rituals' (lines 3-20), referred to by their keyword or brief description of the basic ceremony (del O l m o L e t e 1992a, 174-7; 1999, 2 5 9 - 6 4 , for its identification and comparison with extant rituals). O n the reverse (lines 21-34) are

24

Cf. in general DEL O L M O L E T E 1999, 7-23. Cf. V I R O L L E A U D 1965, 7-10; F I S H E R 1970, 49Iff.; DE M O O R 1 9 7 2 / 1 1 , 26-8; X E L L A 1979, 833; 1981, 335-6; D E T A R R A G O N , 1989, 174-7; D I E T R I C H - L O R E T Z 1988, 321-2; DEL O L M O L E T E 1992a, 171-7 = 1999, 254-64. 25

noted the places which are to provide the product mentioned, even distinguished as to category (jyn: msb/hsp). T h e final section (lines 35-36) gives the sum total of what has been provided. Although not really a ritual, it provides information about the elements of ritual (time, type, subject, predicate, object/material) and so forms part of the corpus of cultic literature. From it can be deduced clearly, unless the contrary is stated, that everyday wine of ordinary quality, the so-called msb was used even in the cult. As confirmed by the administrative texts ( d e l O l m o L e t e - S a n m a r t i n 1 9 9 8 , 1 9 2 - 4 ) , this was a product normally used in the Ugaritic cult and will again be mentioned in other lists of offerings considered here. Wine to be consumed . . . in the king's 'sacrifices': the 'sacrifice of Sapunu', Ibnm (shall provide) ten ('gallons') of wine hlb gngnt " " three " " of wine bir ( " " ) ten (kdm) of mz[b (wine) and two] kdm of hsp (wine) hpty " " two kdm of m^b (wine) Total (amount of) wine . . .".: sev[enty four kdm], and of hsp (wine): two, as one lot. (KTU 1.91 = RS 19.015:1-4, 21-36) As it is an administrative text, its interest lies in the material and the provider. Even so, it is not possible to be deceived about the value of this type of text. Not only do such texts give information but they also define a fiscal-cultic situation, and as such have prescriptive value, both for what refers to the centres providing material and for its use in the rituals mentioned. As such, the text could belong either to the archive of the civil or palace warehouse, or to the cultic or temple archive. In fact, there was only a single economy in these kingdoms. 26

26

T h e text was found in room 81 of the South-Western archive of the royal palace of Ugarit; cf. V I R O L L E A U D 1 9 6 5 , 7. On the relationship between both administrations in general cf. LIPINSKI 1 9 7 9 .

b) Similarly, K T U 1.48 = R S 1.019 27 records one material for offering, in this case 'birds' (csrm), intended for a particular type of sacrificial ritual ([dbh] tph bcl), noting in one case the divine recipient and in others the type of ritual use in it, and also noting at the end a pair of centres which provide it (lines 18-19). It does not preserve, however, the strict nature of a record which the previous text had, but instead is already close to the lists of cultic offerings to be seen below. We do not think, then, that it is a ritual of popular cult, (it was found in the house of the 'High Priest'), nor is it from a small group with meagre resources ( X e l l a 1984a, 165-8; P a r d e e 1988c, 185, n. 19). It is rather an administrative record related to the official cult, determined by the importance which birds had as an offering to the ins ilm, 'the divine peoples' in a sacrifice of a dynastic type (tph b'l) ( P a r d e e 1988c, 185; d e l O l m o L e t e 1987a, 66). T h e mention of the respective rituals and their divine recipients gives the text its 'ritual' character, at least as an 'agenda'. It is a pity that the poor state of preservation does not permit more definite conclusions. [Book-keeping record] of birds [(for) the sacrifice] of the stock of the 'Lord'/sovereign: 28 three birds for29 the 'Lady of the Mansions', three [. . .] for Dagānu, [ . . . . ] [ . . . .]; one tpš sn't game bird, one tr bist30 bird whose owner put hgm ρ [ ]rf;31 27

Cf. X E L L A 1981, 113-6; 1984a, 165-8; P A R D E E 1988b, 173-91; 1989, 43, n. 10; DE T A R R A G O N 1989, 167-8; DEL O L M O L E T E 1992a, 6 8 - 9 = 1999, 89-90. 28 Here, b'l is a royal title whereas in Ugaritic, the 'family of Ba'lu is pf)r/dr b'l. For the empirical meaning of Iph/s'ph cf. K T U 1.14 = RS 2.[003]+ i 24 (= bt, line 7). 29 P A R D E E , 1988c, 178, 186, suggests the restoration [/] before the I. Although not impossible, reference to a 'ram' would be strange if our text is understood as a record of 'birds'. 30 They would be two unidentified types of birds, with feminine adjectives ('disagreeable', 'foul-smelling'), the grammatical gender denoted here by the noun 'sr\ cf. X E L L A 1984a, 168; P A R D E E 1988C, 188. However, reference to a 'bull' would be odd in this and the other cultic texts, although the translation 'a bull which his master placed on the fire (b išt)' cannot be excluded; cf. DE T A R R A G O N 1989, 168, η. 89. 31 Possibly an unidentified type of plant which 'sweetens' 'wild' birds for offerings and sacred repast. T h e restoration l[al]rt would fit the context but is purely hypothetical; cf. P A R D E E 1988c, 182, 188.

two32 [birds] for [ . . . . ] ; one in the hmn [ . . . . ] two(?). On the (very/same) day of the sacrifice of the stock of the 'Lord': one for the 'libation'" of [. .], one for the offering to the [divine ]34 mountain, four birds, the 'estate' of trmn\ one, Salhu. c) It is probable that K T U 1.87 = R S 18.056:58-61 (DEL O L M O L E T E 1992a, 69 = 1999, 91) is also an account record in which are noted the contributions to the cult made by individuals, although this time we do not know what type of offering was involved. T h e fact that it is added to text K T U 1.87 (but not however to its duplicate K T U 1.41 = R S 1.003+), indicates the purely administrative value put on these texts and allowed them to be supplemented archivally with other notes of similar significance when space on the tablet allowed. d) A list of multiple offerings as a simple record with no divine recipient is also provided by K T U 1.148 = R S 24.643:18-22 (cf. below). T o complete the picture of pure cultic records, the various administrative texts have to be cited (category 4 in KTU) which record material for offerings, since they do not differ in genre and scribal structure from those above mentioned; they also mention festivals and gods who receive the offerings (cf. K T U 4.149 = RS 15.039 (wine; cf. K T U 4.213 = R S 16.17:24), K T U 4.168 = R S 15.082 (clothing), K T U 4.182 = R S 15.115 (clothing), K T U 4.219 = R S 1 6 . 1 7 9 : 1 - 3 , K T U 4 . 2 7 9 = R S 17.156(?), K T U 4 . 2 8 0 = R S 17.236:13-4 (silver), K T U 4.284 = R S 17.285:5-6(?) (wine and oil), K T U 4.781 = R I H 8 3 / 2 8 + : 1 - 2 (oil). However the accepted

32

Cf. P A R D E E 1988c, 182, 188; cf. G O R D O N , UT, 4 3 - 4 , on the ambiguous use of tn in respect of agreement of gender. 33 In view of the context and perhaps its parallelism with mkt, the meaning of which is quite definitely 'immolation' (cf. K T U 1 . 4 0 = R S 1 . 0 0 2 : 2 4 and par.; X E L L A 1981, 116), I prefer to understand mzy as referring to a type of offering, although difficult to justify etymologically (cf. Akk. maza'u, 'to squeeze', Arab, mazza, in connection with sour or semifermented drinks; cf. VON S O D E N , AHW, 6 3 7 ; CAD M / l , 4 3 9 ; L A N E , AEL, 2 7 1 0 ) ; see also D I J K S T R A 1 9 9 5 . In any case, the presence of a P N here is hardly likely in terms of context and morphology; cf. P A R D E E 1 9 8 8 C , 1 8 8 9 . There are several possible restorations for the end of the line; e.g. bn [il(m)/bt]. 34 Cf. K T U 1.3 = RS 2.[014]+ iii 29 (btk gry il spn).

distribution of the texts obviates that; texts which we have otherwise grouped together and analysed elsewhere ( d e l O l m o L e t e - S a n m a r t i n 1998). 2.3.1.2 Lists of assigned offerings ( K T U 1.90 = R S 19.013, K T U 1.164 = R I H 7 7 / 0 2 B + , K T U 1.168 = R I H 7 7 / 1 0 B + , K T U 1.162 = R S [Varia 20], K T U 1.148 = R S 24.643, K T U 1.105 = R S 24.249, K T U 1.109 = RS 24.253, K T U 1.46 = R S 1.009, K T U 1.130 = R S 24.284, K T U 1.43 = R S 1.005, K T U 1.41 = R S 1.003+, K T U 1.87 = R S 18.056). However most of the lists of offerings are to be found a m o n g the texts labelled rituals, the ritual nature of which is generally reduced to listing victims and offerings for each deity, with a generic description (in the heading or beginning of each section) of the cultic occasion and even the place where the offering takes place. This implies that these texts are basically lists of offerings and lists of gods (cf. above). a) As an intermediate stage between pure records of offerings and lists of assigned offerings can be cited a series of texts which in their fragmentary condition give offerings to one or two deities only: K T U 1.90 = R S 19.013, K T U 1.164 = R I H 7 7 / 0 2 B + , K T U 1.168 = R I H 7 7 / 1 0 B + . T h e y belong to the genre of 'visits'(?) (id yph mlk) and thus are close to the group of divination texts and oracular consultations in which we have already noticed a reduction in the number of gods mentioned. T h e king is expressly m e n t i o n e d as the officiant of the ritual. Otherwise, the texts do not specify the place of worship ( K T U 1.164 = R I H 7 7 / 0 2 B + 1: b hmrì), which has to be assumed, as does its moment in the cultic calendar. As most of the texts come from Ras ibn Hani, it is taken for granted that these rituals are celebrated in the cultic installations of the palace. 35 T h e textual agreement a m o n g them is surprising, with only the relevant verb (ydbh/yph) changing. T h e initial, single introductory offering is followed by the double šrp/šlmm ritual, all dedicated to the deity w h o m the king 'visits'/ 'sacrifices', sometimes accompanied by others in the repetition šlmm: ršp, cnt, ilib || -, cnt, ins ilm, il, šmn(?). T h e r e is also great uniformity 35

For a general bibliography on the texts from RIH cf. 212 n. 74 = 1999, 316 n. 75.'

DEL O L M O L E T E

1992a,

in the the distribution of the victims in the parallel sections of these texts: hgb/slh/0, ap wtips, ksp whrs || 0/hi, alp ws || / [ csrm\. K T U 1.164 = R I H 7 7 / 0 2 B + , which does not apparently pertain to a visit but to a sacrifice, is the one showing the greatest variation within the same pattern. T h e bad state of preservation does not allow all the information which these texts provided concerning Ugaritic liturgy in its various rituals to be extracted. O n l y K T U 1.164 = R I H 7 7 / 0 2 B + : 19-20 and K T U 1.90 = R S 19.013:20-2 provide the end section, which in both cases seems to refer to the desacralization process and the conclusion of the king's liturgical activity, a moment whose precise definition was seemingly important. b) O n e of the most outstanding examples of this type of offering list is provided by K T U 1.162 = RS [Varia 20], Here, the syntactic structure used (/ + D N + X), except in the first offering (to ilib), forces the series of divine names (cf. above) into first position, followed by the list of victims: invariably one ' r a m ' (s), except in the first case ('two head of cattle and two rams' to ilib) and the last ('one calf', to Sapunu). T h e beginning and end of the text provide ill-defined ritual elements. Entitled dbh il bldn, it quite definitely deals with one of the dbh mlk enumerated in K T U 1 . 9 1 = RS 1 9 . 0 1 5 (cf. line 6 : il bldn), possibly known as nskt qlc (cf. line 2 ) (DEL O L M O L E T E 1998a, 164-7). c) However an even more striking and complete example of a 'list of offerings', sacrificial in type, is provided by K T U 1.148 = R S 24.643, as is shown when set out as a table (DEL O L M O L E T E 1992a, 8 9 - 9 1 = 1999, 131-3). 36 T h e text is complex and contains several rituals, in the m a n n e r of the familiar anthology genre found in other texts of this type. Horizontal lines separate the said rituals. T h e chief concern of this a r r a n g e m e n t of the text into sections a n d cultic moments is the attribution and recording of the offerings. This does not prevent it from also providing important supplementary information concerning ritual (type of sacrifice, place, occasion . . .). T h e syntax is asyndetic (DN + X), which gives the 'god list' pride of place. 36

Cf.

1966, 279-82; V I R O L L E A U D 1968, 580 4; L A R O C H E 1968a, 517-518; 1970b, 306-12; F I S H E R 1970, 493-4; M I L A N O 1977, 23-4; C A Q U O T 1979, 1406; DE T A R R A G O N 1980, 109-10, 201 (index); 1989, 224-8; X E L L A 1981, 91-312; D I E T R I C H - L O R E T Z 1988b, 305 8; P A R D E E 1992, 153-70; D E L O I . M O L E T E 1992a, 88-94 = 1999, 130-8; W Y A T T 1998c, 427-9. ASTOUR

DE M O O R

T h e first section, corresponding to the dbh spn (lines 1 - 9 + 10-12; cf. K T U 1.91 = RS 19.015:3) presents the series of divine names which we call 'List A' (cf. above) and undoubtedly corresponds to the principal liturgy carried out by the dynasty in honour of the gods of Ugarit. T h e corresponding offerings are divided into three groups: 'one head of cattle and one r a m ' (alp wš) for the first four invocations, as well as for the following seven of Ba'lu\ for the remaining gods of the p a n t h e o n , 'one r a m ' , with an addition, with no specific recipient, of 'two head of catde, two birds and one cow'. T h e repetition šlmm (lines 10-12) mentions only 'one r a m ' for ilib and ilu, and 'one head of cattle for the other six b'lrrì (alp, kmm), but the logic of this distribution is not evident. Probably this second section represents only one incipit and the ritual šlmm included identical victims (alp) for the rest of the pantheon. 3 7 T h e following festival, 'when 'Attartu of the steppe enters the royal palace' (lines 18-22; cf. K T U 1.91 = RS 19.015:10) provides a simple record of offerings, with no divine recipient, apparently ceremonial rather than sacrificial in nature ( d e l O l m o L e t e 1992a, 93 = 1999, 136):38 articles of clothing and vegetable products. Lastly, the section on the reverse (lines 23-45) is a unit corresponding to the ritual of the 'gods of (the month) Hiyarvü and in this case the series of divine names corresponds to list B (cf. above). T h e victim offered is basically the 'ram', alternating with 'head of cattle and r a m ' offered to Ba'lu(?) (lines 26-45). e) Tablet K T U 1.105 = R S 24.249, 39 also concerning a ritual of the month Hiyaru, has the clear structure of a 'list of offerings'. Aside from the possible inverted order of the two faces, it presents a series of blocks of offerings distinguished by place, time and type of offering. However this is not a 'list of gods'. By the interchange of semantic patterns, victims are assigned to various gods and in particular, many 37 The third section (lines 13 17) is written in Hurrian, which is unusual for a mere account listing offerings, although the Hurrian insertion does not seem to be one. 38 The kind of offerings makes this section coincide partly with the one presented by the ritual of the 'Entry of 'Attartu of the tomb(s)' (KTU 1.43 = RS 1.005:1-5), although there animal victims were also offered. 39 Cf. V I R O L L E A U D 1968, 588-92; H E R D N E R 1978, 11-5; C A Q U O T 1977, 461-2; 1979b, 1408-9; F I S H E R 1970, 485-501; 1975, 142-7; DE M O O R 1970b, 318-22; D I E T R I C H - L O R E T Z - S A N M A R T I N 1975b, 158; DE T A R R A G O N 1980, 21-2, 165-76, 201 (index); 1989, 181-4; X E L L A 1981, 35-42; D I E T R I C H - L O R E T Z 1988b, 314-5; D E L O L M O L E T E 1992a, 206-8 = 1999, 307-10.

others are recorded without any specific assignment, in terms of the cultic installation in which the sacrifice takes place. Accounts are thus given for the following victims, according to cultic situation (including the purification of the king) and various recipients: on the day of the new moon, 'one head of cattle and one r a m ' (alp wš) to the b'lt bhtm; day 14, two 'rk-m ('puff-pastry cakes'[?]) to Ba'lw, [the 18th] in the gb of Sapunu as a /^-sacrifice, 'gold and silver' for t', 'two rams' for btbt as a holocaust, 'one head of cattle' as a peace-offering for Ba'lu, 'one bird' for Sapunu, 'one (piece o f ) offal' and 'one r a m ' for Raspu, 'two birds' for the ins ilm and (for another deity) 'two ewes'; in the gb of ršp, as a holocaust 'one r a m ' and 'one white ewe'(?); in the gb of Hiyaru, '38 sheep' (sin) and 'seven head of cattle'; in the temple of Ba'lu, 'two rams' (to the same?), 'one head of cattle and one r a m ' (alp ws) to Raspu, 'rams' from three different places to the b'lt bhtm and 'two rams' to Kôtaru as a tzgsacrifice, 'two rams' and 'one bullock' (pr) from a specified place. T h e gods venerated do not form any particular list. T h e restricted and domestic nature of the deities commemorated is striking. In this respect, the agreement with K T U 1.106 = RS 24.250+ and K T U 1.112 = RS 24.256 is impressive: b'lt bhtm, btbt, ins ilm, on the one hand, and ršp, b'l, spn [+ ktr], on the other. It belongs, therefore, to the range of palatine texts with their own pantheon and in connection with celebrations which are strongly funereal in nature. This was already to be assumed not only from the place of the offerings (gb), but also from the name of the month (hyr). f ) According to the heading, the Ugaritic text K T U 1.109 = RS 24.253 is a list of offerings made during the full moon in various sacred locations and according to a different type of sacrifice, 40 preceded by the well known ritual of 'purification of the king'. T h u s we have in this full m o o n ritual the well known multiplicity of sacrificial units on the same day in various sacred places, in palace

40

Cf.

1968, 592-3; C A Z E L I . E S 1969, 505; DE M O O R 1970b, 322 6; 1978, 16-221; X E L L A 1981, 49-54; D I J K S T R A 1984, 69-76; D I E T R I C H - L O R E T Z 1988b, 316 8 ; DE T A R R A G O N 1989, 188-91; DEL O I . M O L E T E 1992a, 182-5 = 1999, 271-7. T h e most important study of these texts (KTU 1.109 = RS 24.253 and K T U 1.46 = RS 1.009) is by Dijkstra, who completes it with other fragments (KTU 1.28 = RS I 11[021], K T U 1.56 = RS 1.044, K T U 1.31 = RS 1-11 [022], K T U 1.27 = RS 1.[064]; and perhaps K T U 1.134 = RS 24.294 and K T U 1.160 = RS 28.059); K T U 1.130 = RS 24.284 is a duplicate/partial variant of them. CAQUOT

VIROLLEAUD

1977, 462;

HERDNER

and town, mostly, also in this case, to the great tutelary gods of the dynasty and with the cult of the 'dead' predominant. T h e text, with alternating syntax (X + / + D N type of sacrifice D N + X), is arranged into two sections, separated epigraphically by a horizontal dividing line and defined by two 'types' of 'generic' cultic action which incorporate c o m m o n and known sacrificial actions. T h e first, of the 'šrt type (line 5), includes two series of sacrifices srp wšlmm (lines 10, 15), identical in one case (kmrrì) and different in the other. This twofold series, most probably carried out in two different sanctuaries (palace and town), closes with another šlmm-sacrifice (line 23), celebrated in a particular installation in an unknown sacred place (burbt, line 19). Whereas the second section, of the šnpt-type (line 24), also includes a repetitive series of srp wslmm-sacrifice (line 28, kmrrì), which in turn is followed by what can be considered a new series of šrp wslmm (lines 36-7), with an interlude (lines 29-31) in a new cultic installation of unspecified location (qgrt tlhn, lines 29-31). This ritual symmetry forces us to take into account the specific nature of the generic Cšrt/šnpt rituals, which should not therefore be placed at the same cultic level as the 'sacrificial' rituals. T h e y represent 'ceremonies of ceremonies', mixed rituals. Accounts for the following sets of victims are given: as a holocaust (srp): 'two cows one month old' (yrh) for Ba'lu of Sapunu, along with 'two ewes' and 'one dove' for Ba'lu; 'two loins' and 'one r a m ' for rml; 'one liver' and 'one r a m ' for šlm; 'one liver of one head of cattle' and again 'one r a m ' for Ba'lu of Sapunu; 'one ewe' for Sapunu. As a peace offering (šlmm), the same again. Again, as a holocaust: 'two livers' and 'one (piece o f ) offal' (for Ba'lu?); 'one cow' for ilib; for Ilu, Ba'lu and Pidrayu, 'one r a m ' each; for 'Anatu of Sapunu, 'one head of cattle and one r a m ' (alp wš). As a communion sacrifice: for ilib, Ba'lu of Ugarit, Ba'lu of hlb and Tarhu, one ' r a m ' each; for 'Anatu of Sapunu, 'one head of cattle and one ram'; for Pidrayu and ddms, 'one r a m ' each. A new series of communion victims includes: 'one r a m ' for each of the gods ilib, Dagānu, il t'dr b'l, Ba'lu, 'Anatu and Rašpu, as well as 'one head of cattle and one r a m ' for Ba'lu. T h e first series of the šnpt ritual includes: 'one r a m ' for Ilu, 'two rams' for 'Anatu of his; and for the gtrm, 'the left gsb of two head of cattle' plus 'one head of cattie and one ram'. T h e second part, besides 'thirty rams' spiced with 'tamarisk' for Ba'lu in 'the burner of the table of the Lady/ies of the Mansions', includes by double entry (kmrri), holocaust and communion: 'two calves [?]' (7m) for Ba'lu of Sapunu; 'one r a m ' for il bt[(?), Sapunu and Ba'lu of Ugarit, one each;

'one cow' for ilib·, and for Ba'lu of Ugarit and 'Anatu of Sapunu, 'one head of cattle and one r a m ' jointly. However, the smaller n u m b e r of deities c o m m e m o r a t e d in this liturgy is evident, with emphasis on the three great deities: Ilu, Ba'lu, 'Anatu, under various invocations. In the first section they are canonical deities (+ rms), whereas in the second section gtrm, il bt and b'lt bhtm are mentioned, invocations which place us squarely in the context of the dynastic ancestor cult (on the relationship of this series of gods occurring in these texts, K T U 1.109 = RS 24.253, K T U 1.46 = RS 1.009, K T U 1.130 = RS 24.284, cf. above). g) O n the other hand, the fact that the full moon ritual included in K T U 1.109 = RS 24.253:1-14 occurs literally in the sequence of a menology (K TU 1.46+ = R S 1.009+) shows the solid structure of each cultic unit, which remains unchanged. K TU 1.46:10-7 is, then, a word for word repetition of K T U 1.109:1-14, as we have just seen. Instead, the beginning ( K T U 1.46 = R S 1.009:1-10) belongs to another cultic unit, indicated by time (new moon) and is, more clearly than the following, a pure list of offerings. 41 It tells us nothing, however, about the local situation of the rites. Otherwise, the literary structure and the syntax are like those of the previous text ( K T U 1.109 = R S 24.253). O n the 1st day we have, apparently, the classic duo šrp wšlmm: 'one sltf and 'one (piece of) offal' (npš) for t' and 'two livers' for the 'god of the mansion' (il bt); equivalently, 'one r a m ' for ilib, Ilu, Ba'lu and Dagānu; 'one cow' for 'ttr-'ttpl and 'Anatu; 'one ewe' for Sapunu; several other victims are dedicated to recipients now missing. And also on the 3rd, this time with absolute clarity, and by double entry (kmm): 'one r a m ' for ilib, Ilu, Ba'lu, Atiratu and Yammu; 'one cow' for b'l knp and another missing deity, as well as 'one ewe' for Sapunu. O n this day in addition, a 'supplement' to the standard dual pattern, as was the case in K T U 1.109 = RS 24.253:19ff: '[one ram and] one head of cattle' for b'l and 'ttrt; 'two birds' to the ins ilm; 'two cows' to the two bbtm. In the group of deities mentioned in this coda to the 3rd day there occur the ins ilm, the deified dead of the dynasty, honoured by their habitual offering of 'two birds' (esrm), corresponding to the gtrm who occur in K T U 1.109 = RS 24.253:26. Reference to the bbt-m, which follows immediately, also gives the ritual the significance

41

Cf. GRAY

1989, 164-6;

1 9 6 5 , 9 0 - 2 ; XELLA DEL O L M O

IJÎTE

1 9 8 1 , 5 5 - 8 ; DIJKSTRA

1984, 69

1992a, 186-7 = 1999, 278-80.

7 6 ; DE T A R R A G O N

of royal cult which we had noted in this text, and mention of te (line 1) only corroborates this. h) Unfortunately fragmentary, K T U 1.130 = RS 24.284 42 is considered by Dijkstra to be a schoolboy copy of K T U 1.46 = R S 1.009:11-7 || K T U 1.109 = R S 24.253:3-18 and he suggests reading the sides of the tablet in reverse order. T h e sacrificial sequence of K T U 1.109 supports this. In spite of their variants, these texts are interesting inasmuch as they can throw light on the royal character of these rituals. These texts, then, fix the liturgy of the most sacred days of the lunar calendar, new moon and full moon, carried out in various sacred places of the Ugaritic cult, the urban and palace sanctuary, under the highest officiant, the king. T h e basic rite described here is that of sacrifice, in its various forms, accompanied as is usual by the purification/desacralization of the officiant. W h a t is not clear in this ritual, as in other rituals, is the principle which determines the choice of deities and their epithets, as well as the reladonship between each type of victim and the deity to whom it is offered. i) Another text which is, in fact, a pure list of offerings, apart from the non-sacrificial ritual elements at its close, is K T U 1.43 = R S 1.005. 43 T h e whole text is a single ceremonial procession in honour of various deities ('Attartu, the gtrm) who 'enter' the palace where they receive the offerings and a banquet is held (Cšrt). This 'processional' aspect connects it with K T U 1.148 = R S 24.643:18-22, where another rite of the same kind occurs. K T U 1.91 = R S 19.015:10-1 refers to these texts as dbh mlk rituals and our text could be one of them. T h e nature of the offerings are in line with this ritual, mosdy of the type 'clothes' or 'precious metal', as we have seen (cf. above), whereas the animal victims seem to be relegated to the background. Because of this we saw a reference to it perhaps in K T U 1.91 = R S 19.015:8 (izr) ( d e l O l m o L e t e 1992a, 175 = 1999, 260).

42

C f . M I L I K 1 9 7 8 , 1 3 5 - 8 ; X E L L A 1 9 8 1 , 1 0 1 - 4 ; DIJKSTRA 1 9 8 4 , 7 4 - 5 ; DE T A R R A G O N

1989, 43

216-8;

DEL O L M O L E T E

C f . FISHER

1970, 4 9 2

TARRAGON 1 9 8 0 , 9 8 - 1 0 7 ,

8 6 - 9 0 ; DE M O O R O L M O LETE

357-9.

1992a,

1992a,

188-9

4; DIETRICH 111, 113; 1989, =

1999,

280-2. SANMARTIN

1975g,

525-8;

1 6 1 - 3 ; CAQUOT 1 9 7 9 , 1406; XELLA

1987, 1 6 8 - 7 1 ; DIETRICH 189-94

=

LORETZ -

1999, 2 8 2 - 9 1 ;

LORETZ

1988b, 3 2 6 - 7 ; 1 9 9 1 , 8 7 ;

PARDEE

1993, 301

17; W Y A T T

DE

1981, DEL

1998C,

T h e inclusion of the genre (kt'rb) in K T U 1.91 = RS 19.015 already guarantees its royal character, which is also clear from the actual text as it mentions the place where the rite takes place (lines 2, 10: bt mlk) and from the behaviour of the king at its close (lines 23-26). This ritual colophon provides the situation and defines the 'genre' of the preceding cultic actions. T h e y are processions in which the king takes part, 'following' and 'carrying' the 'gods' (statues), going out to meet them and accompanying them as they are moved to the chapel or palace cultic installation, when they 'enter' the palace. These, then, are rituals, though we cannot say whether they are successive or separate, of procession or transfer of sacred images from their usual place (the temple) to the royal palace; rituals which are accompanicd, as is usual, by sacrifices and other offerings, once the gods have arrived and been enthroned in their new quarters. However, they do not provide us with an indication of the 'moment' of this ritual (or these rituals), no doubt well known to those using them as significant moments of the annual liturgical cycle, like Mediterranean pilgrimages and patronal feasts, which need no further specification in the calendar. T h e structure of the ceremonial supposed in the text is very clear, arranged into two explicit rites and another six which are merely mentioned, presumably partial repetitions of the second (double) rite. Of these eight rituals, the first (lines 1-8) forms a separate unit, with its own final rubric (lines 7-8: repetition 'seven times') which marks it off and makes it parallel to the seven other sections, which in turn are given this rubric at the end of the text (line 26: repetition 'seven times'). It all refers to the gtrm or cntm (the seventh is fragmentary). In its first section (lines 1-8) the ritual supposes three different types of offering-rite (Cšr, trmt, šlmm), directed to the (astral) gods and to Kôtaru (or Gataru), seven times to each of the two groups. It is strange to find that there are also 'seven' elements or materials for the offerings indicated here (lbš, ktn, ušpgt, hrs, s, alp, sin), which may not have any special significance, although it could correspond to the series of 'seven' offerings, instead of supposing a sevenfold repetition of the whole thing. In that case, the rubric would be explanatory. T h e 'time' (parrì) refers to the action and not to its object, as has also to be supposed in line 26. This way of glossing the sacrificial rite and the listing of its elements in the other rites where they are mentioned should be borne in mind as a peculiarity of these Ugaritic rituals.

It is most probably, then, a procession ritual which takes place in the cultic installations of the palace when the goddess ' 'Attartu of the tomb(s)' (DEL O L M O L E T E 1996a, 47-52) makes her 'entrance' into them. This invocation, in connection with the following section centring on the gtrm, as well as the reference to the 'Temple of the astral gods' make this text parallel to K T U 1.112 = R S 24.256, with its reference to the 'ascent' of the gods and of the royal family to the hmn of the palace' seven times' ( K T U 1.112:6-8), the offering to the goddess "Attartu of the tomb(s)' ( K T U 1.112:13; the only two references to this goddess in Ugaritic literature) as well as the appearance of the gtrm on the day of the full moon ( K T U 1.112:18). T h e second section of the ritual (lines 9-16), also an 'entry rite', has a clear twofold structure with respect to the deities (špš-yrfo, gtrc nt/ilt bt) and offerings (fors, ksp, ap-nps). W e do not know whether the final sacrificial element (alp wš) refers to the previous text, as a complementary peace offering, or to the following damaged text. T h e first hypothesis seems the more likely. At all events, it is clear that the epicentre of this section focuses on the gtrm and that in principle there are 'two' of them, which occasions the repetition of the ritual. This repetition of the ritual enables us to reach the number of seven cultic actions presupposed by line 26 and also mentioned in the first ritual (lines 7-8). With this it also partly agrees in the 'material' for the offerings (precious metals and animals, hrs/alp-š), besides the peculiarity of the funerary offering typical of holocaust, ap wnps. However, the type of offering is not specified, and nor of course, is the moment, although it is included as in the previous ritual, in the occasion commemorated ('When the . . . e n t e r . . .'). Instead, the place remains the palace, i.e. the ritual continues to be royal, as the deities celebrated indicate. In what remains of the lines in the second part of this section (lines 17-21) the repetition of the previous ritual is stated (without elaboration) twice more per pair, six in all (plus the cultic action presupposed in line 21). Finally, with line 22 a ritual of royal behaviour begins (lines 22-26), the third part of the second section of the text, which complements the previous section on offerings. Syntactically, the heading and the ending are predicative in structure, (DF. T A R R A G O N 1980, 98, 110), whereas the central section is of the standard nominal type: X + I, I + X, type + X. This double syntactic pattern agrees with the twofold sacred action or rite which makes up this ritual: procession and sacrifice.

j) A virtually 'pure' list of offerings with a duplicate (a rare phen o m e n o n in Ugaritic literature) is provided by K T U 1.41 = R S 1.003+ and K T U 1.87 = RS 15.130, with slight indications of time, place and ceremonial type; 44 they comprise one of the clearest witnesses of a monthly liturgy or menology (a sequence of sacred, especially sacrificial, celebrations, corresponding to particular days of one month), which has been transmitted to us among the rituals of Ugarit. It refers to the month of rišyn ('of the first/new wine') and of course has the format of a firstfruits ritual ('bunches/clusters are cut for Ilu') of 'New Year' (autumn). However, this aspect is blurred in the sequence of sacrificial rites. Unlike other texts of the same genre, here there is a surprising confusion and mix of the parameters temporal and local by which it is usually organized. Thus, for example, the numerical sequence of days is not followed. T h e reason here could be that attention has been given to the importance of the rites, which here as elsewhere correspond above all to ceremonies celebrated on the d a y / n i g h t of the 'full moon', and these have therefore been placed in first position. Full moon and new moon appear to be decisive dates for the 'temporal' parameter, whereas, the computation of the week also retains its significance, as we shall see, as is evident in other menological texts. Within this to some extent temporal confusion, the parameters of time and type of offering acquire special significance for understanding the overall structure of the text. It is divided into 6 sections and 2 appendices, which differ in each text; the one of K T U 1.41 = R S 1.003:50-5 is a sacrificial ritual text; the one of K T U 1.87 = R S 15.130 is, in turn, twofold: lines 5 4 - 5 7 , sacrificial ritual, lines 5 8 - 6 2 , a record (cf. above). T h e list of offerings is the following, distributed over the days of the month (is it one of the hdtm of K T U 1.91 = R S 19.015?): on the 1st day: 'one bunch of grapes (utkl)' for Ilu and 'two rams' for 'Attartu (cf. line 49; apparently in different ceremonies). T h e 14th day comprises 4 or 5 sacrificial rites of different type and recipient, accompanied by another group of non-sacrificial cultic actions, with the king as officiant, although the sacred place where they occur is not specified. It opens (lines 4 - 6 ) with the 'offering of the firstfruits' (ris argmn), possibly to the gods in general,

44

Cf. 157-65;

1963, 105ff.; DE M O O R 1972, 13-7; 1987, 1975e 143-6; X E L L A 1981, 5 9 - 7 5 ; D I E T R I C H - L O R E T Z 1988b, 311 4; DE T A R R A G O N 1989, 152 60; DEL O L M O L E T E 1992a, 72-87 = 1999, 96-128; L E V I N E - DE T A R R A G O N 1993, 76-115; L E V I N E DE T A R R A G O N R O B E R T S O N 1997a; W Y A T T 1998C, 348-56. HERDNER

DIETRICH

1956, 104-12; -

LORETZ

-

LEVINE

SANMARTIN

or to the supreme god Ilu, and refer no doubt to the products of the season of the year, primarily to wine (cf. lines 1~2: ris yn || ris argmn). It is accompanied (lines 5 - 6 ) by a fourfold sacrificial offering to the same n u m b e r of deities, according to standard alternating syntax (X + I + D N ; X + DN), with a total of 'six' victims ('four rams' and 'two birds'). 45 M o r e significant, however, is the group of these deities: b'lt bhtm, ins ilm, ils, ilhm (cf. above for their sequence and relationship to what is offered in K T U 1.39 = R S 1.001). If we except ils, who is apparently out of place (? absent in lines 23-39), the other invocations are typical of the 'palace' liturgy, a n d are inserted into the cult of the dead and deified kings and transformed into gods of the palace. T h e r e follow two rituals of the king, one of expiation and one of proclamation of this great day of the full moon of the month of ris yn, with no ritual of offerings specified. Next we have a new sacrificial act (srp) in the 'gml and the urbt; the second installation is already known from other texts. T h e recipients of the sacrificial victims are five or six deities: the great gods (?), Ilu and 'Anatu, on the one hand, and again the ilhm, together with tkmn wšnm and Rašpu, on the other, in two distinct groups, thus forming the same series that appears in K T U 1.39 = RS 1.001:2-4, clearly a palace text. Next (lines 13-17), the corresponding sacrificial action of the šlmm type is introduced, as in K T U 1.39 = R S 1.001:4-8. T h e r e follows a simple listing of a series of offerings and recipient deities, using the syntactic patterns X + D N , D N X, which alternate. O n c e again the most striking aspect is the reappearance in pride of place of the invocation ilh/ilhm, now forming an opening 'triad' (ilh, ilhm, ilhm), with five items of offering ('two ewes', 'one head of catde and one ram', 'one cow'). T h e r e follows a list of eight deities/invocations, 4 6 with seven victims ('five rams' and 'two cows'), which in fact represent a synthesis or summary of the official 'pantheon' in its various sections (cf. 'List A', above): b'l/atrt, 'nt/ršp, dr il, phr b'l, šlm, with the ubiquitous tkmn wsnm included a m o n g the goddesses, which does not occur there. Possibly it is a type of sacrifice of 'general communion', offered to the whole pantheon, as is made clear in K T U

15 T o these animal victims must be added 'the cruet of oil', so that we would have 'seven' units of offering, as a group complementing the offering of 'firstfruits'. Curiously, here dr il and pfrr b'l are grouped together whereas they are separate in the official 'pantheon'; cf. 'List A', above.

1.148 = RS 24.643:1-9. T h e precedence given to ilh or ilhm only attempts to ensure the ancestral meaning: positioned at the head of the other gods, ilh is here synonymous with ilib, just as ilib is synonymous with il at the head of the pantheon; the primordial 'divine being' of the gods and so of the actual divinized dynasty or mlkm, also present in the pantheon. T h e unity and peculiar nature of the following ritual action (lines 17-19) is confirmed as a new type of sacrifice by K T U 1.39 = R S 1.001:8-10, whereas in both texts we have different continuations. This time the sacrifice of 'burning' (urm) and the three combustible materials (entrails, cereal or early grass and grain) are again offered to the ilhm, this time called b'lm, who thus retain pre-eminence as recipients of offerings in the five sacrificial actions of this section. In this case the syntactic pattern is X + D N + X. new type of offering (m'rb) ( D i e t r i c h - L o r e t z - S a n m a r t i n 1975e, 145; X e l l a 1981, 66; L i p i n s k i 1985, 216-7; d e M o o r 1972, 15 n. 47; 1987, 161, n. 36)47 also begins the third ritual section (lines 19-22), comprising a dbhw of vegetables (oils and bread) in unspecified quantities, and of animal produce (honey and dove), remarkably unusual in Ugaritic liturgy and not dedicated to any particular deity. It is a global offering which yqh bt mlk, to be understood as a 'tribute-offering which the palace takes'. A

This meaning is confirmed by the next rite (lines 22-23), which specifies that another offering of the same vegetal kind 'wine and flour', and perhaps as part of the same m'rb sacrifice, has to be poured out, without indicating to which deity, in the gr. T h e offering of 'wine' returns us to the beginning of the text and together with the 'flour' 49 provides a suitable funerary communion offering. This closes the series of rites—seven in all—which are celebrated on the 14th day of the month in the palace and its cultic installations.

4/ Related to the root *'rb, Heb. ma'arāb, Akk. irbu, aibtu, the word seems to mean approximately 'entry', in a commercial sense; however it could also have a 'funerary' connotation in connection with the 'setting' said of the 'sun' which 'enters' (the underworld). 48 Here dbh denotes the type of offering, intended for a 'festal banquet'. Text and context are reminiscent of the dbh mlk of K T U 1.91 = RS 19.015:2; cf. DE M O O R 1987, 161, η. 38; DE T A R R A G O N 1989, 155, η. 54; DEL O L M O L E T E 1992a, 172 = 1999, 256. 4!) In the previous line 'leavened loaves' were mentioned, their correlative, just like wine is the correlative of oil-honey.

Next, the text takes us to another cultic scenario on the same date, the temple of the 'goddess'///«^, on the 'steps' of whose altar (lines 23-38) three series of sacrificial rituals take place. T h e first sacrificial ritual (lines 23-9), expressed according to the syntactic patterns X + / + D N || X + D N , comprises twelve sets of victims ('two birds' twice, 'two rams', 'four cows' and 'one ewe') and nine divine recipients: spn/glmt, yrh/nkl || b'lt bhtm, ins ilm, ilhm || sps, rsp. Even so, the particular meaning of the group and reason for each offering elude us; the second sequence repeats the beginning of the text, lines 5 - 6 , but without ils. T h e second series (lines 29-33) has the syntactic pattern X + D N and five deities celebrated with six victims ('two cows' and 'four ewes'): ilh, ilhm, il, tkmn wsnm, ilt, like that of lines 10-13, with 'nt for b'lt bhtm and without rsp. T h e sacrificial series closes with the offering of two victims ('ewes') with no recipient expressed, perhaps omitted by mistake, although the precise place of the ceremony is indicated and the sacrificial series is divided into two. In the third series (lines 33-38), there are again five/six recipient deities for the offerings: b'l spn, spn, b'l ugrt, ilib, atrt, ri-(?), to which six or seven sacrificial units correspond ('two cows', 'one ewe', 'two birds' and ' o n e / t w o rams'), assigned according to syntactic pattern X + / + D N . T h e group of gods is almost the same as in the following section, but in a different sequence and with some variations: in this last section Ilu has to be supposed for Ilib, as is usual, and considered as being offered the 'shekel of silver'; in line 41 Sapunu has been omitted. This section divides the sequence into two subgroups in respect of the place of offering: il, atrt, ins ilm, b'l Spn, Spn, b'l ugrt, who also occur in K T U 1.109 = R S 24.253:32-5. T h e break in the text prevents us being able to determine how the surprising remark 'thirty times' (which comes next and occurs in other texts) is connected with the preceding. In all likelihood it refers only to the last sacrificial offering, as can be inferred from the multiple parallelism with K T U 1.39 = R S 1.001:20-2, a parallelism which probably suggests reading the damaged In- as l(r)i[nš ilm], to w h o m inevitably there correspond 'srm, 'two birds'. T h e triple series closes with a unit of ritual which again mentions (lines 36-8) the sacred place in which it occurs and refers us back to the beginning (lines 23-4), a perfect literary inclusio: the altar of the temple of the 'Lady/ies of the Exalted Mansions'/Ilatu. T h u s far,

the rituals which take place on the 14th day, the day of the full moon, the sacred climax of the month ris yn, in two cultic locations: the palace sanctuary and the temple of Ilatu (possibly also located within the palace). T h e final section (lines 38-48) records the offerings corresponding to the first week of the month, on its last days, which are 'festive' days: the 5th, 6th and 7th. T h e ritual of the 5th day takes place (lines 38-45) in the temple of Ilu, so going back to the heading of the text, which presupposed the offering of the firstfruits to that god. T h e text is broken, but a first series of offerings is discernible: 'one shekel of silver' and various animal victims offered to Ilu, Atiratu and to the ins ilm, with the known pattern X + / + D N . T h e series next continues on 'the altar of ^a'/w'-presumably in the same temple (?)-with a series of sacrifices ('one cow' and 'two ewes') offered to various invocations of that deity, using the same pattern. T h e correlation of this series with the one occurring in lines 3 3 - 6 is surprising: both reproduce the two groups of deities, of Ba'lu and of Ilu, but in reverse order, and in both, the repetition of its last element a specified n u m b e r of times (30/22) is assumed, with or without a particular recipient. In this case it seems that the offering which has to be repeated is specified: 'one ram', 'one cruet of oil' and 'one cow'; an offering which is taken up again on the 6th day (lines 45-6): 'two cruets of oil' and 'one cow'. T h e only new element is a reference to the exact culdc installation, the cly, the 'sacrificial (place)', in which the ceremony is carried out and which has already occurred before (line 37), in connection with the temple of Ilatu/'hady of the Mansions', the location of the previous series of ritual actions. T h e sacred and sacralizing ritual ends on the 7th day (lines 47-48) with the desacralization of the (feast) day and of the king officiating at sunset. This refers us back, by literary inclusio and ritual sequence, to the new purification, which will take place on the 13th and 14th days (lines 3 and 6-7), and to the following 'proclamation' of the day (lines 7-8), related to the beginning of our text. We have already anticipated above how this ritual ends (lines 48-9) with the return to the ceremonial of the 1st day of the month (new moon) in order to complete all its elements (lines 1-2), thus closing the text almost as a literary unit with a final inclusio. T h e liturgy of the month of ris yn is thus outlined in its first two weeks between the climactic limits of new moon and full moon. T h e rites of the first week (1st, 5th, 6th and 7th days) take place, apparendy,

in the temple of Ilu, whereas those of the end of the second week (13th and 14/15th days) take place in the cultic installations of the palace and in the temple of Ilatu/'Lady of the Mansions', possibly part of them. T h e appendices K T U 1.41 = R S 1.003+:50~5 and K T U 1.87 = R S 18.056:54-7 include two specific rituals which simply mention the victims required. T h e first describes with sufficient detail, 'the Canaanite ritual of huts' in which are offered 'one r a m ' in holocaust and as a peace offering 'one head of cattle and one r a m ' (alp ws), seven times, to an unknown deity (prgl.sqrn) (on this cf. d e l O l m o L e t e 1992a, 84 = 1999, 123-4). T h e second simply notes the offering of 'one r a m ' on the 14th day of a certain month (for K T U 1.87 = R S 18.056:58-61; cf. above § 7.2.2.1.1 'Record of sacrificial material', c). As a whole, this text has the structure of a multiple and complex mixed sacrificial ritual (temporal-local-typological), i.e. an ordo mensualis or ritual 'menology' of the 'New Year' liturgy in Ugarit. 2.3.1.3 Lists of offerings in context ( K T U 1.112 = R S 24.256, K T U 1.115 = R S 24.260, K T U 1.43 = R S 1.001, K T U 1.106 = RS 24.250+) a) Instead, in the ritual K T U 1.112 = R S 24.256 50 the offerings are set within more explicit cultic contexts. Only lines 2 2 - 7 are in the nature of a 'list of offerings' which have been allocated, with a syntactic variant. T h e text is a collection of ceremonies which take place also 'in the month of Hiyaru(?Y, and within it, ten 'feast days' are specified. These, calculated from the new moon ( d e V a u x 1961, 468ff. ' T h e Liturgical Calendar. T h e new moon'), are the 1st, 3rd, 7th, 8th, 11th, 13th, 14th, 15th, 16th, 17th days. In this text, various rituals of a sacrificial nature, of offerings, are joined with others which are non-sacrificial, such as rituals of purification, procession or oracle, which require special consideration. T h e first fact that stands out is that each type of offering ritual (hdrgl, iyn, šnpt) begins on the first day of the corresponding first three weeks of the m o n t h (1st, 8th, 15th). 50

Cf. H E R D N E R 1978, 21-6; X E L L A 1981, 43-8; 1989, 197-9; D I E T R I C H L O R E T Z 1988b, 315-6; DEL 1999, 232-53.

1980, 113-36; 1992a, 156-70 =

DE T A R R A G O N OLMO LETE

T h e offerings are arranged into three categories: animals (sheep, cattle, birds), vegetables (grain/bread, wine), minerals (silver, pottery), with a clear tendency to grouping them into sets of 7 and 3 units of offering ( 7 / 3 / 7 / 7 / 1 4 / ? / 3 ? ) . 5 1 T o this cultic and sacrificial arithmetic must be added the 7 'ascents' made by the king's sons on the first day of the month, perhaps in connection with the 7 sets of offering. Distributed according to their recipients, these offerings correspond to the tutelary gods of palace and dynasty. In reality they receive, as we have seen, the offerings of the royal family on the 1st day: b'lt bhtm, the 'Lady of the Mansion(s)' ( d e M o o r 1970a, 199 ('Anatu?); 1971, 8 5 - 6 n. 4, 95 ('Mistress of the Mansion'); d e T a r r a g o n 1980, 163ff.),52 'one shekel (of silver)' and 'one platter', as well as 'one r a m ' and 'two birds' t o / f o r the ins ilm.™ T h e 8th day: the offering of 'grain/bread', 'one shekel (of silver)' and '(one j a r of) wine' is directed to 'Attartu hr, whereas on the 14th day the mysterious gtrm receive 'two rams' ( d e l O l m o L e t e 1992a, 161-2 = 1999, 239-40.), most probably belonging to this sphere of dynastic deities. O n the 15th day, the climax of the festival, the following are offered from the sacrificial hecatomb: 'one r a m ' to each of the gods Ilu, Ba'lu of Sapunu and Ba'lu of Ugarìt; 'two' to btbt and the same n u m b e r to Atiratu; lastly, to the il msd/ilt mgdl, 'one platter/one ram' (depending on the reading), plus 'seven cows' and 'fourteen ewes'. O n the 16th day, 'one r a m ' is again offered to btbt (there follows a lacuna in the text). Finally, on the 17th day, the deity iln receives three(?) unspecified offerings. As is evident, the deities venerated form a small group, even if the lost sections of the text arc taken into account. Apart from the classic group of the great gods of Ugarit (Ilu, Atiratu, Ba'lu, 'Attartu], we also have a collection of deities, generally under obscure descriptive names, who form part of what we may call the 'tutelary gods' of palace and dynasty (cf. List C above): b'lt bhtm, ins ilm, gtrm, btbt. This limitation of divine recipients in turn corroborates the 'private', 51 Similar series of 2 / 7 / 1 4 animal victims can also be noted in the Hebrew Bible (cf. e.g. Num 29:2ff.) and in other Ugaritic texts; on this, also in connection with the materials for offerings (silver, grain/bread, wine), cf. DE T A R R A G O N 1980, 34ff., 43ff. Of course here 'bread' and 'wine' stand for the respective unit, as specified elsewhere: 'one measure/loaf o f . . .', 'one jar o f . . .'. The offering of king Keret also includes bread and wine (cf. K T U 1.14 = RS 2. [003]+ ii 16-9). 52 O n the other hand, the empty space in line 2 suggests that another deity received the two rams mentioned first. 53 For the various interpretations of this epithet cf. D E L O L M O LF.TE 1987a, 66; 1992a, 92 n. 134, 160 n. 68 = 1999, 134 n. 167, 239 n. 70.

palace nature of this ritual ( d e l O l m 5 9 - 6 2 ; d e T a r r a g o n 1980, 162-9).

o

L e t e

1992a, 4 4 - 7 = 1999,

b) K T U 1.115 = R S 24.260 5 4 has particular importance for the information it provides about the royal pantheon, as well as for its clear structure, which almost suggests that this text is literary. Ultimately, however, it is a list of offerings in a clear context. This royal ritual takes place in the palace, more specifically, in its sanctuary (line 7: qdš), as we already know from other texts. It is of the dbh type (line 1) and includes two banquets (lines 8 / 1 9 : t/ylhm), one connected to a hll y dm-ritual (lines 6-8) and the other forming part of a slmm sacrifice (lines 9-10). T h e m o m e n t is given in the opening statement ('When the king sacrifices to Ušharayú'), although we do not know when it actually took place. Its basic syntax is sacrificial, with alternating patterns: X + / + D N / X + / + D N + type / X + type / / + D N + X , with two prescriptive-explanatory verbs: ydbh, t/ylhm. T h e most striking and peculiar thing about this ritual is its dedication to only one pair of deities (usfyr, bbt), who seem to be connected with the palace cult as tutelary or 'patron' gods of the palace, whereas they do not belong to the official pantheon of Ugarit (del O l m o L e t e 1992a, 5 4 - 8 = 1999, 71-8) with the specific invocation under which they are c o m m e m o r a t e d here. However, they are the first two deities of the dynastic pantheon (list C; K T U 1.102 = R S 24.246:1-2). This divine couple is the structural element of the text, which is arranged in repetitive series in a concentric chiasmus. T h e text shows a marked concentric structure with a symmetrical twofold distribution of offerings, like the one seen in K T U 1.43 = R S 1.005, there centred on the deities gtr and 'Anatu: 'ram', 'dove', 'ram' || ' r a m ' || 'ram', 'ram', 'dove'. T h e ritual ends with an indication of time which claims to specify, in line with the normal semantics of ahd, the occurrence of the double sacrificial-banquet rite on one and the same day.

54

Cf. V I R O L L E A U D 1965, 586-8; F I S H E R 1969, 197-205; 1970, 49Iff.; 1975, 139-41; DE M O O R 1970b, 316-7; L E V I N E 1974, 9-11; D I E T R I C H - L O R E T Z - SANMARTIN 1975i, 1975j; C A Q U O T 1979, 1407; J A N O W S K I 1980, 246-9; DE T A R R A G O N 1980, 87-91; 1989, 200-2; X E L L A 1981, 105-8; D I E T R I C H - L O R E T Z 1981, 85-8; 1988b, 319-20; A A R T U N 1984, 6-7, 18-9, 24-5, 32-3; 1985, 21; D E L O L M O L E T E 1992a, 177-8 = 1999, 264-6.

c) K T U 1.39 = R S 1.001 provides various series of'sacrifices', holocausts and communion sacrifices, probably offered to the dead kings under their various invocations: t\ ilh, ilh b'l, and of course, ins ilm. T h e r e is an actual moment, 'the night of Sapšu pgr wtrmnm' (of the ' d e a d / f u n e r a r y offering' and of the 'sovereigns'), in which these trmnm are honoured with a sacrificial ritual which involves all the gods of the dynastic pantheon, as we saw above (List C; lines 12~9). It is a text-list of sacrificial liturgy in which a series of offerings/ victims is itemized following the known formulae of sacrificial syntax. 55 T h e r e are very few temporal and local parameters and they are of no help in determining clearly the structure of the text. At all events, from line 12 it seems to be divided into two parts each with the same n u m b e r of lines, which in respect of the cultic ' m o m e n t ' and the list of gods which the second provides, are structurally and intentionally different. T h e ritual of the second part (lines 12-22) takes place III, 'at night', and in bt mlk, 'the palace'. These specifications and the systematic use of the god list of K T U 1.102 = R S 24.246:1-14, referred to above, indicate that the whole thing is a royal dynastic ritual of the funerary cult, presented autonomously, independent of the day of the month in which it is to be celebrated. Instead, the text has another parameter which is much better for determining its structure. This is the one connected with the ritual typology which is specified with unusual precision and in this matter is very like K T U 1.112 = R S 24.256. Its first part, (lines 1 - 9 , 9 - 1 0 , 10-11) is thus arranged according to four different types of offering/sacrifice (šrp-šlmm, urm, šnpt), each one with its particular series of gods, some even repeated, as might be expected a prion. This sequence has a parallel in K T U 1.41 = R S 1.0034:11-17, 17-19, which gives it its 'standard' character (cf. above 'Lists of gods in sacrificial texts' § 2). T h e second part (lines 12-19, [19]-20-22), instead, with a list which is itself already a unit, as we have seen, presents one single type of sacrifice, unspecified. T h e end of the text (lines [19]-20-22, reverse of the tablet) is a supplementary ritual, which could be understood as the 'thirty times' repetition, i.e. over a ' m o n t h ' (pgr?), of the offering/sacrifice to 'two' new deities (b'lt bhtm, ins ilm), and thus as an extension of the standard

55

Cf. D I E T R I C H - L O R E T Z - S A N M A R T I N 1975e, 1 4 1 - 3 ; X E L L A 1979-1980, 147ff.; 1981, 76-80; DE T A R R A G O N 1980, 65ff.; 1989, 135-9; D I E T R I C H - L O R E T Z 1988b, 3 0 8 - 1 0 ; D E L O L M O L E T E 1992a, 145-9 = 1999, 213-9.

list of K T U 1.102 = R S 24.246:1-14, as a synthesis of the royaldivine list of the reverse (cf. above 'List of Names of Divine Kings'). T h e distribution of offerings is as follows. As a burnt sacrifice: to the t\rri) as a whole, 'two ewes', 'one dove', 'two loins and one liver from a head of catde'; to Ilu, 'one ram'; 'one cow' to the ilhm·, 'one ewe' to tkmn-šnm and another to Raspu. As a peace offering: to the ilhm as a whole, 'two ewes', 'one head of cattle', 'one ram', 'one cow'; to Ba'lu, Atiratu, tkmn-šnm, 'Anatu and Raspu, 'one r a m ' each; 'one cow' to the groups dr il and phr b'l and another to šlm. In a burnt sacrifice (urm): to the ilhm b'lm, 'a burnt heart' and 'fifteen measures of forage and spelt'. 56 In a sacrifice of 'presentation' (half of it?): one ' r a m ' each to b'l spn, b'lt bhtm, ilt mgdl and ilt asrm. Lastly, as a /)gr(?)-sacrifice: to the pantheon of K T U 1.102 = R S 24.246:1-14, 'one cow' to ten of its deities, 'one ewe' to another four, 'one r a m ' to one deity and two unspecified offerings (dbhrrì) to two others, without it being possible to determine the meaning and significance of this variety of victims. In the final 'coda': 'thirty times one cow' to the b'lt bhtm and 'two birds' to the ins ilm. W e can, then, understand this text as a 'sacrificial agenda' which lists the various types of sacrifice related to the royal funerary liturgy: šrp-šlmn, urm, šnpt, pgr and the 'daily sacrifice', with the time and place of celebration unspecified. d) K T U 1.106 = R S 24.250+ is also a royal funerary ritual of a particular month, the month of gn;57 there victims are offered to all the 'divine peoples' and to some dead kings (the most recent?) in particular, under their 'divine' names: yarsil, ydbil, 'mtr (cf. 'Lists of Names of Divine Kings', above § 7.2.2.2). T h e sacred time corresponds to days 8, 22 and 25 (cf. lines 18, 24, 25-26). This implies that the ceremonies mentioned in lines 1 - 1 7 (18) took place on the first or on subsequent days of the first 'week' of the month in question. As for the sacrificial rite, we have the known type šrp, '(as a) holocaust' (cf. lines 2 and 7), repeated twice, apparently corresponding to the twofold mention of the recipient deities (ršp/ ins ilm/group of three || ršp/inš ilm/[?]), and of the victim (hgb/npš wš + 'srm). T h e 56 This sacrificial offering is unique in all the cultic literature of Ugarit. On the meaning of the types of rituals termed urm and šnpt and their possible relation to šrp wšlmm cf. DEL O L M O L E T E 1 9 9 5 , 4 5 , nn. 4 9 , 5 1 . 57 Cf. H E R D N E R 1978, 26-30; X E L L A 1981, 8 1 - 5 ; D I E T R I C H - L O R E T Z 1988b, 324-5; DE T A R R A G O N 1989, 185-7; DEL O L M O L E T E 1992a, 149-56 = 1999, 219-32.

remainder or second part of this initial ritual of sacrifice has elements remarkably in parallel with those in K T U 1.112 = RS 24.256: 1) the participation of the royal family, 'sons' and 'daughters', and then of the king himself (cf. lines 9 - 1 0 , 17); 2) the performance of the cultic action in the cult installations of the palace: the sacrifice of 'one r a m ' to Pidrayu and 'seven ewes' in the hmn, 'one r a m ' in the qds/'ly/hmn of Nikkalu, another in the kbm and 'one cow' 'at the door of the royal mausoleum' ( d e l O l m o L e t e 1992a, 153 = 1999, 227); 3) the sevenfold rhythm of the ceremonial (cf. K T U 1.106 = R S 24.250+: 12-3; K T U 1.112 = R S 24.256:6-7); 4) in both texts the 'cantor' takes part (lines 15—7), and there is an 'oracular response' (ttb rgm) in both, at a later stage (cf. lines 23 and 32). Also in both, but at different times, the desacralization/ purification rite of the king is mentioned, although this fact is not of itself significant. O n the 8th day the sacrificial rite includes, together with the usual victims numbering 'seven cows', 'fourteen ewes' and 'one (piece o f ) offal' (cf. K T U 1.112 = RS 24.256:26-7), a new class of offering, unknown in the other rituals of Ugarit: 'fish stew' (šbšlt dg). Similar ritual elements recur and close the cultic action of the third week (cf. lines 32-3). T h e decisive day of the ritual is the 25th. It comprises a sacrificial-banquet) at night before the king, seated on his throne, in honour of and in communion with arsy and 'gods of the underworld' (ilm ars) with 'provisions' (šlhmt) selected and prepared beforehand (cf. lines 25 and 28: 'nine ewes', 'one head of cattle' a n d 'one goose'), offered to the gods, and 'two ewes' and 'one cow' offered to the goddess. As for place, the cultic action shifts on day 8 from the cult installations of the palace mentioned in lines 13-4, to a new sacred location, known only from this text and giving its name to the month of the ritual: the 'garden' (gn) ( d e l O l m o L e t e 1996a, 103-4, for the identification of the gn), which fits in well with the 'infernal'/'funerary' nature of the deities honoured in this ritual. Apart from the sacrificial moment (lines 1 - 1 5 , 19—22, 30-31) in honour of and communion with deities either specifically mentioned or assumed, it also includes the desacralization/purification rite of the king, already known from other texts (lines 24, 26, 34), and two

new royal ceremonies, which can be considered peculiar to it and to which we have already referred: the singing before the king and the preparation of the funerary banquet. Its interpretation must, however, be left for another systematic analysis which goes beyond consideration of the text as 'list of offerings and gods'. e) Together with mere lists of offerings like those above, a text such as K T U 1.119 = R S 24.266 58 provides a more complex context, a mixture of list a n d elements of recitation; more specifically, a psalm of prayer in the event of a national threat, the only one in any text of this type. It poses the question, at least in this case, of whether they are administrative (cf. above). However it has to be remembered that the tablet was found in the 24th campaign (1961) in the library of the priest bn agptr, i.e. in principle in an archive more concerned with ritual than with administration. T h e relationship of this cultic element to the foregoing list of offerings is not clear. A 'list' which itself already provides 'rituals' that transcend the simple record of a victim: exact sacred place and time, type of ceremony and in particular, repeated mention of the royal desacralization rite, which may or may not have included some sort of victim to justify the allusion in a simple record of offerings. N o r can we be sure whether both sides of the tablet belong to the same text or to two different texts. T h e second (the reverse) could quite well be an authentic liturgical ritual, even though it also begins like a list of offerings (lines 18-22). In fact the lower part of the tablet is broken, which means that we have two separate texts, obverse and reverse, corresponding perhaps to two separate rituals, either of the same m o n t h or of different months. T h e obverse (lines 1-17) has a list of offerings which corresponds to the sacrificial ritual of the month of ib'lt, specified for various days: 7th, 17th and 18th; this datum divides the text into three sections: lines 1 - 4 , 4 - 1 1 , 11-17. T h e first two are clearly distinguished by the sequence desacralization/purification of the king, which means that between days 7 and 17 the king does not celebrate ritually, whereas the ceremony of days 17 and 18 forms a single unit. This

58

Cf. H E R D N E R 1973, 693-703; 1978, 31-9; A V I S H U R 1978, 254-62; X E L L A 1978, 127-36; 1981, 25-34; D E T A R R A G O N 1980, 17ff.; 1989, 206-11; M A R G A L I T 1981b, 62-83; S A R A C I N O 1983a, 263-369; 1983b, 3 0 4 - 6 ; DE M O O R 1983, 251-2; 1987, 171-4; M I L L E R 1988, 139-55; D I E T R I C H - L O R E T Z 1988b, 819; D E L O L M O L E T E 1992a, 197-205 = 1999, 292-306; P A R D E E 1997a, 283-5; W Y A T T 1998C, 416-22.

liturgy takes place in the temple of Ba'lu, bt b'l ugrt (lines 3, 9-10), mgdl b'l ugrt (line 12), and then moves to the temple of Ilu, qds il, bt il (lines 6, 14) for the second and third rituals. Even the various areas of the temple of Ba'lu where the offerings are made are specified, which will not be discussed here. Also specified are the various sacrificial actions of the syntactic pattern ' X + I + D N [+ type of offering + officiant'] with, exceptionally, more verbal than nominal syntax. Victims: besides the usual ones (large and small flocks: ram, cow, bull, lamb), there occur others which are rarer or specific ([piece o f ] offal, dove, donkey). T h e officiant: when specified, he is the king, also the subject of other non-sacrificial rites (purification). However the strangest datum of this text comprises the (divine) recipients of the offerings, together with their type, an element which is by no means clear. In the first section (lines 1-4) 'two rams' are delivered for b'l and sacrificed in the temples mentioned. In the second section (lines 4-11) 'three cows' are delivered to the sanctuary of Ilu and to the b'lm and to the 'hero' (glrri), 'two sheep' ('ewes') and 'one cow' to the Damsels (glmt), plus 'one lamb' and 'one dove' which 'are burned (tmr) It',59 in the testimonial of the temple of Ba'lu of Ugarit'. As in other cases, the ritual of the 17th day forms a continual unit with the ritual of the 18th day, section three (lines 11-17), and so the desacralization of the king is not specified. O n this day the offering of 'one head of cattle' (alp) is made in the temple of Ba'lu; in the temple of Ilu, 'one (piece of) offal' (npš) for Ušharayu(?), another for Ba'lu and 'a donkey' for whom or why we do not know. T h e fourth section (lines 18-36, reverse) has two clearly differentiated parts: a ritual (lines 18-24 + 25) and a prayer (lines 26-36). In it are noted the victims which correspond to a 'week' of offerings (apparently reserved for Ba'lu), of which have been preserved those of the 'two birds' (twice), 'one liver' and 'one "little chain"(?) of a liver'. This complex of victims ('srm, kbd, ššrt) belongs more to divination or summoning rituals. It is possible, then, that although it takes place in the same sacred place, the ritual on the reverse is completely 59 This lexeme, t'(y), has normally been given the meaning of 'offering' in one of its types (which is not defined); however closer analysis makes it preferable to see here and in other texts a divine-cultic title of the dead and divinized kings of Ugarit, a title which is already attributed in epic to the legendary King Keret ( K T U 1 . 1 6 = R S 3.325+ i 24 and par.); cf. DEL O L M O L E T E 1988a. Cf. Wyatt 1998c, 200 n. 115.

different from the ritual on the obverse. It is not sacrificial, but a magical entreaty, with the king as subject in both functions. T h e weekly sequence of this section has a structure like that of the 'graded numerical sequence' in myth and epic ( d e l O l m o L e t e 1981a, 60-1), with a climax on the seventh day. T h e syntax is nominal as usual, on days (3 and) 4 and 5 (and 6) together (num. + X + [/ + D N + specification of place]), whereas it is verbal (ykbd) on days 1 and 2; on the 7th, also, as a recitation, not as sacrificial (dn). Analysis of these elements is left to the discussion of these texts as 'ritual' as are the place and mode of the action. T h e same applies to the fifth and final section (lines 26-36), which is a prayer. As a whole, this double complex ritual refers to only a limited range of 'deities' to w h o m offerings are presented (b'l, il?, b'l?/il?, b'lm, glm, glmt, t'/b'l?, isf}iy?/b'l? ? ?/b'l). From this point of view, the ritual is largely 'related to Baal' in respect of recipient and 'royal' in respect of officiant. At all events, this whole set of ritual and recited elements has already appeared in other Ugaridc texts, those called 'cultic myths' ( K T U 1.23 = R S 2.002, K T U 1.24 = RS 5.194; 'mythological rituals' instead are K T U 1.108 = R S 24.252 and K T U 1.161 = R S 34.126). In practice, this mix of cultic elements was probably c o m m o n e r than appears from their inadequate distribution in literature and cult (a category which is largely a matter of lists). Some scholars already speak of the character of myths as recitations, an element of ritual, and m a n y cultic texts must have had elements which presumably were recited, as is the case in the text analysed here. Ritual was never movement without speech, nor was 'myth' recited without movement: gesture and speech belong together. Here we conclude this survey of the texts of cultic practice in Ugarit in respect of 'lists of gods and offerings', a literary form under which many of them are given, although it allows glimpses of a more complex liturgy than the act of sacrifice sketched out in these lists and was its culminating sacred act. In this handbook another section is devoted to disentangling and explaining the complex ritual involved in the liturgy of Ugarit (see § 7.1). (Translation:

W.G.E.

Watson)

3

T h e

O m e n

P a o l o

3.1

T e x t s

X e l l a

Mesopotamian tradition

T h e great M e s o p o t a m i a n tradition of divination was well known a n d accepted in Ugarit, a city whose archives have preserved several tablets (mostly in alphabetic cuneiform) which belong to the same f a m o u s tradition. In particular, we have on the one h a n d some texts with teratological omens which concern predictions derived f r o m deformed or monstrous births of animals or h u m a n beings. O n e fragm e n t a r y d o c u m e n t instead reports several astrological omens, while a short tablet concerning an eclipse of the sun m a y also be noted; there are also several small models of livers a n d an inscribed model of a lung which derive f r o m the practice of extispicy. In general one can say that in Ugarit M e s o p o t a m i a n tradition was followed quite slavishly, even if there are some indications of adaptation and specific formulations connected with local requirements, both political and social. O n e significant fact, in this respect, is that it was sometimes felt necessary to compose the texts of the o m e n s in the local language, as witness to a certain desire for cultural appropriation.

3.2

The teratological omens

T h e teratological omens, that is, those which contain presages d r a w n f r o m the observation of d e f o r m e d births (or abortions) of animals a n d h u m a n s , are represented in Ugarit by three alphabetic texts, K T U 1.103+ = R S 24.247+, K T U 1.145 = R S 24.328A+ a n d K T U 1.140 = R S 24.302. T h e y all c a m e to light to the south of the acropolis, in w h a t has been called the Cella aux tablettes, which was part of the house of a famous diviner (here, incidentally, were also f o u n d models of livers a n d of a lung as well as the difficult [necromantic?] text K T U 1.124 = R S 24.272). T o this can be a d d e d a f r a g m e n t of a tablet in syllabic cuneiform (RS 7.001 = A O 18.892), as yet unpublished, which was f o u n d on the surface. 1

1

Essential bibliography: Η KRONER 1978;

DIETRICH

LORETZ

SANMARTIN

1975d;

This small group of texts is of the greatest interest because it documents translation into the local language of a type of divination text which is typically Mesopotamian (the Babylonian series Summa izbu studied by Leichty). 2 T h e documents in question, which are approximately contemporaneous with those found in the great Syrian archive of Meskené-Emar, 3 also allow us to examine further the history of the spread of this genre. In fact they date back to a period earlier than the corresponding documents in Akkadian and Hittite found at Bogazköy ( R I E M S C H N E I D E R 1 9 7 0 ) and represent the most direct chronological link with the Old Babylonian tablets which circulated in series before their final 'canonization' in 24 tablets, evidence for which is provided from Ashurbanipal's library at Nineveh. T h e Geburtsomina of Ugarit are typologically related to those from Bogazköy and, if we follow the classification proposed by Riemschneider, 4 these too can be divided basically into two types: K T U 1 . 1 0 3 + = R S 2 4 . 2 4 7 + and K T U 1 . 1 4 5 = R S 2 4 . 3 2 8 A + refer to anomalous animal births (which are to be set alongside Tablet V of the canonical series), whereas K T U 1 . 1 4 0 = R S 2 4 . 3 0 2 concerns h u m a n births (and seems to derive from Tablets I - I V of the canonical series). T h e teratological omens from Ras S h a m r a are a fairly faithful reflection of the 'principles' 5 of what is termed deductive divination, such as the classic opposition right—left, where the right refers to everything concerning the country and its inhabitants, whereas the left concerns the king and enemy countries. 6 In this perspective, every defect or anomaly found on the right comprises an unfavourable omen, but if it is found on the left it is interpreted favourably. Every 'increase'—even if strange or monstrous—of the right side, therefore, corresponds to events which are positive for the king and his kingdom. T h e absence of limbs or organs seems instead to portend bad omens.

XELLA

-

CAPOMACCHIA

1990a, 87-165; = 1999, 353-9.

LORETZ 2

LEICHTY

3

Cf.

4

RIEMSCHNEIDER

1979;

XELLA

DIETRICH -

1981, 191-206;

LORETZ

PARDEE

1986;

DIETRICH

1990C; DEL O L M O L E T E 1992a, 2 3 7 - 4 1

1970.

1985-87, V I / 1 - 2 and V I / 4 , 686-93. 1 9 7 0 , who correcdy distinguished on the basis of the protasis, texts introduced by the clause lák-ku-zu SAL-ζα . . . a-a-ši (human births) from those introduced by the clause ák-ku IZ-BU (animal births). 5 See, in general, B O T T É R O 1 9 7 4 . fi Cf. W Y A T T 1996c and below, § 13.2.2. ARNAUD

In line with the typology of these omens in the earliest period of their circulation, there are no references to the individual and private sphere: every prediction is of an official nature and in fact always refers to the king or the country. In the Ugaritic omens also one catches a glimpse of this association of ideas based on analogy: note for example the interpretation of certain bodily features of the foetus. A m o n g the examples that can be adduced the following can also be noted: if the foetus looks like a bird the prediction is that the gods will fly away from the land; the head of the foetus symbolizes the king or the governor of the land; missing or deformed feet or paws (which therefore prevent walking) are interpreted in the sense that the land will not be able to progress as usual in its existence due to epidemics or internal strife. In Ugarit, then, the dichotomy between (a) anomalies and malformations, and (b) similarities with animals has also been adopted. From the limited material available it is not possible to verify the presence and consequences of multiple body-parts and of a whole range of cases of deformity a n d / o r defects, but even so everything suggests complete agreement with the Babylonian model. In Ugarit as in Bogazköy there was, then, full acceptance of Mesopotamian tradition, even if the Old Babylonian originals were certainly slightly adapted for local needs which were exclusively of public concern. W h a t are called the 'stock apodoses' display a limited range of events which evidently reflects military politics on a small scale, as in Ugarit.

3.3

Astrological omens

Already attested in the Old Babylonian period, the Mesopotamian collections of astrological omens ( W e i d n e r 1 9 4 1 - 4 , 1 7 2 - 9 5 ) were circulated widely around the mid-second millennium bce and as a result we find copies scattered in 'peripheral' centres such as Bogazköy, Alalah, E m a r and even Ugarit. 7 At Ras Shamra in particular, besides a few as yet unpublished syllabic texts (RS 2 3 . 3 8 , R S 2 5 . 4 4 0 A and RS 2 5 . 1 4 1 + 4 5 4 f ) , a broken tablet in alphabetic cuneiform has been found in the northern palace of Ras Ibn Hani ( K T U 1.163 = RIH 7 8 / 1 4 ) . 8 It contains presages based on meteorological observations, 7 8

Detailed bibliography in D I E T R I C H L O R E T Z 1990a, 165 95. B O R D R E U I L - C A Q U O T 1980, 352; D I E T R I C H L O R E T Z 1990a, 165-95 (cf. 168-70);

of the sun and the moon, and is connected with the Mesopotamian series Enûma Anu Enlil ( W e i d n e r 1941-44). As usual, the sentences comprise a protasis and an apodosis and the presages concern the well-being of the king, the ruling house and their possessions. T h e text is too broken to supply information or provide specific parallels with the omens of Mesopotamian tradition. T h e most detailed analysis available, also at the comparative level, is by D i e t r i c h - L o r e t z (1990a) and for the time being it is not possible to progress further. T h e widespread belief in astrology in Ugarit is documented not only by references to the '(divine) stars' in some ritual texts and from allusions in texts such as K T U 1.23 = R S 2. [024] and K T U 1.19 = RS 3.322 + ii 1-3, 9 but also by an unusual document, K T U 1.78 = R S 12.61. It concerns an eclipse of the sun, the simultaneous appearance of the planet Mars and the presages derived from it.10 This astral conjunction seemed to take on a fatal and dangerous character, confirmed by the consultation of livers expressly carried out. At all events, the precise date of the event (probably March 5th, 1223 bce) is still debated by specialists and there is no unanimous opinion." T h e exceptional nature of the document, which lies halfway between astrology and hepatoscopy, further complicates its translation and interpretation. T h e translation provided here is still provisional and hypothetical: 1 2 3 4 5 6

bit. ym.hdt hyr 'rbt špš Igrh ršp kbdm tbqrn skn

DEL O L M O L E T E

-

LORETZ

In the si1ence(?)12 (of the month) hyr down went the Sun: her gatekeeper was Rašap. (For this) (1/2) liver(s) was/were examined: danger!

1992a, 236 = 1999, 352;

DIETRICH

LORETZ

1988b, 94-5;

DIETRICH

1990C.

9

As correctly noted by DEL O L M O L E T E 1992a, 236, n. 68 = 1999, 352, n. 69. Essential bibliography: DEL O L M O L E T E 1992a, 235 = 1999, 350; D I E T R I C H L O R E T Z - SANMARTIN 1974b, 464-5; F U L C O 1976, 38ff.; X E L L A 1981, 171; D I E T R I C H L O R E T Z , 1988b, 99, 100; D I E T R I C H - L O R E T Z 1990a, 39-85. 10

11

KUDLEK -

1987-8;

MICLER

1989;

1 9 7 1 ; SAWYER -

STEPHENSON

1 9 7 0 ; DE J O N G -

VAN S O L D T

1990a, 2 8 1 - 5 (contribution by W . C . S E I T T E R - H . W . D U E R B E C K ) . P A R D E E - S W E R D L O W 1993; W Y A T T 1998C, 366-7. 12 For a discussion of b It see D I E T R I C H - L O R E T Z 1990a, 50ff.; the meaning proposed here is hypothetical. WALKER

DIETRICH

-

LORETZ

3.4

Extispicy

Also typical of Mesopotamian divination tradition are the texts of omens written on some models of livers and on one model of a lung, found in the house of the so-called diviner-priest (which was probably also a sort of school), also quite an important person in his official capacity. T h e five models of livers we now have ( K T U 1.141 = R S 24.132; K T U 1.142 = R S 24.323; K T U 1.143 = R S 24.326; K T U 1.144 = RS 24.237; K T U 1.155 = RS 24.264) 13 report on the results of hepatoscopic analysis concerning various events and it cannot be excluded that they record independent developments in respect of the tradition from which they come. Whereas K T U 1.155 = RS 24.654 is too damaged to provide any information at all, the other texts are very interesting. K T U 1.141 = RS 24.312 refers to the acquisition of a youth (a slave? gzr) from an Alashiote (= a Cypriot) by Agptr, a well-known person in Ugarit; 14 K T U 1.142 = RS 24.323 mentions a sacrifice—dbh—in honour of a dead person (a sort of kispum?), on the occasion of which the commissioner asks the diviner for a response. T h e contents of K T U 1.143 - RS 24.326, on the other hand, are rather vague, as the text refers only to the performance of an extispicy requested by someone, without going into the causes which led to the sacrifice. Finally, K TU 1.144 = RS 24.327 is an omen which was probably requested by a prince with regard to the hupsu-mcrcenanes who must have caused worries at the political level ( D i e t r i c h - L o r e t z 1990a, 15). T h e model of a lung ( K T U 1.127 = RS 24.277), which merits a separate study, is also connected with a Mesopotamian tradition. 15 T h e incised text is divided into sections marked off" by lines which probably correspond to the anatomical regions of the lung. It seems likely, nevertheless, that the various parts of the text are connected to each other and comprise a unity, even if it is difficult to understand (cf. for example the uncertainty concerning the term nat which

13

Essential bibliography: 1990a, 24 Iff.; DEL

LORETZ 14

O n

15

KRAUS

him

liography in 1999, 91-5.

cf. COURTOIS

1985, 185 7; DIETRICH

XELLA OLMO 1979,

MEYER LORETZ

1981, 184-90; M E Y E R 1987, 218ff.; 1992a, 232-4 = 1999, 347-9.

DIETRICH

-

LETE

1271.

in D I E T R I C H L O R E T Z 1990a, 255ff. Specific bib1990a, 18; add DEL O L M O L E T E 1992a, 69 72 =

could even be translated 'removal'). 16 T h e most plausible hypothesis is that this model has been inscribed with several divination exercises, performed at different times, but concerning the same event and with the same aim: to avert the danger of an enemy attack by means of a series of rites, including also a ceremony belonging to the well-known typology of the 'scapegoat' (lines 29-32). It should be stressed, then, that the lung model provides us with evidence of the fusion of two different religious traditions, one of Mesopotamian divination and the other of West Semitic atonement ceremonies, for which the oldest documentation is now supplied by the texts from Ebla in the 24th century b c e ( X e l l a 1996b). (Translation: W . G . E .

16

DIETRICH -

LORETZ

Watson)

1990a, 25Íf., with a discussion of the various hypotheses.

CHAFFER

EIGHT

THE CORRESPONDENCE OF UGARIT

1

T h e

U g a r i t i c

Jesus-Luis

1.1

L e t t e r s

C u n c h i l l o s

Introductory

T h e origin of letter-writing is most probably to be sought in oral messages transmitted through an intermediary. By the mid-second millennium bce, communication by letter had developed considerably. Likewise, the role of the intermediary between sender and recipient. D e p e n d i n g on the culture, the intermediary would be called mar šiprì in Akkadian, wpwty2 in Egyptian and ml'P in Northwest Semitic. T h e intermediary carried a short written text which served him as a letter of accreditation in the presence of the recipient a n d allowed him to expand the message, replying to any requests for explanation the recipient might pose. 1.1.1

The texts

T h e 86 letters or fragments written in alphabetic cuneiform and in Ugaritic with their various collations are as follows: K T U 2.1 = R S 3.427, K T U 2.2 = R S 3.334, K T U 2.3 = R S 1.013 + 1.043, K T U 2.4 = R S 1.018, K T U 2.5 = R S 1.020, K T U 2.6 = R S 1.021, K T U 2.7 = R S 1.026 + 2.[025], K T U 2.8 = R S 1.032, K T U 2.9 = R S 2. [026], K T U 2.10 = R S 4.475, K T U 2.11 = R S 8.315, K T U 2.12 = R S 9.479a, K T U 2.13 = R S 11.872, K T U 2.14 = R S [Varia 4], K T U 2.15 = R S 15.007, K T U 2.16 = R S 15.008, K T U

1

1956, 68-110; B O T T E R O 1977, 333-4, 233-6 and 211 sub 3; 1964, 212-3; O T T E N 1964, 213-4; L U K E R 1968, 55-77; H O L M E S 1975,

MUNN-RANKIN

EBELING

376-81. 2

S e e VALLOGGIA

3

CUNCHILLOS

1976.

1981a, 1982, 1991.

2.17 = R S 15.098, K T U 2.18 = R S 15.107, K T U 2.20 = R S 15.158, K T U 2.21 = R S 15.174, K T U 2.22 = R S 15.191 [a], K T U 2.23 = R S 16.078+, K T U 2.24 = R S 16.137[bis]+, K T U 2.25 = R S 16.196, K T U 2.26 = R S 16.264, K T U 2.27 = R S 16.378a, K T U 2.28 = R S 16.378b, K T U 2.29 = R S 16.378c, K T U 2.30 = R S 16.379, K T U 2.31 = R S 16.394, K T U 2.32 = R S 16.401, K T U 2.33 = R S 16.402, K T U 2.34 = R S 17.139, K T U 2.35 = R S 17.327, K T U 2.36+ = R S 17.435+, K T U 2.38 = R S 18.031, K T U 2.39 = R S 18.038, K T U 2.40 = R S 18.040, K T U 2.41 = R S 18.075a, K T U 2.42 = R S 18.113A, K T U 2.43 = R S 18.113[b], K T U 2.44 = R S 18.134, K T U 2.45 = R S 18.140, K T U 2.46 = R S 18.147, K T U 2.47 = R S 18.148, K T U 2.48 = R S 18.285[a], K T U 2.49 = R S 18.286+, K T U 2.50 = R S 18.287, K T U 2.51 = R S 18. [312], K T U 2.52 = R S 18.[364], K T U 2.53 = R S 18.[380], K T U 2.54 = R S 18.[386], K T U 2.55 = R S 18.[387], K T U 2.56 = R S 18.[400], K T U 2.57= R S 18.[443], K T U 2.58 = R S 18.[482], K T U 2.59 = R S 18.[500], K T U 2.60 = R S 18.[528], K T U 2.61 = R S 19.011, K T U 2.62 = R S 19.022, K T U 2.63 = R S 19.029, K T U 2.64 = R S 19.102, K T U 2.65 = R S 19.158b, K T U 2.66 = R S 19.181a, K T U 2.67 = R S 19.181b, K T U 2.68 = R S 20.199, K T U 2.69 = R S 24.660G, K T U 2.70 = R S 29.093, K T U 2.71 = R S 29.095, K T U 2.72 = R S 34.124, K T U 2.75 = R S 34.148, K T U 2.76 = R S 34.356, K T U 2.77 = R I H 7 7 / 0 1 , K T U 2.78 = R I H 7 7 / 2 1a, K T U 2.79 = R I H 7 7 / 2 5 , K T U 2.80 = R I H 7 8 / 2 1 , K T U 2.81 = R I H 7 8 / 3 + 30, K T U 2.82 = R I H 7 8 / 1 2 , K T U 2.83 = R I H 7 8 / 2 5 , K T U 5.9 = R S 16.265, K T U 5.10 = R S 17.063, K T U 5.11 = R S 17.117. 4 T o date we know of 20 other unpublished letters. 5

1.1.2

Structure

Usually, a letter comprises heading, m a i n message a n d ending. T h e study of letters written in Ugaritic has not progressed m u c h since C u n c h i l l o s 1989a; see there for lengthier discussion of matters raised here.

4 C U N C H I L L O S - V I T A 1993a, 239-306, 8 6 3 - 4 ; the texts, in corrected form are available in Sapānu, Publicaciones en Internet, at http://wwvv. labherm.filol.csic.es. 5 See M ALB R A N - L A B A T 1995a, 103 n. 2.

1.2

Heading

T h e heading, also called the introduction, comprises the address, proskynesis, greetings a n d wishes. 1.2.1

The address can take on various forms:

(a) / recipient rgm + thm sender (b) thm sender + I recipient rgm T h e n a m e of the more important person preceded the name of someone of lower rank, unless out of politeness the n a m e of the recipient had to come first ( L i v e r a n i 1979a, 1328). In other words, the second formula shows that the sender has higher rank than the recipient, whereas the first formula shows the higher rank of the recipient or else is an indication of politeness between persons of the same rank. If this is the case, Ugaritic usage would be the same as in other regions ( L i v e r a n i 1979a, 1328 with bibliography). However, not all scholars share this opinion. For some, hierarchy is the only criterion that explains the precedence of the n a m e in both cases ( K n u t s o n 1975, 199; K r i s t e n s e n 1977, 144-5; C a q u o t 1979b, 1414). Ugaritic usage would then differ from that of other areas, except for El A m a r n a ( K n u t s o n 1975, 199). As we shall see later on (§ 8.1.2.2), proskynesis occurs only when the first formula is used. This is the one, then, that is used when an inferior addresses a superior. T h e personal names of both the sender and the recipient are quite often replaced by terms denoting relationship. For example, adty, 'My L a d y ' , occurs in K T U 2.12 = R S 9 . 4 7 9 a . 2 ; K T U 2.24 = R S 16.137[bis]+:2; K T U 2.33 = R S 16.402:1; K T U 2.56 = R S 18.[400]:1; K T U 2.68 = R S 20.199:1; K T U 2.82 = R I H 78/12:2; adtny 'our Lady', K T U 2.11 = R S 8.315:1, adny 'my lord', K T U 2.64 = R S 19.102:2; b'ly 'my lord', 'my master', K T U 2.40 = R S 18.040:1; K T U 2.42 = R S 18.113 a: 1 ; K T U 2.61 = R S 19.011:2; K T U 2.63 = R S 19.029:1; K T U 2.64 = R S 19.102:10; bclh 'his master', K T U 2.47 = R S 18.148:2; b'lny 'our master', K T U 2.70 = R S 29.093:1; b'lkm 'your (pi.) master', K T U 5.10 = R S 17.063:3; umy 'my mother', K T U 2.11 = R S 8.315:1; K T U 2.13 = R S 11.872:2; K T U 2.16 = R S 15.008:2; K T U 2.30 = R S 16.379:1; K T U 2.34 = R S 17.139:2; K T U 5.10 = R S 17.063:3; K T U 2.82 = R I H 78/12:1; buy 'my son', K T U 2.14 = R S [Varia 4]:3; φ 'my brother', K T U 2.14 = R S

[Varia 4]:3; K T U 2.38 = R S 18.031:2; ihy 'my brother', K T U 2.44 = R S 18.134:2; ahty 'my sister', K T U 2.21= R S 15.174:3; ahth 'his sister', K T U 5.10 = R S 17.063:1; 'bdk 'your servant(s)', K T U 2.11 = R S 8.315:4; K T U 2.12= R S 9.479a.5; K T U 2.24 = R S 16.137[to]+:4; K T U 2.33 = RS 16.402.2; K T U 2.40 = R S 18.040.4; K T U 2.42 = R S 18.1 13a:3; K T U 2.64 = R S 19.102:12; K T U 2.68 = R S 20.199:3; K T U 2.75 = R S 34.148:4; K T U 2.81 = R I H 7 8 / 0 3 + 30:5), bnk 'your son', K T U 2.13 = R S 11.872:4; K T U 2.30 = RS 16.379:3; K T U 2.64 = R S 19.102:5; ahk 'your brother', K T U 2.38 = RS 18.031:3. T h e terms 'father', 'mother', 'brother' and 'son' do not have the literal meaning of blood-relationship a m o n g those concerned. T h e y are terms of respect. T h e king is called 'father' and the queen, 'mother' (see C u n c h i l l o s 1989a, 247-8). T h e noun thm, 'message', differentiates the Ugaritic formula from contemporary Akkadian formulae (El A m a r n a and even Ugarit itself) which begin with umma, 'thus'. 6 Sometimes an unbroken line separates the address from what follows (see C u n c h i l l o s 1989a, 249). N o account is taken of this line by epigraphists, except for TU. T h e unbroken line is one of the punctuation marks of Ugaritic. It corresponds to our full stop, new paragraph (see C u n c h i l l o s 1989a, 248-9) except when it is used as a writing guide. In that case it recurs regularly in all or nearly all the lines of the text. I.2.2 The proskynesis is a formula of prostration and respect towards the recipient. It is used only with the first type of address formula as given above. T h e r e are four different types. T h e simplest is / p'n Ν qlt, 'at the feet of Ν (the recipient) Í fall' ( K T U 2.13 = R S I I . 8 7 2 : 5 - 6 ; K T U 2.30 = R S 16.379:4-5; K T U 2.64 = R S 19.102 obverse:6-7; K T U 2.79 = R I H 7 8 / 3 + 30:5-6; K T U 2.80 = R I H 7 8 / 1 2 : 2 - 3 ; see C u n c h i l l o s 1989a, 249, n. 26). Somewhat more

6

It occurs in K T U 2.4 = RS 1.018; K T U 2.10 = RS 4.475; K T U 2.11 = RS 8.315; K T U 2.12 = RS 9.479A; K T U 2.13 = RS 11.872; K T U 2.16 = RS 15.008; K T U 2.21 = RS 15.174; K T U 2.24 = RS 16.137[to]+; K T U 2.26 = RS 16.264; K T U 2.30 = RS 16.379; K T U 2.33 = RS 16.402; K T U 2.34 = RS 17.139; K T U 2.39 = RS 18.038; K T U 2.40 = RS 18.040; K T U 2.42 = RS 18.113A; K T U 2.44 = RS 18.134; K T U 2.46 = RS 18.147; K T U 2.49 = RS 18.286+; K T U 2.61 = R S 19.011; K T U 2.63 = RS 19.029; K T U 2.64 = RS 19.102; K T U 2.71 = RS 29.095; K T U 2.76 = RS 34.356; K T U 2.78 = RIH 7 7 / 2 1a.

distant is I pcn Ν mrhqtm qlt 'at the feet of Ν I / w e fall from afar' K T U 2.11 = RS 8.315:5-7; K T U 2.33 = R S 16.402:3-4; K T U 2.45 = RS 18.140:11-2 (see C u n c h i l l o s 1989a, 249, n. 27). More solemn is / pen Ν tnid sbcd mrhqtm qlny 'at the feet of Ν twice seven (times) I / w e fall' K T U 2.64 = R S 19.102 obv. 13-16; K T U 2.70 = R S 29.093:8-10. O r also I p'n Ν šbCd wsb'id mrhqtm qlt 'at the feet of Ν seven and seven (times) from a distance I fall' K T U 2.12 = RS 9.479a:6-11; K T U 2.24 = RS 16.137[Äw]+:5-7; K T U 2.40 = RS 18.040:5-8; K T U 2.42 = R S 18.1 13a:4-5; K T U 2.51 = R S 18.[312]:2-3; K T U 2.68 = R S 20.199:4-7 (see C u n c h i l l o s 1989a, 249, n. 29). T h e two last expressions could have the same meaning although it is not completely certain and they could have different nuances (see C u n c h i l l o s 1989a, 250). Ί fall' is a translation of qlt from Vqyl, cognate with Akkadian qâlu (see C u n c h i l l o s 1989a, 250). 'From afar' is a translation of mrhqtm or mrhqm, an adverb derived from Vrhq. 1.2.3 Greetings. T h e sender usually greets the recipient with a jussive: yšlm I- 'Peace to N ! (see C u n c h i l l o s 1989a, 251 and n. 3 6 ) , which occurs in several letters.' Its variants a r c y š l m ly8 and InyšlmP T h e nominal use of šlm cannot be excluded in some cases. 10 1.2.4 Wishes. T h e greeting can be followed by a wish with a religious content as expressed by the sender in favour of the recipient. W h e n both elements are present, the action seems to be progressive. Indeed, with the greeting the sender wishes the recipient to enjoy good health when receiving the letter, while with the wishes he expresses the hope that the gods may continue preserving the recipient's good health. T h e same progression is glimpsed in the body of the letter. T h e greeting, then, can be differentiated from the wishes or

7

K T U 2.1 = RS 3.427; K T U 2.4 = RS 1.018; K T U 2.6 = RS 1.021; K T U 2.10 = RS 4.475; K T U 2.13 = RS 11.872; K T U 2.16 = RS 15.008; K T U 2.21 = RS 15.174; K T U 2.30 = RS 16.379; K T U 2.33 = RS 16.402; K T U 2.34 = RS 17.139; K T U 2.38 = RS 18.031; K T U 2.41 = RS 18.075; K T U 2.44 = RS 18.134; K T U 2.46 = RS 18.147; K T U 2.52 = RS 18.[364]; K T U 2.63 = RS 19.029; K T U 2.67 = RS 19.181B; K T U 2.71 = RS 29.095; K T U 2.81 = RIH 7 8 / 3 + 30. 8 K T U 2.30 = RS 16.379; K T U 2.33 = RS 16.402; K T U 2.68 = RS 20.199; K T U 2.72 = RS 34.124. See C U N C H I L L O S 1989a, 252. 9 K T U 2.79 = R I H 7 8 / 3 + 3 0 . 6 , see C U N C H I L L O S 1989a, 2 5 2 ~ 3 . 10 K T U 2.70 = RS 29.093; K T U 2.72= RS 34.124; K T U 5.10 = RS 17.063.

expression of wish/desire 1 1 although there are different opinions (see C U N C H I L L O S 1989a, 2 5 3 - 4 ) . T h e religious expression normally used in the wishes (see C U N C H I L L O S 1989a, 2 2 9 - 3 4 ) is ilm tgrk tslmk, ' M a y the gods protect you and keep you healthy!'.1'-' Ugaritic uses two verbs because two actions are meant to be represented (see C U N C H I L L O S 1989a, 254). T h e verb tgr, ΛIngr, denotes the action of 'protecting'. 1 3 Theoretically, tšlm could be the šaphel causative or the intensive D stem. However, both Akkadian and H e b r e w normally use the verb Sim in the intensive stem. Ugaritic probably does the same. 1 4 T h u s , the expression manifests the wish that the gods continue preserving the health of the recipient. T h e formulation of the wishes or desires can change. T h e letter writer has freedom of expression which is evident, for example, in K T U 2.16 = R S 15.008:4-6, when he writes ily ugrt tgrk tslmk, ' M a y the gods of Ugarit protect you a n d keep you healthy!', an expression which has its equivalent in the Akkadian of Ugarit: ilānu ugarit ana šulmāni lissurūka or else ilānu sa ugarit ana šulmāni lissurūka (see C U N C H I L L O S 1989a, 256). T h e a u t h o r of the letter K T U 2.4 = R S 1.018 sends wishes to the High Priest ilm tslmk tgrk t'zzk, ' M a y the gods keep you healthy, protect you and keep you hardy!' (lines 4 - 6 ) . A longer version occurs in K T U 5.9 = R S 16.265:2-6: ilm tgrk tslmk t'zzk alp ymm w rbt šnt b'd clm, ' M a y the gods protect you, keep you healthy, keep you hardy for a thousand days a n d ten thousand years for ever!'. Finally, there is the formula b'lysul šlmk, ' M a y Ba'al take care of your health!' of K T U 5.11 = R S 17.117:2, which becomes intelligible if c o m p a r e d with the H e b r e w expression THWHys'l Islmk, of Arad Letter 18 a n d the Akkadian expressions ilānu šulumka . . . lisal of EA 9 6 : 4 - 6 , ilānu lišalū šulumka of T a a n a k 1.5-6 and Aramaic clhyc yšlw šlmk of A P 56:1 (see C U N C H I L L O S 1983a).

11

Further details and discussion in C U N C H I L L O S 1989a, 2 5 1 - 2 . With some variations it occurs in K T U 2.1 = RS 3.427:1-2; K T U 2.4 = RS 1.018:4-5 (ilm tslmk tgrk)· K T U 2.6 = RS 1.021:5-6 (ilm tgrkm ts'lmkm)·, K T U 2.11 = RS 8.315:7-9; K T U 2.13 = RS 1 1.872:7-8; K T U 2.14 = RS [Varia 4]:4-5; K T U 2.21 = RS 15.174:5-6 (ilm tslmk tgrk)] K T U 2.30 = RS 16.379:6-7; K T U 2.34 = RS 17.139:3-4; K T U 2.38 = RS 18.031:4-5; K T U 2.41 = RS 18.075:1-2; K T U 2.44 = RS 18.134:4-5; K T U 2.46 = RS 18.147:4-5; K T U 2.63 = RS 19.029:5-6; K T U 2.68 = RS 20.199:9-10 (collation: Pardee); K T U 2.70 = RS 29.093:6-7; K T U 2.71 = RS 29.095:4-5 (ilm tslmk tgrk); K T U 2.75 = RS 34.148:6. 13 Others consider the root to be gyr; see C U N C H I L L O S 1989a, 2 5 5 . 14 See C U N C H I L L O S 1989a, 254-5; T R O P P E R 1990a, 164. 12

1.3

Body of the letter or message

1.3.1 Structure of the message. T h e second part of the letter is laid out in dialogue form, marked in some cases 15 by the adverbs hnny, 'here' and tmny, 'there' (further details in C u n c h i l l o s 1989a, 257, n. 66). Generally (see C u n c h i l l o s 1989a, 258, n. 68) the sender uses hnny,lfi or its equivalent hlny17 to open the section where he provides news of his circumstances, limited to the expression 'everything is (very) fine with 18 (me/us)'— ( mn(y) kll (mid) šlm in various forms (see C u n c h i l l o s 1989a, 259, n. 70). T h e term tmny is used to begin the expression (see C u n c h i l l o s 1989a, 259, n. 73) by the sender that all may go well with the recipient cm Ν mnm šlm. T h e two expressions introduced by hnny and tmny in the body of the letter are the counterpart of the greeting and wishes in the structure of the text. T h e greeting, like the expression introduced by hnny, denotes the situation as at present, whereas the wishes, like the expression introduced by tmny, are set in the future. In both cases there is an element of progression. In cm Ν mnm šlm, the verbal form Sim is probably an optative qatala, and in cmn(y) kll (mid) šlm the qatala is in the present. 19 1.3.2 Syntax. Within the dialogue section, the key to the wording and of course to understanding is syntax. Unfortunately, this is the least known part of Ugaritic grammar. T h e research undertaken by us in Madrid has not yet provided results. T h e key to syntactic structure frequently lies in particles which are still largely not understood, although their mysteries have been penetrated to some extent. Thus,

15 In K T U 2.1 = RS 3.427; K T U 2.11 = RS 8.315; K T U 2.13 = RS 11.872; K T U 2.21 = RS 15.174; K T U 2.24 = RS 16.137[£w]+; K T U 2.30 = RS 16.379; K T U 2.34 = RS 17.139; K T U 2.36+ = RS 17.435+; K T U 2.38 = RS 18.031; K T U 2.46 = RS 18.147; K T U 2.56 = RS 18.[4001; K T U 2.65 = RS 19.1 58B; K T U 2.67 = RS 19.181B; K T U 2.68 = RS 20.199; K T U 2.70 = RS 29.093; K T U 2.71 = RS 29.095; K T U 2.72 = RS 34.124; K T U 2.76 = RIH 7 7 / 2 1a; K T U 2.83 = R I H 78/25. 16 K T U 2.11 = RS 8.315:10; K T U 2.38 = RS 18.031:6; K T U 2.46 = RS 18.147:6; K T U 2.56 = RS 18.[400]:4; K T U 2.65 = RS 19.158 b:1; K T U 2.71 = RS 29.095:5; K T U 2.72 = RS 34.124:4. 17 K T U 2.1 = RS 3.427:3; K T U 2.13 = RS 11.872:9; K T U 2.21 = RS 15.174:7; K T U 2.24 = R S 16.137 [Aw] + :8; K T U 2.30 = RS 16.379:8; K T U 2.36+ = RS 17.435+:2; K T U 2.47 = RS 19.181 B:4; K T U 2.70 = RS 29.093:11; K T U 2.76 = RIH 77/21a:4; K T U 2.83 = RIH 78/25:2. 18 See C U N C H I L L O S 1989a, 258, n. 69. 19

See

CUNCHILLOS

1986,

259

60; 260,

n.

75.

ht, usually translated by 'now' 2 0 has the peculiarity of always marking the passage to the second part of the discourse 21 or argumentation. It means 'also', in the first case and 'but' 2 2 when followed by the conditional particle hm, 'if'. O t h e r particles, such as w which begins an apodosis, also contribute to the recognition of structural elements. 23 Ugaritic w, as in Hebrew, can also denote the beginning of a conditional clause or a subordinate conditional clause. 24 In this way, step by step the syntax is revealed, but the picture is still incomplete. If we consider the letter as a whole and not just the body of the message, we can add that epistolary style may be marked by the use of a large n u m b e r of jussive, imperative and optative forms. T h e third person jussive is used frequentiy, for example in yslm Ik in the greeting and ilm tgrk tslmk in the wishes, but also in the body of the message, for instance in tmny mnm šlm and sometimes throughout the whole letter as in K T U 2.16 = RS 15.008. Perhaps this is one of the characteristics of epistolary style,25 which is due to the involvement of a third person in the communication, an intermediary who reads the written message aloud. Scholars who do not accept the role of the mlak have great difficulty in understanding the use of the third person. 2 6 It cannot be excluded that the use of the third person is an expression of respect. 1.3.3 Literary aspects of the letters. T h e letters are the work of various authors preserved in a single copy, as short unvocalized texts full of hapax legomena which allude to contexts which are unknown to us. These are the historical contexts in which the letters were written, the archaeological contexts in which they were found. We also lack the essential communication link of the 2nd millennium BCE, the maVak who held the key to interpretation, the additional explanations d e m a n d e d by the text so as to be fully understood by the recipient. So much for the inherent difficulties of the letters. However, it

20

See C U N C H I L L O S 1989a, 262, n. 77; DLU, 169-70. For example in K T U 2.4 = RS 1.018 and K T U 2.14 = RS [Varia 4]. See C U N C H I L L O S 1989a, 273 n. 6 and 295 n. 12 respectively. 22 See C U N C H I L L O S 1989a, 337, n. 36; 278, n. 10. 23 C U N C H I L L O S 1989b, commentary on the letters K T U 2.10 = R S 4.475; K T U 2.30 = RS 16.379; K T U 2.33 = RS 16.402. 24 See C U N C H I L L O S 1989a, 338 n. 4 0 . 21

25

PARDEE -

26

KRISTENSEN

WHITING 1977,

1987. 145

in respect of

K T U

2.47

=

RS

18.148.

has to be accepted that on the other h a n d it presents several advantages c o m p a r e d to other literary genres. T h e letter has spatial boundaries which cannot be said of more complex literary texts such as myths and legends. This factor assures us b e f o r e h a n d that if the tablet is intact, the m e a n i n g of the letter will be complete. It has been claimed that the correspondence in Ugaritic is only a translation f r o m Akkadian texts. T h e arguments for and against have been set out elsewhere. 2 7 T h e r e is a strong possibility that Ugaritic correspondence was original.

1.4

Closing formula

T h e body of the message is followed by a request for information concerning the recipient and his surroundings. T h e formula is usually rgm ttb I-, 'Send a reply to (lit. return a word to)!'. T h e verbal form ttb is a Shaphel imperative of the verb twb, with assimilation of the p h o n e m e / š a / typical of the Shaphel to / t a / . W h e n the letter is written in the third person, tttb is used in the closing formula, 3rd pers. fem. sing, jussive: ' M a y she return a word to her servant!'. This occurs in K T U 2.12 = R S 9.479A: 1428 and K T U 2.16 = R S 15.008:19. 29

1.5

Conclusion

Historical and social importance of the correspondence. T h e sender of most of the letters preserved is a k i n g / q u e e n or a prince/princess. Some seem to come f r o m lesser court officials. O f all the letters preserved, the one by queen Pudugibat or Puduhepa ( K T U 2.36 = R S 17.435+), is undoubtedly the most interesting for its historical significance, but also for the information it provides on diplomatic relations and even on the political tactics used by Niqmaddu, king of Ugarit. P u d u h e p a was the wife of Hattusilis, the m o t h e r of Tudhaliya and a c o n t e m p o r a r y of Ramesses II. T h e kingdom of Ugarit was a vassal of Hatti, and the king of Ugarit was obliged to present himself each year in the Hittite court, bringing a tribute established beforehand. As the letter shows, the king

27 28 29

See See See

CUNCHILLOS CUNCHILLOS CUNCHILLOS

1989a, 264-7. 1989a, 286 n. 6. 1989a, 3 0 2 n. 2 5 .

of Ugarit had to bring a quantity of gold for the House of the king, but the same a m o u n t also for the House of the queen. N i q m a d d u sent the required a m o u n t of gold that was due to the king by adopting the ruse of sending it to the House of the queen. Similarly, Niqmaddu, king of Ugarit, had to send a certain quantity of precious stones which had been worked in his kingdom to the Hittite court. N i q m a d d u pretended not to have any gem cutters. In the rest of the letter, he asks the Hittite court for help in assuring that the caravans going to Egypt would pass through Ugarit, no doubt thinking of the benefits such commercial traffic would entail. In another letter, ( K T U 2.38 = R S 18.031), the king of Tyre informs the king of Ugarit that the fleet he had sent to Egypt found itself in the middle of a great storm which forced the fleet to split, half going to Tyre and the other half to Acre. T h e fleet was carrying wheat. T h e captain m a d e them take the wheat out of the amphoras and lay it out to dry, but first he delivered it to the king of Tyre, the one responsible on land. T h e king of Tyre informed the king of Ugarit that the cargo had again been loaded on the boats and that the fleet had resumed its journey. Another letter, ( K T U 2.68 = RS 20.199), by Urfctešub, seems likely to deal with the old Hittite king who had fallen into disgrace and had had more than friendly relations with Ugarit. Each of the letters is a remnant of someone's life and reflects the society to which he belongs.

1.5

Appendix: the vocabulary of the letters

Α . - T h e total n u m b e r of Words with their Morphological Display (or WMD)30 in the Ugaritic corpus published so far is 6521. 31 T h e n u m b e r of WMD exclusive to a single literary genre is as follows: Administration: 2031; Myth: 1164; Epic: 495; Ritual: 438; Correspondence: 332; Assorted Fragments: 81; School Exercises: 44; Law: 44; Inscriptions: 40; Hippiatric texts: 37. 876 of the 6521 WMD contained in the published Ugaritic texts occur in the Correspondence. Of these, 544 are also c o m m o n to other genres, leaving 332 as exclusive to the Correspondence.

30 Palabras en Alorfologia Desplegada or PAID in Spanish. Here the English abbreviation is used for convenience. See § 16.2.2. 31 C U N C H I L L O S - V I T A 1993a, 1-2191.

B . - T h e Vocabulary of the C o r r e s p o n d e n c e shared with other literary genres: 1.- C o r r e s p o n d e n c e and Administration: adc; adty, ahty, alyyn; anyt; argmnm; atlg·, ibm\ iwrdn; iwdr, iwrpzn; ihy; lytlrrv, ilšpš; iltkm; itn; ittl; ubr'y; ully; urgttb; bdhm; blym\ bcly\ bclyskn\ b'rm; gzl; gib; grgs; dym; dmn; drdm, drm; hyil; hpr.; hlpn·, hpn\ hpnm; ybnn; ymz', kbr.; ky, kkr, klby; ksphm; ktt; Iby; Iqht; mz'/m; wz^wA; mkr, mlbš] mly; mlkytn; mrdt, sny; cbdbcl; cbdyrh; cbdm; c yn; 'ky, 7; 7r, 7«; cttry, plsy, pm\ psn; spy, rhbrv, škn; šlmy; šml·, sm't; s't, stn; tyt; tlmyn; tmtt; fry; trgds; trtn; tby; tpbtcl\ tpn. 2 . - Correspondence and Myth: ad\ adnh; ahš; ams\ arš; ilak; iph\ irstk; bnk\ dbr, dtm; hdd; hdt\ hin·, hmk\ hš; hsri\ y bit, ycmsn\ klm\ kpr, lah, lit, lp\ mla; mlak; mgt; mrt, mtt; n'mm; sip", cbdk\ cnn\ pcnk; slm; qbt, qdt, qlh\ qlt; qmh; rgmt, šbCd, šdt; škb; šntm\ špr, thmk) tm; t'tqn; tgrk; tttb; tbh\ tmt. 3 . - C o r r e s p o n d e n c e a n d Epic: adnk; ahtk; amrk; argm\ aškn; aštn; ikm; itt; bClny; d'\ dt\ hndt, ytbC mndC\ Crym\ tadnv, tdbr, tmgy; ta; tqb; tnnth. 4 . - C o r r e s p o n d e n c e a n d Ritual: ατηπ, aspt\ dbhn\ dr'·, hwth] hrdh\ htm] y dbh; ysal; mzn\ mit.; msqt, mrhqm·, ndr, si. 5 . - C o r r e s p o n d e n c e and Law: iky, brt; mgsh, cmnk. 6 . - C o r r e s p o n d e n c e and Assorted Fragments: hdr·, mli; tzn; tsc. 7 . - C o r r e s p o n d e n c e and School Exercises: h; t; mtr. 8.— C o r r e s p o n d e n c e a n d Vocabularies: ikt. 9 . - C o r r e s p o n d e n c e and the Hippiatric Texts. C . - Vocabulary exclusive to the C o r r e s p o n d e n c e , as has been seen, comprises 332 WMD. A preliminary analysis results in the following: Verb forms: Λlabd 'to be eternal', 'to perish': tubd. Vahb 'to love': ihbt.

\lawd 'to weigh, pay': udh\ tud. ^1awš 'to make a gift, give': nul. ^azr 'to gird': yšizr. ^lahd 'to grasp, seize': did] didn. •Jahr 'to delay': yšihr; tšiffrhm.

^aty 'to come': nitk. ^arš 'to ask, desire': aršt; yirš.

^lbll 'to moisten, ferment': iblblhm. V67 'to work, make': ysb'l·, ib'ltn. •Jb'r 'to burn', 'to abandon, destroy': ibcr; b'r·, yb'm. ylbqt ' to seek': bqt\ ybqj.

yIbtt 'to move away', S 'leave aside': ybt\ ušbt\ ušbtm. ^lgd' 'to cut': agdc.

ΛIgrr 'to move': igr.

Adhl 'to fear, be scared': tdhl. ΛIdhs 'to be excited/nervous': tdhs. \Ihbt 'to humble, demolish': hbt; yhbt\ thbt. 1 ihdy 'lacerate, gash oneself': hdy. ^lwhl 'to be discouraged': twhln. "Jwzn 'to weigh': yzn. •Jwpt 'to spit': pt?2 Vzwd 'to provision, supply', Nifal 'to be provisioned, supplied': nzdt. Ahdy 'to see, look', 'rejoice': hd. λIhwy 'to prostrate oneself, 'to live': hw, hytn. "Jhlq 'to perish', 'to be absent': hlqt. yhpn 'to cover, protect, wrap': yhpn. 'to snatch, conquer, defeat': hti; nhtw, f}tat. •Jycf 'to know': yd'm; Idc.33 ^y'b 'to be spacious': y'b. Λlysa 'to go out': assu; yssa; šsu. ylyra 'to fear, be frightened': yritn. \lyrd 'to descend': yrdnn. yytn 'to give, sell': atnnk) ytnh, ytnnn; štnt; štnth; štntn. ^lytb 'to sit, dwell in': ytbt; ytbtn. *ikwn 'to be establish', Š 'to create, make preparations': tšknn·, tšknnnn\ yškn. ^lkhd 'to erase', 'to hide': khdnn. Mhm 'to eat', 'to fight', 'to be suitable': ilhmn. vlik 'to entrust with, commission, send': ilakh, labrr, likt, lut, tlikr, tlikn. λÍlwy 'to follow, enclose, go round': aslu (?). Mwt 'to mould (clay)': Itt (Gt). ymkr 'to trade': amkrn.

^lmla 'to fill': mlit (part. act. G. fem.pl.). >Imgy 'to arrive, come, reach': ymgyk; mgyy; mgyk\ tmgyy. ^lnad 'to raise': nitk. yinhr 'to cause to flow': /Ar.34 ^nw' 'to move', 'ask, request': n'kn. AIns' 'to uproot': //«. ΛIngr 'to protect': tgrkm; tgm. ΛIntb 'to pull up': ntb. Vsyr 'to travel': syr. yspr 'to write, tell, relate': yspm. Vewr 'to be naked', 'to go blind', 'to worry': t'wr. "J'zz 'to be strong, prevail', D 'to preserve strength': t'zzk. Vemt 'to hit, tie, wrap': cmt. VVy 'to empty, unload', 'to turn': cryt. " Hapax. It comes from a pupil's exercise. Hapax. Text in bad condition, but it is a preposition + inf. const, of yd'. 34 Hapax. Possible but not certain. 33

V'/r 'to invite': t'sr. Vctq 'to pass, move forward', 'to become old': tctq\ s'tq. ^gzr 'to fortify': agzr. ^lphy 'to discern, see, perceive, experience': iphn\ phn'\ phnn. •4pit 'to set free, save': npl. Vsba 'to grow, to rise (of the sun)', 'to decrease, to set (of the sun)': sb\ \Iswq 'to squeeze': šsq 'to press'. AIshq 'to laugh': tsshq. \Ispy 'to cover, put on', 'observe, keep guard': aspy\ tspy. ΛIqwr 'to flow', 'be curved': šqr. ^qyl 'to fall': qlny. ^lqšr 'to knot, join': qšr; tqšr. ^Jqty 'to end': tql. ylrhl 'to worry, travel': trhln. ^lriš 'to swell up, have a headache': rus. vrws 'to be poor': yrs. TO 'to destroy': yrs. všay 'to wait, lie in wait': asi. ^šal 'to deal with, watch over/supervise': yšul. ^Išil 'to question, ask, request': šil; silt·, išai, išalhm·, tšal; yštal; yššil. ^lšivr 'to besiege, enclose': yšm. •Jšyt 'to place, set': ištn. 1 1škh 'to meet/find': tškh; nškh. yšlh 'to send', 'to found, build': nšlh. y Ilm 'to be well, be in peace': šlmt; šlmtn\ tšlmk; tšlmkm. ^lšmh 'to rejoice': ššmht. ^lšm' 'to hear': sm'fi; tšrrínv, mšrrít. ^lšqp 'to strike, break': yšqp. 4šty 'to drink, taste': ištn. yltbr 'to break, smash': ttbrn. ^ltkp 'to be demolished, resist, repel': ntkp. Vt'y 'to correct, inspect', 'to offer': tt'y. Nouns: adn 'Lord, father, master, foreman': adny; adnty. idn 'permission, consent': idn. adt 'Lady': adtny. az 'fringed carpet': azt. amt 'maid, maidservant, slave; forearm, elbow': amtk. any 'fleet': anyk\ anykn. argmn: 'tribute' argmny. ins 'people': inšk. udr 'nobility, the most noble'; 'camels'; 'slope, skirt (of a mountain)': udrh\ udrk. udr 'courier, messenger'; 'salvation': udrh. u/} 'brother': uhy. un 'sorcery, magic': unk.

urk 'length': urk. bnny: bnny (or preposition bn 'between' + suff. -ny 'us'). bs 'linen': bs. g 'voice': gy. gšm 'rain': gšm. dbr 'plague, pestilence; thing, word'; 'pastureland, pastures, hut': dbrm. dn 'receptacle, amphora': dnh\ dntm. dr' 'arm'; 'wheat seed': dr'hm. hwt 'life'; 'house, dynasty, country, territory, district, region'; 'animal, serpent': hwtk, hwtm. hmy 'wall'; hmy. hbt 'freeman, fugitive'; 'soldier, proletarian': /)btm. hrd 'guard, sentinel'; 'part of the conscript army': hrdk. kw 'liquid measure': kw. ksu 'throne': ksiy. ksp 'silver': kspym 'silversmiths'. lb 'heart': Ibk. lbs 'garment': Ibsk. Ig 'gallon': Igk.35 Idt 'receptacle, cellar': Idtk. Ih 'cheek, jaw, snout, fauces'; 'tablet, writing, dictation, message': Ihy. mid 'abundance': midy. md 'surveyor': mdym. mhr 'surveyor'; 'dung'; 'courtier'; 'coat, cloak': mhrk. mkr 'trader': mkrn. mlakt 'mission, assignment, embassy': mlakt] mlaklh\ mlakly; mlaktk. mli 'full, complete' (qatil ν mla 'to fill'): mlit. mlg 'a type of sacrifice': mlgt. msgr 'casde, fortress': msgr. m'n 'reply': m'nk. mrdt 'carpet': mrdtt. mrkbt 'chariot': mrkbtk. mtn 'repetition, reply': mtnn. n'r 'boy, lad, soldier, knight/horseman, cadet'; 'a type of roasted or dry flour'; 'mix, mixture': n'ry. npš 'throat, cheeks, desire, person, animal, living being': npšy. npšn 'abundance'; interjection: 'by (my) life!'. npt 'type of offering': nptn. nr 'yoke': nrm. nt 'inner part': nt; nty. sglt 'property': sglth. shl 'driller (of precious of semi-precious stones)': shlk; sf}lm. 'ps 'frontier': 'ps. 't 'moth, louse': 'tty.

Hapax. T h e text is very corrupt.

gb 'sacrificial pit': gbny. gl 'depression, hollow', 'reedbed': glhm. qnu 'lapis lazuli': qnim; qnuym. qrt 'city': qrtn. rgm 'word, speech, matter, claim, complaint': rgmh; rgmy. rc 'shepherd', 'friend, companion': r'. M 'field': sdk. sin 'gift, peace offering': sink. šlm 'peace, appeasement, peace offering'; 'satisfied'; 'victim of communion sacrifice': šlmk. šm 'name': šmy. ty 'water, river': ty; th; 0>ny.'iR thm 'errand, resolution, decree, message': thmhy. Personal names: agzr] ahtmlk; ibrkd; iwrphn; ihqm; illdr; irrtrm; gnryn; ddyn: yrmhd: ytrhd; nmy\ nmry; 'bdmlk; pgn; pdgb; pzy\ pzy, plz', pnht\ pgsdb\ ptmy\ ttyy. Divine names: atrty, 'Athirat'. Toponyms: grgms 'Carchemish'; Iwsnd 'Lawasantiya'; mgšh 'Mukiš'; ngt 'Nuhašše'; gblm 'Byblos' (unless it is a gentilic: 'Byblians'); ns. Gentilics: ugrtym 'Ugaritians'; šmny: smnyk. Names of professions: kspym 'silversmiths'; mdym 'surveyors'. Numbers: snp 'two-thirds'. Personal pronouns: ankn T . Demonstrative pronouns: hndn\ hnt, hnkt. Indefinite pronouns: mhy; mhk\ mhkm\ klklhm\ kllh. Adverbs: ikmy; inm\ innm; hnny; midm; mrhqt; mrhqtm; sb'id 'seven times'; tmny; tnid 'twice'; 'šrid 'ten times'. Prepositions: 'mrv. 'mny; 'mnkm; 'Un 'against'; gmt 'according to, in accordance with'; bn 'between': bnny (with suff. -ny 'us'). Conjunctions: uy; uk; uky. Excluded from the above classification are: 1. T h e WMD which are susceptible to different analyses. T h e list is as follows: akin, ally, ankm, ptn, tittm, tbt, tnlh, ttk. 2. Also excluded from the previous grammatical classification are the WMD for words which as yet remain unexplained. T h e list is as follows: ahnnr; itrhw; ul'nk; us'nk; bas; bby; bnptn; hzb; hhb; hkm; htm; ysin; ysunn; stn; t'pr; pkdy; prhn; tlhmy.

36

Hapax. It occurs in K T U 5.11 = RS 17.117.

T h e next step is to determine whether other noun or verb forms, i.e. the same noun with different suffixes or other forms from the same root are attested in other literary genres besides Correspondence. T h i s a n d other supplementary information can be found in the Concordance of Ugaritic Roots, in preparation (see § 16.2). (Translation:

W.G.E.

WATSON)

2

T h e

Akkadian

John

2.1

Letters

Huehnergard

Introduction

Some 354 letters or fragments of letters have been found at Ras Shamra. O f these, 134, or just over one-third of the total, were discovered during the 1994 season of excavations in the 'House of Urtenu', and have not been published as of this writing. 1 Of the 220 letters or letter fragments found before 1994, 24 remain unpublished; thus, the total n u m b e r of letters available for study is 196. While many of the published Akkadian letters were written within the kingdom of Ugarit, many others were sent to Ugarit from cities and states across the Near East. Perhaps as m a n y as 43 of the letters may be said with some degree of confidence to have originated within Ugarit itself, or to be copies of letters sent from the Ugaritian court. But about twice that number, roughly 85 of the letters, give some indication of their point of origin outside Ugarit, and a m o n g these at least 17 places of origin are attested: A1ašiya, Amqu, Amurru, Assyria, Aštate, Beirut, Carchemish, Egypt, Hatti, Mari, Mukiš, Parga, Qadesh, Sidon, Siyannu (and Ušnatu), Tarhudašše, and Tyre (perhaps also Alalah, E m a r and M a ' h a z , as well as a second town n a m e d Beirut/Bi'rut; see the listing at the end of this chapter). It is not surprising, perhaps, that the largest numbers of letters come from Ugarit's overlord, Hatti, and Hatti's provincial capital, Carchemish. O f course, there are a great many letters between individuals, in which the sender did not specify his or her location at the time of writing, or in which in any case such information can not be recovered.

2.2

Format and Formulae

Most of the Akkadian letters exhibit a c o m m o n format (see A h l 1 9 7 3 ; Y a m a d a 1 9 9 2 ) . T h e y begin with an address, which gives the identities of the sender and the addressee, in the form of a c o m m a n d to the scribe/messenger: in its simplest form, 'Say to [the addressee];

1

See

M A L B R A N - L A B A T 1995; L A C K E N B A C H E R 1995a; and especially 1995b for overviews of the 1994 Akkadian tablets.

BORDREUIL -

MALBRAN-LABAT

thus/word-of [the sender]' or the inverse, 'Thus/word-of [the sender]; say to [the addressee]'. Normally the party that appears first in the address is the one with a higher social rank ( N o u g a y r o l 1955, 2-3). T h e r e usually follows a salutation, which may contain any or all of the following elements: a statement of obedience if the sender is of lower social rank than the addressee; a wish for the well-being of the addressee; an invocation of divine blessing. A salutation is omitted in letters from the Hittite overlord. Following the salutation there may be a stereotyped statement of the sender's well-being and a polite request for news of the well-being of the addressee. T h e n comes the body of the letter. These various elements—address, salutation, statement/inquiry of well-being, body—are often separated from one another on the tablet by a ruled horizontal line. T h e scribe may also use such a line to mark off separate topics in the body of the letter. T o illustrate the introductory formulae, the following letters may be presented. In the first, R S 20.238 = t/g 5 no. 24, the king of Ugarit writes to a king of superior rank; after the address, there is a statement of obedience (prostration) and a lengthy wish for the well-being of the addressee and his household; then follows the body of the letter. Ana šar mat Alašiya abi-ya qibi-ma; umma šar mat Ugarit māri-kā-ma:

Say to the king of the land of A1ašiya, my father; message of the king of the land of Ugarit, your son:

Ana šēpī abī-ya a[mq]ut. I f[a1]1 at my father's feet. Ana muf}hi abī-ya lū šulmu. Ana bītātī- May my father be well. May your ka hīrātī-ka sābī-ka, ana gabbi mimmû houses, your wives, your army, everyša šar māt Alašiy[a] abī-ya d[a]nniš danništhing of the king of the land of lū šulm[u]. A1ašiy[a], my father, be v[e]ry, very weip]. Abī, anumma eleppētu ša nakrī Ulaka; ālānī-ya ina išāti išairip, u amāt [Í]ā banīta \in\ a libbi māti īte[p]šū. Abūya ul ī[d]e kī gabbu sābī . . . -ya ina māt Hatti ašbū u gabbu e[lepp]ētū-[y]a ina mât Lukkâ ašbu? \Ad~\īni ul ikšudan-ni. U mātu kâm-ma nadât. Abū-ya amāt annīta lū īde. Inanna 7 ekppētu ša nakrī ša illakan-ni, u amāt mašikta ītepšū-nâši. Inanna šumma elepp[ētu] ša nakrī šanâtu

My father, now then, the enemies' ships have been coming; they have been setting fire to my towns, and so they (the enemies) have do[n]e something [u]np1easant [i]n the land. Does my father not know that all my . . .—troops are situated in Hatti and all [m]y s [hip] s are situated in Lycia? They have not [ye]t reached me. The land could be overthrown

ibašši-mi, tē[m]a \ayy\akâmma šupr[a]nni, u lü īde.

this way. M y father should be aware of this. N o w the enemies' ships that have been coming against me are seven, and they (the enemies) have done something heinous against us. N o w , if there are more of the enemies' ships, send [m]e ne[w]s [so]mehow, that I may know.

In the next letter, R S 19.070 = PRU 4, 294, the king and queen of Ugarit write to a man of lesser rank; after the address there follow a brief expression of well-wishes and an invocation of divine blessing. Message of the king of the land of Umma šar mat Ugarit u šarrat mat Ugarit] Ugarit and the queen of the land of ana Kila'e abī-ni qibī-ma: Ugarit; say to our father Kila'e:

Lu šutmu ana muhhi-ka. Ilānū ana šulmāni lissurū-ka.

M a y you be well. M a y the gods keep you in good health.

W e have sent herewith our mesAnumma Ili-milku mar-šiprī-ni ana senger ΙΐΓ-milku to inquire after the šc?āli Mmi ša šatri bētx-ni naltapar-šu. [. . . welfare of our lord the king. [ . . .

Finally, a letter to the king of Ugarit from one of his agents, R S 17.383 = PRU 4, 22Iff., illustrates the stereotyped statement of the sender's well-being and inquiry concerning that of the addressee; we quote here only the introductory part of the letter: Ana šar mat Ugarit beh-ya qibi-ma; umma Taguhli ardī-kā-ma:

Say to my lord the king of Ugarit; message of your servant Taguhlu:

I fall at my lord's feet from afar Ana šēpī bēlī-ya ištu rūqiš šinîšu sebîšu amqut. twice seven times. N o w then all is well with the king Enūma itti sarri u ittī-ya gabba šulmu. and with me. Is everything well diere Ašranu itti šarri bēlī-ya mīnummê šulmānu? with my lord the king? M a y news Tema litenūni. be sent back to me.

2.3

Topics of the Letters

A wide range of topics is addressed in the Akkadian letters. Most of the letters from Ugarit's overlords, Hatti and Carchemish, concern military and political matters, such as movements of enemy troops; requests for troops, arms, or information; border disputes; the activities of merchants; and other legal matters. Others, however, have to do with the sending of gifts and tribute, or accompany the sending of a high official. Letters from other courts often discuss gift exchange, alliances, and good relations, or present requests for commodities. Letters between individuals concern economic matters, offer or request news, or simply bear a greeting from sender to addressee, with a request for a return letter.

2.4

Grammar

T h e letters found at Ras Shamra, like the other Akkadian texts, are for the most part written in the western peripheral type of Middle Babylonian that is termed Hurro-Akkadian or Syro-Anatolian. This type of Akkadian usually exhibits a certain a m o u n t of confusion in the writing of stops and sibilants; an enclitic particle, -me, which tends to replace normative Akkadian -ma on prepositions, conjunctions, adverbs, and pronouns (e.g., in the conjunction kīmê instead of kīma; the adverb anummê instead of anumma\ and the indefinite pronoun mannummê instead of mannumma); the frequent appearance of the conjunction u to introduce main clauses after protases of conditional sentences, after relative clauses and other subordinate clauses, and after instances of casus pendens. Peripheral Akkadian texts also tend to exhibit an admixture of several core Mesopotamian dialects into the Middle Babylonian matrix, such as Old Babylonian and Old and Middle Assyrian forms. T h e y also betray substrate influence, i.e. features of the scribes' native languages, including lexical items, phonological and morphological patterns, and syntactic structures. Finally, the peripheral texts also show a certain a m o u n t of grammatical simplification and reduction, the result of creolizing tendencies as the scribes attempted to write their imperfecdy-learned Akkadian. T h e features just listed are common to all Syro-Anatolian Akkadian texts, including the Ras S h a m r a Akkadian letters. Texts written in a particular locale, however, will exhibit each of the various features to a greater or lesser extent than texts written elsewhere. In

other words, the g r a m m a r and orthography of peripheral texts vary according to their provenance. For example, letters from Sidon tend to have a large n u m b e r of Assyrian forms ( A r n a u d 1 9 9 2 ) , as in the following: RS 34.149 = RSO 7, no. 38. Umma Adad-isme šar māt Sidūni;

Message of Adad-išme king of the land of Sidon;

ana šar mat Ugarit ahi-ya qibi-ma:

say to the king of the land of Ugarit, my brother:

Alterne lā [išp]ura ahu-ya an[a muhhi]- I have heard that my brother [wro]te t[o m]e: '. . .' ya: . Ittalak ana āl Sid[ūn]a ilī-ya ana' epāš' šipn-šu. Aptiqis-su ina aklē ina bēt[ē -s]u ina ūmē ša [uši]b annaka. A[numm]a ana muhhī-[ka utta"]er-šu.

He arrived at Sid [on], to me, to produce his message. I provided him with bread in [h]is hous[e] for the days that he [stay]ed here. [I have now retur]ned him to [you].

T h e following distinctions between texts written at Ugarit proper and texts written at Carchemish have been noted, inter alia ( H u e h n e r g a r d 1979):

UGARIT

CARCHEMISH

sporadic instances of incorrect broken writings intervocalic / w / written either <W> or <M> š does not always > I before dental; utu and ultu inandin, less often inaddin attu- does not occur 3fs prefix normally t-, rarely isubordination marker -u virtually absent mixing of Ill-weak forms: ileqqe/ ilaqqe, iqtabi/ iqtebi many clauses not verb-final yānu is common lū is rare preterite and perfect interchangeable for past tense

unexpected broken writings virtually absent intervocalic / w / written <M> only š > I before dental always, except for iltu (never ultu) inaddin, never inandin attu- does occur 3fs prefix i- (t- in one text only) subordination marker appears about half the time Ill-weak forms all normative Babylonian nearly all clauses verb-final yānu unattested lū is common perfect in main clauses, preterite in subordinate

T h e following two letters, the first from the king of Carchemish and the second from the king of Ugarit, illustrate several of these distinctions: RS 17.423 = PRU 4, 193. Umma sarrim-ma; ana Ibīrāni šar Ugarit qibī-ma:

Message of the king; say to Ibīrānu king of Ugarit:

Lu šulmu ana muhhi-ka.

May it be well with you.

Enūma Misra-muwa ašrānu itti Kuzvšarru-ma ašābi illaka. U atta ištu zīttīšu kī tābiš lū tēteneppuš-šu. Appūnama ana sisî-šu â tibna lū tattanaddin-ma. Summa atta ul tīdē-šu, ahū-šu ša Upparmuwa šūt; mār šanim-ma šūt. U ištu ZĪttî-šu kī tābiš lū tēteneppuš-šu.

Now then, Misra-muwa is coming there to stay with Kuzi-Sarruma. You must treat him consistently well, according to his due. In addition, you must keep his horses supplied with grain (and) straw. In case you do not know him, he is the brother of Uppar-muwa; he is the king's own son. So you must always treat him consistently well, according to his due.

RS 20.184 = Ug 5 no. 28. Say to my lord Hešmi-Teššup; message of your servant Ammištamru: I fall at my lord's feet. Ana šēpī bēlī-ya amqut. May my lord be well. May your Ana muhhi bēlī-ya lū šulmu. Ana bītīka hīrātī-ka, ana gabbi mimmu ša bēlī- house, your wives, everything of my lord's, be very, very [well]. ya danniš danniš lū [šulmu]. Ana Hešmi-Teššup bēlī-ya qibī-ma; umma Ammištamri ardī-kā-ma:

Bēl-ī, enūma kī ušebbal šarru bēlī-ya My lord, now then, whenever my sisî ana ardī-šu ina qāti ša Taguhli [...], lord the king would send horses to u hataddi arad-š[u . . .]. Kī šaknak- his servant in the charge of Taguhlu k[u. . .?]. [...], h\is\ servant would always be glad [...]. How does it seem toyo[u? ...] Now, horse [s . . .] Inanna sis[û. . .] . . . several lines missing. . . . . . several lines missing . . . [Anumma'] Amu[tara' ana mu]/}hi šarri [Herewith] I have sen[t] Amu[taru bēlī-ya altap[ar]. Bēlī-ya Amutara [ana\ t]o my lord the king. May my lord pānī šani bēlī-ya lušmbšu. Amātē-šu kī present Amutaru [t]o my lord the ša tābiš bēlī-ya ana pānī šarri bēlī-ya king. May my lord discuss his conlidbub u kī damqiš lišēr[ibšu]. cerns before my lord die king as well as possible, and prese[nt him] as properly as possible.

U bēlī-ya ana pānī šani bēlī-ya liqbi u 2 sisî damqüti u 1 qašta damiqta sa māt Hanigalbat lušēbûa ana ardī-šu ina qāti Amutari Š(arKu pānī-ya.

Also, may my my lord the king, two good horses from Hanigalbat the charge of my

U atta bēl-ī 1 qašta damiqta ša mat Hanigalbat šūbila ana ardī-ka ina qāti Amutari ^arKu pānī-ya. Anumma ana šulmāni bēlī-y[a] ištēn kitâ rabītu damiq[tu] u ištēn kitâ šanû damqu ultēbilakku.

Also you, my lord, send one good bow from Hanigalbat to your servant in the charge of my .. . Amutaru. Herewith, as a greeting gift for m[y] lord, I have sent you one goo[d] large linen garment and one other good garment.

Such grammatical peculiarities exhibited by be used as evidence for the provenance of origin is otherwise uncertain, or at least, in gest that a given letter is probably not from

2.5

lord speak before that he may send and a good bow to his servant in . . . Amutaru.

individual dialects m a y a letter whose point of some instances, to suga given place.

Akkadian Letters according to Place of Origin

Letters discovered at Ras S h a m r a before the 1994 season are listed below according to where they were written. T h e assignment of provenance is frequently uncertain; this is reflected in the list by ? or ?? before a R S n u m b e r . Letters for which the place of origin is unknown a p p e a r at the end of the list. 2.5.1

Alalah

Perhaps R S 4.449 ( V i r o l l e a u d also B u n n e n s 1987. 2.5.2

1936a, 23ff.): Mukiš or Alalah?; see

Alasiya

R S 20.018 = Ug 5, no. 22; perhaps also R S [Varia 16] ( R S L 1) = Ug 5, no. 23: Alasiya or Carchemish; see B e r g e r 1969, 219; Y a m a d a 1992. 2.5.3

Amqu

?RS 17.424e + 397b = PRU 4, 2 1 9 - 2 0 .

2.5.4

Amurru (see

Izre'el

1991)

?RS 15.024 + 050 = PRU 3, 18b; ?RS 16.111 = PRU 3, 13-4; ??RS 16.116 = PRU 3, 10b; R S 17.116 = PRU 4, 132ff; R S 17.152 = PRU 4, 214; R S 17.286 = PRU 4, 180; ??RS 19.006 = PRU 6, no. 1; ? R S 20.033 = Ug 5, no. 20; see I z r e ' e l - S i n g e r 1990; R S 20.162 = Ug 5, no. 37; R S 23.023 (unpublished; see M a l b r a n - L a b a t 1995b, 35-6); R S [Varia 27] ( F a l e s 1984). 2.5.5

Assyria

??RS 6.198 ( T h u r e a u - D a n g i n 1935); ??RS 34.165(+) = R S O 7, no. 46; perhaps also R S 18.054a = PRU 4, 2 2 8 - 9 : Assyria or Sidon; see M a y e r 1971, 2; A r n a u d 1992, 193 (5.4). 2.5.6

Aštate

?RS 34.134 = R S O 7, no. 31; ??RS 34.141 = R S O 7, no. 32; ??RS 34.173 = R S O 7, no. 33. 2.5.7

Beirut

R S 11.730 = PRU 3, 12-3; R S 34.137 = R S O 7, no. 37; R S 8 6 . 2 2 1 2 (unpublished; see A r n a u d see also the next entry. 2.5.8

1992,

192

(3.5));

Bi'rut

R I H 8 1 / 4 ( A r n a u d 1 9 8 4 , who suggests that this Bi'rut is the Ugaritian port rather than the Phoenician city). 2.5.9

Carchemish (see

Huehnergard

R S 8.333 = PRU 3, 7 - 8 ; R S 15.077 = PRU 3, 6 - 7 ;

1979)

R S 16.003 = PRU 3, 3 - 4 ; ?RS

17.078

=

PRU

4,

196-7;

R S 17.289 = PRU 4, 192; RS

17.292

=

PRU

4,

188;

R S 17.385 = PRU 4, 194; R S 17.423 = PRU 4, 193; R S 20.022 = Ug 5, no. 2; ?RS 20.174a = Ug 5, no. 25; ??RS 20.182c = Ug 5, no. 63; ?RS 20.216 = Ug 5, no. 35; R S 20.237 = Ug 5, no. 31; ?RS

25.461

(Lackenbacher

1989,

317-9);

R S 34.136 = R S O 7, no. 7; R S 34.138 = R S O 7, no. 8; ?RS 34.143 = R S O 7, no. 6; R S 34.145 = R S O 7, no. 9; R S 88.2013 (unpublished; see M a l b r a n - L a b a t in 1995b, 39); perhaps also R S 13.007b = PRU 3, 6a: Carchemish or Hatti; perhaps also R S [Varia 16] ( R S L 1) = Ug 5, no. 23: Alasiya or Carchemish; see B e r g e r 1969, 219; Y a m a d a 1992. 2.5.10

Egypt

86.2230 (unpublished; see A r n a u d 1992, 181 η. 6); R S 88.2158 (unpublished; see L a c k e n b a c h e r 1995b); perhaps also R S 26.158 = Ug 5, no. 171: Hatti or Egypt. RS

2.5.11

Emar

?RS [Varia 26?] (Latakia 88) = R S O 7, no. 30; see 105 with n. 33. 2.5.12

Hatti (see

Hagenbuchner

1989)

?RS 15.033 = PRU 3, 15-16; R S 17.130 = PRU 4, 103ff. (letter/treaty); R S 17.132 = PRU 4, 35ff); R S 17.133 = PRU 4, 118-9 (letter/legal); ??RS 17.144 = PRU 6, no. 6; R S 17.247 = PRU 4, 191; ?RS 17.429 = PRU 4, 2 2 7 - 8 (letter?);

Arnaud

1982a,

R S 18.003 = PRU 4, 103ff. (letter/treaty); R S 20.212 = Ug 5, no. 33; R S 20.255A = Ug 5, no. 30; R S 34.129 = R S O 7, no. 12; ?RS 34.133 = R S O 7, no. 36; ??RS 22.216 ?RS 88.2009 perhaps also perhaps also perhaps also perhaps also 2.5.13 ?RS

Ma'fraz

10.046

2.5.14

(unpublished; see M a l b r a n - L a b a t in 1995b, 36); (unpublished; see M a l b r a n - L a b a t in 1995b, 39); R S 13.007b = PRU 3, 6a: Carchemish or Haiti; R S 17.383 = PRU 4, 221ff.: Hatti or Ugarit; R S 17.422 = PRU 4, 223ff.: Hatti or Ugarit; R S 26.158 = Ug 5, no. 171: Hatti or Egypt.

(Virolleaud

1941,

1-2;

see also PRU

3,

9-10).

Man

?RS 34.142 = R S O 7, no. 47. 2.5.15

Mukiš

R S 20.003 = Ug 5, no. 26; perhaps also R S 4.449 ( V i r o l l e a u d 1936a, 23ff.): Mukiš or Alalah?; see also B u n n e n s 1987. 2.5.16

Parga

R S 15.019 = PRU 3, 13b. 2.5.17

Qadesh

R S 20.016 = Ug 5, no. 38; R S 20.172 = Ug 5, no. 39; ?RS 20.200b = Ug 5, no. 40; R S 34.146 = R S O 7, no. 15; R S 80.387 ( A r n a u d 1982a, 221-2). 2.5.18

Sidon (see

Arnaud

1992)

R S 11.723 = PRU 3, 9b; RS

25.430a

RS

34.149

RS

86.2208

(unpublished; see =

R S O

7, n o .

Arnaud

1992,

193

(5.7));

Arnaud

1992,

193

(5.11));

38;

(unpublished; see

RS

86.2221

Arnaud

+

86.2225

1992,

+

+ 8 6 . 2 2 4 0 (unpublished; see much of text transliterated and trans-

86.2226

1 7 9 - 9 4 (5.12);

lated in footnotes); ?RS 8 6 . 2 2 3 4 (unpublished; see A r n a u d 1 9 9 2 , 1 9 3 ( 5 . 1 3 ) ) ; perhaps also R S 1 8 . 0 5 4 A = PRU 4 , 2 2 8 - 9 : Assyria or Sidon; see M a y e r

2.5.19.1

1971,

2; A r n a u d

1992,

193

(5.4).

Siyannu

RS 17.083 = PRU 4, 216; R S 17.143 = PRU 4, 217-8; ?RS 17.288 = PRU 4, 215; R S 17.425 = PRU 4, 218b; ?RS 20.017 = Ug 5, no. 43; RS 20.021 = Ug 5, no. 42; ??RS 21.183 = Ug 5, no. 41. 2.5.19.2

Siyannu/Ušnatu

R S 34.158 = R S O 7, no. 16; perhaps also R S 20.219 = Ug 5, no. 44: Ugarit or Siyannu. 2.5.20

Tarhudašši

R S 34.139 = R S O 7, no. 14. 2.5.21 R S

Tyre

[Varia 25] = Latakia

2.5.22

Ugarit (see

7 (Arnaud

H u e h n e r g a r d

1982b).

1989)

Aphek 5 2 0 5 5 (TA 8 , 7 - 8 [ O w e n ] ; see also S i n g e r ?Ber1in 1690 (EA 48); Berlin 1692 (EA 45); ?Ber1in 1693 (EA 47); ?Ber1in 1694 (EA 46); Cairo 4783 (EA 49); ? C K 7 (private collection; unpublished; see A r n a u d ?RS 15.011 = PRU 3, 19; ?RS 15.014 = PRU 3, 5; ??RS 15.063 = PRU 3, 20a; RS 16.112 = PRU 3, 4b;

1983);

1992,

192

(3.6));

?RS 17.239 = PRU 6, no. 8; R S 17.455 = PRU 6, no. 3; R S 19.070 = PRU 4, 294; ?RS 19.080 = PRU 6, no. 2; ?RS 20.013 = Ug 5, no. 49; ??RS 20.015 = Ug 5, no. 53; ??RS 20.019 = Ug 5, no. 48; ?RS 20.023 = Ug 5, no. 54; ??RS 20.141b = Ug 5, no. 34; ?RS 20.158 = Ug 5, no. 51; R S 20.168 + 195p = Ug 5, no. 21; R S 20.178 = Ug 5, no. 55; ? R S 20. 182a(+)b =

Ug 5 , n o .

36;

R S 20.184 = Ug 5, no. 28; R S 20.200c = Ug 5, no. 29; ??RS 20.232 = Ug 5, no. 58; R S 20.238 = Ug 5, no. 24; ?RS 20.239 = Ug 5, no. 52; ?RS 20.243 = Ug 5, no. 32; ??RS 22.347 (unpublished; see

Malbran-Labat

?RS

25.131

(Lackenbacher

1989,

?RS

25.138

(Lackenbacher

1989,

318,

in 1995b, 35);

320);

318-20);

?RS 32.204 = R S O 7, no. 19; ?RS 34.135 = R S O 7, no. 17; ?RS 34.140 = R S O 7, no. 11; ?RS 34.150 = R S O 7, no. 10; ?RS 34.151 = R S O 7, no. 13; ??RS 34.153 = R S O 7, no. 35; ?RS 34.154 = R S O 7, no. 18; ?RS 34.180,17 (34.180f) = R S O 7, no. 26; ?RS [Varia 10] (1957.2) = A n O r 4 8 , 2 3 - 4 ; perhaps also R S 17.383 = PRU 4, 22 Iff.: Hatti or Ugarit; perhaps also R S 17.422 = PRU 4, 2 2 3 f f : Hatti or Ugarit; perhaps also R S 20.219 = Ug 5, no. 44: Ugarit or Siyannu. (Ušnatu: see Siyannu § 2.5.23 RIH RS

8.2.5.19)

Provenance unknown

77/17 1.056

(Arnaud-Kennedy

(Virolleaud

1929,

1979, pl.

318);

76/1);

R S 1 .[057] ( V i r o l l e a u d 1929, pi. 76/2); R S 1-11.[028] = A O 19.952 (unpublished; pardal transliteradon in n. 1); PRU 3, R S 11.794 = PRU 3 17-8; (letter?); R S 11.834 = PRU 3 17b; R S 12.005 = PRU 3 16-7 R S 12.033 = PRU 3 1 4 - 5 R S 15.018 = PRU 3 1 1 - 2 ; R S 15.108 = PRU 3 2 0 - 1 (letter?); R S 15.124 = PRU 3 21b (letter?); R S 15.178 = PRU 3 8 - 9 ; R S 17.142 = PRU 6 no. 4; R S 17.148 = PRU 6 no. 7; R S 17.315 = PRU 4 111; R S 17.390 = PRU 6 no. 10; R S 17.391 = PRU 4 226a; R S 17.393 = PRU 4 2 2 6 - 7 ; R S 17.394 + 427 = PRU 4, 220b; R S 17.398 = PRU 6 no. 11; R S 17.428 = PRU 6 no. 9; R S 17.451 = PRU 6 no. 12; R S 17.452 = PRU 6 no. 5; R S 17.456 = PRU 4 228b (letter?); R S 18.057 = PRU 6 no. 13; R S 18.089 = PRU 6 no. 15; R S 18.268 = PRU 4 229b; R S 18.281 = PRU 6 no. 17 (letter?); R S 19.050 = PRU 6 no. 14; R S 19.053 = PRU 6, no. 18 R S 19.115 = PRU 6, no. 19 RS RS RS RS RS RS RS RS RS RS

20.095 = Ug 5, no. 65; 20.130 = Ug 5, no. 46; 20.141a = Ug 5, no. 76; 20.150 = Ug 5, no. 56 20.151 = Ug 5, no. 50 20.159 = Ug 5, no. 74 20.182d = Ug 5, no. 67; 20.189d = Ug 5, no. 73; 20.191 = Ug 5, no. 75; 20.194 = Ug 5, no. 62;

RS RS RS RS RS RS

20.196d = Ug 5, no. 77 (letter?); 20.200A = Ug 5, no. 78; 20.2 1 4 d = Ug 5, no. 79; 20.225 = Ug 5, no. 45; 20.227 = Ug 5, no. 57; 20.242 = Ug 5, no. 72;

20.244 = Ug 5, no. 61 (same tablet as 1974]); R S 20.246 = Ug 5, no. 68; R S 20.248 = Ug 5, no. 59; R S 20.426,14+ = Ug 5, no. 70; R S 20. [438] = Ug 5, no. 47; R S 21.006c? = Ug 5, no. 80 (letter?); R S 21.007c = Ug 5, no. 71; R S 21.054b = Ug 5, no. 66; RS

RS

21.063f

1974]); R S 21.064 R S 21.072 R S 21.201 R S 22.006 RS RS RS RS

RS RS RS RS RS RS RS RS RS RS RS RS RS RS RS

=

= = = =

Ug 5, no. 60 (same tablet as

RS

RS

21.063f?

[Kühne

20.244?

[Kühne

Ug 5, no. 64; Ug 5, no. 69; PRU 6, no. 20; PRU 6, no. 16;

23.031 (unpublished; see M a l b r a n - L a b a t 23.365 (unpublished; see M a l b r a n - L a b a t 2 3 . 3 6 8 (unpublished; see M a l b r a n - L a b a t 28.017 (unpublished; see M a l b r a n - L a b a t 34.070 = R S O 7, no. 29; 34.152 = R S O 7, no. 40; 34.155 = R S O 7, no. 21; 34.160 = R S O 7, no. 24; 34.161 = R S O 7, no. 22; 34.163 = R S O 7, no. 39; 34.164 = R S O 7, no. 34; 34.167 + 175 = R S O 7, no. 25; 34.170 = R S O 7, no. 23; 34.171 = R S O 7, no. 20; 34.174 = R S O 7, no. 41; 34.180,12 (34.180e) = R S O 7, no. 28; 34.180,5 (34.180g) = R S O 7, no. 42; 34.180,60 = R S O 7, no. 27; 88.2011 (unpublished; see M a l b r a n - L a b a t

in in in in

1995b, 37); 1995b, 35); 1995b, 3 7 ) ; 1995b, 35);

in 1995b, 39);

RS RS RS RS RS RS

86.2216 (unpublished; see R S O 5 / 1 , 357); 86.2232 (unpublished; see R S O 5 / 1 , 359); 86.2236 (unpublished; see R S O 5 / 1 , 359); 86.2241a (unpublished; see R S O 5 / 1 , 360); 86.2249a (unpublished; see R S O 5 / 1 , 360: letter?); [Varia 34] = A O 29.507 (unpublished; see R S O 5 / 1 , 381).

C H A P T E R NINE T H E

L E G A L

T E X T S

Ignacio

1

1.1

F R O M

M a r q u e z

U G A R I T

Rowe*

Introduction

Definition and scope

What constitutes a legal document? And what is the difference between a legal and an administrative text, both often associated and sometimes even confused? These are two basic questions we should attempt to answer before we begin to discuss the evidence. Strictly speaking, a legal document serves mainly as written proof that a transaction has taken place. Since the transaction itself can be validated if, and only if, it is performed before a body of witnesses, the legal document, as opposed to administrative records, must mention the n a m e or names of the witnessing party. It should also be noted in passing that although the practice of signing or sealing, i.e. affixing a personal mark, is almost a prerequisite in the composition of a legal document (for it provides the proof that the person who signs is present or has assumed the obligation undertaken in the transaction) it is by no means confined to this genre (administrative texts or letters may also bear such marks). However satisfactorily these definitions may have answered the opening questions, anyone who has dealt with this kind of source knows well that the problem is not so simple. Indeed, the dividing

* I wish to express my most sincere thanks to Pierre Bordreuil and Dennis Pardee, epigraphists of the Mission de Ras Shamra, who kindly made available to me their transliterations and copies of the unpublished Ugaritic texts discussed in this paper. I am also grateful to Mme Florence Malbran-Labat, also epigraphist of the Mission de Ras Shamra, who kindly provided me with the contents of the newly excavated Akkadian legal documents as well as with her transliterations and copies of other unpublished texts. My especial thanks go to Dennis Pardee, who read critically an earlier draft of this paper, making several corrections and annotations. It goes without saying that any mistakes remain my own responsibility.

line between a legal and an administrative document is far from ideal. For example, as early realized, bills and receipts that are usually classified a m o n g administrative records may sometimes mention that the act has been performed before witnesses; and, on the other hand, some contracts may unexpectedly omit any reference to them. This general problem is also found in the corpus of legal texts from Ugarit. As already remarked by Nougayrol in publishing the first important group of clay tablets of legal content unearthed at Ras Shamra, three out of his thirty-eight texts lay in fact 'à la limite du domaine juridique'. 1 As a result of these attested cross-reladonships (in addition to the uncertain identification of badly damaged fragments) the n u m b e r of legal documents of Ugarit obviously varies in accordance with the individual scholar's point of view. Although this is no place to discuss every controversial case, mention will be made here of the criteria followed and the consequent inclusion or exclusion of documents. Given the broad scope of a handbook it has been considered more convenient here to frame our material not according to its linguistic provenance but rather according to its find-spot. This means that we shall include not only the m a j o r group of domestic legal texts (over 250 texts and fragments) but also the international legal documents comprising edicts and treaties (about 100 texts and fragments) that have been found at Ugarit, which stem namely from the courts of Hatti, Carchemish, Amurru, Siyannu as well as that of Ugarit itself.2 Regardless of the place of the composition of the texts, legal documents found so far at Ugarit are inscribed without exception on single clay tablets (and called accordingly in Akkadian tuppu, and in Ugaritic spr) without envelopes.

1.2

The find-spot

Due to the ever-increasing corpus of texts excavated at Ras Shamra, only a general, but not complete, picture of the archaeological distribution of the written material of Ugarit can be attempted. Another

1

2

NOUGAYROL

1952,

182 n.

2.

This same scope was also applied in the last survey of the legal texts published by D . Pardee and P . Bordreuil in the ABD article on Ugarit ( P A R D E E - B O R D R E U I L 1 9 9 2 , 718FF.). In contrast, in the corresponding contribution to SDB, Sznycer dealt only with the legal texts written in the Ugaritic language (SZNYCER 1 9 7 9 , 1417FF.).

limit to the interpretation of the archaeological context of tablets and fragments is the presence of stray finds: intrusions a m o n g the different archives of Ugarit do exist as shown by joins between far distant pieces. 3 T h e bulk of the legal texts of Ugarit was unearthed from the royal palace archives at Ras S h a m r a (around 300 texts and fragments). It was soon observed after the excavation of the palace that the concentration of these records followed a deliberate pattern according to their genre. Indeed, the royal domestic legal texts, in which the king appears regularly either as the person presiding or as one of the parties of the transactions, come mainly from the so-called 'central archive', while legal texts dealing with international affairs were kept as a rule in the palace 'southern archive'. Several private archives outside the royal palace have also yielded legal material. Each of them contains a relatively small n u m b e r of texts (they rarely exceed 10 records) and apparently deal with the private business of the owners of the households over some generations. As reported from the last epigraphic finds, the archive in the house of Urtenu kept copies of three international legal documents. 4 In this regard, one should note that the owners of these archives were presumably high officials of the kingdom of Ugarit, so that both the existence and content of the archives may be explained to a certain extent on account of their political and social status.

1.3 Chronology Dating texts was not practised at all by Ugarit scribes. Strictly speaking, not even legal records were dated, in contrast to the c o m m o n legal tradition. This is also true of the international documents drawn up by scribes of the courts of Hatti, Carchemish, Amurru or Siyannu. T h e only chronological reference is the n a m e of the king (obviously normally attested in royal documents) or a possible prosopographical connexion, such as for example the name of the scribe. In any case, the most accurate date we can give for any document stands grosso modo for the span of one particular reign. T h e chronological frame of our material covers the reigns of seven successive kings, from N i q m a d d u II down to c Ammurapi', in all like3

Cf.

4

MALBRAN-LABAT

VAN S O L D T

1991a, 50. 1995b, 1 1 0 - 1 1 .

lihood the last m o n a r c h of Ugarit. T h i s means that this corpus belongs to a period, in absolute terms, of about 150 years, from around 1330 bce down to the destruction of Ugarit. 5

1.4

Language and script

N o doubt Ugarit has provided us with very rich textual material, both in n u m b e r and variety, attesting up to seven different languages and five different scripts. Scribes at Ugarit, however, were mainly biscriptal:6 they basically wrote Ugaritic, their own tongue, in alphabetic cuneiform, and Akkadian, the lingua franca of the time, in syllabic cuneiform. W h a t was noted for the archival distribution of the legal texts is also true for the language and script. T h e genre in question seems to be closely connected with the language employed. Indeed, except for a few cases, legal texts are drawn up in Akkadian in syllabic cuneiform. T h e few exceptions correspond to several Ugaritic documents and one Hittite text. T h e reason why the treaties and international legal documents found at Ras S h a m r a were composed in Akkadian can be easily explained on account of the aforementioned diplomatic nature of the language. Concerning the domestic texts, however, the explanation has to be sought elsewhere, especially because scribes did chiefly choose Ugaritic to write their local administrative records. Admittedly, this question seems to be intimately related to the genre of the material. In coining the expression and discipline of 'cuneiform law' Koschaker m e a n t to make clear that the strong appeal of Mesopotamian or, more exactly, Babylonian law to other cultural centres of the ancient Near East was fixed at a formal level, of which script and language constituted the basic aspects.' A closer look at the internal evidence shows furthermore that the few Ugaritic examples are fully influenced by their Akkadian counterparts. Because the division of legal texts in this section is made according to the language of the texts the order will follow the linguistic conventions of this genre in Ugarit.

5 6 7

We follow here the absolute dates proposed by VAN S O L D T (1991a, This designation was coined by VAN S O L D T (1995a, 186). KOSCHAKER

1935,

27.

1-46).

2

T h e

Akkadian

Legal

Texts

W e have mentioned above that Akkadian was the language par excellence of legal documents. W e also stated that this Akkadian corpus consists of two main different genres or sub-genres, namely records dealing with domestic affairs and records that concern foreign affairs. Accordingly, we propose here to deal with them separately. It should be noted that scribes at the royal court of Ugarit were probably familiar with this distinction; not only because of the content of the documents they could read or the characteristic physical aspect of the tablets (such as their shape or seal impressions) but also because they deliberately filed them separately within the royal archives, as pointed out above.

2.1

The domestic legal texts

Texts dealing with domestic transactions, and thus written at Ugarit, are by far the major group of legal documents. Nougayrol suggested classifying these records according to the nature of the witnessing party (as we have seen, the main defining element of this genre). H e distributed them into three different categories: 1) acts performed before witnesses, 2) those performed before the king and 3) acts of the king, in which no mention of the witnessing party is made. 8 Concerning the latter type, it is generally agreed that the active role of the king must have rendered unnecessary the mention of witnesses in the text. In other words, one must assume that the king presided over these transactions too. This, as we shall see, can be supported by the seal impression of the tablets in question. As a result, we suggest dividing the domestic legal texts of Ugarit into two main groups: 1) records describing acts presided over by or performed before the king, and 2) records describing acts performed before witnesses. In spite of this division, both royal and non-royal documents were written by Ugaritian scribes (in fact, we know of scribes who wrote both kinds of texts) and present, therefore, several c o m m o n features that distinguish them from other, foreign texts. At first sight, it is possible to distinguish a legal text composed at Ugarit on the basis 8

See

NOUGAYROL

1952,

182ÍT. a n d

1955,

23.

of both the shape of the tablet and the sealing practice. T h e prototype is the c o m m o n tablet (about 90 x 70 m m , the thickness exhibits more variation, reaching sometimes about 40 mm) with one single seal impression placed on top of the obverse, hence at the head of the text (sometimes divided by a ruling). A close look at the tablets shows that the seal was rolled (very seldom stamped) before the text was written. T h e text, on the other hand, also presents several characteristic features of this local legal practice. N o doubt the most typical example is the invariable opening adverbial phrase 'From today' (ištu ūmi annîm).

2.2 2.2.1

The royal legal texts

Definition and scope

This group consists of those legal texts in which the king of Ugarit presides, explicidy or not, over the transactions involved. These 'actes royaux', as Nougayrol called them, are as a rule characterized by one main feature: they consistently bear the same seal impression, that of the so-called 'dynastic seal' of the kings of Ugarit. This feature is of course essential to define a legal text, for the presence of this impression implies that the owner of the seal, i.e. the king, was present at the transaction. Therefore, the preservation of the royal seal impression, or the corresponding seal identification formula, quickly characterizes a royal document. O f course, the form and content which describe the nature of the transaction are also good indicators for identifying these texts. I have counted altogether about 170 texts and fragments in Akkadian that describe domestic transactions presided over by the king. 2.2.2

Find-spot

All but three of the documents were found within the walls of the royal palace. This should not be surprising when dealing with royal documents. Most significant, however, is the distribution of these texts within the palace proper. O f the approximately 170 texts and fragments, at least 147 come from the same royal archive, n a m e d by Schaeffer 'les Archives Centrales'. A closer look at the archaeological context has shown that these records were in all likelihood stored on file in a room located on the upper floor of the northern

wing of this area. 9 Also of interest is the fact that almost all the legal texts discovered in this archive are royal documents. Therefore it is reasonable to assume that this archive was probably meant to keep royal domestic legal documents. 2.2.3

Form

T h e physical characteristics of these documents have been referred to above. O n e should note, however, the existence of exceptionally large tablets like the Sammelurkunde R S 15.109+. W e also mentioned that the seal impressed at the head of these tablets, except for just a few cases, is the 'dynastic' or state seal of Ugarit. This cylinder seal, as the legend reads, belonged to 'Yaqarum, son of N i q m a d d u , king of Ugarit', very likely the Amorite ancestor and founder of the Ugaritic dynasty. All the members of the dynasty from N i q m a d d u II down to the last monarch c Ammurapi' used as a rule this seal as well as an official replica of it to sign their records. O t h e r royal seals did exist and were occasionally impressed on royal documents like the stamp ring-seal of N i q m a d d u (presumably N i q m a d d u II) on RS 17.147. T h e form of the royal documents follows a very regular and concise pattern. T h e first section consists of a description of the completed part of the transaction; it is thus phrased in the past (preterite or perfect) and in objective style. For example, the king has granted, the buyer has purchased. Characteristic of this part is the opening adverbial expression 'from to-day', mentioned above, usually combined with the closing expression '(and) for ever' (ana/adi dāntim is the most frequent formula). T h e king appears then after that introductory phrase either as having presided over the transaction, 'before R N ' (ana pāni RN), or as first and active party. Apart from the king the other parties involved are properly identified too by name with or without further qualification. Sometimes a summary or the result of this event marks the end of this section. Next follow the so-called 'final clauses'. These deal with the presentfuture stipulations or obligations of the transaction such as the guarantee, eviction or services inherent to the contract, and are therefore phrased regularly in the durative, also in objective style. For example,

9

See

VAN S O L D T

1991a, 88-93.

no-one will take the estate granted or purchased from the grantee's or buyer's hands; the grantee will or will not have to perform the ilku/pilku-servìce. Finally, the document closes with the formula that identifies the dynastic seal, either as '(great) seal of the king' or as 'seal of R N ' , often followed by the name of the scribe who is sometimes qualified as witness. Schematically, then, the structure of royal d o c u m e n t s can be described as follows: 10 a) b) c) d)

dynastic seal impression, operative section, seal identification formula, name of the scribe.

O n the other hand, a royal document can describe one or more transactions. These c o m p o u n d transactions deal as a rule with the same party involved as beneficiary. At least in some cases it seems that each of them may have been previously written in one single document (the most illustrative example is the above mentioned document RS 15.109+, already qualified as Sammelurkunde). These transactions may belong to the same category, called accordingly by Nougayrol 'actes multiples homogènes' (e.g. RS 15.109+, RS 15.122, RS 16.248), but also to different ones. O n e example of the latter 'actes multiples hétérogènes' is RS 15.85 which contains a royal transaction (in lines 1-10) beside a non-royal one, i.e. performed 'before witnesses' (in lines 11-21). Before closing this chapter, it should be noted that the legal practice of Ugarit hardly had any recourse to the divine action against transgressors of contracts; on the other hand, there are no examples of promissory oaths. 2.2.4

Content

Almost all the royal documents that preserve a more or less complete and intelligible text are deeds of conveyance. Gifts (e.g. RS 15.88, RS 15.145, RS 16.142, RS 16.150), sales (e.g. RS 15.136,

10

196

Other schemes have been proposed by N O U G A Y R O L 1 9 5 5 , 2 4 , H A A S E 1 9 6 7 , (for the sale documents), K I E N A S T 1 9 8 0 , 5 3 3 (also for the sale documents) and

KIENAST

1979,

433.

R S 16.137, R S 16.156), purchases (e.g. R S 15.119, R S 16.284), barters (e.g. R S 15.123+, R S 16.158, R S 16.246) or divisions of inheritance (e.g. R S 15.90, R S 15.120), and even royal verdicts that concern litigations on landed property (e.g. R S 16.254c, R S 16.356) record the resulting transfer, usually heritable, of title to real estate. Real estate in the legal texts of Ugarit basically consists of houses and land which include, as the most characteristic cultures of this Mediterranean area, vineyards and olive groves. T h e description of the property conveyed is as a rule very concise providing only the name of the previous tenant; no measurements or boundaries are given and sometimes a brief geographical reference of its location is mentioned. By far the largest group of documents is composed by deeds of royal gifts, with or without countergift (i.e. in return for a sum of money like R S 15.126, R S 16.135, R S 16.167 or R S 16.174). In a few records the gift concerns entire villages (e.g. R S 15.114, R S 15.147) a n d / o r some of their taxes (e.g. R S 16.153, RS 16.244). A m o n g the few transactions that do not concern, strictly speaking, the law of property one should mention several deeds of adoption (e.g. R S 15.92, R S 16.200, R S 16.344) and manumission (e.g. R S 16.267) as well as grants of privileges like franchises (e.g. R S 16.238+) or promotions (e.g. R S 16.348). As for the parties involved in these transactions, no doubt the king stands as the main protagonist. Documents from the time of c Ammittamru II are by far the most attested. T h e donees of royal gifts are often found not only benefiting from several grants but also as the contracting parties of deeds of sale, purchase and barter. Accordingly, Nougayrol grouped these records under different 'dossiers'. Among these individuals, who are not always well identified, we find members of the royal family and officials of the court of Ugarit. 2.2.5

An example

Transactions described in royal deeds obviously differ from text to text, not only in contents but also in form due to the particular style of each scribe. Nevertheless, for the sake of illustration, we have chosen one of these documents, namely R S 16.275, a gift of real estate from N i q m a d d u II to his brother Nuriyānu, to show the pattern described above.

Obv

Impression of dynastic seal

is-tu u4-mz [an-ni-(i-)im] l

d

níq-ma- iškur

dumu

['Ammittamru] aru lugal u-ga-n-i[t] it-ta-ši é-/[tt]

5

Rcv

sa lpa-be-y[i] i-na uru lul] -l[a-mi] ù id-d[in-šu] a-na ^nu-r[i-ia-na] ù a-na dumu.me[š-rá]

10

a-na da-n-tim[-ma] ma-am-ma-an la-a i-l[eq-qè(-šu)] is-tu šu-ti lnu-n-ia-[na] ù iš-tu šu-ft dumu[.meš-iu]

a-na da-n-tim-rn[a ] 15

na4

kišib s'a 1uga[Ī]

2.3 2.3.1

From today Niqmaddu, son of 'Ammittamru, king of Ugarit, has taken the house of P N (which is located) in T N and has given it to Nuriyānu and to his sons for ever. No-one will take it from the hands of Nuriyānu or from the hands of his sons ever. Seal of the king.

The non-royal legal texts

Definition and scope

This second group of domestic legal texts consists of those records which describe transactions p e r f o r m e d before witnesses a n d that were not presided over by the king. As opposed to royal deeds, these documents are regarded as 'private deeds'. Following o u r definition above, we have excluded f r o m this corpus m e m o r a n d a (e.g. R S 15.41) or bills (e.g. of personal security like R S 15.81 or R S 16.287) in which no mention of a body of witnesses is m a d e . O n the other h a n d , a text like R S 16.354 which recapitulates a series of debt-notes with their respective testimonies, lies precisely on the borderline of the legal domain. All in all the material consists of about 65 relatively well-preserved d o c u m e n t s a n d of several fragments. 2.3.2

Find-spot

As one would expect, most of these d o c u m e n t s have been f o u n d in private archives. T h e m a j o r concentrations c o m e f r o m the so-called 'Residential Q u a r t e r ' : nine d o c u m e n t s f r o m the house of R a s a p ' a b u a n d eight m o r e or less complete d o c u m e n t s a n d several fragments f r o m the house of R a p ' ä n u . O n e should also mention the recently

discovered archive of Urtenu which counts now some seven Akkadian legal records. O f interest is the fact that the royal palace has also yielded private documents. T h e reason for this find-spot is not easy to explain. It has been suggested that the king was highly involved in the legal life of Ugarit. But other explanations are also possible. It is plausible, for example, that some of the documents could actually belong to court officials or other personnel closely connected with the palace. In this regard, one should note that three of these deeds (RS 17.86+, RS 17.102, RS 17.325) belonged to the queen's 'dossier' and at least two of them (RS 11.856, R S 15.182) to the sākiniCs. 2.3.3

Form

As already mentioned above, the physical appearance of these tablets is not essentially different from that of royal documents. T h e seals impressed at the head of the documents were usually anepigraphic cylinder seals and belonged to the private people involved in the transactions, especially those undertaking obligations. A small number of deeds, however, are without seal impression (e.g. RS 11.856, R S 20.176) and some of them show the sealing space in blank (like R S 20.146). T h e schema and form of these texts are also similar to the ones described above for royal documents, and as was mentioned above, we know of scribes like Nu C me-rašap, Iltahmu or M u n a h h i m u who wrote down both royal and private transactions. T h e text begins with the phrase 'from to-day', followed by the mention 'before witnesses' (ana pāni šībūti) which thus determines the 'private' character of the deed. After these introductory expressions the operative section describes in the past and in objective style the fulfilled part of the transaction. Next follow the obligation clauses drawn up in the durative and also in objective style. N o stipulations on services are attested in private deeds; on the other hand, penalty clauses occur rather frequently in these texts. T h e note that identifies the seal impression and the list of witnesses, which often includes the scribe, close the document. T h e following scheme can be outlined:"

" See previous note.

a) b) c) d)

private seal impression operative section seal identification formula list of witnesses (often including the n a m e of the scribe)

2.3.4

Content

Private archives contain a relatively restricted n u m b e r of deeds and they are concerned with the activities of the members of the family. A m o n g them we find deeds of conveyance like sales of land (e.g. RS 15.182, R S 17.61) and persons (e.g. RS 11.856, RS 20.236), purchases (RS 15.37, R S 17.22+) and gifts especially in contemplation of death (RS 8.145, R S 17.38, R S 17.378a). As with royal documents, houses and lands are described in a very concise m a n n e r (note the exceptional reference to measurements in RS 17.22+). Very c o m m o n a m o n g such family transactions are deeds that concern the law of persons, such as deeds of emancipation (e.g. RS 8.279bis, RS 16.129, RS 20.176, RS 27.53) and deeds of adoption (RS 17.21, RS 20.226, R S 21.230, R S 25.134). Little is known of the identity of the contracting parties. As we have already mentioned, the queen appears in some of these documents (deeds of sale) and one can argue that many of the other contractors must have belonged to the relatively high social class in Ugarit. O n the other hand, the n u m b e r of witnesses is not uniform; three to five witnesses seems to have been the general tendency but examples of texts with as few as two and as many as fourteen witnesses (e.g. R S 14.16) are attested. 2.3.5

Publication and studies

T h e majority of these Akkadian texts was published by Nougayrol in three volumes of the French Mission de Ras Shamra: N o u g a y r o l 1955, 1968 and 1970. T h e former contains most of the royal legal texts and also includes an excellent study of the material. In the same volume, B o y e r (1955) offers a juridical interpretation of the documents as compared to other cuneiform legal sources. O t h e r scattered non-royal texts were published by T h u r e a u - D a n g i n (1937), V i r o l l e a u d (1941c no. I I , 1951 no. V I ) and, more recently, L a c k e n b a c h e r (1991) and Malbran-Labat ( B o r d r e u i l et al. 1991, no. 29). A few still remain unpublished. Apart from some scattered partial studies, two dissertations have

dealt with the legal texts of Ugarit, chiefly those written in Akkadian: M i l l e r 1 9 8 0 and L i b o l t 1 9 8 5 . T h e former basically consists of an English translation with a brief commentary of all the domestic legal documents (including also some administrative records); and the latter exclusively deals with the royal land grants, presented in transliteration and translation in chronological order, and aims at determining the evoludon of Ugaridan society through a thorough examination of the changes in language, namely the clauses on heritability in these texts. A study by the present author of both the language and the content of the royal deeds of conveyance is forthcoming.

2.4 2.4.1

The international legal texts

Definition and scope

U n d e r this category we include all legal documents unearthed at Ugarit that are concerned with international matters such as treaties, edicts, verdicts or any other transactions that involve parties of more than one state. This material consists of approximately one hundred texts and fragments. More often than not it is the state or states proper that are involved, in which case the parties stand for the rulers themselves or also sometimes their plenipotentiaries. Therefore the division of the evidence in accordance with the witnessing party corresponds mutatis mutandis with the palace, viz. the court of the state of origin. A verdict promulgated by the Hittite suzerain usually took place at his own court, and the same holds true for arbitrations or agreements drawn up by the scribes of the kings of Carchemish, Ugarit, A m u r r u or Siyannu (the provenance of a small n u m b e r of texts and fragments cannot be determined due to the obscure or fragmentary context). 2.4.2

Find-spot

International documents found at Ugarit obviously concern in one way or another the state of Ugarit proper. Therefore, as one would expect, most of these records have been discovered in the archives of the royal palace. As we have already mentioned, the distribution of these texts follows a clear pattern. Over three quarters of this material come from the same concentration of tablets, the so-called 'southern archive'. Like the domestic royal legal texts found in the palace central archive, international documents were probably stored

on file on the upper storey, 12 and they seem to have been kept there separately. Moreover, some of them were identified by means of labels or tags attached either to the tablets themselves or the baskets in which they were placed. 13 Although some of the few finds outside the royal palace could have been misplaced, it is nonetheless probable that several private archives belonging to high state officials would also have contained some of the international legal documents (as is certainly true of the epistolary material). This seems indeed to be the case of the archive of Urtenu which, according to the last epigraphic report already mentioned above, has yielded three documents of Hittite origin.

2.5 2.5.1

The texts from Hatti

Definition and scope

It goes without saying that the nature of the foreign records found at Ugarit is intimately connected with the political and diplomatic relations existing between the states involved. Indeed, the fact that Ugarit was annexed to the Hittite empire as a vassal state during the reign of Niqmaddu II explains why the approximately thirty legal documents stemming from the Hittite court discovered at Ras Shamra basically consist of vassal treaties, edicts and verdicts establishing Ugarit's foreign relations. Dated texts range from the days of the annexation of Ugarit by Suppiluliuma I down to the reign of T u d haliya IV. In opening the section on legal texts we referred to the general problem of drawing the line that divides legal texts from administrative records. With Hittite documents the controversy is further raised mainly due to the form of the texts. As is well known, Hittite scribal practice often opens the text with the epistolary Akkadian formula umma P N / R N 'Thus (says) P N / R N ' , regardless of the basic nature of the document (e.g. in treaties or annals). Furthermore, the use of subjective or objective style in the body of the text is not conclusive either since both are attested in letters and legal documents, and they may also occur side by side in the same text. As a matter of fact, one may wonder whether the Hittite Great King made

12 13

See VAN S O L D T 1991a, 97. See labels nos. 27-32 in VAN

SOLDT

1989b.

any distinction at all: he decreed and proclaimed, the document was issued, and next it was presented or sent to the vassal king. Therefore, what really mattered was the point of view in which the text was couched (and accordingly sealed) rather than the formulae used. O n e should recall here the three duplicates of a text called by Nougayrol 'lettres-édits' (RS 17.130 and dupl.). T h e introductory formula ana R N qibīma, 'speak to R N ' , leaves no doubt about the epistolary character of the documents; the text, however, describes Hattusili Ill's regulations conccrning the business activities of the merchants of U r a in Ugaritic territory. T o the same domain of ambiguity belongs the suzerain's letter-verdict R S 17.133 sealed with queen Puduhepa's seal about a litigation over a shipwreck. Finally, RS 17.132 should be mentioned, a letter containing the early proposal from Suppiluliuma I to N i q m a d d u II for a military alliance which includes the legal stipulations following the description of the contractual agreement. 2.5.2

Form

In spite of these concurrent problems of definition, a Hittite tablet of legal content is in principle not very difficult to identify at first sight. O n e can even speak of prototype if by this we understand the more c o m m o n appearance of documents. This would be a relatively large tablet (larger than the prototype of Ugarit: the thickness averages some 40 mm) and 'cushion'-shaped showing a highly raised surface on the centre of the obverse where the deep impression of the stamp seal (in our corpus especially the T a b a r n a seal) is to be found. O t h e r types are of course attested; the above mentioned verdict and edict-letters, for example, bear the seal impression on a sort of appendix protruding from the left upper edge (or 'shoulder', as Nougayrol would put it) of the tablet. T h e exception to this kind of tablets are the duplicates of the treaty between Murši1i II and N i q m e p a ' (RS 17.79+ and dupl.), which possibly represent copies of an original metal tablet kept at Hattuša (in fact, some of the duplicates could have been made at Ugarit). Their shape is already unusual: they are extraordinarily large and flat on both sides; also exceptional is their fine script as well as the absence of any seal impression. As for the text, apart from the already mentioned c o m m o n opening letter-formula (in documents issued by the Great King we normally find umma Samši R N 'thus [says] "the S u n " R N ' followed by

his titles and filiation), one should mention the usual recourse to divine action to secure and guarantee the undertaken agreements. 2.5.3

Content

O f course the basic 'binding' text (Akk. tuppu ša nksi/úkilti) for the above-mentioned international state of affairs is the vassal treaty. T w o such texts have been found at Ugarit. O n e is the treaty between Šuppiluliuma I and N i q m a d d u II (RS 17.340, RS 17.369), and the other is the one already mentioned between Murši1i II and Niqmepa'. Like other Hittite vassal treaties, the stipulations contained in these texts concern the extradition of fugitives, the frontiers of the kingdom of Ugarit, military assistance and the vassal's loyalty. O t h e r documents complete this main agreement, for example the edict establishing the detailed inventory of the tribute due to the Hittite overlord (RS 17.227 and dupl., cf. also R S 17.380+) or the decree regarding extradition of fugitives concluded by Hattusili III (RS 17.238). Complaints of, or disputes between vassal states concerning one of these basic stipulations were arbitrated by the Hittite Great King himself; so the decision given by Murši1i II on the dispute of border territory between Ugarit and Mukiš (RS 17.62+ and par.) and between Ugarit and Siyannu (RS 17.235+ and par.), or also the above mentioned verdict promulgated by Hattusili III regulating the business activities of the merchants of U r a in Ugaritian territory. Problems concerning the dynastic succession of the kings of Ugarit also belonged to the jurisdiction of the Hittite overlord; so, for example, the case of divorce of 'Ammittamru II and the daughter of the king of Amurru, who was also niece of the Hittite king (RS 17.159, RS 17.365+), or the case of conspiracy against this same king apparently plotted by the sons of the queen of Ugarit (RS 17.35+-), both adjudicated by Tudhaliya IV.

2.6 2.6.1

The texts from Carchemish

Definition and scope

N o doubt, the role of the king of Carchemish as viceroy of the Hittite Great K i n g — a n d thus overlord—of the Syrian vassal states accounts not only for the considerable n u m b e r of documents issued from this court and found at Ugarit, namely about thirty texts and

fragments, but also their significant nature: most deal with inter-state arbitrations. Although a few are dated to Ta1mi-Teššub most of the material belongs to the reign of his father Ini-Teššub. Here also special reference should be made to one text that may defy classification. Like R S 17.132, R S 17.334 opens with the expression umma šarrima 'thus [says] the king' and contains the proposal from Šarru-kušuh to N i q m a d d u II for a military alliance against the king of Nuhašši and includes the resulting stipulations of the agreement. O f particular interest is the fact that this text is preserved in a copy sealed by Ini-Teššub of the original, broken document. 2.6.2

Form

Characteristic of the legal texts from Carchemish is their physical appearance. All arbitrations and most other deeds are written on oblong formatted tablets and the cylinder-seal is rolled in the middle of the reverse. O n the other hand, the text of verdicts begins invariably with the formula 'before P N ' or more often 'before R N ' (ana pāni RN). Only three records (RS 17.146, R S 17.230, R S 18.19) differ from this pattern and the three are shaped after the Hittite prototype (note that they are all deeds of agreement). T h e difference, however, is that the name and titles of the king of Carchemish are not introduced by umma. Arbitrations of lawsuits are phrased after a clear standard pattern. After the opening formula 'before R N / P N ' , the two litigant parties coming to court are n a m e d and the statement of one or both of them is presented, usually in subjective style. Next the verdict of the king is mentioned also in subjective style, followed by the result or execution of the royal decision phrased this time in objective style and in the past. T h e text ends with the clauses that forbid any future claims coming from either of the two parties and the final clause that states the irrevocable character of the document. Deities are invoked in only one of the deeds of agreement as guarantors of the transaction. 2.6.3

Content

By far the largest group of documents from Carchemish concern international arbitrations. Almost all these transactions are presided by king Ini-Teššub. A m o n g the parties involved one should mention the king of Ugarit and his sākinu as well as the whole community

of Ugaritian citizens. These verdicts concern different matters such as, for example, an action for a debt (e.g. RS 17.314, R S 17.346) or a complaint about the death of a prisoner (RS 19.63); but the most frequendy attested are cases of assault, theft and especially the murder of merchants in Ugaritic territory (e.g. RS 17.128, RS 17.145, R S 17.158, R S 17.234). This major problem is precisely the subject of the four preserved deeds of agreement between the states of Carchemish and Ugarit (RS 17.146, RS 17.230, R S 18.19, R S 18.115). O n the other hand, the role of the king of Carchemish as viceroy of the Hittite Great King over Ugarit is reflected in other inter-state arbitrations such as R S 17.59, in which Ini-Teššub presides and enacts Tudhaliya IV's exemption of 'Ammittamru II from his military obligations in the war against Assyria, or the same cases that were presided by the Hittite suzerain himself, such as the divorce between 'Ammittamru II and the daughter of the king of A m u r r u (RS 17.396, RS < 2 0 . > 1957.1) or the family conspiracy against the same king of Ugarit (RS 17.352).

2.7 2.7.1

The texts from Ugarit

Definition and scope

Scribes at Ugarit also drew up transactions undertaken by parties from different states and some of these documents (around ten) have been found at Ugarit. T h e problem, however, is again one of definidon, for a few of them are said to have been presided over by foreign officials. T h e fragmentary verdict R S 17.371+, for example, which is written by Nu'me-Rasap, the well-known scribe of domestic legal texts of Ugarit, is arbitrated by the kartappu of Carchemish. How, then, are we to classify these records? Admittedly these cases do not agree with the criterion stated above, namely correspondence between witnessing party and place of composition of the tablets. T h e r e seems to be little doubt that in these arbitrations the court of the overlord (Hatti or Carchemish) was transferred for some reason to the seat of the vassal (note e.g. that in R S 17.371+ one of the parties is the king of Ugarit); in such cases the overlord himself would then appoint at least one of his officials as authorized arbitrator. Following in a strict sense the tide of this chapter we have resolved to include these documents here.

2.7.2

Form

T h e main physical characteristics of documents composed at Ugarit have already been described above. It is interesting to observe that R S 16.170 in which the king of Ugarit acts as overlord is shaped and styled in exactly the same way as any other royal legal document (one should further note that this document was found a m o n g the royal documents in the palace central archive). By contrast, the documents that describe acts presided over by a superior, foreign authority at Ugarit significantly show a distinctly eclectic aspect. For example, R S 18.02 in which the king of Ugarit redeems certain people from the Hittite priest of Istar of Zinzaru, is shaped after the Carchemish fashion (i.e. an oblong tablet) and sealed with the Hittite stamp seal of the priest in the Hittite fashion (i.e. in the middle of the obverse); nevertheless, since the scribe is N u ' m e - R a s a p the text clearly follows the local form of Ugarit (as in, for example, the opening phrase 'From to-day'). 2.7.3

Content

In spite of its fragmentary state, R S 16.170 describes in all likelihood a verdict of king N i q m e p a ' concerning the long-standing conflict of border territory between the kingdom of Ugarit and Siyannu, then its vassal state, thus previous to its defecdon to the king of Carchemish. O n e tablet that belongs to the large dossier of the divorce of king 'Ammittamru II and the daughter of the king of Amurru (RS 16.270) bears the impression of two seals of 'Ammittamru, the dynastic seal and his personal one, and seems to have been written by a scribe of Ugarit. As for the arbitrations of foreign commissioners at Ugarit, RS 17.371+, which is somewhat damaged, probably refers to a litigation involving the merchants of Ura, and R S 18.02, as already stated, is a deed of redemption. O t h e r deeds written at Ugarit record private international transactions. For example, R S 16.180 and R S 17.251 are deeds of sale; the former records the sale of a horse from the huburtanūru of the king of Carchemish to the king of Ugarit (note the possible badly preserved Ugaritic summary at the end of the text), 14 and the latter, written by Burqānu, describes the sale of an individual by two Hittite brothers to the sākinu of Ugarit. 14

See

MARQUEZ

ROWE

1996a, 457f.

2.8 2.8.1

The texts from Amurru

Definition and scope

A few legal texts written by scribes from the chancellery of A m u r r u have been found at Ugarit. T h e two matters that are recorded reflect on the one h a n d the neighbouring situation of both kingdoms and, on the other, the intermarriage that was relatively frequently practised between both dynasties. T h e former is represented by one text, R S 19.68, a treaty between N i q m a d d u II and Aziru, which is possibly one of the oldest agreements—if not the oldest—recovered from the royal palace archives. T h e latter is constituted by four documents that belong to the large dossier concerning the divorce ease between ' A m m i t t a m r u II and the daughter of Bentešina and sister of Šaušgamuwa. 2.8.2

Form

Some formal features of the treaty R S 19.68 have led to the origin of the tablet being called into question. Indeed, the mention throughout this parity agreement of the royal n a m e N i q m a d d u in first position as well as the interference of subjective style within the text phrased from the viewpoint of the Ugaritic king could suggest that the text was composed by a scribc of the royal court of Ugarit. O n linguistic and epigraphic grounds, however, it seems more likely that A m u r r u was the original provenance of the scribe; 15 on the other h a n d , the seal impression at the head of the text corresponds to the cylinder seal of Aziru. T h e later documents of his descendant Šaušgam u w a show the strong Hittite influence that developed through the years. Indeed, not only are the tablets in the 'cushion' shape but, more significandy, the cylinder seal of Aziru gives way to Sausgamuwa's stamp seals inscribed with Hittite hieroglyphs, of which the impression can be found in the middle of the raised surface of the obverse of the tablets (so in 'tablette G. Badr',' 6 R S 17.228, R S 17.318+, and R S 17.360a+, which also bears the cylinder seal impression of Aziru on top of the obverse). As for the text, the scribes of the chancellery of A m u r r u , like the ones of Ugarit, used the formula 'from to-day' to open the d o c u m e n t

15

See

16

Edited by

NOUGAYROL

1956,

ARNAUD

281,

- SALVINI

and

IZRE'EL

1991-2.

1991,

21.

(RS 17.228, R S 17.360A+, R S 19.68). Divine curses are attested only in the treaty R S 19.68. 2.8.3

Content

T h e treaty between N i q m a d d u II and Aziru describes the latter's r e n o u n c e m e n t of the f o r m e r claims of A m u r r i t e rulers to adjacent Ugaritian territory (including Siyannu); the text declares that Aziru accordingly received satisfaction, namely 5,000 shekels of silver. Furt h e r m o r e , military aid f r o m the A m u r r i t e king in case of an e n e m y attack against his n o r t h e r n neighbour is stipulated. T h e divorce between 'Ammittamru II and the daughter of Bentešina obviously provoked a serious international diplomatic conflict. This, as we have seen, is well reflected not only by the n u m b e r of records that were needed to regulate the affair but also by the authorities w h o h a d to supervise it, such as the Hittite e m p e r o r (who was in turn the divorcee's uncle) or his viceroy in Syria, the king of C a r c h e m ish. T h e documents that were issued a n d sealed by the king of A m u r r u a n d b r o t h e r of the divorcee describe his obligations in the agreem e n t : 'tablette G . B a d r ' concerns her expulsion, a n d R S 17.228, R S 17.318+, a n d R S 17.360A+ concern her final extradition (for execution) a n d resulting compensation of 1,400 shekels of gold.

2.9 2.9.1

The texts from Siyannu

Definition and scope

O n l y two legal d o c u m e n t s stemming f r o m the court of Siyannu have been f o u n d at Ugarit. T h e y are both related to the long-standing disputes of b o r d e r territory between both states (see above the verdict of king N i q m e p a ' or the final reguladons decreed by the e m p e r o r Mursiii II). R S 18.01 was issued by king Padiya a n d R S 17.123 by king 'Abdi-Ninurta. 2.9.2

Form

T h e physical a p p e a r a n c e of both tablets is hardly distinguishable f r o m the Ugarit prototype (possibly as a result of political influence). T h e cylinder seal impressed by Padiya at the head of the d o c u m e n t is that of his predecessor, Sassi (used thus as a 'dynastic seal'); c Abdi-

Ninurta, in turn, rolled another, anepigraphic one. Also like the documents composed at Ugarit, they are introduced by the opening phrase 'from to-day' and they are couched in objective style. 2.9.3

Content

Both deeds concern real estate property. More precisely, they describe the final result and state of affairs after litigation concerning the ownership or jurisdiction of land that belonged to frontier districts between Ugarit and Siyannu. This is explicitly mentioned in R S 17.123 in which the arbitrator was the Hittite overlord. T h e districts involved in this case are Šukši and H a r m a n a . Vineyards of the former frontier district are also the object of partition in R S 18.02, this time between the religious associations (marzihu) of Ugarit a n d Siyannu. 2.9.4

Publication and studies

T h e bibliographical picture is almost the same as we have seen for the domestic texts. T h e great majority of the international material was magisterially published by Nougayrol in two volumes of the French Mission de Ras S h a m r a : N o u g a y r o l 1956 and 1970. T h e former contains most of the documents as well as an excellent study of them. O t h e r scattered texts have been published by F i s h e r (1971), Malbran-Labat

( B o r d r e u i l et al.

1991, no.

1) a n d A r n a u d

-

Salvini

(1991-2), and a few still remain unpublished. T h e treaty between Murši1i II a n d N i q m e p a ' was thoroughly analysed by d e l M o n t e (1986), and B e c k m a n (1996c) has recently provided us with a h a n d b o o k of Hittite international documents. Mention must also be m a d e of two other m o n o g r a p h s that offer a c o m p r e h e n s i v e picture of the international relations as described a m o n g others by the documents of Ugarit: K e s t e m o n t 1974a (more restricted to the juridical aspects) a n d L i v e r a n i 1990.

3

T h e

Ugaritic

Legal

Texts

Since Akkadian was the language par excellence of legal documents it is hardly surprising that this genre is far less represented in the Ugaritic corpus. This is illustrated, for example, by the fact that the chapter of legal texts in the standard edition of this corpus (indexed

with the digit ' K T U 3.') runs to barely six pages. 1 ' Indeed, we should regard these few legal documents written in alphabetic cuneiform as exceptions. Probably on this account and also because the phraseology of some of the d o c u m e n t s was obviously inspired by the Akkadian pattern, it has been suggested that these texts could be translations of Akkadian originals. 18 This seems to be true of the one international document, namely the edict that establishes the inventory of the tribute due to the Hittite overlord. However, this is not so obvious for the rest of the material, not even for royal documents that clearly show a parallel phraseology. 19 Indeed, concerning the latter, how could one explain that the so-called 'translated copies' were found a m o n g other original Akkadian documents in the palace central archive? And where, then, were the Akkadian 'originals' accordingly kept which, on the other hand, have so far not been found? In fact, it seems reasonable to ask why scribes who were admittedly biscriptal would not have written, if only exceptionally, legal texts in Ugaritic (note that Burqānu, for example, drew up at least three documents in Akkadian [RS 8.145, R S 25.137 and R S 17.251] and one in his own tongue [ K T U 3.8 = R S 19.066]). 20 T h e fact that these texts are written in Ugaritic implies that they were most likely drawn u p at Ugarit. In this chapter we shall follow the plan proposed for the Akkadian corpus. Accordingly we shall divide Ugaritic legal documents into two sub-genres: on the one hand, the domestic texts and, on the other, the one international document.

3.1

The domestic legal texts

All Ugaritic legal texts but one deal with domestic transactions. According to the nature of the witnessing party these documents can be classified into two categories, the same ones by which we classified the Akkadian domestic texts, namely the royal and the non-royal documents.

17

One should note that in spite of the corrections and additions to this edition LORETZ SANMARTÎN 1 9 9 5 ) that will be pointed out throughout this paper this preliminary appreciation is not significantly changed.

(DIETRICH 1R

19

See

KIENAST

1979, 433,

444,

or

LIBOLT

1985,

359.

For the parallel phraseology see e.g. N O U G A Y R O I . 1962, 29, R A I N E Y 1969, 132ff., K I E N A S T 1979, M I L A N O 1980, 182ff. 20 Of course one may still question with VAN S O I . D T (1991a, 2 7 n. 213) whether this text is not a Ugaritic copy of an Akkadian original.

3.2 3.2.1

The royal legal texts

Definition and scope

As defined above, this group consists of those legal texts in which the king of Ugarit presides, explicitly or not, over the transactions involved. These royal documents (in Ugaritic: spr mlk as attested in K T U 2.19:13 21 include four published examples ( K T U 2.19 = RS 15.125, 22 K T U 3.2 = RS 15.111, K T U 3.4 = R S 16.191+ and K T U 3.5 = RS 16.382) and one unpublished text (RS 94.2168) found in the recently excavated archive of Urtenu. From this same archive came another unpublished record (RS 94.2965) arranged similarly, and containing clauses typical of royal legal documents; although no reference of the witnessing party is made and no royal seal is impressed (nor is there a space left free on the tablet for sealing) it is possible to understand implicitly that the king presided over this transaction. (One should note that R S 94.2168 also does not bear the royal seal impression: were these two documents actually private 'copies'?) Furthermore, an unclassified fragment ( K T U 7.63 = R S 15.117) that has often been taken to list the royal tides of N i q m e p a ' could also belong under this category. 23 3.2.2

Find-spot

As already mentioned, one or two royal documents have been found outside the royal palace, namely in Urtenu's house. T h e other four deeds (as well as the fragment K T U 7.63 = RS 15.117) come from the palace archives. O f interest is the fact that these four tablets come entirely from the same concentration, the so-called palace central archive where, as we have seen, most Akkadian royal domestic legal documents were kept ( K T U 7.63 was discovered in the palace eastern archive).

21 V I R O L L E A U D ( 1 9 5 7 , 15FF.) used the designation 'actes royaux' differently, for he also included royal letters. 22 Although still included under the epistolary category with the resulting digit in D I E T R I C H LORETZ S A N M A R T I N 1 9 9 5 , 1 6 8 , it is now generally agreed that this fragment is of legal content (see already K I E N A S T 1 9 7 9 , 4 3 3 , 444ff). 23 Note that VAN S O L D T (1991a, 5 6 4 ) suggests cataloguing the text as epistolary or legal.

3.2.3

Form

We have referred to the fact that royal legal texts drawn up in Ugaritic reflect, and presumably imitated the Akkadian pattern. Actually it is reasonable to assume that the same scribes shaped and wrote both the Akkadian and Ugaritic tablets. For example, as regards the shape of tablets, it is interesting to observe that although written in the shorter alphabetic script Ugaritic royal documents have the same average measurements attested for their Akkadian counterparts (e.g. K T U 3.5 = R S 16.382 measures 90 x 70 x 35 mm). T h e sealing practice is also identical: the royal seal impression is placed on top of the obverse and could be divided from the text by a ruling. Note that the dynastic seal of the kings of Ugarit was rolled on K T U 3.2 a n d K T U 3.5, whereas the ring-seal of N i q m a d d u (presumably N i q m a d d u II) was stamped twice in order to fill the sealing space in K T U 3.4 (although not preserved, K T U 2.19 was certainly sealed by the king as stated in 11. 6ff.).24 As for the text, the phraseology clearly reproduces the Akkadian formulae. Thus, the text begins with the temporal expression 'From to-day' (I ym hnd), which could also be combined with the closing expression '(and) for ever' ('d 'Irri). T h e first section of the text describes the completed part of the transaction, obviously phrased in the past (perfective) and in objective style. T h e king appears in first position either as active party (so in K T U 3.2 and K T U 3.5) or as having presided over the transaction (so in R S 94.2168, omitted in K T U 3.4), introduced by the preposition 'before' (/ pn—perhaps to be restored before the royal n a m e in K T U 7.63:1). Next follow the final clauses that deal with the obligations of the transaction, phrased this time in the future (imperfective) and also in objective style. Although the seal identification clause is only preserved in K T U 2.19, its restoration at the end of texts K T U 3.2 and K T U 3.5 seems very plausible. Perhaps the same holds true for the name of the scribe which is not attested in these texts. T h e scheme of the Ugaritic royal documents is thus not very different from their Akkadian counterparts. 3.2.4

Content

T h e contents of these texts is as follows: K T U 3.2 = R S 15.111 and K T U 3.5 = R S 16.382 are both deeds of royal gift; R S 94.2168 24

As shown after collation, the upper part of the fragment K TU 7.63 is lost.

concerns the rights of succession of c Abdi-milku's children and in particular the inheritance of the property he received from the king; and R S 94.2965 lists estate property acquired by a Yabninu together with the rights and obligations b o u n d to it. K T U 2.19 = RS 15.125 is a deed of franchise (note the Ugaritic designation spr d tbrrt in lines 9-10), namely from the ?7/t«-service (called in Ugaritic unt). T h e final document, K T U 3.4 = RS 16.191+, concerns the redemption of seven persons from the Berutians by one Iwrikili for 100 shekels of silver and the resulting obligations imposed on them. As for the parties involved, 'Ammittamru II is the donor king of K T U 3.2 and K T U 3.5 as well as the donor and guarantor of the rights of succession in R S 94.2168, and N i m q a d d u (II/III) is the grantor of privileges in K T U 2.19. Unfortunately it is difficult to idendfy the other contracting parties due to the want of more solid prosopographical evidence. Still, it is possible to relate some of the Ugaritic documents to the Akkadian 'dossiers'. For example, amtm, the donee's father in K T U 3.2, could be identified with the wellattested Amutarunu. And the c Abdi-milku whose inheritance is the topic of RS 94.2168 is very likely the homonymous main party of one of the largest dossiers identified by Nougayrol; moreover, we learn from this Ugaritic text that he was married to the king's daughter. This explicit information supports the assumption that many of the beneficiaries of these dossiers were members of the royal family as well as court officials (with this regard, one should note that the protagonist of the fragment K T U 7.63 = RS 15.117 is the sākinu of the palace and door-keeper of the king). 3.2.5

An example

In order to illustrate the similarities in the scheme and form between these Ugaritic documents and their Akkadian counterparts, we present here in transliteration and translation the royal document K T U 3.5 = R S 16.382: Impression of dynastic seal

5

I ym . hnd 'mttmr . bn nqmpc . mlk ugrt . ytn šd . kdgdl

From today 'Ammittamru, son of Niqmepa', king of Ugarit, has given the field of P N h

uskny . d . b šd[y] Lo.E.

10 Rev.

15

20

x m χ χ . [y\d gth yd . zth . yd . [A] rmh . yd [k]lklh w ytn.nn I . b'ln . bn kltn . w . 1 bnh . 'd [.] 7m Ihr . elmt bnš bnšm I . yqhnn . bd bcln . bn . kltn w . bd . bnh . ed 'lm . w . unt in . bh

from TN ι, (which is located) in the territory of TN 2 , together with its factory, and its olive grove, and its vineyard, and whatever belongs to it, and he has given it to PN2, son of PN:), and to his sons for ever. In the future no one will take it from the hands of PN 2 , son of PN3, and from the hands of his sons ever. And there is no service on it.

(end broken off)

3.3 3.3.1

The non-royal legal texts

Definition and scope

W e include under this second group of domestic legal texts those records which describe transactions performed before witnesses (Ugaritic yph) and that were not presided by the king. Such definition immediately excludes from our corpus texts like K T U 3.3 = R S 15.128, a bill concerning personal security, K T U 3.7 = R S 18.118, a list of Egyptian persons entering the ilku (Ugaritic «n/)-service, 25 and K T U 3.10 = R I H 8 4 / 3 3 , a list of debtors, 26 in which no mention of a body of witnesses is made. O n the other hand, K T U 3.6 = R S 17.[468] is too fragmentary for classification. 27 As a result, only two other K T U 3. series texts can be included under this category, namely K T U 3.8 = R S 19.066 and K T U 3.9 = R S [Varia 14] = R S 1957.702, which do list the witnesses of the transaction. In addition,

25

See more recently M A R Q U E Z R O W E 1993b. Note that it partially duplicates K T U 4.791 (see B O R D R E U I L 1995a, 4-5) listed thus within the administrative records; other debts lists are e.g. K T U 4.123 = RIH 8 4 / 0 4 obv., K T U 4.283 or K T U 4.310. 27 Note that even D I E T R I C H - L O R E T Z - S A N M A R T I N ( 1 9 9 5 , 2 0 2 ) admit that its content is 'uncertain'; the presence of the sequence (> yph [?]) at the end of 1. 2 is in a completely fragmentary and obscure context. 26

reference should be made to other texts that have been classified as administrative and could in fact lie on the borderline of the legal domain. Like the above mentioned Akkadian text RS 16.354, K T U 4.386 = RS 18.111 (restoring [y]ph at the beginning of 1. 3 and also possibly [yp]h at the beginning of 11. 7 and 9), K T U 4.632 = R S 19.073, K T U 4.778 = R I H 8 3 / 1 2 with its duplicate K T U 4.782 = R I H 84/08, 2 8 K T U 9.477 = R I H 8 3 / 3 9 + 2 9 and perhaps also K T U 4.258 = RS 17.007+ (sec 1. 5) record a series of debt-notes together with the witnesses involved. As a matter of fact, the explicit mention of witnesses is what distinguishes these texts from other debtnotes tablets like K T U 3.10. Significantly, the same holds true for K T U 3.8 vis-à-vis K T U 3.3. Both are bills of personal security and use a similar phraseology (also parallel to the Akkadian counterparts RS 15.81 or RS 16.287 referred to above). 30 However, as already stated, neither the latter nor the Akkadian texts mentions any witnesses. 31 Stronger doubts about the legal nature of K T U 3.8 and also K T U 3.9 have been expressed basically because both lack seal impressions and do not use phrases typical of legal texts (like the introductory formulae 'from to-day' or 'before witnesses'). As a result, it has been suggested that these texts should rather be identified as 'memoranda'. 3 2 Despite the doubts of classification, which may indeed question once more the accuracy of our definitions, both documents will be included under this category following thus our basic premise, viz. the more or less explicit mention of witnesses in the texts. 3.3.2

Find-spot

Very little can be said about the find-spots of K T U 3.8 and K T U 3.9. T h e former was found in the open space between the royal palace and the so-called 'southern palace'. As recently pointed out,

28 See B O R D R E U I I , 1987, 29 ΗΓ. Note that some of the debt-notes in this text do not mention the witness. 29

30

See

B O R D R E U I L et at.

1984,

430.

See M I L A N O 1980, 186ff. and H O F T I J Z E R - VAN S O L D T 1991a, 189ff. 31 The list of persons on the reverse (11. 10ff.) correspond to the sureties (cf. 1. 1 : spr. 'rbnm, see most recently H O F T I J Z E R VAN S O L D T 1 9 9 1 , 1 8 9 ) and not to the body of witnesses (as stated by M I L A N O [ 1 9 8 0 , 1 8 8 ] ) . 32 So K I E N A S T 1979, 432f, with critical comments upon the former inclusion of this and other K T U 3. series texts within the category of 'Verträge' in D I E T R I C H LORETZ SANMARTIN 1976a (see also S Z N Y C E R 1979, 1 4 1 8 ) .

tablets unearthed in this area 'could have come from different parts of the surrounding buildings', 33 which means either from the palace or from a private house. As for the latter, it is almost certainly from illegal excavation at Ras Shamra. 3 4 It is not without interest that all witnessed debt-notes come from the palace archives and that two of them (the two duplicates) were found in the palace at Ras Ibn Hani. 3.3.3

Form

We have already mentioned that K T U 3.8 and K T U 3.9 lack some of the characteristic features of legal documents. T o begin with, neither of them bears a seal impression. O n the other hand, the formulas that usually introduce a legal text, like 'from today' or the one attested in the Akkadian counterparts 'before witnesses', are absent. Nevertheless, examples with these same atypical features can be found within the Akkadian corpus (compare also the Ugaritic royal document R S 94.2965). R S 11.856, for instance, describes the transaction without any such introductory formulas, lists the witnesses at the end of the text after a ruling and bears no seal impression. These are exactly the same characteristics and scheme as in K T U 3.8 (significantly, the unusual shape of both tablets is similar). 35 As for the text itself, K T U 3.8 follows the general pattern of legal documents. It is phrased in objective style, the completed part of the transaction is in the past (perfective) and the final clauses in the future (imperfective). After listing the three witnesses, the name of the scribe is mentioned at the very end of the document. T h e text of K T U 3.9, on the other hand, is exceptional in that it combines both objective and subjective style, an unusual feature in domestic contracts. T h e names of two witnesses close the document. 3.3.4

Content

K T U 3.8 is a deed of security: five persons assume surety for two debtors. K T U 3.9 concerns the inauguration of a religious associaV A N S O L D T 1991a, 146. Note that it has recently been possible to trace back the find-spot of another Claremont Ras Shamra tablet, namely from the palace archives ( A R N A U D - SALVINI 1991-2, 8 η. 5). 35 Viz. they are small and nearly square: K T U 3.8 measures 48 X 42 X 28 and RS 11.856, 40 χ 40 χ 22 mm. 33

34

tion (mrzh) by a Š a m u m a n u and stipulates the binding obligations for both its founder and leader and the rest of its members.

3.4 3.4.1

The international legal text

Definition and scope

It is hardly surprising to find only one Ugaritic text, K T U 3.1 = RS 1 1.772+, under this category since Akkadian was not only the legal but also the diplomatic language par excellence at that time in the N e a r East. As a matter of fact, what is truly surprising is precisely the very existence of one d o c u m e n t written in alphabetic cuneiform that describes a transaction between two states, namely Hatti and Ugarit. T h e discovery in the palace southern archive of the Akkadian parallel text, namely the edict that establishes the inventory of the tribute of N i q m a d d u II owed to his overlord Suppiluliu m a I (RS 17.227 and dupl., see above § 9.2.5.3), clearly showed that we are dealing here with an Ugaritic version. 36 This interpretation, however, has been recently challenged arguing that the language and formulae of K T U 3.1 do not correspond to a legal document but, rather, that they reveal the epistolary nature of the text. 37 This opinion, however, takes into consideration neither the ambiguous language used by Hittite scribes (as discussed above under § 9.2.5.1), also present in the Akkadian original edict, nor proper Ugaritic epistolary structure. 38 3.4.2

Find-spot

T h e different fragments of this tablet were found scattered within and around the palace western archive. O f interest is the fact that the Akkadian tribute list RS 11.732[β] was discovered at this same archive and that the content thereof is predominantly administrative. O n this account, it has been reasonably suggested that the Ugaritic version of the edict was kept there for administrative purposes. 39

36

See e.g. 1990a, 354ff. 37

38 39

NOUGAYROL

1956, 37ff.,

DIETRICH

LORETZ

K N O P P E R S 1993, followed by T R O P P E R 1995b, 235. See also the remarks of P A R D E E ( 1 9 9 8 ) . See VAN S O L D T 1991a, 57-8.

1966a, or

VAN

SOLDT

3.4.3

Form

As already m e n d o n e d , the text is preserved from several fragments (RS 11.772+11.780+11.782+11.802). T h e original tablet must have been relatively large. Unlike the Akkadian duplicates, however, K T U 3.1 is not 'cushion'-shaped and its width, namely 30 m m , and rather flat surface seems to rule out a prion the original presence of the Hittite seal impression in the lost middle part of the obverse. T h e scheme of the text corresponds to the Akkadian original, the first preserved paragraphs to the historical introduction, and the following to the detailed definition of the tribute. Although not preserved, the end of the text may have contained the invocation to the gods as witnesses and guarantors of the agreement, as in the Akkadian text. Finally, it has been pointed out that the order of some of the tribute items as well as that of the Hittite officials follows the text of the treaty of Murši1i II (RS 17.380+) rather than the edict of Suppiluliuma I.40 3.4.4

Content

T h e historical introduction so typical of Hittite edicts and agreements relates how N i q m a d d u II refused to take part in the antiHittite coalition and thus stresses his loyalty towards his overlord Suppiluliuma I. T h e text then describes in detail the inventory of the tribute that as a result the king of Ugarit owed to the Great King, the queen his wife, as well as his high officials.

4

The

Hittite

Legal

T e x t

T h e excavations at Ras Shamra have so far yielded only two cuneiform texts written in the Hittite language. O n e is part of a trilingual literary composition (RS 25.421+) and the other is a complete legal document (RS 17.109). T h e latter, with which we are here concerned, is thus unique in that it is the only Hittite original document found at Ugarit. Shortly after its discovery, Hittitologists, too, agreed with the uniqueness of RS 17.109 for not only was this tablet the first Hittite legal document discovered outside the Anatolian core of 40

See

VAN S O L D T

1990a, 341.

the Hittite empire but also the only source which could shed light on the form of Hittite private legal practice.

4.1

Definition

R S 17.109 was catalogued by E. Laroche with the n u m b e r C T H 296 as one of the Hittite 'procès' within the brief chapter of law. H e titled it 'témoignage écrit de Pallariya'. Indeed, the text describes the tesdmony of a witness of an inter-state transacdon, namely between Hatti and Ugarit; the deposition is made before two witnesses and properly authenticated by means of a seal impression. O n epigraphic, linguistic and also prosopographic grounds, the tablet can be safely dated to the second half of the 13th century bce, and thus it fits into the general chronology of the Ugarit archives. O n the other hand, the fact that the text is written in the classical script of Bogazköy does not necessarily mean that the tablet was drawn up at the Hittite capital. As we have already seen, it is possible, and in this case even probable, that Hittite officials would arbitrate international affairs in foreign states or courts like Ugarit.

4.2

Find-spot

T h e tablet was found in the palace southern archive. This find-spot agrees well with the nature of RS 17.109 since, as we have mentioned in several occasions, this royal archive was meant to keep the legal documents concerned with foreign affairs.

4.3

Form

T h e shape of RS 17.109 is rather peculiar. In fact, it belongs to a category of Hittite tablets we have already seen above (§ 9.2.5.2). T h e tablet bears a kind of appendix that protrudes from the middle of its upper part, on top of which the stamp-seal impression is to be found. Also typical of Hittite scribal practice (and of deposition documents), the text opens with the word umma (here as an Akkadogram). In this case it introduces the statement of one of the parties involved. T h e n his witness declares, also phrased in subjective style and in the past. Next follow the clauses that settle the affair and validate the written document, mainly provided for the other, absent party.

After a ruling, the mention of the two witnesses (of whom one probably acted as judicial officer) close the text.

4.4

Content

As catalogued by Laroche, R S 17.109 is a deed of deposition. It contains the declaration of Pallariya, the witness of a transaction, namely the receipt of a large a m o u n t of silver by the Hittite tax collector from the sākinu of Ugarit. T h e same state officials, especially the latter, are often protagonists in other international legal cases that were most often arbitrated, as we have seen above (§ 9.2.6), by the king of Carchemish.

4.5

Publication and studies

T h e tablet was published in transliteration and translation by L a r o c h e (1968b no. 1); recently, S a l v i n i (1995a) has provided us with the copy and photos, and also a few reading improvements. Apart from Laroche's first interpretation, one should also mention the contributions by H a a s e (1971), v o n S c h ü l e r (1971) and M a r q u e z R o w e (1999).

T H E

E C O N O M Y

Michael

O F

U G A R I T

H e l t z e r

T h e economy of the kingdom of Ugarit is considered here under three headings: the administrative texts, commerce and crafts and industries.

1

T h e

Administrative

1.1

Texts

General

T h e administrative texts written in both Ugaritic and Akkadian belong to the time span from the middle of the 13th century bce to the destruction of the kingdom in about 1180 bce or a few years later. It is impossible to separate the Ugaritic and Akkadian texts, for they deal with the same issues, and often the same persons are mentioned in the texts in both languages. It is also impossible to determine the exact n u m b e r of the texts, for (a) there are hundreds of tablet-fragments, which rarely can be used to explain anything; (b) many texts and text fragments in the administrative economic field are still unpublished. 1 T h e r e is also the problem of dividing the texts according to content. Some of the administrative texts are economic, others are agrarian, etc. In all, we possess approximately 550 clearly and fairly legible texts of this kind in Ugaritic and 160 in Akkadian, a ratio of 7:2. We can divide these texts, albeit not absolutely, into primary tablets made on the spot or at the time of the administrative (economic or fiscal) act, and tablets which are monthly or even annual summaries.

' For the overall number of the texts and their editions, cf. TEO. The Akkadian texts are generally cited according to their publication in PRU 3, 4, 6 and Ug 5. Other occasional publications are quoted separately. On unpublished texts cf. KTU2, 9.

In some instances it is also difficult to draw a dividing line between administrative and commercial texts (in cases when the prices of the goods are not given). W e also include here the texts excavated in the territory of the kingdom of Ugarit (Ras-ibn Hani (RIH)). In general, all these texts reflect the structure of Ugaritic society, which was divided into two sets. O n e comprised the 'sons of Ugarit' (Akk. mârēM Ugarìt), i.e. the villagers, who had a certain communal organization ( H e l t z e r 1976). T h e other comprised the 'servants of the king' (Ug. bnš mlk, Ugaritic Akk. bunusu milki, Akk. ardēU šarri) or royal servicemen, who received various deliveries for their service from royal stores and also, for performing their duties, land allotments on c o n d i t i o n a l h o l d i n g ( H e l t z e r 1982). T h i s social stratification is known from the text PRU 4, 107 = R S 17.238, which also mentions the 'servants of the servants of the king' (Akk. ardeU¥& ardêm'& šarri). From this text, as well as from the administrative texts, it is clear that some of the royal servicemen were in a position to have underlings. T h e royal servicemen were organized into professional groups and at the head of every group stood a royal nominee-official or rb (lit. 'great', 'big'). They performed their duties (taxes, conscription and corvées). 2 T h e various professions of craftsmen, agricultural workers, different military groups, priesdy professions, officials and palace personnel were organized in groups sometimes of ten.

1.2

Classification of the administrative texts

T h e r e is some difficulty in classifying the administrative texts as they are largely of mixed contents, but in general it is possible. 1.2.1

Lists of villages

We shall begin with the lists of villages (Ug. qrt, Akk. âlu). This category contains the texts which merely provide the names of villages (since the numbers which follow them are sometimes broken, as are the first and last lines, where the purpose of a text was given). 3 T h e y can be further classified as:

2

Therefore it is impossible to define them as 'guilds', as this would presuppose at least some degree of internal self-government which is not apparent in the bnl mlk texts. 3 K T U 4.94 = RS 11.832; K T U 4.303 = RS 17.370[B]; K T U 4.365 = RS 18.073; K T U 4.414 = RS 18.251 [c]; K T U 4.621 = RS 19.048[A]; K T U 4.622 =

a) Lists of mobilization of villagers for military purposes (or sometimes for the corvée); sometimes arms were delivered to them: 'bows' (qstm), 'shields' (Akk. gabābūMEè), 'slings' (qlcm) and 'spears' (mrhm).4 b) A list of payments of tribute in silver to the Hittite king (professional groups of bnš mlk also appear in this text, K T U 4.610 = R S 19.017). c) Payments by the villagers—not individually but as a unit—of silver, wool, cereals, olive oil, wine and cattle to the royal treasury. 3 d) Tablets recording the distribution of 'food' (akl) or 'rations' (hpr) to the villages in corpore at the time of performing their corvée and other texts where the villages are listed ( K T U 4.41 = RS 8.280). 1.2.2

The gt

W e learn from the administrative texts that there was a well developed royal economy in the kingdom ( H e l t z e r 1979a). It was divided into units (Ug. gt, Akk. dimtuf which were dispersed over all the kingdom. 7 Here too certain groups of texts shed light on the economic activities on the gt's. Texts include:

19.048[B]; K T U 4.629 = RS 19.061; K T U 4.661 = RS 19.167A: K T U 4.676 = RS 19.174i; K T U 4.684 = RS 19.257; K T U 4.685 = RS 19.258; K T U 4.686 = RS 19.259; K T U 4.693 = RS 20.157; K T U 4.770 = RIH 77/27; PRU 6, 97 = RS 19.118; PRU 6, 169 = RS 18.279; Ug 5, 102 = RS 20.207A; 103 = RS 20.143B; 104 = RS 20.144; etc. 4 K T U 4.68 = RS 11.716; PRU 3, = RS 11.841; PRU 6, 131 = RS 19.35A; cf. also K T U 4.63 = RS 10.052 where mobilized people of various villages receive their arms; PRU 6, 95 = RS 19.74; K T U 4.683 = RS 19.256; and Bordreuil 4 = RS 34.131 etc. 5 K T U 4.73 = RS 11.724+; K T U 4.93 = RS 11.776 + 11.800; K T U 4.95 = RS 11.836+; K T U 4.108 = RS 12.003+; K T U 4.113 = RS 12.018; K T U 4.232 = RS 16.355; K T U 4.267 = RS 17.103; K T U 4.308 = RS 17.386; K T U 4.553 = RS 18.[479]; K T U 4.609 = RS 19.016; K T U 4.611 = RS 19.018; PRU 3, 194 = RS 11.830; PRU 3, 191 = RS 11.841 [a]; PRU 3, 191 = RS 15.020; PRU 3, 191 = RS 15.179: PRU 3, 192 = RS 15.183; etc. PRU 6, 104 = RS 19.43; PRU 6, 105 = RS 19.117; PRU 6, 106 = RS 19.119; Ug 5, = RS 19.80; Ug 5, 111 = RS 19.129; Ug 5, 113 = RS 19.26; Ug 5, 187 = RS 19.27 (although it is not always certain whether these are payments or deliveries). 6 T h e literal meaning of gt is 'oil or wine press' ( H E L T Z E R 1979b). 7 T h e texts concerning the more than 70 royal gts are: K T U 4.89 = RS 11.796; K T U 4.96 = RS 11.840; K T U 4.97 = RS 11.841; K T U 4.103 = RS 11.858:45; K T U 4.110 = RS 12.006; K T U 4.122 = RS 13.012; K T U 4.125 = RS 14.001; K T U 4.139 = RS 15.017; K T U 4.141 = 15.022+; RS 4.142 = RS 15.023; K T U 4.175 = RS 15.096; K T U 4.200 = RS 15.189; K T U 4.213 = RS 16.127; K T U 4.243 = RS 16.395; K T U 4.271 = RS 17.115; K T U 4.296 = RS 17.031; K T U 4.297 = RS 17.326; K T U 4.307 = RS 17.384; K T U 4.313 = RS 17.399; K T U 4.320 = RS 17.444: K T U 4.345 = RS 18.033; K T U 4.358 = RS 18.048; K T U

a) T h e tablets concerning villagers, villages and royal service people whose service fields were in the area of a certain gt and connected with it, who had to deliver their share of their own produce to the gt,a where it was processed. b) Texts concerning stocks of agricultural tools on various gts.9 c) Lists of the state of cattle on the gts. These include beasts of burden (oxen and donkeys), cattle for fattening (mrì) and poultry, 10 listed together with the working teams. d) Texts concerning agricultural products (cereals, wine, oil, etc.) which are at the gt, including fodder and products delivered by the villagers and (non-agricultural) craftsmen to the gt stores. Sometimes the name of the month is given, in which case these are monthly reports." e) Tablets concerning 'royal servicemen' (bnš mlk), who had agricultural professions: 'ploughmen' (hrtm),12 'vine dressers' (gpnym), '(sheep-) shearers' [gzzm), 'shepherds' (nqdm and r'ym), and other personnel who were partly referred to as ' m a n of the gf (bnš gt) and 'belonging to the gf (bdl gt).n 4.365 = RS 18.073; K T U 4.380 = RS 18.099; K T U 4.382 = RS 18.106+; K T U 4.386 = RS 18.111; K T U 4.397 = RS 18.136; K T U 4.400 = RS 18.139; K T U 4.405 = RS 18.143[c]; K T U 4.409 = RS 18.146 bis; K T U 4.424 = RS 18.296; K T U 4.618 = RS 19.045; K T U 4.625 = RS 19.052; K T U 4.636 = RS 19.097; K T U 4.638 = R S 19.100B; K T U 4.750 = RS 29.094; PRU 6, 122 = RS 21.203; Ug 5, 95 = RS 20.01; Ug 5, 96 = RS 20.12, etc. 8 K T U 4.110 = RS 12.006; K T U 4.244 = RS 16.396; K T U 4.643 = RS 19.105, etc.; cf. also L I V E R A N I 1979c. 9 K T U 4.625 = RS 19.052; K T U 4.632 = RS 19.073, etc.; cf. also PRU 6, 141 = RS 19.112 (?). 10 K T U 4.89 = RS 11.796; K T U 4.128 = RS 14.176; K T U 4.175 = RS 15.096; K T U 4.275 = RS 17.125; K T U 4.296 = RS 17.140; K T U 4.358 = RS 18.048; K T U 4.367 = RS 18.076; K T U 4.618 = RS 19.045; K T U 4.636 = RS 19.097; Ug 5, 95 = RS 20.01. 11 K T U 4.143 = RS 15.031; K T U 4.213 = RS 16.127; K T U 4.243 = RS 16.395; K T U 4.269 = RS 17.106; K T U 4.271 = RS 17.115; K T U 4.345 = RS 18.033; K T U 4.397 = RS 18.136; K T U 4.400 = RS 18.139; K T U 4.636 = RS 19.097. 12 K T U 4.65 = RS 11.602:1. 13 K T U 4.35 = RS 8.183+; K T U 4.65 = RS 11.602; K T U 4.71 = RS 11.721; K T U 4.87 = RS 11.789; K T U 4.96 = RS 11.840; K T U 4.99 = RS 11.845; K T U 4.106 = RS 12.001; K T U 4.122 = RS 13.012; K T U 4.125 = RS 14.001; K T U 4.126 = RS 14.084; K T U 4.129 = RS 15.001; K T U 4.141 = RS 15.022+; K T U 4.153 = RS 15.046; K T U 4.175 = RS 15.096; K T U 4.183 = RS 15.116; K T U 4.213 = RS 16.127; K T U 4.243 = RS 16.395; K T U 4.269 = RS 17.106; K T U 4.296 = RS 17.031; K T U 4.307 = R S 17.384; K T U 4.320 = RS 17.444; K T U 4.332 = RS 18.010+; K T U 4.355 = RS 18.045; K T U 4.358 = RS 18.048; K T U 4.374 = RS 18.082; K T U 4.378 = RS 18.087; K T U 4.382 = RS 18.106+; K T U 4.440 = RS 18.[317]; K T U 4.491 = RS 18.[389]; K T U 4.493 = RS 18.[391];

f ) M a n y texts also deal with the deliveries from the gts for certain persons. T h e goods are wheat, emmer, wine, oil, wood, cattle, textiles, metal artifacts, etc. 14 At times it is difficult to differentiate between storage stocks and deliveries. It should be noted that these texts may overlap with commercial ones and we cannot draw a clear dividing line. Furthermore, many of the texts are of mixed content and it is sometimes impossible to classify them precisely. 1.2.3

The royal servicemen

These are texts dealing with the 'royal servicemen' (Ug. bnš mlh, Akk. ardē šani) according to their professional groups. Again the divisions are not definitive. a) Texts where the professional group is mentioned as a unit, the personal names of its members not being given, and where the whole group receives deliveries of some kind. 1 ' b) Lists of royal servicemen of different professions (hrs anyt, 'shipwrights', hrs bhtm 'house-builders', nskm '(metal) casters', etc.) where each professional is listed and identified only by his name plus his father's name, or by his father's n a m e alone (bn X 'son of (PN) X'). Sometimes they also receive certain deliveries in silver, or in kind and sometimes in these texts give a total of the amount of the delivery. These texts also tell us in which village of the kingdom of Ugarit they resided. 16 K T U 4.608 = RS 19.014; K T U 4.609 = RS 19.016; K T U 4.618 = RS 19.045; K T U 4.630 = RS 19.062; K T U 4.636 = RS 19.097; K T U 4.729 = RS 24.301; K T U 4.740 = RS 24.660e; PRU 6, 118 = RS 18.116; Ug 5, 96 = RS 20.012. 14 K T U 4.36 = RS 8.208; K T U 4.38 = RS 8.272; K T U 4.128 = RS 14.176; K T U 4.131 = RS 15.003; K T U 4.144 = RS 15.032; K T U 4.149 = RS 15.039; K T U 4.153 = RS 15.046; K T U 4.175 = RS 15.096; K T U 4.213 = RS 16.127; K T U 4.216 = RS 16.165; K T U 4.230 = RS 16.341; K T U 4.243 = RS 16.395; K T U 4.263 = RS 17.049; K T U 4.269 = RS 17.106; K T U 4.352 = RS 18.042; K T U 4.377 = RS 18.086; K T U 4.378 = RS 18.087; K T U 4.387 = RS 18.112; K T U 4.495 = RS 18.[394]; K T U 4.609 = RS 19.016; K T U 4.630 = RS 19.062; K T U 4.636 = RS 19.097; K T U 4.705 = RS 21.056, etc. 15 K T U 4.105 = RS 13.014[Aw]; K T U 4.126 = RS 14.084; K T U 4.139 = RS 15.017; deliveries: K T U 4.29 = RS 3.320; K T U 4.36 = RS 8.208; K T U 4.38 = RS 8.272; K T U 4.47 = RS 10.043; K T U 4.99 = RS 11.845; K T U 4.217 = RS 16.176; K T U 4.312 = RS 17.397[A]; K T U 4.416 = RS 18.252; K T U 4.485 = RS I8.[381]; K T U 4.745 = RS 25.417 and K T U 4.125 = RS 14.001 (cereals); K T U 4.216 = RS 16.165 (wine); K T U 4.609 = RS 19.016 (hpr monthly 'ration's); K T U 4.610 = RS 19.017 (argmn špš 'tribute to the Sun', i.e. the Hittite king). 16 K T U 4.35 = RS 8.183+;' K T U 4.43 = RS 9.011; K T U 4.44 = RS 9.453; K T U

c) T h e bnš mlk can be considered as a special category, designated as being at the disposal of the king, queen (or queen-mother), the 'vizier' (Ug. skn, Akk. sākinu) or other functionaries who are defined as overseers. T h e formula