Quantity in Czech: A dialectical and historical analysis. Ph.D. dissertation

THE UNIVERSITY OF CHICAGO

QUANTITY IN CZECH: A DIALECTAL AND HISTORICAL ANALYSIS

A DISSERTATION SUBMITTED TO THE FACULTY OF THE DIVISION OF THE HUMANITIES IN CANDIDACY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY DEPARTMENT OF SLAVIC LANGUAGES AND LITERATURES & DEPARTMENT OF LINGUISTICS

BY MARK JEROME PISARO

CHICAGO, ILLINOIS JUNE 2002

UMI Number: 3048356

________________________________________________________ UMI Microform 3048356 Copyright 2002 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. ____________________________________________________________ ProQuest Information and Learning Company 300 North Zeeb Road PO Box 1346 Ann Arbor, MI 48106-1346

Copyright © 2002 by Mark Pisaro All rights reserved

For Michaela

Let no one be surprised if, in speaking of entirely new principalities as I shall do, I adduce the highest examples both of prince and of state; because men, walking almost always in paths beaten by others, and following by imitation their deeds, are yet unable to keep entirely to the ways of others or attain to the power of those they imitate. A wise man ought always to follow the paths beaten by great men, and to imitate those who have been supreme, so that if his ability does not equal theirs, at least it will savour of it. Let him act like the clever archers who, designing to hit the mark which yet appears too far distant, and knowing the limits to which the strength of their bow attains, take aim much higher than the mark, not to reach by their strength or arrow to so great a height, but to be able with the aid of so high an aim to hit the mark they wish to reach. — Niccolò Machiavelli, The Prince

v

TABLE OF CONTENTS List of Maps................................................................................................................. viii List of Tables .................................................................................................................. x List of Symbols............................................................................................................. xii Abbreviations .............................................................................................................. xiii Cyrillic Transcription ................................................................................................ xviii Introduction .................................................................................................................... 1 Notes on Major Sources ................................................................................................. 5 Dialectological material from the Czech Academy of Sciences. ......................... 5 Dialect Regions .............................................................................................................. 7 Major Czech Dialectal Divisions ................................................................ 7 Dialects of Chodsko (chodský úsek) .................................................................... 8 Archive material for the Staročeský Slovník ........................................................ 9 Chapter 1 -- Historical Background............................................................................. 11 Beginnings ......................................................................................................... 11 Czech Literary Language (CLC)........................................................................ 20 Chapter 2 -- Synchronic Quantity ................................................................................ 26 Phonetics ............................................................................................................ 26 Morphology........................................................................................................ 30 Chapter 3 -- Quantity in Old Czech ............................................................................. 32 Detecting quantity in Old Czech documents ...................................................... 32 Chapter 4 -- Slavic Accentology & Czech Length...................................................... 36 The Need for a General Theory of Slavic Accentology ..................................... 36 Proto-Slavic Length ........................................................................................... 38 The Role of the Dialects..................................................................................... 43 Slavic Accentual Paradigms (a.p. = accentual paradigm) .................................. 45 Czech Quantity and Proto-Slavic Accent........................................................... 47 Dynamic Ictus and Fixed Ictus........................................................................... 51 Acute a.p. (a)...................................................................................................... 53 Oxytonic a.p. (b) ................................................................................................ 55 Mobile a.p. (c) .................................................................................................... 55 Chapter 5 -- Feminine Disyllabic (j)ā-stems................................................................ 57 Expected reflexes of root vowels in the Slavic Languages ................................ 57 Feminine Disyllabic j(ā)-stem paradigms in Contempory Literary Czech......... 70 vi

Acute Feminine (j)ā-stems ................................................................................. 74 The Source and Distribution of Czech Shortened Acute Intonation ......... 83 Czech Dialectal Data ................................................................................ 83 The Czech Lengthening? .......................................................................... 86 Interpretation of the Data.......................................................................... 87 Instrumental Singular................................................................................ 90 Genitive Plural .......................................................................................... 92 Oxytonic Feminine (j)ā-stems ............................................................................ 93 Mobile Feminine (j)ā-stems ............................................................................. 105 Relative Chronology............................................................................... 108 Neo-Acute Feminine jā-stems .......................................................................... 109 Feminine monosyllabic i-stems........................................................................ 115 Chapter 6 -- Masculine monosyllabic stems.............................................................. 117 Expected reflexes of root vowels ..................................................................... 117 Acute monosyllabic masculine (j)o-stems........................................................ 127 Oxytonic monosyllabic masculine (j)o-stems .................................................. 131 Mobile Monosyllabic Masc. (j)o-stems............................................................ 135 Chapter 7 -- Neuter Disyllabic substantives .............................................................. 146 Expected reflexes of root vowels ..................................................................... 146 Neuter Disyllabic j(o)-stem paradigms in Contempory Literary Czech........... 151 Acute Disyllabic Neuter (j)o-stems.................................................................. 153 Chapter 8 -- Conclusions............................................................................................ 166 Bibliography ............................................................................................................... 173 Major Reference works employed – Czech...................................................... 173 Major Reference works employed – Other Slavic Languages ......................... 174 Monographs and articles cited.......................................................................... 174 APPENDIX A – AV DIALECT SURVEY LOCALES........................................... 182 APPENDIX B – DIALECT MAPS........................................................................... 187

vii

LIST OF MAPS Map 1 – Lengthened /í/ in the Nom. Sing. /*jьskra/ (a.p. a) ‘spark’ ...........................................................188 Map 2 – Short /a/ in the Instr. Sing. of /*korva/ (a.p. a) ‘with cow’ ...........................................................189 Map 3 – Long /á/ in the Instr. Sing. of /*korva/ (a.p. a) ‘with cow’............................................................190 Map 4 – Reflex of Short Vowel in the Nom. Sing. /*griva/ (a.p. a) ‘mane’................................................191 Map 5 – Reflex of Shortened /e/ in the Nom. Sing. /*berza/ (a.p. a) ‘birch’...............................................192 Map 6 – Long /í/ in the Inst. Sing. of /*berza/ (a.p. a) ‘with birch’.............................................................193 Map 7 – Reflex of Retained Long Vowel in the Nom. Sing. /*repa/ (a.p. a) ‘with beet’ ............................194 Map 8 – Distribution of Length & Brevity in Nom. Sing. /*solma/ (a.p. a) ‘straw’ ...................................195 Map 9 – Reflex of Lengthened /ó / in the Nom. Sing. of /*kopa/ (a.p. b) ‘mound’ ....................................196 Map 10 – Lengthened /ó/ in the Nom. Sing. /*vosa/ (a.p. b) ‘wasp’...........................................................197 Map 11 – Long /ú/ in the Nom. Sing. of /*duha/ (a.p. a) ‘arc, rainbow’ .....................................................198 Map 12 – Lengthened /ó/ in the Nom. Sing. of /*krosna/ (a.p. a) ‘backpack for hay’ ................................199 Map 13 – Lengthened /í/ in the Nom. Sing. /*zima/ (a.p. c) ‘cold, winter, malt beverage’ ........................200 Map 14 – Lengthened /ó/ in the Gen. Pl. of /*noha/ (a.p. c) ‘leg’ ...............................................................201 Map 15 – Reflex of Length in the Gen. Pl. of /*louka/ (a.p. b) ‘meadow’..................................................202 Map 16 – Reflex of Length in the Gen. Pl. of /*bouda/ (a.p. b) ‘shack’ ....................................................203 Map 17 – Lengthened /é/ in the Nom. Sing. of /*dъždь/ (a.p. b) ‘rain’ ......................................................204 Map 18 – Reflex of Lengthened /ó/ in the Nom. Sing. /*rojь/ (a.p. b) ‘swarm’..........................................205 Map 19 – Lengthened /ó/ in the Nom. Sing. /*kosъ / (a.p. b) ‘the bird, Turdus Merula’............................206 Map 20 – Lengthened /ó/ in the Nom. Sing. of /*stohъ/ (a.p. b) ‘haystack’ ...............................................207 Map 21 – Long /á/ in the Gen. Sing. of /*gorxъ/ (a.p. a) ‘peas’..................................................................208 Map 22 – Lengthened /ó/ in the Nom. Sing. /*košь / (a.p. b) ‘basket’........................................................209 Map 23 – Lengthened /u/ in the Nom. Sing. of /*hus(a)/ (a.p. a) ‘goose’ ...................................................210 Map 24 – Lengthened /á/ in the Nom. Sing. of /*sazje/ (a.p. a) ‘soot’........................................................211 Map 25 – Long /y:/ in the Nom. Sing. of /*lyko/ (a.p. a) ‘phloem, cellulose layer of a tree’......................212 viii

Map 26 – Reflex of length in the NS of /*melko/ (a.p. b) ‘milk’ ................................................................213 Map 27 – Reflex of length in the Nom. Sing. of /*bolto/ (a.p. a) ‘mud, marsh’ .........................................214 Map 28 – Long /í/ in the Nom. Sing. of /*žito/ (a.p. a) ‘rye’ ......................................................................215 Map 29 – Shortening of Acute Long Vowels in Prehistoric Czech.............................................................216

ix

LIST OF TABLES Table 1 – Transference of ictus to 2nd syllable ..............................................................................................27 Table 2 – Non-transference of ictus to 2nd syllable .......................................................................................27 Table 3 – Expected correspondences of Proto-Slavic intonational patterns..................................................58 Table 4 – Acute disyllabic feminine (j)ā-stems in CLC with original long root vocalism............................59 Table 5 – Oxytonic disyllabic feminine (j)ā-stems in CLC with original long root vocalism.......................63 Table 6 – Circumflex disyllabic feminine (j)ā-stems in CLC with original long root vocalism ...................64 Table 7 – Neo-acute disyllabic feminine jā-stems in CLC with original long root vocalism........................66 Table 8 – Neo-acute disyllabic feminine jā-stems in CLC with original short root vocalism.......................68 Table 9 – Feminine disyllabic (j)ā-stem paradigms in CLC..........................................................................71 Table 10 – Acute disyllabic feminine ā-stems (all with original long root vocalism)...................................76 Table 11 – Acute disyllabic feminine jā-stems (all with original long root vocalism)..................................80 Table 12 – Oxytonic disyllabic feminine ā-stems with original long root vocalism .....................................94 Table 13 – Oxytonic/Mobile feminine disyllabic jā-stems..........................................................................100 Table 14 – Oxytonic disyllabic feminine ā-stems with original short root vocalism ..................................101 Table 15 – Mobile disyllabic feminine ā-stems with original long root vocalism ......................................105 Table 16 – Mobile disyllabic feminine ā-stems with original short root vocalism......................................106 Table 17 – Mobile feminine disyllabic jā-stems with original long root vocalism .....................................109 Table 18 – Mobile feminine disyllabic jā-stems with original short root vocalism ....................................109 Table 19 – Neo-acute disyllabic feminine jā-stems with original long root vocalism.................................109 Table 20 – Neo-acute disyllabic feminine jā-stems with original short root vocalism................................110 Table 21 – Feminine third declension substanival paradigms in CLC ........................................................115 Table 22 – Monosyllabic masculine paradigms in CLC .............................................................................117 Table 23 –Acute monosyllabic masculine (j)o-stems in CLC with original long root vocalism.................120 Table 24 – Oxytonic monosyllabic masculine (j)o-stems in CLC with original long root vocalism ..........124 Table 25 – Circumflex monosyllabic masculine (j)o-stems in CLC with original long root vocalism .......125 x

Table 26 – Oxytonic monosyllabic masculine (j)o-stems in CLC with original short root vocalism..........126 Table 27 – Acute monosyllabic masculine (j)o-stems with original long root vocalism.............................127 Table 28 – Oxytonic monosyllabic masculine (j)o-stems with original long root vocalism .......................131 Table 29 – Oxytonic monosyllabic masculine (j)o-stems with original short root vocalism ......................133 Table 30 – Mobile monosyllabic masculine (j)o-stems with original long root vocalism...........................135 Table 31 – Mobile monosyllabic masculine (j)o/u-stems with original short root vocalism.......................137 Table 32 – Acute disyllabic neuter (j)o-stems in CLC with original long root vocalism............................146 Table 33 – Oxytonic disyllabic neuter (j)o-stems in CLC with original long root vocalism.......................148 Table 34 – Circumflex monosyllabic neuter (j)o-stems in CLC with original long root vocalism .............149 Table 35 – Neuter disyllabic (j)o-stem Paradigms in CLC .........................................................................152 Table 36 – Acute disyllabic neuter (j)o-stems, all with original long root vocalism...................................153 Table 37 – Oxytonic disyllabic neuter (j)o-stems with original long root vocalism ...................................158 Table 38 – Oxytonic disyllabic neuter (j)o-stems with original short root vocalism...................................161 Table 39 – Mobile disyllabic neuter (j)o-stems...........................................................................................163 Table 40 – Indeterminate neuter (j)o-stems.................................................................................................164

xi

LIST OF SYMBOLS a.p. (a)

a.p. (c) // [ ] { } V C ~ * # ъ

Proto-Slavic Accentual paradigm with fixed stress on final syllable of the stem Proto-Slavic Accentual paradigm with fixed stress on the first syllable of the desinence Proto-Slavic Accentual paradigm with mobile stress pattern phonemic transcription phonetic transcription morphophonemic transcription vowel consonant morpheme boundary 1.) alternates with, 2.) above vowel = neo-acute intonation reconstructed or non-attested form 1.) fleeting vowel, 2.) end of phonological word or morpheme back jer [u9]

ь

front jer [ıª]

> <

becomes historically comes from historically

a.p. (b)

xii

ABBREVIATIONS Languages & Dialects

B BR BS ChSl. CR CS CSR CZ CLC Čak. ENG EPSL ES GMC IT Kaj. LAT LC LG LPS LSo M

MCZ MHG OCS OCZ OHG OR OSC P PCZ PIE Plb. PSL R RChSl S S/C SK SLN Slv. Štok. Ukr. USo

Bulgarian Belarussian Balto-Slavic Church Slav(on)ic Croatian Common Slavic Contemporary Standard Russian Czech Contemporary Literary Czech Čakavian Serbo/Croatian English Early Proto-Slavic East Slavic Germanic Italian Kajkavian Serbo/Croatian Latin Literary Czech Low German Late Proto-Slavic Lower Sorbian Macedonian

xiii

Modern Czech Middle High German Old Church Slav(on)ic Old Czech Old High German Old Russian Old Serbo/Croatian Polish Proto-Czech Proto-Indoeuropean Polabian Proto-Slavic Russian Russian Church Slav(on)ic Serbian Serbo/Croatian Slovak Slovenian Slovincian Štokavian Serbo/Croatian Ukrainian Upper Sorbian

Grammatical 1. 2. 3. adj. aor. a.p. AP arch. AS cf. DP DS fem. GP GS imp. imper. infin. IP ipf. IS

1st person 2nd person 3rd person adjective, adjectival aorist accentological paradigm accusative plural archaic accusative singular compare dative plural dative singular feminine genitive plural genitive singular imperfect imperative infinitive instrumental plural imperfective instrumental singular

LP LS masc. neut. NP NS p.a.p. pf. pl. p.p.p. pres. p.t. sg. sup. voc.

xiv

locative plural locative singular masculine neuter nominative plural nominative singular past active participle perfective plural past passive participle present tense past tense singular supine vocative case

Major Reference Sources ČJA I Český jazykový atlas, vol. 1 ČJA II Český jazykový atlas, vol. 2 ČJA III Český jazykový atlas, vol. 3 ESSL Etimologičeskij slovar’ slavjanskix jazykov. Gebauer Historická mluvnice jazyka českého. LF Listy filologické. Machek Etymologický slovník jazyka českého. Praha: Academia, 1971 PSSČ Příruční slovník spisovné češtiny, 1937-1956 SS Niederle, L. Slovanské starožitnosti. (referencing the following): Oddíl I. Oddíl kulturní. Díl I. Díl II. Díl III.

Život starých slovanů. Život starých slovanů. Život starých slovanů.

Oddíl II. Oddíl historický. Díl I. Díl II. Díl III. Díl IV. SSČ StČS Vas.

Původ a počátky slovanského národa. Původ a počátky slovanů jižních. Původ a počátky slovanů západních. Původ a počátky slovanů východních.

Slovník spisovné češtiny, 1995 Staročeský slovník. 1967Etimologičeskij slovar’ russkogo jazyka, t. 1-4 (Vasmer)

xv

Old Czech Sources Abbreviation

Description

AlbRáj

Ráj duše, a collection of Old Czech treatises attributed to Albert the Great. 2nd half of the 14th century.

Arch Č 6

(ed. F. Palacký, Prague 1872) Selected historical papers from 1401 to 1420.

Arch Č 18

(ed. J. Kalousek, Prague 1900) Selected historical papers from 1498 to 1502.

BiblNymb

Nymburk bible, 3rd edition. Manuscript, Prague, Muz IV B12, 731 fol.: 1462.

Budyš

Budyšinský manuscript, containing Husite prose and verse. 1420.

ChelčPost

Postila of Peter Chečický, Prague 1522, 282 fol.

EvZimn

Winter readings (čtenie zimnieho času), evangelical readings from the 2nd half of the 14th century.

HustPostH

Postila or Vyloženie svatých čtení nedělních. Prague Muz IV C. 18, fol. 1a-1961.

HusPostTN

Postila or Vyloženie svatých čtení nedělních. Postila svaté pamětí M. Jana Husa. Prague 1564, vol. 1, fol. 1a-1141.

Kruml

Krumlovský rukopis from the 1st quarter of the 15th cent.

LegDuchM

Fragments of legends in verse, circa 1300. Muz 1 Ac 50.

LegJidD

Fragments of legends in verse, circa 1300.

LegJidM

Fragments of legends in verse about Judas from sometime around 1306.

LegPil

Fragment of a legend in verse about Pilate from the beginning of the 14th century.

Otc

Životy sv. Otcův, 15th century.

Pror

Translation of the prophet Izaiah, Jeremiáše a Daniela from the end of the 14th century.

RokJanB

Výklad na evangelium sv. Jana: České Budějovice, Regional Library. Be A 24, 334 fol., 1492.

RokPostB

Postila, Brno, SA G 10, č. 115, fol.1a-324a, circa 1503.

SlovKlem

Latin-Czech dictionary, slovník Klementinský from 1455.

ŠtítSvátA

Sunday and saint day sermons. Praha, UK XVIIC15, 254 fol; beginning of the 15th century.

TovHád

Ctibor Tovačovský z Cimburka (†1494), Hádání Pravdy a Lži o kněžské zboží a panování jich; 1467.

VýklKruml

Krumlovský výklad, latin-Czech biblical dictionary from the1st half of the 15th century.

VýklŠal

Výklad Piesniček Šalomunových (Canticum Canticorum), 1st half of the 15th century. xvi

Abbreviations are taken from the StČS. This list includes only the most often quoted Old Czech sources. The abbreviations are identical to those used in the Staročeský slovník, úvodní stati, soupis pramenů a zkratek, Praha Academia, 1968. For more detailed descriptions of the works quoted, see Gebauer (1906, etc.).

xvii

CYRILLIC TRANSCRIPTION Because of the many orthographic systems in use in the Slavic world, it is necessary to use several systems of transcription in this paper in order to accurately represent sounds and letters. When transliterating Cyrillic the following system has been used, reproduced from the Slavic and East European Journal with some minor changes: A Б B Ґ

a b в ґ

a b v (U, BR) g

Г

г

(U, BR) h (all others) g (M) g

Ѓ Д E Ё Є w Ж З S

ѓ д e ё є w ж з S

И

и

(U) y (all others) i

I Ї Й

i ї й

i (U) ji j

d e ё (U) je (CS) je ž z (M, CS) dz

J K Ќ Л Љ M Н Њ O П C T Ћ Y  OY

j k ќ л љ m н њ o п c t ћ y 
oy

j k ќ l (SC, M) lj m n (SC, M) nj o p s t (SC) ć u (CS) u (CS) u

Ч Џ Ш

ч џ ш

č (SC, M) dž š

Щ

щ

(B, CS) št (all others) šč

Ъ

ъ

(B) â (all others) "

Ы Ь ý

ы ь ý

Э Ю Я Ф = 0

э ю я ф = 0

y ' ě é ju ja f i ę

Ϋ

ÿ

(BR) w

2

2

(CS) ję

Ф X Ц

ф x ц

f x c

1

1

(CS) o (B)

8 3 ÿ 6

8 3 ÿ 6

o (CS) jo (OR) ja (OCS) dz

Long vowels in Czech and Proto-Czech dialects will be marked by the acute accent in accordance with Czech orthography. A distinction between ú from lengthened

xviii

/*ú/ and ů1 from lengthened /*ó/, will be made throughout. Italicized forms represent spelling. When phonetic detail is called for, the International Phonetic Alphabet (IPA) is used between brackets denoting phonetic transcription: e.g. [bæ] ‘badger.’ Old Czech manuscripts written before the adoption of a standard orthographic tradition present special problems of transcription that are discussed in detail in Chapter 3.

1

The circle or kroužek over the ů was intended by Jan Hus, the inventor of Czech orthography, to indicate the etymological source of long ů from /ó/ as opposed to long ú from /ú/. In general, long ú from /ú/ is preserved in word initial position. Elsewhere, long /ú/ > /uo/ > /ou/ (e.g. CLC ústa ‘mouth’ vs. CLC sousto ‘morsel’). xix

INTRODUCTION

The purpose of this doctoral thesis is to analyze vowel quantity in Czech1 synchronically and to trace its development from a comparative historical and dialectal perspective. By vowel quantity what is meant is the phonemic distinction2 in Czech of long versus short vowels. Vowel quantity is one of the characteristics of the Czech sound system that distinguishes it from other Slavic languages. A phonemic quantitative difference between long vowels and short vowels in Czech plays an active and important role in distinguishing between meanings of words. For example, the words {dal-Ø#} ‘he gave,’ and {dál#} ‘further,’ with a markedly long vowel, differ only with respect to the length of the vowel /a:á/, yet these two words are semantically and etymologically unrelated. Similar examples illustrate the same relationship for the other Czech vowel pairs: e.g. /e:é/ {lek-Ø#} ‘fear’: {lék-Ø#} ‘medicine’; /i:í/ {bit-Ø#} ‘apartment’: {bít-Ø#} ‘to be’; /u:ú/ {ú-tlum-Ø#} ‘drowsiness’ : {u-tlum-Ø#} imper. 2. sg. of infin. {u-tlum-it#} ‘to muffle’; /o:ó/ {bor-Ø#} ‘pine forest’ : {bór-Ø#} ‘borium.’3

1

Throughout this paper the word Czech unless otherwise noted refers to the dialects spoken roughly within the boundaries of the Czech Republic and the people who speak these dialects. For a more detailed description of these dialects, see Bělič, 1971. The Czech dialects spoken abroad (e.g. Vojvodina, Serbia; Texas, U.S.A.; Volyn, Ukraine) although interesting in their own right have not been included in this thesis because of a general lack of data for the core words and word forms under investigation. 2

In the dialects of Silesia (slezská/lašská nářečí) as well as the Czech dialects transitional with Polish (nářečí polsko-českého míšeného pruhu) the opposition between long and short vowels has been lost, although evidence of prehistoric length shows up as /u/ for /ó/ and /i/ for /ē/. 3

In Standard Czech (i.e. the literary language based on a 16th century central Bohemian dialect) /ó/ is somewhat marginal as a phoneme since it occurs only in borrowed words. In several Moravian dialects, however, as well as at an earlier stage in the dialect upon which the literary language is based, the opposition /o/:/ó/ is phonemic. 1

2

Comparative historical refers to two distinct yet complementary methodologies: internal reconstruction and the comparative method. Internal reconstruction sensu stricto strives to understand the history of a language from data contained within the language itself, whereas the comparative method compares two or more languages that share a common ancestor. Both methods are employed here in analyzing data from the history of Czech and other Slavic languages to support hypotheses about the development of vowel quantity in Czech and the relationship of vowel quantity in Czech to the Proto-Slavic accentological system. Comparative dialectal refers specifically to the dialect geography of the Czech Republic and what can be gleaned from the Czech dialects about the development of quantity in Czech in light of some newly available dialectal material in addition to several reliable studies compiled on individual dialects. Making use of what is known from these existing studies, together with supporting evidence from selected manuscripts dating from the 13th century,4 and further aided by the advances that modern historical linguistic methods have achieved in the reconstruction of Proto-Slavic, the present work will analyze the development of quantitative vocalic relationships in the Czech dialects, including the dialect upon which the literary language is based, beginning with the period just before the differentiation of Czech from Proto-Slavic and leading up to the present day. A major goal of this work is to make Czech data about quantity useful to other linguists working on Proto-Slavic accentology. For this reason, this thesis will concentrate on selected lexical items that have good Proto-Slavic pedigrees and that are

4

Czech manuscripts from the 9th and 10th centuries, or at least those which are copied from manuscripts this old, are not typically considered to be, in the strictest sense of the word, Czech, but rather Common Slavic (Old Church Slavonic) of a Bohemian rescension. From the 10th to the early 13th century, most of the documents that originated in the Czech lands were written in Latin. Therefore, the documents most pertinent to this study originate from the 13th century.

3

relevant for answering the questions posed by Slavic linguists. Topics investigated include the distribution of neo-acute, acute, and circumflex reflexes in the Czech dialects, morphological and phonological levelings of ancient accentological paradigms. The idea has been to pay particularly close attention to lexical items that are very old, that is, that are reconstructable as Common Slavic words5 used by the group of people that spoke Proto-Slavic, collect data about the form which these lexical items occur in throughout the Czech dialect region and in the oldest Czech manuscripts, and analyze the data from a historical, dialectal, and theoretical perspective. Impetus for this thesis can be traced directly to my dissertation advisor, Prof. Bill Darden, who in reply to the question, “What should I write a dissertation about?” answered, “Czech quantity.” Without Prof. Darden’s unflagging support and advice, not to mention the countless hours he spent teaching me about the intricacies of Slavic accentology, this work would not have been possible. At the University of Chicago, I benefited from meetings with many individuals who helped in one way or another with this project. I am grateful to Prof. Howard Aronson for his encouragement and suggestions on how to make this thesis a better work, Prof. Victor Friedman whose advice regarding the presentation of dialectal material has been invaluable, Jim McCawley (†) just for being Jim and lending me his brilliant mind on a subject that does not even represent a major interest, Tony Buccini for teaching me what historical dialectologists do, Prof. František Svejkovský who is an inspiration to nearly everyone he meets and who guided me through Czech Literature, old and not so old. At the Dialectological Division of the Institute of the Czech Language, Academy of Sciences of

5

Of particular help in selecting words with Common Slavic pedigrees was a work by Kopečný, F. 1981, Základní všeslovanská slovní zásoba. Praha. [Basic Common Slavic vocabulary list].

4

the Czech Republic, many scholars provided me with access to information and were never too busy to answer questions I might have about quantity in Czech. Igor Němec, PhDr., Milada Nedvědová, PhDr. Stanislava Kloferová, PhDr. Karel Fic, PhDr. Jarmila Bachmannová, Prof. Pavel Jančák, PhDr. Jiří Kouba (†), Prof. Jan Havránek, the Fulbright Commission, Prof. Jiří Kraus. Further impetus came from Edward Stankiewicz (1993), Chr. Stang (1965), G. Shevelov (1965), Alan Timberlake (1983a, 1983b) and other scholars who have made reference in their work to the lack of descriptions of Slavic dialects that treat accentual features in a systematic, exhaustive, and unified way. The absence of such analyses has led to as many theories about Proto-Slavic intonation and accent as there are people working on the problem, as well as a tendency to beg for postponement of final judgment until the dialects can be reckoned with and included in the analysis. And quite rightly so. Most dictionaries and reference sources for Czech, for example, adhere strictly to the literary language (CLC), which is based on the Central Bohemian dialect. But this dialect is radically different in some respects from other Czech dialects and cannot by itself be expected to provide a complete and accurate description of the development of quantity in Czech any more than e.g. the dialect spoken in Belgrade can by itself be expected to provide a complete and accurate description of the development of accentual features in Serbo-Croatian. The Accentual Patterns of the Slavic Languages (Stankiewicz 1993) brings us a steady leap forward in providing a systematic, exhaustive, and unified analyses of the accentual patterns of Slovene, Serbo-Croatian, Bulgarian, Russian, Ukrainian, Byelorussian, and Slovincian. Now, with the completion of the first three (of five) volumes of the Český jazykový atlas (ČJA) and the availability to scholars of the data that have been collected over the past 50 or so years in order to produce the ČJA, the dialect geography of the Czech speaking community is sufficiently documented to allow a

5

systematic, exhaustive, and unified comparative historical and dialectal analysis of the development and status of the feature the Czech language that plays a role in Proto-Slavic accentual patterns – namely vowel quantity.

Notes on Major Sources Dialectological material from the Czech Academy of Sciences. The Dialectical Division of the Institute of the Czech Language, Czech Academy of Sciences (Dialektologické oddělení, Ústav pro jazyk český, ČSAV) provided unrestricted access to the material that they have been gathering beginning in 1947 for the Český jazykový atlas. This material, which has so far culminated in the publication of the first three volumes of the ČJA (two more are to follow), is located in the archives of the Dialektologické oddělení, Ústav pro jazyk český, ČSAV in Brno and contains all the completed questionnaires from the project. The questionnaires,6 which were designed and created with a view towards the historical development of the Czech dialects, were filled out in duplicate by trained dialectologists who spent years in the field collecting data from informants who were carefully selected to best represent the most archaic stage of the dialect being recorded. These informants were often the oldest people in the community who had lived in the very same communities for most if not their entire lives. The data have been valuable enough to be designated a national monument (národní památník) and are kept in metal drawers organized by the numbers which correspond to the cities, towns, and villages that were selected to be part of the survey (ČJA I 1992:39-

6

The questionnaires went through several different phases over the years and there were also questionnaires that were sent out early in the study that were specific to certain regions. For this study, the questionnaire that was exclusively consulted was the final version of the one that was used for the entire project, the Dotazník pro výzkum českých nářečí (Český jazykový atlas) 1964.

6

47). The duplicates are then arranged by the entry number from the questionnaires, which correspond to specific lexical items, or grammatical forms that were researched for the study. For each word or word form relevant for the present thesis, the number of the location and the form the word was attested in at that location was copied down, noting any local or idiosyncratic variation. This information has been used to map the words under investigation. Appendix A lists the locations of the towns and villages where the dialectal survey was conducted. The geographical placement of these Academie Věd (AV) survey locales is evident from each of the maps attached in Appendix B.

7

Dialect Regions Throughout this study reference will be made to the dialect regions of the Czech speaking lands. The following map shows the major divisions of the Czech dialects. The divisions are somewhat arbitrary, based on a major isogloss or a bundle of isoglosses and morphoglosses that separate one dialect region from another. The Northeastern Bohemian and the Silesian dialects include several villages in Poland (118 Slané, 119 Jakubovice, 801 Bránice, 818 Velké Petrovice, 819 Křenovice).

A

C

B H

E D

F

Major Czech Dialectal Divisions Key A. Northeastern Bohemian dialects (severovychodočeská nářečí) B. Central Bohemian dialects (středočeská nářečí) C. & D. Southwestern Bohemian dialects (jihozápadočeská nářečí) C. Dialects of Chodsko (chodský úsek) D. Dialects of Doudlebsko (doudlebský úsek) E. Czech-Moravian dialects (českomoravská nářečí) F. Central Moravian dialects (středomoravská nářečí) G. Eastern Moravian dialects (východomoravská nářečí) H. Silesian dialects (slezská nářečí)

G

8

Dialects of Chodsko (chodský úsek) We will also frequently refer to the dialects of the Chod region. The map below shows details of this region. Of particular interest are the Chod towns that figured prominently in the Czech dialectological atlas project: 301 Postřekov, 302 Chodov, 303 Draženov, 304 Mrákov, 305 Zahořany, 306 Sulislav, 307 Všekary, 308 Lštění, 309 Černíkov, 310 Chodská Lhota, 311 Újezd.

9

Archive material for the Staročeský Slovník The Archive material from the Czech Academy of Sciences for the Staročeský slovník provides details about the development of quantity in Czech over approximately a 200 year period, from the earliest preserved manuscript fragments of the oldest Czech legends in verse (dating from very soon after 1300 AD.), to the oldest printed texts of the early part of the 16th century (in some cases later if the edition is a reprint from an earlier copy). This material, therefore, is concentrated on the period of arguably the most intense historical development of Czech. The information comes in the form of 4” x 6” cards arranged in an enormous card catalogue alphabetically by word. Each section in the card catalog represents all the collated material about that word and is used to create entries for the Staročeský slovník (StČS). The work to compile entries for the StČS was begun by Jan Gebauer in the 1880’s and began being published in fascicles in 1901 (see Syllaba, 1983, p.66-74 for further details regarding the publication of both of Gebauer ’s great works, Historická mluvnice and StČS). Gebauer saw the publication of only the fascicles from the letter A up to the letter J before his death in 1907, although he had collected citations for words beginning with all Old Czech letters of the alphabet. The project was not renewed until 1966 by the Ústav pro jazyk český, oddělení pro dějiny českého jazyka, Československá Akademie Věd with B. Havránek as the chief editor. Although they are similar in intent, the StČS of 1903-07 and the StČS of 1966-99 are quite different in design, scope and approach. The new StČS has benefited from 40 more years of additional compiling of relevant citations by specialists in the field. A much larger staff, improvements in printing technology, and the type of conceptual organization that vigorous modern lexicological methodology provides (largely thanks to Igor Němec) has also contributed to making the

10

new StČS a most reliable source. However, the new StČS team, after reprinting the original two volumes of Gebauer ’s StČS in 1977, has only just realized the publication of fascicle No. 21 povolánie – pravý (1996). Because the Staročeský slovník is still not complete and probably will not be for another decade, and because there is much more information for the entries from A to J than is available in the published StČS of 19031907 and the reprint, it was necessary to cull information from the card catalog of citations at the Ústav pro jazyk český in Prague for this study on the development of quantity in Czech. In addition to the above, an unpublished, internal document written for the Academy of Sciences by Dr. M. Nedvědová on quantity in Czech as attested in the oldest manuscripts proved to be an invaluable guide to interpreting indications of quantity in Old Czech manuscripts and printed texts.

Chapter 1

HISTORICAL BACKGROUND

This chapter is intended to provide a brief historical context for the settlement of the Czech lands by the Proto-Czechs and the formation of the Czech literary language or Contemporary Literary Czech (CLC). Although our knowledge of the Slavs’ early history is incomplete and sketchy in some areas, a great deal of attention has been given to the subject and a lively debate continues to this day. The discussion presented below synthesizes a fraction of the material relevant for establishing the Proto-Czechs in the Czech Republic with possible dialectal divisions suggested by early historical data. The next section outlines the formation of the Czech literary language. This discussion is important for the present work in establishing two basic facts: 1) the Czech literary language is based on the Central Bohemian dialects of the 13th through 16th centuries and should not be used exclusively for generalizations about the development of linguistic features throughout the entire Czech speaking territory and; 2) historical documents that reliably mark length are available several hundred years after the appearance of the Czechs in the Czech Republic, therefore any application of data to earlier periods concerning Czech quantity – be they synchronic, diachronic, or dialectal – involves careful extrapolation based on disparate and often controversial facts. Beginnings While the Slavs still lived together as a homogenous group of people, although pre-divisional regional dialectal features in all likelihood already existed, the language that they spoke was also more or less homogenous. The language of these early Slavs, 11

12

Proto-Slavic, has been largely reconstructed with precision and sophistication by Slavic linguists in an effort to trace the development of the various Slavic peoples and languages back to their collective roots. The investigation has taught us much about many aspects of their lives: where they lived, the tools they used, the plants they cultivated, their political struggles, cultural activities, religious beliefs, economy, legal system, and movements. Of particular interest to scholars has been the question of the Slavs’ original homeland and their early migrations. According to one commonly held and well supported theory, based, among other things, on the high concentration and etymologies of archaic Slavic toponyms in the region, the ethnicons mentioned in historical Greek sources (see Goląb 1991:236-309 for an excellent analysis of the problem), and the study of trees for which the Slavs have common non-borrowed words (Rostafiński 1908, Moszyński 1957), the early homeland of the Slavs (6th century B.C. to the 1st century a.d.) was located north of the Carpathian Mountains in the marshy region of the upper Dniepr River known as the Pripet’ Basin (Polesie or Volyně) in present-day Byelorussia and Ukraine.1 As the Slavs prospered and increased in number, they outgrew their homeland. Groups of Slavs – including those who later crossed the Elbe and Odra rivers and settled in Moravia, the Czech lands and Slovakia – began to migrate in search of more arable soil to support their brand of agriculture, avoiding heavily forested and mountainous regions that were not as fertile. In this connection we would like to quote Goląb (1991:300):

1

Dvorník (1956:3-13) argues for a broader region that extends West into Poland across the Odra River and covers the territory of the so-called Lusatian culture which archeological evidence shows was destroyed around 500 B.C.

13

So, the main thrust of the prehistorical Slavic expansion seems to have been west across the Bug River into the Vistula basin. Here I would like to emphasize the following points: many Slavic hydronyms in the Vistula basin have been transferred there from the Dniepr basin (Visla, etc., Bugъ, Dunajь, etc.); and the dendric boundaries of typical western trees with etymologically non-Slavic names, which on our map are represented by the eastern boundary of the beech tree [Sl. bukъ, borrowed from GMC] indicate clearly that the Central Vistula basin, with the inclusion of the Central Warta basin (which already belongs to the Odra system), must have been a relatively old zone of prehistorical Slavic colonization, which at the beginning avoided sub-Carpathian forested areas (rich in beech woods) and sub-Baltic areas (also with beech trees). These territories north of the Carpathians in all probability served as the launching pad for the migration of the Czechs and Slovaks in the 5th and 6th centuries, perhaps as early as the 3rd century (see below). The paths that the proto-Czecho-Slovaks followed over the Carpathian Mts. probably led through the Dukla Pass in the Eastern Beskyd Mts. between Poland and Slovakia, and the Jabluňkov Pass in Silesia between the Jeseník Mts. and Moravian-Silesian Beskyds. Some scholars believe that Avar attacks in 562 and 566 chased the West Slavs over the Odra and the Elbe rivers (Schenker 1995:21) and that the settling of Moravia and Bohemia thus took place after the Avar attacks. Niederle cites several German historians who expand on this theory and presuppose a chasing from the southern reaches of the Danube (SS II.3:182), but dismisses it on the grounds that the Avars did not arrive in Pannonia until the 560’s, whereas the West Slavs had certainly migrated beyond the Vistula to the Elbe River prior to this. He also argues that the civilization of the Western Slavs by the Avars could not have taken place on such a large scale. Gimbutas (1971:116) offers archeological findings, which, if accurate, would indicate that the Czecho-Slovaks began migrating through the Dukla Pass as early as the 3rd century. The area referred to, east of Košice at Prešov, where “cultural elements of

14

the ‘Przeworsk type’ … from the third to fifth centuries AD, which possibly represent the earliest Slavic occupation in this area,” is directly south of the Dukla Pass. There is some evidence for a separate surge into Southern Bohemia based on archeological and cultural differences between the north and south Bohemians during pagan times. Shevelov (1964:373) also seems to think that at least one Slavic tribe, the Dúdlebi (SLN Dudlebi), migrated with the Avars to southwest Bohemia and northeast Slovenia in the mid 6th century from Volyně. The tribal name is preserved to this day in the designation for the region of Bohemia called Doudlebsko, but this migration did not necessarily have to be accompanied by the Avars. Nor did the Dúdlebi necessarily migrate independently of the other Czecho-Slovaks. The tribal name might well have been preserved amidst a larger group of migrating Western Slavs, most of whom also kept their tribal designation. Whatever the truth may be, historical sources are mute on the question of when and how the Czechs settled Bohemia, Moravia, and Lower Silesia.2 It is evident that other Proto-Slavic tribes migrated at about the same time as the ProtoCzechs and Slovaks (some perhaps in unison) and were either assimilated when early tribal unions were strengthened or later tribal unions were formed. Some groups moved on further south (Croatians),3 southwest (Slovenians), west (Sorbians), north

2

The place name Bohemia is from the Germanic name Bai(a)-haimon (in Tacitus Boihemum) meaning the home of the Boi – the Celtic tribe that inhabited the Czech lands before the Czechs occupied their present territory (Trávníček 1935:3). The name Moravia is from the river Morava that flows south from the Jesenik Mts. through Eastern Moravia to the Danube River. And the name Silesia is in all probability derived from the Vandalian tribe name Sillingi, who inhabited Silesia before the Slavs arrived and who must have been later absorbed by the Slavs (Goląb 1992:303-305). 3

According to Dvorník (1956:63, fnt.) “It seems that the Croats who migrated to the south on the invitation of Heraclius were the tribes which were settled in western Galicia, round Cracow and the upper Vistula, although some Slavs, still bearing their name, may have stayed in their old home and seem to be mentioned in the foundation charter of the bishopric of Prague in the tenth century.” This is not to say that the tribal name referred to an already Southern Slavic branch of the early Slavs, but rather that outshoots of a larger Proto-Slavic tribe settled in

15

(Pomeranians, Slovincians, Polabians) or expanded their territory in all directions (Poles). We can also surmise that the formation of larger unions or alliances that allowed its members to make use of fortresses or who traded among one another characterized this early period. Niederle (SS, Oddíl II, díl III:183) sums the situation up succinctly: Zbývá tedy jenom filologie, ale ta ovšem mluví ve prospěch vlasti a cesty severní, nebot´nám ukazuje řetěz slovanských řečí tak navzájem úzce a starobyle spojený, že se tentýž řetěz, totéž territoriální spojení předpokládati musí i pro poměry prvotních dialektických středisk ve slovanské pravlasti, která byla na severu Karpat. Už v ní centrum pračeské bylo vedle srbského, polského a slovinského umístěno tak asi, jako dnes a Čechové se nemohli do nynější své vlasti dostati jiným způsobem než tím, že se posunuli ku předu v stálém dotyku na jedné straně s lužickými Srby a na druhé se Slovinci. Přišli tudíž ze severovýchodu, čímž ovšem nevylučuji, že část, dostihnuvší přes Moravu Dunaje, mohla se do jižních Čech dostati odtud, od Dunaje, o čemž se dále ještě zmíním. [All we have then is philology, which certainly attests to a northern enclave and a path of migration from the north, because it shows us a chain of Slavic dialects so interrelated, tight, and anciently connected, that this same chain must also entail a territorial connection for the conditions of the first dialectal centers of the Slavic homeland, which was in the northern Carpathians. The Proto-Czech center was already defined along with the Serb, Polish, and Slovene centers positioned the same way as today, and the Czechs could not have reached their present lands in any other way than by moving forward in constant contact on one side with the Lusatian Serbs and on the other side with the Slovenes. They came, therefore, from the northeast, by which I do not exclude the possibility that a part of them, extending across Moravia to the Danube, came from the Danube region to Southern Bohemia, which I will make reference to again].

territories west of their original homeland before the migrations of the South Slavs to their present locations.

16

Unfortunately, there are no direct historical sources about the life of neither these early Czechs nor the paths they took to reach Bohemia. The best evidence we have for the early dialectal divisions come from the present dialectal geography of the region. The extent to which dialectal features, particularly as regards vowel quantity, can be claimed to represent early dialectal divisions is difficult to establish. Some dialectal isoglosses seem to be very ancient indeed, and others are obviously of later provenance. Historical evidence for these tribal divisions begins in the 11th century. The most important sources are the founding charter of the Prague bishopric from 1086 (which exists in a copy from the 12th century), and Cosmae Pragensis Chronica Bohemorum, which confirms the old privileges from the year 973. In the charter of the Prague bishopric the borders of its territory are defined in terms of the tribes that lived there: Termini autem eius occidentem versus hii sunt. Tugust, quae tendit ad medium fluminis Chub, Zedlza et Lusane et Dazana, Liutomerici, Lemuzi usque ad mediam silvam, qua Boemia limitatur. Deinde ad aquilonem hi sunt termini: Pssouane, Chrouati et altera Chrowati, Zlasane, Trebouane, Pobarane, Dedosize usque ad mediam silvam, qua Milcianorum occurunt termini. Inde ad orientem fluvios habet terminos: Bug scilicet et Ztir cum Gracouua civitate provinciaque cui Uuag nomen est cum omnibus regionibus ad praedictam urbem pertinentibus, quae Gracouua est. Inde Ungroroum limitibus additis usque ad montes quibus nomen est Triti dilatata procecdit. Deinde in ea parte, que meridiem respicit, addita regione Morovia usque ad fluvium cui nomen est Wag et ad mediam silvam cui nomen est More et eiusdem montis, eadem parochia tendit, qua Bauuaria limitatur. [The boundaries (of Bohemia) to the west are the following. The Tugust stretches to the middle of the river Chub (Kamb), the Sedličané, Lučané and Děčané, the Litoměřici and Lemuzi without interruption stretch to the forest that bounds Bohemia. Then to the north the following boundaries obtain: the Pšované, Chorvati, and other Croats, Zlasane, Trebovane, Pobarane, Dodosize,

17

continue to the middle of the forest where the Milčane meet them. From there to the east the rivers form the boundaries: the river Bug it is certain and the Stir, where Kracov is, are civilized as is the province named Vag with the entire region already mentioned stretching out to Kracow. From the Hungarian boundary, their territory extends to the mountains named the Tatras. Moravia occupies this region east to the river Vag and to the mid Morava, which flows to the mountains, that bound Bavaria.] The names mentioned above in part refer to large tribes and in part to smaller groups or fortresses. The Tugust settlement occupied the territory from Domažlice south through the Bohemian Forest to the Cham in Bavaria and may be the ancestral territory of the Chods of Southern Bohemian.4 The Sedličané (Zedlza) are named after the settlement Sedlec located near Karlovy Vary (Carlesbad) in Northern Bohemia. Due to the gradual encroachment upon this territory by Germanic speaking peoples throughout history, the later resettlement of the same territory by the Czechs, and the intermingling of the two peoples, the dialects attested in and around Karlovy Vary are considered impure from a historical linguistic standpoint and tell us little if anything about the prehistoric dialect of the region. The Litoměřici (< *Liutoměrici/Lutoměrici) were situated at the mouth of the Ohře where Litoměřice is located today. Along the Elbe further north near Děčín, the

4

Mauer (1984:7-10) summarizes the historiographic references to the Chods and their towns and villages. There is not, however, any clear evidence for determining from whence they arrived nor when they settled in Chodsko. Speculation ranges from Poland, to the lower Danube, to Southern Russia and even Switzerland, and various authors date their arrival from the Bronze age to the 11th century. The first mention of the town, Chodov, is in 1195, but Mauer (Ibid.:10, fnt.) points out that this toponym is attested around Prague in 1185 and therefore offers little evidence for establishing the Chods as an ethno-linguistic group in place or time. Lutterer et al. (1988:123) gives further details regarding a third Chodov near Karlovy Vary and presents the etymology most often associated with the Chods – those that walked (chodili) along the border and guarded the territory against intruders.

18

Děčané settled and apparently also the Lemuzi. The Lusatian Serbs were separated from the Czechs by the forest north of the Krušné Mountains and by a line east of the Elbe that extended from Česká Kamenice through Česká Lípa to the peak Ještěd southwest of Liberec. South of this line past Bezděz to Mělník the Pšované lived. The Lučané, one of the largest tribes mentioned, occupied a wide expanse along the Ohře River from Žatec to Karlovy Vary and eventually extended their territory to the Mže River in the south. The mention of the Chorvati and other Croats in the Prague founding charter has been the topic of considerable debate for decades. Most scholars agree that the ablative altera means here ‘the other side of some boundary,’ but the question is the other side of what? Their territory in Bohemia was probably bounded in the north by the Krkonoše Mountains and in the south by the Elbe River and in the west by the Jizera where the Pšované met them, but in Silesia the Croats are also known to have occupied the upper Vistula and Upper Nisa basins so the eastern boundary is difficult to determine. At any rate, in all probability, these groups represent outshoots of a larger tribe mentioned in the Nestor Chronicles as occupying the Upper Vistula basin, which completely lost their original dialectal roots and were absorbed by the local Czech dialects over a period of 1000-1500 years. The Czechs are first mentioned as the Cichu-Windones in a manuscript from southern France, the Moissac Chronical, in connection with a Frankish expedition to the Czech lands in 805. From Kosmas’ manuscript we know that the tribe with the name Čechové occupied a territory which stretched to the south and west to the forest by Křivoklát past Libušín, along the Žuřin River by Mt. Osek and to the north along the lower Ohře to the Elbe. To the east, they reached to Kouřim beyond which the Zličané settled. Even though this tribe did not occupy the largest territory of the early Czechs, their rulers succeeded in unifying kindred tribes in Bohemia, Moravia and Slovakia

19

whose administrative center was in Prague (Praha). These rulers gave the ethnonym Čechové to the unified entity whose power and prestige grew with the significance of Prague as its center. The name Praha first appears in historical documents in the 10th century most often in the High German. An interesting description of the city was given by Ibrahim ibn Jakub here quoted after Niederle (SS, Oddíl II, díl III:198, fnt). The city Frága is constructed of stone and limestone and is one of the richest trade cities. Russians and Slavs arrive from Kracow, and the Turkish Musilmans, Jews as well as Turks arrive with goods and byzantine objects and trade them for flour, zinc, and various furs. Their land is the best of the northern territories and they have the richest means of existence: they sell for one “knšár” enough wheat to last for a month, and one “knšár” will buy enough barley to last a horse 40 days or will buy 10 chickens. In the city, one can also obtain saddles, bits, and shields that are made use of in this land. It is therefore evident and understandable that the literary activity of the Czechs began and developed primarily around the culturally and economically most significant center in Bohemia with Prague as the only major contender. The dialects that were spoken by e.g. the Lučané, the Litoměřici, and the other tribes distinct from the Čechové most certainly continued to develop separately from the Central Bohemian dialect that was elevated to the status of the Czech national literary language. On the other hand, many dialectal features were without a doubt erased throughout history having normalized to the pattern of the literary language. To what extent quantitative features are preserved in the dialects or erased altogether is one of the subjects of this study and will be addressed in a later chapter.

20

Czech Literary Language (CLC) This section is a brief introduction to the history of the contemporary Czech literary language (CLC), otherwise known as spisovná čeština, in very broad terms and is limited in scope to matters relevant for the philological study at hand. Our intention is to establish basic facts concerning the development of the literary language and Czech orthography in order to assess the reliability and relevance of Old Czech (OCZ) manuscripts and printed sources for determining the phonological history of the Czech language, particularly with respect to vowel quantity. It would be simple to determine the quantitative vocalic relations in pre-divisional Czech dialectal forms of Late ProtoSlavic if 6th to the 10th century Proto-Czech texts existed with markings for vowel quantity and ictus, but fate did not provide philology with such straightforward evidence. In fact, up until the 9th century there is scant written documentation of any Slavic language that was later connected with Czech, and there is a gap of about 500 years between pre-divisional Late Proto-Slavic dialectal Czech (Proto-Czech) and a firm Czech literary tradition with documents reliably marked for vowel quantity. From the very beginning of a literary tradition in the Czech lands, OCS competed with Latin as the language used in all aspects of religious and literary life. The glagolitic OCS tradition that had been established in Moravia by the mission of St. Cyril (†869) and his brother Methodius (†885) in 862/863 was no doubt based on their own Makedonian speech. This OCS tradition was more dear to the Slavic inhabitants of the Great Moravian Empire than the Latin tradition and for a short while dominated the fray, but the use of OCS in church services met with immediate disapproval from Rome, which was under political and economic pressure from the Bavarian bishops already established in Central Europe to use exclusively Latin for religious activities. The original mission lasted only two decades, but it was nevertheless crucial in establishing OCS as the

21

liturgical language in the Great Moravian Empire at the end of Ro/astislav’s5 reign and enabled its resurgence in Bohemia in the 10th century under Prince Václav who had been schooled in both liturgy written in Latin and OCS. The predecessors of the original mission were chased out of Moravia to Bulgaria in 885 following Methodius’ death and the persecution, imprisonment, and the torture of Methodius’ disciples. Although three disciples, Clement, Naum, and Angelarius, escaped on a raft they built which floated them down the Danube to Belgrade, there does not seem to be any evidence that direct disciples of Cyril and Methodius fled to Bohemia. In 1027, nevertheless, with the prince’s support, Prokop, the monk, founded the Benedictine Sázavský Klašter (in Sázava, southeast of Prague) which became the progenitor of the glagolitic OCS tradition in the Czech lands for the better part of the 11th century. Among the manuscripts attributed to the Sázavský monks are the Legend of St. Václav, Besědy sv. Řehoře Velikého, and the Lives of St. Ludmila, St. Prokop, and St. Vít. There is a good deal of speculation as to the origin and age of several Old Church Slavonic glagolitic manuscripts with language of the 9th and 10th centuries that have been shown to contain certain phonological and morphological features that are West Slavic in reflex – notably the Kiev Missal and the Prague Fragments.6 According to Vajs

5

There is somewhat of a polemic concerning Ro/astislav’s linguistic identity that involves the spelling of his name as either Rastislav which would indicate affinity with a Central Slovak dialect or Rostislav which would indicate Moravian or Bohemian affinity. Palacký (1936, vol. I.:80-81) refers to the prince using both spellings, but by implication acknowledges that the historically more correct form is Rastislav. He also mentions that the Great Moravian Empire (830-906 a.d.) extended far into Slovakia and Pannonia, and we know that the Slovak town, Nitra, played a very important role in the formation and expansion of the Great Moravian Empire. 6

The other early sources of Czech words are economically described by Travníček (1935:9): “Bohemika jsou jazykové drobty, ojedinělá slova nebo vazby, v cizím kontextu, v památkách psaných německy nebo hlavně latinsky. Jsou to především vlastní jména osobní a místní, někdy také obecná, která písař nebo autor neuměl nebo nechtěl přeložiti; často mají zlatinisovanou podobu anebo jsou zapsána neuměle. Nalézáme je od sklonku století 10. v lat. legendách o sv. Ludmile (např. Slavibor), o sv. Václavu, o sv. Vojtěchu, dále v různých starých listinách,

22

(Rukověť hlaholské paleografie, 1932:118) several leading scholars differ as to their conclusions regarding the origin of these documents. Sreznevskij (1871:490-3) believes that the Kiev Missal is Moravian or Czechoslovakian. Geitler (Die slav. u. albanes. Schriften 153) thought it arose on Makedonian soil. Jagić (1890) who agreed with Sreznevskij, and Vondrák (O původu Kijevských listů a Pražských zlomků v Praze, 1904:4-5) believes that it is to be attributed to Pannonian Slovenian because Methodius’ disciples are believed to have continued their work in Sirmium on Lake Balaton after fleeing Moravia. Durnovo (Slavia I, 1922:219-227) admits that the author might have been a Czech whose language preserved nasal vowels in place of denasalized full vowels. There is similar speculation regarding the origin of the Prague fragments and at least one reputable scholar, Jagić (Entstehungsgeschichte 106), attributes them to the Sázavský Klašter. The dates of these documents are also widely disputed. What is not disputed is the fact that no markings for quantity are present in these works (Havránek 1936:5) and that the above-mentioned manuscripts already testify to OCS’s loss in prestige by the many mistakes they contain. The OCS tradition in the Czech lands came to an abrupt end in 1097 when the monks at the Sázavský Klašter were chased out and the Prague monks from Břevnov burned down the monastery. To this day, no one knows what happened to the Sázava monks.

otíštěných hlavně v těchto sbírkách: Regesta Bohemiae et Moraviae, Codex diplomaticus et epistolavis regni Bohemiae a Codex diplomaticus et epistorlaris Moraviaie.” [Bohemianisms are linguistic tidbits, isolated words or phrases, in a foreign context, in monuments written in German or primarily Latin. These are, for the most part, proper personal and place names, sometimes also common words that the scribe or author did not know how or did not want to translate; often they have a latinized form or they are transcribed incompetently. We find them from the turn of the 10th century in the Latin legends about Saint Ludmila (for example Slavibor), Saint Václav, Saint Vojtěch, and also in various papers printed in the following collections: Regesta Bohemiae et Moraviae, Codex diplomaticus et epistolavis regni Bohemiae and the Codex diplomaticus et epistorlaris Moraviaie.]

23

From the end of the 11th century, the literary tradition began to shy away from Slavic/Czech in favor of Latin in both the religious and secular spheres of life. Apart from a handful of Latin/Czech glossaries and manuscripts with Czech marginalia (GlosBrit, GlosJag, GlosJer, GlosMV, GlosOl, GlosOpat, etc.) there is virtually no written record of the Czech language of this period. The Roman Catholic Church, whose influence had considerably been strengthened in Bohemia, did not tolerate Slavic/Czech. Even the first secular historical treatise of Bohemia and the Přemyslid dynasty, Cosmae Pragensis Chronica Bohemorum, written in the 2nd half of the 12th century by Kosmas (1045-1125), a deacon of the Prague Church, is in Latin with only a few words and proper names in Czech. This work, nevertheless, marked a broadening of literary interest, which supported the budding transition from Latin to Czech that was not to culminate in the establishment of Czech as the literary language until after the Hussite wars in the mid 15th century. When in the later 12th and early 13th centuries the theatre arose as a new venue for literary activity, a growing and more diverse population (i.e. not only the clergy and monks) became involved in a new verbal art, which must have strengthened the status of the vernacular. Even though we have no direct records of it, we have to assume that a secular theatre, with magical scenes, criticisms of the church, paganism and diabolic references existed because the clergy specifically prohibited these practices. Since these subjects have no contact with ancient tradition, we can assume that the Czech literary tradition continued to develop despite its restricted use.

24

The Czech literary language in a restricted sense can be dated to sometime before the year 1300 with the beginning of a continuous body of Czech vernacular literature7, concerned with life in the city centered on the activities of the inhabitants of the castle – especially the prince or king and his court as described in the series of court documents known under the collective title Klaret. Since the city of greatest cultural importance and activity at the time was Prague, it is natural that the Czech literary language from the beginning was based on the Central Bohemian Dialect spoken in Prague and its environs. During the reign of Charles I, King of Bohemia (Charles the IVth, Holy Roman Emperor) the Czech literary language expanded considerably to include prose texts used for scholarly and entertainment purposes. In addition, Charles the IVth, partly because of his Czech ethnicity (his mother was Czech), moved the capital of the empire from Aachen to Prague. Thriving from its new status as capital of a vast empire, international influences began to enter the country, leaving ample evidence on the language, a subject of considerable interest but with arguably little impact on the phonology or morphology of Old Czech. It was not until Jan Hus devised a method of rendering Czech sounds with a combination of Latin letters and diacritics in De orthographia bohemica8 in the early part of the 15th century that quantity began to be marked. Yet, even after Hus introduced his orthography, it took another two centuries before his ideas became a standard that was adhered to by the literate few. We have tried to show the difficulty of the task of tracing the development of quantity in Czech by the use of early documents. The reader might rightly ask at this

7

Prof. Frantíšek Svejkovský (personal communication, 2002) points out that the true literary language must actually seek its source in the sermons, prayers, and songs that were used in church services well before the end of the 13th century. 8

For a recent discussion concerning the authorship of De orthographia bohemica, see Vidmanová (1982:75-89).

25

point questions such as: Is there any way to deduce the historical developments of Czech quantity during the 500 years of Proto-Czech for which no historical documentation exists? To what extent can the present dialect geography of the Czech language be relied upon to tell us of the distant past? These are questions, which we hope to answer in the following chapters of this study.

Chapter 2

SYNCHRONIC QUANTITY

The purpose of this section is to describe the current state of Czech quantity in the literary language and summarize briefly the results of inquiries into the phonetics, phonology, and morphology of quantitative phenomena in Czech. Phonetics The first significant attempt to measure Czech quantity in a laboratory occurred in 1893 by Jos. Král and Fr. Mareš.1 The goal of the study was to ascertain whether Czech metric verse depended on an accurate alternation of long and short syllables. Král, a classical philologist, recognized that the problem could not be solved with the naked ear, so he turned to Mareš, a physiologist, for help. Mareš constructed a special instrument, which was essentially a kymograph2 that consisted of a freshly prepared frog muscle tied to a telephone mouthpiece. The membrane of the mouthpiece reacted to speech by vibrating, sending a signal along the frog muscle which was attached at the other end by means of a lever to a pen that then etched the analog signal onto blackened paper wrapped around a turning cylinder. Although the measurements were not particularly

1

Král, J. and Fr. Mareš. “Trvání hlásek a slabik dle objektivné míry.” Listy Filologické XX (1893):257-290. 2

ky.mo.graph n [Gk kyma wave + -graph] (1872): a device which graphically records motion or pressure (as of blood) -- ky.mo.graph.ic adj -- ky.mog.ra.phy n (Webster’s Dictionary). 26

27

exact (the frog muscle reacted too slowly to the impulse), the authors concluded that the relationship of Czech long versus short vowels was not 2:1 (as had previously been stated by Hattala 1854), but rather that long vowels were merely somewhat longer than short vowels. Later phonetic studies confirmed their results. In 1900 a short article by two French linguists working in Le Laboratoire de phonétique expérimentale du College de France was published3 which tried to measure the effect of ictus (always on the initial syllable of the phonological word in the Czech literary language) on the intensity of the following syllable. The study showed that under certain circumstances, the ictus transfers some of its energy to the second syllable. The authors only tested 8 types of utterances and the results cannot be considered to be conclusive, but if correct, then a carry over of intensity from the first syllable to the second syllable occurs in the word types in Table 1 but not in the word types represented by the Table 2. Table 1 – Transference of ictus to 2nd syllable Syllable 1 short short short short

Syllable 2 short long short heavy

Syllable 3 n/a n/a long short

Examples bude, ona milá, kabát zeleným hubička

Syllable 3 n/a n/a short short

Examples blíží vrána zavítal žaludy

Table 2 – Non-transference of ictus to 2nd syllable Syllable 1 long long short short

3

Syllable 2 long short long short

Gauthiot and Vendrys. “Note sur l'accentuation du tchèque,” MSL, XI (1900).

28

In other words, they found that in disyllabic and trisyllabic words, an initial short vowel was the key for the intensity of the ictus carrying over to the next syllable. In trisyllabic words, the weight of the 2nd syllable also played a role. It was only in trisyllabic words with a 2nd heavy syllable that the ictus transferred some of its energy to the 2nd syllable. What this suggests is that an initial long syllable would not contribute to the preservation of length in the 2nd syllable of a phonological word, yet an initial short syllable followed by a long syllable in a disyllabic word would. In word final position, the authors note that quantity is unstable and this has been historically true as well. All word final lengths in Czech that are not the result of later contractions of VjV syllables have become short syllables. Because Král and Fr. Mareš examined the lengths of words and syllables rather than individual segments with an instrument decidedly inadequate, Chlumský made an early analysis based on work in Rousselot’s Paris laboratory published in French,4 and translated into Czech in 1911,5 which followed the Scriptur method (Chlumský 1911:1). This method involved a gramophone hooked up to a needle that magnified the recorded sounds by scratching lines onto a piece of carbon paper. The title of the second published report itself, Pokus o měření českých zvuků a slabik v řeči souvislé, belies the experimental nature of the study and Chlumský admits that at the time the experiment was conducted, he was just getting used to the methodology. He spends more effort on

4 5

Chlumský, J. “L'analyse du courant d'air phonateur in tcheque.” La Parole, 1902.

Chlumský, J. Pokus o měření českých zvuků a slabik v řeči souvislé. [An attempt to measure Czech sounds and syllables in connected speech]; Praha, 1911.

29

the consonantal system than on the vocalic system, but he discovered, in general, that short vowels often are pronounced longer at the end of the phonological word – sometimes even equaling the duration of the pronunciation of long vowels. He also refers to the effect of ictus emphasis on the duration of all vowels, but with little detail. The most comprehensive phonetic study of Czech quantity, which surpasses all previous studies, was conducted by Chlumský in 1925 and 1926 and published in 1928.6 This study is devoted to measuring vowel quantity in more detail than Chlumský's previous studies and has to be called the definitive work on the subject for literary Czech (spisovná čeština). The informants that he used were all from Prague, nevertheless the results give empirical evidence for several phonetic tendencies generally noted in the dialects. Když shrneme, co bylo řečeno, tedy u všech tří osob vidíme, I. že krajní samohlásky í, ú, (i, u) jsou zpravidla kratší než samohlásky ostatní, jmenovitě než á (nebo a) . . . Toto zjištění je důležité, neboť vysvětluje, proč tak snadno se krátí samohlásky í, ú daleko snadněji než á, zvláště když jsou v slabikách nebo slovech méně závažných. A v tomto krácení je zase pramen známých potíží pravopisných ve škole i mimo školu. (Chlumský, 1928:21) [When we summarize what has been said about all three subjects we see that 1.) the vowels í, ú, (i, u) are as a rule shorter than the other vowels in the system, and markedly shorter than á (or a) … This discovery is important, because it explains why the vowels í and ú are much more likely to shorten that á, especially when they are in syllables or words that carry less emphasis. And this

6

Chlumský, J. Česká kvantita, melodie a přízvuk. Praha: Ceská akademie věd a umění (1928).

30

tendency to shorten is also the source of the well known difficulties encountered in grammar school and elsewhere.] Morphology Morphological processes in CLC that involve vowel quantity are not too numerous to mention, but vary with respect to the historical layer the alternation reveals. Strictly speaking, to qualify as a synchronic phonological vowel lengthening we should only allow a/á, e/é, i/í, y/ý, u/ú, o/ó. Yet the system is complicated by the historical development of the vocalic system, and is further distorted by the present orthography which still uses the é although this vowel is normally pronounced [í] in obecná čeština. For example, some masculine diminutives derived by the suffix {–ek#} are accompanied by lengthening of the root vowel7: Vocalic Alternation

Masc. Substantive

Diminutive

Gloss

a/á

hlas

hlásek

‘voice’

e/é

oheň

ohének

‘fire’

e (< ě)/í

řez

řízek

‘cut, cutlet’

i/í

list

lístek

‘leaf’

y/ý

jazyk

jazýček

‘language’

u/ou

zub

zoubek

‘tooth’

o/ů

boh

bůček

‘god, idol’

u/ú

N/A

N/A

N/A

o/ó

atom

atómek

‘atom’

telefon

telefónek

‘telephone’

7

This rule is far from regular, however, in CLC, and there are yet other masculine substantives whose dimunitives derived with {–ek#} exhibit shortening of the root vowel – stůl

→ stolek ‘table,’ dům → domek ‘house.’

31

The productivity of this derivational pattern is illustrated by the examples atom → atómek, telefon → telefónek, but the alternation between /o/ and /ó/ is limited to certain environments – /o/ does not lengthen to /ó/ when the base word is monosyllabic or ends in a consonant other than /l, m, n, r/ (cf. koh-i-nórek), and even this rule is not steadfast (zlom → zlomek). The productivity of the lengthening in this class of words is compromised by the lack of examples where /u/ lengthens to /ú/. Historically, /ú/ diphthongized to /ou/ beginning in the 13th century. The alternation of /u/ and /ou/ is well preserved in masc. diminutives (zub:zoubek ‘tooth,’ puk:pouček/puček ‘seam’) as is the alternation between /o/ and /ů/ (roh:růžek ‘horn,’ boh:bůžek ‘god, idol’). Other derivational processes include the shortening of the root vowel as in žába:žabka ‘frog.’ There is also a tendency to lengthen /o/ in fem. ā-stems of foreign origin. This lengthening is a morphological process that is current in the modern language. An interesting example is the sign at the Prague metro station (recorded in 1996) on the B line named Flora. At one of the exits, some evil graffitist has scratched a čárka into the plastic above the /o/ in Flora to reflect the common pronunciation of this word as [flóra]. No doubt [flora] has become [flóra] because /ó/ is the mark of a foreign word as is /f/ and in this case an original /o/ has lengthened to mark the foreignness of the word.

Chapter 3

QUANTITY IN OLD CZECH

Detecting quantity in Old Czech documents This section devoted to vowel quantity in Old Czech originates from linguistic material which captures the development of quantity in Czech over roughly a 200 year period as attested in the oldest manuscripts of legends in verse which appeared shortly after the beginning of the 14th century and continues with old printed works from the period around 1500 (in some cases from the 16th century when the work is from an older period). The data is therefore concentrated primarily in the period of the historical development of Czech that is used for the Old Czech dictionary project that Gebauer used in large measure for his dictionary. The documentation for this data makes use of material Gebauer collected (found both in his dictionary as well as in sheets of paper from his archives, left to the Ústav pro český jazyk and made use of for the new dictionary project which is publishing fascicles of the dictionary from the letters N to Z), as well as specially excerpted material collected from old Czech manuscripts, in some cases from old printed works, in which vowel quantity to various degrees of specificity were marked (see the attached list of relevant Old Czech documents). The various degrees of detection of marked quantity and the different functions of diacritic marks especially over /i/ (as well as /ie/) and sometimes over /u/ is something which is important to take into consideration in some of the manuscripts. The vowel /i/ (short) is written either with a /y/ or with a /i/ but which as a rule are marked with a thin diacritic line (in addition to the common occurrence without the dot) in manuscripts 32

33

which do not mark quantity at all. In these cases the čárka has only the function of the dot over the /i/ in contemporary Czech -- that is, to visually emphasize the letter. The use of this mark (thin čárka) in manuscripts that which do mark quantity regularly therefore makes it difficult to clearly identify /í/ (long) as well as differentiate between the two. Unlike other well attested phonological changes whose various stages of development are reliably documented by old Czech manuscripts and texts (for example the change of /ie/ > /í/, /ó/ > /uo/ > /ů/, /šč/ > /št’/ etc.), vowel quantity is problematic to trace. From the viewpoint of old Czech scribes who were bound by the tradition of Latin texts where the length of vowels was not graphically marked, vowel quantity was a phenomenon that was of less importance than other phonological characteristics. The functional load of vowel quantity in differentiating words was light. The failure to mark the quantity of a vowel rarely interfered with the interpretation of the text being transcribed. Just as in the orthographic systems of a number of contemporary literary languages that do not mark quantity (Slavic and non-Slavic), differences in quantity arose within the context of the words and forms used. As far as the actual notation used to mark quantity in old Czech manuscripts and printed documents is concerned, according to the orthographic system known in Czech as “sprežkový” (e.g. combinatory), quantity in Czech was marked with double letters. For example: negmaam (nejmám) EvZimn 14a, naa§§ (náš) t. 18a, modlaam (modlám) Pror 8a, waazzyy (váží) t. 24a, §taal ŠtítSvátA 63a, (stál) naarod (národ) t. 10a; §weey (svéj) EvZimn 23a, lidee (lidé) t. 1b, obylee (obilé) Pror 57a, k kraloweey (k kralovéj) t. 16a, podlee (podlé) ŠtítSvátA 91a, naleezaa (nalézá) t. 98b; zieezlyw (žiezliv) EvZimn 9a, hnutyee (hnutie) t. 10a, napowijedaa (napoviedá) ŠtítSvátA 76a, wijera (viera) t. 176a; od lidij (od lidí) EvZimn 11a, nuzij (nuzí) t. 6a, na vdolyy (na údolí) Pror 12b, wuodczyy (vuodcí) t. 78a, wygijty (vyjíti) ŠtítSvátA 37a, pij§mo (písmo) t. 40a, Kyy (ký) t. 56b; z

34

oonu (z ónů, ex illis) EvZimn 12a, nooh (nóh) Pror 26a; ne§uu (nesú) EvZimn 7a, v mvu u§§ij (u mú uší) t. 2b, §uv§yedee (súsědé) Pror 24a, rzkuvcze (řkúce) t. 5a, mohvv (mohú) ŠtítSvátA 11b, popuudcze (popúdce) t. 125b, §uvzen (súzen) t. 207a. In documents where the diacritic markings devised by Hus have been more systematically put to use, vowel quantity is indicated in a similar fashion to Contemporary Czech—that is with a line or “čárka” above the vowel. For example: pán HustPostH 76b, k žádo§ti (k žádosti) t. 81a, bázní (bázní) Budyš 77a, blátem t. 50a, hledá§é BiblNymb 56b, ptáka t. 63a; hodné HustPostH 82a, §vatého (svatého) t., chléb Budyš 81b, mordéré (mordéré) t. 5a, gméno (jméno) BiblNymb 2b, zlaté t. 11b; viéry (viery) HustPostH 77b, bíédié (biedě) Budyš 85b, míé§to (miesto) t. 45a; nepr.í§lu§íe (nepříslušie) HustPostH 77b, §viedc.íme (svědčíme) t. 106a, chvíle … píle Budyš 27a, lúpez.níky t. 15b, wína (vína 2.sg.) BiblNymb 103b, odyímage (odji’maje) t. 76b, pa§ty.r.i HustPostH 81a, býwá Budyš 7a, blý§ká §e (blýská se) BiblNymb 51b, §wým (svým) t. 74b; bóh HustPostH 3a, vóli t. 82a, móg (mój) Budyš 22b, dolów (dolóv adv.) t. 2b, wókol (vókol adv.) BiblNymb 51b, wódce t. 75b; o §mlúwu HustPostH 81a, ne§út. 112b, bezúmý Budyš 35am búr.e t. 41b, lúpez.e BiblNymb 82a, w§túpiti §te (vstúpili ste) t. 93a. As a distinguishing feature for vowel quantity a dot above the letter is also often encountered. A period or dot over consonant letters was used to differentiate palatal phonemes. The orthographic systems of the cited documents however are not complete and simple remnants of the so-called combinatoric orthography. Diacritical orthography was not, as shown by B. Havránek and others, reliably put into use in any of the preserved old Czech documents. In addition to the above-described systems, in old Czech manuscripts we also meet with occurrences where quantity is indicated in a different manner. In this case, the markings are different from the markings for consonantal markings. These are for the

35

most part found in the orthographic systems of the oldest legends in verse. Quantity of vowels is marked in them as a double letter with the second element written or transcribed as a superscript letter, for example: leepe (le’pe) LegJidD 1a 15, liubee (l’ubé) t. 1b 6, W§§iee (všie) t. 1a 23, k ooku (k óku) LegJidM 113. In some cases an arc over the vowel letter marks quantity, e.g. tbaa LegJidM 119, raad LegDuch 72, o gineem LegPil 2b 13, keezh, zdrawiee, wzdyychanyee, w buducziich, ziimu, rsiekuu, nemaluu, nuumiely. An arc above /y/ however without a doubt does not have the same meaning, as J.Cejnar showed in the introduction to his edition of the oldest legends in verse. From a complete analysis of the marked quantity in the oldest Legends, we can also conclude that the frequency of occurrences of marked quantity (by one means or another) is not the same for all vowels. Another difficulty arises with manuscripts that mark quantity sporadically such as the Bible Olomoucká, rukopis Rokycanova Výkladu na evangelium sv. Jana (from 1492). In the latter, for instance, length is mark with a čárka (e.g. práwie) 74b, od mudrák. For this study, no OCZ texts have been cited that are known to not mark quantity. In instances where there is a question as to whether certain marks in the manuscript represent length or not, the practice has been rather to err on the conservative side and not assume quantity when there is doubt. Consequently, the number of OCZ manuscripts dating up to the middle of the 13th century that are reliable with respect to quantity is quite limited.

Chapter 4

SLAVIC ACCENTOLOGY & CZECH LENGTH

The Need for a General Theory of Slavic Accentology Tracing the origin and describing the development of phonemic length, intonationally based paradigms, ictus patterns, and vowel changes in Proto-Slavic and Czech is a complex and incomplete field of inquiry. The objectivization of linguistic data is perhaps only possible in phonetic analysis with the increasing precision of digital recording technology and computer hardware and software that quantify these data and allow for their quantifiable manipulation and study. Dialectologists, in particular William Labov (1994), have been harnessing this technology for the synchronic identification of linguistic change in process. For this study, however, we must rely on somewhat less precise data recorded by trained Czech dialectologists (typically natives from the dialect region under investigation) who wrote down, and to a lesser extent recorded on magnetic media, the word forms they heard in the field with as much phonetic precision as their ears, tape recorders, and the notation system could distinguish. We must therefore lean on another scientific principle, relativization, to bolster arguments for the historical development of Proto-Slavic and Czech based on imperfect synchronic dialectal data. The notion that humans are capable of discovering facts that they cannot directly observe has proven to be possible many times over in the sciences and the social 36

37

sciences. The study of Slavic accentology forces the historical Slavic linguist to believe in the same principle in order to support hypotheses that cannot be directly observed. Not only is it impossible to hear or read the Proto-Slavic1 language, but the scholar is equally as hampered – and at the same time challenged and lured – by the incredible complexity of the problems in reconstructing Proto-Slavic accentology. Because of a natural instability of prosodic features (Shevelov 1965:40), the reflexes of Proto-Slavic intonation have been subject to many levelings and adjustments in the individual Slavic languages, including literary Czech, which further obstructs the problem and makes it devilishly difficult to discover with any certainty how the true accentological system of Proto-Slavic was actually structured and what path of development the accentological system followed to reach the stages and systems represented by each of the Slavic languages and dialects alive today. Since our task is to discover what, if anything, Czech in its entire dialectal geographical expanse can tell us about Slavic accentology, we will address accentological issues that Czech can truly illuminate. Rather than attempt to refute or support the much debated phonological laws and morphological processes that remain at the center of studies on Slavic accentology, intonation, and prosody at this stage of the study, we will instead work within an acceptable framework of the phonological laws and

1

Following Birnbaum 1963, Proto-Slavic (PSL) refers to the reconstructed language that emerged out of the Balto-Slavic unity and lasted up until the earliest Old Church Slavonic manuscripts. When specification that is more precise is called for, reference will be made to the three stages of Proto-Slavic (PSL I, PSL II, PSL III) as defined by Goląb, 1994. The term, Common Slavic, will be used to refer to the language of the Slavs during the period when Old Church Slavonic was used chiefly as the literary language shared by all Slavs – roughly from the mid 9th century through the 12th century.

38

morphological processes attributed to Proto-Slavic, with the intention of sorting out the contribution that the Czech dialects can offer, where appropriate, to the arguments in favor of or against the operation of these laws and their corollaries. We will begin quite abruptly with several conventional tenets that affect the relationship Czech quantity has to Proto-Slavic prosody and accentology. Proto-Slavic Length The origin of length in Czech can be traced circuitously by commodius vicus back to Proto-Indo-European,2 which is not to say that the phonemic opposition of length in Czech is directly descended from Proto-Indo-European (PIE).3 Classic examples that show PIE length oppositions and their Slavic reflexes are for example: PIE /ā:a/ Slavic mati lat. māter ‘mother’ vs. Slavic nosъ lat. nares ‘nose,’ for PIE /ō:o/ Slavic darъ lat. dōnum ‘gift’ vs. Slavic oko lat. oculus ‘eye.’ These words are typically cited to show that Early Proto-Slavic (EPSL) merged the PIE long back vowels /ā/ and /ō/ as Slavic /ā/ and the PIE short back vowels /a/ and /o/ as Slavic /o/. PIE oppositions in length are evident in all Slavic languages in the distribution of what are today morphophonemically conditioned ablaut alternations which in pre-dialectal Proto-Slavic include the alternations /o/ ~ /a/, /e/ ~ /ě/, /#/ ~ /i/, /#/ ~ /y/.

2

The question about whether an opposition in intonation (rising pitch:falling pitch) was inherited from Indo-European is beyond the scope of this study and irrelevant to the development of prosodic patterns in Proto-Slavic and Czech. Whether accent was fixed on the penultimate (or some other) syllable at the dissolution of Indo-European and the later development of free ictus occurred when the rise of phonemic intonation caused a redistribution of accent patterns in BaltoSlavic (or again only in Slavic) paradigms is admirably treated in Shevelov (1965). 3

With the exception of Balto-Slavic vowels lengthened before voiced consonants according to Winter’s law (See Collinge 1985:225-227).

39

The point of departure for quantity in Czech is the earliest stage of (post BaltoSlavic) Proto-Slavic which we take to have inherited an opposition in length from ProtoIndo-European (PIE). The development of phonemic length in Czech must seek its source and explanation in the Proto-Slavic period when through a complex relationship of ictus, tone, morphology, morphophonemics, and vowel quantity – phonemic length emerged once again in Late Proto-Slavic (LPSL) only to be reshuffled in certain ProtoSlavic dialects, fortified in others, and obliterated in still others. Most prominent among the changes involve two general tendencies. The first was the redistribution of the PIE vowel system from one based on quantitative oppositions to a system based on qualitative distinctions. The second was the tendency towards open syllables, which, among other things, contributed to the elimination of nasal and oral diphthongs. The inherited PIE quantitative oppositions in Early Proto-Slavic (EPSL) developed into qualitative distinctions, which allowed quantity in EPSL to become nondistinctive (redundant) until new quantitative oppositions evolved in LPSL through the interplay of ictus shifts, the shortening of long vowels and diphthongs under certain conditions (phonological as well as morphological), as well as the lengthening of previously short vowels and short diphthongs under certain conditions (phonological as well as morphological). Nevertheless, vowel quantity seems still to have been a characteristic of the qualitatively differentiated Proto-Slavic vowel system that persisted into LPSL. Without going through all the details, which are well documented (Shevelov 1965, Horálek 1963, etc.), we should say that by LPSL (Goląb’s PSL III), the vowel system had radically changed. PIE /*ē/ changed qualitatively by EPSL to /*æ/ and

40

continued as /ěî/ [æ] in LPSL. PIE /*e/ became /*æ/ in EPSL and raised to /e/ in LPSL which is the end result. PIE /*ū/ by LPSL > /y/ (either by push or pull chain in connection with the change of PIE /*au, *ou > LPSL /ū/) and PIE /*u/ lowered and became lax to /ъ/ by LPSL. PIE /*i/ / lowered and became lax to /ь/ in LPSL, and PIE /*ī/ remained in LPSL as /ī/ but the length became redundant phonemically when PSL /*i/ lowered and became lax to /ь/. LPSL /ū2/ developed from PIE and PSL /*au, *ou/ diphthongs, and /ěî2/ (phonetically identical to /ěî/) developed from PIE /*oi/ diphthongs, except in certain morphological categories (e.g. o-stem masc. anim. NP desinence {-oi#}) where /*oi/ > /ī2/. The PIE /*ei/ diphthong by LPSL had become /ī2/ as well. Nasal diphthongs /*en, *em, *in, *im, *em, *on, *om, *an, *am/ were simplified in PSL in tautosyllabic positions to /ę, o/, the timbre of the new nasal vowel reflecting the timbre of the initial segment of the nasal diphthong (i.e. original rounded nasal diphthongs > /o/ [], original non-rounded nasal diphthongs > /ę/). By the LPSL vocalic system, /e, ь, o, ъ/ are considered short vowels, and /ā, ěî, ěî2, ę, o, , ī, ī2, y, ū 2 (< *ou, *au)/ are considered long vowels. The liquid diphthongs (or, er, ol, el) are also considered long at this stage. Liquid vowels /l•, r•/, originally long in EPSL, developed in post-dialectal LPSL with long and short variants. This distinction is preserved somewhat in dialects of Czech and under certain conditions following palatal consonants and dentals in CLC. Prosodic features that characterized the LPSL phonological system included: 1) intonation (+high tone:±tone), 2) quantity (+long: ±long), and 3) force (+ictus: ±ictus).

41

Length could occur on any syllable of the word, thus, there were words with one, more than one, or zero long syllables. Ictus, on the other hand, could fall on only one syllable, and was automatically assigned to the first syllable (counting from the left) with acute (high) tone. If there were no syllables in a word with acute intonation, circumflex intonation (stressed) developed on the first syllable of the word. Tone in Proto-Slavic, therefore, only contrasted on the initial syllable. New quantitative oppositions were formed in LPSL because of phonetic shortenings under the following conditions: Word final long vowels in disyllabic and polysyllabic word forms, but not in monosyllabic word forms (see page 47); In unstressed syllables that were not followed by a short stressed syllable4; Stressed long syllables that were at least two syllables from the word final syllable, which often means stressed initial long syllables in polysyllabic words; e.g. CZ ulice, lahoda, lastovice, jahoda, palice, měsíc. Under stress in word internal syllables (however inconsistently); e.g. motyka, lopata, rodina, kopyto, kobyla, poleno, hromada, otava, beseda, rokyta, koleno, koryto, hovado. Post-tonic syllables under certain conditions (see Shevelov 1965:518-520). The new shortenings once again created phonemic oppositions based on length, but dialectal features no doubt existed even during the earliest period of a more compact Slavic unity, and further dialectal distinctions seem to have understandably accelerated

4

In pretonic position before a short or long PSL vowel, length seems to have been retained; e.g. útroba, útěxa, koupel, koukol, stoupati, tráviti, kouřiti, žádati, vrátiti.

42

tangentially with the migration of the Slavs. The phonological systems of the individual Proto-Slavic dialects slowly began to emerge along the dialectal geopolitical plane. The development or abolishment of LPSL phonemic quantity, together with the development of paradigmatic patterns based on the LPSL intonational system with distinctions between acute, circumflex, and neo-acute intonation, and the consolidation or abolishment of free ictus in each of the Proto-Slavic dialects must have been interconnected. One might argue that such a cumbersome vocalic system was hardly sustainable, and in fact there are no contemporary Slavic languages that preserve as robust a system of quantitative and qualitative oppositions as in the period of Slavic dialectal dissolution. The East Slavic languages of Russian, Byelorussian, and Ukrainian show no residue of this original opposition. New lengthened short vowels in Russian and Byelorussian may provide evidence for the persistence of pitch or intonation distinctions on short vowels in East Slavic, but are of little comparative interest for Czech. The West Slavic languages, Czech, Slovak, and to some extent Upper Sorbian, have richly developed phonological and morphological systems which make use of oppositions in vowel quantity, whereas in Polish phonemic length survived until the fifteenth century and in Kashubian survived until the nineteenth century. Polish and Slovak are of limited help because the reflex of vowel length is detectable in few environments; in Slovak only in neo-acute syllables (Horálek, 1962:122, fnt.) not in acute syllables, and pretonically (SK trúba ‘pipe’). Reflexes of pretonic length in Literary Polish show up only in nasal vowels and /o/ in certain phonological environments, e.g. P trąba ‘pipe,’ krół ‘king.’ Certain Polish dialects show reflexes of long /ē/ and long /ā/. Upper and Lower Sorbian

43

are cited occasionally for their reflexes of the *TorT and *TelT groups, and Slovincian has gathered new significance in the equation since Stankiewicz’s analysis of 1993. In South Slavic, length is preserved under certain conditions in Serbo-Croatian and Slovene, in Bulgarian and Macedonian it is not. Serbo-Croatian and Slovene data will be cited often to support the classification of Proto-Slavic words into the Proto-Slavic accentual paradigms to which they belong. The Role of the Dialects Precisely when in history it is possible to speak about Czech as an independent language as opposed to a dialect of Proto-Slavic depends on one’s interpretation of what a dialect is versus what a language is. Some scholars argue that because the loss of the weak jers took place in all the Slavic languages at roughly the same time and ended the tendency towards syllable structures with increasing sonority, it is not possible to talk about individual Slavic languages until after the loss of weak jers (Birnbaum, 1963:2). Others argue that the difference between a dialect and a language is a political determination dependent upon national boundaries with certain other requirements such as a codified grammar based on a common body of literature, mutual intelligibility of all dialects of the language, a sense of ethno-linguistic identity, and the weight and number of isoglosses separating the dialect in question from its neighboring dialect(s). There does not seem to be a universal solution to the problem short of declaring a moratorium on the term language and referring only to dialects. Yet inevitably we would have to confront the question “dialect of what?” and we are back to talking about languages and dialects.

44

Most Slavic accentological studies (with some notable exceptions)5 focus on the literary languages for accentological arguments, for, as Stankiewicz (1993:2) explains “the selection of the contemporary literary languages as the primary object of study is motivated by the fact that most descriptions of these languages (in both grammars and dictionaries) provide exhaustive material for a structural and functional analysis of their accentual patterns, whereas no such materials are available for the accentual study of the dialects and the older stages of their development.” The implication here is not that one should depend exclusively on the literary language, but that often the most complete descriptions of relevant phenomena are available about the literary language and not about most other dialects. The same is true of Czech. Despite the excellent dialectological work being continually published by Czech dialectologists in the form of the Český jazykový atlas, complete structual analyses of the Czech dialects are still lacking.6 The methodological faults and fissures that arise from using data from one narrow dialect (be it the literary language, a composite of dialects, or a remote dialect) are many in number. In the context of the present study, one of the obstacles Slavic accentologists run into who rely on the Czech literary language alone to draw conclusions about the entire Czech speaking territory or to generalize about Proto-Slavic accentuation, is that the source of CLC is anything but a unified homogeneous (either

5 6

Dybo (1993), Bulaxovskij (1983).

That is not to say that structual analyses of the Czech dialects do not exist. Bělič (1972), although modestly titled as Nástin české dialektologie [A sketch of Czech dialectology] is in reality extremely detailed and reasonably thorough. In addition, a number of excellent and adequately complete dialectal studies – Voráč (1955), Bělič (1953), Vydra (1923), to name but a few, have been published which add measurably to the accuracy of historical studies.

45

historically speaking or synchronically speaking) entity. The literary language is an artificial construct and is in fact an amalgam of various dialects, the most dominant of which are the Central Bohemian Dialects. The end result of relying solely on CLC for historical studies is that the view of the developments becomes blurred by the narrowness of the data set. To be sure, quoting one form from CLC as the normalized form of a word makes the task of representing the form for accentological comparisons less onerous, but by analyzing the development of quantity in the dialects (including the literary language), we are able to make more precise use of Czech data to help answer the various questions that remain puzzles for historical Slavic linguists and philologists. For example: Did the so-called neo-circumflex exist as a distinct Proto-Slavic intonation, Proto-Czech/Slovene intonation, or neither? What are the likely territorial origins of compensatory lengthening and the neo-acute intonation and how does Czech quantity help support or refute a centerperiphery theory as shown by Timberlake (1983a, 1983b)? What relationship is there, if any, between the fixing of ictus on the initial syllable of the phonological word in Czech and the rise of the neo-acute? Was the shortening of vowels under the original acute intonation a Proto-Slavic phenomenon or was it already dialectal? Slavic Accentual Paradigms (a.p. = accentual paradigm) The consensus view of Slavic accentology (Stang 1965, Shevelov 1965, Bulaxovskij 1983, Dybo 1990) posits four accentual patterns or paradigms (a.p.) for Proto-Slavic.7 These four accentual paradigms are referred to variously, but the

7

Dybo et al. (1993) believe they have identified a fifth type (d), a variation of the oxytonic a.p. with accent on the NAS root rather than on the desinence.

46

important characteristics can be summed up by the reconstructed paradigmatic accent alternations of the Proto-Slavic system. There were two columnar fixed accent patterns – one on the root [a.p. (a)] and one on the first syllable of the desinence [a.p. b)]. There was one mobile pattern [a.p. (c)]. In the (j)ā-stems there is a fourth type – one with fixed accent and neo-acute intonation on the stem. The neo-acute intonation is undoubtedly a later development that is characterized by the retraction of accent one syllable to the left. In Czech, as evidence from dialectal data suggests, there seems to be a correlation between lengthening of vowels and the neo-acute intonation. The identification and classification of words within these accentological paradigms is based on somewhat regular correspondences in the Slavic languages. In combination with place of ictus in Russian, pitch and length in Serbo-Croatian and Slovene, and to some extent place of ictus in Bulgarian, Czech length or the absence of it is often cited as evidence of the belonging of a lexeme to one or another of the four accentual paradigms. In a.p. (a) with a fixed stem stress pattern, acute intonation was correlated with fixed stress on the stem of non-derived disyllabic words (in contrast to derived words, e.g. R. gotov, with a morphologically derived fixed stress and not a fixed stress as a reflection of the correlation between fixed stress and acute intonation). The a.p. (b), according to Illič-Svityč and Dybo, is the result of a Balto-Slavic shift from a fixed short or circumflex intonation on the stem to the initial syllable of the desinence. From the point of view of Slavic, however, a.p. (b) can be considered a morphological class. In a.p. (b) with a fixed columnar desinential stress pattern, acute intonation was also correlated with fixed stress on the initial long syllable of the desinence. In the mobile

47

a.p. (c) pattern, circumflex intonation is found on the word initial syllable (including proclitics) when the stress falls on that syllable and the syllable is long. If the stress falls on the initial syllable and this syllable is short (o, e, ъ, ь), a non pitch accent is posited. Stang (1965:42-45) claims that there were some cases of non-initially accented circumflexes, but all are in suspiciously morphological environments and the clear cases are instances of oxytonic accent on an internal syllable. These vowels lost their ictus through retraction with the resulting neo-acute intonation. Among the jā-stems (e.g. CZ vůle) there is another type of stem with neo-acute intonation characterized by retracted stress. The neo-acute introduced a new rising intonation which probably encouraged the loss of intonation on the old acute. Except for the Czech (and perhaps Upper Sorbian) periphery, all Slavic languages neutralize the old acute by shortening the long vowel. Correspondences in the contemporary Slavic languages provide details by which these four accentual types are identified for Proto-Slavic words. Czech Quantity and Proto-Slavic Accent With respect to these accentual types in Modern Czech, Gebauer was first to treat the historical development of Czech quantity as a reflex of Proto-Slavic accentuation (Gebauer 1894:586-613). He based his assumptions on the accepted theory still valid today (Miklosich, Über die langen Vocale in den Slavischen Sprachen, 1879:312) that Proto-Slavic syllables with front and back jers, /ъ/, /ь/, and the vowels /e/,and /o/, were short and syllables with /i/, /y/, /ě/, /a/, /u/, front and back nasal vowels, /ę/ and /o/, and the liquid diphthongs /er, el, or, ol, ъr, ьr, ъl, ьl/ were all originally long. Citing examples such as *mąka ‘flour’ (CZ mouka) vs. *mąka ‘torture’ (CZ muka), *nesú indic. 3. pl.

48

(CZ nesou) vs. *nesą indic. 1. sg. (CZ nesu), Gebauer concludes that Proto-Slavic quantity was in flux but he does not attempt an in depth analysis of the relationship between Late Proto-Slavic intonation and Czech quantity beyond recognizing that there is some identification between the two. It is interesting to note, however, that Gebauer leans towards a view of Czech quantity that places the burden for Czech length more on the historical development of Czech than on the reflexes of Proto-Slavic prosodic features. He interprets, for example, length in derived imperfectives such as vídati (cf. viděti ‘to see’), slýchati (cf. slyšeti ‘to hear’), and nakládati ‘to load’ (cf. first singular kladu ‘I put’), not as the preservation of Proto-Slavic length, but as secondary Czech lengthenings. He also states quite clearly that contemporary Czech quantity is for the most part the same as Old Czech quantity as represented in the Old Czech manuscripts which mostly originate from the Central Bohemian dialects. "Je tedy kvantita česká původu většinou podružného. Ale tím není řečeno, že by to byl původ pozdní a nedávný. Naopak, jednak změny hláskoslovné, které jsou účinkem délky (úžení a rozšířování), a jednak stará forma psaná dosvědčují bezpečně, že kvantita novočeská je celkem t.j. valnou většinou táž, jako v historické době staročeské" (Gebauer 1894: 338). [Quantity in Czech is therefore from various sources. But this does not mean that the origin is a later, more recent development. On the contrary, in part phonological changes, and in part the old written forms clearly show, that new Czech Quantity is for the most part, that is, in most cases, the same as in the Old Czech historical period.] This seemingly mundane observation by arguably the greatest authority on Old Czech ever to walk the planet turns out to be very significant. If Gebauer is right, it

49

means that the development of Czech quantity from Proto-Slavic accent was primarily prehistoric. Although our view of the development of prehistoric Czech is compressed by centuries of invisible development, the comparative method and internal (dialectal) reconstruction hints at a variety of demonstrable changes that occurred in the centuries between the dialectal dispersion of the Slavs and the first Czech manuscripts. Concerning changes in Czech quantity, Gebauer distinguishes between old and new differences in quantity. The older distinctions, he points out, show up qualitatively, which presupposes the shifting of the Early Proto-Slavic vowel system from one based on quantity to the Late Proto-Slavic vowel system based on qualitative distinctions. The development of the second stage of the Late Proto-Slavic vowel system however, has been the subject of conjecture and debate. The argument for the Late Proto-Slavic development of a new opposition in length stems from the belief that “all long final vowels in di- and polysyllabic words, whether stressed or unstressed, were shortened” and that “no Slavic language now reveals any original length in this position” (Shevelov 1966:508).8 For Czech and Slovak, however, this statement may not be true as shown by Old Czech and attested forms in Southwestern Bohemian dialects. Leaving aside for the moment lengthening in auslaut position from contractions (e.g. /oje > é/ etc.), which are for the most part later lengthenings, Old Czech indicates that word final long vowels did

8

Dybo (2000:37-43) states that only unstressed word final long vowels were shortened in Proto-Slavic and identifies ten (10) environments in a.p. (b) and a.p. (c) with stressed acute endings. He presents arguments that reflexes in various Slavic dialects, mostly in S/C and SLN, either directly in the form of long vowels or indirectly by triggering the neo-circumflex provide evidence that the shortening of word final long vowels was phonological in unstressed desinences only.

50

in fact remain long in some cases, e.g. dial. obá, dvá, nikdý Háj herb 69b, někdý t. 191b, nihdá Btch. 446, sotvá Bart D. 8, OCZ nom. třie, ctyřie, adv. silnějie (Gebauer 1894:339). Strong aorist forms in the 3rd person pl. for several verbal paradigms also preserve length (e.g. vedú, bolú, jidú, nesú, etc.). Trávníček (1935:249) argues that the length in the 3rd pl. aorist forms is by analogy to the present 3rd person pl. of the same paradigms where length was presumably preserved before a final soft /-t’/ that was later lost. We find this hypothesis difficult to accept. Syncretism destroyed the tense opposition in prehistoric Czech between the 3rd person pl. aorist and the 3rd person pl. indicative forms, but it is unlikely that the aorist form would have tended towards the indicative form and adopted by analogy the long desinence because by doing so the tense distinction would have been neutralized. We find it more plausible to believe that the length in the aorist form is preserved (stressed) length and that the 3rd person pl. aorist and the 3rd person pl. indicative forms coalesced when the soft /-t’/ was lost. In other words, the syncretism of the 3rd person pl. aor. and the 3rd person pl. indicative forms was caused by the loss of final /-t’/ in the 3rd person pl. indicative and not by analogical extension of length from the indicative to the aorist. Moreover, in Slovak dialects we find length in the o- and u-stem masc. IP forms such as chlapý, vratý9 and even ā-stem IP forms such as cestamí. Trávníček attributes these lengths either to some local innovation or to the local spread of neo-acute intonation, which determines the Slovak preservation of length. Central Slovak dialects also show length in neuter o-stem

9

Instrumental plural in {–ý} (long) is widespread and according to Stang (1965:31) occurs in Čakavian with the neo-acute accentuation.

51

nominative/accusative forms such as slová, mestá, okná, poliá. The origin of length in these forms is questionable. One explanation is the possible transfer of neo-acute length to this class of substantives only in the Central Slovak dialects that the Slovak literary language is based on. Slovene, however, also shows the reflex of neo-acute (rising intonation in the final syllable) on several cognate endings. So, perhaps the length in these forms is truly a preservation of Proto-Slavic neo-acute length that spread in endings before the loss of acute length and was shortened in other Slavic dialects including the dialects that literary Czech is based on. Nevertheless, one detail that is worth noting in this connection is the fact that only original long vowels in either Slovak or Czech may be dialectally long in word final position. We know of no instances where original short vowels in word final position show up as long, which may provide additional evidence that the lengths in question are preserved Proto-Slavic lengths. Dynamic Ictus and Fixed Ictus In Proto-Czech, as in LPSL, ictus was still dynamic and could fall on any syllable of the Proto-Czech word. Hujer (1936:9) notes that there probably were not many words in Proto-Czech that had the ictus fixed on the same syllable throughout the paradigm. He does not explain why he believed this, but subsequent scholarship has adequately demonstrated that the fixed ictus types [a.p. (a), a.p. (b), neo-acute] are very common. It is impossible to determine how long dynamic ictus lasted in Proto-Czech since ictus is not marked in any Old Czech texts as far as we can tell. We have some historical evidence that in the late 8th and the early 9th centuries OCZ still had dynamic ictus. The Slavs by way of Western Slavic dialects borrowed their word for king from the Franks as

52

an end-stressed substantive *karlь (from Karl Grosse) which later developed under the neo-acute intonation with retracted ictus in the nominative case. We also know from Old Czech words borrowed into Germanic at the time that in the 13th century, Old Czech had already fixed the ictus on the first syllable of the word. In Contemporary Czech dialects we find quite a few exceptions to fixed ictus on the first syllable of the word. In the Silesian,10 Oravian, and Eastern Slovak dialects the ictus is on the penultimate syllable as it is in Polish. It would logically follow, therefore, that Polish has influenced these dialects, but there are other dialect regions in the Czech lands and Moravia that are not contiguous with Polish dialects where the ictus is also on the penultimate syllable (hránické nářečí, starojické, na Karlovsku a Rožnovsku, na Zlínsku, u Holešova, Kroměříže, Přerova). The same process which fixed the ictus in Polish on the penultimate syllable most likely acted independently in the Czech dialects mentioned above, but we will probably never be able to determine whether or not this is true. At any rate, there can be little doubt that the loss of dynamic ictus in prehistoric Czech and the fixing of ictus on the initial syllable of the phonological word influenced the interplay of intonation and quantity. We surmise that since in many word forms the ictus would have been retracted in order to fall ultimately on the initial syllable, that this retraction would have tended to have, if anything, a lengthening effect on the syllable onto which the ictus was being retracted.

10

In other texts (e.g. Hujer, 1936), these dialects are sometimes referred to as the lašská nářečí from the word lach of uncertain origin (but used in the Primary Chronicles). Bělič (19671:12, fnt.) notes that the term has a negative connotation in CZ (ničema ‘good-for-nothing,’ darebák ‘fool’) and therefore prefers the term slezká nářečí for the Silesian dialects.

53

Acute a.p. (a) Literary Czech is often generally described as preserving acute intonation as length on the first syllable of original disyllabic forms of the fixed accent class a.p. (a). That is not to say that there are not other sources of length in Czech (e.g. the coalescence of –VjV– clusters pás < *pojasъ ‘belt;’ neo-acute length kůň < *konь ‘horse,’ compensatory lengthening dial. plouh < *plugъ ‘plow’: morphological lengthening déle/dýl, comparative of dlouhý ‘long;’ derivational lengthening Dáša dim. of Dašenka), or that the preservation of acute length is regular in any grammatical sense of the word regular. In fact, the most that can be said of original acute intonation is that a number of PSL words of the fixed stem-ictus a.p. (a) have long vowels in Literary Czech. According to this traditional view, if we examine, for example, a lexical item such as Czech sláma (< PSL *solma) ‘straw,’ the length of the vowel /á/ in the Czech word agrees with other evidence that the Proto-Slavic word contained a root vowel or diphthong whose intonation was acute, (sometimes referred to as high-pitch or rising). In Serbo-Croatian (and South Slavic in general), however, Proto-Slavic acute intonation shows up as brevity. S/C slama with a short falling accent shows the expected reflex of Proto-Slavic acute intonation in S/C. In Russian, which does not distinguish between long and short vowels, a remnant of the Proto-Slavic acute intonation can be seen in the paradigmatic ictus patterns of the words with polnoglosie. Russian solоma with columnar ictus on the second syllable of a polnoglosie word supports the notion that the place of ictus was on the initial syllable and in this particular instance the intonation was acute. Slovene sláma with long-rising intonation is further evidence that the Late Proto-

54

Slavic term for ‘straw’ contained a diphthong /ol/ with acute intonation later identified with fixed place of ictus in Russian. Similarly in CZ rádlo ‘plow,’ which occurs in a representative selection of contemporary Slavic languages in different forms, the length of the vowel /á/ in the Czech word provides evidence that the Proto-Slavic word contained a root vowel or a diphthong that was high in pitch (acute) and that had the ictus fixed on that syllable throughout the paradigm.11 Serbo-Croatian (S/C) ralo tells us that the vowel in LPS fell under the acute intonation since /a/ was a long vowel and there is no rule to shorten a long vowel in PS except under the original acute intonation. The vowel could have been long or short as far as the S/C reflex is concerned. The short falling intonation, however, is the expected reflex of Proto-Slavic12 (stressed) acute intonation. Russian ralo indicates that the place of ictus was on the first syllable and since the ictus is fixed on the first syllable throughout the paradigm, we have additional support for the claim that the word was accompanied by acute intonation in Proto-Slavic. Slovenian (SLN) rálo with long-rising intonation is further evidence that the LPSL term for ‘plow’ contained a long low vowel with acute (rising) intonation, which was identified with fixed place of ictus.

11

The actual quality of the intonation, whether it was rising and then falling, steadily rising, or falling then rising is irrelevant and probably impossible to determine. 12

The importance of distinguishing between the periods of Common Slavic with respect to the length of vowels is discussed in Birnbaum (1963) and Goląb (1994) etc. In this thesis, we agree that length became phonemic in LPSL after the shortening of long vowels in various positions.

55

Oxytonic a.p. (b) Czech also presumably preserves pretonic length in disyllabic words, which means that this class is impossible to distinguish from a.p. (a) by simply looking at the Czech form. A good example of an oxytonic ā-stem in Slavic that shows the preservation of pretonic length in Czech is *tromba ‘pipe,’ CZ trouba (GP trub), SK trúba, S/C trúba, R. truba (AS trubu), SLN tróba, P. trąba. A good example of an o-stem oxytonic word with preserved pretonic length in Czech (and Slovak) is *bykъ ‘bull, ox,’ CZ býk, SK býk, R byk (AS byka), S/C bîk (GS bíka), SLN bìk, AS bíka. Mobile a.p. (c) The Czech word hrad ‘castle,’ R gorod, S/C grâd, SLN grâd ‘city,’ with a short /a/ in Czech, ictus on the first syllable of a R polnoglosie word in the NS, and falling intonation in S/C and SLN indicates that the Proto-Slavic word *gordъ contained a diphthong /or/ that did not fix the accent on that syllable, but rather adhered to a mobiletype accentual pattern.13 More simply put, acute intonation in LPSL shows up in CLC often as length in disyllabic forms of the paradigm, circumflex intonation in LPSL shows up in CLC as brevity throughout the paradigm. This statement, however, as we shall see turns out to be far from formulaic for CLC and there is a great deal of divergence in the dialects. Nevertheless, the dialectal geography and distribution of length vs. brevity in

13

The actual circumflex (long falling) intonation occurred in this mobile type of word only when the ictus fell on the first syllable.

56

the fixed stress stems seems to indicate some patterns that provide a clue to the conundrum of the relationship between PSL accent and Czech quantity. To take another example, the Proto-Slavic word *golva ‘head,’ Czech hlava, Russian golovа, accusative golovu, Serbo-Croatian gláva, Slovene gláva; with the pattern short /a/ in Czech, mobile ictus in the Russian polnoglosie paradigm, and long rising intonation in Serbo-Croatian indicates that the word contained a diphthong /ol/ that that did not fix the accent on that syllable, but rather adhered to a mobile-type accentual pattern and therefore had circumflex intonation when the ictus fell on the word initial syllable in Proto-Slavic. The circumflex intonation is defined as the intonation on the initial syllable of a mobile class stem when that syllable is under stress. There is a sort of puzzle to be worked out between Czech and the other Slavic languages that preserve a distinction between long and short vowels in that circumflex length is preserved in S/C and Slovene whereas in Czech the reflex of circumflex length is brevity.

Chapter 5

FEMININE DISYLLABIC (J)Ā-STEMS

Expected reflexes of root vowels in the Slavic Languages Quantitative patterns in the (j)ā-stem feminine substantival paradigms in the Czech dialects reveal some striking facts about the development of quantity in Czech. Table 3, adapted for use from Dybo (1990:13) with several amendments, illustrates the expected reflexes of Proto-Slavic intonations in the most useful Slavic languages for accentological studies that preserve a distinction either in intonation, quantity, or in both. These are the patterns that we have good evidence for believing represent the various Proto-Slavic accentual types. For Czech, the table implies that in monosyllabic and disyllabic forms, length is possible in all stems except on circumflex (mobile) stems. For trisyllabic forms, length is possible in pretonic originally long initial syllables and neoacute pretonic originally long syllables (initial and internal). Length under the neo-acute intonation in trisyllabic forms developed in Czech after the ictus was retracted (Stang’s law), but under the acute or circumflex intonation, length was lost. We believe that the neo-acute intonation that arose from the retraction of accent one syllable to the left in Proto-Slavic conditioned lengthening of the short vowels in Proto-Czech dialects in certain classes of substantives and morphological desinences that spread to the roots of other substantives with short vowels of the oxytonic class.

57

58 Table 3 – Expected correspondences of Proto-Slavic intonational patterns Proto-Slavic Intonational Paradigm Language

No. of Syllables

Acute

Neo-acute on long syllable

Neo-acute on short syllable

Circumflex on long syllable

Circumflex on short syllable

S/C

1 syllable 2 syllable 3 syllable

|  | |  |  | |  |  |  |

| -- | | -- |  | | -- |  |  |

|  | |  |  | |  |  | |

| -- | | -- |  | |  |  |  |

| -- | |  |  | |  |  |  |

S/C Čak.

1 syllable 2 syllable 3 syllable

|  | |  |  | |  |  |  |

| - | | - |  | | - |  |  |

|  | |  |  | |  |  | |

| -- | | -- |  | |  |  |  |

| -- | |  |  | |  |  |  |

S/C Old Croatian

1 syllable 2 syllable 3 syllable

|  | |  |  | |  |  |  |

| - | | - |  | | - |  |  |

|  | |  |  | |  |  | |

| -- | | -- |  | |  |  |  |

| -- | |  |  | |  |  |  |

S/C Kajkavian

1 syllable 2 syllable 3 syllable

|  | |  |  | |  |  |  |

| - | | - |  | | - |  |  |

|  | |  |  | |  |  | |

| -- | | -- |  | | -- |  |  |

| -- | | -- |  | |  | -- |  |

Slovene

1 syllable 2 syllable 3 syllable

|  | | - |  | | - |  |  |

| - | | - |  | | - |  |  |

|  | | - |  | | - |  |  |

| -- | |  | -- | |  | -- |  |

| -- | |  | -- | |  | -- |  |

Czech

1 syllable 2 syllable 3 syllable

|- | |- | | | | | |

|- | |- | | |- |- | |

|| || | || | |

|| || | || | |

Slovak

1 syllable 2 syllable 3 syllable

|| || | || | |

|- | |- | | |- | | |

|||| | | |

|| || | || | |

|| || | || | |

S/C, S/C Čak, Old Croatian, S/C Kajkavian |  | = short falling

| | | | | | |

| | | |

Slovene

Czech and Slovak

| -- | = long falling

| -- | = long falling

| - | = long vowel

| - | = long rising

| - | = long rising

|  | = short rising

|  | = short rising

|  | = short vowel without distinctive intonation

|  | = short vowel without distinctive intonation

|  | = short vowel | | = unspecified for length – could be either a long or a short vowel.

59

One of the peculiar differences between Czech and the other Slavic languages with oppositions in quantity is that Czech seems not to show length where the others do, and does show length where the others do not, except in positions where the ictus was retracted. When the ictus was retracted as in the oxytonic stems of a.p. (b) and the neoacute originally long jā-stems, Czech (with Slovak) seems to have preserved length rather more consistently than the other Slavic languages, and in the neo-acute originally short jā-stems, Czech consistently lengthens the short vowel. Deriving the CLC quantity patterns for (j)ā-stem paradigms from the PSL accentual paradigms is instructive in understanding the Czech developments. If we look at the four LPSL accentual paradigms: acute [fixed root stress = a.p. (a)], oxytonic [fixed stress on first syllable of the desinence = a.p. (b)], circumflex [(mobile = a.p.(c)], and neo-acute (retracted stress), we can initially derive the following paradigms for the feminine (j)ā-stem disyllabic substantives according to the historically determined rules for the development of length in Czech: Table 4 – Acute disyllabic feminine (j)ā-stems in CLC with original long root vocalism Case / Number

Proto-Slavic Form according to Stang (1965:56)

Rule for Preservation or Loss of Length in original Long Root Vowel

Expected CLC Form

Old Czech Form

Literary Czech Form

NS

*ko!rv a

Acute length preserved

KRÁVA

kráva

kráva

GS

*ko!rv y

Acute length preserved

KRÁVY

krávy

krávy

DS

*ko!rv ě

Acute length preserved

KRÁVĚ

krávě

krávě

AS

*ko!rv o

Acute length preserved

KRÁVU

krávu

krávu

IS

*ko!rv ojo

Acute length lost in word greater than two syllables

KRAVOU

kravú krávú

kravou krávou

LS

*ko!rv ě

Acute length preserved

KRÁVĚ

krávě

krávě

60

Case / Number

Proto-Slavic Form according to Stang (1965:56)

Rule for Preservation or Loss of Length in original Long Root Vowel

Expected CLC Form

Old Czech Form

Literary Czech Form

NP

*ko!rv y

Acute length preserved

KRÁVY

krávy

krávy

GP

*ko!rv ь

1). Acute length preserved, 2). Morpho-phonemic loss of length in GP

KRAV

krav / kráv

krav1

DP

*ko!rv am ъ

Acute length lost in word greater than two syllables long

*KRAVAM

kravám

kravám

AP

*ko!rv y

Acute length preserved

KRÁVY

krávy

krávy

IP

*ko!rv am i

Acute length lost in word greater than two syllables long

KRAVAMI

kravami

kravami

LP

*ko!rv ax ъ

Acute length lost in word greater than two syllables long

*KRAVACH

kravách

kravách

We should note here that there are no paradigms in CLC that match the expected result of the PSL acute (j)ā-stem paradigm for every grammatical case, although the CLC F Type paradigm with shortened root vowel in the IS, GP, DP, IP, and LP comes very close. The only deviation from the expected results in Type F is in the DP and LP desinences. Although phonologically we would expect length on the root in the GP (attested in many examples from OCZ), a morphophonemic rule operated for a time across the (j)ā-stem paradigms that shortened the root vowel in the GP. Wherever there is an alternation of length in a CLC (j)ā-stem paradigm with a long root vowel (Types D, E, and F above), the GP participates in the alternation and the root vowel is short. Gebauer (III.I.:179) lists GP strák, múk, rýb, attested in OCZ from the acute a.p. (a), which means that either there was some residue of phonologically retained length in the 1

Stang (1957:96 & 168) reconstructs {*-ōm/n} despite Meillet (MSL 22, p. 258) maintaining that the Slavic ending {–ъ} must be from {*-om}.

61

GP, or length from other forms of the paradigm spread to the GP. However, when one considers that even in the OCZ literary language (of Central Bohemian origin) NS straka, NS muka, and NS ryba had short root vocalism throughout the paradigm, it becomes obvious that the exceptional length in the GP attested in OCZ for these words is from a different source. The behavior of length in the GP has been attributed to two major factors, either: 1) the neo-circumflex shortening of original acute length, or 2) the generalization of an alternation in quantity between the GP and the rest of the paradigm in order to differentiate the GP. For the OCZ attested length in the GP of originally acute (j)ā-stems (OCZ GP strák, múk, rýb) that lost length in the rest of the paradigm, the best guess is that the length, even though it may have originated from a phonological retention of acute length in the GP of the fixed stem stress a.p. (a) class, by the time we have historical documents must be attributed to the generalization of an alternation in quantity between the GP and the rest of the paradigm in order to differentiate the GP. It is important to notice additionally for ā-stem paradigms in CZ, that the DP desinence {-ám} and the LP desinence {-ách} were generalized very early in OCZ to most ā-stems.2 The source of the length in these desinences is uncertain, but the length is

2

Gebauer (III.I.:181) observed that LP {–ách} in OCZ sometimes is shortened, especially in the Chod dialects and that the DP {–om} from the o-stems is also attested in the Chod dialects Chod DP kravom. It is possible that the brevity in these desinences represents a residue of the expected CZ phonological reflexes and are not shortenings at all. Jančák (1970:129) makes the following observations;“O stáří těchto krátkých západočeských koncovek –am, -ach nelze říci nic určitého. Jejich zeměpisné rozložení shodující se s typem –om však ukazuje, že impuls k těmto nářečním novotvarům mohl vyjít právě od krátkých tvarů na – om, zvlášt’ u anologických podob, např. v žen. voda, nastala totiž kolize mezi krátkým typem ‘kravom’ a původním typem ‘ženám.’ Její vyrovnání si pak ovšem vynutilo obdobnou změnu; u korespondujících tvarů, a to jak v témže paradigmatu, tak v tvarech analogických u jiných vzorů

62

restricted to Czech, Slovak, and Polish dialects and might very well be connected to the loss and vocalization of jers or to a spread of the neo-acute intonation in these desinences. Length in Czech that is not PSL acute length or neo-acute length is often called compensatory lengthening. In the u-stem LP where the desinence was {*-ъ/ь-xъ} (e.g. *dom-ъ-xъ), the vocalized jer does not become a long vowel (cf. CZ domech). Original short /o/, however, in the o-stem DP {*-o-m-ъ} does become long (e.g. chlapomъ > chlapóm > chlapuom > chlapům) as does the original o-stem GP {*-o-v-ъ > -ó-v > -uo-v > ů-v > ů} that has been generalized to nearly all masculine paradigms in CLC. We conclude from these observations that the lengthening of the stem vowel in the substantival paradigms coincided with the loss of the weak jers but in all probability began operating before the vocalization of the strong jers. Yet there is still the possibility that it was the neo-acute intonation from the neo-acute class that spread to these desinences and that the length developed here as a continuation of the neo-acute intonation.

(pánom – kravom – ženam – ženach – klubach – klukam). Přitom jsou tvary s krátkým –am na této oblasti zcela ve shodě s krátkými koncovkami –um a –im v 3. p. u ostatních typů skloňování (pánum, mužum, městum; nášim, kost’um).” [Regarding the age of these short endings, –am, – ach, in the Western Bohemian dialects, nothing certain can be said. But their geographical distribution agrees with the distribution of the –om endings [DP], which shows that the impetus towards these dialectal innovations could stem precisely from forms with –om, especially when compared to a similar situation such as with the paradigm of voda. A collision between the short type, kravom, and the original type ženám took place. This levelling then caused the same analogical change in corresponding forms in the same paradigm but also in analogical forms of other paradigms e.g. (pánom – kravom – ženam – ženach – klubach – klukam). Therefore the forms with short -am in these dialects are agree with the short desinences –um and –im in the DP of the other declensions (pánum, mužum, městum; nášim, kost’um)].

63 Table 5 – Oxytonic disyllabic feminine (j)ā-stems in CLC with original long root vocalism Case / Number

Proto-Slavic form adapted from Stang (1965:60)

Rule for Preservation or Loss of Length in original Long Root Vowel

Expected CLC Form

Old Czech Form

Literary Czech Form

NS

*trob à

Length preserved pretonically

TROUBA

trúba

trouba

GS

*trob ỳ

Length preserved pretonically

TROUBY

trúby

trouby

DS

*trob ě

Length preserved pretonically

TROUBĚ

trúbě

troubě

AS

*trob o

Length preserved pretonically

TROUBU

trúbu

troubu

IS

*trob ojo

Length preserved pretonically

TROUBOU

trúbou

troubou

LS

*trob ě

Length preserved pretonically

TROUBĚ

trúbě

troubě

NP

*trob y

Length preserved pretonically

TROUBY

trúby

trouby

GP

*trob ъ

1). Length preserved pretonically, 2). Neo-acute retraction, 3). Morpho-phonemic loss of length in GP

TRUB

trub

trub

DP

*trob am ъ

Length preserved pretonically

*TROUBAM

trúbám

troubám

AP

*trob y

Length preserved pretonically

TROUBY

trúby

trouby

IP

*trob am i

Length preserved pretonically

TROUBAMI

trúbami

troubami

LP

*trob ax ъ

Length preserved pretonically

*TROUBACH

trúbách

troubách

The CZ reflexes of the PSL oxytonic (j)ā-stem paradigm exhibit the expected phonological reflexes throughout the paradigm except for in the GP and the desinences of the DP & LP. Gebauer (III.I.:179) lists GP pát, kóp, attested in OCZ from the oxytonic a.p. (b). These forms may represent a remnant of the reflex of the neo-acute intonation in OCZ, but in the GP, any phonological process that may have once operated has been nearly obliterated by the generalization of brevity in the GP, particularly in the oxytonic

64

class. In OCZ there is a paucity of original oxytonic forms that alternate in the grammatical cases that we find the acute substantives alternating in. There are very few examples in OCZ of shortenings of original oxytonic root vowels in the IS, DP, LP, IP. The natural conclusion to reach is that words of the oxytonic class were not subject to the loss of pretonic length in three syllable forms of the paradigm (in other words, pretonic length was not treated the same way as acute length) and that the spread of brevity in the GP was a later morphological development that affected nearly all long (j)ā-stem lexemes. Spurious OCZ shortenings in the IS, DP, LP, and IP for this class of words must be attributed to sporadic mirroring of the acute a.p. (a) pattern which was quite well preserved in OCZ. Table 6 – Circumflex disyllabic feminine (j)ā-stems in CLC with original long root vocalism Case / Number

Proto-Slavic form according to Stang (1965:62)

Rule for Preservation or Loss of Length in original Long Root Vowel

Expected CLC Form

Old Czech Form

Literary Czech Form

NS

*golv à

Length preserved pretonically

*HLÁVA

hlava

hlava

GS

*golv ỳ

Length preserved pretonically

*HLÁVY

hlavy

hlavy

DS

*gôlv ě

Circumflex length lost

HLAVĚ

hlavě

hlavě

AS

*gôlv o

Circumflex length lost

HLAVU

hlavu

hlavu

IS

*golv ojo

Length lost in word greater than two syllables long

HLAVOU

hlavou

hlavou

LS

*golv ě

Length preserved pretonically

*HLÁVĚ

hlavě

hlavě

NP

*gôlv y

Circumflex length lost

HLAVY

hlavy

hlavy

GP

*golv ъ

Length preserved pretonically

HLÁV

hlav / hláv

hlav

DP

*golv am ъ

Length preserved pretonically

*HLÁVAM

hlavám

hlavám

AP

*gôlv y

Circumflex length lost

HLAVY

hlavy

hlavy

IP

*golv am i

Length preserved pretonically

*HLÁVAMI

hlavami

hlavami

LP

*golv ax ъ

Length preserved pretonically

*HLÁVACH

hlavách

hlavách

65

CLC is quite consistent in generalizing brevity in this class of substantives. Nearly all (j)ā-stems that we have classified as a.p. (c) (see Table 15 thru Table 18) are short throughout the paradigm and there are few dialectal exceptions. Map 13, however, which shows the dialectal distribution of length in CZ zima ‘cold, winter’ illustrates the fact that attested dialectal length in the NS from an original a.p. (c) root is dramatically different (heavily concentrated in the Northeastern Bohemian dialects) from dialectal length in the NS of ā-stems that belonged to other accentual paradigms [e.g. CLC duha / CZ dial. NS douha from a.p. (b) shown in Map 11, CLC NS bříza / CZ dial. březa from a.p. (a) shown in Map 5]. If it weren’t for the many attestations of zíma in OCZ (e.g. SlovKlem 34a Bruma zíma, SlovKlem 48a Frigus zíma which shows that the word had two meanings – and in some dialects these two meanings are distinguished by an opposition in length 680 [v zimňe] ‘in the cold,’ [v zẹ@mje] ‘in the winter’), one might be inclined to explain this form as a local CZ dialectal innovation. It is possible that the dialectal forms with length are actually preserved pretonic length in the NS, or alternatively, that PSL *zima showed early variation of its accentual paradigm. If the stress was on the stem vowel /-a-/ in the DP, IP, and LP of the mobile paradigm, we would expect to see some residue of this length preserved in OCZ and the dialects , but there does not seem to be any. From the Czech data at least, it appears that the ictus in the DP, IP, and LP would have been on the final vowel of the desinence and not the stem vowel. We would also expect there to be some residue in OCZ and the dialects of the preservation of length in the NS, GS, and LS in the circumflex (j)ā-stems, but there are not many examples in the data we have looked at. The only other

66

alternative is to imagine that circumflex length in Proto-Czech was lost early enough for the brevity to spread throughout the paradigm creating a class without length considerably before the appearance of manuscripts and texts with markings for length. In the GP, Gebauer (III.I.:179) lists strán, hláv, nóh, vód, hór, rúk, dúš with length from the mobile paradigms. Map 14 shows the dialectal distribution of lengthened /ó/ in the GP of /*noha/, a.p. (c) ‘leg.’ The map shows that only a small area in the northwestern Czech territory preserves the length. Length in the GP of (j)ā-stems is not a morphological generalization in these dialects, and if it weren’t for the fact that there are attested forms in OCZ with length in the GP of originally mobile stems, and that the phonologically correct form is one that we expect to have length, it would be difficult to rest a case on these examples as preserving old length in the GP. Nevertheless, we must regard them as old. Whether the explanation for this length is phonological (e.g. preserved pretonic length) or morphological (e.g. archaic tendency to generalize length in the GP when the rest of the paradigm is short in order to differentiate the GP from the rest of the paradigm) is not possible to determine from our data. Table 7 – Neo-acute disyllabic feminine jā-stems in CLC with original long root vocalism Proto-Slavic form

Rule for Preservation or Loss of Length in original Long Root Vowel

NS

*gor̃dj a

1). Pretonic length preserved, 2). Neo-acute retraction of ictus, Desinence 3). Přehláska /a > ä > e/

HRÁZE

hráze

hráze

GS

*gor̃dj æ

1). Pretonic length preserved, 2). Neo-acute retraction of ictus, Desinence 3). Přehláska /a > ä > e/

HRÁZE

hráze

hráze

Case / Number

Expected CLC Form

Old Czech Form

Literary Czech Form

67 Proto-Slavic form

Rule for Preservation or Loss of Length in original Long Root Vowel

DS

*gor̃dj i

1). Pretonic length preserved, 2). Neo-acute retraction of ictus

HRÁZI

hrázi

hrázi

AS

*gor̃dj o

1). Pretonic length preserved, 2). Neo-acute retraction of ictus, Desinence 3). Přehláska /u > ü > i/

HRÁZI

hrázi

hrázi

IS

*gor̃dj ejo

1). Pretonic length preserved, 2). Neo-acute retraction of ictus, Desinence 3). Contraction /ejo > ú/,

HRÁZÍ

hrází

hrází

Case / Number

Expected CLC Form

Old Czech Form

Literary Czech Form

4). Přehláska /ú > ü@ > í/ LS

*gor̃dj i

1). Pretonic length preserved, 2). Neo-acute retraction of ictus

HRÁZI

hrázi

hrázi

NP

*gor̃dj æ

1). Pretonic length preserved, 2). Neo-acute retraction of ictus, Desinence 3). Přehláska /a > ä > e/

HRÁZE

hráze

hráze

GP

*gor̃dj ь

1). Pretonic length preserved, 2). Neo-acute retraction of ictus

*HRÁZ

hrází

hrází

DP

*gor̃dj am ъ

1). Pretonic length preserved, 2). Neo-acute retraction of ictus, Desinence 3). Přehláska /a > ä > e/

*HRÁZEM

hrázím

hrázím

AP

*gor̃dj æ

1). Pretonic length preserved, 2). Neo-acute retraction of ictus, Desinence 3). Přehláska /a > ä > e/

HRÁZE

hráze

hráze

IP

*gor̃dj am i

1). Pretonic length preserved, 2). Neo-acute retraction of ictus, Desinence 3). Přehláska /a > ä > e/

HRÁZEMI

hrázemi

hrázemi

LP

*gor̃dj ax ъ

1). Pretonic length preserved, 2). Neo-acute retraction of ictus, Desinence 3). Přehláska /a > ä > e/

*HRÁZECH

hrázích

hrázích

The neo-acute on an original long vowel nearly always shows up in Czech and Slovak as length. Neo-acute on an original long vowel in the masculine stems also

68

preserves length throughout the paradigm (e.g. král, krále, královi, krále, královi, králem, králové, králů, králům, krále, králi, králích). The neo-acute on an original short vowel, however, has a different effect on the feminine stems than in the masculine stems. In the feminine jā-stems with neo-acute (see Table 8) the root vowel has length throughout the paradigm, but in the masculine oxytonic (j)o-stems with lengthened /o/ in the NAS (see Table 29), the length does not appear in any other case forms. Table 8 – Neo-acute disyllabic feminine jā-stems in CLC with original short root vocalism ProtoSlavic form

Rule for Preservation or Loss of Length in original Long Root Vowel

Expected CLC Form

NS

*võlj a

1). Neo-acute retraction, 2). Lengthening of /o/ under neoacute, Desinence 3). Přehláska /a > ä > e/

VŮLE

vóle

vůle

GS

*võlj æ

1). Neo-acute retraction, 2). Lengthening of /o/ under neoacute, Desinence 3). Přehláska /a > ä > e/

VŮLE

vóle

vůle

DS

*võlji

1). Neo-acute retraction, 2). Lengthening of /o/ under neo-acute

VŮLI

vóli

vůli

AS

*võlj o

1). Neo-acute retraction, 2). Lengthening of /o/ under neoacute, Desinence 3). Loss of nasalization 4). Přehláska /u > ü > i/

VŮLI

vóli

vůli

IS

*võlj ejo

1). Neo-acute retraction, 2). Lengthening of /o/ under neoacute, Desinence 3). Contraction /ejo > ú/,

VŮLÍ

vólí

vůlí

VŮLI

vóli

vůli

Case / Number

Old Czech Form

Literary Czech Form

4). Přehláska /ú > ü@ > í/ LS

*võlj i

1). Neo-acute retraction, 2). Lengthening of /o/ under neo-acute

69 ProtoSlavic form

Rule for Preservation or Loss of Length in original Long Root Vowel

Expected CLC Form

NP

*võlj æ

1). Neo-acute retraction, 2). Lengthening of /o/ under neoacute, Desinence 3). Přehláska /a > ä > e/

VŮLE

vóle

vůle

GP

*võlj ь

1). Neo-acute retraction, 2). Lengthening of /o/ under neo-acute

*VŮL

vól

vůlí

DP

*võlj am ъ

1). Neo-acute retraction, 2). Lengthening of /o/ under neoacute, Desinence 3). Přehláska /a > ä > e/

*VŮLEM

vólím

vůlím

AP

*võlj æ

1). Neo-acute retraction, 2). Lengthening of /o/ under neoacute, Desinence 3). Přehláska /a > ä > e/

VŮLE

vóle

vůle

IP

*võlj am i

1). Neo-acute retraction, 2). Lengthening of /o/ under neoacute, Desinence 3). Přehláska /a > ä > e/

VŮLEMI

vólemi

vůlemi

LP

*võlj ax ъ

1). Neo-acute retraction, 2). Lengthening of /o/ under neoacute, Desinence 3). Přehláska /a > ä > e/

*VŮLECH

vólích

vůlích

Case / Number

Old Czech Form

Literary Czech Form

The few examples of shortening in the IS in jā-stems, e.g. volí, koží, s-těží, Gebauer (1960, III.I:215-126) seem to come from the neo-acutes (not from the acute or oxytonic jā-stem paradigms) and indicate that there was some spread of brevity in the IS from original acutes into the neo-acute paradigms. There are, however, attested OCZ shortenings in the plural as well for several original neo-acute lexemes. All in all, there is not sufficient data at this time to suggest that a separate paradigm with shortened vowel only in the IS and not in the plural forms ever existed.

70

Feminine Disyllabic j(ā)-stem paradigms in Contempory Literary Czech Following Verweij (1993:494), it is useful to classify paradigms in the Czech substantival system according to patterns in vowel quantity alternations in the root vowels of originally disyllabic words. We see no need to improve dramatically on the system developed by Verweij and therefore reproduce this classification for convenience in reference. The only addition we have made is to offer a table for each gender rather than combining genders in the same table as Verweij has done. It should be noted that these paradigmatic classifications apply to Contemporary Literary Czech (CLC). Czech dialects may or may not exhibit the same patterns as the literary language. A. Short root vowel in all forms; B. Long root vowel in all case forms; C. A long root vowel in the NAS, short root vowel in the remaining case forms; D. Short root vowel in the GP, long root vowel in the remaining case forms; E. Short root vowel in the GP, DP, IP, LP, long root vowel in the remaining case forms; F. Short root vowel in the IS, GP, DP, IP, LP, long root vowel in the remaining case forms. G. Slovak pattern with length in the GP; X. Pattern ambiguous. Dialectal material (in the form of maps attached in the appendices, attested forms from a variety of published dialectal studies, dictionaries, and forms collected from the card catalogue at the Institute for the Czech Language, Czech Academia of Sciences) whose data differ considerably from the facts found in CLC will be introduced in the

71

following discussions to support the membership of words in the accentological classes recognized for PSL. Table 9 – Feminine disyllabic (j)ā-stem paradigms in CLC

Case / Number

A (all short)

muka muky muce muku mukou muce muky muk mukám muky mukami mukách

NS GS DS AS IS LS NP GP DP AP IP LP

B (all long)

bříza břízy bříze břízu břízou bříze břízy bříz břízám břízy břízami břízách

C (long NAS, else short)

N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A

3

D (short GP, else long)

chvála chvály chvále chválu chválou chvále chvály chval chválám chvály chválami chválách

E (short GP, DP, IP, LP, else long) kráva krávy krávě krávu krávou krávě krávy krav kravám krávy kravami kravách

F (short IS, GP, DP, IP, LP, else long) rána rány ráně ránu ranou ráně rány ran ranám rány ranami ranách

Italicized forms in Table 9 show vowel quantity alternations within a given paradigm. As a shorthand way of indicating the forms of all the cases for a particular CLC word, the scheme shown above will be referred to by noting the letter of the declension type for the Czech words in the column labeled ‘Czech’ in the tables that follow (see Table 10 thru Table 19) which list the substantives according to Proto-Slavic accentual class. These classification tables represent Proto-Slavic (j)ā-stem substantives, compiled from the works of various authors (Bulaxovskij, Dybo, Gebauer, Hujer, Kopečný, Shevelov, Vasmer, Trávníček, and others), that exhibit the patterns of

3

Although the jā-stem (soft paradigm = měkký vzor) desinences in CLC have been dramatically altered by the so-called Czech umlaut (přehláska), the alternations of quantity in the root vowel of disyllabic words occurs in the same grammatical cases, so that the classification scheme may still be applied.

72

correspondences in the contemporary Slavic languages that are generally associated with acute, oxytonic, circumflex and neo-acute accentual paradigms. From Table 9, Type A with short vowel throughout the paradigm applies to original long vowel stems as well as to original short vowel stems. Many of the PSL mobile type (j)ā-stems of a.p. (c) in CLC have a short vowel throughout the paradigm. As far as the acute PSL (j)ā-stems of a.p. (a) and oxytonic PSL (j)ā-stems of a.p. (b) are concerned, a number of them in CLC also belong to this paradigm even though we have reason to expect acute long vowels and pretonic length to be preserved in CLC. Ā-stems with original PSL long vowels that belong to the A type paradigm in CLC with shortened long vowel include e.g. : baba, brada, cena, cesta, chmura, děva, duha, duma, hlava, hnida, hřada, huba, hvězda, jikra, jiskra, jizva, klika, kuna, kupa, lava, lichva, měna, mluva, muka, něha, niva, pata, pěna, piha, pila, plena, pleva, ruda, ruka, ryba, řada, řasa, řeka, slina, sluha, sluka, snaha, stěna, straka, strana, středa, střecha, střela, štika, uzda, věda, vina, vydra, zima, žluna, žluva. Ā-stems that belong to the A type paradigm with brevity throughout the declension in CLC with original PSL short vowels include e.g.: deska (< PSL *dъska), hora, jehla (< PSL *jьgъla), kopa, kosa, koza, mora, noha, nora, roba, rosa, rota, sestra, socha, sosna, stopa, voda, vosa,4 žena.

4

But see Map 10 for the dialectal distribution of lengthened /ó/ in this word.

73

Jā-stems in CLC with originally long PSL vowels that belong to this type with brevity throughout the declension include e.g.: duše, kaše, kleště, paže, saze, sukně, věž, višně, zmije. Jā-stems in CLC with originally short PSL vowels that belong to this type include e.g.: mez (< OCZ mezě), nozdra, rez (< PSL *rъzjā), země, zora (< PSL *zorjā). Type B with long vowel throughout the paradigm applies to original long vowel stems as well as original short vowel stems. Ā-stems that belong to the B type paradigm with long vowels throughout the declension in CLC with original long vowel stems include e.g.: bříza, hlína, hlíva, hlíza, hříva, jícha, jíva, kláda, kýla, kýta, láska, líska, máma, máta, prouha, roura, routa, síra, skýva, sláva, stoupa, sváda, škvára, štˇáva, vláha, brázda, céva, dýka, hlísta, jícha, jízda, krása, křída, lícha, líska, míza, mouka, pýcha, slouha (m.), střída, třída, tříska, vláda, žláza. Jā-stems in CLC that belong to this type include e.g.: báň(e), boule, bouře, číše, díž(e), dýně, hloub(-i), houně, houštˇ, hráz(e), chůze, chýše, káně, kdoule (< kъduňa), koupě, louč, mříž(e), nouze, nůše, píce, pouštˇ, příze, skráň, skříň, souše, stáj, stráň, stráž, šíje, šíř(e), tíž(e), tůň/tůně, vůle, vůně, výše, zář(e). Once again, however, the fact that these words have a long vowel throughout the paradigm in CLC is not sufficient to prove that they were acute stems in Proto-Slavic. A number of the jā-stems with long root vowel are in fact neo-acute stems (see Table 19). Type C with long vowel throughout the paradigm except the NAS does not occur with (j)ā-stems. This pattern is reflected by the (j)o-stems of the type NS dům, GS domu, DS domu, AS dům, VS dome!, IS domem, LS domě.

74

Type D with long vowel throughout the paradigm except in the GP characterize bába, bída, bouda, houba, hrouda, jáma, moucha, sláma, víra, skála, vrána, žába, díra, chvála, kroupa, touha, trouba, louka, strouha, mísa, slíva, tráva, brány. Jā-stems in CLC that belong to this type include: hrábě, chvíle, svíce. Type E with long vowel throughout the paradigm except in the GP, DP, IP, and LP include several ā-stems: e.g. kráva, skála, vrána, and zero jā-stems. All of these words are original acute stems (see Table 10). We regard this pattern as the second most archaic pattern for original acute substantives. Type F with long vowel throughout the paradigm except in the IS, GP, DP, IP, and LP is the most archaic. All of these words are original acute stems (see Table 10). The difference between the two paradigms, E and F, is reflected in the preservation of the reflex of the morphophonemic rule that shortened the root vowel in the trisyllabic IS form. Substantives of Type E have leveled out by analogy the alternation between the IS and the rest of the forms of the singular paradigm. There are several (j)ā-stems that preserve the alternation in the IS in CLC rána, síla, and several words that allow for type F as an option either in CLC or OCZ hrouda, kráva, OCZ dráha, lípa, míra, sláma, strouha, žíla, bríza, brána ‘gate,’ blána. Acute Feminine (j)ā-stems It should be mentioned outright that the lists of words in the tables are not intended to be exhaustive, though effort has been made to include as many relevant lexemes as possible. The idea throughout this project has been to focus on words with solid PSL pedigrees, although we have included words borrowed into PSL as well if they

75

appear to illuminate our knowledge of the PSL accentological system. Lexemes that were originally trisyllabic (e.g. CZ jehla < *jьgъla)5 have been generally excluded from the scope of this analysis. Slovak forms are quoted as a point of reference for the Czech data and indicate where the length in the Czech form is innovative or preserved in contrast to the Slovak, and where Czech and Slovak share the length or lack of length. Because Slovak consistently preserves pretonic and neo-acute length, it is nearly diagnostic for these PSL accentual paradigms. Slovak also exhibits a great deal of regularity in the shortening of acute length in the (j)ā-stems as well as in other classes of stems. Although Slovene is not particularly useful in helping to establish the PSL intonational paradigm for the feminine (j)ā-stems because of the fact that SLN lengthened all stressed vowels, some clues are provided by the quality of originally short /e/ and /o/ type vowels. The closed variants [/ọ@, ộ, ẹ@, ệ/ (in the orthography used by Pleteršnik, 1894)], whether from an originally long vowel, diphthong, or short vowel in SLN indicate that the ictus was originally on that vowel, whereas SLN long open /ó, é/ indicate that the ictus was retracted onto that syllable and later lengthened when all stressed vowels in SLN were lengthened. For place of stress, Russian proves to be most reliable, although for the identification of the oxytonic class versus the mobile class, we have discovered that what Stang alluded to (Stang, 1965:67) regarding a different type of mobile paradigm without retracted stress in the AS but with retracted stress in the NP (e.g. R sloboda, slobodu, slobody) may indeed be supported by the Czech data. 5

For trisyllabic words that have lengthened short vowels to the right of weak jers that were lost, however, such as CZ lhůta/lhota < PSL *lьgota see footnote 37.

76 Table 10 – Acute disyllabic feminine ā-stems (all with original long root vocalism) PSL

S/C

GLOSS

SLOVENE

CZECH

SLOVAK

RUSSIAN

*rana

‘wound’

rana

rána

rána F

rana

rana

*sila6

‘power’

sıla

síla

síla F

sila

sila

*bolna

‘membrane’

Ø

blána

blána D/F

blana

bolona AS bolonu / bolonu U bolona bolonь7

‘hunk’

gruda

grúda

hrouda, GP hrud D/F8 hruda (ESSJ)

hruda

gruda

*krava

‘cow’

krava

kráva

kráva E/F

krava

korova

*dorga

‘road’

draga

drága

draha OCZ dráha D/F

draha

doroga

*lipa9

‘linden’

lıpa

lípa

lípa D/F

lipa

lipa

*měra

‘measure’

mjera

mẹ@ra

míra, D / OCZ F10

mera miera

m’era

*groda/ *gruda

6

Dybo (2000:50) a.p. (a).

7

Although not attested in S/C and despite the fact that Russian is indeterminate, we believe *bolna is a clear example of an a.p. (a) ā-stem based on the CZ, SLK evidence. R bolona ‘tree epiphyte’ is probably the secondary form derived from a posited earlier Proto-Russian *bolona a.p. (a) ‘tree membrane’ by way of an opposition in accentual paradigms. Vas. (1986, I:189) states that the R stress in bolona is different from the Lith. bálnas ‘white,’ which indicates original acute (ā-stem). 8

In OCZ in several texts from the 15th century where vowel quantity is marked there is evidence that other cases in the paradigm had the shortened root form: “By kto tobě zlatú hrudu (IS) dal u bok” Kruml. 244b “hrudu ohennú ignitam strumam” Otc. 352b. Also in the PSSČ the word is given as alternatively F with /u/ in the IS, GP, DP, LP, and IP. 9

Dybo (2000:50) a.p. (a).

10

(Gebauer, III. I.:182).

77 PSL

S/C

GLOSS

SLOVENE

CZECH

SLOVAK

RUSSIAN

*solma

‘straw’

slama

sláma

sláma D/F

slama

soloma

*struga

‘ditch, canal’

struga

strúga

strouha D / OCZ F11

struha

dial. struga

*žila12

‘vein’

žıla

žíla

žíla D/F

žila

žila

*skala

‘cliff’

skala

skála

skála D/E

skala

skala13

*vorna14

‘crow’

vrana

vrána

vrána D/E

vrana

vorona

*baba15

‘old lady’

baba

bába

bába D

baba

baba

*jama

‘ditch’

jama

jáma

jáma D

jama

jama

*mucha

‘fly’

mùha mùhy Vas. muha

múha

moucha, GP much D16

mucha

muxa

*berza

‘birch tree, betula’

breza

brẹ@za

bříza B/F

breza

ber’oza

*glina

‘clay’

glína Vas. gnıla

glína

hlína B

hlina

gl’ina

*griva

‘mane’

grıva

gríva

hříva B

hriva

gr’iva

*iva

‘yew, salix caprea’

ıva

íva

jíva B

iva

iva

dial. struga

11

(Ibid.:182). Gebauer also comments on the coexistence in CLC of a new formation with a stem struh-, formed from the GP, that is short throughout the paradigm. 12

Dybo (2000:50) a.p. (a).

13

The R form indicates end stress. OR plural forms also indicate end stress (ESRJ:IV, 201), however a fairly old loan word exists in Russian skala ‘musical scale’ with fixed root stress that could have very easily influenced skala ‘cliff’ to adopt an end-stress paradigm. 14

Dybo (2000:50) a.p. (a).

15

Ibid.

16

In OCZ in at least one text from the 15th century where vowel quantity is marked there is evidence that other cases in the paradigm had the shortened root form: “Belzebub se vykládá jako buoh much neb nad muchamy” [Beelzebub is described as the lord of the flies or over the flies] RokycPostBrň. 182 (StČS I.: 416).

78 PSL

S/C

GLOSS

*kolda 17

*kyla

SLOVENE

CZECH

SLOVAK

RUSSIAN

‘log’

klada

kláda

kláda B

klada

koloda

‘hernia’

kıla

kíla

kýla B

kyla

kila18 NP kily

*laska

‘tenderness’

laska

láska

láska B

láska < Czech?

laska

*slava

‘glory’

slava

sláva

sláva B

sláva < CZ

slava

*sliva

‘plum’

slıva

slíva

slíva B/D

sliva

sl’iva

*děva

‘girl’

djeva

dẹ@va

děva A

deva

d’eva AS d’evu

AS djevu 19

*gnida

‘nit’

gnjıda

gnída

hnida A

hnida

gn’ida

*jьskra

‘spark’

ıskra

ískra

jiskra A

iskra

iskra

jazva

jazva

*ězva

‘sore, cut’

jazva

jâzba Vas.

20

jizva A

dial. jazva *klika

‘latch’

kljuka

kljúka

klika A

klika

AS kljuku

kl’uka AS kl’uku Ukr. kl’uka

*kupa 21

‘60, mound’

kupa A

AS kupu

kùp GS kúpa

kupa

kupa

kupa AS kupu

*lava

‘bench’

Ø

láva

lava A

lava (ESSL)

lava

*měna

‘change’

mijéna AS mijénu

mẹ@na

měna A

mena

m’ena

muka

múka

*moka

‘torture’

AS m’enu muka A

muka

muka AS muku

17

Dybo (2000:50) a.p. (a).

18

Ukr kila. Also the Dict. of 1704 has kila throughout according to the 17-volume Akedemačeskij slovar’ russkogo jazyka. 19

Dybo (2000:50) a.p. (a).

20

OCZ jiezva indicates length, and jizva itself indicates original PSL length ěî > ié > í > i.

21

Dybo (2000:50) a.p. (a).

79 PSL

S/C

GLOSS

*něga

‘tenderness’

njega

SLOVENE

Ø

CZECH

něha A

SLOVAK

neha

‘field’

njıva

n’ega AS n’egu

AS njegu *niva

RUSSIAN

njíva

AS njıvu

niva A Vas. níva

niva

n’iva AS nivu

*pěna

‘foam’

pjena

pẹ@na

pěna A

pena

p’ena

*pěga

‘freckle’

pjega

pẹ@ga

piha A22 OCZ pieha Vas. píha

peha

p’egij

plẹ@va

pleva A Vas. pléva Chod. plíva

pleva

polova

AS pjegu *pleva

‘chaff’

pljeva AS pljevu

AS polovu p’el’ova Vas. p’el’ovu Vas. plěva, pьlěva Srez.

*rěpa

‘beet’

repa

rẹ@pa

řepa A

repa

r’epa

*ryba

‘fish’

rıba

ríba

ryba A Vor. rejba

ryba

ryba

*sorka /svorka23

‘magpie’

svraka

sráka

straka A

straka

soroka

*tlapa24

‘paw’

dial. lapa

lápa

tlapa A25

tlapa

lapa

*vydra26

‘otter’

vıdra

vídra

vydra A

vydra

vydra

*vьlna

‘wool’

vuna

vólna

vlna G

vlna

volna

22

OCZ pieha indicates length, and piha itself indicates original PSL length ěî > ié > í > i.

23

Ibid. (op. cit.)

24

Ibid. (op. cit.)

25

Preservation of /tl/ cluster in word initial position for CZ & SK is a rarity.

26

Dybo (2000:50) a.p. (a).

80 Table 11 – Acute disyllabic feminine jā-stems (all with original long root vocalism) PSL

S/C

GLOSS 27

SLOVENE

CZECH

SLOVAK

RUSSIAN

*banja

‘spa’

banja

bánja

báň(e) B

báň, baňa

banja

*kēsja

‘goblet’

čafiša

čáša

číše B

čaša

čaša

*dynja

‘cucumis melo, pumpkin’

dınja

dínja

dýně B

dyňa

dynja

*guna/ *gunja

‘gown’

gunja

OSLN gúnj

houně B

huňa

dial. gunja

*kapja

‘drop’

kapl’a

káplja

kápě B

kaplina

kaplja

AS dynju

GS dınji

AS kaplju *mergja

‘fishing net’

mreža

mrẹ@ža

mříže B

mreža

mer’eža

*pitja

‘food’

pıħa Vas.

píča

píce B

Ø

pišča

pıća Rejzek *xysja/ *xyzja

‘shack’

hiša/ OSC hıža

híša híža

chýše/chyše chýže B/A

SK dial. chyža28

RChSl xyža dial. xyža

*kasja

‘porridge’

kaša

káša

kaše A

kaša

kaša

29

*sadja

‘soot’

dial. sađa

sája

saze A dial. sáze

sadza

saža

*věgja

‘tower’

véža dial.

véža30

věž(e) A

veža

Vas.OR věža

*višnja

‘cherry (tree)’

vıšnja

vîšnja

višně A

višňa

višňa

Proto-Slavic acute intonation is traditionally associated with length in Czech on initial syllables in disyllabic words, which includes many forms of the feminine (j)āstems of the a.p. (a) class.31 If we accept the Proto-Slavic paradigm for (j)ā-stem acute

27

For an etymological explanation of West Slavic meaning ‘cupola’ see Rejzek (2001:69).

28

ESSJ, v. 8, p. 165-166.

29

Lith. súdžiai indicates acute intonation. Map 24 shows the dialectal distribution of length versus brevity in this word. 30

South Slavic points to oxytonic, SLK, R point to acute.

31

The IS, DP, IP, LP are originally trisyllabic forms.

81

words as proposed by Stang (1965:56) and as illustrated in Table 4 with acute intonation on the root vowel of all forms in the paradigm except the GP,32 we expect the ProtoCzech retention of length on the root vowel as well until the phonological rule that shortened (stressed) acute length in forms greater than two syllables long began to operate. The observation that CLC on a semi-regular basis shows length on the vowel of the acute stem has led linguists to think of this length as preserved Proto-Slavic length despite the fact that Czech is the only Slavic language to presumably preserve acute length.33 In Serbo-Croatian and Slovene, stressed acute length was lost.34 In Slovak as well, stressed acute length was lost. Yet in more than half of the CZ (j)ā-stems identified as acute, the reflex of the vowel in the CLC stem is short throughout the paradigm (marked with A in Table 10 and Table 11). The source of this brevity could be due to the relatively high number of forms in the acute disyllabic (j)ā-stem paradigms that are three syllables long where brevity in the root is to be expected phonologically since there is other internal evidence that (stressed) acute length in trisyllabic forms was shortened in Czech (in addition to all other Slavic languages with the long:short opposition except

32

In Čakavian we find evidence of the neo-circumflex in the GP. The Slovenian pl. forms show neo-circumflex in the root vowel before a long vowel in the final syllable. The endings with long vowels were GP {*-ōvь}, NP {*-ī}, LP {*- īxъ}. 33

Although hesitant to declare that acute length is preserved in USo, Schaarschmidt (1997:49) does give evidence in USo for the preservation of acute and neo-acute length in addition to evidence for the lengthening of PSL short vowels /e, o, ъ, ь/ under the neo-acute. For RP (Rising Pitch = acute) USo brěza ‘birch’ indicates the preservation of length, as do USo dróha ‘road,’ CZ draha/dráhy, USo blóta, CZ bláto ‘swamp.’ For NRP (New Rising Pitch = neo-acute) he lists USo jěža ‘hedgehog,’ USo GS hrěcha ‘sin,’ CZ hřích, USo dial. stróža, CZ stráž(e) ‘guard.’ 34

Length was later restored in Slovene, however, on all non-final stressed vowels.

82

SLN). Examples cited by Shevelov (1965:512-517) include: 1) initial stress: S/C ulica, CZ ulice, SK ulica, R ul’ica ‘street;’ S/C lagoda, CZ lahoda, SK lahoda, U lahoda ‘delicacy;’ CZ jahoda ‘strawberry,’ R jagoda ‘berry,’ etc., 2) internal stress: CZ malina, R mal’ina ‘raspberry;’ CZ beseda, R b’es’eda ‘conversation;’ CZ lopata, SK lopata, S/C lòpata, R lopata ‘spade,’ etc. Pretonic length, on the other hand, in initial syllables (CZ útroba ‘womb,’ R utroba ‘maw’) and internal syllables (CZ kolbása, R kolbasa ‘sausage’) in trisyllabic words is generally preserved in Czech. With these conditions in mind, we can then say that the paradigm in CLC closest to what we expect for feminine acute (j)ā-stems is represented by Verweij’s F-type words (e.g. rána) – the class with shortened length in the trisyllabic forms (IS, DP, IP, LP).35 There are, however, only two ā-stem a.p. (a) roots that exhibit exclusively this pattern in CLC – rána, and síla. There are no (j)ā-stem roots that exhibit this pattern in CLC. The fact that in OCZ a number of CLC D-type a.p. (a) ā-stems belonged to the F-type paradigm (dráha, hrouda, kráva, míra, strouha) indicates that the phonologically expected a.p. (a) paradigm was more widespread than it is today. Of special interest are the acute ā-stems with root diphthong /-ou- < ú/, hrouda, strouha, moucha. These words show that the shortening of the vowel /ú/, which would have been phonological in the IS, GP, DP, LP, and IP is relatively unstable. All of these words in OCZ indicate the pattern of shortening in the cases we would expect from a.p. (a) stems, and all of them have

35

The shortening of acute length in the GP is be discussed separately (see p. 92).

83

retained the /ou ~ u/ alternation in CLC only in the GP. Other words show variation in CLC and the dialects (lípa, sláma, žíla, skála, vrána, bříza, jiskra, etc.) and they help to explain the development of the alternations in the (j)ā-stem paradigms. The Source and Distribution of Czech Shortened Acute Intonation In accordance with well established principles of historical linguistic change, we expect the NS to be the form most likely to serve as the source of analogical leveling in Slavic and Czech. By looking at original a.p. (a) (j)ā-stems that belong to Type A in CLC (brevity throughout the paradigm) and Type B (length throughout the paradigm) Type D (brevity in the GP only), Type E (brevity in the GP, DP, LP, and IP) it appears that the pressures of analogical change have acted on the CLC paradigms in favor of generalized length or brevity. It is also clear that the general tendency throughout the historical period has increased the number of shortened roots. The question is whether the length in these paradigms represent preserved Proto-Slavic length, or whether length was reintroduced by some other phonological or morphological process. First we will present the Czech dialectal data, next we will present the explanation offered by the Dutch linguists Kortlandt (1978) and Verweij (1993), then we will present our own interpretation of the data. Czech Dialectal Data There is no distinguishable relationship between the phonological or phonotactic structure of the root and the generalization of either length or brevity in Czech. The fact that many of the CLC disyllabic feminine a.p. (a) (j)ā-stems have a short vowel throughout the paradigm in CLC does not necessarily mean that they were not a.p. (a)

84

stems in Proto-Slavic. For example, NS jiskra ‘spark’ (< *jьskra) is an original a.p. (a) stem (S/C ıskra, SLN ískra, SK iskra, R iskra). Verweij (1991:503) correctly identifies the word as a.p. (a) and the comparative evidence also points to a.p. (a). Map 1 shows the dialects with attested length in the NS of /*jьskra/. NS jískra is concentrated along the periphery of the Czech lands – namely, in the Chod dialects (AV Survey locales 301311) the Southwestern Bohemian dialects in general, as well as the Northeastern Bohemian dialects (101-161) with some attestations in other Southern Bohemian dialects. PSL /*griva/ CLC hříva also undoubtedly belonged to a.p. (a). Map 4 shows that brevity in the NS of /*griva/ is restricted primarily to the Czech-Moravian, Central Moravian, Eastern Moravian, and Silesian dialects. In the Silesian dialects, there is no opposition of short and long vowels.36 The brevity in the Moravsko-Slovensko dialects (near the Czech-Slovak border in southeastern Moravia) and in the north central Moravian dialects illustrates the diffusion of the PSL shortening of original acute length that we will present a full argument for below. In CLC hříva was not one of the lexemes affected by the diffusion of shortened acute lengths, but other lexemes were. Map 5 shows the dialectal distribution for brevity in the root vowel of CLC NS bříza ‘birch.’ The distribution of CZ dialectal brevity (most often as březa) is nearly identical to the CZ dialectal pattern of brevity in hříva. We can safely say that the

36

Bělič (1971:84) identifies the shortness of originally long /í/ (together with the shortness of originally long /ú/) in the southwest and northern parts of the Central Moravian dialects as a relatively recent phenomenon which he attributes in part to the short phonetic duration of these vowels and the fact that the long and short variants are closer together in duration than the long and short variants of other long:short vocalic pairs. But the brevity attested for this word in other dialects is not a result of this tendency.

85

process responsible for the short variants in the dialects for these two words must come from the same source. Map 7 shows the dialectal distribution for length (řípa) in the root vowel of CLC řepa ‘birch.’ Here we see pretty much the same pattern emerging, but in this example CLC has been affected and the root vowel was shortened. CLC jáma (ČJA II.414-415) with brevity throughout Moravia, and Silesia (length however in the larger towns such as Brno 63, Moravský Krůmlov 62, Prostějov 66, Boskovice 64, Nové město na Moravě 53 no doubt influenced by the literary language) shows the same pattern, as does CLC draha pl. tant. ‘fallow field’ / dráhy ‘ruts, railroad tracks’ (ČJA II.322, map on 325). This word, however, has been influenced by a separate tendency to differentiate lexemes that can be construed as collective substanives or pluralia tantum. The neuter NP desinence {–a} accompanied by short root vowel is most likely responsible for the form pl. tant. draha. The original fem. pl. dráhy is common in the Southwestern Bohemian dialects with length. Bulaxovskij (1983:161) mentions other examples of original acute (j)ā-stems where brevity in Moravia, Silesia, and Slovakia is the rule: klada, skala, rana, vrana, žila, lipa, hlina, sila, hruda, mřeža, etc. We therefore believe that the Southern Bohemian Dialects and particularly the Chod dialects preserve a more archaic form of the acute (j)ā-stems with length in the root vowel and that the shortening of the root vowel in acute (j)ā-stems in the Central Bohemian dialects is an innovation – but not one that spread by the power of the literary language.

86

The Czech Lengthening? Kortlandt (1978) and Verweij (1991) have come up with an alternative solution to lengthened vowels in CZ that involves the lengthening of vowels that were originally rising (e.g. originally acute or with fixed internal stress on short vowels). The rule is formulated in the following terms: “A short rising vowel in an open first syllable of disyllabic words is lengthened unless the second syllable contains a long vowel” (Verweij, 1990:505). This rule is formulated to explain the quantity alternations in the CLC paradigm marked as Type F (e.g. rána, síla) and other lengths in CLC that are difficult to explain (e.g. lhůta < lъhota ‘deadline, term < freedom from obligation’),37 and to argue for a Proto-Slavic shortening of acute length that left a rising tone on the acute 37

Part of the merit of Kortlandt and Verweij’s analysis rests on its ability to explain the anamolous length in this type of CZ word with the structure Cъ/ъ(C)Co/eCV, itself a rarity in Czech. CZ vzhůru < vъzhoru, shůry < sъhory will serve as two other examples. In order to make their rule of Czech lengthening work with all its conditions, it is necessary to assume fixed accent on the /o/ in these words including lhota/lhůta. Despite the fact that almost every example of {-ota} in CSR used as a deadjectival suffix is end-stressed, Dybo (1981:124) argues fairly convincingly for the derivational suffix {-ota} having fixed accent in precisely the OR word lgota (derived from an end-stressed adjective), an argument he bases on OR manuscripts and South Slavic manuscripts from two areas far removed from eachother. We believe, however, that there is another more likely explanation for the length in CZ lhůta. First of all it is necessary to mention the fact that in CZ place names, one finds almost exclusively lhota (Sedláček 1908), which designates an ancient feudal parcel that became free of obligations to the lord. OCZ then made use of the variation in quantity to lexically distinguish lhůta ‘deadline, term.’ It is far from certain, however, that the length in lhůta is attributable to a stressed PSL /o/ in {-ota}. Trávníček (1928:178) lists a number of forms which also had stressed internal /o/ but which show no lengthening, dialectal or otherwise: CZ kolo, oko, moře, boha, boje, boku, boru. In addition, EESL lists the OCS form of the word as ьlьgota, which suggests that the /l/ was a long syllabic /ĺ/. The word may have had stress on the /o/ in the OR word lgota, but it is likely that the PSL stress was on the syllabic /ĺ/. The transference of length from long syllabic /ĺ/ is not difficult to accept if one considers that Czech does indeed preserve a distinction in long syllabic /ĺ/ vs. short syllabic /l/ when preceded by consonants. OCZ dĺhý > dlouhý, but OCZ dlhъ > dluh. The length in vzhůru and shůry can also be explained in various ways (Trávníček, 1928:177-179). In short, there are simply too many uncertainties to form a theory to explain several rare forms when a much more logical and verifiable scenario can be presented.

87

stems in CZ that was later lengthened by their rule except when followed by a long vowel as in IS ranou, DP ranám, LP ranách, or in trisyllabic and monosyllabic forms such as IP ranami, GP ran. An additional condition for this sound law is that the vowel that was lengthened had to have been originally stressed. This condition is added in order to explain the failure of their sound law to affect the oxytonic class in the same way. In the oxytonic class, the root vowel is long throughout the paradigm except in the GP. The socalled Czech Lengthening therefore presupposes that the preservation of pretonic length and the lengthening of old acute vowels is unrelated, and that all of the acute lengthenings occurred after the loss of the weak jers. Aside from the fact that the logic is circular and largely unverifiable, we believe there is a more tenable explanation for the alternations in the (j)ā-stem paradigms, to which the dialectal and historical data lend greater credence. Interpretation of the Data One point that is clear from OCZ regarding original long roots is that there have been quite a few analogical levelings in the (j)ā-stems throughout the history of Czech. Since we believe that the so-called Czech Lengthening is untenable and that the evidence supports the hypothesis that CLC a.p. (a) (j)ā-stems with original long root vocalism retained original acute length, our task is to try and explain the many cases in CLC where these roots have lost their length. One possible interpretation is to attribute the loss of length in acute (j)ā-stems to some local Central Bohemian variation spreading from Prague as the center of the literary language. In several Southern Bohemian dialects (particularly those of Chodsko and

88

Doudlebsko), a number of these acute (j)ā-stems – as well as end-stressed ā-stems (albeit inconclusive for end-stressed jā-stems) and the neo-acute jā-stems – have a long vowel in the root (Voráč, 1955:28-29, cites pína, plíva, slína, hnída, jískra, sáze, káše, rejba, houba, strána, vína, douha), whereas in the literary language these forms have a short root vowel. The persistence of long vowels in these (j)ā-stems in the Southern Bohemian dialects has suggested to some scholars that length in the roots of (j)ā-stems in general spread as a morphological marker in these dialects (as opposed to representing the preservation of acute length). But we do not believe this is the case. All of the words noted by Voráč were either original acute or oxytonic words except for strána which was mobile.38 In truth, a long root vowel in (j)ā-stems in the Southern Bohemian dialects is not very regular at all. Circumflex (j)ā-stems with lengthened root vowels in the Southern Bohemian dialects are a rarity (strána is the only example we have seen). Circumflex (j)ā-stems, put another way, preserve the same pattern found in CLC (Chod Postřekov jehla, koza, husa, kosa, kopa ‘kupa sena,’ země, mez, zima, ruka, řeka, noha, struna), so if there had been a selective morphological extension of the lengthened vowel to (j)ā-stems in the Chod dialects, it is impossible to discover the morphophonemic environment to which the extension of the lengthened vowel was made without reference to the accentual paradigms. Moreover, the expected alternations in the IS or original a.p. (a) roots in these dialects is well preserved (cf. Map 2 & 8). The GP plural of many acute

38

The attested length in strána in several Southwestern Bohemian dialects (Voráč, 1955 Map Va (e.g. 5a) indicates two locations – one near Domažlice, and one near Sušice) might very well be the only example of retained length in the Czech dialects in the NS, (GS, LS) of a mobile lexeme that has provided the source of the analogical leveling throughout the paradigm. One is indeed surprised not to find other circumflex words with residual long vowels in the NS (GS, LS).

89

lexemes in the Chod dialects exhibit shortening of the root vowel (e.g. Chod Postřekov NS kníha, GP knih; NS mísa, GP mis). The question, then, is when and how did there come to be so many original a.p. (a) roots in CLC with a shortened original long vowel? The center-periphery pattern of shortenings shown by the maps cited above suggests that the forms with length are older and that the Central Bohemian dialects (including the Prague dialect) show a newer pattern with respect to the brevity in these lexemes. The dialectal patterns for the NS of jískra/jiskra, dráhy/drahy, jáma/jama, řípa/ řepa, hříva/hřiva, bříza/březa, kláda/klada, skála/skala, rána/rana, vrána/vrana, žíla/žila, lípa/lipa, hlína/hlina, síla/sila, hrouda/hruda, mříže/mřeža are indicative of what we believe to be a specific phenomenon characteristic of the development of a.p. (a) feminine (j)ā-stems and a.p. (b) feminine (j)ā-stems in Czech (and other acute and oxytonic stem classes). All of Moravian and Silesian (and Slovak too) exhibit a short vowel in these lexemes. We believe that the actual innovation was not the lengthening of original acute (residually rising) vowels in Czech (as Verweij and Kortlandt propose), nor shortenings in Moravian dialects that spread from Prague as the center of the literary language, but rather shortenings of acute vowels that spread from the East through lexical diffusion perhaps during the 9th and early 10th centuries when the Great Moravian Empire was at its zenith of power and prestige. Luckily, there is comparative Slavic evidence to support this view too. The fact that the Czechs and the Upper Sorbians were at the periphery of the Slavic expansion and theirs are the only Slavic languages to retain acute length suggests

90

that the Slavs had spread out too far for the phonological change [+acute +length > +acute ±length] to be extended to the periphery of the Slavic territory. Upper Sorbian does indeed provide evidence of retained length precisely in original acute ā-stems, although the reflex of length appears to be restricted to polnoglasie words: USo wróna, króva, brěza (vs. plesa) , klóda (but slama). The Proto-Slavic shortening of acute long vowels represents a dialectal sound change that was simply not carried out to the periphery of the Proto-Slavic dialects, and CLC (Pragocentric) and Upper Sorbian occupy the periphery. Instrumental Singular Analogical levelling of length in the instrumental singular of original a.p. (a) (j)āstems is prominent. Dialectal evidence for the preservation of an alternation in the IS is abundant, but it is obvious that there is no simple formula or dialectal-geographic pattern that can identify a preserved short vowel. The alternation is nearly exclusively restricted to original a.p. (a) roots. Gebauer (III. I.:183) points out that in OCZ not all fem. ā-stems with a long vowel in the NS shorten the vowel in the IS. The examples that he gives are all with root vowel /á/: CZ kára, krása, láska, páska, válka. These words are all either loan words borrowed into CZ, or internal CZ derivations (with the exception of CZ krása, R krasa – end-stress which we would not expect to shorten the long vowel in the IS). OCZ kára < OHG karre ‘cart’ < Lat. carrus was probably borrowed with length. None of them show any tendency to join the OCZ paradigm with brevity in the IS. We note that other (j)ā-stems with a long vowel that are loan words from GMC do not typically indicate that the shortening in the IS, (or the GP, DP, LP, and IP). CZ roura ‘pipe’ <

91

MHG rōre ‘reed, pipe’; routa OCZ rúta ‘Ruta graveolens’ < MHG rūte < LAT rūta; stoupa ‘thrashing machine’ < Proto-GMC *stamp-; píka ‘ancient stabbing weapon with a long handle’ < Germ. Pike; pípa ‘tap, pipette’ < LG Pipe; OCZ skýva ‘slice of bread’ < OHG scība ‘slice, round’ – are attested only with shortening in the GP. These facts lead us to believe that the F type paradigm was not productive even at the earliest stages of borrowing from Germanic and further supports the hypothesis that this paradigm represents the most archaic (and historically phonologically derivable) of all the ā-stem paradigms. Map 2 and 3 together show the distribution of long and short /a/ in the IS of kráva (kravou/krávou ‘with cow’). CLC does not allow the shortened form. Map 6 shows the locations of the Czech dialectal survey that have attested /í/ in the IS of bříza (břízou/břizou). By a comparison of these three maps, what becomes clear is that the cases where /í/ is long in IS břízou cannot be related to the same phenomenon as the locations of the Czech dialectal survey that have attested length in krávou. The IS form břizou is a secondary shortening from original /ěî > ie > í > i/. Only the IS forms březou can be attributed to shortening under the acute in trisyllabic words. IS březou is attested in AV survey locales in the Northern Bohemian dialects (118, 119, 130) in the Southern Bohemian dialects (307, 320, 424, 448, 457, 460, 462, 463, 464) and in the majority of the Central and Eastern Moravian. Map 3 also shows that there are in fact three phenomena associated with length in IS krávou. One is the innovation of the analogical levelling of long /á/ in IS krávou in the Central Bohemian and Southwestern Bohemian dialects, the second is what appears to be

92

a separate innovation of the analogical levelling of long /á/ in the IS krávou in the Central Moravian and Eastern Moravian dialects, and finally we see here illustrated the lack of opposition in quantity in the Silesian dialects. Genitive Plural It is noteworthy that where there is any alternation exhibited between brevity and length (CLC types D, E, F) within a given paradigm, the GP is the form most likely to exhibit an alternation. As mentioned above (Table 5), the generalization of brevity in the GP is also quite common in the a.p. (b) class of ā-stems (NS bída, bouda, díra, chvála, kroupa, touha, trouba, louka, tráva, brána, čára, váha, kůra: GP běd, bud, děr, chval, krup, tuh, trub, luk, trav, bran, čar, vah, kor). But there are no a.p. (c) words in CLC that have adopted an alternation between length and brevity in the GP. The shortened vowel in the GP has been attributed to the so-called neo-circumflex, whose environment is still not precisely defined. In the class of a.p. (a) (j)ā-stems, there is no phonological reason to posit the shortening of long vowels to the left of a weakened or lost jer. In the a.p. (b) oxytonic ā-stems, the GP is an environment where the neo-acute operated and we would expect preserved length in this environment. It has been suggested, however, that the spread of brevity in the GP is a morphophonemic change that is motivated by a principle whereby the GP is differentiated from the rest of the forms by the generalization of brevity when the direct forms are long, and the generalization of length when the direct forms are short. There does not, however, seem to be much generalization of length in the GP except marginally at a very early stage of OCZ. In Slovak, length in the GP is consistent for all (j)ā-stem paradigms. It is logical therefore to imagine that most of the

93

OCZ GP forms with length are long as a result of Slovak influence, particularly in light of the effect Slovak appears to have had on a.p. (a) paradigms in Czech. Oxytonic Feminine (j)ā-stems The fixed end-stress a.p. (b) (j)ā-stem substantives in Czech also reveal some interesting patterns when compared to the acute a.p. (a) and mobile a.p. (c) (j)ā-stem patterns. Table 12 through Table 14 represent Proto-Slavic words that Dybo (1993:13), Trávníček (1932:150) or others, identify as columnar end-stressed (j)ā-stem substantives in Proto-Slavic, as well as words that we are endeavoring to classify here for this study as oxytonic. For the long vowel (j)ā-stems, the identification of oxytonic words is complicated by the fact that in the mobile a.p. (c), the NS form had end stress (therefore we would expect retained length) which makes it the logical source of analogical leveling in the direction of the end-stressed paradigm. We would also expect neo-acute jā-stems to have the lengthened grade in the root in the NS (and throughout the paradigm) making the detection of neo-acute vs. oxytonic jā-stems in Czech nearly impossible. Length alone in the NS of any (j)ā-stem substantive in CLC tells us very little about the PSL accentual class the word belonged to. Words that belonged originally to the mobile a.p. (c), however, show remarkable consistency in their failure to preserve length in any grammatical case forms, and show remarkable regularity of generalized brevity throughout the paradigm.

94 Table 12 – Oxytonic disyllabic feminine ā-stems with original long root vocalism PSL

*běda

S/C

GLOSS

‘poverty’

bijèda AS bijèdu

SLOVENE

bẹ@da

CZECH

SLOVAK

bída,39 GP běd D

bieda

RUSSIAN

b’eda AS b’edu NP b’edy

*borna

‘harrow’

brána AS bránu

brána

OCZ brána D CLC NP brány40

brána

borona AS boronu / boronu NP borony

41

*buda

‘booth’

Ø

Ø

bouda, GP bud D

búda

Vas. OR buda42

*děra

‘hole’

déra AS déru Vas. dìra

dẹ@ra

díra, GP děr dial. d’oura D

diera

d’ira, dyra

guba

gọ@ba

houba, GP hub D

huba

*goba

‘mushroom’

AS dyru NP dyry Vas. OR guba AS gubu

39

CZ běda arch. ‘bad luck, woe’ is used adverbially e.g “Běda jim, běda těm cizincům z daleké země.” [Woe to them, woe to those foreigners from a far away land] (Vančura, 1939:68). 40

CLC brána ‘gate’ a.p. (a) and NS brána / NP brány ‘harrow’ a.p. (b) seem to be etymologically unrelated. Brána ‘gate’ shows variation between the D and the F paradigms in CLC characteristic of a.p. (a) roots, and brány ‘harrow’ has a short vowel in the GP, LP, DP, and IP which is peculiar for an original a.p. (b) root but is probably due to confusion with brána ‘gate.’ The conclusion to be reached is that there are two independent lexical items that are not etymologically related but which have nonetheless been differentiated grammatically to avoid lexical collision. Mach. (1971:64) indicates that brána ‘gate’ is based on a derivative *bornь from *bort- ‘to fight, struggle.’ The switch of this word to the ā-stem paradigm might be related to vrata ‘gate,’ even though vrata is neuter pl. The fact that in OCZ brána ‘gate’ is most often short in the IS, GP, LP, DP and IP, (Gebauer III. I:182) and in CLC the root vowel is optionally long or short in these forms indicates that the word joined the a.p. (a) paradigm in Proto-Czech. The word brána meaning ‘gate’ only occurs in CZ, SLK, and dial. SLN. We propose that the word was derived from *borniti ‘to protect’ in dialectal Proto-Slavic at a time when there was a certain degree of unity between CZ, SLK, & SLN. 41 42

See Map 16 for the dialectal distribution of length in the GP.

Vas. I: 230 says R buda is borrowed from P buda < OHG buode ‘tent’ rendering the place of ictus in R meaningless.

95 PSL

*krupa

GLOSS

‘grain of barley’

S/C

krúpa, AS krúpu

SLOVENE

krúpa

CZECH

kroupa D

SLOVAK

krúpa

RUSSIAN

krupa AS krupu NP krupy

*loka

‘meadow’

lúka AS lúku

lúka

43

louka / luka GP luk A/D

lúka

luka AS luku NP luki

*moka

‘flour’

múka AS múku

mọ@ka

mouka, GP muk D

múka

muka AS muku NP muki

*trava

‘grass’

tráva AS trávu

tráva

tráva, GP trav D

tráva

trava AS travu NP travy

*troba

‘trumpet’

trúba AS trúbu

trọ@ba

trouba, GP trub D

trúba

truba AS trubu NP truby

*tuha

‘desire’

túga AS túgu

túga

touha, GP tuh D

túha

tuga Vas. adj. tug, tuga, tugo

*xvala

‘praise’

hvála AS hválu

hvála

chvála, GP chval D

chvála

xvala AS xvalu NP xvaly

*borzda

‘furrow’

brázda AS brázdu

brázda

brázda B OCZ na brazdách

brázda

borozda AS borozdu and arch. AS borozdu NP borozdy

*jucha

‘fish soup’

júha AS júhu

júha

jícha B

jucha

uxa AS uxu NP uxi

43

The shortened form luka is the result of a morphological process whereby the –a adopted as the NP desinence from the o-stem neut. plural paradigms and shortened long vowel becoming the marker of collective nouns Geb. (III. I.:178). Other examples include mouka NP mouky but also muky,

96 PSL

*glista

GLOSS

‘worm’

S/C

glísta

SLOVENE

glísta

CZECH

hlíst, hlísta Vas. NS hlísta

SLOVAK

hlísta

RUSSIAN

glista, glisty glist, glista m.

*črěda

‘group’

čréda AS črédu

črẹ@da

třída B

trieda

čereda AS čeredu NP čeredy

*duha

‘rainbow’

dúga AS dúgu

Ø

duha A Chod dúha

dúha

duga AS dugu NP dugi

*hoba

‘mouth’

gubica

gôbec

huba A OCZ húba

huba

guba AS gubu NP guby

*kuna

‘weasel’

kúna AS kúnu

kúna

*pelna *pelena

‘wrap’

pelèna peleníca pèlen ‘bylica’

kuna A

kuna

Ukr. kuna

plena A Vas. pléna

plena

pelena AS pelenu NP peleny

*pila

‘saw’

Čak. pīla

píla

pila A

píla

AS pîlu

pila AS pilu NP pily

*rędsa

‘eyelash’

résa AS résu BG resa

rẹ@sa

řasa A OCZ řása

riasa

r’asa / r’asa AS r’asu44 NP r’asy

*ruda

‘ore’

Čak. ruda

rúda

AS rudu

ruda A OCZ rúda

ruda / rúda ruda AS rudu NP rudy

*sluga

‘servant’

slúga AS slúgu

slúga

sluha, A dial. slouha

sluha

sluga AS slugu NP slugi

44

Cf. P rzęsa with reflex of length (pretonic).

97 PSL

*strěla

S/C

GLOSS

‘arrow’

stréla, strijèla

SLOVENE

strẹ@la

CZECH

střela A

SLOVAK

strela

RUSSIAN

strela AS strelu NP strely

*strěcha

‘roof’

strexa

strẹ@xa

střecha A

strecha

AS strexu

str’exa AS str’exu NP str’exi

*vina

‘guilt’

Ø

Ø

vina A

vina

v’ina AS v’inu NP v’iny

*zvězda

*žlna

*vrba

‘star’

zvijèzda AS zvijèzdu Čak. zvêzdu

zvẹ@zda

hviezda

zv’ezda AS zv’ezdu NP zv’ezdy

‘woodpecker’ žúna (žúnja) AS žúnu

žólna

‘willow’

vŕba

vŕba, GS vrflby

hvězda A45

žluna A

žlna

želna AS želnu NP želny

vrba G dial. vŕba

vŕba

verba AS verbu

The development of the doublet běda / bída with lexical (and grammatical) differentiation formed off the same PSL word was made possible after the phonemicization of length in CZ. The root form běd- in the paradigm of OCZ bieda would have been morphonological only in the GP. It is appears, however, that the adverbialized form is built on the NS, which would have retained the long vowel pretonically. The R paradigm may give us a clue to a possible source for the stem běd-.

45

Bulaxovskij (1983:158) believes that the shortening in hvězda is by analogy to the GP. But if retraction of stress to the first syllable in the NP in Russian identifies a mobile type without retraction of stress to the first syllable in the AS, hvězda ‘star’ might be an example of a paradigm that was leveled analogically to the NP. Without the influence of Hollywood on Proto-Slavic, it is natural to imagine NP *zvězdy being used more often than the NAS. This same argument might be used for řasa, střela, střecha, sluha, but does not work for vina, short also in SLK.

98

Stang (1965:67) noticed that there seem to be other types of mobile words in the separate Slavic languages with stress on the first syllable only on the NAP but not in the AS. In Russian NS b’eda, NA b’edu, NAP b’edy follows this pattern. If Stang is correct in believing that this type in Russian (and Serbo-Croatian) is a [prehistoric] simplification of the mobile class with stress on the first syllable in the AS, and we assume that the word had circumflex intonation in the NAP, this would be the likely source for the shortened form of the root in Czech. CZ houba < *goba ‘mushroom’ (R dial. guba) and CZ huba < *goba ‘mouth’ (R guba) were probably both oxytonic and may be semantically related to one another because of the connection between the sponginess of the mouth and of mushrooms in general (Vas.). In several OCZ documents we find the etymological form with length (preserved pretonically) for CZ huba < *goba ‘tlama, mouth’; GS húby Budyš 58b, NP húby HrubLobk 67b. CZ lexically differentiated the two homonyms by utilizing a distinction in the length of the root vowel (hub- arising as a stem alternate from the GP) that other Slavic languages were not able to utilize. The only grammatical case in which the two lexemes are indistinguishable in CLC is in the GP where the root vowel shortens in houba GP hub and remains short in huba GP hub. The collision here in the GP, however is not likely to have ever caused any difficulty in understanding, and with the subsequent diphthongization of /ú/ /uo/ > /ou/ in OCZ further distinguishing the two words, the two lexemes continued to co-exist side by side, whereas even if such a lexical differentiation was initially present in other Slavic languages, the two lexemes as

99

homonyms might have not been able to persist because the collision of the two words in languages where a distinction in vowel quantity was not preserved, or where the distinction between /u/ and /ú/ was not transformed into a qualitative distinction, would not have supported the survival of one or the other lexemes. Slovene seems only to have preserved a word for ‘sponge’ gọb@ a although the root meaning ‘mouth’ is also evident in gôbec ‘mug, snout.’ S/C added a suffix gub-ica ‘mouth’ retaining guba as ‘leprosy, shelf-fungus.’ Russian has preserved the two lexemes but has differentiated them by preserving guba ‘mouth’ primarily used only in the plural guby ‘lips’ and guba with a very specific meaning ‘vid griba iz semejstva trutovikov, rastuščego na stvolax derevjev i vyzyvajuščego gnienie drevesiny’ (Ušakov, I.:635). Map 15 shows the dialectal distribution of length (primarily /ú > ou/ except in the Central Moravian dialects with a raised closed /ọ@/ for etymological /ú/) for the GP of CZ louka < lúka ‘meadow.’ The concentration of lengthened forms appears to indicate a local phenomenon that is not necessarily related to the generalization of length in the GP from Slovak, however Map 16 showing the reflex of length in the GP of bouda shows a much wider distribution, very concentrated along the Czech-Slovak border, with many more attested forms with length in the Southern Bohemian and Northern Bohemian dialects. It is hard to say whether the spread of length in the GP boud should be attributed to an ancient period of close political and social ties with Slovakia, to the generalization of length from the rest of the paradigm, or to some other factor without further evidence, but the maps suggest that the GP with length in these two forms took

100

separate paths. What is most notable about the development of oxytonic stems with original long vowels in Czech is the correlation between the Type D pattern with shortened vowel in the GP (a morphological change) and the lack of attested short vowels (OCZ and CLC) in the IS, DP, LP, and IP. Table 13 – Oxytonic / Mobile feminine disyllabic jā-stems PSL

*dězja

*nudza *medja

GLOSS

S/C

SLOVENE

‘wooden bowl for mixing dough’

dîžva46

‘need’

nuħa Vas. nužda

núja, nọ@ja

mèđa GS mèđi / međi

méja

‘boundary’

dẹ@ža

CZECH

díž OCZ dieže B

SLOVAK

dieža

‘serpent’

deža AS dežu U diža

nouze B

núzda

zmìja

nužda / nuža nuždu/nužu

mez(ě) A

medza

meža AS mežu NP meži / meži

AS među *zmija / zmeja

RUSSIAN

zmíja

zmije A

zmija

zmeja / zmija AS zmeju / zmiju NP zmeji / zmeji

The paucity of words that belong to this class make any generalizations about the development of quantity in Czech difficult, as do the difficulties in assigning these words to any particular accentual paradigm. CZ nouze could be either acute, oxytonic, or neoacute, SK núzda could be oxytonic or neo-acute, but S/C nužda indicates acute because of the short falling intonation, and Russian also indicates either acute, oxytonic, neo-

46

Stang (1965:59) suggests that words with the suffix {-va} might also have been associated with neo-acute intonation, although he points out that only *kletva had a PSL stem in {-va} while others are due to later transformations. At any rate, without a S/C word that is identical to the other Slavic words, there is no evidence for neo-acute.

101

acute. The word *medja originally had a short vowel and there do not appear to be any examples in the Czech dialects of a lengthened form. S/C AS među, CZ & SK without length argue for the mobile paradigm. The R NP alternate meži might also indicate mobility. The word *zmija / *zmeja is also unclear with respect to the accentual paradigm it belonged to. Vasmer (II:100) suggests that it may be an old ju-stem and if etymologically related to *{zem-} might be reconstructed as a PSL trisyllabic word *zьm(i/ь)j(a/u). In OCZ (Gebauer III.I:154) there are forms attested that show the reflex of the Czech contraction, zmí, but this is hardly surprising and is inconclusive with respect to quantity of the root in this word. Table 14 – Oxytonic disyllabic feminine ā-stems with original short root vocalism PSL

*groza

*kora

S/C

GLOSS

‘threat’

‘bark’

SLOVENE

gròza AS gròzu also gróza dial. groza (ESSL)

gróza

kora

kóra

dial. kòra

CZECH

SLOVAK

hrůza, GP hrůz47 B

hrôza

kůra, GP kůr B OCZ also kora

kôra

RUSSIAN

groza AS grozu NP grozy kora AS koru NP kory

*kopa

‘sheaf’

kúpa / kupa

kópa

(SČS)

kopa / kúpa A/B

kopa

kopa AS kopu NP kopy

*kosa

‘scythe’

kòsa AS kosu

kósa

kosa / kůsa A/B

kosa

kosa AS kosu / kosu NP kosy

47

According to Bulaxovskij (1983:280) the length in the CZ form is a result of the expressive nature of the word which means ‘horror, terrible.’ This explanation would perhaps be acceptable, if not for the fact that the simpler and more likely explanation of a phonetic lengthening due to retraction off the original end stressed syllable makes more sense in this case.

102 PSL

S/C

GLOSS

SLOVENE

CZECH

SLOVAK

RUSSIAN

*mora

‘nightmare’

mòra AS moru

móra

můra, GP můr B

mora/mura

_____

*smola

‘tar’

smòla AS smolu48

smóla

smola / smůla A/B

smola

smola AS smolu NP smoly

*sova

‘owl’

sova

sóva

AS sovu

sova / sůva A/B

sova

sova AS sovu NP sovy

*stopa

‘step’

stòpa

stópa

stopa A

stopa

stopa AS stopu NP stopy

*žena

‘woman’

žèna

žéna

žena A

žena

žena AS ženu NP ženy

*osa

‘wasp’

òsa

ósa

vosa A Chod vůsa

osa

osa AS osu NP osy

CZ kopa/kúpa ‘60 pieces of something, pile of hay’ is problematic for a number of reasons. The S/C form does not support oxytonism in this case, and there is no doubt lexical confusion as well with kupa ‘pile of hay.’ The word kůra ‘bark’ in OCZ is attested with the shortened form korou in the IS, presumably indicating shortening of the root vowel and not the absence of length throughout the paradigm (Gebauer I: 600), although other OCZ texts (Prešp, Rozk, BohFl.) show that there was variation very late and that at least some dialects either by analogy adopted NS kora or never adopted NS kůra. What is interesting to note is that not only do pretonic long vowels remain long in the oxytonic class of Czech stems, but also pretonic short /o/ lengthens in many roots as if mimicking the lengthening effect of the neo-acute on short vowels. This fact gives us a

48

Stankiewicz (1986).

103

potential way of identifying pretonic (j)ā-stems as opposed to mobile (j)ā-stems in ProtoSlavic and perhaps even of identifying neo-acute from original oxytonic stems. Slovak does not lengthen pretonic /o/ in this class as consistently as CZ, but does lengthen /o/, as well as /e/ and front and back jers, under the neo-acute. The behavior of the a.p. (b) (j)āstem substantives in Czech is parallel to the lengthening of short /o/ in various other word classes in Czech. Neo-acute classes that exhibit the lengthening of /o/ in Czech include the class of jā-stem substantives represented by words like vůle ‘will,’ růže ‘rose,’ stůně ‘moan;’49 o-stem substantives with Proto-Slavic ictus retracted from a jer, such as vůl ‘ox,’ důl ‘mine.’ The details surrounding lengthened /o/ in these classes point to several peculiarities that are worthy of note because they prove that the lengthened vowel has been transferred to roots that are members of the same morphological class but not necessarily related in the way that analogical changes are typically transferred – e.g. phonologically or phonotactically. Further factors influencing the retention of Proto-Slavic length in Czech (j)ā-stem substantives relate to the fact that in non-initial position there could be no opposition between circumflex and acute intonation. Since circumflex intonation occurs only on the initial syllable of the mobile paradigm, acute intonation and circumflex intonation were distinctive in word initial position only. This fact might have influenced Czech insofar as it made length in PSL in the initial syllable of the word redundant because the opposition between acute and circumflex intonation in the initial syllable carried the phonemic

49

CZ dial. vůlše ‘Alder orchard’ (< *jelьxa / *olьxa), SLK jelša, S/C jóxa, SLN jélša SLN dial. ólša, R although not a jā-stem neo-acute, shows olьxa R dial. vωlьxa, also shows Czech lengthening of neoacute retraction onto a short vowel.

104

distinction. The fact that Czech provides three distinct paradigms in answer to the three PSL paradigms, fixed root stress, mobile, and neo-acute, means that at the time (e.g. before the development of phonemic length in Proto-Czech) there were three distinct intonations: acute, circumflex, and neo-acute. Phonemic redundancy begets variation, and variation encourages innovation. With intonation being contrastive in this environment, the phonemic load for the distinction between long and short vowels to carry would have been reduced and therefore not as critical for lexical or grammatical distinctions as quantity might have been were it to have been the only distinctive feature in this environment. But the variation in vowel length independent of intonation would have strengthened the opposition of phonemic quantity in Proto-Czech, and with the loss of a phonemic distinction in intonation (triggered by the rise of the neo-acute), the role of quantity as a distinctive feature gained in importance. The implication here is that during the early Proto-Czech period when intonation still played the primary role in distinguishing like roots from one another (e.g. mouka:muka), quantitative distinctions might have been less stable and more likely to vary dialectally or from one speaker to the next. This might partially explain the tremendous variance in the Czech dialect geography with respect to quantity. Despite this variance, however, what is remarkable is the fact that patterns associated with original Proto-Slavic intonation and the phonological reflexes of these intonational patterns persisted into the Proto-Czech period.

105

Mobile Feminine (j)ā-stems Table 15 – Mobile disyllabic feminine ā-stems with original long root vocalism PSL

GLOSS

S/C

SLOVENE

CZECH

SLOVAK

*borda

‘beard’

bráda AS brâdu

bráda

brada A

brada

*cěna

‘price’

céna, cijèna AS cijènu Čak. cēna

cẹ@na

cena A

cena

*cěsta

‘trip’

cesta AS nà cestu

cẹ@sta

cesta

*golva

‘head’

gláva AS glâvu

gláva

cesta A OCZ ciesta51 hlava A

hlava

*gręda

‘perch’

gréda AS grêdu

gréda

hřada A

hrada

*pęta

‘heel’

péta AS pêtu

pẹ@ta

pata A

päta

*rěka

‘river’

réka AS réku/rêku

rẹ@ka

řeka A

rieka

*slina

‘saliva’

slına AS slınu

slína

slina A Chod slína

slina

*stěna

‘wall’

stijèna AS stıjenu

stẹ@na

stěna A

stena

*storna

‘side’

strána AS strânu

strána

strana A Chod strána

strana

*zima

‘winter’

zíma, AS zîmu

zíma

zima A

zima

RUSSIAN

boroda AS borodu cena AS cenu AS cenu50 cěsta Srez. golova AS golovu gr’ada AS gr’adu p’ata AS p’atu r’eka, AS r’eku / r’eku Vas. OR sl’ina / sl’ina st’ena AS st’enu storona AS storonu z’ima AS z’imu

50

CZ cena (along with zima and žena) are identified as oxytonic by Trávníček (1932a:150). Bulaxovskij (1954:153, 161) believes that the end-stressed Russian AS forms attested in the 1st half of the 19th century are secondary, however the comparative evidence suggests cena was oxytonic. Illič-Svityč includes this word (1963:26) among his list of Indo-European roots with mobile-oxytonic accent because of the fact that kainà belongs dialectally to class 4 in Lithuanian and because the Greek cognate is end-stressed. Literary Lithuanian, however, treats káina as a class 1 substantive. For Illič-Svityč (1963:100), the word was mobile in Slavic. Since there does not seem to be any evidence in the Czech dialects or in OCZ for length, even though S/C indicates a.p. (b), we believe it was mobile, a.p. (c). 51

See Trávníček (1935:258) for a possible explanation of length in OCZ ciesta.

106 Table 16 – Mobile disyllabic feminine ā-stems with original short root vocalism PSL

*doba52

GLOSS

‘period’

S/C

doba neut.

SLOVENE

dóba

CZECH

doba A

SLOVAK

doba

RUSSIAN

doba AS dobu

indecl.

ukr. doba *gora *koza

*loza

*noga *rosa

‘mountain’ ‘goat’

‘rod’

‘leg’ ‘dew’

gòra AS goru

góra

kòza AS kozu

kóza

lòza AS lozu

lóza

nòga AS nogu

nóga

ròsa AS rosu

rósa

hora A dial. hůra53

hora

koza A

koza

gora AS goru koza54 dial. AS kozu

loza (OCZ) A

loza

noha A

noha

loza (Ukr.) AS lozu (Ukr.) noga AS nogu

rosa A 109 rúsa is puzzling

rosa

rosa AS rosu AS rosu (arch.) 55

*soxa *voda

‘dry twig’ ‘water’

sòha AS sohu

sóha

vòda AS vodu

vóda

socha A

socha

soxa AS soxu

voda A

voda

voda AS vodu

Identification of this class rests heavily on the alternation in the AS of the R paradigms. As noted by Comrie (1978:58-59) however, CSR cannot be relied on as the

52

Vas. (1986, I.:518) indicates that this word was originally neut. plural, which would explain the stress pattern in R and S/C. Nevertheless the CZ and SLN words behave as mobile stems. 53

CZ hůra ‘garret’ (Bartoš. Slov. 111, Bělič. Dolsk. 174) hora, hóra (Vážný. Středověk. list. 20), East SK dial. hura ‘mountain’ húra ‘garret’ (Kálal 191). 54 55

R NP kozi DP kozam is a partial argument as well for mobility.

Bulaxovskij (1954:160). Further evidence that rosa belongs to a.p. (c) is the fact that there is only one CZ dialect with *rusa/růsa attested.

107

only source for the original ictus pattern in R, for “In the nineteenth century, there were rather more words in this [mobile] class, including v’esna, zola, izba, nora, ovtsa, rosa, soxa.” One cannot help but notice that most of these words are of the short root vowel variety. The mobile feminine (j)ā-stems are generally characterized by brevity in the root vowel in Czech. Roots with short vowels, by virtue of the fact that short vowels could not carry acute intonation in Proto-Slavic unless the accent was fixed by a morphological process, exhibit the pattern expected for the mobile paradigms. Examples of lengthening of original short vowels for this class of circumflex stems is extremely rare. CZ zima is noteworthy since the lengthened root vowel can be found in a number of dialects in the North and Central Bohemian dialect regions and several Central Moravian dialects (see Map 13). The dialectal distribution of length in this word together with the fact that the Southern Bohemian dialects do not indicate any preservation of length suggest two things: first of all, it shows that lengthening in general in the (j)ā-stems in the Chod dialects and Southern Bohemian dialects is not a generalization made for all (j)āstems regardless of PSL class, and secondly, it indicates that there were some roots whose PSL intonation was in variation. A.p. (b/c) řeka is probably another example of a PSL word with intonational variation in the PSL dialects . We regard Trávníček’s (1932a:150) classification of zima as oxytonic to be puzzling unless he considered the word to show intonational variation in PSL. Since the distribution of lengthened circumflex in zíma is much different from the preservation of length in pretonic a.p. (b) words and the preservation of acute and neo-acute length, there must have been a difference in intonation.

108

Relative Chronology The fact that there is not more residual length in the end-stressed forms of original mobile words might suggest that the shortening of circumflex length is an older, albeit dialectal, phenomenon in Proto-Czecho-Slovak than the shortening of acute length is in dialectal Proto-Slavic. Most of the irregularities that the strictly phonological shortening of circumflex intonation would have created in the paradigm left no trace on the language by the historical period, whereas the irregularities created by the shortening of acute length in trisyllabic forms are preserved precisely in the ā-stem paradigm of CLC Type F. If it is true that the Proto-Czecho-Slovak shortening of circumflex length predated the shortening of acute length in PSL, since S/C and SLN retain circumflex length, we are faced with a situation whereby an opposition between acute length and circumflex length became an opposition between length and brevity in dialectal PSL Czecho-Slovak. Then when the shortening of PSL acute length began to operate (brought on by the rise of the neo-acute), circumflex and acute intonations essentially merged in Slovak, Moravian, and South Slavic. We imagine then two waves of sound change: the first wave that shortened circumflex length in Proto-Czecho-Slovak but which did not extend to South Slavic, and a second wave that shortened original acute length that extended to South Slavic but excluded Western Proto-Czech (and Upper Sorbian). The later operation of the shortening of acute length coincided with the greater expansion of the West Slavs to the West. The historical data corroborates the linguistic evidence in this case.

109 Table 17 – Mobile feminine disyllabic jā-stems with original long root vocalism PSL

*dusja

S/C

GLOSS

‘soul’

dúša AS dûšu

SLOVENE

dúša

CZECH

duše A

SLOVAK

RUSSIAN

duša

duša AS dušu

Table 18 – Mobile feminine disyllabic jā-stems with original short root vocalism PSL

*zemja

S/C

GLOSS

‘earth’

zèmlja AS zèmlju DS zemlji

SLOVENE

zémlja

CZECH

země A

SLOVAK

RUSSIAN

zem

zeml’a AS zeml’u

There are not enough examples of these paradigms to make any real generalizations about the development of quantity in the Czech dialects. We can only observe that circumflex length has been lost in the one example we researched and that is attested in enough Slavic languages to be considered Common Slavic. There may be other mobile feminine disyllabic jā-stems with original short root vocalism that can be considered Common Slavic. Neo-Acute Feminine jā-stems Table 19 – Neo-acute disyllabic feminine jā-stems with original long root vocalism PSL

GLOSS

S/C

SLOVENE

CZECH

SLOVAK

RUSSIAN

*gostja

‘thick liquid’

gušta ‚ Vas. ‘čašča’

gọ@šča

houšť/houště OCZ húšč/húščě

húšť

gušča

*gordja

‘barrier’

grâđa

grája

hráz(e)

hrádza

goroža

*kupja

‘buy’

kûpl’a

kûpnja

koupě

kúpa

kupl’a Ukr. kupl’a

*suchja

‘drought’

sûša

súša

souš (OCZ súše)

súš

suša

*storgja

‘guard’

strâža

stráža

stráž(e)

Ø

storoža

*tęgja

‘weight’

têža/téža

téža

tíž(e)

tiaž

t’až

*žęd’a

‘thirst’

žêđa

žéja

OCZ žiezě

_______

žažda

110 Table 20 – Neo-acute disyllabic feminine jā-stems with original short root vocalism PSL

GLOSS

S/C

SLOVENE

CZECH

SLOVAK

RUSSIAN

*xodja

‘gait, walk’

xòđa

hója

chůze

OSK chôdza

xožij

*xvoja

‘evergreen’

xvója/xvòja

hvôja

chvojě/chvůjě56

chvoja

xvoja dial. xvoja

*kozja *nosja

‘skin, leather’

kofiža

‘rucksack’

nòšnja

kóža

kůže

koža

koža dial. kωža

nóša

nůše

noša

noša dial. nωša

*volja

‘will’

vofilja

vólja

vůle

vol’a

vol’a dial. vωl’a

*von’a

‘smell’

vofinj, vònja

vónja

vůně

vôňa

von’ dial. vωn’a

Neo-acute intonation, strictly a Slavic development, arose from the retraction of ictus onto the preceding syllable. In Czech, the neo-acute intonation often shows up as length, which is most visible in its effect on originally short vowels when they show lengthening. The most prominent and widely accepted environment for this retracted ictus intonation is in words that ended in a stressed jer (e.g. *korl'ь > kor'ôl’). Yet it is difficult to make a case for the rise of the neo-acute intonation as a phonetic development. The neo-acute, on the other hand, can be identified in a number of specific morphophonemic environments with evidence for the neo-acute intonation in Czech. Here we are most concerned with the class of jā-stems that can be identified as neo-acute, either from comparative evidence, or from the lengthened /o/ that is preserved in CLC

56

CZ chvoj, chvoje, chvůj, chvůje together with the S/C and R forms indicates that there was variation in PSL with regards to the intonation. It seems that there might have even been variation with respect to the place of accent.

111

and the dialects. And in fact, the neo-acute forms a class of stems in the substantival system only for the jā-stems. A special group/class of constant root stressed feminine jāstem substantives was identified by Stang (1965:57) as neo-acute. In Czech these roots almost always contain a long vowel or lengthened short vowel throughout the paradigm. Verweij and Kortlandt believe that pre-tonic length is preserved length in CZ and that acute length is relengthened. If this is right, we would expect length to be retained in all cases of neo-acute stems such as *volja including the IS. If there is a difference in the dialectal distribution of *volí / vůlí vs. břízou / břizou, then there is a difference in intonation. And in fact there is a difference, we have not uncovered any dialects with the IS volí, although this form is attested in OCZ. Stang posits for the neo-acute a retraction off circumflex endings such as the NS in certain end-stressed jā-stems such as *volj-a. Stang (1965:13) also believes that the neo-acute resulted from the retraction from a stressed internal circumflex syllable to the preceding syllable – even off of forms he identifies as neo-circumflex: e.g. godîn > godin. Dybo (1993:15) rejects this theory based in large part on what he calls the law of Dybo and Illič-Svityč, which posits a Common Slavic shift of accent to the right off of short-circumflex syllables in Balto-Slavic ortho-tonic forms. Phonologically under the neo-acute retracted ictus from a word final jer, short /o/, short /e/ and even jers show up as long in Czech and Slovak (e.g. CZ kůň ‘horse,’ déštˇ ‘rain,’ SK jéž ‘hedgehog’). In the GP, the Czech literary language has leveled out most lengthenings of short /o/, /e/ and jers, but the dialects sporadically indicate that there was lengthening under the neo-acute in the GP (for example dial. nuh ‘legs,’ okén dial. ‘windows’). Some classes of words

112

that Stang (Slavonic Accentuation, 1965) says are neo-acute on short vowels with retracted ictus never show length in any language. PSL fixed theme vowel stressed verbal stems in Czech show the preservation of the long vowel under the neo-acute, for example vrátiti ‘to return’ is long and follows the PSL pattern of fixed theme vowel stressed conjugation (cf. ztratiti ‘to lose’ which is short and mobile), but reportedly short vowel neo-acute verbs such as the i-stems voditi, nositi, choditi which have retracted ictus in the 2nd and 3rd persons (cf. Russian ношу, носишь) exhibit no trace of lengthening neither in Old Czech nor in the Czech dialects57. This means that there was no general lengthening of short /o/ under the neo-acute retracted stress. Stang also argued that the motivation for neo-acute retraction in the jā-stems such as vůle was the presence of circumflex long vowels in word final position. He cited evidence from Polish dialects that retained length on the word final /a/ in this class. Some scholars believe that there is no basis for Stang’s belief that the cited Polish dialects indicate circumflex intonation on the desinential /a/. If this is true, it leaves no motivation for a phonologically based neo-acute in this class of words. A possible explanation is that this particular neo-acute development is a later, different development of the neo-acute and that the lengthening of the /o/ in Czech reflects a morphological development that is specific to Czech and Slovak. Czech lengthening of /o/ in the pretonic B-type ā-stems

57

In Czech there is some residue of e-stem neo-acute lengthening (e.g. mohu, můžeš, beru, béřeš, etc.), which might indicate that the lengthening of /o/ under the neo-acute in the verbal estem class was phonological but not in the verbal i-stem class. Another explanation put forward originally by Vondrák (Gebauer III, 1894:33-34) is that the lengthening in these e-stems is somehow related to the old optative. At any rate, it is clear that the putative neo-acute intonation in the i-stem verbs is not related to whatever is going on in the e-stem verbs.

113

(see Table 14) is probably an extension of a class marker which is length, phonologically based only in the lengthening of short /o/ off the neo-acute in masculine substantives such as kůň. It is interesting that it is precisely the pretonic short /o/ of a.p. (b) fixed end stress ā-stems to which this class marker is extended. Phonologically lengthened short /e/ from the neo-acute retraction does not become a morphological marker extended to the a.p. (b) ā-stems as shown by the lack of length in Czech žena ‘wife.’ The extension of this lengthened /o/ class marker in Czech then serves to distinguish between a.p. (b) pretonic jā-stems and a.p. (c) mobile (j)ā-stems, a distinction that would have been seriously blurred if not obliterated by the fixing of ictus on the initial syllable of the phonological word in Czech. The solution of this problem also has bearing on a more general question of Slavic accentology, namely, the question as to whether Proto-Slavic acute length was shortened everywhere including Czech and length was then reintroduced secondarily in Czech, or whether acute length in Czech is preserved Proto-Slavic length. As noted above, the argument for the shortening of acute length in Czech is bolstered by the shortening of acute length in all other Slavic languages with phonemic quantity including Slovak. The problems in formulating a comprehensive theory for the existence of sound laws that operated to restore length in Czech and in defining the conditions and factors responsible for a reintroduction of length (Verweij, 1994:505) have proven formidable. This theory of Czech Lengthening, repeated here for convenience, is based on speculation that the original rising intonation persisted on long and short vowels: “A short rising vowel in an open first syllable of disyllabic words is

114

lengthened unless the second syllable contains a long vowel.” This theory was originally expounded in Kortlandt (1975:19) with the only difference in his formulation the stipulation that the law was operative in early Czech. Verweij additionally notes that Stang (1965:35) first speculated about the existence of some such sound law when he wrote: “As we have seen, in Czecho-Slovak, in the first syllable of an old disyllabic word, acute appears as long and circumflex as short. But it is possible that this may be due to a secondary lengthening of a rising vowel.” Stang (1965:36), however, states that he finds no convincing argument that acute was shortened in Proto-Slavic, and supports this idea by referring to the Latvian prosodic system with a three way opposition in tone (rising = Dehnton, falling = Fallton, broken = Stosston). The Czech Lengthening presumably operated in Czech after the loss of weak jers which can be discerned from the examples given by Kortlandt (kráva NS, but krav). Although there is some merit to the Czech Lengthening, for the masculine monosyllabic stems with a lengthened short vowel, the theory seems to fall apart. Verweij (1994, p. 526) is forced to admit that the paradigm to be expected where the Czech Lengthening is to have operated (lengthened vowel in the GS, DS, IS, LS, NP, AP, and IP) is nowhere attested in Czech. In fact quite the contrary, instead of lengthening of short rising (in their view) vowels in the GS, DS, IS, LS, NP, AP the only case forms with lengthened vowels are the NS and the AS. In other words closed syllables in predominately fixed end-stress paradigms. Verweij has no way of explaining the failure of the old acute to lengthen in the NAS either.

115

Feminine monosyllabic i-stems Table 21 – Feminine third declension substanival paradigms in CLC Case / Number NS GS DS AS IS LS NP GP DP AP IP LP

A (all short)

moč moče moči moč močí moči moče močí močím moče močemi močích

B (all long)

louč louče louči louč loučí louči louče loučí loučím louče loučemi loučích

C (long NAS, else short) hůl hole holi hůl holí holi hole holí holím hole holemi holích

D (short GP, else long) N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A

E (short GP, DP, IP, LP, else long) N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A

F (short IS, GP, DP, IP, LP) N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A

Monosyllabic feminine i-stems follow the same patterns as the masc. monosyllabic stems for the most part, although the number of feminine monosyllabic stems in Czech is far fewer than for the masculine stems. The originally long vowel stems that remain long in CLC louč ‘kindling,’ pout ‘trip’ (masc. > fem.), ciev ‘kindling,’ pied ‘kindling,’ tvář ‘cheek,’ dásn ‘gum tissue,’ písn ‘song,’ tísn ‘kindling’ lén, čést, kád, mlád, pied, sín, síť, básň, bázň, kázň, plísň, přízň, vášň, žízň. Original long vowel stems that are short in CLC include běl, (s)běr, bran, hrud, hus, chutˇ, lat, lut, mast, směs, mysl, myš, nat, nit, past, pěst, řeč řit, san, sěč, strast, trest, věc, vlast. Short vowel stems for the most part remain short in CLC and include kost ‘bone,’ moc ‘power,’ moč (m./f.), noc ‘night,’ zlost ‘night,’ čest ‘night,’ lest ‘night,’ lež ‘lie,’ pec ‘night,’ plet ‘night,’ rež ‘night,’ ves ‘night,’ zed ‘night,’ žen ‘night.’ At least two feminine i-stems with original short vowels have lengthened the root vowel – hůl, sůl. There are isolated examples from OCZ manuscripts of other long vowels in forms that occur primarily with short vowels;

116

krev/krév, lež/léž, ves/vés, rez/réz, but not attested in more than one manuscript or consistently enough to deserve mention.

Chapter 6

MASCULINE MONOSYLLABIC STEMS

Expected reflexes of root vowels Table 22 – Monosyllabic masculine paradigms in CLC

Case / Number NS GS DS AS IS LS NP GP DP AP IP LP

A (all short)

dub dubu dubu dub dubem dubu duby dubů dubům duby duby dubech

B (all long)

žák žáka žákovi žáka žákem žákovi žácí žáků žákům žáky žáky žácích

C (long NAS, else short)

dům domu domu dům domem domě domy domů domům domy domy domech

D (short GP, else long)

N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A

E (short GP, DP, IP, LP, else long) N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A

F (short IS, GP, DP, IP, LP, else long) N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A

There are fewer possible patterns of long versus short vowel alternations in the roots of masculine monosyllabic (originally disyllabic) substantives than there are for the feminine (j)ā-stems. Even the historically trisyllabic forms (IS, DP, LP) show less dialectal variation in the masculine substantival sub-system than the trisyllabic forms in the (j)ā-stem substantival sub-system. The masculine monosyllabic substantives are original o-, jo-, u- and i-stems, originally disyllabic (i.e. in the NS before the loss of weak jers), with weak /–ъ/ and /–ь/ in word-final position in the NAS (if animate NS only) and GP cases. Among them we find both words very basic in nature and of very old deverbal derivation together with several loan words. The following categorization is 117

118

organized by PIE stem class and the Proto-Slavic accentological type to which the words belonged. The accentological type has been determined by comparing relevant forms in the modern Slavic languages. As noted above, R has been used to determine the place of ictus,1 and the combination of S/C, SLN, CZ and SLK has been used to determine the accentual paradigm: fixed root [a.p. (a) = acute intonation] fixed end-stress [a.p. (b) = oxytonic] mobile stress [a.p. (c) = circumflex intonation]. Some reference is made to Dybo’s d-type accentual paradigm [a.p. (d)] (Dybo, 1993:28) that is essentially a variation on the oxytonic a.p. (b) – the chief difference being in the NAS where instead of fixed end stress, the forms are believed to have had an intonation that Dybo calls enklinomennaja nisxodjashchaja. Since the acute fixed stem stress paradigm a.p. (a) represents forms with stressed intonation that could occur only on long vowels (in monosyllabic stems), as we would expect, no words with original short vowels can be reliably assigned to this paradigm. Serbo-Croatian forms and Slovene forms quoted are from the literary language except as noted otherwise. As far as the classification of stems into the accentual paradigms a.p. (a), a.p. (b), and a.p. (c) is concerned, we have leaned heavily on Dybo (1993), Illič-Svityč (1963), and Stang (1965). Since the arguments made in this study rely to some degree on the accent class to which the lexical items belonged, caution is used to classify each word as accurately as possible. Since, however, there is considerable disagreement about the membership of words to Proto-Slavic accentual

1

In some known cases, CSR does not preserve the original place of ictus and will be noted.

119

paradigms, when a word is not clear about its membership to a specific accentual paradigm, we have tended towards the view that best represents the data from the entire analysis of the masculine monosyllabics and the classification that allows for the most comprehensive view of the languages and dialects involved. Where there is conflicting data from other Slavic languages or dialects, or disagreement among Slavic accentologists regarding the classifications below, we have tried to indicate the divergence. If we look at the three LPSL accentual paradigms: acute [fixed root stress=a.p. (a)], oxytonic [fixed stress on first syllable of the desinence=a.p. (b)], circumflex [(mobile=a.p.(c)], we can initially derive the following paradigms for the masculine (j)ostem disyllabic substantives according to the historically determined rules for the development of length in Czech:2

2

Tables 24 thru 27 apply to the u-stems as well for the most part. By early Czecho-Slovak, the o-stem and u-stem paradigms were virtually indistinguishable.

120 Table 23 –Acute monosyllabic masculine (j)o-stems in CLC with original long root vocalism

Case / Number

Proto-Slavic Form according to Stang Reconstruction (1965:68-69)

Rule for Preservation or Loss of Length in original Long Root Vowel

NS

*dym ъ

Acute length preserved

DÝM

dým

dým

GS

*dym a

Acute length preserved

DÝMA

dýma/u

dýmu

DS

*dym ě

Acute length preserved

DÝMĚ

dýmu

dýmu

AS

*dym ъ

Acute length preserved

DÝM

dým

dým

IS

*dym ъmъ3

Acute length lost in word greater than two syllables

DYMEM

dýmem

dýmem

LS

*dym ě

Acute length preserved

DÝMĚ

dýmě

dýmě

*dym i

Acute length preserved

DÝMI

dýmy

dýmy

GP

*dym ъ

Acute length preserved

DÝM

dýmóv

dýmů

DP

*dym omъ

Acute length lost in word greater than two syllables

DYMOM

dýmóm

dýmům

AP

*dym y

Acute length preserved

DÝMY

dýmy

dýmy

IP

*dym y

Acute length preserved

DÝMY

dýmy

dýmy

LP

*dym ěxъ

Acute length lost in word greater than two syllables

DYMECH

dýmech

dýmech

NP 4

Expected CLC Form

5

Old Czech Form

Literary Czech Form

3

Although the older o-stem IS desinence is undoubtedly {-omъ}, West Slavic generalized {-ъmъ} quite early, and the oldest West Slavic manuscripts already illustrate this fact. The origin of {-ъmъ} is almost certainly the u-stem declension. 4 5

For the history of this case form, see Gebauer (III.I:54-58).

There are several dialects that we know of that have the expected reflex in the DP desinence of acute stems. Dušek (1894-1908, III:24) names Kouto, Kolinec, Chrást, Bušovice, Měčín as villages in Southern Bohemia with either {-om} or {-um} as the regular DP o-stem desinence. The dialectal {-um} desinence is undoubtedly a much later shortening {-um < -ům < -uom < -óm}, which must have taken place after the diphthongization of /ó/ that operated from the end of the 14th to the beginning of the 16th centuries, but it is possible that the dialectal desinence {-om} represents an archaicism of the original phonologically correct o-stem form. Gebauer (III.I:58-59) assumes that {-om} attested in the Chod dialects (Domažlice) and in Doudlebsko represents shortening of the original {-óm}, but we do not see any compelling reason that the shortened form may not represent the original desinence and that the lengthened form (with Czech neo-acute or compensatory lengthening) was later generalized. The phonological source of the lengthening in this desinence is problematic.

121

There are no paradigms in CLC that match the expected result of the PSL acute (j)o-stem and u-stem paradigm for every grammatical case. Leaving aside the analogical irregularities of the grammatical case morphemes, the expected shortening of acute length in the root of the trisyllabic forms (IS, DP, LP) does not survive as a declensional pattern. There are a handful of original disyllabic a.p. (a) o-stems with long vowel in the NAS and shortening throughout the rest of the paradigm in CLC (chléb, mráz, hrách, práh, vítr) and OCZ (Ján, mák,6 pán, pás < *pojasъ), but there is doubt as to whether this pattern reflects an archaic a.p. (a) paradigm. This same alternation pattern is common for original a.p. (b) roots with a short vowel that was lengthened in the NAS such as dům. Unfortunately, attestations of shortenings in only the IS, DP, and LP (the expected paradigm) for a particular acute lexeme in OCZ or even in a particular OCZ text are nowhere to be found in the literature or in OCZ manuscripts that we know of. Perhaps closer examination of OCZ manuscripts with an orthographic distinction of long and short vowels would provide documentation of the expected paradigm. The alternation in NAS pás G pasu gives us a clue about the diachronic behavior of this alternation. In the NAS as well as in the other grammatical cases we expect the phonological result of the contraction to be a long vowel (e.g. pásъ, < *pojasъ, pásu < *pojasu). The word pás, pasu must have adopted the alternation subsequent to the completion of contraction in the Czech lands, which means sometime after the beginning of the 12th century (Marvan, 2000). Furthermore, since we imagine that the most likely source for this alternation is in

6

CLC mák no longer preserves the pattern with short vowel in the indirect cases.

122

the words of the type NAS dům, GS domu, etc. where the length developed after the loss and vocalization of the jers, the alternation must be a development in pás that occurred independent of PSL intonation. Naturally, it is impossible to prove that the alternation in pás has the same source as in mráz and hrách or vice versa, but the possibility remains. In the GP, Gebauer (III.I:55) lists OCZ čás, hód, bóh, lék, hřiech, pár, plát, ráz with the old o-stem zero ending and length in the root. He notes that these forms (GP in {-Ø} with or without length) are much more common in the oldest texts and are nearly non-existent in CLC (except in the occasional place name, e.g. Hradčan), but there is no correlation between length in the root in the GP and the original PSL accentual paradigms. Lengthening or preserved length in masculine disyllabic roots in the GP with {-Ø < -ъ/ь} would have been phonological in stems of a.p. (a), a.p. (b), and in a.p. (c) where the stress was on the jer. The tendency to generalize length or brevity in the root in the GP to distinguish this from from the NS does not seem to have been characteristic of the masculine substantives as it was with the (j)ā-stems, nor does SK preserve the original (j)o-stem GP {-Ø < -ъ/ь} desinence. Both Czech and Slovak generalized the ustem {-ovъ} in the GP. Lengthening in Czech (but not Slovak!) and an anomolous loss of the final /v/ gives us the CLC desinence {-ovъ > -óv > -uov > -ův > -ů} with no alternation in the root. Yet the OCZ forms cited by Gebauer above are likely to have their source in early Czecho-Slovak where neo-acute lengthening or the preservation of original length in the GP of a.p. (b) and a.p. (c) (j)o-stems would have occurred. In a.p. (a) the length should have been preserved in the GP.

123

It is important to notice additionally for o-stem paradigms in CZ, that the CLC LP desinence {-ech} was generalized very early in OCZ to most o-stems and u-stems, however, the forms that end in a velar consonant show the reflex of the original long desinence as in {*-ěch > -ích} (e.g. LP jazycích).7 The source of the length in these desinences is uncertain, but it is found in Czech, Slovak, Polish dialects, Čakavian S/C, Slovene and might very well be connected to the loss and vocalization of jers or to a spread of the neo-acute intonation. Length in Czech that is not PSL acute length or neoacute length is often called compensatory lengthening. In the u-stem CLC LP where the desinence was {*-ъ/ь-xъ} (e.g. *dom-ъ-xъ), the vocalized jer does not become a long vowel (cf. CZ domech). Original short /o/, however, in the o-stem DP {*-o-m-ъ} does become long (e.g. chlapomъ > chlapóm > chlapuom > chlapům) as does the o-stem GP {*-o-v-ъ > -ó-v > -uo-v > ů-v > ů} that has been generalized to nearly all masculine paradigms in CLC. We conclude from these observations that the lengthening of the stem vowel in the substantival paradigms coincided with the loss of the weak jers but in all probability began operating before the vocalization of the strong jers. Yet there is still the possibility that it was the neo-acute intonation that spread to these desinences and that the length developed here as a continuation of the neo-acute intonation. We note, however, that if the neo-acute intonation spread to the LP desinence as well, it failed to lengthen the jer in the morpheme {*-ъ/ь-xъ}.

7

Gebauer (III.I.:181) observed that LP {–ách} in OCZ sometimes is shortened, especially in the Chod dialects and that the DP {–om} from the o-stems is also attested in the Chod dialects Chod DP kravom. It is possible that the brevity in these desinences represent a residue of the expected CZ phonological reflexes and are not shortenings at all.

124 Table 24 – Oxytonic monosyllabic masculine (j)o-stems in CLC with original long root vocalism

Case / Number

Proto-Slavic Form according to Stang Reconstruction (1965:68-69)

Rule for Preservation or Loss of Length in original Long Root Vowel

Expected CLC Form

NS

*sod ъ

1). Length preserved pretonically, 2). Neo-acute retraction

SOUD

soud

soud

GS

*sod à

Length preserved pretonically

SOUDA

souda/u

soudu

DS

*sod ù

Length preserved pretonically

SOUDU

soudu

soudu

AS

*sod ъ

1). Length preserved pretonically, 2). Neo-acute retraction

SOUD

soud

soud

IS

*sod ъmъ8

Length preserved pretonically

SOUDEM

soudem

soudem

LS

*sod ě

Length preserved pretonically

SOUDĚ

soudě/u

soudu

NP

*sod ı

Length preserved pretonically

SOUDI

soudy

soudy

GP

*sod ъ

1). Length preserved pretonically, 2). Neo-acute retraction

SOUD

soudóv

soudů

DP

*sod omъ

Length preserved pretonically

SOUDOM

soudóm

soudům

AP

*sod y

Length preserved pretonically

SOUDY

soudy

soudy

IP

*sod y

Length preserved pretonically

SOUDY

soudy

soudy

LP

*sod ěxъ

Length preserved pretonically

SOUDECH

soudech

soudech

Old Czech Form

Literary Czech Form

The CZ reflexes of the PSL oxytonic (j)o-stem paradigm exhibit the expected phonological reflexes throughout the paradigm. We were unable to find any examples in OCZ of shortenings of original oxytonic root vowels in the trisyllabic forms IS, DP, LP,

8

Although the older o-stem IS desinence is undoubtedly {-omъ}, West Slavic generalized {ъmъ} quite early, and the oldest West Slavic manuscripts already illustrate this fact. The origin of {-ъmъ} is almost certainly the u-stem declension.

125

IP. The natural conclusion to reach is that words of the oxytonic class were not subject to the loss of pretonic length in three syllable forms of the paradigm. Table 25 – Circumflex monosyllabic masculine (j)o-stems in CLC with original long root vocalism

Case / Number

Proto-Slavic form according to Stang Reconstruction (1965:72-77)

Rule for Preservation or Loss of Length in original Long Root Vowel

NS

*bêrg ъ (?)

Circumflex length lost

BŘEH

břeh

břeh

GS

*bêrg a

Circumflex length lost

BŘEHA

břeha/u

břehu

DS

*bêrg u

Circumflex length lost

BŘEHU

břehu

břehu

AS

*bêrg ъ

Circumflex length lost

BŘEH

břeh

břeh

IS

*bêrg ъm ъ (?)

Circumflex length lost

BŘEHEM

břehem

břehem

LS

*bêrg ě (?)

Circumflex length lost

BŘEZĚ

břehu

břehu

NP

*berg i (?)

Circumflex length lost

BŘEZI

břehy

břehy

GP

*ber̃g ъ

1). Length preserved pretonically, 2). Neo-acute retraction

BŘÍH

břehóv

břehů

DP

*berg õm ъ

Desinence 1). Neo-acute retraction, 2). Lengthening of /o/ under neo-acute, 3). /ó > uo > ů/

BŘEHŮM

břehóm

břehům

AP

*bêrg y

Circumflex length lost

BŘEHY

břehy

břehy

IP

*bêrg y (?)

Circumflex length lost

BŘEHY

břehy

břehy

LP

*berg ěx ъ

Desinence Length preserved pretonically

BŘEZÍCH

březích

březích

Expected CLC Form

Old Czech Form

Literary Czech Form

CLC is quite consistent in generalizing brevity in this class of substantives. If Stang (1965:72-77) is correct and the jers were stressed in the DP9 and LP of the mobile paradigm, we have found a possible historical phonological source for the lengthened vowels in the CLC masculine DP desinence {-ům} of hard and soft stems alike, and the

9

Stang is ultimately unsure about whether the stress was on the desinence in the DP, but offers evidence in support of this view.

126

LP desinence {-ích} of soft stem and velar stem masculines. As far as the roots are concerned, nearly all (j)o-stems that we have classified as a.p. (c) are short throughout the paradigm and there are few dialectal exceptions. The few examples of lengthened vowel in the old GP with zero ending of original mobile words (e.g. hód, bóh) must be explained as lengthening of /o/ under the neo-acute in Czech. But the GP in zero is very rare even in Old Czech – the u-stem desinence {-ovъ} dominated the fray very early. Table 26 – Oxytonic monosyllabic masculine (j)o-stems in CLC with original short root vocalism Case / Number

Proto-Slavic Form

Explanation for Reflex of Original Short Root Vowel

Expected CLC Form

Old Czech Form

Literary Czech Form

NS

*kõnj ь

1). Neo-acute retraction, 2). Lengthening of /o/ under the neo-acute

KŮŇ

kůň

kůň

GS

*konj æ

Non-lengthening environment without retraction

KONĚ

koně

koně

DS

*konj u

Non-lengthening environment without retraction

KONI

koni

koni

AS

*konj æ

Non-lengthening environment without retraction

KONĚ

koně

koně

IS

*konj òm ъ

Non-lengthening environment without retraction

KONĚM

koněm

koněm

LS

*konj ı

Non-lengthening environment without retraction

KONI

koni

koni

NP

*konj ı

Non-lengthening environment without retraction

KONI

koně

koně

GP

*konj ь

1). Neo-acute retraction, 2). Lengthening of /o/ under neo-acute

KŮŇ

koní

koní

DP

*konj òm ъ

Non-lengthening environment without retraction

KONÍM

koním

koním

AP

*konj æ

Non-lengthening environment without retraction

KONĚ

koně

koně

IP

*konj òm i

Non-lengthening environment without retraction

KONĚMI

koněmi

koněmi

LP

*konj æx ъ

Non-lengthening environment without retraction

KONÍCH

koních

koních

127

The analogical adoption of the i-stem desinence in the GP of this paradigm has obliterated any trace of the expected lengthening in this form. The expected lengthening in the NS (and AS of inanimate substantives) is not preserved in all lexemes. Various attempts have been made to phonologically explain CLC forms such as dům, půst, růst, etc. We will examine this question below. Another major observation to be made about this paradigm is the striking fact that in contrast to the feminine end-stress paradigm with original short root vowel that was lengthened in all of the grammatical case forms in certain dialects (e.g. kůlna, kůpa, vůsa, kůra), the length in the masculine substantives shows up only in the N(A)S. Acute monosyllabic masculine (j)o-stems Table 27 – Acute monosyllabic masculine (j)o-stems with original long root vocalism PSL

GLOSS

S/C

SLOVENE

CZECH

SLOVAK

RUSSIAN

*xlěbъ

‘bread’

hljeb, hljeba

hléb, hléba hlệb, hlèba ?

chléb, chleba C

chlieb, chleba

xl’eb, xl’eba

*cěpъ

‘tip, point’

cîp, cîpa

cìp, cípa

cíp , cípu B

cep, cepa

cep, cepa

*dymъ

‘smoke’

dım, dıma

dìm, díma

dým, dýmu B

dym, dymu

dym, dyma

*gorxъ

‘peas’

grax, graxa

gràh, gráha

hrách, hrachu C OCZ GS hráchu Chod. GS hráchu

hrach, hrachu

gorox, goroxa

*klinъ

‘lap’

klın, klına

klìn, klína

klín, klínu B

klin, klina

kl‘in, klina

*makъ

‘poppy’

mak, màka

màk, máka

mák, máku B OCZ GS maku

mak, maku

mak, maka

128 PSL

GLOSS

S/C

SLOVENE

CZECH

SLOVAK

RUSSIAN

*morzъ

‘frost’

mraz, mraza

mràz, mráza

mráz, mrazu C Chod GS mrázu

mráz, mrazu

moroz, moroza

*plaktь

‘crying’

plač, plača

plàč, pláča

pláč, pláče B

plač, plaču

plač, plača

*porgъ

‘hearth’

prag, praga

pràg, prága

práh, prahu C

prach, prachu

porog, poroga

*rajь/rojь

‘paradise’

râj, raja

ràj, rája also râj, râja

ráj, ráje B

raj, raja

raj, raja

*skirъ

‘scorpion’

štır, štıra

štîr, štîra

štír, štíra B

štúr, štúra Vas. št’ur

ščur, ščura

*syrъ

‘cheese’

sır, sıra

sìr, síra

sýr, sýra B

syr, syra

syr, syra

*tynъ

‘fence’

tın, tına

tìn, tína

týn, týnu B

týň ‘tynina, stake’

tyn, tyna

*větrъ

‘wind’

vjetar, vjetra

vệter, vệtra

vítr, větru B

vietor, vetra

v’et’er, v’etra

*bičь

‘whip’

bıč, bìča

bìč, bíča

bič, biče A

bič, biča

b’ič, b’iča

*brat(r)ъ

‘brother’

brat, brata

bràt, bráta

bratr, bratra A

brat, brata

brat, brata

*časъ

‘time’

čas, časa

čàs, čása

čas, času A OCZ GP čás

čas, času

čas, časa

*dolnь

‘palm’

dlan, dlana

dlàn, dlànî OSLN dlàn

dlaň, dlaně A Chod (301305) dláň

dlaň, dlane

ladon’ Vas. dial. dolon’

*gadъ

‘snake’

gad, gada

gàd, gáda

had, hada A

had, hada

gad, gada

*gněvъ

‘anger’

gnev, gneva

gnẹ$v, gnẹ@va

hněv A OCZ hniev

hnev

gn’ev, gn‘eva

*jugъ

‘south’

jug, juga

jùg, júga

jih, jihu A

juh, juhu

jug, juga

*krajь

‘edge’

krâj, kraja

kràj, krája

kraj, kraje A Chod (301305) kráj, u kráje, na kráji

kraj, kraja

kraj, kraja

*plugъ

‘plow’

plug, pluga

plûg, plûga also plùg, plúga

pluh, pluhu A plouh / plouhu B

pluh, pluha/u

plug, pluga

129 GLOSS

S/C

*rakъ

‘crayfish’

rak, raka

ràk, ráka

rak, raka A

rak, raka

rak, raka

*zętь

‘daughter’s husband’

zet, zeta

zèt, zẹ́ta

zet’, zet’a A

zat’, zat’a

z’at’

*dědъ

‘grandfather’

djed, djeda

dẹ$d, dẹ@da tudi dèd

děd, děda A

ded

d’ed, d’eda

*svatъ

‘son in law’s father’

svat, svata

svât, svâta

svat, svata A

svat

svat, svata

*volxъ

‘Romanian’

vlax, vlaxa

vlàh, vláha Vas. làh

vlach A

vlach

volox, voloxa

*xolpъ

‘boy’

xlap, GS xlapa (Vas. also GS xlâp)

hlâp, hlâpa

chlap A

chlap

xolop, GS xolopa

PSL

SLOVENE

CZECH

SLOVAK

RUSSIAN

Preserved length in Czech is often the mantra of Slavic accentologists for evidence of PSL acute intonation. Although it is true that preserved length in Czech shows up more often in original acute stems than mobile stems, more than half of the original acute masculine monosyllabic (j)o-stems listed in Table 27 do not retain the expected length in the roots. Variation is characteristic of every language. S/C cîp, cîpa, indicates mobility, and other peculiarities such as S/C mak, màka; pràg, praga; bıč, bìča require language specific internal/dialectal clarification. SLN svât, svâta; hlâp, hlâpa; vệter, vệtra; štîr, štîra throw a little doubt on whether these words were original acutes or not, and SK chlieb, mráz, štúr, týň, vietor, with length transgress the SK rule shortening of acute length. But no Slavic language used for accentological arguments is as unreliable as Czech is. Explaining the numerous shortened acute forms in Czech has been taken up by other scholars. Bulaxovskij (1983:164-165) offers ad hoc explanations for the shortness in bratr (‘fonetičeskij fakt’) děd (from dědík) svat (parallel to starosvat)

130

vlach (shortness in the plural forms) pan (allegro speech) jih (from na jíhu) hněv (deverbal from hněvati). Trávníček (1935:259) attributes shortness in the acute stems to metatony in certain morphological environments: “Tak v gen. pl. vznikla metatonií intonace novocirkumflexová, která se projevila v pčsl. krátkostí, čas atd. Metatonie vznikla bez pochyby také v nom. ak. sg., čas, děd ... atpod. Většinou vznikla při tomto vyrovnávání buď veskrze krátkost nebo veskrze délka.” [So in the GP through metatony the neo-circumflex intonation arose, which shows up in Proto-Czech as brevity, e.g. čas etc. Metatony without a doubt also arose in the NAS čas, děd ... etc. In general, leveling then took place either by the generalization of brevity throughout the paradigm or the generalization of length]. This explanation is unsatisfactory. We have already indicated that we believe the shortening in the GP of (j)ā-stems is better explained as the tendency to differentiate the GP from the rest of the paradigm than as a separate hypothetical metatonic intonation called the neo-circumflex. Furthermore, there is no reason to believe that there was a change of intonation in the NAS of any stems. Finally, the shortening in the root of the GP of (j)o-stems in {-Ø < -ъ/ь} is nowhere a regular feature as far as we can tell from the scant evidence, dialectal or otherwise, and in fact we expect neo-acute (=preservation of length) phonologically in the oxytonic and mobile paradigms for the original GP.

131

In the Chod dialects particularly in this group we find deviations, especially chleb, chleba; hrách, hráchu (301-311, 313-337)10 s mákem (301-321); GS práhu. In Old Czech, we find little variation. Even vítr in OCZ always shows up with length. Oxytonic monosyllabic masculine (j)o-stems Table 28 – Oxytonic monosyllabic masculine (j)o-stems with original long root vocalism PSL

*dělъ

S/C

GLOSS

‘part, share’

dıjel, GS dıjela

SLOVENE

CZECH

SLOVAK

RUSSIAN

Vas. dẹ@ł

díl, GS dílu B

diel

d’el, d’ela

Vas. Čak. dél, dēla *grěxъ Vas. ustem

‘sin’

grıjex, GS grijèxa

grêx

hřích, GS hříchu B

hriech

gr’ex, GS gr’exa

*xlěvъ

‘stable’

hlêv, LS hlévu Vas. xlıjev

hlẹ@v, GS

chlév, GS chléva B

chliev

xlev, GS xleva

hlẹ@va

*xrustъ

‘brown beetle that eats vegetation’

hrûšt, GS hrúšta ‘type of cherry’

hrûst, GS hrûsta

chroust11 B

chrúst

xrust, GS xrusta

*klobъ

‘joint’

Vas. klupko

klộb ? Vas. klộbko

kloub, GS kloubu B

kĺb

klub, GS kluba

*ključь

‘key’

kljûč, kljúča

kljúč

klíč B

kl’úč

kl’uč, kl’uča

*kotъ

‘corner’

kût, GS kúta

kọ@t, -a

kout, GS koutu B

kút

kut, GS kuta

*križь

‘cross’

krîž, kríža

kríž

kříž B

kríž

kr’iž, kr’iža

*pěstъ

‘fist’

Ø

pẹ@sta

píst (m.) B Vas. písta (f.) pěst

piest (f.)

p’est, GS p’esta

pệst, -î (f.)

10

Refer to Map 21 for the dialectal distribution of length in the GS hrachu/hráchu.

11

OP chrąst.

132 PSL

S/C

GLOSS

SLOVENE

CZECH

SLOVAK

RUSSIAN

*plaščь

‘overcoat’

plâšč, plášča

plášč

plášť B

plášť

plašč, plašča

*prodъ

‘current’

prûd, GS prúda

prọ@d, -a Vas.

proud, GS proudu B

prúd

prud, GS pruda

*stъlpъ

‘column’

stûp, GS stúpa

stólp, -a

sloup, B

stĺp

stolp, GS stolpa

*směxъ

‘laughter’

smêh, GS smêha / sméha

smệh, -a

smích, B

smieh

sm’ex, GS sm’exa

*sodъ

‘court’

sûd, GS súda

sód, -a

soud, B

súd

sud,

šîp, GS šípa Vas. šîp

šîp, šípa

štît, GS štíta

*šipъ *ščitъ

‘arrow’ ‘shield’

suda šíp, GS šípu B

šíp

ščìt, GS ščíta

štít B

štít

šip, šipa ščit, ščita

*virъ

‘eddy,’ ‘whirlpool’

vîr, GS víra

vîr, víra

vír B

vir

vir, vira

*vusy12

‘beard, mustache’

Ø

vộs, vộsi NP

vousy NP B

fúzy

us, usa (dial. usa) NP usy

*plastь

‘layer, mound (of hay)’

plâst, GS plásta ‘kupa’

plâst, plâstî

plást, GS plástu B

plást

plast, plasta / plasta

The Czech reflex on a long vowel in the oxytonic paradigm is extremely consistent. All of the oxytonic stems that we have examined have generalized the long vowel throughout the paradigm. Even oxytonic words, such as hřích, that are inconsistent in its reflex pan-slavonically are at least consistent in CLC (and Slovak for the most part). The situation is somewhat different, however, for the oxytonic roots with original short vowel. In a number of the words in Table 29, there is a good deal of

12

Polish wąs indicates end stress.

133

evidence for lengthening of the original short vowel in CLC. There is often further dialectal evidence for lengthening of original short vowels in oxytonic stems. 13

Table 29 – Oxytonic monosyllabic masculine (j)o-stems with original short root vocalism PSL

S/C

GLOSS

SLOVENE

CZECH

SLOVAK

RUSSIAN

*dъždь

‘rain’

dažd, dàžda

dèž, dežjá

dešt’/déšt’, deště14 C/B

dážd, dažda

dožd’ dožd’a

*dvorъ B

‘yard’

dvôr, dvóra

dvòr, dvóra

dvůr, dvoru C

dvor, dvor / dvôr

dvor, dvora

*kolъ B

‘stake’

Kaj. kôl, kola

kòł, kóła

kůl, kolu / kůlu C

kôl, kola

kol, kola

*konjь

‘horse’

konj, kònja

kònj, kónja

kůň, koně C

kôň, koňa

kon’ kon’a

*nožь

‘knife’

nôž, nóža

nòž, nóža

nůž, nože C

nôž, noža

nož, noža

*postъ B (/D)

‘fast’

pôst, posta

pòst, pósta

půst, postu C

pôst, pôstu

post, posta

*stolъ B

‘table’

stô, stòla

stòl, stóla

stůl, stolu C

stôl, stola

stol, stola

*volъu B

‘ox’

vô, vòla

vòl, vóla

vůl, vole C

vôl, vola

vol, vola

*bebrъu B

‘beaver’

dabar, dabra

Plet. bọ@ber,

bobr, bobra A

bobor, bobra

bobr, bobra

*bobъ B (/D)

‘bean’

bob, bòba

bòb, bóba

bob, bobu Mor. bób15 A

bôb, bôbu

bob, boba

*borъu B/D

‘pine tree’

bôr, bora

bòr, bóra tudi bộr, -a

bor, boru

bor, bora

Mor. bór A

bor/bôr, boru/bôru

*gvozdъ B

‘forest, nail’

gòzd, gózda17

hvozd, hvozdu A

hvozd, hvozdu

gvozd’, gvozd’a

bọ@bra

gvozd, gvòzda

16

13

Original u-stems are included here marked with a superscript /u/.

14

See Map 17 for Czech dialectal distribution of lengthened /e/ (< *ъ).

15

Gebauer (I.III:609).

16

Gebauer (I.III:609).

17

Cf. Slvn. gvộzd, gvộzda ‘nail, wedge.’

134 PSL

GLOSS

u

S/C

SLOVENE

CZECH

SLOVAK

RUSSIAN

*kosъ

‘Turdus merula’

kôs, pl. kosovi

kộs, kộsa

kos, kosa dial. kůs, kosa18 A

Ø

Ukr. arch. kis, kosa

*košь

‘horse’

koš, kòša

kòš, kóša

koš/kůš, koše19 A

kôš, koša

koš, koša

*rojь

‘swarm’

rôj, roja

ròj, rója

roj, roje dial. růj A

roj, roja dial. rôj

roj, roja dial. rôj, roja

dial. rej, rèja *skotъ

‘cattle’

skot, ?

skòt, skóta

skot, skota A OCZ skót

Ø

skot, skota

*slojь20

‘layer’

slôj, sloja

slòj, slója

sloj, sloje A sluj < slůj21

sloj, sloja

sloj, sloja22

*slonъ B (/D?)

‘elephant‘

slon, slòna

slòn, slóna

slon, slona A

slon, slona

slon, slona

*stogъ D

‘haystack’

stôg, stòga

stòg, stóga Plet. stộg, stộga

stoh, stohu23 A OCZ stóh

stoh, stohu

stog, stoga

*stropъ B/D

‘ceiling’

strôp, stròpa

stròp, strópa

strop, stropu A

strôp, stropa

strop, stropa

18

See Map 19 for Czech dialectal distribution of lengthened /o/.

19

See Map 22 for Czech dialectal distribution of lengthened /o/ in this word.

20

Illič-Svityč (p. 134) attributes the S/C reflex and transference to the mobile class as secondary and a result of lengthening of /o/ before /j/. In Czech, Gebauer believed that all /o/’s before /j/ were lengthened (Gebauer, 592-593). He quotes OCz forms from BiblA. Ex 14, 13 imper. bój sě, nč. dial. bůj sa, OCz boju sě (ty) sě nebuoy Levšt. 150b, bůj se Suš. 89, matcy se nic nebůj. To avoid lexical collision or by analogy to the forms in the paradigm that did not end in ь, the phonologically correct *bój (m.) was levelled to the forms of the oblique cases where there would have been no lengthening. In addition, there are many attestations of roj with length (Map 18). In the Silesian dialects the reflex of lengthened /o/, namely /u/ in roj is widely attested. Sporadic relics of length in roj are also attested in several Central Moravian dialects and Southern Bohemian dialects. The word voj ‘army,’ (cf. vojsko, vojín, etc.) is not attested with length. This fact however is due to it’s infrequent usage outside of compounds and by no means nullifies what was obviously a general Czech development which lengthened /o/ before /j/ at the end of the phonological word. 21

Machek (1971:555) “Sem náleží i č. sluj < *slůj původně téhož významu jako sloj, ale v bratrské bibli s významem dutina, díra v zemi, jeskyně, jenž vešel i do spisovného jazyka.” 22

This word was probably originally a mobile stem with variation to the oxytonic paradigm.

23

See Map 20 for Czech dialectal distribution of lengthened /o/ in this word.

135 PSL

GLOSS

S/C

SLOVENE

CZECH

SLOVAK

RUSSIAN

*xvostъ B

‘tail’

xost, xòsta

hvộst, GS?

chvost, chvostu A

chvost, chvosta

xvost, xvosta

*prьstъ

‘finger’

prst, prsta

prst, pŕsta

prst, prstu X

prst, prsta

p’erst, p’ersta

The Czech reflexes of the oxytonic paradigm a.p. (b) with original short root vocalism are discussed together with the mobile paradigm below as the two are interrelated. Mobile Monosyllabic Masc. (j)o-stems Table 30 – Mobile monosyllabic masculine (j)o-stems with original long root vocalism PSL

GLOSS

S/C

SLOVENE

CZECH

SLOVAK

RUSSIAN

*běgъ

‘run’

bêg, bêga Vas. bıjeg

bệg

běh A

beh

b’eg,

*blěskъ

‘lightening’

blêsak, blêska

blêsk

blesk A

blesk

bl’esk (3a)

*borvъ

‘small cattle, castrated pig’

brâv, brâva

brâv

brav A

brav

borov

*bergъ

‘shore’

brıjeg, brıjega

brệg, brệga

břeh A

breh

b’er’eg, b’er’egu

*brusъ

‘blade’

brûs, brûsa

brûs

brus A

brús

brus (1a)

*krugъ

‘circle’

krûg, krûga

krộg

kruh A

kruh

krug, kruga

*květъ

‘blossom’

cvêt, cvêta Vas. cvıjet

cvệt

květ A

kvet

cv’et, cv’eta

*darъ

‘gift’

dâr, dâra

dâr

OCZ dar A

dar

dar,dara

*dъlgъ

‘debt’

dûg, dûga

dộłg

dluh A

dlh

dolg, dolga

*duxъ

‘spirit’

dûx, duxa

dûh

duch A

duch

dux, duxa

*goldъ

‘hunger’

glâd, glâda

glâd, glâda

hlad A

hlad

golod

*golsъ

‘voice’

glâs, glâsa

glâs

hlas A

hlas

golos (1c)

borova

136 PSL

S/C

GLOSS

SLOVENE

CZECH

SLOVAK

RUSSIAN

*gordъ

‘town’

grâd, grâda

grâd

hrad A

hrad

gorod (1c)

*ězъ

‘weir’

jâz, jâza

jệz

jez A

jaz / jez

jaz, jaza Vas. OR ězъ

*listъ

‘leaf’

lîst, lîsta

lîst

list A

list

l’ist, l’ista

*měxъ

‘goatskin thermos’

mêh, meha

mệh, mệha

měch A

mech

m’ex, m’exa NP m’exa ‘bellows’

mıjex Vas.

*koipъ

‘thrasher’

cêp, cêpa

c’êp

cep A

cep

c’ep, c’epa

*volsъ

‘hair’

vlâs,

lâs

vlas A

vlas

volos (1c)

*vorgъ

‘murderer’

vrâg

vrâg

vrah A

vrah

voroh (UK)

As expected, in CLC and dialectally, all the original mobile paradigms that we examined with original long root vocalism have shortened vowel throughout the paradigm. Original mobile words with short root vocalism, however, are subject to lengthened vowel in the NAS. Since the ictus in the NAS for the mobile paradigm was undoubtedly on the root, the environment cannot have been identical to the oxytonic paradigm. The lengthening in the NAS only occurs on /o/, which is rather suspicious. Original long vowels do not re-lengthen, so lengthened /o/’s must be the result of either compensatory lengthening in front of a lost jer, or by analogy to the oxytonic paradigm.

137 24

Table 31 – Mobile monosyllabic masculine (j)o/u-stems with original short root vocalism PSL

GLOSS

S/C

SLOVENE

CZECH

SLOVAK

RUSSIAN

*bogъ C

‘God’

bôg, boga

bóg

bůh, boha C

boh, boha

bog, boga

*dolъu D

‘ditch’

dô, dola

dộł, dộła

důl, dolu C

dol, dolu

dol, dola

*domъu C

‘house’

dôm, doma

dộm, dộma

dům, domu C

dom, domu

dom, doma

*gnojь

‘manure’

gnôj, gnoja

gnôj, gnója

hnůj, hnoje C

hnoj, hnoja

gnoj, gnoja

*lojь

‘lard’

lôj, loja

lôj, lója

lůj, loje C Chod. loj 301-306

loj, loja

loj, loja

*orstь

‘growth’

râst, râsta

râst

růst, rostu C

vzrast, vzrasta

rost, rosta

*vozъ C/D

‘wagon’

vôz, voza

vộz, vozâ Plet. vộz, vộza,

vůz, vozu C

voz, voza

voz, voza

*bodъ D

‘point’

bôd, boda

Plet. bòd, bóda

bod, bodu A

bod, bodu

Ø

*bokъu D

‘side’

bôk, bôka

Plet. bộk, bộka

bok,boku A

bok, boku

bok, boka

*brodъ D/C

‘boat, ford, ferry’

brôd, broda

brộd, brộda

brod, brodu A

brod, brodu

brod

*gonъ B/D

‘hunt’

gon25

gòn, góna

hon, honu A

hon, honu

gon, gona

*grobъ B/D

‘grave’

grob, gròba

gròb, gróba

hrob, hrobu A

hrob, hrobu

grob, groba26

*gromъ D/C

‘thunder’

grôm, gròma

grộm, grộma

hrom, hromu A

hrom, hromu

grom, groma

*krovъ B/D

‘roof’

krôv, krova

kròv, króva

krov, krovu A

krov, krova

krov, krova

*ledъu C

‘ice’

lêd, leda

lệd, lệda

led, ledu A

l’ad, l’adu

lёd, l’da

24

Original u-stems are included here marked with a superscript /u/.

25

Dybo (1991:194), Illič-Svityč (1963) classifies as oxytone.

26

See Dybo (1991:195) for explanation of ictus pattern B/D.

138 PSL

GLOSS

S/C

SLOVENE

CZECH

SLOVAK

RUSSIAN

*lovъ

‘the hunt’

lôv, lova

lòv, lóva

lov, lovu A

lov, lovu

lov, lova

*mostъ B/D

‘bridge’

môst, mosta

mộst, mộsta

most, mostu A

most, mosta

most, mosta/

*nosъ D

‘nose’

nôs, nosa

nộs, nộsa

nos, nosu A

nos, nosa

nos, nosa

*plotъ

‘fence’

plôt, plota

plột, plộta

plot, plotu A

plot, plota

plot, plota

*potъ D

‘sweat’

pôt, pota

pột, pộta

pot, potu A

pot, potu

pot, pota

*rodъu C

‘kin’

rôd, roda

rộd, rộda

rod, rodu A

rod, rodu

rod, roda

*rohъ D

‘horn’

rôg, roga

rộg, rộga

roh, rohu A OCZ róh, rohu

roh, roha

rog, roga

*voskъ D

‘wax’

vosak, voska

vộsk, vộska

vosk, vosku A

vosk, vosku

vosk, voska

*xodъ

‘walk’

hôd, hoda

hòd, hóda

chod, chodu A

chod, chodu

xod, xoda

mosta

In Literary Czech there are a number of monosyllabic substantives which are characterized by an alternation between [long] /ů/ in the NAS and [short] /o/ in the other case forms. In rare cases, long /e:/ alternates with short /e/ in the same morphological categories. This alternation is a result of the phonological lengthening of an original short vowel (o, e, ь, ъ) of the root in the NAS forms. These substantives are typically masc. o-stems (kůl, stůl, půst, bůh), jo-stems (kůň, déšt’, nůž, hnůj, lůj), or u-stems (důl, vůl), however several feminine i-stems may be cited as exhibiting a similar lengthening pattern (sůl, hůl). Attempts to explain this length as either compensatory lengthening as the result of the loss of a following weak jer, lengthening as the result of ictus retraction and the rise of the neo-acute (Stang’s law), phonologically or phonetically conditioned lengthening, morphophonemic lengthening, or a combination of some or all of the above

139

have been generally incomplete because of the complexity of the problem and the use of incomplete dialectal data or the unsystematic use of data from Old Czech. We will try to sort out the facts for Czech to determine if possible which factors and conditions contribute to the lengthening of the original short vowels, which attestations of the lengthened short vowels can be considered phonological, which attestations of the lengthened short vowels should be thought of as morphological extensions, and the reasons for such conclusions. Czech data both from the literary language as well as the Czech dialects will be utilized in an effort to bring as much relevant information as possible to bear on this complicated problem. Slovak, and other Slavic languages will be cited to support the theory that what on the surface may seem to be a single phenomenon is in fact several quite distinct but interrelated phenomena. The Czech forms which regularly exhibit lengthening in the NAS are: oxytonic a.p. (b): dvůr, kůl, půst, stůl, déšt’, kůň, nůž, vůl; mobile c-type: bůh, růst, vůz, lůj, hnůj, důl, dům. We also include sůl and hůl in this discussion since the alternation between /o/ and /ů/ (NAS sůl, hůl oblique sol- , hol-) in these words is almost certainly related. The somewhat regular lengthenings found in the Czech dialects for stoh ‘haystack’ (Map 20) kos ‘Turdus Merula=blackbird’ (Map 19) koš (Map 22) ‘basket’ illustrate what are very good examples of length that has not been preserved in the Czech literary language but whose dialectal distribution forces us to attribute the length to something more than idiosyncratic variation. It is important to mention that whereas mobile substantives with original short /o/ do exhibit lengthening, mobile substantives with original long vowel do not exhibit preservation of length. The pattern of dialectal retention of lengthened /o/

140

shows clearly that the loss of length in these forms of the original oxytonic paradigm was innovative in the Central Bohemian dialects upon which the literary language is based. It is natural to find wide dialectal variation of a feature that alternates within a paradigm. Historically speaking, since the only forms in the oxytonic paradigm for masculine monosyllabic inanimate stems (o-stems, jo-stems, and u-stems) that would have been affected by the retraction of ictus from a jer that lost the ability to carry ictus were the NAS (and the GP for o-, jo- stems), the paradigmatic pressure for analogical leveling in favor of the original short vowel would have been considerable. One would indeed be very surprised that the alternation had been preserved at all if it weren’t for the fact that the source of the lengthened grade is precisely the NAS, in other words, the morphological categories that most often serve as the source for other targets of the analogical leveling processes. For animate o-stem and jo-stem substantives (such as *kos, *skot, *slon, *konj) the only form susceptible to a phonological compensatory lengthening of /o/ would have been the NAS and perhaps the GP if the GP {–ъ} was still extant at the time that the pattern was established and if the (j)o-stems adopted the u-stem GP after the phonemicization of length in Czech. For the u-stems (e.g. důl, vůl, bobr) only the N(A)S would have provided the correct phonological environment for compensatory lengthening in Czech and Slovak. It’s no wonder then, that if the compensatory lengthening of /o/ in animate *kosъ ‘Turdus merula,’ and *slonъ ‘elephant’ took place, that the lengthened vowel was leveled out by analogy to the rest of the paradigm without compensatory lengthening or neo-acute lengthening and ictus retraction. Map 19 shows that the dialectal distribution of neo-acute length for *kosъ is

141

restricted to what many consider to be the most archaic dialects of the Chod region in the territory of the Southwestern Bohemian dialects. The word *skotъ is attested with length in OCZ, but we were unable to collect dialectal data for *skotъ from the survey done by the Czech Academy of Sciences. The word *bobrъ probably does not even belong here because /r/ was no doubt syllabic and carried the ictus which would have blocked the effects of compensatory lengthening. The word *slonъ, then, is the only end-stressed animate form that shows no residue of the lengthened /o/ under neo-acute lengthening conditions in this class of words. Perhaps this word along with *hvozdъ and *chvostъ were not end-stressed in dialectal Proto-Slavic in the Czech and Slovak territories and were therefore not subject to neo-acute lengthening. The dialectal distribution of length in déšt’ (Map 17) illustrates by the concentration in the Central Moravian dialects and the lack of attestations in the more archaic Chod dialects that the lengthening in this case is a separate phenomenon from the lengthenings represented by stoh and koš. The remaining examples concern only lengthened /o/. Other sporadic attestations of lengthened /o/’s in Czech include: Výklad kruml 132b (sir 38, 29) vapor znuoij horkoft, from the first half of the 15th century; CLC trůn27 ‘throne’ SK trôn < Lat. thronus < Gr. θρόνος.

27

The word tron was no doubt borrowed without length from Latin but acquired length to mark it in Old Czech as a foreign word, much in the same way that words in contemporary standard Czech are often marked by long /ó/ to distinguish them from native words as loan words or words of foreign origin that have not yet been assimilated into the literary language, cf. móda, dóm, flóra, bonbón, neón, vagón, stadión, milión, balkón, medailón, pavilón, Bablón, Arón, tón, Platón, maratón, zón.

142

One difficulty with interpreting the operation of compensatory lengthening and neo-acute lengthening has been the interest in equating the phonological environment with the morphological environment. The fact that Slovak shows a regular reflex of length in the GP (even showing the expected lengthening in liquid vowels!) has led to the belief that in Slovak, lengthening under the neo-acute in general is absolutely regular. A more likely explanation, however, is that lengthened vowels in the Slovak GP has become a morphological marker with generalization of lengthening to many paradigms where the lengthening effect would not have arisen phonologically. The source of this length may very well have been retraction off a stressed jer in the oxytonic accentual paradigm a.p. (b) and the GP of u-stems and ā-stems of a.p. (c) proceeded by a neo-acute lengthening in Czecho-Slovak, but that is not to say that GP length in Slovak is equivalent to neo-acute length since the effects have spread morphologically so extensively. But why are some /o/’s lengthened under the retraction from a stressed jer but not others? As far as the words under investigation in this study are concerned, Slovak agrees with Czech for oxytonic a.p. (b) kôl, pôst, stôl, kôn, nôž, vôl. There is also a strong possibility that Czech dial. kůš/kuš, Slovak kôš reflect the same lengthening which must be considered neo-acute lengthening. In the Chod dialects we find further evidence (Hruška, 1907:47 “kůš, ale ost. pády pravidelně: koše, atd.”) that this word retained the short vowel in the other grammatical forms as expected for words that had neo-acute lengthened short vowel in the NS. But Slovak has no length for the words where /o/ was lengthened in the NAS for Czech of the mobile class. In fact, Slovak shows no

143

lengthening of any short vowel in words of the mobile accentual pattern. Two additional oxytonic Slovak substantives, bôb, strôp, show neo-acute lengthening. Czech bób with long /o/ must be considered archaic. As far as the CZ dialects are concerned, we were unable to find data to confirm whether bób shows up with length anywhere today. For the masculine monosyllabic stems, however, the Czech data indicate that whereas Slovak may have been the center of a center/periphery phonological phenomenon, Czech has extended the effects of neo-acute lengthening beyond it’s phonological source which is almost certainly the lengthening of /o/ in Late West Slavic or Czecho-Slovak in a special morphophonemic environment: /o/ > /ó/ /__C# when the ictus was retracted off of a jer as a result of the advent of the neo-acute intonation on the syllable one place to the left. It may appear that there is a phonotactic constraint on the lengthening of /o/ under the neoacute in Czech as espoused by Černý, Vondrák, Trávníček et al. since oxytonic forms such as hvozd, slon, strop, chvost in CLC lack the short /o/ long /o/ alternation. It is possible, however, that the original neo-acute lengthening occurred everywhere in Czech but was phonetically influenced by the intervening consonant before length became phonemic and therefore the alternation of long /o/ and short /o/ was only preserved in paradigms ending in consonants where the phonetic lengthening was reinforced. The environment before sonorants is particularly supportive of the lengthening of /o/ under the neo-acute, especially before /l/. There cannot be a general lengthening of /o/ under the neo-acute or by compensatory lengthening unless one assumes a kind of lexical selectiveness in the neoacute retraction, but whenever there was a morphological based accent on an /o/, it tended

144

to lengthen it. If there were two separate retractions from auslaut they could have retracted at different times –one retraction associated with the neo-acute as in the jāstems with lengthened /o/ such as vůle, důše, růže, etc. and the other retraction off of stressed jers which also no doubt caused phonetic lengthening of /o/ in CZ as in důl. It seems that indeed most /o/’s were lengthened when the conditions were met for compensatory lengthening and in the grammatical categories that the neo-acute obviously became morphologized in such as the GP in Slovak. The spread of the alternation created by compensatory lengthening to a.p. (c) mobile monosyllabic substantives in Czech (e.g. dům, bůh, sůl) was subject to the same phonotactic conditions as the preservation of the original alternation. The alternation tended to spread to lexemes whose stems ended in a sonorant or voiced fricative bůh, vůz, důl, dům, sůl. That the lengthening is not phonological in the mobile class is proven by the lexemes that end in the requisite consonant, /l/ for instance, that show no hint of length dialectal or historical such as mol ‘moth,’ bol ‘pain.’ The word růst ‘growth’ is the only example of the alternation where the environment is not before a sonorant or voiced fricative, but could be explained as an analogical change based on půst or is perhaps better explained by the fact that it is a deverbal derivation of růsti which has the same lengthening as nésti under the neo-acute intonation. The extension of the alternation of long /o/ and short /o/ to the environment /___ j, (e.g. hnůj, lůj, dial. růj) however, appears to be a separate phenomenon whose source we find in the pronominal paradigm (můj, moje). We see from the dialectal distribution

145

of lengthened /o/ in /roj/ (Map 18) that the lengthened vowel is not phonological everywhere before /j/ in the mobile class, but a morpho-phonemic analogical change. We have tried to show that what may appear to be explainable by a single phonological rule is in fact three separate linguistic phenomena. The first and most regular is the lengthening of short vowels – primarily short /o/ – in Czecho-Slovak under the retraction of ictus off a weakened jer and the rise of the neo-acute intonation. In the case of masculine monosyllabic substantives, this lengthening is sometimes referred to as compensatory lengthening, but this term is something of a misnomer that implies a phonological lengthening of a short vowel in the syllable in front of a weak jer when in fact the lengthening is attributable not so much to the loss of the weak jer as to the neoacute intonation that developed in the NAS in specific cases of certain PSL paradigms [NAS of a.p. (b) monosyllabic masc. (j)o-stems with short root vowel, GP of a.p. (a), a.p. (b), & a.p. (c) monosyllabic masc. (j)o-stems, etc.]. The second phenomenon is the reinforcement of the phonetic alternation based on paradigmatic influences and phonotactic constraints leading to the possible retention of the alternation within a given paradigm. Finally, there is the analogical spread of the resulting morpho-phonemic alternation to the mobile class substantives that end in consonants which reinforced both phonetically and morphologically the lengthening of /o/.

Chapter 7

NEUTER DISYLLABIC SUBSTANTIVES

Expected reflexes of root vowels If we look at the three LPSL accentual paradigms: acute [fixed root stress=a.p. (a)], oxytonic [fixed stress on first syllable of the desinence=a.p. (b)], circumflex [(mobile=a.p.(c)], as we have done for the feminine ā-stem paradigms in Chapter 5 and the masculine (j)o-stem paradigms in Chapter 6, we can derive the expected paradigms for the neuter (j)o-stem disyllabic substantives in Czech in a similar fashion to assess whether or not there appears to be a correlation between the actual CLC paradigms and the reflexes that we would expect when taking into consideration the phonologically and morphologically conditioned rules for the preservation of length, the shortening of original length, or the lengthening of original short vowels in Czech. Table 32 – Acute disyllabic neuter (j)o-stems in CLC with original long root vocalism Case / Number

Proto-Slavic Form according to Stang (1965:82)

NS

*lět o

Acute length preserved

LÉTO

léto

léto

GS

*lět a

Acute length preserved

LÉTA

léta

léta

DS

*lět u

Acute length preserved

LÉTU

létu

létu

AS

*lět ъ

Acute length preserved

LÉTO

léto

léto

IS

*lět ъmъ1

Acute length lost in word greater than two syllables long

LETEM

létem

létem

LS

*lět ě

Acute length preserved

LÉTĚ

létě

létě

1

Rule for Preservation or Loss of Length in original Long Root Vowel

Expected CLC Form

Old Czech Form

Literary Czech Form

See Chapter 6, fnt. 8 above for an interpretation of the source and history of this desinence. 146

147

Case / Number

Proto-Slavic Form according to Stang (1965:82)

Rule for Preservation or Loss of Length in original Long Root Vowel

Expected CLC Form

Old Czech Form

Literary Czech Form

NP

*lět a

Acute length preserved

LÉTA

léta

léta

GP2

*lět ъ

1). Acute length preserved, 2). Morpho-phonemic loss of length in GP

LET

let

let

DP

*lět omъ

Acute length lost in word greater than two syllables long

LETOM

létóm / letóm

AP

*lět a

Acute length preserved

LÉTA

léta

létům / letům léta

IP

*lět y

Acute length preserved

LÉTY

lety

lety

LP

*lět ěxъ

Acute length lost in word greater than two syllables long

LETECH

létech / letech

létech / letech

Just as we discovered with the masculine acute paradigm and its reflexes in CLC, there are no neuter paradigms in CLC that match the expected result of the PSL acute (j)o-stem paradigm for every grammatical case. We do, however, find evidence for an older pattern that reflects the phonological and morphophonemic reflexes that we would expect from the original acute paradigm. The word that is closest to what we would expect is precisely one of the words most likely to preserve an older pattern, léto ‘year, summer’ because of the frequency of its usage and core vocabulary nature of the meaning. Although the paradigm in CLC for this word is entirely unique, we have cited it here in order to illustrate what we believe must be thought of as the historical regularity of the prosodic sound changes that acted upon the Proto-Czech substantival system.

2

For the history of this case form, see Gebauer (III.I:54-58).

148 Table 33 – Oxytonic disyllabic neuter (j)o-stems in CLC with original long root vocalism Case / Number

Proto-Slavic Form according to Stang (1965:83)

Rule for Preservation or Loss of Length in Original Long Root Vowel

Expected CLC Form

NS

*vin o

Length preserved pretonically

VÍNO

víno

víno

GS

*vin à

Length preserved pretonically

VÍNA

vína

vína

DS

*vin ù

Length preserved pretonically

VÍNU

vínu

vínu

AS

*vin o

Length preserved pretonically

VÍNO

víno

víno

IS

*vin ъmъ

Length preserved pretonically

VÍNEM

vínem

vínem

LS

*vin ě

Length preserved pretonically

VÍNĚ

víně/u

víně

NP

*vin à

1). Length preserved pretonically, 2). Analogical generalization of neo-acute (from LP), 3). Neo-acute retraction

VÍNA

vína

vína

GP

*vin ъ

1). Length preserved pretonically, 2). Neo-acute intonation & preservation of length, 3). Morpho-phonemic loss of length in GP

VIN

vin/vín

vín

DP

*vin omъ

1). Length preserved pretonically, 2). Analogical generalization of neo-acute (from LP), 3). Neo-acute retraction

VÍNOM

vínóm

vínům

AP

*vin à

1). Length preserved pretonically, 2). Analogical generalization of neo-acute (from LP), 3). Neo-acute retraction

VÍNA

vína

vína

IP

*vin y

1). Length preserved pretonically, 2). Analogical generalization of neo-acute (from LP), 3). Neo-acute retraction

VÍNY

víny

víny

LP

*vı‚n ěxъ

1). Length preserved pretonically 2). Neo-acute retraction

VÍNECH

vínech

vínech

Old Czech Form

Literary Czech Form

149

Stang (1965:82) believes that the stress in the LP was retracted producing [sic] the neo-acute intonation on the preceding syllable and that the generalization of this accentuation to the other grammatical forms of the plural was a development that was made possible in an effort to differentiate the oxytonic paradigm from the mobile paradigm that had the opposite pattern – that is, root stress in the singular and end stress in the plural. Even though OCZ must be considered mute on the question pertaining to the origin of initial stress in the plural grammatical cases of the oxytonic paradigm, the CZ reflexes of the PSL oxytonic neuter (j)o-stem paradigm exhibit the expected phonological reflexes throughout the paradigm under either scenario. The neo-acute intonation would have resulted in preserved length on the root vowel in the plural, and pretonic length would also have been preserved as length, but we have shown in Chapters 5 and 6 that short root vowels (particularly /o/) under the neo-acute intonation become long vowels as a rule. We would expect, therefore, more evidence for lengthened /o/’s in words from this paradigm if the neo-acute truly spread throughout the plural as Stang suggests. Table 34 – Circumflex monosyllabic neuter (j)o-stems in CLC with original long root vocalism Case / Number

Proto-Slavic form according to Stang (1965:84)

Rule for Preservation or Loss of Length in original Long Root Vowel

Expected CLC Form

Old Czech Form

NS

*zolto

Circumflex length lost

ZLATO

zlato

zlato

GS

*zolta

Circumflex length lost

ZLATA

zlata

zlata

DS

*zoltu

Circumflex length lost

ZLATU

zlatu

zlatu

AS

*zolto

Circumflex length lost

ZLATO

zlato

zlato

IS

*zoltъmъ

Circumflex length lost

ZLATEM

zlatem

zlatem

LS

*zoltě

Circumflex length lost

ZLATĚ

zlatě

zlatě

Literary Czech Form

150

Case / Number

Proto-Slavic form according to Stang (1965:84)

Rule for Preservation or Loss of Length in original Long Root Vowel

Expected CLC Form

Old Czech Form

Literary Czech Form

NP

*zolta

Length preserved pretonically

ZLÁTA

zlata

zlata

GP

*zoltъ

1). Length preserved pretonically, 2). Neo-acute retraction

ZLÁT

zlat

zlat

DP

*zoltomъ

Root 1). Expect shortening Desinence 1). Neo-acute retraction, 2). Lengthening of /o/ under neo-acute, 3). /ó > uo > ů/

ZLATŮM

zlatům

zlatům

AP

*zolta

Length preserved pretonically

ZLATA

zlata

zlata

IP

*zolty

Length preserved pretonically

ZLATY

zlaty

zlaty

LP

*zoltěchъ

Root 1). Expect shortening Desinence 1). Length preserved

ZLATÍCH

zlatech

zlatech

If the disyllabic neuter paradigm with mobile stress pattern was characterized by a stress shift to the final syllable of the desinence in the plural as Stang (1965:84) implies, and there was ever preservation of pretonic length in Czech, the effects have been all but completely lost. Nedvědová (unpublished ms., page 64) cites several OCZ examples of ancient mobile stems with length throughout the paradigm as doublets that she attributes to analogical leveling “břícho, tielo, máso, etc.” It is difficult to know whether these spurious examples represent a pattern based on phonological reflexes of the NAP forms with analogical leveling to the singular or are just random dialectal forms because the sources are not indicated. Gebauer does not mention them at all, to our knowledge, nor

151

are there entries in the OCZ dictionary project card catalogue or the published fascicles of the StČS that indicate any pattern of length in these words. Neuter Disyllabic j(o)-stem paradigms in Contempory Literary Czech The disyllabic neuter (j)o-stem paradigms in CLC exhibit a mix of analogical and phonological levelings, dialectal influences, and morphological readjustment just as the masculine and feminine substantival subsystems that we examined earlier do. Many of the same general historical tendencies are supported by the synchronic and diachronic data for this class of word: the shortening of circumflex length in Czech and Slovak; the preservation of pretonic, acute, and neo-acute length; the generalization throughout the paradigms of either brevity or length; evidence for the morphophonemic substitution of the contrastive quantity in the root vowel of the GP to differentiate the GP from the other forms of the paradigm; widespread lexical shortening of original acute length that is preserved as length along the periphery of the Czech dialectal landscape – particularly in the Chod villages of the Southwestern Bohemian dialect region and the northern dialects of the Northeastern Bohemian dialect region (see Map 28).

152 3

Table 35 – Neuter disyllabic (j)o-stem Paradigms in CLC

Case / Number

A (all short)

břicho břicha břichu břicho břichem břiše břicha břich břichům břicha břichy břiších

NS GS DS AS IS LS NP GP DP AP IP LP

B (all long)

bláto bláta blátu bláto blátem blátě bláta blát blátům bláta bláty blátech

C (long NAS, else short)

N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A

D (short GP, else long)

dílo díla dílu dílo dílem dílě díla děl dílům díla díly dílech

E (short GP, DP, IP, LP, else long) léto léta létu léto létem létě léta let letům léta lety letech

F (short IS, GP, DP, IP, LP, else long) N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A

Almost all neuter (j)o-stems have generalized a short vowel (Type A above) or a long vowel (Type B above) throughout the paradigm. Several minor groups of words deviate from the normalized patterns: Type D as in dílo ‘piece of work’ with GP děl along with several other words is a somewhat expected reflex pattern of original acute stems (although there is not a strong correlation between the two). Additionally, although we will not endeavor to reconstruct the accentual paradigms for consonantal neuter stems, original n-stem jméno ‘name’ has GP with shortened vowel jmen. Type E as in léto ‘summer’ with GP let and an optionally short vowel in the rest of the oblique plural cases: LP letech ~ létech, DP letům ~ létům, lety ~ léty, is represented by only one word, but we believe there is evidence for this paradigm as a pattern (phonologically and morphophonemically close to what is expected from original acute stems) at an earlier

3

The letter G represents the Slovak paradigm with lengthening of the root vowel in the GP. The letter X represents a CLC paradigm with syllabic /r/ or /l/ in the root vowel.

153

stage of the language. There is also dialectal and historical evidence for further variation which we will discuss in the appropriate sub-sections below. Acute Disyllabic Neuter (j)o-stems Table 36 – Acute disyllabic neuter (j)o-stems, all with original long root vocalism PSL

*lěto

S/C

GLOSS

‘summer’

ljeto

SLOVENE

lẹto@

CZECH

léto D/E

SLOVAK

leto G

*dělo

*jetro

‘marsh, swamp’

blato

‘action, piece of work’

djelo

‘liver’

jêtra

jẹ@tra

játra (pl. tant.) D GP jater

jatrá (pl. tant.)

jatra (pl. tant.)

‘phloem, cellulose layer under tree bark’

lıko

líkọ

lýko B

lyko G

lyko

‘butter’

maslo

blátọ

blata dẹ@lọ

*maztlo

bláto B Chod IS blatem4

blato GP blát

boloto

dílo D

dielo

d’elo

bolota

d’ela

djela

NP d’ela

(PSL pluralia tantum) *lyko

l’eto l’eta

ljeta *bolto

RUSSIAN

lyka NP lyki máslọ

máslo B

maslo

maslo masla NP masla

*město

‘place’

mjesto

mẹ@stọ

místo B ‘place’

miesto

m’esto m’esta NP m’esta

*mydlo

‘soap’

mılo

mílọ

mýdlo B

mydlo

mylo, myla NP myla

4

Hruška (1923) cites Chodov (AV locale 302) dialect with IS blatem “Děti, vy děláte s tím medem jako s blatem!” [You kids are making mud out of that honey.]

154 PSL

*pasmo

S/C

GLOSS

‘band, belt’

pasmo

SLOVENE

pásmọ

CZECH

pásmo B

SLOVAK

pásmo

RUSSIAN

pasmo pasma NP pasma

*poto

‘bond’

puto

pọ@tọ Vas.

pouto B

puto G

puto puta NP puta

*puzdro

‘pouch, paunch’

puzdro

Ø

pouzdro B

puzdro

puzdro / puzdro Vas.

*pravo

‘right, law’

pravo

prâvọ

právo B

právo

pravo prava NP prava

*ordlo

‘plow’

ralo

rálọ Vas.

rádlo B

radlo

ralo rala NP rala

*rano

‘morning’

rano

rân m. Vas. rána GS Plet.

ráno B

ráno, raný

rano adv.

*ruxo

‘robe, clothing, stolen from the enemy (Vas.)’

ruxo

rúho

roucho B

rúcho

ruxo

‘lard’

salo

*sadlo

ruxa Vas.

sálọ Vas.

sádlo B5

sadlo

salo sala NP sala

*sito

‘sieve’

sıto

síto

6

síto B

sito G

sito sita NP sita

5

In OCZ the IS is attested without length in ChelPost 156a, “když jeho lénem pohnú a břichem posrostlým sadlem.” 6

Several OCZ sources indicate shortening in the IS: VýklŠal 59b4, “mají váženy a šetřěny býti a jako sytem velikú pilností prosievány.” HusSlovník “sieto, síto: mají jako sytem weliku pilností (býti) prosyeweny.”

155 PSL

S/C

GLOSS

*stado

‘herd’

stado

SLOVENE

Ø

CZECH

stádo B

SLOVAK

stádo

RUSSIAN

stado stada NP stada

*věko

‘eyelid’

Ø

véko Vas.

víko B

veko G

v’eko v’eka NP v’eki

*brjuxo

*dělo

‘belly’

‘cannon’

Ø

Ø

Ø

Ø

břicho A dial. břícho B

brucho G

dělo A 7

delo

br’uxo br’uxa NP br’uxa OR d’elo d’ela8

*jutro

‘morning’

jutro

jútro

jitro A dial. jítro B

jutro

utro utra / s utra NP utra

*kreslo9

‘arm chair’

Ø

Ø

křeslo A

krieslo Vas. kreslo Ver.

kr’eslo kr’esla NP kr’esla

*město ‘city’ WSL only with this meaning

mjesto

*nědro

njedro

‘breast’

mẹ@stọ

město A ‘city’

mesto G

m’esto m’esta NP m’esta

jádro

ňadra NP A plur. tant.

ňadrá NP plur. tant.

OR n’edro n’edra CSR plur. tant. NP n’edra

7

OCZ GP díl (Gebauer III, I:139).

8

Vasmer (I:497) believes the R is borrowed from P działo ‘cannon’ and that the Czech is too, but we believe the rise of quantitative oppositions in PCZ allowed for lexical differentiation from dílo ‘piece of work’. This same process differentiated město ‘city’ from místo ‘place’ as well as a number of other 9

Lith. krẹs@ las mobile acute. Dybo (2000:51) classifies as a.p. (a).

156 PSL

*rydlo

S/C

GLOSS

‘shovel, spade’

rılo

SLOVENE

rílọ Vas.

CZECH

rydlo A

SLOVAK

rydlo

RUSSIAN

rylo ryla NP ryla

*žito

‘rye grain’

*bьrdo

‘shack’

žıto

brdo

žíto

bŕdo

žito A Chod dial. žíto10

žito G

brdo X

brdo G

žito žita NP žita b’ërdo b’ërda b’ërda

*gъrdlo

‘throat’

grlo

gŕlo

hrdlo X

hrdlo

gorlo gorla NP gorla

11

*jьgo s-stem

‘yoke’

*zьrno

‘grain’

ižeso zrno

jígo Vas. igộ

jho X

zŕno

zrno X

Ø

igo iga12

zrno G

z’erno z’erna NP z’ërna

Stang (1965:82) points out that Russian is not very reliable for identifying original neuter acutes because many of them have merged with the mobile paradigm with stress on the desinence throughout the plural. Be that as it may, most of the examples above can be clearly identified as original acutes based on the S/C, SLN, and Czech/Slovak

10

See Map 28 for the dialectal distribution of length in this word. CJA III (2000:154, with accompanying map) gives a full lexical treatment of this word. In Western Moravia, although another lexeme raž has taken over for žito in meaning, wherever the earlier lexeme is attested, the root vowel is short. This is the identical pattern we noted for original acute disyllabic masculine and feminine roots that are short in CLC. 11

Lith. jùngas mobile acute. Dybo (2000:51) classifies as a.p. (a). There is no telling what the effect of the acute should be on an original jer in Czech if the intonation was truly acute in PSL. 12

Zaliznjak (1980:516) indicates that formation of the plural for this word is difficult to fathom.

157

correspondences. Slovak appears to be less consistent in shortening original acute intonation in the neuter paradigms than in the feminine and masculine paradigms. The length that we find in SK pásmo, právo, ráno, rúcho, stádo, is the result of factors we cannot quite identify. We will point out, however, that /a/ is by far the most frequent vowel to be long in these original acute stems.13 Map 25 [Long /y:/ in the NS of /*lyko/ (a.p. a) ‘phloem’] shows a pattern that we have identified earlier as the spread of the shortening of acute length from the East to the West. Compared with Map 27 [Reflex of length in the Nom. Sing. of /*bolto/ (a.p. a) ‘mud, marsh’] and finally Map 28 [Long /i:/ in the Nom. Sing. of /*žito/ (a.p. a) ‘rye’] we can see that this spread has been carried out to a greater extent with bláto/blato and nearly entirely with žíto/žito. In OCZ and CLC acute pl. tant. játra ‘liver’ shortens the root vowel in the GP jater. Oxytonic jádro ‘core, nucleus’ also shortens the root vowel in the GP jader. The shortening in these two words is more likely a relict of the morphological shortening in the GP within a paradigm that has a long root vowel in the other grammatical cases than the reflection of any phonological rule. However, we cannot completely rule out the possibility that the short root vowels in GP jader, jater, indicate that after the loss of weak jers and the vocalization of strong jers, the rule that shortened acute length in trisyllabic forms still operated at some level, phonological, paradigmatic, or perhaps even phonotactic or had spread to other paradigms after the loss of distinctive intonation took 13

Utěšený (1960:61) comments quite appropriately that the shortened original neuter acutes in the Moravian dialects he studied behave differently from the feminine acutes with shortened long vowels.

158

away the original conditions under which the trisyllabic shortening rule has been shown to have operated. Table 37 – Oxytonic disyllabic neuter (j)o-stems with original long root vocalism PSL

GLOSS

S/C

*čit tlo

‘number’

číslo NP čísla

SLOVENE

číslo

CZECH

číslo B

SLOVAK

číslo

RUSSIAN

čislo, čisla NP čisla

‘woodpecker’

*dolto

dlijèto NP dlijèta

dléto

dláto B

dláto

doloto dolota NP dolota

*gnězdo

‘nest’

gnijèzdo NP gnijèzda

gnẹ@zdo

hnízdo B

hniezdo

gn’ezdo, gn’ezda NP gn’ëzda

‘core’

*jedro

Ø

jédrọ

jádro D GP jader

jadro

jadro, jadra NP jadra

*kuzlo

‘smithwork’

Ø

Ø

kouzlo B ‘magic’14

kúzlo

OR kuzlo (Vasmer)

*kridlo

‘wing’

krílo NP kríla

krílo

křídlo B15

krídlo

krylo kryla NP kryl-ja

*liko

‘face’

líce NP líce

líce

líce B

líce

l’ico l’ica NP l’ica

*melko

‘milk’

mlijèko NP mlijèka

mlẹ@kọ

mléko B

mlieko

moloko moloka NP moloka

*myto16

‘toll’

míto NP míta

míto

mýto B

mýto

myto myta NP myta

14

OCZ GP kuzl (Gebauer III, I:139).

15

OCZ GP krzidl (Gebauer III, I:139).

159 PSL

*runo

S/C

GLOSS

‘fleece’

rúno NP rúna

SLOVENE

rúno

CZECH

rouno B17

SLOVAK

rúno

RUSSIAN

runo runa NP runa

*čerslo

‘loins’

Ø

črẹ@slo

tříslo B

trieslo

čer’eslo NP čer’esla

*usta pl. tant.

‘mouth’

ústa

ústa

ústa B

ústa

usta

*vino

‘wine’

víno NP vína

víno

víno B

víno

v’ino v’ina NP v’ina

*vědro

‘bucket’

vèdro NP vedra

vẹ@drọ

vědro18 A

vedro

v’edro, v’edra NP v’ëdra

*vorta pl. tant.

‘gate’

*žězdlo

‘sceptor’

vráta pl. tant.

žèzlo NP žezla

vráta pl. tant.

žẹ@zlọ

vrata pl. tant. A Chod GP vrát

vráta pl. tant.

pl. tant.

žezlo A

žezlo

Ø

vorota

From the Russian data and the S/C data it is evident that in the (j)o-stem neuter disyllabic oxytonic paradigm, the stress was on the initial syllable in the NP. Czech paradigms of Type B with length throughout the paradigm represent the expected reflex. In OCZ we do not find original oxytonic roots whose vowel quantity alternates in the

16

According to Vasmer (III:26), this word was borrowed from OHG mûta. R stress might indicate borrowing into R at later date from West Slavic. 17

The OCZ dictionary project has not reached the letter R yet, but from the card catalogue at the Academy of Sciences we found quite a few attestations of NS runo as well as GP run: e.g. AlbRaj 67a „takéž modlitba bez náboženstvie jako runo volové;“ RokJan 171b, 2, “owce ma dwie sukni, gednu wzdy lidem dáwá a druhú zachowává, to gest runo, kterez dáwá;“ VýklKruml 41b2, “vellus lane runo volnij (rolnij?)”, VýklKruml „De vellibus z run.“ 18

Bulaxovskij (1983:169) points out that not only is the shortening in CZ puzzling, but the SLN variant vẹd@ rọ / védro is also odd given the evidence for oxytonic from R and S/C.

160

grammatical cases that we find the acute substantives alternating in, although there was some tendency to shorten long root vowels in the GP and sporadically in originally trisyllabic forms (e.g. Tkač 44 “já jsem jemu ... vin nechtěl naložiti” Tkač 38 “měl se mnú spolky a viních a vjiných kupectví”). The shortenings in the GP we interpret as further evidence that the morphophonemic rule that changed a long vowel to a short vowel in the GP of an otherwise long root vowel paradigm and changed a short vowel to a long vowel in the GP of an otherwise short root vowel paradigm operated at some period in the history of Czech. The shortenings in trisyllabic forms of the oxytonic paradigm, however, are hardly numerous enough to deduce a pattern from or base a theory upon. We have from the feminine neo-acute paradigm several examples of shortened /o/ in the IS in OCZ (volí, koží) and other trisyllabic forms. It is possible that there was some mirroring of the acute paradigm with shortened long vowels in the trisyllabic forms or that there was an overlapping in time of the shortening of length in trisyllabic forms independent of intonation, but we may never know for sure how the intonation affected this rule or its possible spread to other environments.

161 Table 38 – Oxytonic disyllabic neuter (j)o-stems with original short root vocalism PSL

GLOSS

*pero

‘feather’

*sedlo / selo

‘seat, settlement’

S/C

SLOVENE

pèro pèra NP pèra / pera

pẹ@ro

sèlo NP sela

sélo

CZECH 19

péro B / pero A

SLOVAK

pero G

RUSSIAN

p’ero p’era NP p’er-ja

sídlo B sedlo A

sídlo sedlo G

s’elo s’ela NP s’ëla

*stьblo

‘stalk’

stáblo

stéblo

stéblo B

steblo / stéblo

st’eblo st’ebla NP st’ëbla

‘small of back’

*bedro

bèdro NP bedra

bédrọ

bedra pl. tant. A

bedro

b’edro b’edra NP bëdra

*čelo

‘forehead’

čèlo NP čela

čélọ

čelo A

čelo

čelo čela NP čëla

*pьklo

‘hell’

pàkao GS pàkla

péklọ

peklo A

peklo

p’eklo p’ekla NP p’ekla

*stegno

‘thigh’

stègno NP stegna

stégnọ

stehno A

stehno G

st’egno st’egna NP st’ëgna

*vez-tlo

‘oar’

vèslo NP vesla

véslọ

veslo A

veslo

v’eslo, v’esla NP v’ësla

One tendency that was noted for the feminine stems with neo-acute intonation and masculine monosyllabics grammatical cases with neo-acute intonation (for feminine stems the pattern was noted as well in a number of originally oxytonic words), seems to be less

19

OCZ GP per (Gebauer III, I:146).

162

pronounced but present, nevertheless, in the neuter paradigms – we have found several attested lengthenings of original short vowels. As far as we can tell, the neuter words with attested lengthened short vowel in the Czech dialects are the following: CLC lůno ‘lap’ (indeterminate a.p.); CLC kůzle (a t-stem, in variance with kozle) ‘goatling, kid’; CZ dial. kůtě (CLC kotě ) ‘kitten’ (a t-stem); in most CZ dialects including CLC stéblo (< stьblo) ‘grain stubble, chaff’ [a.p. (b)]; péro ‘feather’ [a.p. (b)]; sídlo ‘settlement’ [a.p. (b)]; CZ dial. vůje (CLC oje pl. tant.) ‘wooden rod that connects the horse harness to the wagon’; and CLC pole ‘field’, which is attested in only four of the towns used in the Czech dialectal atlas project. All of these towns (103, 105, 106, 108)20 are located in the Northern Bohemian dialect region. These examples do not lend themselves to the same interpretation. They did not all belong to the same paradigm and are difficult to make any generalizations about. We note, however, that stéblo, péro, sídlo were end-stressed and it is possible that despite the fact there are a number of oxytonic neuter (j)o-stems with short vowel that remained short in CLC and the dialects (bedra, čelo, peklo, stehno, veslo), there may have been a tendency to lengthen short vowels of original oxytonic words when the ictus was retracted as there was under the neo-acute retraction of ictus.

20

ČJA III (p. 100, map on p. 101) shows other concentrated dialect groups with póle that must be later lengthenings that are phonologically unrelated to půle even though the dialects are in the same general region, but not contiguous.

163 Table 39 – Mobile disyllabic neuter (j)o-stems PSL

S/C

GLOSS

SLOVENE

CZECH

SLOVAK

RUSSIAN

*bolgo

‘goodness’

blâgo

blagộ

blaho A

blaho

OR (ESSJ) bologo dial. bologo

*dervo s-stem

‘tree, wood’

drvo

drẹvộ

dřevo A

drevo G

d’er’evo, d’er’eva

NP dròva

NP d’er’evja *mąso

‘meat’

mêso

mesộ

maso A

mäso G

m’aso, m’asa

*pivo

‘beer’

pîvo

pívo

pivo A

pivo

p’ivo p’iva NP p’iva

*proso

‘millet’

proso

prosộ

proso A

proso

proso prosa NP prosa

*seno

‘hay’

sıjeno

sẹnộ

seno A

seno G

s’eno s’ena NP s’ena

*slnce *tělo s-stem

‘sun’ ‘body’

sûnce pod sūncem

sôłnce

tıjelo,

tẹlộ

slunce A

slnce G

solnce solnce

NP tıjela

tělo A OCZ GP tiel

telo

t’elo t’ela NP t’ela

*těsto

‘dough’

tıjesto

tẹstộ

těsto A

cesto G

t’esto t’esta NP t’esta

*věno

‘trousseau’

Ø

Ø

věno A

veno G

v’eno v’ena NP v’ena

*zolto

‘gold’

zlâto zlâta

zlatộ

zlato A

zlato

zoloto zolota NP zolota

*krosno

‘rucksack’

krosna pl. tant.

krósna pl. tant.

krosno A

krosná

krosno krosna NP krosna

164

From the reflexes of original circumflex [a.p. (c)] (j)o-stem neuters with original long root vocalism and original short root vocalism in CLC and the Czech dialects, it is clear that the predictions made by the traditional rule that posits shortening of circumflex length is exactly what we see as the result. In general, the phonological shortening of original circumflex length and the generalization of the paradigm pattern with brevity throughout is one of the most consistent historical accentual principles that we find in CLC and the dialects. This principle applies to masculine, feminine, and neuter disyllabic substantives regardless of stem theme vowel and there appears to be very little evidence for dialectal variation. The one example of an original circumflex root with dialectal long vowel in the NAS is pivo ‘beer’ which is attested in a small region of Moravia as described by Utěšený (1960:67, Map II.D), and appears to be isolated and localized.21 Table 40 – Indeterminate neuter (j)o-stems PSL

*dno

S/C

GLOSS

‘bottom’

dno

SLOVENE

dnọô

CZECH

dno X

SLOVAK

dno

RUSSIAN

dno dna NP donja

*stьklo

‘glass’

sklo

steklọô

sklo X

sklo

st’eklo st’ekla NP st’ëkla

*lono

‘lap’

lono ESSJ

Ø

lůno B

lono G

lono, lona NP lona

A few neuter (j)o-stems are indeterminate with respect to Proto-Slavic accentual paradigm, but illustrate reflexes of other root shapes that might be relevant for other

21

The colloquial Prague pronunciation [pívo] is most likely expressive and unrelated to the dialectal form.

165

studies. PSL *dno was monosyllabic and shows no phonological divergence in Czech across the dialects. PSL *stьklo looks like an original oxytone but the Czech & Slovak reflex might be expected with vocalization of the jer (as in stéblo). Theoretically, the lengthening in PSL *lono CZ lůno might be attributable to the neo-acute intonation if it could be proven that the word was originally oxytonic. The entry in the StČS attests to the fact that the length is quite old – Gebauer cites two manuscripts from the early 15th century, with luono, lónu.

Chapter 8

CONCLUSIONS

By examining the geographical distribution of a single phonological binary feature, vowel quantity with the opposition [+long:±long], as a synchronic cross-section in selected ancient lexical items across the contemporary dialectal Czech landscape, and analyzing the variance of this feature against the diachronic phonological, morphophonemic, and morphological paradigmatic progression of its development, we have made use of a method that has provided evidence for sound changes, analogical morphological and morphophonemic developments that correlate to ancient intonational patterns which were characteristic of the Proto-Slavic common ancestral language. The summary enumerated below of the chief conclusions reached by this study is intended to show the importance of including dialectal data in future inquiries whose aim is to detect the traces of the Proto-Slavic language on the development of the individual Slavic languages. 1.

Arguments in favor of the so-called Czech lengthening, whereby “a short rising vowel in an open first syllable of disyllabic words is lengthened unless the second syllable contains a long vowel” (Verweij, 1990:505) are circular and cannot be satisfactorily supported. The short rising intonation is supposed to have applied to both original acute vowels that were shortened prehistorically (while keeping their rising intonation), as well as 166

167

to internally stressed original short vowels. An argument for the shortening of acute length in Czech is seen as desirable in order to force Czech to agree with the other Slavic languages where acute length was lost, but we have tried to show that it is at least as plausible that the shortening of acute length was simply not entirely carried out on the Czech and Upper Sorbian periphery of the Proto-Slavic territory. Additionally, the Czech Lengthening rule is supposed to have operated on /e, o, ъ, ь/ as well, for which there is little evidence outside the oxytonic and neo-acute paradigms which are not supposed to have been covered by the rule – a clear contradiction to the proposal. The condition that the vowel lengthened only if it was originally stressed is designed to explain the fact that pretonic vowels in general retained their length in Czech, but in no way offers an explanation for original unstressed short vowels that lengthened either in CLC or dialectally such as kůpa, sůva, růsa, vůje, lůno, déšt’, hrůza, kůra, můra, vůsa, etc. The restraint on the Czech Lengthening to environments not followed by a long vowel is required to explain IS ranou, DP ranám, LP ranách, a pattern that is only characteristic of a few words in CLC. According to the rule, GP ran does not lengthen because the originally acute /a/ was in a closed syllable, but the comparative evidence for a later morphophonemic rule in the GP offers a much clearer explanation of the interplay of length and brevity across the substantival system. In terms of Contemporary Czech, in

168

paradigms that have an alternation between a long and a short vowel, the alternation is always described as a shortening of the long vowel and not vice versa which is the exact opposite of what we would expect if the putative Czech lengthening were to have left its mark on the Czech paradigmatic system. 2.

The traditional views of the retention or loss of length in Czech based on the original PSL intonations and their reflexes offer the greatest explanatory power. Acute length in literary Czech is preserved PSL length. The copious examples of original acute stems with shortened long vowel in CLC are to be attributed to influence from Moravian and Slovak dialects during the geo-political prominence of the Great Moravian Empire. These shortenings progressed by lexical diffusion and affected the Central Bohemian dialects, but the sound change was not carried out completely on the periphery. Southwestern dialects, and particularly the Chod dialects which formed a small but politically and socially cohesive group, were the most resistant to the shortenings of original acutes. The shortenings did not affect original neo-acute length, so there must have still been a distinction between neo-acute and acute intonation when the acute lengths were shortened. Evidence is abundant for feminine, masculine, and neuter acute paradigms. We have created the following Maps for this study that illustrate these principles: Map 1 – Lengthened /í/ in the Nom. Sing. /*jьskra/ (a.p. a) ‘spark’; Map 4 – Reflex of Short

169

Vowel in the Nom. Sing. /*griva/ (a.p. a) ‘mane’; Map 5 – Reflex of Shortened /e/ in the Nom. Sing. /*berza/ (a.p. a) ‘birch’; Map 7 – Reflex of Retained Long Vowel in the Nom. Sing. /*repa/ (a.p. a) ‘with beet’; Map 8 – Distribution of Length & Brevity in Nom. Sing. /*solma/ (a.p. a) ‘straw’; Map 11 – Long /ú/ in the Nom. Sing. of /*duha/ (a.p. a) ‘arc, rainbow’; Map 21 – Long /á/ in the Gen. Sing. of /*gorxъ/ (a.p. a) ‘peas’; Map 23 – Lengthened /u/ in the Nom. Sing. of /*hus(a)/ (a.p. a) ‘goose’; Map 24 – Lengthened /á/ in the Nom. Sing. of /*sazje/ (a.p. a) ‘soot’; Map 25 – Long /y:/ in the Nom. Sing. of /*lyko/ (a.p. a) ‘phloem, cellulose layer of a tree’; Map 27 – Reflex of length in the Nom. Sing. of /*bolto/ (a.p. a) ‘mud, marsh’; Map 28 – Long /í/ in the Nom. Sing. of /*žito/ (a.p. a) ‘rye’. This hypothesis is further supported by data from Upper Sorbian. A synthesis of the original disyllabic acute substantives and the Czech dialectal shortening of the long vowels in the lexical items examined can be found in Map 29 – Shortening of Acute Long Vowels in Prehistoric Czech. 3.

Traces of ancient accentological features can be detected in Czech, the dialects, and Old Czech. A correlation between the CLC ā-stem paradigm Types E/F and the Proto-Slavic acute accentual paradigm (a.p. a), characterized by shortening of the original acute long vowel in the trisyllabic forms IS, DP, LP . There is also a correlation between the CLC ā-stem paradigms Types B/D and the Proto-Slavic oxytonic accentual

170

paradigm (a.p. b). A majority of the original oxytonic ā-stems have retained length throughout the paradigm. The CLC ā-stem paradigm Type A and the Proto-Slavic mobile accentual paradigm (a.p. c) coincide with one another rather neatly. Nearly all original circumflex words have generalized brevity throughout the paradigm. This correlation applies to masculine and neuter circumflex paradigms as well. 4.

Proto-Slavic identification with a single accentual paradigm was not an absolute regularity: PSL words could vary dialectally / intonationally resulting in mixed reflexes among the Slavic languages -- sometimes even dialectally within a given individual Slavic language. A good example is represented by zima ‘cold, winter’ (see Map 13) with a distribution of length and brevity in the Czech dialects unlike any other word.

5.

The best evidence in Czech of the Proto-Slavic neo-acute intonation is the effect the neo-acute had of lengthening short /o/ in CLC and the dialects. Numerous examples include chůze, chvojě/chvůjě, kůže, nůše, vůle, vůně, which form a morphological class in Proto-Slavic. The manner in which Czech has extended the effects of the neo-acute to other paradigms is discussed in Chapter 6 and 7 and is described as a morphological extension, most prominent in original oxytonic ā-stems.

6.

Loss of circumflex length in Czecho-Slovak preceded loss of PSL acute length. The regularity that Czech and Slovak show in the shortening of circumflex length and the generalization of brevity in the words of a.p. (c)

171

provides clear evidence that this sound change was carried out in its entirety, and when compared and contrasted with the behavior of original acute paradigms with much greater dialectal variation, and neo-acute paradigms with separate but predictable reflexes, support for a distinction between circumflex intonation, acute intonation, and neo-acute intonation in Proto-Czecho-Slovak is provided. 7.

There is excellent evidence in support of the morphophonemic rule changing a long vowel to a short vowel in the GP of an otherwise long root vowel paradigm and changing a short vowel to a long vowel in the GP of an otherwise short root vowel paradigm. The effects of this change, however, have largely given way in the Czech lands to the analogical process generalizing length or brevity throughout a substantival paradigm. All that remains of this rule is a morphological tendency to shorten the root vowel in the GP. Whenever there is an alternation between a long vowel and a short vowel in a CLC paradigm, the GP takes place in the alternation and is short. If we were to wager a guess and could see into the future, we would either predict that the generalization of brevity in the GP will continue to spread to other words, or will be analogically leveled to the rest of the paradigm for the lexical items that keep the alternation.

8.

In contrast to the feminine oxytonic ā-stem paradigms with original short root vocalism that exhibit the reflex of lengthened /o > ó > uo > ů/ throughout the paradigm as the result of a morphophonemic spread of the

172

effect of the phonological neo-acute lengthening of /o/ in the soft-stem paradigms such as vůle, růže, etc. (e.g. Chod vůsa ‘wasp’), there is no similar pattern for the oxytonic neuter substantives with original short root vocalism, although there are several oxytonic words that appear to have lengthened original short vowel (stéblo, péro, sídlo). 9.

Relative chronology Loss of circumflex length in Czecho-Slovak (and Upper Sorbian) Neo-acute retraction that feeds Czech (Slovak) lengthening of /o/ under the neo-acute Shortening of acute length in trisyllabic words (may operate on neo-acute lengths as well: cf. OCZ volí < *volьjo but does not operate on contracted VjV sequences) Phonemicization of quantity in Czech and Slovak Onset of contraction Loss of weak jers Vocalization of strong jers Morphological generalization of short vowel / long vowel in GP in order to differentiate the GP from other forms of the paradigm

173

Bibliography Major Reference works employed – Czech 1908- (2001) Slovník staročeský. Praha. 1936-1958

Příruční slovník jazyka českého. Praha.

1978

Slovník spisovné češtiny pro školu a veřejnost. Praha.

1982

Srbocharvátsko-český slovník. Praha.

1999

Velký obrazový slovník. Montreal.

Bartoš, František. 1906 Berštejn, S.B. et al. 1967

Dialektický slovník moravský. v Praze. Karpatskij dialektologičeskij atlas. Moskva, I. 272 str., II 8+ 212 str.

Bibliografický katalog ČSFR. 1989České disertace a autoreferaty. Praha: Narodní knihovna v Praze. České disertace 1979-1986. Praha: Statní knihovna CSFR. Československé disertace 1900-1978. [Praha]: Statní knihovna CSFR. ČSAV. Ústav pro jazyk český. 1995 Český jazykový atlas. 1. díl. Praha: 1997

Český jazykový atlas. 2. díl. Praha:

1998

Český jazykový atlas. 3. díl. Praha:

Disertace Pražské univerzity 1882-1953. [published in] 1965. Praha: Univerzita Karlova. Etymologický slovník slovanských jazyků (rkp.). v. III (nabat-obьtь). Red. Kruh. F Kopečný, E Havlová, J. Němec, interní red. H. Plevačová. Autoři: V. Čapková, E. Havlová, F. Kopečný, H. Plevačová, V. Polák, Ž. Šarapatková, V. Šaur, P. Valčáková. Rec. pro nakladatelství: Ved. red. A. Lamprecht. Machek, Václav. 1971

Etymologický slovník jazyka českého. Praha.

174

Rejzek, Jiří. 2001 Sedláček, A. 1908

Český etymologický slovník. Voznice. Místopisný slovník historický království českého. Praha.

Major Reference works employed – Other Slavic Languages 1894-1895

Slovensko-nemski slovar / uredil Maks PletersnikV Ljubljani.

1967

Słownik górnołużycko-polski. Warszawa.

1974-

Etimologičeskij slovar’ slavjanskix jazykov. Moskva.

1991-1999

Historický slovník slovenského jazyka. Bratislava.

Snoj, Marko. 1997 Sreznevskij, I.I. 1893-1912 Vasmer, Max. 1987

Slovenski etimološki slova. Ljubljana. Materialy dlja slovarja drevne-russkogo jazyka. Moskva. Etimologičeskij slovar’ russkogo jazyka. Moskva.

Monographs and articles cited Bartek, H. 1933

“Slovenské výsledky praslov. přízvučných akútových dĺžok.” In: Sborník na počest J. Škultétyho. Turč.

Bělič, J. – V. Křístek. 1964 Moravskoslovenská nářečí. Olomouc. Bělič, J. 1968

“Naléhavé úkoly české dialektologie.” Slovo a slovesnost 29:287-294.

1971

Nástin české dialektologie. Praha.

1954

Dolská nárecí na Moravě. Praha.

175

Biblia Slavica: Herausgebeben von Hans Rothe und friedrich Scholz unter Miterbeit von Christian Hannick und Ludger Udolph. Serie I: Tschechische Bibeln. Band 1 Dresdener Bibel / Drážďanská Biblie Blahoslav, J. Gramatice česká 1991 Gramatika česká Jana Blahoslava. Masarykova Univerzita v Brně. Brněnský dialektologický kolektiv (J. Chloupek, et al.) 1954 “K diferenciaci moravskoslovenských nářečí a k jejich vztahu k českému narodnímu jazyku.” Sborník prací fil. fak. Brno. Bulaxovskij, L. A. 1954 1983 Collinge, N.E. 1985

Russkij literaturnyj jazyk pervoj poloviny XIX veka; fonetika, morfologija, udarenie, sintaksis. Moskva. Vibrani pratsi vpiati tomax. vol. 5. Slovjanskaja aktsentologija. Kiev. The Laws of Indo-European. Philadelphia.

Comrie, B. & G. Stone. 1978 The Russian Language since the Revolution. Oxford. Cuřín, Fr. 1967

Chlumský, Josef. 1911 1928 Černý, Fr. n.d.

Studie z historické dialektologie a toponomastiky Čech. Praha. mapa IV, a výklad spolu s doklady na s. 68n. Praha Pokus o měření českých zvuků a slabik v řeči souvislé. Praha. Česká kvantita, melodie a přízvuk. Praha. “Studie o české kvantitě.” Listy filologické, XXIV:346-353. Praha.

Dialektologická komise. Dialektologická komise české akadeimie věd a umění. 1951 Pravidla pro vědecký přepis dialektických zápisů českých a slovenských. Praha. 1943 Dušek, V.J. 1894

Pravidla pro vědecký přepis dialektických zápisů Věstník české akademie věd a umění, 52. Praha. Hláskosloví nářečí jihočeských I. Praha

176

1896

Hláskosloví nářečí jihočeských II. Praha

1908

Hláskosloví nářečí jihočeských III. Praha.

Dybo, V.A. et al. 1990

Osnovy slavjanskoj aktsentologii. Moskva.

1993

Osnovy slavjanskoj aktsentologii: slovar’. Moskva.

2000

Morfonologizovannye paradigmaticheskie aktsentnye sistemy: tipologija i genezis, t. 1. Moskva.

Dzendzelivskij, J.O. 1958 1960

Linhvistyčnyj atlas ukrajinskych narodnych hovoriv zakarpatskoji oblasti URSR I. Užhorod. Linhvistyčnyj atlas ukrajinskych narodnych hovoriv zakarpatskoji oblasti URSR II. Užhorod.

Česká dialektologická komisa. Ústav pro jazyk český, ČSAV. Praha. 1964 Dotazník pro výzkum českých nářečí (Český jazykový atlas). [Internal Questionnaire]. Flajšhans, V. 1903 1957

Folprecht, J. 1905 Friedrich, Paul. 1970

Nejstarší památky jazyka i písemnictví českého. Díl I. Prolegomena a texty. Praha. Klaret a jeho družina. – Sv. 1 Slovníky veršované, Praha, 1926. – Sv. 2. Texty glosované, Praha 1928. – viz. též výbor z čes. lit. (akad.) I, Praha 1957, 772. Includes summaries and descriptions of works attributed to Klaret Sv. I. Vokabulár grammatický Sv. II. bohemář III. Klaret Glossář. Příspěvky k mluvě lidu slováckého na moravském Podluží. Výr. zpráva reálky v Plzni. s. 15n. Proto-Indo-European Trees. Chicago.

Gauthiot and Vendrys. 1900 “Note sur l'accentuation du tchèque,” MSL, XI. Paris. Gebauer, Jan. 1894 1906-1912

Historická mluvnice jazyka českého. Praha. Slovník staročeský. Praha.

177

Gimbutas, Marija. 1971 Goląb, Zbigniew. 1991 1994 Hála, Bohuslav. 1962 Hattala, Martin. 1854

The Slavs. New York. The origin of the Slavs: a linguist’s view. Columbus, Ohio. “Three stages in the developoment of Common Slavic.” [unpublished paper] Uvedení do fonetiky češtiny na obecně fonetickém základě. Praha. Zvukosloví jazyka staro- i novo-českého a slovenského. Praha.

Havránek, Bohuslav. 1936 Československá vlastivěda. Řada II. Spisovný jazyk český a slovenský. Praha. Horálek, Karel. 1962 Hujer, Oldřich. 1936 Hruška, J.F. 1907 Illič-Svityč, V.M. 1963 Jagić, Vatroslav. 1913 Jakobson, Roman. 1929

Ůvod do studia slovanských jazyků. Praha. “Vývoj českého jazyka.” In: Československá vlastivěda, díl III – Jazyk. Praha. Dialektický slovník chodský. Praha. Imennaja aktsentuatsija v Baltijskom i slavjanskom. Moskva. Entstehungsgeschichte der kirchenschlavischen Sprache. 2nd edition. Berlin. “K odstraňování dlouhých samohlásek v češtině.” Slavia 7. Praha.

1963

“Opyt fonologičeskogo podxoda k istoričeskim voprosam slavianskoj akcentologii.” In: American Contributions to the Fifth International Congress of Slavists, Sofia, 1963. The Hague.

1964

“The Byzantine mission to the Slavs.” Dumbarton Oaks Papers. No. 19:257-265.

178

Jančák, Pavel. 1970

“K jazykovému zeměpisu Čech: některé případy západočeského krácení v koncovkách.” Naše Řeč, 53:129-137.

Kašík, A. 1908

Popis a rozbor nářečí středobečevského. Praha.

Kopečný, František. 1981

Základní všeslovanská slovní zásoba. Praha.

Král, J. and Fr. Mareš. 1911 “Trvání hlásek a slabik dle objektivné míry.” Listy Filologické XX. Praha. Kortlandt, F.H.H. 1975 Kučera, Henry. 1961 Kul’bakin, S.M. 1903 Labov, William. 1994

Slavic accentuation: a study in relative chronology. Lisse / Netherlands. The Phonology of Czech. ‘S-Gravenhage. “K istorii i dialektologii pol’skogo jazyka.” SORJAS LXX. Petrohrad. Principles of linguistic change: internal factors. Cambridge, USA.

Lutterer, Ivan; Milan Majtán; Rudolf Šrámek. 1988 Zeměpisná jména Československa: slovník vybraných zeměpisných jmen s výkladem jejich původu a historického vývoje. Praha. Marvan, Jiří. 2000 Mauer, Eduard. 1984 Moszyński, K. 1957 Palacký, František. 1936

Jazykové milénium: slovanská kontrakce a její český zdroj. Praha. Chodové: historie a historická tradice. Praha. Pierwotny zasiąg języka prasłowiańskiego. Wrocław. Dějiny národu českého: v čechách a v moravě. V Praze.

179

Přemysl, Janota, and Pavel Jančák. 1972 “An investigation of Czech vowel quantity by means of listening tests.” Phonetica pragensia 2:3168. Praha. Rostafiński, J. 1908

O pierwotnych siedzibach i gospodarstwie Slowian w przedhistroycznych czasach. Cracow.

Schaarschmidt, Gunter. 1997 The Historical phonology of the Upper and Lower Sorbian Languages. Heidelberg. Schenker, Alexander M. 1995 The Dawn of Slavic. New Haven. Sedláček, A. 1920 “Snůška starých jmen, jak se nazývaly v Čechách řeky, potoky, hory a lesy.” Rozpravy ČA, tř.I, č.60. Praha. Shevelov, George Y. 1965 A Prehistory of Slavic; the historical phonology of Common Slavic. New York. Sköld, H. 1923

„Sledy praslavijanskogo udarenija v češskom jazyke.” Slovanský sborník F. Pastrnkovi. Praha.

Slovenska Akademija Znanosti in Umetnosti. 1950 Slovenski pravopis. Ljubljana. Stang, Christian. 1957

Slavonic accentuation. Oslo.

Stankiewicz, Edward and Dean S. Worth. 1966 A Selected bibliography of Slavic linguistics. The Hague. Stankiewicz, Edward. 1968 Declension and gradation of Russian substantives. The Hague. 1986

The Slavic languages: unity in diversity. Berlin.

1993

The Accentual patterns of the Slavic languages. Stanford, California.

Syllaba, Theodor. 1983

Jan Gebauer na Pražské univerzitě. Praha.

180

Tichá, Zdeňka. 1984 Timberlake, Alan. 1983a

1983b

Cesta starší české literatury. Praha. “Compensatory lengthening in Slavic: 1: conditions and dialect geography.” In: From Los Angeles to Kiev [V. Markov and D.S. Worth, eds.]. Columbus, Ohio. “Compensatory lengthening in Slavic: 2: phonetic reconstruction.” In: American contributions to the Ninth International Congress of Slavists, Kiev, September 1983, vol. I, Linguistics (293-319). Columbus, Ohio.

Trávníček, František. 1908 “K voprosu o češkom količestve glasnyx” izd. Otd. russk. jazyka i slovesnosti, AN, 1918 XXIII, 2 izd. 1921:16-17. Moscow. 1928

“Česk. vzhůru” Listy filologické:VIII:178.

1932a

“O kvantitě dvouslabičných ā- a jā- kmenů s původním přízvukem na koncovce a s praslovanskými dlouhými vokály.” Časopis pro moderní filologii a literatury, 7. 1921, p. 150-152.

1932b

Úvod do české fonetiky. Prague.

1934

“Ke kvantitě mužských kmenů na -o-, -jo-, a -u-, v češtině.” Listy filologické, 48:104-107.

1935

Historická mluvnice československá. Praha.

Ústav pro jazyk český, ČSAV 1945 Slovníkový dotazník pro nářečí českého jazyka I. Praha. 1948 Utěšený, Slavomir. 1958 1960 Vaillant, A. 1923

Slovníkový dotazník pro nářečí českého jazyka II. Praha. “K otázce původu českých nářečních oblastí.” Slavia XXVII:188-208. Nářečí přechodného pásu česko-moravského. Praha. “Le problème des intonations balto-slaves.” Bulletin de la Société de linguistique, 37. Paris.

181

1958 Vajs, Josef. 1932 Vančura, Vladislav. 1959 Verweij, Arno. 1982

Grammaire comparee des langues Slaves, t. II: Morphologie. Paris. Rukovět’ hlaholské paleografie: uvedení do knižního písma hlaholského. v Praze. Obrazy z dějin národa českého. Praha. “Ke spisku Orthographia Bohemica.” Listy filologické 105:75-89. Praha.

Vidmanová, Anežka. 1994 “Quantity patterns of substantives in Czech and Slovak.” In: Dutch contributions to the Eleventh International Congress of Slavists, Bratislava, August 30 – September 9, 1993. Amsterdam. Vondrák, V. 1904

O původu kijevských listů u Pražských zlomků. V Praze.

1906

“Slavische Akzent- und Quantitätsstudien.” Beiträge zur Kunde der indogermanischen Sprachen, 30. [n.p.]

1924

“Další příspěvky k nauce o praslovanském přízvuku.” Spisy filosofické fakulty Masarykovy university, 9. Brno.

Voráč, J. 1955

Česká nářečí jihozápadní: studie jazykově zeměpisná. část první. Praha.

1962

“K českému jazykovému atlasu.” Informační bulletin pro otázky jazykovědné, 3:62-65. Praha.

1976

Česká nářečí jihozápadní: studie jazykově zeměpisná. část druhá. Praha.

APPENDIX A – AV DIALECT SURVEY LOCALES

182

183

1 2 3 4 5 6 7 8 9 10 11 12 13 14 21 22 23 24 25 26 27 30 31 32 40 41 42 43 44 45 49 51 52 53 59 61 62 63 64 65 66 67 71 72

Cheb Tachov Karlovy Vary Kadaň Most Ústí nad Labem Česká Lípa Rumburk Frýdlant Trutnov Nová Paka Hradec Králové Vysoké Mýto Mladá Boleslav Roudnice nad Labem Praha Kolín Rakovník Příbram Benešov Ledeč nad Sázavou Železná Ruda Plzeň Klatovy Vyšší Brod Strakonice Tábor Jindřichův Hradec Prachatice České Budějovice Slavonice Jihlava Třebíč Nové Město na Moravě Svitavy Znojmo Moravský Krumlov Brno Boskovice Litovel Prostějov Kroměříž Lipník nad Bečvou Vsetín

73 74 75 79 81 82 83 84 91 92 93 94 95 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 124(a) 125 126 127 128 129 130

131 Kyjov 132 Uherský Brod 133 Břeclav 134 Mikulov 135 Opava 136 Ostrava 137 Příbor 138 Český Těšín 139 Šumperk 140 Bruntál 141 Moravský Beroun 142 Jeseník 143 Osoblaha 144 Držkov 145 Paseky nad Jizerou 146 Poniklá 147 Horní Dušnice 148 Mrklov 149 Hoření Paseky 150 Všelibice 151 Frydštejn 152 Lestkov 153 Loukov Nová Ves nad Popelkou 154 155 Horní Kalná Rtyně v Podkrkonoši 156 157 Slatina nad Úpou 158 Horní Rybníky 159 Hlavňov 160 Machov 161 Slané 162 Jakubovice 201 Mukařov 202 Všeň 203 Střeleč 204 Úlibice 205 Pecka 206 Šárovcova Lhota 207 Bílá Třemešná 208 Vilantice 209 Dolany 210 Slavětín nad Metují 211 Česká Čermná 212 Kounov 183

Trnov Libel Liberk Žampach Klášterec nad Orlicí Mistrovice Horní Heřmanice Obruby Údrnice Myštěves Neděliště Sedlice Bělečko Lány u Dašic Jenišovice Skořenice Řetůvka Kozlov Březinka Úherce Nouzov Lužec nad Cidlinou Vlčí Habřina Lipoltice Jezbořice Lukavice Horní Bradlo Studnice Otradov Morašice Poříčí u Litomyšle Trstěnice Raná Chrášťany Siřejovice Libotenice Pnětluky Jimlín Hříškov Orasice Brníkov Bechlín Stražníce Sedlec

184

213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256

Krpy Všejany Semice Chleby Vrbová Lhota Ohaře Mikuláš Přibyslavice Běstvina Střížov Mšecké Žehrovice Dobrovíz Dolín Budihostice Kly Bašť Zápy Královice Tismice Krymlov Losiny Chlístovice Kozlov Tis Lašovice Chyňava Korno Kosoř Těptín Ostředek Popovice Strojetice Rejčkov Dobrohostov Broumy Radouš Velká Lečice Drevníky Svatý Jan Křečovice Smilkov Šebiřov Bratrice Velký Rybník

257 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 401 402 403 404 405 406

Šimanov Postřekov Chodov Draženov Mrákov Zahořany Sulislav Všekary Lštění Černíkov Chodská Lhota Újezd Potvorov Hvozd Žebnice Nekmíř Hromnice Vochov Čižice Zemětice Vlčí Ježovy Habartice Rovná Nemilkov Přílepy Šípy Kozojedy Vejvanov Volduchy Lhůta Chocenice Kramolín Velenovy Drozdov Těně Číčov Čížkov Drahlín Vysoká u Příbramě Horní Hbity Zalužany Kovářov Ředice

407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450

Božejovice Sedlečko u Chotovin Prasetín Myslotín Rohozná Věšín Drahenice Cerhonice Oslov Borovany Lom Těmice Polesí Chanovice Čekanice Stará Dobev Albrechtice nad Vltavou Březí Březnice Borkovice Tučapy Pluhův Žďár Kamenný Malíkov Mosty Nezamyslice Kraselov Čepřovice Křtětice Podeřiště Zahájí Drahotěšice Ponědraž Mláka Lásenice Jiřičná Nezdice Stachy Onšovice Lažiště Hracholusky Jáma Rojšín Čakov Vrábče

185

451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 601 602 603 604 605 606 607 608 609

Hůrky Kojákovice Staňkov Přísečná Doudleby Ločenice Soběnov Slavče Třebeč Buková Rapšach Hodice Světlá Volířov Dyjice Pustá Rybná Rohozná Hrbov Velké Losenice Světnov Kadov Ubušín Věstín Vojtěchov Jámy Bobrůvka Krásněves Bohdalov Řehořov Velký Beranov Příseka Hladov Červený Hrádek Báňovice Mladoňovice Heraltice Cidlina Nové Syrovice Oslnovice Pavlínov Hodov Sedlec Příložany Rešice

610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653

Slatina Ctidružice Olbramkostel Mašovice Plaveč Morašice Tasovice Věžná Lomnice Heřmanov Holubí Zhoř Zbraslav Chudčice Žebětín Čučice Hlína Jezeřany Syrovice Nosislav Ivaň Horní Poříčí Zboněk Bořitov Šošůvka Lažany Deblín Ořešín Habrůvka Habrovany Telnice Hodějice Bošovice Žarošice Boleradice Písařov Jedlí Chromeč Dlouhomilov Lupěné Bohuslavice Šumvald Palonín Bohdalov Roubanina

654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 701 702 703 704 705 706 707 708 709 710 711 712

Biskupice Knínice u Boskovic Hrochov Suchý Otinoves Ruprechtov Radslavičky Otaslavice Ohrozim Třebčín Pěnčín Hvozd Příkazy Hnojice Dolany Velký Týnec Charváty Hrubčice Tvorovice Tištín Morkovice Staré Hvězdlice Velký Újezd Vinary Záříčí Zdounky Radotín Nahošovice Loukov Hlinsko pod Hostýnem Němčice Mysločovice Spálov Olšovec Milotice nad Bečvou Lhota u Kelče Hošťálková Rusava Fryšták Lhota Kvasice Huštěnovice Cetechovice Buchlovice

186

713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756

Polešovice Vřesovice Jestřabice Vlkoš Strážovice Kobylí Velké Bílovice Šakvice Bernartice nad Odrou Petřkovice Juřinka Velká Lhota Trnava Klečůvka Částkov Halenkovice Popovice Ostrožská Lhota Vnorovy Radějov Rohatec Mutěnice Mikulčice Lanžhot Hlohovec Hodslavice Zubří Valašská Bystřice Ublo Újezd Pozlovice Přečkovice Nezdenice Boršice u Blatnice Veřovice Prostrední Bečva Velké Karlovice Halenkov Lužná Študlov Návojná Bohuslavice nad Vláří Žitková Strání

757 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836

Javorník Bránice Neplachovice Hlavnice Domoradovice Větřkovice Lukavec Mokré Lazce Bohuslavice Vřesina Hlubočec Lubojaty Petřvald Pustějov Rybí Kozlovice Čeladná Trojanovice Velké Petrovice Křenovice Kobeřice Děhylov Petřkovice Stará Bělá Staříč Sedliště Vojkovice Janovice Morávka Staré Hamry Hať Dolní Lutyně Hradiště Oldřichovice Nýdek Bukovec Dolní Lomná

APPENDIX B – DIALECT MAPS

187

0

W

S

N

d1

E

d2

d3

d312

ÿ

d327

336

d223

d207

d248

d401

334

ÿd

d335

d25

d225

d403

d230

ÿ

ÿ

d26 d243

d242

d231

112

d233 d244

d232

d217

d216

d150

d214

d14

138

ji:skra

ÿd

154

221

ji:skra

12

ji:skra

157

130

501

159

161

160 ji:skra

ÿ

137

ÿd

ji:skra

d59

162

148

147 145 13 ji:skra

146

d135 ji:skra d134 d136

133

132 ji:skra

ji:skra

158

119 i:skra 118

129

ji:skra 131 ji:skra

ji:skra

144

143

d116

i:skra 117

i:skra

ji:skra

128 ji:skra

127

d

114 115 ji:skra

113 ji:skra

156 ji:skra

ji:skra

142

155

d10

141

ji:skra

126

125 ji:skra

ji:skra

153

124

d124

ji:skra

236

220

ji:skra

152

140

123

i:skra

219

235

234

23

218

151

ji:skra

139

ji:skra

d122

ji:skra

ÿd ÿd ÿd d d d ÿd ÿd ÿdÿ ÿd ÿd ÿd d d ÿd ÿdd d ÿd d d ÿd ÿd d d d d ÿ d d ÿd d d ÿd d d d ÿ d d ÿ d d ÿ ÿ ÿd ÿd d ÿd d d d d d ÿ d

ÿ

d102 d 104 d101d d103 d105 d108 ji:skra 110 d109 i:skra 120 121 d111d11 d

ÿd ÿd ÿd ÿd

i:skra 107 i:skra

d215

d149

d213

d241 d252

d22

d228d229

d250 d251

d249

d240

d224

d226

d227

ÿ ÿ

ÿ

ji:skra

ji:skra d204 d212 ji:skra d d21 210 d211

d239

d209

d237 d238 d247

i:skri

d328

326

d329 d330 d

d316

d317 d31

ÿd

ji:skra

d d32524

d205

206

208

d201d

ÿd

ji:skra

d202d203

d7

d106

d9

d650

d652 d651 d d 65 667 d666 d668 d654

d645 d648 d649

d91 d647 d646

d644

d94

d93

d92

d

ÿ

d819

d818

d805

d804

d810

d809

d d802 d 820 d d803 81d 808 807

d801

d95

d d d822830 831 d821 d84 d832 d82

Czech Regions

Roads

Rivers

For the names of AV dialect locales, refer to appendix a.

Legend

Major Dialect Boundaries

AV dialect survey locales

Lengthened /í/ in NS /*j'skra/

90

ÿ ÿÿÿÿ ÿ ÿ ÿ ÿÿ

d318

180

d412

ÿ

d40

270

ÿ ÿ ÿ ÿ

ÿ

ÿ

360

d245

27

222

450 Kilometers

d502

d736

d811 d812 d823 d825 d833 d701d806 d d813 d824 826 d253 d834 d402 d721 d506 d307 d d d406 d653 d83 702 505 d254 d d507 d319 d d d 827 d814d d404 ji:skra d337 d509 d508 676 246 504 d 815 d255 d828 53 d413 ji:skra d ji:skra d722 d663 d d738 ji:skra d405 d d d d d d d d320 331 d835 d ji:skra d d256 703 630d 816 836 655 677 71 d321 408 d d 656 d 669 631d d407 d d739d747817 d305 308 64d d616 d503 d513510 d680 d d409 d d66 670 657 d303 d617 d257 d704 723 ji:skra d748 d829 d301 d332 d512d511 d671 d665 662 d309ji:skra d410 d d i:skra d724 d672 d42 632 681d d633 d658 d d661 d414 ji:skra d302d304 d514 d51 d515 682 d32d d618 d673 d418 d619 ji:skra d415 ji:skra 664 322 ji:skra d420 d d749 d411 d605 d d d659 d634 d d333 d72 740 d678 d417 d684683 d516 d310 d416 706 i:skra d421 637 d635 ji:skra d750 ji:skra d705 d67 d620 d419 i:skra d422 d660 d431 d d d d d674 d d517 d d323 324 d638 d423 d621 622 d636 d427 d601 606 d679709 d685707 d725 d462 i:skra d43241 d425 d441 d d d52 d675 d741d751 d428 d429 d708 726 d63 d640 d623 d607 i:skra 442 d424 d711 d d463 728 d433 d d d d d d30 d430 426 443 444 43 d639 d742 752 d465 d434 d d d d641 d753 743 602 d438 d440 608 d624 d712d710 d727 d437 d518 d464 d445 d439 d715 d435d d446 d d 436 754 625 d713 d729 d44 d49 d d642d d714 d603 d451 d627d d447 519d d755 d449 d d d d716 73 d717 d610 520 628 d730 d74674 744 d453 d731 45 d609 d d626 d452 d611 62 d d d d629 d448 450 d604 d643 745 d734 d455 d733 d612 614d d756 d 615 d732 d d459 d720718 d454 d613 d61 d461 757 d456 d719 d d460 d79 735 d75 d457 d458 d737

d306

d315

d311 d313d314

d4

d5

d6

d8

Map 1: Lengthened /í/ in the nom. sing. /*j'skra/ (a.p. a) 'spark'

188

40

W

S

N

1d

E

0

3d

2d

8 d9

ÌÌÌÌÌÌ Ì Ì ÌÌ Ì Ì

d

d

d

Ì

Roads

Rivers

Major Dialect Boundaries

AV dialect survey locales

Shortening in the IS form /kravou/

7d

106d 107 d

40

80

120

d

40

160

d d 458

200

240

280 Kilometers

d

737736

Legend Ì ÌÌ Ì Ì ÌÌ Ì Ì ÌÌ For the names of AV Ì Ì dialect survey locales, refer to appendix a. Ì Ì ÌÌÌ Ì Ì Ì Ì Ì Ì Ì Ì Ì ÌÌ Ì Ì Ì Ì ÌÌ Ì Ì Ì Ì Ì Ì Ì Ì Ì Ì ÌÌ ÌÌÌ Ì Ì Ì Ì Ì Ì Ì Ì Ì Ì Ì Ì Ì ÌÌ Ì Ì ÌÌ Ì Ì ÌÌ Ì Ì Ì Ì ÌÌÌ Ì Ì Ì Ì Ì Ì Ì Ì Ì Ì Ì Ì ÌÌÌÌ Ì Ì Ì Ì Ì Ì Ì ÌÌ Ì Ì Ì Ì Ì Ì ÌÌ ÌÌÌ Ì Ì Ì Ì Ì ÌÌ Ì Ì Ì ÌÌ Ì Ì Ì ÌÌÌÌ ÌÌ Ì Ì ÌÌÌÌÌ Ì ÌÌÌ Ì Ì Ì Ì Ì Ì Ì ÌÌ Ì ÌÌ Ì Ì ÌÌ ÌÌÌ ÌÌ ÌÌ Ì Ì Ì ÌÌ Ì Ì Ì Ì Ì Ì ÌÌ Ì Ì Ì Ì Ì Ì ÌÌ Ì Ì Ì Ì Ì Ì ÌÌ Ì ÌÌÌ Ì ÌÌ Ì ÌÌÌ ÌÌ Ì ÌÌ Ì Ì ÌÌ Ì Ì Ì Ì Ì Ì ÌÌ Ì Ì ÌÌ Ì Ì ÌÌ Ì Ì Ì Ì

6d

Czech Regions 102d 104 d 101 108d d d d103105 d 120121 d110 11 112 d d 10d 113 116 5d 109 d111d d 149d d d 122 138d 124d d 114 d115d d117 212d 202d203d 204d 210 d 124 201d d119 14d d 211 127 21 d 125 118 139 123 d d d 128 d d208209 d d d 129 d 213 4d 126d 150 206d d 226d 227d d 207d 140d 214d d 225d 95d 205d 130d 151d 141d 228d229 d 94d 12 216d 131d d 215d 142d 143 133d 24 223d d 818d 152 224d 325d d d 801d 217d 218 d 153d 132 135d 22d 144d 231d 146 819d d 136 d 326d 237 238 d 23d 230d 312d d 147 134 d 644d d 240d 219d 14513 92d 247 232d 820d 137d 155d d d d 91 154 311d 802 d 242d d d d81d d 148 241d 314 327d 328334d248 830831 313 160d d 239d d 645d d d647 d 156d d d d 234 807 821d d d d 84 803 26d d 159 233d d 646 335d 401 249d 252 243d 804d 810 d 220d 221d 158d d 161d 162 dd 82 d d 316 93 809 d d d d d 832 d d 805 d d811 d 648 649650 329d d 306d 315d 59d 235d 157 250d 651 244 d d31 236d 222d 652d d 65667 336 27 d 501d 251 833d 701d d806 812d 823d825 25d 403d d330 d 317 d d d 668 d d 245 d 412d 402 502d d 834 d d d d 654 d 826 506d 307d 319 d318 702 d 406d 253d 254 d 653 824 d d 83 507 d 721 d815 d 666 676d 246d 504d 53 509 d d 255 404d d 508 d 827 d d 331d 337d 413d d d 722 d 663 669 d d 405 630 703 71 655 677 320 408d 835 d 256d d 321d d d d d d d738 d d 64 d 656 d d 828 d 407d d 510 305308d 836 d d670 680 409d d d d748829 257d 503d 513d d 511 617d 616 d 631 d d657 665662d 66 301d d d 309 d d 332d 410d 681 dd682704 d 723 739 d d d d d 42d d d 32 d 632 414 514 51 d 618d d d d 418d d661 d672 d d d d304 415d d322 420d 740 749 d d 411d d 515 d 512 619d 634d 633 d d664 333d 673 d 683 d d 72472d d750 d d 310d 302 416d 417d 421d d637 659 706 705 d d516 605 67 620 d 419d 422 d 431 622 324 684 635 d d d 517d 601 606d 41 d d660 674 323d d 638 d 423d 621d d 427d 685 d d 725d d d 462d d 432 d 425 441 751 d 52 d d 675 428 429d d d 442 d726 741 640 709 d d 607 623 d 63 424d d d d d d 463 d d d 679 d d 433d d 426d d 443444 752 708 43d 430 30d 639 465 d d 711 710 434d 743d742d d 602d 608 d 641d d 438d 440d 624 437d d 464d d 518 445d d727 d753 715 d d d 625 435436 d d d d d 439 44d d446 d d 49 d 519520 603 642d73 d 712 d d 744 451d 627628 755d 754 d 74 d d d d d d d 713 610d d447 449 d 45 626 453 d d 609 d 452 611 614 d 62d d 629 d 716 d d d 745 448 d d 604d 643d 717 d 756 d d 450455 d d 731 d 720d d718 d d 732d 757 612 d 61d d615 454d d456459 d 460 461 d d d 733 79 d d d d 613 d 75 719 735 457

d

Map 2: Short /a/ in the instr. sing. of /*korva/ (a.p. a) 'cow'

189

40

N

0

1d

3d

40

d

2

8

d

d9

d

E Rivers

Legend

Major Dialect Boundaries

AV dialect survey locales

Long /a:/ in the IS form /kra:vou/

80

120

6d

7d

160

200

240

280

320 Kilometers

106d 102d 104 107 d106 101 d d 108 d 107108 103105 d d 109 d 120121 120121 d110 11 112 d d 10d 113 116 5d d d d 149 109 149d d d 122 111 122 138 204 138 212 124 203 204 d 202 114 d115d d117 212 203 202 d 210 d 210 d 124 d d d d 201 211 127 201 14d d 211 127 118d119 139 123 d 125 118 139 209 21 d d d 209 d d 128 d d 208 d 213 d 129 d 150 206 4 126 150 227 213 206 d 226 227 d d d 226 207 214 d d 207 d 140 214 d d 225 d d 205 225 95 151 130d 205 151d 141 229 d d 229 d 228 d 228d d 94d 216 12 216 131d d d 215 143 215d 142d 223 133d 24 d 818d 224 d 152 224d 325 d 325 d d 801d 217 153 d 218 153 217 132 135d 22d 144 218 d d 231d d 146 d 819d 136 d 326 326 237 238 d 23d 219 312 d 230d 312 d d 240 d 14513 d147d 134 137 644d 240 219d d 92d 247 232 820d d d 91 154155d 802 311 d d d 327 81 242 d d d 313314 327 d d d 148 241 334 d 314 830831 313 d 160d d d 328 239 328334d248 d647 d 645 d 156 d d d d 803 d807 821d d d d 84 d 26 d 335 d 159 315 233 234d 646 335 249 252 d243 804d 810 d 220d 221d d 158 d 161d 162 dd 82 d 316 d 316 d 401 93 d d 809 d d d d d 832 d d d 329 805 329 d 648 d649650 d 306 315 59d 235d 811 d 157 250 651 d d d 330 d 336 222 244 d 652651 d 236 222 652 667 31 d 336 823 812 27 65 825 833 501 d 667 701 251 d d 701 d d 25d 403d d d 317 d d666 d 668 d d 330 d d d 668 d d 806 245 d 412 402 502 654 721 d318 d 506 d d 834 d 654 d 826 506 505 d 307 319 702702 653 406 253d 254 d 254 d 319 653 824 d721 83 d 507 d d 507 d 676 d 504d 666 d 676 504 246 d 337 509 508 815 d d d827 d 404 d 508 509 255 d 53 d 331 337 722738d d 413d 669677 d 677 331 d d d 405 722 d 703 d 630 630 631 655 663 669 655 656 d d 703 71 656 320 408 835 d d d d 256 828 d 747 321 d d d d d d d 680 510 d 407 d 510 d 64 d d 616 723 738 d d d d 670 305308 d 836 d 680 704 617 616 d 409 d 657 511 662 66 d670 704 d d 617 748 829 257 503d 513d d 511 332 d d 631632 d657 301301 332 d 665 d d671 d 309 410 724 739 748 d 681681 d682 d 682 d 723 d dd d d d d 633 658 d 42 d d 32 d632 618 414 d d 51 514 d 618 d672 d d d 322 418 619 512 d 740 d d d d304 415 d d 749 d d d 740 749 d 661 d 683 d 411 d 633 333 420 d 637 d d 634 333 684 d 706 72472 310 515 d 673 678 d d 310 416 417d 302 635 421 d706 750 d d d67 683 d637 d659 664 d d516 605 d d d 422 705705d 750 620 419 622 d635 d 431 622 684707 d 636 d d d 725 638 d 674 709 517 601 606d 41 d 323324 725 638 d 685 d 423 d 660 621 427 d 685 d d d 462 d 441 d d 751 432 d 425 d d d 751 52 d d d d 675 741 708 442 d 428 429 d 726 741 d d 711 709 640 675 d 607 62362363 728 d d726 424 d d d d d d 463 d 679 d d d d 752 d 433 d 426 d 443444 708 43 430 639 30 742 752 d 639 d 742 d 753 d d 465 743 d 711712710 434 743 d 608 d 641 710727 d d 518 602 d 624 d d d 437 438d 440d d d d 464 445 d 715 d d 625 727 d 754 d753 435436 d 625 d d d d 729 d d 439 712 44 d446 d d 49 642 627 754 603 519 451 744 d 627628 d d74744 755 d d73 d 717d716 d d 520 d 610 626 d 628 d447 449 d 45 d d 731730 d d d d d 713 d 452 453d 611 d614 609d 62d626 d 745 629 716 d d d746745 629 448 450 d d 604 643 717734 733 d450455 d 756 d d d 718 731 d d d 732 d 459 720 d d 732 d d d 720 757 61 612 615 757 718 735 454 456 d 461 d d 719 d d d d d 460 733 79 d d719 d d 613 d 735 457 75 457 737d d d d 458 736 737736 d 40

EE EE EE E E EEEEE E E EE E E For the names of Av E E dialect survey locales, E E E E refer to appendix a. E E E E EE E E E E E E E E E E E E EEE E E EE EE E EE EEE EE E E E E E EE E E EEEEEEEE E EE E EEEE EE E E EE E E EE EEEEEE E E E E E E E E E E EEEE E E EE EEEEEE EEEE EE E E E EEE EEE EE EEEEE EE EE EEE E EEE EEE E EE E E E

d

Map 3: Long /a:/ in the instr. sing. of /*korva/ (a.p. a) 'with cow'

190

40

W

S

N

d

E

Ý1

0

Ý3

Ý2

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

Ý106

d

Ý9

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

Ý

Ú

d

d

d

d

d

d

d

d

d

d

d

d

Rivers

d

d

d

d

d

d

d

d

d

d

d

Legend

Major Dialect Boundaries

Long /a:/ in NS of /*solma/

Short /a/ in NS of /*solma/

d

d

40

d

d

d

d

d

d

d

d

d

80

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

120

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

160

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

Ý40 200

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

240

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

280

d

d

d

d

d

d

d

d

d

d

d

d

d

320

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

360 Kilometers

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

Ý736

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

d

For the names of AV dialect survey locales, refer to appendix a.

Ú Ý 102 Ý7 Ý103104 Ú Ý6 Ú 107 Ú Ú 101 Ú 110 108 Ý Ú Ý 105 ÚÚ 120121 109 Ý10 Ú Ý5 11111 Ý112 Ý Ý149 113Ú 116 Ý Ú Ý 122 Ú Ú Ú Ý 117 Ú Ú Ý202 138 204Ý 123Ý 124 Ý119 114115 212 203 Ú Ý14 Ý 210Ú211 124 Ý125 Ý127 Ú Ý139 201 Ý128 Ý118 208 Ý209 21 Ú Ú Ý4 Ý126 Ý Ú Ú 213 129 Ú 206 Ú Ú Ý214150 226 227 140 Ú Ý205 207 Ý225 Ý130 Ú Ú 95 Ý229 Ý228 Ú Ý141 Ý216 151 Ú 94 12Ý 131Ú Ý Ú Ý Ý24 223 215 142 143 Ú Ý 133 Ú Ý224 Ý325 818 132 Ú Ú Ý217 Ú 152 Ý153 Ý22 Ú 801 135 Ú Ú 144 Ý 218 Ú Ý 134 231 Ý Ú 146 819 Ý Ý 136 Ú Ý312 326 237 238 Ý147 Ý644 23Ý Ý 230 Ý Ý240 Ý Ú Ý Ú Ú 13 219 145 Ý 92 Ú Ý154155 Ý311 Ú 247 820 232 137 Ú Ú Ý 91 802 Ú Ý241 242 808 Ý647 Ú Ý334 Ú Ú Ý646 Ú Ú Ý Ý148 Ú Ý248 Ú239 80381 807 Ú Ý645 830 Ú Ú 313314 327Ú 831 328 Ý26Ý 822 Ú 821 Ý233 Ý234 Ú Ú221 Ý156Ý Ú159Ý 160 Ú Ý Ú Ý315 Ý 84 Ý Ú 832 648 809 Ú Ý649 Ý 335Ý401 249 Ý252 243 Ú Ý316 Ú 804 220 161 Ý 82 157158 93 810 Ú 162 Ý 650 Ý306 Ú317 Ý Ú 805 Ú Ý250 Ý31 329 59 811Ú Ú Ú Ý Ý222 Ý652Ý651Ý65Ý Ú Ý Ý235 244 Ú Ý501 Ý Ú Ý701 Ý Ý 812 823 27 236 825 Ú Ú 667 806 833 Ý318 330 336 Ú Ý668 Ý666 Ý245 Ú Ý412 Ý 25 403 251 Ý Ý 813 Ú Ý 824 826 Ý Ý 502 Ý307 Ý Ý Ý 834 253 Ú 654 402 Ý 721 506 Ú Ý Ú Ý 702 505 406 Ý 653 Ý Ý ÚÚ 83 254 827 319 Ý 507 Ý Ú Ý Ý 676 Ý Ú 50453 509 246 814 Ý 815 404 Ú 337 Ý 508 Ú Ý Ý 255 828 Ý405 Ý Ý Ý 413 Ú Ú Ý Ý Ý320331 Ú Ú Ý 722 Ú 738 663 Ú 669 Ý 630 703 71 Ú Ú 816 836 835 Ý747817 Ý407 408 Ý64Ý655 656 Ý Ý Ý670 677 Ý 631 256 Ý 308 321 Ý Ý ÝÝ510 Ý723 Ý739 Ú Ý617 Ý616Ú 657 Ý 662 Ý Ú ÝÚ Ý748 66 680 Ý704 Ú 409 Ý671 303 305 Ý Ý410 Ý257 Ý503 513 Ý681 829 Ý724 332 512511 Ý672 Ý Ý Ý 665 309 Ú 632 Ý Ý Ý Ý301 Ý Ý Ý Ý Ý 658 42 682 515 Ý Ý418 Ý619 618 661 Ý 414 Ú 633 302304 32322 Ý411 51Ý 514Ý605 415Ú Ú Ý634Ý637 Ý659 664 673 Ú678 Ý Ý683 Ý333 Ý420 Ú Ý Ý740 Ý749 Ú Ý310 Ý635 684 706 Ý 72 Ý750 516 416 417 421 Ý Ý Ý Ú Ý Ý Ý Ú 419 Ý517Ý Ý606 620 Ý 622 Ý636 Ý638 660 Ý674Ý Ý67 Ý Ý707Ý725 705 Ý 422 Ý Ú Ý427 323324 709 Ú 423 Ý425 Ý441Ý 431Ý43241 621 685Ý726 Ý Ý751 601Ý52 Ý428Ý429 Ý Ý462 Ý63 Ý640 Ý675Ý 679 Ý623 Ý607 Ý424 442 741 711 Ú 728 708 463 Ú Ý444 Ú433 Ý443 Ý Ý30 Ý Ý639 Ý 426 Ý742ÝÝ752 Ú 43 430 Ú Ý743 Ý602 Ý608 Ú Ý 465Ý Ú Ú Ý Ý Ú 641 710 Ý Ý445 Ú 434 727 753 440 Ú 438 Ý715 712 624 437 ÚÚ 464 518 Ú Ý625 439 436 Ý714 Ý713 754 Ý729Ý Ý44Ý446 435 Ý49 Ý519Ý Ý642 Ý603 Ý451 Ý Ý Ú Ý Ý Ý Ý 73716 ÝÚ Ý717 Ý610 Ý Ý Ý 627628 744 447 520 755 Ý730 449Ý45 Ú Ý 74674Ý 731 60962626 Ý 452 453 Ý448Ú450 Ý604 Ý611 ÝÝ643 Ý734 745 629 Ú Ý Ú Ý612Ý614 Ý 455Ú 733 Ý Ú 756 Ý 718 Ú Ý613Ý61 615 732 757 459 Ú Ú 454 Ý719 Ý735 Ý79 720 456 Ý460 461 Ý ÝÚ Ý737 75 457 458

d

Ý8

Map 8: Distribution of Length & Brevity in the nom. sing. of /*solma/ (a.p. a) 'straw'

195

50

W

S

N

0

E

d

1

50

d

2

d

3

8

d

106 107 d

d9

d

ß

100

150

200

d

6

d

250

d

7

300

120

350 Kilometers

For the names of AV dialect survey locales, refer to appendix a.

Legend

Reflex of Long Vowel in NS /*repa/ AV dialect survey locales Major Dialect Boundaries Rivers Roads

102 104 101 d d 108 d d 105 d d d 110 11 112 10 113 5 d d 121 d 116 d d d 149 109 d d d 117 d 122 111 124 d 138 114 212 202203 204 210 d d d d d 124 d 125 127 d d 115119 d 201 d d 211 14 d 139 123 d d d d 208 209 d 21 d d d d 128 d 213 d 126 4 d 129 d 118 150 206 227 d d 140 d d 214 d d 207 d 226 d 225 95 d d 130 205 d 151 141 d d 228229 d d 94 d 12 216 d 131 d d d d 143 142 215 d d 133 24 223 818 152 224 d d d d d 325 d 801 135 d d d 132 153 217 218 d 22 144 d 231 d 136 d 146 819 d d d d d 326 237 d 23 238 d 230 d 147 312 d d 644 d 13 134 d 145 240 219 92 d 247 d 232 d 137 d 91 154155 80281 820d 311 d d d d d d d d 242 d d 148 327 241 d d d d 314 d 334248 830 831 160 d 239 d d d 645 328 647 d d 156 d d d d 807 821 d d 84 26 d d d d 233 234 d d 646 d 335 d d 158 159 d d 82 803 249 252 d 243 220 221 d 313 316 d 93 d d dd d 649650 161 d d 401 d d d 810 d 809 d d 804 d d d 832 d d d d d 811 648 d 59 306 315 d 329 235 250 157 d d 244 162 236 222 d d d 31 d 336 d d d 823 812 27 d 65 825 501 667 251 833 701 d d 25 403 d d 317 d d 330 d d d d d 668 d d 245 d d 412 402 d d d 806 502 d d 826 d 834 652 651 506 d 307 319 d 318 702 d d 824 406 253 d 653 d 654 666 d d d 254 d d d d 721 83 676 246 504 d 53 d 509 507508 d d 404 d 337 d d 255 828 d d d 413 d 331 d d d d d d 405 703 d 655656 663 669 677 d 71 d 408 d 630 d d 836835 256 d 321320 308 d d d d 407 d 64 510 d d d d d d d d d 409 d 305 d 816 d d 66 d 670 680 d d d 617 616 631 257 503 d d 511 d d d d d 657 665662 301 332 d 309 410 d 681 d 682 d d 723 d 739 d d 829 dd d d 51 d 42 d dd 414 d d d d d d 32322 d 418 d 619 618 d d 704 d 415 d 748 d d 633 d d d 661 d d 672 d d 304 411 d d d 514 512 d d 333 420 d 632 634 d d 72 d 416 417 515 d 605 683 310 673 d d 740 d 749 302 421 d d d d 67 d d d 705 d d 637 d 659 d 620 419 d d d d 422 431 622 d d 750 517 516 323324 638 d 660 674 d 684 706 d d d 621 d 635 423 427 d 432 d d 462 d d d 725 425 d 751 52 d 606 d d d 441 d 640 41 d d 675 741 d 428429 d 685 d d d 63 623 d 424 d d d d 442 d d 463 d d d 601 679 d d d 708 433 d 426 dd 752 d 607 d 43 430 d d 30 d 465 639 641 434 743742 d d 711 710 602608 d 438 d d d d d 624 d 464 d 518 d d 435436 437 d d d 753 443 445 d 715 d d d d 625 d d d 727 d d d d 44 d d d d 49 439 d 642 603 519 712 446 451 d 627 744 755 d 754 610 d d d 520 d d d 713 d d 74 d d d d 628 453 d 447 449 d 45 d 452 d d d 611 d 614 609 62626 d d d 745 d 448 604 716 d dd 643 717 d d d 629 d 455 d 731 d d d d d d 720 dd 718 d d 733 612 d 61 d 615 d 756 450 d 456459 461 d 732 d 757 d 460 d d d 79 d d 613 454 d 457 d 75 d 719 d 735 d d d d 458 d 737736 40 d

ßß ß ßß ß ßßßßßßß ß ßß ßß ßß ß ß ßßß ßß ßßß ßß ß ß ß ßß ß ß ßßßß ß ß ßßßßß ß ß ßßßß ß ß ß ßßßßßßß ß ßßß ß ß ß ß ß ß ßß ßßßß ß ß ß ß ß ßßßßßßßßßß ßß ß ßß ß ßß ßßß ß ß ßßßß ß ßß ß ß ß ß ß ßßßßß ß ßß ß ßß ß ßßßßß ßß ßßß ß ßßßßß ßß ßß ßßß ß

d

Map 7: Reflex of Retained Long Vowel in the nom. sing. /*repa/ (a.p. a) 'beet'

194

90

W

S

N

1

d

E

d

8

&

&&

d

9

& & & & & & & & &

&

&

d

&

&

Rivers

Legend

For the names of AV dialect survey locales, refer to appendix a.

Major Dialect Boundaries

AV dialect survey locales

Length in the IS of /bri:za/ (a.p. a) 'birch'

& &

&

d

7

&

&

0

90

180

d

270

360 Kilometers

& & & & & & & & & & & && & & & & & & & & & & && & & && & & & & && & & &

&

&

&

d

6

&&&

106 102104 107d 101 d d d 108 d d d103105 d 120121 d 110 11 d 112 d d 10 113 116 5 d d d 109 d111 d 149 d d d 138122d 124 204 114 115d117 212 202203 d d 210 d 124 d d d d 201 d d 211 14 125 127 21 d d d119 118 d 139 123 d d 209 d d d 128 d 208 d d 213 d 129d 4 126 150 206 d d 226 227 d d d 140 d 207 214 d d d 225 d 95 205 d 130 151 141 d d d d 228229 d12 94 216 d 131 d d d 3 d 143 d 215 d 133 142 223 d 24 818 152 d d d d 224 d 325 d d 801 d 153 217 218 d d 144 146d132 135 22 d d 231 d d 819 d d 136 d d 326 237 d 238 23 d147 d d 230 312 644 d d 134 d 219 14513 d 240 92 d 247 d 232 820 137 d 91 155 154 d d d d 311 d 802 d d d d 81 d 242 d 148 d d 241 327 d d 334 314 d 830831 d d 313d 160 d 239 d 328 d 248 d 645d d647 d 156 d d 803 d807 821 d d 84 d 26 d 646 d 159 d d 233 234 d 335401 249 252 d 243 804 810 d809 d d82 220 221 158 d 161 162 d d dd d d 316 93 d 2 d d d d d d d d 832 d d 805 d811 d 648d649650 d d 59 306 315 d 329 235 d 157 250 d d 651 244 d d d31 d d27 236 d 222 d 652 d 65667 823825 833 501 d 701 d806 812 25 403 251 d d 317 d330336 d d d d d d d d d d d 668 245 d 412 402 d d d826 502 834 d d 506 d d d824 307 319 d318 702 d d 406 253 653654 d d d 254 d721 83 d d 507 666 d d d 676 246 504 d 815 d 53 404 d 337 413 509 d 508 d d827 d 255 d d d 331 d d 722 663 d d d d 669677 71 d 630 405 703 655 320 d 408 d828 835 d 656 256 d d 321 d738 d d d d d 64d d d 407 d 409 d d d 305308 d 513510 66 d670 d d 680 704d d d 617 616d 631 d d d657 257 503 829 836 301d d d 309d 332 d d 511 d 665662 410 d d d 681 d682 d d723 d739 748 d d d d 42 d d d d d d 632 414 415 d 32322 d 514 51 d 618 d d d 418 619 d 512 d d d d d661 672 d d d304 420 d d 749 411 d 515 d d673 d 72472 740 634 d633 d659d664 d d d 683 310 d 333 302 416 417 d d605 421 d d 750 d d67 d d d706705 d d d d d516 d 620 419 d 422 d635 637 431 d 622 324 684 606 674 d d 517 41 d d 621 d d 323 d 638 d660 d 423 425 427 685 d 725 d 462 d 601 d d d d d 441442 d 432 751 d d d 428429 d 52 d d d d 640 675 623 63 607 d d726 741 424 d d 463d d d d d d d d 679 709 d d d d d 426 d 752 443444 433 708 d 43 d 30 430 639 742 465 d 711 d d 434 d d d 743 602 608 641 710 d d d 438 440 624 d d d 464d 518 d 445 d d d727 715 d753 d 435436 437 d d d d 625 d d d d d439 d d d 44 d446 d d 712 49 642 754 603 519520 627 73 744 755 449 45 451 d d dd d d d 610 d d d628 713d d74 d447 d d d d d d d 62626 629 d 452 453 d d 716 d dd 745 614 609 448 d450 604 611 d d d 643 d d 717 455 731 d 756 d d d d d d 732 d 720 d718 d 612d 61 d615 757 d 454 456459 d 460 461 d d d d d 79 d d719 d 733 d d 613 d 735 75 457 d d d d d 458 737736 d 40

Map 6: Long /í/ in the instr. sing. of /*berza/ (a.p. a) 'birch'

193

W

S

N

E

0

1#

3#

2#

106 # 107 #

9

#

ÞÞ Þ Þ Þ Þ Þ

Þ

Czech Regions

Roads

Rivers

50

100

150

200

250

300

350

400 Kilometers

Þ Þ Þ Þ Þ Þ Þ ÞÞÞ Þ ÞÞÞÞ Þ Þ Þ Þ Þ Þ ÞÞ Þ Þ Þ Þ Þ Þ Þ Þ ÞÞÞ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ ÞÞ ÞÞ Þ Þ Þ Þ Þ Þ ÞÞ Þ Þ Þ Þ Þ Þ Þ Þ ÞÞ Þ Þ Þ Þ Þ Þ Þ ÞÞ Þ Þ Þ Þ Þ ÞÞ Þ Þ Þ Þ Þ Þ ÞÞ Þ Þ Þ Þ Þ ÞÞ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ ÞÞ ÞÞ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ ÞÞ Þ Þ Þ Þ Þ ÞÞ Þ Þ Þ Þ ÞÞ Þ Þ Þ ÞÞ Þ Þ ÞÞ ÞÞ ÞÞÞ Þ ÞÞ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ Þ ÞÞÞ Þ Þ Þ ÞÞ Þ Þ Þ

For the names of av dialect survey locales, refer to appendix a.

Legend

AV dialect survey locales

#

Major Dialect Boundaries

Shortened acute long vowel in NS /*berza/

Þ

102 # 104 # 7# 6# 101 # # #103105 108 # # 120121 #110 11# 112 # # 10# 113 116 5# # # # # 109 111 149 # 122 # # 138 # 114 #115 212 # # 210 # 124 202#203 # 117# # # 204 124 # # 201 211 14# 125 # # 209 21 # # 127 #128 118# 119 # 139 # 123 # 129 208 # 213 126 4# # 150 # 206 # # # 207 # 226 # 227 140 214# # # 225 95# 205 130 # 151 141 # # # # 229 # 228 94# # 12 216 # 131# # 143 142 215 223 133 # # # # # 24 818# 152 224 # # # 325# # 801 # # 153 217 132 135 22# 144 218 # # # # 231 # 136 146 819 # # # # 326 237 23 238 # # 230 312 # # 147 # # 644 219 # 14513 240 92# 247 # # # 134 137 # # 232 820 # 155 # # 154 91 # 311 802 # # # # # 81# # 242 148 327 241 # # # # 334 314 830# 831# 313 # # 160# 239 #248 # # # # 645 328 # 156 # # # 647 # 234 803 807 821## 26# 84 # 159 233 221 646 # # # 335 ## 249 243 # 804# 810 # 220 # 316 161 # 401 # 252 # # # 162 # 93# #158 #82 # 650 # # # # 809 # 832 649 805 # 648 # 811 306 59# 235 # 250 157 # 315# # 329 # # # 651 244 222 236 # 652 31 # 336 # 501 # # 65#667 251 833# #825 # 823 # #27 # # # # 701 # #806 812 25# 403 317 # # 330 # 245# # # # # 668 502 412 # 826 253 834# # 402 # 654 318 # 307 # # # 702 # 406 # 319 653 # 506# # 254 507 # 721 83 #815824# 666 # 676 # # 508 504 246 53 # 404 337 # # 509 # # # # 255 827# # # 413 # # 331 722 # 663 669 # 405 # 630 71 703 655 677 # # 320 835 # 408# # # 656 # # # # # 256 738 # # # 828 #836 # # # 407 64# ## # 510 # 321 # 305308 # 616 # 503 66 # 680 409 # 513# # 511 617 # 704 257 # # 301# # # 309# 332 # 631# 657 665#662# # # 670 # #723 739 748 # # 829 410 # # # 681 # 682 # 42# # # 632 414 # # # 514 51 32#322 618 # # # # #304 # # 418 # 512 619 672 415 # # # 420 # 749# 411 # 633 # # 661 #673 # # 724 740 # 634 333 515 417 # # 310 # 302 416 # 683 # 421 605 # # #706 72# 750 # # # # 422 # 637 659 664 # 705 # 67 516 620 419 # # # 431 # 622 324 684 # 635 606 # 674 # 517 # 41 323 # 638 725 # # # 423 # # 621 # 441 427 # 660 601 # # 462 # # 52# # # 685 # # # 432 425 # 751 # # # 442 675 741 428 # 640 623# 63 # # # # 429 # 607 424 # # # 463 # #679 709 # #708726 # 426 752 443444 43# 430 # # # # 433 # 30# 639 465 742 # 711 434 # 743 602 # 641 710 # # # # # 608 # # 438 624 437 # 440# 464 # # # #727 445 # # 715 # 518 # # # 625 # # # 435436 # 753 # 439 # 44# 446 712 642 49 754 603 # 519 451 # 744 627 # #520 #73 # # # 628 755# # 449 # 45# # # 610 # # # 713# ##74# # # 626 453 447 # 609 452 611 # 62# # 629 # # 614 745 716 448 604 643 # 756 # # #450455 # # # 717 # # 731 # # # # 732# 757 459 720 612 # 61# 615 # #718 454 # # 456 # 460 461 # # # # #733 79# 613 735 75 719 457 # ## # 458 737736 # 40#

8

#

Map 5: Shortened /e/ in the nom. sing. of /*berza/ (a.p. a) 'birch'

192

0

W

S

N

50

E

gd

1

9

gd

gd

Þ

Legend

Shortened Acute /i/ in /*griva/ AV dialect survey locales Major Dialect Boundaries

3

2

100

150

200

250

300

350 Kilometers

102 104 7 6 101 103105 108 120121 109110 11 112 10 5 116 113 149 111 138 122 124 114115 117 212 202203 204 210 119 124 14 201 211 125 127 21 139 123 128 118 208 209 213 129 4 126 150 206 227 226 207 140 214 225 95 205 130 151 141 228229 94 12 216 131 142 143 215 223 133 24 818 152 224 325 801 135 132 153 217 144 22 218 134 231 146 819 136 326 237 238 23 230 312 644 240 219 145 147 92 247 232 137 154 155 91 311 80281 820 13148 242 241 327 647 334 248 830 831 313314 239 645 328 156 803 807 821 26 84 159 160 646 233 234 315 335 249 243 804 220 221 82 157158 252 401 93 316 810 809 161162 832 805 648 649650 329 306 317 59 235 811 250 244 222 652 651 65 667 31 812 823825 336 27 236 501 251 833 701 403 25 826 330 318 806 668 245 412 402 502 834 253 654 307 505 506 702 813 721 83 406 653 254 827 319 507 666 676 504 246 815 53 509 404 337 508 255 828 413 663 669 405 630 631 655656 703722 71 816 836 835 320 331 408 256 738 308 321 747 407 64 305 503 513510511 409 662 66 670 677 680 704 723 617 616 257 657 301 748 829 332 410 309 632 724 739 671 682 42 304 633 658 414 514 512 51 32322 618 418 619 415 672 661 420 740 749 411 681 683 515 634 333 72 302 678 637 417 673 310 664 416 605 421 659 750 516 67 620 419 422 684 706 705 431 622 635636 638 660 606 674 517 41 323 324 707 709 685 423 621 427 601 462 432 425 441 442 725 741751 52 675 428 429 640 623 63 607 424 711 679728 726 463 708 426 443444 433 752 43 430 30 639 465 434 743742 753 602 608 641 438 440 624 712710 727 518 437 464 445 715 435436 754 625 439 713 44 446 49519 642 714 603 451 627 628 520 755 449 45 716 74744 610 453 447 730 609 62626 452 611 745 614 717 629 448 604 643 734 731 450455 756 612 732 459 720 718 615 757 454 613 61 461 456 460 719 733 79 735 75 457 737 458 736 40

106 107

ggd g gdggd gd gd d gd d gdd Þgd gdgd gd gd gd gd g gd For the names of AV g d gd gd gd d ggdd gdgdgdg g dialect survey locales, g d g g d g d d d g g d d gdgd gd gd d gd gd g refer to appendix a. g d g d gd Þ gd gd gd g d gd gd d gd gd gdgd g d g gd ggd gd gd gd gd d gdg gd d g d g d d gd g g g g g g gd g d gd g gd gd d d g gdd Þgd gdgd d Þgd Þd Þgd gd gd d d d g g d d g g g g d d gd g d gd g g d gd gdgdgd gd gd gdgdg gd ggd d d ÞdgdgdÞÞggdd Þgd Þgd g gd Þgd gdgdgÞdgd gd gd gd g Þg d gd d gd gd gd gd d g d g g d g d g g d gd gd gd ggd Þgd Þgd ÞgdÞgdgd Þ Þgdgd gd d gd gd d gd gd d gd ÞgdÞgd g d gd Þ g gd d Þd Þd Þgd gd gdgd ggd g gd gd gdgd gd g g Þgd d d g gd gd gdgd Þgd Þgd gdgdÞgdÞgd Þgd ÞgdÞgd Þgg g g d dd Þd gÞdÞgdÞgdÞgd Þgd Þgd g d Þ g d d g gd Þ g d gd gd d g g d Þ Þ Þ g g d d Þ Þ g d Þd d gd g g g gd gd g Þgd g g gd d gd ÞdgdÞgdd Þgd gÞgd ggdd gd gdÞgÞdgd Þgd gÞgdgd ÞgÞdgd ÞÞgd gd gd ÞgdgdÞgdÞgdÞÞgdd ÞgdÞdÞgdÞgdÞgd Þgd Þgd gd gd gd d gd d gd gd gd gd gd g d g d g g Þ g Þ Þ Þ g Þd g gdgd gd gd g gd gg d gd Þd gd gd d g gd gd Þgd gd ÞgdÞgd ÞÞgdd Þ Þgd gd gdgd ggd ÞÞggdd ÞgdÞgdÞgd ÞdÞgdÞgd g ggdgdgdgdÞd Þgdg Þgd Þd ÞgdÞd gd d d gd gdg gd g gd gd gd g Þ d d d Þ d d g Þ g d gd gdgd g d gdgd d gd gd gd gd gd gd Þgd Þgd Þgd gdÞgd gÞgd Þgd Þgd ggdg Þgd gÞgd gdÞgd gdgÞdgd ÞgdgÞgdÞgdÞgd ÞgdgdÞgdÞd d gg gd g Þ Þ d g Þ d d d g d g d Þ Þd Þd gd g gd g d gd dd g gdÞgd Þgd ÞgdÞgdgdÞÞdgd gd gdgd gd gd gd gd gdgd d ÞgdÞgd Þgd Þgd Þgd Þ Þgd Þgd Þd gd g d Þ Þ gdgd gd g gd gd Þgd Þgd Þgd gd d ÞgdÞgdgdÞÞggdd gÞgÞdgdÞgg Þgd Þgd Þd gdÞgdÞgd Þgd d Þ gd gd d gd gd gd gd ÞgdÞgdÞ ÞgdÞgdÞgd gdÞgd Þgd g g Þ d Þgd gÞgd Þd Þgd g gd gd gd gd gd Þgd gd gd Þd Þgdgd Þgd Þd gd gdgd Þgd Þgd gd

8

gd

Map 4: Reflex of Short Vowel in the nom. sing. /*griva/ (a.p. a) 'mane'

191

30

W

2

3

S

N

0

E

305 301303 302 304

321

ku:pa

319

30

310

30

ku:pka328

327

326

325

320

323

60

332

202 203

90

402

204

210

445

120

44

447

434

414

239

211

150

448

212

gd

149

14

138

122

gd gd gd gd gd gd Á"gd Á"gd Á"gd gd g Á"d

424

11

112

gd gd gd gd g gd d

111

454

180

419

ku:pa

410

ku:pa

418

461

453

210 Kilometers

460

452

459

ku:pa

457

456

455

45

427

ku:pa

49

464

463

519

465

462

ku:pa

411

ku:pa

125

gd

10

516

520

518

113

604

158

144

117

116

146

159

161

160

162

148

147

132

130

131

133

119

118

129

145 13

501

128

114 115

134

59

603

613

611

608

607

61

610

612

606

52

605

614

626

624

625

634 635 622

618

62

621

609

615

620

619

ku:pa 502 506 505 507 ku:pa 504 ku:pa 53 509 508 ku:pa 630 510 513 616 617 512 511

601

602

127

143

157

156

ku:pa

222

142

12

141

126

ku:pa 515 514

503

517

ku:pa 51

155

ku:pa

221

154

246

124

153

124

ku:pa

236

ku:pa

220

ku:pa 257

256

ku:pa

245

27ku:pa

235

ku:pa

234

140

123

152

ku:pa

23 ku:pa 219

218

151

139

428 429 ku:pa ku:pa ku:pa 426 43 430 ku:pa 440 438 ku:pa 437 439 ku:pa 451

425

ku:pa

417

42

ku:pa

244

ku:pa

233

ku:pa

232

ku:pa

217

ku:pa

216

ku:pa 255 ku:pa ku:pa 409

254

ku:pa

243

150

ku:pa

214

ku:pa

242

ku:pa

231

215

408

ku:pa

253

407

252

230

213

ku:pa 26

241

ku:pa

416

450

449

435 436

423

415

22

229

227

228

406

251

250

249

240

405

224

226

21

225

403

404

209

446

422

25

238

ku:pje 433

41

421

413

ku:pje432

431

420

ku:pa

337

248

401 ku:pka

335

207

223

247

237

334

ku:pa 412

24

205

206

201 208

ku:pje 443 444

442

333

441

324

ku:pa

ku:pa

331

ku:pa

318

317 31 ku:pka 329 ku:pa ku:pa 330 336

ku:pa ku:pa ku:pa 309 32 322 ku:pa

308

307

ku:pa

306

ku:pka

ku:pka 313 314 ku:pka ku:pka 315 316 ku:pka

311

312

4

5

gd gd gd gd gd gd gd gd gd gd gd gd gd gd g gd gd gd gd g d d gd gd g d g gd d gd g g d g d gd Á"gd gd gd gd d gd gd gd gd g d gd gd gd gd gd gd gd gd gd gd gd g gd gd d gd gd Á"gd gd gd gd g d g Á"gd d gd gd gd gd Á"gd gd gd Á"gd g Á"gd g " d g Á " d Á g g " d gd gd gd gd Á"d Ád gd g d g g " d Á Á"gd gd gd d Á"gd Á"gd Á"gd Á"gd gd Á"gd Á"gd gd gd gd gd gd gd gd gd Á"gd Á"gd gd g Á"gd Á"ggd Á"gd Á"gd Á"gd "d Á gd Á g "gd d " Á d g g d d "gd Á"gd gd gd gd gd Á"gd Á gd Á"gd Á"gd Á"gdg Á"gd gd g g g " " d d Á Á g g g d " d d g Á d d gd Á"d Á"gd gd gd Á"gd gd Á"gd "gd Á"gd Á"gd Á"gd g gd ggd gd g g d " d d Á g g g Á d " d gd d Á g d Á"gd Á"gd gdg d gd gd gd gd gd gd gd gd gd gd d "gd Á g d g g " d d Á gd g gd g g " d Á gd g d " Ád g gd gd gd d gd gd gd gd gdg Á"gd "g gd d g g g d d " d Á g d gd gd Ád d Á"gd Á"gd Á"gd gd Á"gd gd gd gd gd g gd Á"gd "gd Á gd g gd gdgd gd gd d gd gd gd gd gd d gd gd Á"gd gd gd gd g gd d gd gd gd gd gd g d g gd gd gd gd ggd gd d gd gd gd gd g d d gd gd gd g gd gd gd d

gd

Map 9: Reflex of lengthened /o:/ in the nom. sing. of /*kopa/ (a.p. b) 'mound' gd

Á"

For the names of AV dialect survey locales, refer to appendix a.

Legend

Lengthened /o/ in NS /*kopa/ 'mound' AV dialect survey locales Rivers Major Dialect Boundaries

196

45

50

d

d

d

#d ³

d

d

434

447

446

d

d

vu:sa

44

d

d

d

d

d

d

#d ³

#d ³

450

0

d

40

d

bu:sa vu:sa 454

448

d

d

d

d

#d ³

d

d

lu:ska

d

d

d

d

d

#d ³

d

43

d

50

460 d

461

d

453

439 d 440 d vu:ska d vu:ska

d

428

d

d

#d ³

d

410

d

430

429

d

d

219

d

100

d

419

d

d

d

464 49 d

#d ³

519 d

465 d

d

462

vu:sa

d

d

d

d

d

d

520

518

d

d

516

d

d

d

150

604

d

603

d

602

d

d

d

d

d

d

d

d

610

607

d

d

d

d

133

# ³

132 146

d

131

d

d

507 d

508

d

502

d

d

136

d

135

# ³ ³ #

59

d

134

d

d

653

137

# ³

91 d 646 647

d

d

654

d

652

# ³

d

651

d

d

d

675

65

645 d d d vusa 648 d 649 fusa d 650

d

644

615 d

d

d

626

200 Kilometers

d

d

609 62

d

624

d

d

d

d

d

d

d

628

79

629

625 d 627

639 d

62363

# ³

# ³

d

d

736

d

75

d

719

d d 718

643

737

720 d

d

641

d

640

d

d

733

717

735

d

734

d

642 73 d

715

d

663 630 631 655 656 d d d 64 d 657 d vu:savusa d 616 66266 617 d d d d 665 d 632 658 d 661 d 633 618 d 619 d vusa d d d d 664 659 634 637 d 635 d d d 620 660 622 vus 636 d d 638 d 621

614

d

# ³

147 145 13 d d d fu:sa 148 160 d d fu:sa 159 d 161 d 162 501

612 d 613 61 d

d

611

608 d

52 d

d

512

606

605

d

510 513 d 511 d

601 d

d

d

144

157 158

d

143

156 d

12

d

505 506 d 504 d 53 509 d

222

515514

d

503

517 d

221

vu:sa 51 d 411

#d ³

d

d

153

d

142

154 155

246

257

236 d

220

d

152

463

256

27

235 245

d

234

d

23

d

218

418 vu:sa

vu:sa

#d ³

244

vu:sa

# ³ ³ #

438

d

233

d

d

216 217

255

d

232

427 d

409

254

452 459 d

d

d

242 243

426

d

d

408 42 d

d

26

417

#d 458 ³ d

457

d

456

455

d

d

215 231 230

vu:ska 451

437

d

425

407

d

253

d

241

252

45

424

449

d

d

416

d

d

22

d

228

406

251

435 436

d

423

d

415

d

250

d

249

d

240

405

d

224

239

403

404

414

422

d

25

d

238

433

41

21

3

02

01

48

7

d

223

Map 10: Lengthened /o:/ in the nom. sing. /*vosa/ (a.p. b) 'wasp' d

# ³

N

Legend

For the names of AV dialect survey locales, refer to appendix a. area of detail

Rivers

Major Dialect Boundaries

AV dialect survey locales

Lengthened /o/ in NS /*vosa/

197

60

1

d

d

2

d

3

#d ³

d

0

d

d

309

d

#d ³

d

d

d

60

d

d

d

d

d

d

431

d

d

d

d

d

d

421 d

413

d

402

d

401

248

d

120

445 d

d

d

d

d

d

d

d

d

d

d

d

d

d

448

d

40

d

d

d

d

d

456

455

d

#d 458 ³ d

d

d

d

d

d

d

428

d

d

43

d

d

430

d

d

d

d

d

d

d

49

464

d

d

d

462

519 d

465

463 d

d

411

257

d

d

d

153

d

d

d

d

520

d

d

d

d

604

d

516

d

603

d

602

d

d

d

d

d

d

607 d

d

d

d

d

d

d

160

617

161

620 d

619

159

d

146

d

131

129

d

610 d 60 614 d 615 612 d d 613 61 d

611

608 d

52

606 d

d 512d

605

d

117

145 13

501 506

510 513 d 511 d

601 d

d

144

d

128

157 158

d

143

d

d

113 114 115

505 d 504 d 53 509 d

515514 d

d

127

d

10

156 d

12

d

d

141

222

503

517

518 d

d

51

d

221

d

126

142 d

d

125

154 155 d

d

124 d

124

#d ³

112

105

246

140

236

#d ³

152 d

d

d

123

11

d

104

du:ha 220

256

419

410

245

27

235

d

d

d 219 d

23

218 d

151

234

d

d

139

d

#d du:ha ³ 111

109 122 d

d

d

101 110103 d d

102

240 Kilometers

d

461

d

453

d

429

d

108

418 d

233

d

217

255

d

d

216

d

150

d

439 d 440 d

d

138

244

d

232

d

d

d

120 121

14

d

d

d

107

214

427

438

460 d

d

409

254

452 459 d

d

d

242 243

426 d

417

d

d

d

408 42

451

+douha 457

d

d

437

d

407

d

26

d

215

d

149

231 230 d

d

213

d

212

253

425

d

45

180

d

d

d

241 252

d

416

454

d

d

227

22

d

450

d

d

424

449

d

211

d

7

229 228 d

406

251

435 436

d

423

d

250

d

249

415 d

d

240

405

d

d

d

210

226

224

239

434

446 44 d d 447

d

225

d

21

d

204

403 404

414 422 d

d

25

d

209

238

433

41 432 d

443 444

d

d

335

d

d

223

d

d

207

247

334 d

d

237

412

420

d

337

d

336

328

d

329

332 333 d

331

d

330

d

327

d

d

d

206

d

202 203

201 208d

205

24

325 d 326

441 442 d

324

d d

d

d

320

323

30

d

318

d

32322

321

d

319

d

316

317 31

d

315

313 314

+douha +douha 310

# #d ³ #d ³ ³ #d +douha ³ #d 302304 ³

308 +douha 305 d 301303 d

#d ³

+douha 307

+douha 306

d

311

d

312

d

4

d

5

d

6

106

Map 11: Long /u:/ in the nom. sing. /*duha/ (a.p. b) 'arc, rainbow' d

³ #

N

Legend

Area of Detail

For the names of AV dialect survey locales, refer to appendix a.

Major Dialect Boundaries

Major Cities

AV dialect survey locales

Long /u:/ in the NS of /*duha/ 'rainbow'

198

100

g1d

gd 9

gd

Rivers

Legend

AV dialect survey locales

gd

Major Dialect Boundaries

Lengthened /o/ in nom. sing. /*krosna/

# ³

0

100

g6d

gd

40

g7d

200

300

400 Kilometers

106 102 104 103 107 101110 kru:sna kru:sn'e 105 108 kru:sna kru:sna 109kru:sn'e 120121 kru:sn'e 111 11 112 10 113 5 116 kru:sn'e 149 krusna kru:sna 117 kru:sn'e 138 122 kru:sen' 124 123 114115 212 202203 204 210 krusna 119 124 kru:sna krusna 14 201208 211 125 127 21 118 139 209 128 kru:sna 126kru:sna 213 129 4 150 krusna 206 227 226 kru:sna 207 140 214 krusna kru:sna 151 225 95 205 130 141 228229 94 12 216 krusna 131 krusna 3 215 142 143 223 133 24 818 152 132 224 krusna 325 801 kru:sna krusna 153 217 krusna 135 22 144 218 231 krusna 134 136 146 819 326 237 krusna 238 23 230 312 kru:sna 644 krusna 232 13 147krusna 219 240 145 92 krusna 247 krusna 137 91 154 155 80281808820 311kru:sna 242 kru:sna 148 327 241 krusna kru:sna 647 803 807 334248 830 kru:sna 313314 160 239 krusna 156 645646 328 krusatina 822 831 821 26 kru:sne 233 234 krusna 84 159kru:sna 221 832 648 809 315 335krusna 249 252 243 804 220 161 158 82 157 649 krusna 401 93 316 810 2 162 650 kru:sna 805 kru:sna 329 krusna 306 317 59 krusna 235 811 250 kru:sna krusna 31 kru:sna 244 kru:sna 236 222 652 651 65 667 330 336 812 823 27 825 501 251 701806 25 403 kru:sna kru:sna krusna 826 833 813 318 666 668 kru:sna 245 krusna 824 412 402 502 834 253 kru:sna 654 kru:sna 721 506 kru:sna 307 702 505 kru:sna 406 653 83 254 827 319 krusna 507 kru:sna 676 504 53 509 246 814815 kru:sna 404 337 508 kru:sna 255 kru:sna 828 413 kru:sna 331 722 kru:sna 663 738 669 405 703 630 71 836 835 816 677 408 631 655 656 256 kru:sna kru:sna kru:sna kru:sna 321 320 308kru:sna 670 747817 kru:sna 407 64657 616 503 513510 680 704723 739 409kru:sna 66266 303 305 617 257 748 829 301 332 512511 671 665 kru:sna 410 309 kru:sn'ak 632 681682 724 krusna 672 658 42 515 661 414 633 302304 kru:sna 514 51 32322 kru:sna 618 664 krusne 619 673 418 415 kru:sna 740 420 kru:sna 411 749 krusna 605 706 634 333 72 678 684683 637 659 417 kru:sna 516 310 416 421 635 krusna kru:sna 750 kru:sna kru:sna 67 705 620 419 kru:sna 422 660 431 622 324 krusna 707 674 709 517 601 606 636 41 kru:sna kru:sne 427 323 638 725 kru:sna 423 621 krusna 679 685 462 432 kru:sna 425 441 442 726 52 675 741751 428 429 640 62363 607 711 728 708 kru:sna424 463 kru:sna426 444 433 752 443kru:sna 43 30 430 kru:sna kru:sna 639 742 465 434 kru:sna 743 602 608 641 710 727 753 438 440 712 624 518 437 kru:sna 464 kru:sna 445 439 715 435436 446 754 625 kru:sna 714 713 kru:sna 729 44 49 519 642 73 603 451 kru:sna kru:snakru:sna 627628 447 kru:sna 520 755 74744 449 45 kru:sna 717716 610 730 746 453 731 609 62626 452 611 kru:sna 448 450 kru:sna 745 614 629 kru:sna 604 643 734733 455 kru:sna 756 612 615 718 kru:sna 732 757 720 459 454 613 61 461 456 719 735 460 79 kru:sna kru:sna 75 457 458 kru:sna 737

g #gd ³ #gdg³ ³ #d #d gd #gd gd ³ ³ #gd gd³ #gd ³ #gd³ ³ #gd ³ gd gd gd #gd gd ³ gd #gd gd # ³ g # ³ d g g g g gd gd³#gd gd d g d g d d g d d #gd gd ³ ³ gdgdd gd gd gd gd gd #gd ³ g g # ³ d g For the names of AV gd #d gdd gd gd gd gd gd ³#gd gd ³#gd g dialect survey locales, d gd g gd gd gd gd g g # ³ # ³ d d refer to appendix a. gd d gd g ³#gd gd gd g g g g # ³ # d ³ d d d # ³ d g g g g gd g d gd d gd gd ³#gd d gd gd ³#gd³#d ³#gd ³#g³#gdd d #gd gd ³ g g g g d d g g d d g #d d ³#gd ³ gd g gd gd ³#d gd gd #gd # ³ ³ gd g g g d g d g # ³ d d d #gdg g gd ³ d gd ³#gd #g gd gdgdgd gdgdgd gd gd dd gd gd ³#gd gdggdgd gd g gd ³#ggd g gd g³#gd³d gd ³#gd g g g # ³ d g g d g # ³ d d d d d g g # gd #d d ³d ³ ³ gd gd gd gd gdd #d d gd ³ gd gd g #gd gd ³ # g d g g g g g #d ³ #d gd g g ³ #g ³ d dd ³#gd d gd #gd ³ gd gd gd gdg #d #gd ³ gd gd gd gd ggdd gdgd gd gd ³#gd d³#gd g g g #gd d ³ ³ d d # ³ g d g # ³ gd d d g g g # ³ d g d g ³#d ³#gd g gd #d ³ ³ gd g #gd #gd #gdd gd ³ ³ ³ gd gdd ³#gd gd ³#gd gd gd gd g ggd g gd gd ggd gd d gd gdgd gdgdgd ggdgd g g # # ³ d g d g g d d d d g g gdgd gd d gd g gd gd d ³#gd d g gd dgd gd ³#ggd gdg³#dgd³#gd gd gd d³#gdgdgd dg gd gdgdgdggd gdggdggd ³#d gd³#d gd gd g g # ³ d gd d d g gd ³#gd g d d d g g gdgd ³#gd ³#gd g d # g #d d ³ ³ d gd d #g g gd d gd g gdgd #gd ³ #gd d gd d gd ³ gd gd gdgd gd gd dgd gd #gd #gd ³ ³ #gd ³ ³ #gd³d d ³ g g g g g d d d # g d g g d g g d g d g d gd gd g ³#gd gd gd ³#g ³#g #gd ³ ³ g d gd gdd g gd ggdgd gdgd gd dggdgd gdgdd gdgd gd gd g d gg g dgd d d #gd #d gd ³ gd d d gd #d ³ #d d ³ ³ gd g d gd gd d ggd gd gd gd N gd gd ggdd #gd g g g gdgd gd # g ³ d # ³ d d d g #gd ³ d g ³ g g g d d gd # g ³ # d # d ³ d d g g g d gd³#gd ³#gd g ³#gd gd gd gd gd g g d d g d g d gd gd gddgd gg gdgd gd gd dgd ³#gd ³#gd gd gdgdd gdgdgd gdgd gd ggdd gdgd ggddd gd gd g d gd gd gd ³#gd³#gd ³#gd gd gd gd d ggd gd gd³#gd d gd kru:sna 736

g8d

Map 12: Lengthened /o/ in the nom. sing. of /*krosna/ (a.p. b) 'backpack for hay'

199

80

0

80

160

240 Kilometers

@d106 #@102 ³ #@104 ³ @d7 @ # ³ @ #@107 ³ # ³ @ 103 # ³ 101 #@105 #@ #@ 110 ³ ³ #@120³ ³ #@121 108³ d @ 109³ #@111 11 ³ #@112 @d10 ³ #@116 ³ #@113 ³ #@149 @d ³ ³ @ @ # # @ 122 117 # ³ @ @ @ d @ #123³ ³ #114³ #115 ³ ³ @d210 @d 212 @d 138 #@119 #@124124³ #@125 ³ #@127³ @118 N @139 # ³ 211 14 # ³ @ # 128 ³ d @ @ # @ @ # ³ # ³ @d227 213 @d 150 129 126 #@140 ³ 226 214 @d95 #@130 ³ @ @ # ³ # ³@d141 151 @d228@d229 d @ d @ @ 94 # ³ 12 @d215 216 #@143 131³ #@142 ³ ³ #@133 @152 # ³ #@132³ 4 @d801 @d @135 # ³ d @ @ # ³ @ @d22 d @ # ³ @ 153 217 @d231 # ³ 144 @ #146@ 134³ ³ #@136 ³ @d23218 @d230 @644 # ³ d @ d @ @ @d92 # ³ 147 d @ @137 # 13 219 145 240 # ³ @d154³ @d91 #@155 @d802@d @d @ @d241 @d242 232 # ³ 808 d @ d @ 81 @d807 d @ 148 @ # ³ 647 803 @ 646 @ # ³645@d 160 #156 ³ ³ @d234 @d @d @d26 @d @ @ # # ³ @d804@d 233 221 #@161 ³ 648 @d649 @d #@158 159³ d252 243 #@157³ ³ 249 @ @d93 220 @162 # 810 d @ 650 d @ @ d811 805 @ # ³ 59 d 235 ³ @ @ 250 # ³ @ @ d @ # # ³ @d27 236 d65 @d667 244 @501 d @ @d251 # ³ 222 d @ 652 651 @ d813 701806@ @d666@d668 #@245 ³ @502 # ³ d @ d @ d @ d @ @ @d406 @d253 @d254 # ³ 654 d @ 506 #@507@d508 @d676 702 72 653 #@504505@d53 @d509 ³ #@246 ³ ³ @ # ³ d @ d @ 255 d @d703@d72 @ d631@d @@dd655 @d656 663 @ 05 #@256 ³ 630@ d670669 @d677@d71@d @d407 @d408@d d @ d @ d @ @ d @ 64 # 510 d ³ @ d704@d723 #@257 @d503 513@d512@d511 @d617 616 @d632 657 @d @d665@d 66266@d671@d 680 @ @410 ³ # ³ d @ @d724 d @ @d42 409 ³ d @ 681 d @ d @ d664661 @d673672 658 @ 682 @418 @d619 @d618 633 # 51 515514@ @ d d @ d @ # ³ 5 @d634 @d637 659 @d678 @d684683@d706 @d72 411 @d516 605 @d @d416 @d417 @d635 @d67 @d @d @d705 @419 # ³ d @ d @ 620 d @ d @ @d517 @d 606 d636 @d638 660 622 @ @d #@685707@d 725 d @ @d427@d @d d 674@ d @ d 423 @ 621 d @ 601 679709 ³ 462 d @ d @ 52 @ d @ d @ d @d424 425 @d d @ 675 428 429 @d728 @d708 726 @d741 d @ 63 640 623 607 711 463 d @ d @ 426 43

Map 13: Lengthened /í/ in the nom. sing. /*zima/ (a.p. c) 'winter'

Roads

Rivers

Area of Detail

For Names of AV dialect survey locales, refer to appendix a.

Legend

AV dialect survey locales

d

Major Dialect Boundaries

Lengthened /í/ in the NS of /*zima/

# ³

200

gd

gd

10

113

674

N

707

gd

# ³

60

124

606

0

60

120

180 Kilometers

gd 117 116gd 114 115 g gd gd gd d 119g 124 d 125 gd 127 gd 118 gd 128 g d g d 129 126 gd gd 95 130 141 gd gd gd 94 12 g 131 d gd gd 133 143 142 gd gd 818 132 g #gd gd d ³ 801 135 nuk 153 g g # ³ d 144 d gd 134 gd nuoh 146 gd 136g 819 gd gd d 147 644 g 145 13 g 92 d d 820 gd gd gd 137 91 154 155 802 #gd gd ³ gd gd gd 808 nuh #gd 81d 807 ³ 148 g # ³ d nuch g 647 803 g 646 830 d 160 nuch #gd nuch ³ #gd ³ 645 g gd d gd d 831gd 84 156 821822 g nuch g g g d g # ³ d d d 159 221 832g nuh 648 809 g gd 804 gdd gd649g #gd 82 ³ 157 #gd 810 ³ 93 g gd 158gd d 161gd 162gd nuch d d 650g gd g nuch # ³ d 805 d #gd nuch811 ³ 59 gd nuch #gd ³ 651 222 652 g nuch 823 825 812 gd 501 gd d 65gd 667gd 833 701 806 gd #gd ³ ³ #gd ³ g #gd 826 gd # ³ d g +nuhnuh ³ d nuch 813 gd 666 668 nuch # 824 g 502 g g # ³ d d g d nuh # ³ d 654 g 721 506 d nuch 702 505 nuh #gd ³ 653 g 83 gd gd #gd ³ 507 gd d +nuch 676 46 504 gd 508g d 815gd 827 814g bez nu:h #gd ³ g gd 53gd 509gd g d d nuh 828 d 630 631 655 656 722738 663 #gd 669 ³ 703 g 71 816 nuh 836 677 g g d g d d g g g d g gd g g # ³ d d d 817 d gd d d 670 gd 64g 657 510 +nuch g 739 747 g g d d # ³ d 513 g 616 503 723 680 66 d g 662 g g d 617 d g gd gd 512gd 511gd no:h g g 704 #d d 632 gd ³ 748 gd gd d 724 gd 829 #gd ³ 681 d gd 633g 658gd 665gd664661gd 671gd672gd 515 gd+nuch682 nuh g 51 d g 618 # ³ d 673 619 d gd gd 514gd 605 d gd gd nuh 740 gd gd 749 683 706 659 gd gd 634 gd 72 678 gd 637 d gd 684 g 516 g g d g d 635g d g gd d 750 d g 705 67 d g 620 d 660 622

112

gd

ch 105

Map 14: Lengthened /o:/ in the gen. plural of /*noha/ (a.p. c) 'leg'

gd 835

gd

834

Area of Detail

For names of AV dialect survey locales, refer to appendix a.

Legend

Lengthened /o/ in the GP of /*noha/ AV dialect survey locales Major Dialect Boundaries Rivers Roads

201

70

N

gd 1

gd 9

AV dialect survey locales

gd

Major Dialect Boundaries

Rivers

Length in the GP of Czech /louka/ (a.p. b) 'meadow'

# ³

0

70

140

210

280 Kilometers

g gdg gd Legend gd 106 gdd gd gd 102 gd d 104d # ³ g d 7 103 g 107 6 101 110 gd gd gd105gd gd gd108 109 gd gd gd gd gd 120121 111 11 112 10 113 5 g gd116 d g g 149 g g d gd gd117 # g ³ d g d d g d d 122 gd gd gd 212 gd 138 g 123 gd124 gd gd 114115 gdgd202d 203gd 204gd210 For names of AV gd119 d g 124 125 127 d 211 g 201208 14 21 dialect survey locales, 118 g 139 g gd d g d 209 128 g d d g d206 gd gd 227d 213 gd150 refer to appendix a. g 129 126 4 d gd226 g gd gd gd gd 207 225 gd 140 gdg 214 95 130 gd 205 d 229 151 141 g d g d d gd 228 94 gd 216 gd12 gd 131³#gd³#gd gd gd 3 gd g #gd ³ 215 142 143 gd24gd 223 133 gd 818 d g g 152 132 g g g g 224 d d d d d 325 g # ³ d g 801 gd 135 d g # ³ d 217 153 # ³ d g 144 22 g 218 g g d d g 238d gd231 g 23d gd g gd 134 136³#gd gd g 819 gd147 gd gd #gd146 ³ gd d247 230 g d g gd 312dg gd 326 237 g d 644 d d d 13 219 145 240 g g 92 g gd gd d 232 137 155 g802 gd gd gdd 820 gdgdgd gd 311g d d d808 gd #gd691g ³ #gd148 d d 327 gd 334 ³ gd ³#gd 160 #gd646 241 g 242 ³ gd g 154gd 156 47 803 81807 313314 239 g d g 248 gd84gd gd 821822gd830 831832 g 328 g 645g d g d g d g d g d d 234 g d d g # g ³ 26 d # gd d ³d 159161 gd gd gd g 335401 gd 809 d 243 233 gd 648d g 249 252 804 220 221 g d 82 157158 649 d 93 810 gd 315gdg316 g g g d 2 162 d gd #g ³ dd ³#gd d d 329gd gd 235 805 gd gd 650 #gd ³ gdgdgd gd gd 823 306 31731 g gd gdg 59 gd gd811dgd 812 #gd ³ gd806 gd gd 250251 244 222 g 652 651 65g 330 336g g 27 236 gd825826 833 gd d 501 667d d g d 701 403 # 25 ³ g d d g g g g d253 gd gd gd 318 d # d ³ d g d g 813 668 666 g 245 # ³ d g d 824 g d 412 402 gd g 834 d 654 #gd 502 ³ gd 702 721gg gd83d gd 827 g g 506 307 d g g d 505g g g 406 d g d d # 653 ³ d d g 254 d d gd gd 319 g g 507 d g g d 676 246 504 53 509 gd722d738d g814815 ggd gd 663d g669 gd 71gd gd828 gd 255 d 677 d 404 g d gd gd655 508 g d g g g g d g d 331 337 413 g d d g d d d 405 # g ³ 703 630 d 836 835 816 320 408 g d g g g g g g g 631 d 656 # ³ # 256 g # ³ d ³ gd 305d 308 gd321 d d d g d d g d d 817 g 670 # g ³ gd303 d g 747 # ³ d g g d 407 g 64 #d 511 617 g616 gd ³ gd 257 gd503 ³ 657 g g dgdg662 gd680gd 704d 723gd 739 748d 829d 513510 66 d g 409 g g gd 309 gdgd 332d g d g g d # gd304 d ³ d g 301 512 d # 671 d d # ³ d 665 # g 410 ³ g g d 632 gd g 42g #d ³ d gd gd 658 gd 51 515 gd 664661 673d672 gd gd681gd682 633 302 #gd ³ gd 724gd 740 gd 420d gd 414 415 418 619 618 g gd 32322 333 #gd 659 ³ #gd514605 ³ d 417d 749 411 #634 ³ d³#gdg637 706 g g g 72 g g 678 g d g d 684683 # ³ 310 g g 516 d d 416 g d g 421 d g g 635 d # g ³ d d g g # d ³ gd 705 gd750d d gd67gd gd707gd725 #d ³ ³ #d ³ ³ gd gd #gd 660 g674 d d g431 gd41d 422 423 gd g g 419 ³#gd 517 gd gd606d 620 621 #gd 622 636 ³ # 323324g 638 g g 709 d g g d d 726 741d 751g gd 425 g 427428d g429d #d 675 gd679 gd 685 #gdd ³ ³ #gd ³ gd462 601 52 607 gd g 424 gd gd432 433 640 62363g d gd 441 442 443 d gd 463 711 728 708 gd 752 #d ³ ³ #gd ³ ³ gd742 g g g d g # ³ g 444 g 426 g d # ³ d gdd g # 43 430 d d g g d 30 g g d d # ³ g d d 639 # d d 465 # g ³ d464 518 602 608 gdg 712g710 727 743 d445 gd434 gd gd 437 438439d440 641 g g 753 624³ # d d gdgd gd gd446 gd ³#gd ³#gd ³#gd g gd #gdgd715 ³ gd714 #gd ³ gd 435436 ³ gd754 gd gd 451 gdd gd713gddg729 #gd 44447 #d ³ 49 519520 642 603 g ³ gd625627 gd 73 # ³ d g g 755 gd 453 628 # 716 74744 d d # ³ d 717 610 gd 45gd 452 730 g g gd449450 g 626 ³ d g # ³ # d d # 731746 g ³ g d # ³ d 609 62 d 611 g g g g gd 734733d gd gd 745756d 629 448 604 gd643 gd455g gd d gd 612 #g ³ d³#d gd614615d gd gd 732 757 720718 g 454 d 459g 613 61 461 d 456 719 460 gdgd 79 gd75gd 735 457 458 737 gd gd 736 40

gd 8

Map 15: Reflex of Length in the Gen. Pl. of /*louka/ (a.p. b) 'meadow'

202

70

gd 9

AV dialect survey locales

gd

Roads

Major Dialect Boundaries

Length in the GP of /bouda/ (a.p. b) 'shack'

G

0

70

gd 7

140

210

280 Kilometers

Rivers gd gd gd gd104gd 106 gd d gd 102 gd gd103 G Legend 107 g 6 101110 g g 108 g d d d g105d g g gd 109 d d g d 120 121 g g Gd Ggd116d g g gd 111 11 112 5 gd 10 G113 gd gd d 122 gd gd124 gd Ggd117 gd149d g 138 gd 204 gd203 gd g gdgd202 123 g 114115 212 g g d Gd 127 d Ggd gd119 d 210 211d G gd 124 125 d 201208 g 14 21 g 118 g gd Gd gd gd 209 gd gd 213d Ggd 128 129 Ggd 139 Ggd 4 126 Ggd150 gd226 g227gd gd gd206207 140 g g 214 Ggd d d G G 225 95 gd 205 130 d 229 151 g 141 g d g d G d G gd 228 94 g Ggd 131Ggd133 Ggd 216 Ggd12 143 Ggd g gd 3 gd g Ggd Ggd 142 215 gd24gd 223d gd 818 d g 152 gd 132 g G g g 224 d G d d d 325 G g d g G 801 gd 135 d g d G 153 217 1 d g 22 G 144 g d d g gd gd231 g 23g218 d gd gd gd 134 136g Ggd 819 gd147 G gd146 gd g g G d d247 238 240 230 g d d g g d gd 312g gd 326 237 G g d 644 G d d G d 13 219 145 92 gd gd 232 137 155 g802 Ggd gd gd gdgd820 gdgdgd gd 91 311g d gd808 gd Ggd Ggd647 d d 327 Ggd 334 Ggd148 gdg646 gd Ggd 160 241 g 242 G G gd g 154 803 81807 313314 G 239 g d g g 248 gd84gd gd 821822gd830 831832 g 645 328 g d d 156 G g d g d G g d d gd 26243d gd Ggd 159161 d gd 809 gd Ggd Gd 221 157158 d Ggd G g g G G gd 233 234 220 d d 648 315 g 335 d 249 804 g 82 649 d 252 g 93 810 gd gdg316 g gd gd g g g 2 gd d d 329d gd 401 gd 235 805 gd 236 306 31731 g gdgGgdg Ggd G 222 gdgd gd gd 16259d Ggd Ggd 650 gd gd811dgd 812 d dg825 gd806 G gd gd 250251 244 G g 652 651 65G 330 336g g 27 d 501 667 d d G d 833 g g d 826 701 d gd 25 403 gd gd g d G gd gd gd 823 gd gd 502 d gd gd 318 g G g 813 G 666 668 245 d d g 824 G G 412 402 g d G d 253 834 d g 654 g g G 721 g 506 g d g d g d 307 g G d 505 702 G G g g d 406 d d g d g 653 d d g G G 83 254 d d 827 g 319 g G 507 g d g g G g d 676 g d 504 246 g g g G d g g 814815 g g d g G d G d Ggd gd655 Gdg722738GgdGgd gd816d 828 836d 835gd gd Gd 71d gd 703 Ggdd 656 gd 320d 331 337 413 404 405d Ggd Ggd663gd Gg669 gd 321 gd 408d gd 255 g53d gd509 Ggd 508G630 dG 677 gd308 G gd GgdG gG 631 G 256 g gd 305 d G GdGgd 739747817Ggd gd G g 670 gd303 d g g 407 g 64 G 510 d g G g d g d GdgGgd d d gd 332 GdGgdGg662 Ggd G657 gd 409 gd 410d 257 gd 503 Ggd680682 gd gd gd 658 gd Ggd Gggd665 Ggd 704723724 829 GgdGgd513g512511 Ggd617 Ggd616 632 301 d 66671 309 g 681 d G g 672 42 515514 g d 661 g Ggdg 748749 414 633 302304 51 d d 32 Ggd g 618 g d 664 G g d G706 G 673 678 gdg Ggd683 gd 418 d g72gd 740 d Ggd 659 G gd 322 333d 420 gd g 415 gd 417 gd 605 619 G G g G 411 d G d G 634 g g 637 d d g G 750 310 g g gd 421gd 422d Gd Ggd 622 Gd Ggd Ggd67Ggd 684Ggd707 Ggd635636 N GgdGgd725 Ggd 620 gd 416 gd Ggd516Ggd gd606 d d d g431 Ggdg 638 gd g 419d Ggd 517 Ggdg660 Ggd 705 gd41 323324g gd674Ggd679709Ggd 685 g 423 g d g g432 621 427 g d d g 601 52 G g G Ggd Ggdgd d G d d 462 d d G G d g 425 441 442 726 d G 675 711 741751 gd g 424 Ggd 428gd429Ggd 463Ggd 708 640 62363g 607 d443444d 433 gd G GGgd 728 d g g G g g d g d d g 426 752 G Gggd 753 G 608 gd43 430 Ggd465 Ggd 602 Gd gd gd440 Gd Ggd 712 gd g 639 641 Gd 727 gd710 d G gd 438 30 gd gd434 gd gd 437 G G 743742 624 d G Gd Ggd g g GgdGgdg gd 464 445 g gd Ggd518 d446 gd714 GgdGgd715 Ggd Ggd g Ggd625 gd 435436 G 755 Ggd754 Ggd 603 d 744 Gd Ggdg 45gd 451gd gd 439gd GGgdd Ggd713 gdG729 44447 49 519520 642 g 627628 G d 73 g g 449 74 716 d d G d 717 610 G g 730 gd d g 452 453 746 G d Ggd GgdGgd609 62626 629 Ggd g643 Ggd 611Ggd g614 GGggdd 734 Ggd733 Ggd G731 448 450 d 604 gd gd 745756 gd g gd d 718 gd455gd 459 g612 G d 615 d G G g g 732 720 d d d G 61 g G 757 454 613 d 719 456 gdGgd 460 461 79 Ggd75gd 735 457 For the names of AV 458 737 dialect survey locales, Ggd gd 736 refer to appendix a. 40

gd 8

Map 16: Reflex of Length in the Gen. Pl. of /bouda/ (a.p. b) 'shack'

203

100

gd1

N

106 107

gd9

101

102 104

AV dialect survey locales

gd

Major Dialect Boundaries

Lengthened /e/ in NS /*dest'/

§

149 212

108 120121

Rivers

0

5

100

200

300

400 Kilometers

103105 110 11 112 10 113 116 109 111 138 122 124 114 115 117 202203 204 210 124 211 201 14 125 127 21 139 123 128 118 119 208209 213 129 4 126 150 206 226 227 207 140 214 225 95 205 130 151 141 229 228 94 12 216 131 3 215 142 143 133 24 223 818 152 224 325 801 217 153 132 135 22 144 218 231 146 819 136 326 237 238 23 230 312 219 14513 147 134 137 644 240 92 247 232 820 154 155 91 311 802 81 242 148 327 241 334248 830 831 313314 160 239 647 328 645 156 803 807 821 26 84 159 233 234 220 221 646 335 249 243 804 161 158 82 252 401 93 316 810 809 2 162 832 649650 805 648 315 329 306 59 235 811 250 157 244 222 652 651 65 667 31 812 823825 833 336 27 236 501 251 701 25 403 317 330 806 668 245 412 402 502 826 253 834 654 506 318 666 307 702 406 653 254 319 721 83815 824 676 504 53 509 507508 246 404 337 413 255 828 331 663 722 669 405 630 703 71 836 655656 677 835 408 256 321320 747 407 64 510 305308 816 680 409 662 66 670 617 616 631 257 503 513 511 301 332 657 665 410 309 723 739 681 682 42 829 414 632 51 32322 618 418 415 704 304 420 740 749748 661 672 514 512 619 411 633 634 333 72 515 417 673 683 310 416 302 421 605 620 750 706 705 637 659 664 67 419 516 422 431 622 324 684 635 517 41 323 638 660 674 725 423 621 427 685 462 432 425 441442 751 52 606 675 741 428 429 640 709 623 63 607 424 726 463 601 679 426 443444 433 752 708 43 430 30 639 641 465 711 710 434 743742 602 608 438 440 624 518 437 464 445 715 727 435436 753 625 439 44 446 49 519 712 642 451 627628 744 520 603 755 754 610 45 713 74 453 447 449 609 62626 452 611 745 614 716 448 629 604 643 717 450455 731 756 732 459 720 718 612 61 615 757 454 461 456 460 733 79 719 613 735 75 457 458 737736 40

6

7

§ggdd §gd ggd gdgdgd§gdgd gd gd Legend gd gd d gd gd gd g gd d gd gd g§ § g § d§ gd gd § § gd§ gd§ gd gdg gd g gd d g gd gd § § g § § d g d gd§ g § d d gd gd gdgd gd gd d d For the names of gd § g gd § gd gd AV dialect survey locales, § g d gd§ d g § § gd d g g g § d d d refer to appendix a. gd g gd gd gd gd gd § § gdg § gd d g d g d d gd g g gd d d gdgd gd gd g gdgd gd gd gd gd gd gd gd g g d d g § d g d g d § §gd ggdgd gd gd gd gd gd gd gdgd gd gd §gdgd§§gdgd d§gd §gdgggd gd ggdgdggdgdgd g g gd dg gd gd d d d gd gd d gd gdgd d gd gd gd § § gd gd § gd gd gd gd§ gd gd gd gd d gd g gd gd gd gdgdd gdgd gd gd § gd § gd gd gd§ g g gd g d d d gd gd d gd gdgd g gdg gd ggd gd gdg gd § § gd gd gd gd gdg § gd d d § d gd gd gd gd ggdd gdgd gd d g g § d d g g g d d d gd gd d gd § gd gd g g gd gd gd g § § gd gd d g g d g § § g d d d g d gd d§ gd gd gd gd § gd gd g gd g d § gd gd gd gd gd§ g d g g g g g § d d § g g g g d g g d g d § d gd gdgdd gd dg § gd § g gd gd d d § gd gdgd g d gd gd d gd gd§ gd§ gd§ g g g g g § § § d d g d d d g g d § g § gdggdggd d gdd d g d g d § g § d g d d g d d g d g d § gd § gd d§ gd § g gd gd ggd gdgd gdgd gdg gd gd § d d gd d § gd § g § gd gd gd gdd gd gd d g § § g d d g g § § g d d g § gd d gd gd gd g gd gd gd d § d § d gd d g § d g g g g § g d d d d g d g g d g § g d g d g d gd § gd gd g gd gd d gd§ g gd§ gd g§ gd § gd gd gd § g g§ g§ g d gd gdd g § § d gg § gd d § gdgd d§ d gdg gd gd d gdggd § gdd § d gd dgd§ gd d § d d gd gd gd g d g g g d § d g g d§ d g g g § g d d § § d gdd d § gd g gd d § d § gd gd gd gd gd gd gd gd gdgd g d g § d d gdgd gd § gd gd gd gd §g§dggddg§dgggdd §gd§§ggdd§gdgdgd§§ggdd g §gd gd gd gd g gd gd gd dgd gd gd gd gdgd gdgdgd gdgd gd § ggdd § § d gd § d gd§ § gd § gd gd gdgd gd gd gd gd § gd § g d gdgd gdgd § §gd gd

gd8

Map 17: Lengthened /e:/ in the nom. sing. of /*dest'/ (a.p. b) 'rain'

204

50

106 107

9

AV dialect survey locales

d Major Dialect Boundaries

Lenthened /o/ in the NAS of /*roj'/ (a.p. c))

³ #

1

0

50

100

150

6

200

7

250

300

350 Kilometers

102 104 101 105 108 120 110 11 112 10 113 5 121 116 149 109 111 117 138 122 124 114 212 202203 204 210 119 124 201 211 14 125 127 115 139 123 128 208 209 21 213 129 118 126 4 150 206 227 207 140 214 225 226 95 205 130 151 141 228229 94 12 216 131 3 143 215 142 223 133 24 818 152 224 325 801 153 217 218 144 22 132 135 231 146 819 136 326 237 238 23 230 312 219 240 14513 147 134 137 644 92 247 232 820 155 154 91 802 311 81 242 148 241 314 327 328 334248 830 831 160 239 645 647 156 26 807 821 84 159 233 234 220 221 803 646 335401 249 252 243 313 316 158 82 93 810 809 2 161 804 649650 832 329 648 306 315 59 235 811 250 157 244 162 222 31 336 27 236 65 667 501 251 701 806 812 823825 833 25 403 317 330 412 668 245 502 826 834 652 651 402 506 318 307 319 702 406 253 254 653 654 666 721 83 824 676 246 504 53 509 507508 404 337 413 255 828 331 663 669 405 703 630 71 836 677 655656 835 408 256 321320 407 64 510 305308 816 680 409 662 66 670 617 616 631 257 503 511 301 332 657 665 410 309 723 739 681 682 42 829 414 51 32322 618 418 415 704 748 420 304 661 672 514 512 619 411 634632 633 333 72 740 515 683 310 673 416 417 302 421 749 659 605 620 705 67 419 422 516 431 622 635 637638 750 706 674 517 323324 684 423 621 660 427 462 432 41 606 425 441 751 52 725 741 675 428429 685 640 623 63 424 463 442 679 601 433 607 426 752 43 430 708 30 639 641 465 434 743742 711 710 602608 438 624 437 518 464 443 445 715 727 435436 753 625 439 44 446 642 49 519520 603 712 451 627 755 754 74744 610 453 713 447449 45 628 609 62626 452 611 745 614 629 448 604 716 643 717 455 731 720 718 459 733 756 612 61 615 450 461 732 757 456 460 79 613 454 457 75 719 735 458 737736 40

8

Rivers eb Legend eb eebb ebebeebbebebeb eb ebeb ebebebeb eb eb eb eb e For the names of b e e e ebbe bebeb eb b ebeb ebeb ebebebeb ebeebebebebeb AV dialect survey locales, e e b refer to appendix a. b eb ebeb b eeb eb eb ebeb eb b eb e b eb eb eb eebb ebeb b ebeb e e b b eb e e e e e ebeb b eb eb ebb eb eb eb eb b b eb ebebb ebeeb eb eÙb eÙb eb e e b e e b b e e eb eÙeÙbebeÙÙb eb e ebebebb eb e ebebeb b eb eb ebb ebeb b eb ebeb e ebebebeebb eb eeb ebeebb b eÙb eÙb eb eb ebebebeb ebb ebeb eebb eb ebeb eeb ebeb eb eb ebb eb ebeb ebeb bebeeb eb eb eebÙb eÙbeÙb eb eÙbebb eÙbeb e b ee ee b e eb e e eÙ ee eb eb ebeb ebbebeebebebeb ebeb eb eb eb ebb eb bebebebeebb b ebeebb ebebb beebbeb eb ebeÙb ebeebebebebebebebÙbÙbeÙbeÙbeb eÙb eÙb b e e N ebebebebebeb ebebebeebb b ebe eb eb ebebebeb ebebeb ebebeebbebebeebbeb ebeb ebebebebeÙbebeÙebebb ebebebebebebebbeebbeebbb ebebebebebebebebebbbeebebebebebeÙb eb eÙb e b e e e e b b e eÙb e b b eb eb eb b eb b ebeebeb eÙb eeb eb e eb eb b ebeb eÙb b b ebeb ebeb eb ebebeb ebb ebebebebeb ebebeÙbebebebb ebebeb ebebebebeebbebebebbebebeb eebb ebeb ebebeeÙb eb ebebebeebeb eÙb eebeb eb eb eb ebeb eb eb eebeb eebbeeb eb ebeebb eebb bebebe eb b ebeeb ebeeb eb b b ebbebeb eb eb beb ebebeb ebebeb eebb eb e beebebeb ebebebbebeebb eb e ebebb ebeb eb ebebeb eb eb ebeebb ebbeb eb eb eb ebeb ebbe b eb

eb

Map 18: Reflex of Lengthened /o/ in the nom. sing. /*roj'/ (a.p. c) 'swarm'

205

100

N

gd 1

106 107

gd 9 Czech Districts (okres)

Major Dialect Boundaries

Legend

AV dialect survey locales Rivers

Lengthened /o/ in the NS of /*kos"/ 'Turdus Merulus'

ÿ

gd

0

100

gd 6

200

gd 7

300 Kilometers

g gd gd For the names of AV gd104d gdd gd gd 102 g d dialect survey locales, 103 g 101110 g 105 g 108 g d refer to appendix a. d d g gd gd gd gd d d gd 109 g gd 120121 111 11 112 10 113 5 g g 116 d d gd 122 gd gd gd gd gd gd117 gd g149 138 gd 204 gd203 d 124 gd gd 114115 gd 212d gd gd202 gd 123 g gd210 211 gd119 g d g 124 d d 201208 g 14 125 127 21 gd 118 139 g gd g d 209 128 g g d d g d d g d gd g 129 gd150 4 126 gd226 g227gd 213 214 gd gd gd206207 gd 140d gd 225 95 gd 130 205 d g 151 141 g d g 229 d d gd 228 94 gd 216 gd12 gd g 131g g gd g 3 gd d 133d gd 215 gd24gd 223d gd 818 gd gd 152d gd 142 143 gd gd132 gd 224 gd 135gd 325 g 801 gd d 217 153 g g 22 144 218 g g d 237d g d 134 gd231 g 23d gd gd g 146 819 g 136 d g d g g g 326 d d d 238 g d g d 230 gd gd d 644 ggd gd 312 gd g ggd820d 219 14513147 gd 232d 92 gd gd 242 137 g802 gd247gd 154 155 gd69147d 311g gd gd d d 81gd807 gd148 gd 240 241 d d gd 327 gd 334 808 g g gdgd830 d d g 803 g 646 d 313314 239 160 g d 831g gd 234 gd gd 156gd gd d gd gd 821 gd gd 328 gd 248 645g 822 d84gd g d g d d g d g g 26 g g d g g d d 159 gd d g 233 221 d 832 d 648 809 315 d 335 g 249 243 804 220 g 161 d 82 157158 649 d 252 g 93 810 gd gdg316 gd g g g g 2 162 g gd d d d 329d gÿd 401 gd 235 805 gd gd gd 650 gd 236d 306 31731 g gd gdg 59 gdgd gd gd gd gd811dgd 812 d 823dg825 gd806 gd gd 250 244 222 g 652 651 65g 330 336g g 27 d 501 d 667 d 251 d g 833 g d 826 701 d gd 25 403 gd gd g d gd g g d gd gd 318 d d g g 813 666 668 245 d g d 824 g 412 402 502 g g d d d 834 253 d d g g 654 g 721 g 506 g d d g g d 307 g 702 505 d g g d 406 d d g d g d d 653 g 83 254 d d 827 319 g g 507 g d g g g d 676 d g 246 504 g g g d g669 gd828 gd gd ggd gd gd d gd 255 dg722738d g814815 gd gd655 gd gd816 gd gÿd 320d 331 337 413 404 405d gd817 gd 408 g53 d d 677 gÿd 308 d gd509 gd 508630 71 g gdd 631 gd gd510 gdgd670 d 656 gd gd663 gd 256 g gÿd 305 d g 739d 747 d gd703 321 gd 836 835 gÿd303 g d g g 407 64657 d g d g d g g 513 d d 723 503 616 ÿ 680 66 d g 409 662 g d g g g d 617 g gÿd d 257 g 511 g d g g gÿd 304 d d d 704 g 829 301 748 332 d 512 g d d 671 d d 665 g g g 309 g 410 ÿd g d d 632 g ÿd d d d672 gd g681gd682 gd gd gd 42 gd 51 515 gd 661 673 gd 664 302 gd 724gd 740 gd 420 gd 658 gd 418 619 618 gg d 633 gÿd 32322 gd514605 gd 417 gd gd414 415d 416 d 683706 749 411 d g g 634 333 g g g 72 g 678 637 659 g d d 684 d d 310 g g 516 g d 421 g g 635 d g d gd gd g606d 620 gd 622 gd gd 660 d gd67gd gd707d ggd 705 g750d d d ÿd g 431 gd41gd 422 gd gd g 419 g d d 674 517 d 636 d d 323324g g g 725 638 g g gd 726 741d 751 ÿd gd 675d gd679709gd 685 gd 621 gdgd gd 425 g 427428d g429d gd462 601 52 607 gd g 423 424 gÿd gd432 433 g 640 62363g d gÿd 441 442 443 d gd 463 711 728 708 gd 752 g d d gd742 g g g d g d g 444 g 426 g d gdd g d 43 d g g d g d g 30 gd465 518d 602 608 d624 gd 639 641 gd 712d 710 727 743 d445 gd434 gd gd 437d 438d gd440 430 464 gd 753 gdgdg gd gd446 gd gd gd gd g gdg715 gd 439g gd714 625g gd 435436 d gd gd754 gd gd 451 ggdd gd713 d g 729 d 44447 49 519520 642 g g 603 g d d 627 d 73 d d g 744 g g 755 628 730 gd449gd 45g 452d 453 gd 611d gd610 gd gd609d 62d626 gd 74674 g d d ggdd 717 gd716 gd 731 745 g 629 448 450 d 604 gd g gd gd643 734733 g gd614615 d g gd455gd 459 g 612 d d 718 gd gd 732 757 756 d gd 720719 454 613 61 456 g gdgd 460 461 79 g d75d 735 457 458 737 gd gd 736 40

gd 8

Map 19: Lengthened /o/ in the NS of /*kos"/ (a.p. b) 'Turdus Merulus'

206

100

N

gd 1

gd 9

AV dialect survey locales

gd

Major Dialect Boundaries

Lengthened /o/ in /stoh"/

# ³

0

100

40

200

300

400 Kilometers

Rivers g gd gd gd gd104g 106 gdd gd gd 102 gd g d 7 d 107 6 Legend gd gd108101gd110103 g gd105gd g gd d 120121 109 d 11 g g d d g d 5 gd 111gd 112gd 10 113 gd116 gd 122 gd ³ gd gd117 gd149 138 gd ³ # ³ #gd 204 gd # ³ g For the names of g d # 124 d g gd³ g g 123 114115 212 g g 203 d d d g 202 # # ³ ³ d gd gd119 gd #d #gd 21d 210 211 g 14 av dialect survey locales, 124 125 127 201208 ³ 118 g 139 gd 128 129d gd g #d g150 refer to appendix a. #gd ³ ³ g gd ³ gd 4 126 #gd 209 ³ #d d gd 140 ³ gd226d 227gd 213 214 g g gd206207 # d g d # ³ d gd 225 95 gd 205 gd 141 gd130g #gd 151 ³ gd 228229 94 gd 216 g gd 12 g d d # ³ d 131 g gd g g d 3 g d g 142 143 g 132d 133 gd d g 215 gd24 223 gd 818 gd ³ 152 d g 224 # ³ d d g d 325 # g d g 801 gd d 135 g d 153 gd ³ 22 144 gd #gd 238 d³ gd231 g 217 23g218 gd gd 134 136d g gd 146 819 g g g g g 326 237g d d d d # g d d d g 230 d 312 g g d gd 644 d d d 219 14513147 gd 232 92 gd gd g gd247gd gd 242 137 gdgd81gd808 gdgd820 gdgdgd gd 154 155 gd691g47d 311g 802 gd148 gd 240 241 d gd 327 gd646 g d # ³ d g 803 g 334 807 d 313314 d 239 160 g g g 248 gd84gd gd 821822 gd d 328 gd gd 645g d #d 156gd gd 159d gd gd ³ gd gd 26 gd dg gd gd 809 gd830 831832 #gd 221 ³ gd gd 233 234 220 648d 315 g g 335 243 249 252 804 g 161 d 82 157158 d 649 g 401 93 810 gd g d g 2 162 g gd d #gd 317gd31g316 ³ d ³ gd 235 805 gd gd 650 gd gdgdgd gd gd gd 236d #gd gd gdg 306 ³ 59 gd gd gd811dgd 812 #gd329336 gd806 gd gd 250 244 222 gd 330 652 651 65g g 823 27 gd825826 833 gd # 501 ³ d 667 d 251 d g g 701 d 403 g d 25 g g g g d d g d d d g g 813 gd 253 gd 666 668 245 #d #gd 318 gd412 402 gd d g506g gd 502 d 721 ³ gd654 824 g g d 834 d gd g g g g g d d d 307 ³ d g 505 702 d 406 d gd #d 507508d ³ gd 404 gd ³ 83 827 319 g ³ gdgd814815 gd ggd 676 #gd gd gd g653 gd722 504 53 509 246 gd gd gd gd gd828 gd gd gd254 255 # d g # ³ g g g d g d 331 337 413 # ³ g d d g # d ³ 663 738 d g 669 d d g g 405 d 630 703 71 816 836 655 677 d 320 g 835 408 g g g g g #d 308 ³ gd ³ gd66d gd670 d 513d 510 d gd723d g 739747817gd gd #d 631gd64657d 656g gd 662 ³ gd 503 #gdg305 ³ #gd 616 ³ gd gd511 #gd 617 407 ³ d303 #g³ gd 256 257 ³ # ³ g 680g #gd321gd³ g 409 d g g d d g d g g gd ³ d d d 704 g 829 g 301 d304 309 748 d 332 # g 512 d d g 671 # d g d 665 g d gd672 g 681gd682 #d d632633 g #d gd gd d 410 gd 51 515 #gd ³ ³ gd gd 658 42 661 673 gd 664 302 gd 724gd 740 #gd ³ gd 659 gd 418 619 618 gg gd 417 dgd gd683706 gd g414 415d 416 420 #gd 32322 333 #gd514605 ³ ³ gd 749 411 d g 634 g g 72 678 637 d g d 684 d 310 g g 516 # ³ g d g d 421 gd gd 705 750 gd 620 gd 622d 635³ gd ³ gd gd 660 gd gd67gd gd707 gd # d323324d gd g 431d gd41gd 422 gd g g 419 gd 517 g d # g g g g 606 d 674 gd gd679709gd 685gd 726d725 gd751d 423 g d 432 g gd 675 gd 621 g636 d gd 638640 d 425 gd 427428d gd429d g gd462 601 52d 607 gd gd d gd 424 g 62363g gd 441 442 443 711 728 708 463g gd 752 d d d gd741 g g g gd gd 433 d 426 444 d gdd g gd gdd 43 430 gd465 gd 602 608 gd g 639 641d g gd 438 30 d gd gd434 gd gd 437 g 743742 d 710 g 727 753 440 d 624 712 d 464 445 g gdgdg gd gd gd518 gdg715 gd 439 d446 gd714 #gd gd 625g ³ gd 435436 d gd gd 603 gd754 gd gd 451 ggdd gd713 gd729 d 44447 49 519520 642 gd74d 744 g g 627628 73 d d g g g 755 g d 449 716 d d d 717 610 g g 730 gd gd 45gd 452 453 g g 746 626 d g d d g d 731 # ³ d 609 611g gd gd g 745 gd g 62 629 gd643 gd 734d733 448 450 604 gd455g gd d gd g612 d 756 d d gd614615d gd gd 732 720718 g 757 613 61 454 d 459g 461 d 456 719 gd 735 gdgd 460 79 g d 75 457 458 737 gd gd 736

gd 8

Map 20: Lengthened /o/ in the NAS of /*stoh"/ (a.p. b) 'haystack'

207

40

N

0

1#

3#

2#

106 # 107 #

9

#

#

ç

##

Roads

Rivers

Czech Regions

Major Dialect Boundaries

AV dialect survey locales

7#

40

80

120

160

40#

200

458

##

ççççç ç ç ç ç ççç çç ç çç ç ç ç ç çççç ççççççç ç ç ç ç ç ç ç ççççççç ççççç ç ç ç çç çç çç ç ç ç ç ç ç çççççç çççç ç ç ç ç ç ç ç ç ççç ç ç ç ç ç çç çç çç ç

6#

240

280

320

360

737736 #

400 Kilometers

Legend

For the names of av dialect survey locales, refer to appendix a.

Long /a:/ in the GS of /*gorx"/ (a.p. a) 'peas'

102 # 104 # 101 # # #103105 108 # # 120121 #110 11# 112 # # 10# 113 116 5# # # # # 109 111 149 # 122 # 138 # 114 115 # 212 # # 210 # 124 202#203 # # 204 124# 125 127 # # 117#119 # 211 201 14# # # 209 21 # # # 118 # 139 # 123 128 # 129 # 208 # 213 126 4# # 150 # 206 # # # 207 # 226 # 227 140 214# # # 225 95# 130 205 # 151 141 # # # # 229 # 228 94# # 12 216 # 131# # 143 215 142 223 133 # # # # # 24 818# 152 224 # # # 325# # 801 # # 217 153 132 135 144 22# 218 # # # # 231 # 136 146 819 # # # # 326 237 238 23 # # 230# 312 # #219 147 # 644 240 # 14513 92# 247 # # # # 134 137 # 232 # # # 154 91# # 311 802#81 820 # #155 # 242 148 # 327 241 # # # # 334#248 830 313 # # 160# # 239 # # #314 # 328 645 # 156 821## # 831# 84 # # # 647 # 234 807 803 26# # 159 233 221 # 161 # # 335 ## 249 243 804# 810 # #646 220 # 316 # 401 # 252 # # # 162 # #82 93# #158 # 809 # # # # # 832 805 # 811 # 648 649650 306 59# 235 # 157 250 # 315# # 329 # 236 # # 651 244 222 # # 652 31 # 336 823#825 812 # 27 # 65#667 501 # 251 833# # # # # # # 701 403 # # 25 317 # 806 # # 245# # 330 412 # # # 668 502 # #826 253 834# # 402 # 654 318 # 307 # # # 702 # 406 # 319 653 # 506# 83#815824# # 254 507 # 721 666 # 676 # # 246 504 # 404 # 337 # 53 508 # # 827# # 509 255 # # 413 # # 630 # 331 # 663 669 # 405 ## 703#722 71 655 677 # 320 408 # # 656 # # #835 # # # # # 256 # 828 321 # # 64# 738 # # # # 407 510 # 305308 # 836 # 616 # 503 680 66 # # 409 617 # 704 # 513# # 511 257 # 301# # # 309 # 332 # 631# 657 665#662# # # 670 # #723 739 748 # # 829 410 # # # 681 # # 682 42 # # # 632 414 # # # 51# #514 32#322 618 # # 415 # # #304 # 418 619 512 # 672 # # # # # 740 420 661 # 749# 411 # # 633 # 724 # # 634 333 515 417 673 # 683 # # # 310 302 416 # 421 605 # # #706 72# 750 # # # # 422 # 637 659 664 # 67 516 620 419 # # # 431 # 622 324 684# 725 705 # # # 674# # 635# 517# 601 606 41 323 638 # 423 660 621 # # #441 427 # # 685 # # # 462 # # # # 432 425 # 751 52# # # # 442 # 675 741 428 # 640 # 623# 63 # # 607 # 429 # 424 # # # 463 # #679 709 # #708726 # 426 443444 43# 430 # # # # # 433 30# 639 742# 752 465 # 711 710 434 # 743 602 # 608 641 # # # # 440 # 438 # 624 # 437 # # # 464 # # 445 # # 715 # 518 # 727 753 # # 625 # # # 435436 # # 439 44# 446 642#73 ## 712 # # 744 49 # 519 603 451 627 # #520 755# 754 # 628 74# # # 449 # 45 # # 610 # # 713 # # 626 453 # 447 # # # 609 452 611 # 62# # 629 # # 614 716 # 745 448 604 643 # 756 # # # #450455 # # 717 # # 731 # 459 # 732# 757 720 612 ## 61# #615 # #718 454 # # 456 # 460 461 # 733 # # # # 79 # 613 75 719 735 457

8

#

Map 21: Long /a:/ in the gen. sing. of /*gorx"/ (a.p. a) 'peas'

208

100

N

gd 1

gd 9

AV dialect survey locales

gd

Czech Districts (okres)

Major Dialect Boundaries

Rivers

Lengthened /o: > uo > u/ in *kosj' (a.p. b) 'basket'

ç

0

100

200

300 Kilometers

g gd gd gd Legend gd104d 106 gdd gd gd 102 gd g d 7 103 g 107 6 101110 g 105 g 108 g d d d g gd gd gd gd d d gd 109 g gd 120121 111 11 112 10 113 5 For the names of g gd116 d g g 149 g g d gd gd117 g d g d d g d av dialect survey locales, çd 203dg 204gd210gd gd 212 gd 138 122gd 123 124gd124 gd gd 114115 gd gd gd202 g refer to appendix a. d 21 211 gd d d118d119 201208 14 125 127 g 139 g gd g 209 128 g g d d g d d g d 129 4 gd gd çgd206207çgd 225gd226 gd227229gd 213 214gd150gd 151gd 140gd 126141gdgd 95 gd 130 205 g d gd 228 94 gd 216 gd12 gd g 131g g gd g 3 gd d 133d gd 215 gd24gd 223d gd 818 gd gd 152d gd 142 143 gd gd132 gd 224 gd 135gd 325 g 801 gd d 217 153 g g 22 144 218 ç g g d 237d g d 134 gd231 g 23d gd gd g 146 819 g 136 d g d g g g 326 d d d 238 g d g d 230 d 644 ggd g ggd820d çgdgd312gd gd g gd247gd gd 240 gd 242gd 232d 219 154gd 155gd 14513147 92 gd 137 g802 gd69147d 311ç gd gd d d 81gd807 gd148 d 808 g g gdgd830 327 d 334 241 g d d g 803 646 d 313314 239 160 g d 831g gd 234 gd gd 156gd gd d gd gd 821 gd gd 328 gd 248 645g 822 d84gd g d g d d g d g g 26 g g ç d g g d d 159 gd d g 233 221 d 832 d 648 809 315 d 335 g 249 243 804 220 g 161 d 82 157158 649 d 252 g 93 810 gd g g g g 2 çgd 317gd31g316d gd329çd çgd 401 g g 250çgçdgd 162 g d gd 235 805 gd gd gd 650 gd 236d 306 gd gdg 59 gdgd gd gd gd811dgd 812 d 823dg825 gd806 244 d d 222 g 652 651 65g 330 336g ç g 27 d 501 d 667 d 251 d g 833 g d 826 701 d gd 25 403 gd ç gd ç d gd g g d gd ç d d g g 318 ç 813 666 668 245 d g d 824 g g 412 502 g g g d d d 834 253 d d g g 654 g 402 721 g 506 g ç d d g g d 307 d g 702 505 d g g d ç 406 d d g d g d d 653 g 83 254 d d g 827 319 g ç 507 g d g g g d 676 d g 246 504 g g g gd828 gd gd gd816 gd817 çgd çg321çgd 320çd 331çgd337 413 404 405d 407gd 408gd gd 255 g 256çgd çgd gd gd510gdg53d gd509 gd 508630gdd 631gdg64gd655657ggdd 656gd g gd663gdd gd670g669d d 677gd 71d gd gd703dgd722738d739g814815 gd308 ç d 747 gd 305 ç gd 836 835 gd303 g ç d d g 513 d d 723 503 616 g 680 66 d g 409 662 g d g g g d 617 g gd d 257 g 511 g d g g ç gd304 d d d 704 g 829 301 ç 748 332 d 512 g d d 671 d g d 665 d g g g 309 g 410 d ç d 632 d gd gd 682 çgd 51 515gd514605gd 619gd 618 g634dgd 637633gd 659658gd 664661 673d672 678gdg gd681683gd706 302 gd 724gd 740 418 çgd 32322dgd 333gd 420ggdd 421ggd gd414 415gd 416gd 41742gd 411 g g 72 g d d g 684 d d 310 g ç 516 g d g g 635 g d çd çgd gd431 gd41d 422 423gd g gd 427gd gd gd 419 462gd 517gd 601gd gd606d g 620 621gd 622 g636dgdgd 638gdgd660 gd674dg679gd70967gd 685gdgd707d gd725gd 705gd gd750d 749 323324 d gd 708726 741751gd gd d 728 52 675 711 gd g 424d 425 gd 428gd429gd 463 gd gd432 433 640 62363g 607 gd 441 442 443 g d d gd gd 752 g g g d gd gd d 444 g 426 d g g d 43 g g g g d g 30 d d 639 641 gd 712d 710 çd d446çgd434435gd436çgd 437çd g 438d 439gd440 430 464gdgd465g518d 602g 608 743742 gd 753d 727 624 g g 445 g g g d 715 g d d g d g ç d 754 d 625 gd gd 603 gd gd d714 çgd 449çgdg gd 451d gd gd gd gd gd713 gd73716 gd729 44447 49 519520 642 gd gd744 gd 627gd 628 gd gd626 755 gd 610 gd 609 717 45 730 gd 450 ç gd 611 74674 g ç 453 d d g d g 731 d 62 d g 452 g g ç g g d g d 734733d gd gd 745756d 629 448 604 gd643 g612 çgd455gd 459çgdçgd 461gd d d gd614615d 718 d gd 732 757 720 g g 454 ç 613 61 d 456 719 460 gdç 79 gd gd75gd 735 457 458 737 gd gd 736 40

gd 8

Map 22: Lengthened /o/ in the nom. sing. of /*kosj'/ (a.p. b) 'basket'

209

100

N

g2d

0

gd

302

gd

hu:sa 303

301

çgd 304

çgd

hu:sa

308hu:sa

100

hu:sa 305

çgd çgd

gd 307

gd

306

gd

310

gd

309

gd

321

gd 319

gd 313

gd 323

gd

gd 320

200

324

hu:d'e

çgd

318

gd gd 322

32

gd

gd

gd

333

gd

327

330

gd 331

441

gd 316

gd gd 31

317

gd 315

314

gd

gd 442

gd

332

gd 329

gd

444

gd

412

gd 335

gd

432

gd

334

gd

gd 41

421

gd

413

gd

402

gd 401

gd

248

300 Kilometers

gd

gd

431

gd

420

gd

337

gd

336

gd 328

gd

433

gd

422

gd 25

Map 23: Lengthened /u/ in the nom. sing. of /*hus(a)/ (a.p. a) 'goose'

414

gd

ç

Legend

For the names of av dialect survey locales, refer to appendix a.

area of detail

Czech Districts (okres)

Cities

Roads

Rivers

Major Dialect Boundaries

AV dialect survey locales

Lengthened /u/ in NS /*hus(a)/ (a.p. a) 'goose'

210

100

1

gd

212

149

#gd ·

sa:ze

9

2

3

305 301303 302 304

0

308

307

306

310

318

325

442

336

443 444

431

248

445

447

100

44

434

414

454

457

456

455

45

459

460

438

14

108

43

200

461

453

440 439

428

102

109

104

219

464

49

125

221

10

113

117

116

516

520

518

604

610

607

612 613 61

611

608

52

133

160

614

134

135

59

136

137

644

646

653

654

652

648

645

626

624

622

637

628

79

629

625 627

639

62363

636

634 635

640

720

75

719

643 718

641

638

659

630 631 655 656 64 657 616 632 658 633 618

609 62

615

620

502

162

508

621

617

507

161

619

159

148

300 Kilometers

603

602

601

606

605

510 513 512 511

515 514

503

517

51

146

132

131

130

129

119 118

147 145 13

501

506

157 158

505 504 53 509

222

128

114 115

144

127

143

156

12

141

126

142

154 155

246

519

465

462

411

124

153

124

112

105

257

236

220

152

140

123

463

256

419

410

245

27

235

234

23

218

151

139

111 11

101 110103

122

430

429

418

233

244

255

232

217

216

150

138

120 121

107

214

427

409

254

452

426

417

242

243

408

42

451

437

425

407

26

215

231

230

213

253

252

241

229

227

416

450

449

22

424

435 436

423

415

7

211

228

406

251

250

249

240

405

448

239

224

225

210

226

21

204

403

404

446

422

25

238

433

41

421

413

402

401

432

335

223

207

247

237

334

412

420

337

329

206

202 203 201 208 209

205

24

328

326

332

333

331

330

327

441

324

32322

320

323

30

309

321

319

316

317 31

315

313 314

311

312

4

5

6

106

g gd gd gd gd gdd gd gdgd gd gd gd gd gd gd gd gd gd gd gd gd g d g g g d g gd gd gd gd g d g d d g d gd gdgd dgd gd gd gd d g gd g d d g d gd gdgd g gd g gd gd d d gd gd gd gd gd gd g gd gd gd gd gd d g g d g d d gd gd gd gd g gd gd gd gd gd gdgd d g g gd g g d d d d g gd (GP) gd gd gd gd gd sa:z gd g gd d g gd g g g d ·#gd d d d g gd gd d d gd gd gd g gd g g g d d d g gd d gd gd gd d gd gd gd gd gd gd gd gd gd gd gd gd gd gd gd gd gd gd gd gd gd gd gd gdgd g gd g g g d d d gd gd gd g gd gd d gd g gdd gd gd gd gd d g gd g d g g d gd gd gd gd d gd g gd gdgdgd gdd gd gd d gd gd gd gd gd g gd gd sa:zi: (GP)sa:ze g g d g d d g g d sa:ze · g d #d # · gd sa:ze · sa:ze sa:ze gd gd gd gdgdgd g gd ggd gdgdgdgd #gd #gd d gd g #gdg· · # · g d d g gd gd d gdsa:ze sa:ze d gd gd gd gd g gd gdd #gd · #gd sa:ze · gd gd gdd gd gd gd gd sa:ze gd sa:ze g g d # · d gd gd g gd gdd gd gd g gd sa:ze g g g gd ·#gdgd sa:ze ·#gd sa:ze d d g d d g g # · d gd d gd gd d gd g d gd #gd #gdsa:ze · · gdgd g d d sa:ze g d g g g g g # d #d d sa:ze·d gd · gd sa:ze gd gdgdd gd gdgd gd gd gd gd g gd gd #gd · #gd gd sa:ze · g N g # · d d d gd gdsa:ze gd gd gd gd g gd gd gd gdsa:ze #gd · d g g d d g g g g d g gd gd d gd gd gd gd ·#d gd g gd gdd gd gd gd gd gd gdd gd d gd gd gd

8

Map 24: Lengthened /a:/ in the nom. sing. of /*sazja/ (a.p. a) 'soot'

area of detail

Czech Districts (okres)

Cities

Rivers

Roads

Legend

For the names of av dialect survey locales, refer to appendix a.

AV dialect survey locales

gd

Major Dialect Boundaries

Lengthened /a/ in the NS of /*sazja/ (a.p. a) 'soot'

· #

211

40

N

106 107

9

eb

1

Rivers

Roads

Major Dialect Boundaries

AV dialect survey locales

Long /y:/ in the NS of /*lyko/ (a.p. a) 'phloem'

0

1

3

2

40

80

6

120

7

160

200

240

280

320

360

400 Kilometers

102 104 101 103105 108 120121 110 11 112 10 113 116 5 109 111 149 138122 124 114 115 117 212 202203 204 210 124 125 127 14 201 21 211 139 123 128 118 119 208209 213 129 4 126 150 206 226 227 207 140 214 225 95 130 205 151 141 229 228 94 12 216 131 215 142 143 133 24 223 818 152 224 325 801 153 217 218 132 135 22 144 231 146 819 136 326 237 238 23 230 312 14513 147 134 137 644 219 240 92 247 232 91 154155 311 802 81 820 242 148 241 830831 313314 327 328334248 160 239 645 647 156 803 807 821 26 84 159 233 234 220 221 646 335401 249 252 243 804 810 161 158 82 93 316 809 162 832 805 648 649650 329 306 315 59 235 811 157 250 651 244 222 652 31 27 236 336 65667 501 251 701 806 812 823825 833 25 403 317 330 668 245 412 402 502 826 834 654 506 318 307 319 702 406 253 254 653 721 83815824 666 676 246 504 53 509 507508 404 255 827 331 337 413 722 663 669 405 630 703 71 655 677 320 408 656 256 828 835 321 738 407 64 510 305308 836 680 704 409 257 503 513 511 617 616 631 657 665662 66 670 301 332 410 723 739 748829 309 681 42 682 632 414 51 514 512 619 618 32322 418 415 304 420 740 749 661 672 411 633 724 634 333 72 515 310 302 416 417 421 605 659 664 673 67 683 706 705 750 516 620 419 422 431 622 635 637 638 684 517 601 606 41 323324 423 621 660 674 427 685 725 462 432 425 441442 52 751 675 741 428 429 709 640 623 63 607 424 726 463 679 426 752 443444 433 708 43 430 30 639 641 465 711 710 434 743742 602 608 624 437 438 440 518 464 445 715 727 753 435436 625 439 44 446 712 49 519520 603 642 73 451 627628 744 755 754 449 610 45 713 74 453 447 609 62626 452 611 716 745 614 629 448 450455 604 643 717 731 756 732 757 720 718 612 61 615 454 456459 461 460 733 79 719 613 735 75 457 458 737736 40

8

eb Czech Regions Legend bb eb1eebeb1eb 1e eb e eb 1ee1eb e1b1 eb b b 1b 1ebeb 1eb eb eb eb For the names of eb eb 11 av dialect survey locales, 1 1 e e 1 b b e e e e e e b 1 e b e eb b b 1 bb b 1 refer to appendix a. 1eb1eb1b 11eb1eb1eb 1eb 1eb eb e 1eb 1 1eb 1ebe 1eb11eb1 ebebeb eb e b 1 be1 1eb e1eb 1eb 1 1eb1b eb eb 1eb 1ebeb 1b 1eb1eb 1eb 1b 1 eb e e b b eb 1 1 e e e e e b b b eb eb eb b eb 1b 1eb eb e1b 1eb 1eb 1eb 1eb1 1 1eb e1beb1b 1eb1ebe 1e eb 1 b e e 1 e e b b 1 e e b b 1 1 1 e 1b eb 1b 1eb 1eb eb 1ebeb1ebeb 1eb 1eb ebe 1eb1eb1eb 1eb eb 1eb1ebeb eb 1b 1eb 1eb1eb eebb ebeebb eb ebebeb eb e e e b 1 b b e e b 1 1 b e 1 e 1 1 b 1 e b e e e eb e eb e eb 1b 1 eb 1b e eb 1b eb beb eb eb e bb ebeeb eb ebb 1eb eb e e b1 1ebe1ebb 1eb1ebeb1 1b e e 11ebeb 1b 1 1eb 1ebeb 11eb 1eb1b 1 eb1 1eb eb eb eb ebeebeb ebbeb eb ebebebeb eb e e 1 eb e b 1b 1 eb b 1 e e b b e b 1 b eb e eb 1 1b eb 1eb 1 1eb e 1 1eb 1eb e 1eb1 e bb eb eb eb1ebebebebebeb eebb eb eb beb 1eb 1eb1b ebeb eb eb 1ebeb ebeb eb eb eeb e b 1 b b e 1 e e e 1 b 1 e b b e e 1 b 1 1 b 1 e eb ebebe e eb ebeb eb1b 1 eb eb 1ebebeb eeb b 1eb e 1e 1eb1e e b1 1eb e eeb 1b 1eb1eb 1eb1b eb1eeb eb eebb 1b eb1b 1eb1 ebeb eeb 1eebb ebebebb ebebeb e eebebebebeb ebeb eb ebebeb bb 1eb1 e1eb eb1eb e1beb 1eb b 1b eeb1 1eb 1ebe1eb e 1b 1b eb e1eb1eb1b eb eb eb ebbebeb bebeb1eb eb eb eb 1b 11eb 1eb 1ebb 1 e 1b eb 1eb 1eb1eb 1eb 1 e1eb eb eb ebeb eb eb b e 1bebeb 1eb eb eb eb eb b eebebebeb ebeb 1eb 1eb 1 e1eb e eeb 1eb 1b 1eb e eb eb eb 11 ebeb 1eb ebe ebeb eb b eb eb 1 1 ebe 1b 1b eb1b 1e 1eb eb1eb 11 ebb ebbeb 1eb eb ebebeb1ebb eebebeebebeb eb eb eb e b1 1ebbeeb 1eb eb 11eb e b e e e b b b e e e e 1 1 e b b eb b ebebb 1ebeb 1ebb b eeb eb 1b 1e1eb eb e e 1b b 1 e e b b b 1b 1eb 1 1eb 1b eb 1 e1eebb eb 1eb1eb b 1eb eb

eb

Map 25: Long /y:/ in the nom. sing. of /*lyko/ (a.p. a) 'phloem'

212

40

N

0

eb

Þ

Rivers

Roads

Major Dialect Boundaries

AV dialect survey locales

Reflex of retained length in the NS of /*melko/ (a.p. a) 'milk'

1

3

40

2

106

9

Czech Regions

80

120

6

160

7

200

240

280

320

360

400 Kilometers

102104 101 105 107 11 112 10 113 116 5 149 120 108109 111 124 138 114 212 202203 204210 122 117 14 201 211 127 119 123124 115 208 213 21 4 126 150 118 227 125 209 139 128 140 214 206 207 129 226 225 95 151 141 229 205 94 12 216 131 130 142 143 133 228 215 24 223 818 152 224 325 801 153 22 144 146 132 135 218 231 217 819 136 237 238 23 230 312 219 240 145 92 134 137 644 91 820 232 155 154 802 311 326 81 247 241 242 647 314 327 328 13 148 239 645 156 821 830831 84 234 26 248 233 249 252 243 220 221 160 646 803 804 807 158 313 316 93 809 650 82 832 159 329 335 401 306 315 59 648 250 244 23527 236 222 157 501161 162 805 811812 823825 833 31 65667 25 403 317 336 245 251 502 834 253 826 402 652 651 307 319 318 668 702 701 505 254 721 83 506507 508 676 246 504 404 337412 827 255 653656 663 669 405 406 703 71 836 835 320331 509 256 413 308 407 64 305 408409 816 662 66 670 677 680 257 503 53 510 617616 301 332 410 309 657 723 739 42 414 511 51 32322 618 632 418 415 672 420 304 682 748 411 333 72 681 417 310 416 740 749 421 302 514 605 620619 635 659661 673 67 705 419 422 431 750 637 323 516 423 684 706 427 462 324 432 425 674 52 606607 621 741 428429 725 41 63 638 424 622 463 751 517601 426 43 430 640 675 679 30 441 442 444 433434 708 639 465 742 608 752 711 710 438 624623 518 464 641 443 435436 437 727 753 49 519 602603 439 451 625628 712 74744 743 754 445 44 446 610 45 447 60962 716 452 611 755 520 745 614 453 626 643642 455 756 629 604 732731 612 61 615 718 448 456 459 461 733 757 79 719 75 735 613 454 457 460 458 737736 40

8

eÞb eÞb Legend Þ e Þ e eÞb eÞbb eÞb eÞbeÞeÞbbeÞbeÞbeÞb eÞb For the names of eÞÞ b eb eÞb eÞb eÞb eÞb eÞb eÞb eÞeÞb eÞb av dialect survey locales, Þ e b refer to appendix a. eÞb eÞb eÞb eÞ eÞb eÞb b eÞb eÞeÞb eÞÞ b ebb eÞ eÞb eÞeÞeÞbbeÞbeÞb eÞb eeÞbeÞbb eÞbeÞb eÞb ebeÞb eÞb eÞb eÞbb eÞbeÞb eÞb eÞb eÞbeÞbeÞb eb eÞbeÞb eÞbb b Þ b eÞbeÞb eÞ eÞb eÞb e b eÞb Þ e Þ e e Þ e eÞeÞbebeÞb eÞb eb eÞbb eÞb eÞb eÞb eÞb b b eÞb eÞebÞb b eÞb eÞbeÞb eb ebeb eÞb Þ e e Þ Þ b e e b eÞb eÞbbeÞb eÞb eÞ eÞbeÞbeÞb b eÞb eÞb eÞb b eÞb eÞb b eÞb eÞb eÞb eÞ eÞb ebeÞbeÞbeÞb eÞb eÞbeÞbebeebb eb eebbebebeb eeebeb bb ebebeb bbeb ebeb eÞb eÞb eÞbeÞbeeÞbb eÞbb eÞb eÞb eÞbeÞb eÞb ebeÞb eÞeÞb eÞb eÞb eÞbeÞeÞb eÞbeÞÞ b ebeÞb eÞb eÞb eb ebebeeb eb e e eb eÞ Þ b ee eÞb eÞb eÞb eÞb ebeÞbeÞb eÞeÞbb ebeÞebÞb ÞeÞbeÞb eÞb eÞb eÞb eÞbeb b eÞb b eÞbeÞbebeÞbeÞeÞbb eÞb eÞeÞbb ebebb ebebeebebe eb eeÞÞbb ebeÞbeÞeÞb ebebbebebeebb ebeb eb eÞbeÞeÞbbeÞbeÞb eÞb eÞbeÞbebeÞbeÞb b eÞb eÞ eÞb ebeÞ eÞb eeÞbb eÞb eÞb eÞb eÞb eÞb ebeÞbebebebeeÞÞbbeÞb ebeÞb eÞbeÞebb eÞbeÞbebebeeÞbb ebebebebeebbbebebebebb ebebeÞbeÞbeÞb eÞbeÞbeÞbebbb eÞbeÞeÞbebeÞbebÞeb eb eb eÞb eÞeÞb eÞb ebÞb eÞbeeÞb beÞ eÞb eÞb Þ eÞb eÞb ebeb ebeb eb eeÞbeÞbeÞbeÞeÞb eeÞb eb ebeÞbeebÞ eÞbeÞbeÞbeÞb eÞ eÞbebbeÞb eb b Þ eeÞb b eÞ b eÞb eÞb eÞ ebeÞb eÞb eÞbeb ebeb eb b eÞbbeb b eÞb eb eÞbeÞb beÞb eÞbeÞb eÞbb eÞb eÞb eÞb eb ebeÞb eÞÞ b b eÞbb eÞeÞb eÞÞe eÞb b eÞb eÞbeÞbeb eÞb eÞbeÞb eÞb eb eb bebeÞb b eÞbb eeÞb eÞ ebeÞb eb eb eb ebeebebebeb eÞb eÞbeÞbeeÞbbeÞeÞbb eÞÞbeÞebÞeÞebÞeÞbeebÞbeÞb eÞbeÞbeÞbeÞb eÞb eÞb eÞbeÞbeÞb eÞebÞb beÞ eb ebeebebebebebebb eb eÞbeÞÞb eÞbeÞb eeÞbbbb eeÞb eÞb b eÞbeÞÞ eÞb b b eb eÞebÞebb eÞb b beb eÞb eb

Map 26: Reflex of length in the nom. sing. of /*melko/ (a.p. b) 'milk'

213

40

N

0

Roads

Major Dialect Boundaries

AV dialect survey locales

Preserved long /a:/ in the NS of /*bolto/ (a.p. a) 'mud, marsh'

1

3

40

2

106

9

Czech Regions

Rivers

80

120

6

160

7

200

240

280

320

360

400 Kilometers

102104 101 105 107 11 112 10 113 116 5 149 120 108109 111 124 138 114 212 202203 204210 122 117 14 201 211 127 119 123124 115 208 213 21 4 126 150 118 227 125 209 139 128 140 214 206 207 129 226 225 95 151 141 229 205 94 12 216 131 130 142 143 133 228 215 24 223 818 152 224 325 801 153 22 144 146 132 135 218 231 217 819 136 237 238 23 230 312 219 240 145 92 134 137 644 91 820 232 155 154 802 311 326 81 247 241 242 647 314 327 328 13 148 239 645 156 821 830831 84 234 26 248 233 249 252 243 220 221 160 646 803 804 807 158 313 316 93 809 650 82 832 159 329 335 401 306 315 59 648 250 244 23527 236 222 157 501161 162 805 811812 823825 833 31 65667 25 403 317 336 245 251 502 834 253 826 402 652 651 307 319 318 668 702 701 505 254 721 83 506507 508 676 246 504 404 337412 827 255 653656 663 669 405 406 703 71 836 835 320331 509 256 413 308 407 64 305 408409 816 662 66 670 677 680 257 503 53 510 617616 301 332 410 309 657 723 739 42 414 511 51 32322 618 632 418 415 672 420 304 682 748 411 333 72 681 417 310 416 740 749 421 302 514 605 620619 635 659661 673 67 705 419 422 431 750 637 323 516 423 684 706 427 462 324 432 425 674 52 606607 621 741 428429 725 41 63 638 424 622 463 751 517601 426 43 430 640 675 679 30 441 442 444 433434 708 639 465 742 608 752 711 710 438 624623 518 464 641 443 435436 437 727 753 49 519 602603 439 451 625628 712 74744 743 754 445 44 446 610 45 447 60962 716 452 611 755 520 745 614 453 626 643642 455 756 629 604 732731 612 61 615 718 448 456 459 461 733 757 79 719 75 735 613 454 457 460 458 737736 40

8

ßeb ßeb Legend ßeb ßßeebeb eßebebeeebbßebebeb ßeb ßeb e e eßeb b ebb ebb ebeebebßeb ßeb ßeebeebßeb For the names of ßbebbebbebßeb eb b eb ßeb b ßeb ßeb ßebßebb b ebeb av dialect survey locales, e e b refer to appendix a. ßeb ßeebeb ßebeb eb eb ebeb e eb ebe ßebße e b b b e b b eb ßeb eb eb ßebeb ßeb eeb eßeb ßeebebßebe ßebßebßeb ßeb ß e b e eb ebeb b e ßb b b e b e b b ß e b e b ß b ß e e b e e b e e b ßebebbeb eb ebeßbebebb ßebßeb ßebbeßebeb beßeb ebßeb ßeb ßebebßeßbebßeb eb ebebebeebb eb eebbebebeb ebebebeb ebebßebßeb ßebeb ebeb eb eb e bb ebebeb eb ebeb ßebb eeb ebeb ebeb ßeb b eb ßeebb b ebebeßeb eßß ßeb eb ebebe e e e e b ßb e e e e ee e e b b ee ßeb eb eeb ebb ßbebeb eebb ebeßebb ebßeb eb eßeb ßeb ßebb ßeb ßebebebebßebb ebßeebb e ebeeb bebebbebe eeb eeebb b eebebeb ebebbebbeeebb ebb eebe bb ebebebeeebbebebeebb b bebebebebbebebb ebebebebbeb b b ßebeebbebebeb ebebeßbßbeb ebeb ßeb beb ßeb ebb eb ßebßebßeb eßebebßebebebebeeebbeb ebeb ebebb ebebebbe e ßeb ebebßeb eb ebebeb eeb eb e ßebßbe ebebeb eb eb eebebebebeb eebb ebb bebeebeeb ebeebebebeb ebebebeb eb b e b e b bb b b eb ebebßebebb ßebebeebeeßebbebebebßßeebbebebebßeb ebebebb eb ebebebeb ebeebebeb eebb ebebeb ebebebeebb eb ebebebeebeebb ebßebb b ebb ebßebe ßeb ebebeb eb eeb eeebb ebeb ebebebeebb ebeebbeebbebeb ebb eb beebebebebb b eb ebeb ebeb eb eebeb ßeb ebebeb ebb eb b eßbebeßeb eb eebebb eb b b eb eb

ß eb

Map 27: Reflex of length in the nom. sing. of /*bolto/ (a.p. a) 'mud, marsh'

214

40

N

0

ÿ

eb Rivers

Roads

Major Dialect Boundaries

AV dialect survey locales

Long /i:/ in the NS of /*zito/ (a.p. a) 'rye'

1

3

40

2

106

9

Czech Regions

80

120

6

160

7

200

240

280

320

360

400 Kilometers

102104 101 105 107 11 112 10 113 116 5 149 120 108109 111 124 138 114 212 202203 204210 122 117 14 201 211 127 119 123124 115 208 213 21 4 126 150 118 227 125 209 139 128 140 214 206 207 129 226 225 95 151 141 229 205 94 12 216 131 130 142 143 133 228 215 24 223 818 152 224 325 801 153 22 144 146 132 135 218 231 217 819 136 237 238 23 230 312 219 240 145 92 134 137 644 91 820 232 155 154 802 311 326 81 247 241 242 647 314 327 328 13 148 239 645 156 821 830831 84 234 26 248 233 249 252 243 220 221 160 646 803 804 807 158 313 316 93 809 650 82 832 159 329 335 401 306 315 59 648 250 244 23527 236 222 157 501161 162 805 811812 823825 833 31 65667 25 403 317 336 245 251 502 834 253 826 402 652 651 307 319 318 668 702 701 505 254 721 83 506507 508 676 246 504 404 337412 827 255 653656 663 669 405 406 703 71 836 835 320331 509 256 413 308 407 64 305 408409 816 662 66 670 677 680 257 503 53 510 617616 301 332 410 309 657 723 739 42 414 511 51 32322 618 632 418 415 672 420 304 682 748 411 333 72 681 417 310 416 740 749 421 302 514 605 620619 635 659661 673 67 705 419 422 431 750 637 323 516 423 684 706 427 462 324 432 425 674 52 606607 621 741 428429 725 41 63 638 424 622 463 751 517601 426 43 430 640 675 679 30 441 442 444 433434 708 639 465 742 608 752 711 710 438 624623 518 464 641 443 435436 437 727 753 49 519 602603 439 451 625628 712 74744 743 754 445 44 446 610 45 447 60962 716 452 611 755 520 745 614 453 626 643642 455 756 629 604 732731 612 61 615 718 448 456 459 461 733 757 79 719 75 735 613 454 457 460 458 737736 40

8

Legend

eb eb eb eeÿÿbb ebebeeebbebeÿbeÿb eb For the names of eÿb eb b ebebeÿb eb eb eb av dialect survey locales, eb e b refer to appendix a. ÿ e e ebebeebebeb eÿb eÿb ebeb beb ebebebeb ebeeb eÿb bebeÿb e e b b eb eebeb b eeb eb eÿb ebeb eb eb b eÿb e eb eÿbeb eÿbb b b ebeb e eb eb e b eb e e e e eebebeb eb eb ebb eÿb eÿb eb ebb b eb eebÿb b eÿb eÿbeb eb ebeb eb ÿ e e b e e b ebebbeb eb eebebebb ebeb ebb ebeb b eb ebeb e eÿb ebebebeb eb ebebebeebb eb eebbebebeb eeebeb bb ebebeb bbeb ebeb eb eÿb ebebeebb ebb ebeb ebeb ebebeb eeb ebebebeeb ebebebebeb ebeb ebebeeb eb e e eb e b ee eb eb ebeb ebebeb eebb ebeebb ebeb eÿb eb eb ebeb b eb b ebebebebeebb ebeebb ebebb ebebeebebe eb eebb ebebeeb ebebbebebeebb ebeb eb eÿbeeÿbÿbeÿbeÿb eb ebebebebebb ebe eb ebe eb eebb eb ebebeb ebebebebebebeebbeb ebeb ebeebb ebebebebeebb ebebebebeebbbebebebebb ebebebebebebebebebbb ebeebebebebeb eb eb eb e eb ebe eb e eebeb eb eb eÿb eebeb ebb ebeeb beeb eb b ebeÿb eebb e b ebeb e ebeb ebeb ebeb ebb ebebebebeb ebebebeebbebb ebeebb ebebebebbeebbebbebeebbebeb ebebb ebeb eb eb ebb ebb eebee ebb ebebebeb eb ebeb eb eb eb bebeb b ebb eeb e ebebeb eb eb ebeebebebeb eb ebebeebbeebb ebeebeebebeebbeb ebebebebeb eb ebebeb eebb be eb ebeebebebebebebb eb ebeb ebeb eebbbb eebeb b ebe eb b b eb eebebb eb b beb eb eÿb

Map 28: Reflex of length in the nom. sing. of /*zito/ (a.p. a) 'rye'

215

50

N

0

50

100

ß ßß ß ß

ß ß ß VßVß ß ß ß ß ß ß ß

ß

ß

ß

ß

150

ß

V

V

V

200

Map 29: Shortening of acute long vowels in prehistoric czech

250 Kilometers

ß

Length in /*sazje/

ß

Great Moravian Empire 9th Century (Approximate Center)

Length in /*duha/

V

zi:to/zito

ly:ko/lyko

bla:to/blato

hra:chu/hrachu

ji:skra/jiskra

hri:va/hriva

riipa/repa

briiza/breza

216