The Power of Words: Essays in Lexicography, Lexicology and Semantics. In Honour of Christian J. Kay (Costerus NS 163)

The Power of Words

COSTERUS NEW SERIES 163 Series Editors: C.C. Barfoot, Theo D’haen and Erik Kooper

Christian Kay

The Power of Words Essays in Lexicography, Lexicology and Semantics In Honour of Christian J. Kay

Edited by Graham D. Caie, Carole Hough and Irené Wotherspoon

Amsterdam-New York, NY 2006

Cover photo: University of Glasgow Cover design: Aart Jan Bergshoeff The paper on which this book is printed meets the requirements of “ISO 9706:1994, Information and documentation - Paper for documents - Requirements for permanence”. ISBN-10: 90-420-2121-7 ISBN-13: 978-90-420-2121-1 ©Editions Rodopi B.V., Amsterdam - New York, NY 2006 Printed in the Netherlands

Acknowledgements The editors would like to thank Mrs Flora Edmonds and Mr Ian Hamilton of the Department of English Language, University of Glasgow for their invaluable assistance and technical support. Ms Marieke Schilling of Rodopi has been most helpful, as has Professor Erik Kooper of the University of Utrecht and the series editor.

This page intentionally left blank

List of Contents Introduction

ix

Old English colour lexemes used of textiles in Anglo-Saxon England C. P. Biggam

1

Slang terms for money: a historical thesaurus Julie Coleman

23

‘Huv a wee seat, hen’: evaluative terms in Scots Fiona Douglas and John Corbett

35

Lexical splits and mergers: some difficult cases for the OED Philip Durkin

57

Of fæderan and eamas: avuncularity in Old English Andreas Fischer

67

$ho:fian{*}/vK2: a LAEME-based lexical study Roger Lass and Margaret Laing

79

The rhyme potential of Scots Caroline Macafee

91

Of politeness and people Terttu Nevalainen and Heli Tissari

103

ME douten and dreden Michiko Ogura

117

What did Anglo-Saxon seals seal when? Jane Roberts

131

Notes on the medical vocabulary of John Keats Jeremy J. Smith

159

viii

‘Tell her to shut her moof’: the role of the lexicon in TH-fronting in Glaswegian Jane Stuart-Smith and Claire Timmins

171

Forces of change: are social and moral attitudes legible in this Historical Thesaurus classification? Louise Sylvester

185

Key word in context: semantic and pragmatic meaning of humour Irma Taavitsainen

209

Lexicographical Lyrics James McGonigal

221

Notes on the Contributors

225

Tabula Gratulatoria

Introduction This volume is in honour of Christian Kay, Professor of English Language at the University of Glasgow since 1996. Christian’s career at the university goes back to 1969, when she was appointed a part-time research assistant on the Historical Thesaurus of English (HTE), an ambitious project initiated by Professor Michael Samuels. From 1969 Christian worked also as a part-time editor working on the Collins dictionaries, and the two posts started her on a research career in lexicography and lexicology. Christian’s love of collating and categorising things goes back to her Edinburgh childhood, when she was never happier than when putting buttons or sweets into piles according to colour, shape or size. Her schooling took place at the Mary Erskine School, made famous by Muriel Spark, though there is nothing of Miss Jean Brodie about Christian. Christian thrived there and went on to Edinburgh University, where she graduated M.A. in English Language and Literature in 1962. Like many Scots, she decided to see the world: first she travelled to the States, where she completed a postgraduate M.A. in English at Mount Holyoke College, writing a thesis on “Synonym Clusters in Beowulf”. After that she spent three years teaching English at the Folk University of Sweden in Stockholm, where, inter alia, she learned to speak Swedish. On her return to Scotland she took the diploma in General Linguistics at Edinburgh in 1968-69. Luckily for us she forsook Edinburgh, coming to Glasgow in 1969 to work on the Thesaurus as well as for Collins dictionaries, and in 1979 she was appointed to a full-time lectureship in Glasgow’s English Language Department. She has now worked here for thirty-six years and with effect from September 2005 she will be an Honorary Professorial Research Fellow in the Department. Christian has worked tirelessly and with great enthusiasm on the HTE and she has, since Michael Samuels’s retirement in 1989, been the project’s Director. The HTE, a monumental work, which promises to be such a great benefit to scholars in so many different disciplines, will be completed in the next few years. Already a large number of scholars have benefited greatly from using this project’s archives, as many of the articles in this collection witness. Many more visitors, too numerous to mention, have made pilgrimages to Glasgow to consult

x

the research materials assembled here. The largest work that has so far come out of the HTE project is A Thesaurus of Old English (1995, 2000), which is now about to be published again, this time in electronic form. Christian has written a large number of articles on the HTE, semantics and lexicography. She has edited six volumes of essays, which include collections of conference papers on historical linguistics. At the turn of the century she embarked with Jim McGonigal and other colleagues at Glasgow University on the ambitious LILT project (Language into Languages Teaching) which was commissioned by the Scottish Executive Education Department. After its completion Christian toured Scotland introducing the software to teachers who have found it a major benefit in language teaching. Christian has always been at the forefront of computer assisted learning, producing programs such as English Grammar: an Introduction, The Basics of English Metre and ARIES: Assisted Revision in English Style. She has been Director of STELLA (Software for the Teaching of English and Scottish Language and Literature and its Assessment), a project in the Computers in Teaching Initiative, for many years and along with Jean Anderson has helped the University of Glasgow remain at the forefront of CALL. She is also a prime mover in the highly successful project, the Scottish Corpus of Texts and Speech. Her administrative abilities, including fund-raising, are legendary. When I succeeded Michael Samuels to the English Language Chair in 1990 Christian graciously guided me in the mysteries of Higher Education in the UK and in even more complex topics such as the administrative workings of the University of Glasgow. She made a superb Head of Department in the periods 198992 and 1996-99, demonstrating her flair for financial management and her admirable common sense and no-nonsense approach to some of the illogical or irrational dictates of central university or government. She also has wide experience of examining at all levels in the university system and is much in demand at international conferences. Christian’s teaching has covered a wide range of topics in English language, from her own subjects of semantics and literary and linguistic computing to general linguistics, the history of English, pragmatics and spoken discourse. She is an extremely caring teacher and many scholars today have her to thank for the encouragement and

xi

assistance she gave them. Many of the contributors to this volume acknowledge this help. I think that this quotation from Matthew Arnold which, although intended for classical translators, fits Christian admirably: she is “eminently plain and direct both in the evolution of [her] thought and in the expression of it, that is both in [her] syntax and [her] words; [she] is eminently plain and direct in the substance of [her] thought, that is in [her] matter and [her] ideas; and finally that [she] is eminently noble.” It is not all work and no play with Christian. She is a keen and knowledgeable supporter of opera and classical music, especially Scottish Opera. It is therefore a great honour and pleasure to prepare this volume of essays on her research areas of lexicography, lexicology and semantics. Graham D. Caie

This page intentionally left blank

Old English colour lexemes used of textiles in Anglo-Saxon England C. P. Biggam Introduction This paper is the first of at least two articles to investigate the Old English (OE) vocabulary of textile colours, in relation to both AngloSaxon manufactures, and imported materials.1 While the present paper concentrates on colour words, the next will be principally concerned with dyes and dye words.2 It is hoped that this semantic investigation will complement the valuable work taking place in the disciplines of 1

I would like to record here my gratitude to Prof. Kay for the years of interest and support she has generously given to my efforts in historical semantic research. Colour has long constituted one of our mutual interests. In all my work, including the present paper, her Thesaurus of Old English, produced with Jane Roberts and Lynne Grundy, has proved an invaluable research tool (Jane Roberts and Christian Kay with Lynne Grundy, A Thesaurus of Old English, King’s College London Medieval Studies XI, 2 vols., London: King’s College London Centre for Late Antique and Medieval Studies, 1995). 2 The principal sources of Old English word definitions in this research are the Dictionary of Old English (DOE), and, for those words which have not yet appeared in the DOE (from the letter G onwards), the dictionaries by Clark Hall, and Toller (Angus Cameron, Ashley Crandell Amos, Antonette diPaolo Healey et al. (eds.), Dictionary of Old English in Electronic Form A-F, Toronto: Pontifical Institute of Mediaeval Studies, 2003; J. R. Clark Hall, A Concise Anglo-Saxon Dictionary, with a supplement by Herbert D. Meritt, 4th edition, Cambridge: Cambridge University Press, 1960; T. Northcote Toller (ed.), An Anglo-Saxon Dictionary Based on the Manuscript Collections of the Late Joseph Bosworth, Supplement by T. Northcote Toller with revised and enlarged addenda by Alistair Campbell, 2 vols., Oxford: Oxford University Press, 1898-1972). In cases where I have published word-studies, I use my own definitions, and provide references. For Latin words, my principal authority is the dictionary by Latham and Howlett (DMLBS) and, for those words which have not yet appeared in the DMLBS, the Oxford Latin Dictionary (OLD), and the dictionary by Souter (R. E. Latham and D. R. Howlett, Dictionary of Medieval Latin from British Sources, Oxford: Oxford University Press, 1975- ; P. G. W. Glare (ed.), Oxford Latin Dictionary, Oxford: Clarendon Press, 1982; Alexander Souter, A Glossary of Later Latin to 600 A.D., Oxford: Clarendon Press, 1949). Definitions have been abbreviated where appropriate.

2

C. P. Biggam

archaeobotany and the chemical investigation of dye traces on surviving textiles. 2. Scope of the Research The results presented and discussed in both this and later papers are drawn from a review of all occurrences of hue words and dye words in the Old English corpus.3 All those references which could definitely or reasonably be related to a textile referent were retrieved, and then classified as belonging to one of four different contextual types relating to the geographical location of the textile: English, exotic, generalized and unknown. For these papers on the situation in AngloSaxon England, I have drawn on the English and the generalized contexts. The latter indicates contexts in which the author implies that a colour or dye usage is universally appropriate, and we must assume that an Anglo-Saxon audience or readership would have interpreted this as applicable to them. The research does not include achromatic colours, namely, the range from black, through all the greys to white, since these colours (and also brown) were available without the use of dyes as, for example, in woollen textiles produced from sheep’s fleeces of these colours. Also excluded are words denoting colour features other than hue, such as DARK, PALE, BRIGHT and others, and textile colours which are not the result of dyeing, such as thread of gold.4 3

Antonette diPaolo Healey and Richard L. Venezky, A Microfiche Concordance to Old English, Toronto: Pontifical Institute of Mediaeval Studies, 1980. 4 I have also decided to exclude OE pæll and pællen, unless they are in association with a colour or dye word. Although the older dictionaries are agreed that this noun and adjective refer principally to a type of garment, they also tend to add senses which imply a specific colour, such as ‘purple garment, purple’ (Clark Hall, op. cit., s.v. pæll). The Old English word derives from the Latin pallium which, in the context of early medieval ecclesiastical dress, was white in colour (Janet Mayo, A History of Ecclesiastical Dress, London: Batsford, 1984, pp. 23-24). Some examples of this garment were made of a costly material known as purpura in Latin, but it is clear that purpura was not always purple in colour, as Dodwell quotes examples of red, white, green, black and mixed-colour purpura (C. R. Dodwell, Anglo-Saxon Art: a New Perspective, Manchester Studies in the History of Art 3, Manchester: Manchester University Press, 1982, p. 146). It is clear, therefore, that the use of Modern English ‘purple’ in the definition of OE pæll/pællen is unsafe, whether it is intended to refer to the colour of the pallium, or to the meaning of purpura. Dodwell defines purpura as a textile word, referring to shot-silk taffeta (colour not specific) (op. cit., pp. 147-50).

Old English colour lexemes

3

3. Textile Types 3.1 Silk 3.1.1 (seolc; geolu) In the tenth-century medical work now known as Bald’s Leechbook, jaundice (gealadl or geolwe adl) is said to cause the body of the patient to ‘turn yellow like good yellow silk’ (ageolwaþ swa god geolo seoluc).5 3.1.2 (godwebb; geolu) See Section 5.1.1 under Garments (Unspecified). 3.2 Wool 3.2.1 (wull; hæwen) See Section 4.2.3 under Straining-Cloths. 4. Textile Manufactures (Excluding Clothing) 4.1 Hangings and Coverings 4.1.1 (wagrift, stræl, hwitel, wæstling; brun, brunbasu) In his prose work De virginitate, the Anglo-Saxon scholar Aldhelm (died 709 or 710) stresses that purity is not sufficient, by itself, to achieve perfection, since it must be accompanied by other virtues. To illustrate his point he reminds the reader of hangings and coverings woven in diverse colours which please the eye much more than a monochrome product. Aldhelm, writing in Latin, uses the words cortina ‘curtain, wall-hanging’ and stragula (for stragulum) ‘bed- or couch-cover, rug, blanket’.6 In two manuscripts of this text, cortina is Under these circumstances, it is not possible to interpret any context, which lacks further elucidation, as involving a particular colour. 5 Oswald Cockayne (ed.), Leechdoms, Wortcunning and Starcraft of Early England, Rerum Britannicarum medii aevi scriptores 35, 3 vols., London: Longman, Green, Longman, Roberts and Green, 1864-66, II, 1865, 10, 106. 6 Rudolf Ehwald (ed.), Aldhelmi Opera, Monumenta Germaniae historica: Auctorum antiquissimorum 15, Berlin: Weidmann, 1919, p. 244.

4

C. P. Biggam

glossed in Old English with wagrift ‘tapestry, veil, curtain’, and stragula is glossed with stræl ‘curtain, quilt, matting’, hwitel ‘blanket’, and wæstling ‘sheet, blanket’.7 Certain of the threads in these products are said to be dyed purpureus in colour, that is ‘purple, crimson or sim[ilar] (the exact shade depending on the technique used)’, referring to the dye obtained from the purpura shellfish. Purpureus is here glossed brun in Old English. Both Napier and Goossens suggest that brun is an abbreviation for brunbasu (although Goossens adds a question mark to his comment), and the DOE agrees (s.v. brun-basu). While it is true that abbreviations are common in these glosses, the expansion to brunbasu in this context may be unwarranted. There is no doubt that the hangings and coverings evoked by Aldhelm are of the highest quality, since he describes the embroiderer’s skill in working pictures, and he continues with a second example comprising the Temple hangings in Jerusalem. This high-quality context may have persuaded earlier editors that brun, traditionally defined as ‘brown’ or ‘dark’, indicates too ordinary a colour, while brunbasu is more impressive, meaning ‘dark purple, purple; red-purple, scarlet’.8 The DOE, however, while also defining brun as ‘of a brown hue, dark-coloured’, continues its definition to include this word’s extended range into ‘purple’ and ‘red’ (among other colours) when glossing Latin words with such meanings. Old English had no basic term for PURPLE and so conveyed this colour with various ‘reddish’ words,9 and the use of 7

Arthur S. Napier (ed.), Old English Glosses Chiefly Unpublished, Anecdota Oxoniensia 4, part 11, Oxford: Clarendon Press, 1900, p. 28; Louis Goossens (ed.), The Old English Glosses of MS. Brussels, Royal Library, 1650 (Aldhelm’s De laudibus virginitatis), Verhandelingen van de Koninklijke Academie voor Wetenschappen, Letteren en Schone Kunsten van België, Klasse der Letteren, Jaargang XXXVI, 74, Brussels: Koninklijke Academie voor Wetenschappen, 1974, pp. 218-19. 8 I must express my doubts about this dictionary sense of ‘scarlet’ for brunbasu. Both elements of this compound term need more research, but it is clear that basu denotes warm (red-based) colours which are vivid, rich and eye-catching. I suspect that the addition of brun- creates a darkening effect to such colours, indicating, perhaps especially, violet, dark purple and crimson. This would parallel the effect of blæ- in blæhæwen which darkens hæwen ‘blue (grey)’. (C. P. Biggam, Blue in Old English: an Interdisciplinary Semantic Study, Amsterdam and Atlanta: Rodopi, 1997, pp. 22339.) Scarlet is not dark, and so appears to be in inappropriate company in the DOE’s definition of brunbasu. 9 Roberts, Kay with Grundy, op. cit., I, 146.

Old English colour lexemes

5

brun to fulfil this function in a translation context suggests it may have had a close relationship with the more impressive warm colours even where translation was not involved.10 4.2 Straining-Cloths 4.2.1 (claþ; hæwen) A number of medical recipes in Old English call for a concoction to be strained or wrung through a blue cloth. In Bald’s Leechbook, treatment for ‘a broken head’ involves boiling some herbs in butter and straining them (seohhian) through a hæwen cloth (claþ).11 In Book 3 of the Leechbook, a treatment for palsy involves mixing coriander powder with a woman’s milk, and wringing it (awringan) through a hæwen cloth.12 Also in Book 3, the preparation of an ear-salve involves soaking several herbs in wine or vinegar, and then wringing (wringan) the resulting liquid through a hæwen cloth into the ear.13 The Old English colour word hæwen has been researched, and defined as ‘blue (grey)’ indicating that ‘grey’ is a less common sense than ‘blue’.14 4.2.2 (claþ; linhæwen) Two further references of related interest to those in Section 4.2.1 occur in another medical text, the Lacnunga, at least part of which dates to the late tenth to mid eleventh century, but which contains much older elements. The first reference concerns a treatment for eyes that are ‘stopped up’, and which recommends taking specified herbs, and dripping (drypan) their juice into the eye through a cloth.15 The second reference is a treatment for ‘dimness of the eyes’ which involves soaking a part of wild teasel in honey, pounding it, and then 10

This may explain the references where brun is used of flowers (DOE, s.v. brun, 1.b). 11 Cockayne, op. cit., II, 24. 12 Ibid., II, 338. 13 Ibid., II, 344. 14 Biggam, op. cit., pp. 115-270. 15 Edward Pettit (ed. and trans.), Anglo-Saxon Remedies, Charms, and Prayers from British Library Ms Harley 585: the Lacnunga, Mellen Critical Editions and Translations 6a-b, 2 vols., Lewiston, NJ, Queenston, Ontario and Lampeter: E. Mellen Press, 2001, I, 6.

6

C. P. Biggam

wringing (wringan) the resulting substance through a cloth into the eye.16 In both these cases the cloth is described as linhæwen. Lin can mean ‘flax’ or ‘linen’. It seems unlikely that a cloth would be described as ‘linen-blue’ (or ‘linen-grey’) since such a description would be redundant if the cloth were made of linen (‘linen-blue linen’) and confusing for other textiles (e.g. ‘linen-blue wool’). The flax-plant probably provides the answer, since the most common colour of its flowers is pale blue, so linhæwen is reasonably defined as ‘flax-flower blue’.17 (See also Section 4.2.3 below.) 4.2.3 (claþ; hæwen) In Book 3 of the Leechbook, a treatment is recommended for ‘small eye’ (æsmæl), and all pains in the eye. ‘Small eye’ has been variously interpreted as shrinkage of the eye, or contraction of the eyeball or pupil.18 The problem is to be treated by chewing wild teasel and then wringing (wringan) the juice through a blue (hæwen) and woollen (wyllen) cloth onto the eyes of the patient.19 Although several British plants can produce a blue dye, there is little evidence, either historical or archaeological, that anything other than woad (Isatis tinctoria L.) (OE wad) was exploited for this purpose in Anglo-Saxon England.20 The blue cloths used in medical treatments, therefore, were most likely to have been dyed with woad, and there may have been a very good reason for this. The woad plant has vulnerary and styptic properties, aiding the healing of wounds and the staunching of blood,21 and its appearance in certain Anglo-Saxon medical recipes suggests these properties were understood.22 It is clear, however, that wound-healing qualities are not especially helpful for some of the problems mentioned above, and it may be that certain

16

Pettit, op. cit., I, 8. Biggam, op. cit., pp. 209-10. 18 DOE, s.v. æ-smæl. 19 Cockayne, op. cit., II, 338. 20 Penelope Walton Rogers, Textile Production at 16-22 Coppergate, The Archaeology of York 17/11, York: Council for British Archaeology for York Archaeological Trust, 1997, p. 1766. 21 Malcolm Stuart (ed.), The Colour Dictionary of Herbs & Herbalism, London: Orbis, 1979, p. 80. 22 For example, Cockayne, op. cit., II, 132 (a burn); II, 36 (an eye ulcer). 17

Old English colour lexemes

7

healers had come to consider a woad-dyed cloth as standard medical equipment. 4.3 Binding-Strips 4.3.1 (wræd; wrætt; read, wræteread) In Book 3 of the Leechbook, a treatment for migraine (healf heafod ece) is given which involves binding plantain roots around the head. The instruction reads bind þa moran ymb þ[æt] heafod mid wræte reade wræde. Cockayne interprets this as an instruction to bind [plantain] roots and wrætt (a plant) around the head with a red band (… reade wræde),23 but others have taken wræte read to be a compound term meaning wrætt-red.24 The latter interpretation would, of course, assume the binding-strips to have been dyed with wrætt dye, and would exclude the presence of wrætt plants. Since the term wrætbaso ‘wrætt-red’ (or ‘-purple’, or ‘-crimson’) is extant, there can be no objection to the theoretical existence of wrætread. Another cure for pain in the head (heafodece), however, lends support to Cockayne’s interpretation. In this instruction, the physician is told to take the lower part of wrætt, put it on a red binding-strip, and bind it to the head (genim nioþowearde wrætte do on readne wræd binde þ[æt] heafod mid).25 In this second cure, wrætt is the only plant involved, and the binding-strip is an unspecified red. It would clearly be safer to interpret the first cure along similar lines, with the exception that two plants were involved in that case, plantain and wrætt, perhaps because migraine is a stronger pain than general headache. Both cures would then involve the plants being tied to the head with red binding-strips, but not specifically wrætt-red. It is quite possible that the writer of these two cures was unconcerned about distinguishing between the wrætt plant and wrættdyed cloth, since it may have been simply the presence of wrætt, in any form, which was important. Schlutter makes a good case for the wrætt-plant being a source of red dye, by considering the evidence for the cognate word in Old High German, rezza or riza.26 This lexeme is 23

Ibid., II, 307. For example, O. B. Schlutter, “Anglo-Saxonica”, Anglia, 30 (N. F. 18) (1907), 23960, 248; Toller, op. cit., see Supplement, s.v. wræt-read. 25 Cockayne, op. cit., II, 304. 26 Schlutter, op. cit., 249. 24

8

C. P. Biggam

used to gloss several Latin words which can indicate red dyes, such as coccum ‘scarlet dye etc.’, murex ‘purple dye etc.’, sandix ‘red dye etc.’ and others. The Old English word wrætt also appears to be identified with madder, probably the best-known red-dye plant in England, in a herbal glossary.27 Wrætt-dye will be further discussed in a future paper (see Section 1). 4.4 Patches 4.4.1 (fihl, fogclað; read) In St Matthew’s gospel (9.16), Jesus is asked why his disciples do not fast regularly, and he offers a metaphor to explain. He says that one does not repair an old garment with a piece of “new cloth” (Authorized Version), as this just makes the split worse. The words used in the Vulgate for this phrase are panni rudis, and rudis has many senses, but usually indicates something unfinished or untried. The Old English gloss to rudis in the West Saxon gospels is niwe ‘new’, but, in the Lindisfarne Gospels, the gloss to the phrase is fihles [ve]l fotclað reades. Fihl is defined as ‘cloth, rag’, and fogclað as ‘‘joining-cloth’, patch’ (fotclað is a manuscript error). It would appear that the glossator of the Lindisfarne Gospels mistakenly understood Latin rudis as meaning ‘red’, and has produced a puzzling sentence implying that a red patch makes a split in a garment worse. Clearly, his attention was on the sense of the individual words rather than the meaning of the sentence.28 5. Clothing 5.1 Garments (Unspecified) 5.1.1 (gewæd; geolu) The solution of Riddle 35 in the Exeter Book,29 and the almost identical Leiden Riddle30 is ‘mail-coat’.31 The mail-coat is the 27

J. Richard Stracke (ed.), The Laud Herbal Glossary, Amsterdam: Rodopi, 1974, p. 42. Line 728 reads Grias .i. medere. wrette. See also Stracke’s comments on p. 97. 28 I am grateful to Roy Liuzza for discussing this passage with me. 29 George Philip Krapp and Elliott van Kirk Dobbie (eds.), The Exeter Book, The Anglo-Saxon Poetic Records: a Collective Edition 3, New York: Columbia University Press, and London: Routledge, 1936, p. 198.

Old English colour lexemes

9

narrator, and admits it is a garment (gewæd) but presents a series of statements which are intended to confuse those attempting to solve the riddle, one of which is that it is not made of any textile. The mail-coat further states that it has not been woven by worms which decorate, with great skill, the fine yellow (geolu) cloth (godwebb). Aldhelm’s Latin original of this riddle refers to ‘Chinese worms’ (Seres … vermes), making it sufficiently clear that silk is the fabric in Aldhelm’s mind.32 5.1.2 (hrægl, reaf; hasufag) Riddles use several ploys to mislead and confuse the listener, so that the solution to the riddle is more difficult to guess.33 Riddle 11 of the Exeter Book begins with the two lines ‘My clothing (hrægl) is hasufag; ornaments bright, red and shining [are] on my garment (reaf)’.34 Suggested solutions to this riddle include ‘night’, ‘gold’ and ‘wine’, but only ‘gold’ could result in the ‘garment’ being any form of textile, namely, a purse, although the purse could just as easily be made of leather.35 This riddle, has, however, been included in this paper solely on the basis of its deliberately misleading introduction, which encourages its audience to envisage a person’s clothes, before the later clues cause them to consider an alternative solution. Hasu usually means ‘grey’ but can also mean ‘grey-brown’,36 and fag indicates ‘particoloured, variegated; discoloured, stained, marked’, so the compound word suggests either a garment of various shades of grey and/or brown, or a garment stained with grey and/or brown marks. Anglo-Saxon sheep were not only white, but also brown, black

30

Elliott van Kirk Dobbie (ed.), The Anglo-Saxon Minor Poems, The Anglo-Saxon Poetic Records: a Collective Edition 6, New York: Columbia University Press, and London: Routledge, 1942, p. 109. 31 Krapp and Dobbie, op. cit., pp. 340-41. 32 Ehwald, op. cit., p. 111. 33 For riddling techniques, see Archer Taylor, English Riddles from Oral Tradition, Berkeley and Los Angeles: University of California Press, 1951; Nigel F. Barley, “Structural aspects of the Anglo-Saxon riddle”, Semiotica, 10 (1974), 143-75. 34 For the Old English text, see Krapp and Dobbie, op. cit., p. 186. 35 C. P. Biggam, Grey in Old English: an Interdisciplinary Semantic Study, London: Runetree, 1998, pp. 281-87. 36 Ibid., pp. 272-304.

10

C. P. Biggam

and grey,37 so particoloured garments of these colours could be produced relatively cheaply with no dyeing involved. Alternatively, the ‘discoloured’ sense of fag may be foremost in this passage. In either case, it would seem that the audience of this riddle gained a first impression of cheap clothes, or a working-garment, perhaps stained with dirty marks, and this image would have been immediately disrupted, in typical riddling fashion, by the following mention of apparently costly red and shining adornments. The third possibility is the only one which causes this reference to appear in this paper. The garment may be dyed brown to produce a sober-coloured item appropriate for church (see Sections 5.2.1 and 5.6.2). The following mention of shiny adornments would provide the same confusing contradiction as in the previous interpretation, since they are not appropriate for a religious context. 5.1.3 ((ge)scierpan; read) Much of the content of the Late Anglo-Saxon homily sometimes known as Napier XLIX, concerns those who are wealthy, encouraging them to carry out charitable work, and reminding them of the temporary nature of earthly riches. The examples given of such riches are the brightest gold, the costliest gems and clothes of the reddest fine cloth (þeah we us scyrpen mid þam readdestan godewebbe) (Scragg’s edition).38 Nine versions of this homily survive,39 and two of them40 exhibit an interesting change of wording: ‘brightest gold’ and ‘reddest fine cloth’ have been altered to ‘reddest gold’ and ‘brightest fine cloth’. Scragg suggests that certain scribes thought gold was more likely to be red than fine cloth,41 but the phrase ‘red gold’ is, of

37

M. L. Ryder, “Medieval sheep and wool types”, Agricultural History Review, 32/1 (1984), 14-28. 38 A. S. Napier (ed.), Wulfstan: Sammlung der ihm zugeschriebenen Homilien nebst Untersuchungen über ihre Echtheit, Sammlung englischer Denkmäler in kritischen Ausgaben 4, Berlin: Weidmann, 1883, p. 262; D. G. Scragg (ed.), The Vercelli Homilies and Related Texts, Early English Text Society os 300, Oxford: Oxford University Press, 1992, p. 209. 39 Scragg, op. cit., pp. 191-95. 40 Ibid., pp. xxx-xxxi (sigla J and K). 41 Ibid., p. 217.

Old English colour lexemes

11

course, familiar from other Anglo-Saxon texts and may have come easily to mind.42 Anderson suggests that the reason for this phrase is that OE read (and its Germanic cognates) retained colour senses from their IndoEuropean origin in a term for earth colours, especially ochre and haematite which, depending on their treatment as pigments, can produce colours such as red, reddish brown, orange and reddish yellow.43 I agree with Anderson that OE read would have included ORANGE in its coverage, although I have suggested that fire, rather than ochre/haematite, was the earliest prototype for the warm colours.44 Old English, and its predecessors, had no basic colour term for ORANGE,45 so, until this emerged, the colour orange would have been considered a type of red and/or yellow, depending on the composition of any particular shade. The phrase ‘red gold’ provides evidence that much of the colour denoted by Modern English (ModE) orange was considered a form of red by speakers of Old English. It is clear that the Old English RED category differed in its extent from the equivalent category in the modern language. The crucial question arising from the ‘reddest fine cloth’ in Napier XLIX is, therefore, whether Old English speakers envisaged the ‘reddest’ part of the category as being the same as in present-day English. There is very little evidence to help answer this question, but an indication is found in a glossary entry in which two Old English compound colour terms, geoluread ‘reddish yellow’ and geolucrog ‘saffron yellow’ translate three Latin words: flavus ‘pale yellow, golden’, fulvus ‘tawny (from dull yellow to reddish brown), golden-yellow’ and rubeus 42

Earl R. Anderson, “The semantic puzzle of ‘red gold’”, English Studies, 81 (2000), 1-13. 43 Ibid., 10-11. 44 C. P. Biggam, “Prototypes and foci in the encoding of colour”, in Christian J. Kay and Jeremy J. Smith (eds.), Categorization in the History of English, Amsterdam Studies in the Theory and History of Linguistic Science, Series IV: Current Issues in Linguistic Theory 261, Amsterdam and Philadelphia: John Benjamins, 2004, pp. 1940. 45 Brent Berlin and Paul Kay, Basic Color Terms: their Universality and Evolution, Berkeley and Los Angeles: University of California Press, 1969. The lack of a basic term does not preclude the use of non-basic terms, such as compound terms, or of descriptive phrases, to denote a particular hue, but, in addition, such hues can usually be denoted by an existing basic term with broader coverage than its modern equivalent.

12

C. P. Biggam

‘red’.46 The presence of rubeus is a little surprising, although it could be argued that the Anglo-Saxons would have understood this word was not ruber, the basic Latin term for RED, and might, therefore, have taken it to represent a mixed colour like the others in this glossary entry. The significant point, however, is that a mixture of red with yellow, while apparently included within the boundaries of the red category, was not at the focus of that category, or a compound hue term would not have been necessary. This may suggest that the reddest red for an Anglo-Saxon was an unmixed, fully saturated, or vivid red, as it is today. 5.2 Garments, Female (Unspecified) 5.2.1 (hrægl; brun) An Old English penitential text usually known as the Confessionale pseudo-Egberti, possibly dating to the late ninth century,47 includes in its list of sins and points of guidance for the use of confessors, a statement concerning the colour of clothes (hrægl) which women must wear to Mass, as decreed by St Basil.48 The colour word used is brun. As discussed above (in Section 4.1.1), brun can gloss Latin words meaning ‘purple’ and ‘red’, but, more commonly, it means ‘brown’ or ‘dark’, and the latter pair of meanings would be more appropriate for this context. Corroboration of this view comes from the Latin text of the Penitential of Theodore, one of the sources upon which the Confessionale draws.49 It states that women may receive the host under a veil, as Basil decided, and the Latin colour word used of the veil is niger, meaning ‘black’ or ‘dark’.50 It is debatable whether an 46

Robert T. Oliphant [ed.], The Harley Latin-Old English Glossary Edited from British Museum MS Harley 3376, Janua linguarum, studia memoriae Nicolai van Wijk dedicata, Series practica 20, The Hague and Paris: Mouton, 1966, p. 187. 47 Allen J. Frantzen, “The tradition of penitentials in Anglo-Saxon England”, AngloSaxon England, 11 (1983), 23-56, 42. Frantzen refers to this text as the ‘Scrift boc’ and argues against a previously suggested late tenth-century date (ibid., 49, where this text is referred to as the ‘Confessional’). 48 Robert Spindler [ed.], Das altenglische Bussbuch (sog. Confessionale PseudoEgberti): ein Beitrag zu den kirchlichen Gesetzen der Angelsachsen, Leipzig: Bernhard Tauchnitz, 1934, p. 189. 49 Frantzen, op. cit., 44. 50 A. W. Haddan and W. Stubbs (eds.), Councils and Ecclesiastical Documents Relating to Great Britain and Ireland, 3 vols. in 4, Oxford: Clarendon Press, 1869-71,

Old English colour lexemes

13

Anglo-Saxon reader who was not familiar with the Latin version, would interpret brun in this context as ‘brown’ or as ‘dark’ (of any hue). 5.3 Garments, Royal 5.3.1 (godwebb; tirisc) In Bede’s metrical Vita Sancti Cuthberti, he relates how, after the death of King Ecgfrith of Northumbria (685 A.D.), Aldfrith returned from his studies under Irish scholars to become the next king.51 Bede’s phrase for Aldfrith’s royal state is Tyrio …in ostro, and this has been glossed in Old English as in tiriscum godwebbe.52 The place-name refers to Tyre, now in Lebanon, and a location once famous for the production of so-called ‘Tyrian Purple’ (Latin purpura), a dye extracted from three types of whelk.53 Bede seems to be clearly echoing the Roman phrase for the accession of a new Emperor, namely, ‘taking the purple’, so it is possible that the phrase was not to be taken literally. The following discussion, however, presumes there was a real tirisc royal garment. The Old English glossator translated ostrum ‘purple garment’ (in this context) as godwebb ‘fine clothes’. Owen-Crocker defines godwebb as ‘something made of precious cloth, frequently purple, normally of silk; probably shot-silk taffeta’.54 Since the Old English phrase means something like ‘in Tyrian fine clothes’, the exact force of the qualifier, tirisc, is difficult to assess. It could simply mean that the fabric of the royal garment was believed to have come from the III, 1871, 196; John T. McNeill and Helena M. Gamer, Medieval Handbooks of Penance: a Translation of the Principal libri poenitentiales and Selections from Related Documents, Records of Civilization: Sources and Studies 29, New York: Columbia University Press, 1938, p. 205. 51 Werner Jaager [ed.], Bedas metrische Vita sancti Cuthberti, Palaestra 198, Leipzig: Mayer & Müller, 1935, p. 99. 52 Herbert Dean Meritt, Old English Glosses (a Collection), The Modern Language Association of America: General Series 16, New York: Modern Language Association of America, and London: Oxford University Press, 1945, p. 17. 53 John Peter Wild, “The Eastern Mediterranean, 323 BC – AD 350”, in David Jenkins (ed.), The Cambridge History of Western Textiles, 2 vols., Cambridge: Cambridge University Press, 2003, I, 102-17, 115-16. 54 Gale R. Owen-Crocker, Dress in Anglo-Saxon England, revised and enlarged ed., Woodbridge: Boydell Press, 2004, p. 334.

14

C. P. Biggam

Levant, but the fame of Tyrian dye makes it more likely that it combines semantic features concerned with the high quality and colour of that dye. The final hue resulting from dyeing with Tyrian Purple varied, as has been mentioned already (see Section 4.1.1), according to the particular species of whelk used, whether and how their dyes were mixed, and how long the dyed cloth was exposed to the light when drying.55 The colours most usually associated with this process are those that were the most highly prized in the Classical world, namely, violet, purple, crimson and scarlet, and this range, or part of it, may have been included in the semantics of OE tirisc. 5.4 Garments, Monastic (Inappropriate) 5.4.1 (brunbasu) The scholar Aldhelm wrote both a prose and a metrical Latin text in praise of virginity, and presented the prose version to the nuns of Barking. He makes very clear his disapproval of those nuns who wear colourful and ornamented clothes, describing certain fine garments as ‘coloured with precious dyes of purple tincture’.56 Aldhelm uses the Latin phrase purpureae tincturae,57 and purpureus has been glossed with OE brunbasu.58 Basu ‘purple, red, crimson’ is especially associated with vivid and deep (i.e. darkened vivid) varieties of these hues, occurring in much-admired contexts such as phoenix feathers, illuminated manuscript letters, luxury fabric, and gem-stones. The DOE defines the compound term, brunbasu, as ‘dark purple, purple; red-purple, scarlet’. I have suggested (in footnote 8) that the sense of ‘scarlet’ is the least likely, since the force of brun- in the compound term is probably to add darkness, just as the blue element in the colour purple is seen to have a darkening effect on the red element. It seems, therefore, that purple, violet,59 and possibly crimson,60 are the most likely colours of the garments disapproved of by Aldhelm.

55 Wolfgang Born, “Purple in classical antiquity”, Ciba Review, 4 (Dec. 1937), 11117, 112-13. 56 Michael Lapidge and Michael Herren (trans.), Aldhelm: The Prose Works, Ipswich and Totowa, NJ: Brewer, 1979, p. 124. 57 Ehwald , op. cit., p. 314. 58 Goossens, op. cit., p. 468; Napier (1900), op. cit., p. 130. 59 By violet, I mean a mixture of red and blue in which blue is perceived as dominant.

Old English colour lexemes

15

5.5 Copes and Chasubles 5.5.1 (mæssehacele; grene) The Visio Leofrici is a short text which recounts several vision experiences of the pious Leofric, Earl of Mercia (died 1057). On one occasion, Leofric was praying in Christ Church, Canterbury when he saw himself standing in the middle of the floor, with arms outstretched, and dressed for taking Mass, including a green (grene) cope or chasuble (mæssehacele). The garment is described as brightly shining (beorhte scinende) but this may be a feature of the vision rather than a realistic depiction of a fabric he had seen.61 5.5.2 (mæssehacele; brun) A writ of Edwin, a monk of New Minster, Winchester includes details of an agreement between the Old and New Minsters in the town. The text is not authentic in its present, probably late twelfth-century form, but Keynes believes it “incorporates some convincingly circumstantial information”.62 The writ purports to date from 1031-1032, since it claims to have been written when Ælfwine was abbot of New Minster (1031-1057) and Ælfsige was bishop of Winchester (1012 x 10131032), and it refers to an earlier agreement which it claims took place while Æthelwold was Bishop of Winchester (963-84). This agreement is said to have been ratified by Bishop Æthelwold’s gift of a brun cope or chasuble (mæssehacele) to each monastery. Harmer translates this colour as ‘brown’.63 This is a rather sombre colour for a prestigious, and intentionally impressive garment, so it may be that its 60

Basu as a simplex can certainly denote scarlet, and crimson can be seen as a darkened form of this colour which could reasonably be denoted by the compound term, brunbasu. This is not to say that basu could not also have been used, on its own, to denote crimson, since individual speakers must have differed in their usage, as in modern languages, where colours are closely related. 61 A. S. Napier, “An Old English vision of Leofric, Earl of Mercia”, Transactions of the Philological Society, s.n., (1909 for 1907-10), 180-88, 184. 62 Simon Keynes (ed.), The Liber Vitae of the New Minster and Hyde Abbey, Winchester, British Library Stowe 944 together with Leaves from British Library Cotton Vespasian A.VIII and British Library Cotton Titus D.XXVII, Early English Manuscripts in Facsimile 26, Copenhagen: Rosenkilde and Bagger, 1996, p. 101. 63 F. E. Harmer, Anglo-Saxon Writs, Manchester: Manchester University Press, 1952, p. 403.

16

C. P. Biggam

common alternative meaning of ‘dark’ should be understood here, perhaps indicating a colour similar to violet or crimson. 5.5.3 (mæssehacele; geolu) The will of Bishop Theodred of London and Suffolk (died c.952) survives in a much later manuscript, and it includes four copes or chasubles (singular: mæssehacele).64 Theodred bequeaths a geolu ‘yellow’ mæssehacele, which he bought in Pavia, to Odgar. Pavia, in Lombardy, sold Byzantine silks, and was well-known to Anglo-Saxon merchants,65 but there is no information on the fabric of Theodred’s garment. He leaves another yellow (geolu) mæssehacele to Gundwine, and this garment is further described as ungerenod ‘unornamented’. 5.5.4 (mæssehacele; read) Bishop Theodred (see Section 5.5.3. above) also mentions a read ‘red’ mæssehacele, and this sentence contains an unfamiliar word (spracacke) which has been interpreted as the name of a beneficiary. Whitelock, however, finds no similar name in any Germanic language, and prefers to consider the passage corrupt.66 Theodred’s position in the Church, and his ability to buy items in Pavia, suggest that his vestments are of the highest quality, but it may not be so in all cases. The four bequests involving these garments appear to be listed in descending order of value: thus another Theodred receives a white mæssehacele from Pavia (not discussed here), a chalice and a massbook; Odgar receives a yellow mæssehacele from Pavia; Gundwine receives an unornamented yellow mæssehacele, and, finally, the red mæssehacele is mentioned. It is noteworthy that the last garment is probably not in the admired colours of purple, scarlet, violet or crimson, since, if that were the case, I would expect a more specific colour term, such as basu or brunbasu. Read, the basic term for RED, seems to constitute an unenthusiastic description which may indicate the use of a locallyobtained fabric and dye, in contrast with luxury goods from southern Europe. 64 Dorothy Whitelock (ed.), Anglo-Saxon Wills, Cambridge: Cambridge University Press, 1930, p. 4. 65 Dodwell, op. cit., pp. 149-52. 66 Whitelock, op. cit., pp. 4, 103.

Old English colour lexemes

17

5.6 Dresses Modern English dress is used here to indicate ‘a one-piece garment for a woman or girl that covers the body and extends down over the legs’.67 The Old English garment terms appearing in this section are cyrtel and tunece. Old English cyrtel probably originated as a word for a short garment (OE *(ge)cyrtan ‘to cut off, shorten’) and it is certainly used of a man’s short tunic, but Owen-Crocker identifies it with the long, and sleeved woman’s dress seen in manuscript illustrations, and suggests that the semantics of the word had extended to include this long dress, worn by secular women on most occasions.68 I have avoided the use of ModE gown as the COD defines it as a dress for formal occasions only. Old English tunece, as a woman’s garment, occurs very rarely. Owen describes it as similar to a cyrtel, but she presents evidence that suggests it had religious associations and was, therefore, likely to be dull and/or dark in colour.69 As a woman’s garment, the Modern English definition of OE tunece is not tunic since that cannot be longer than the knee (COD). 5.6.1 (cyrtel; blæwen) Æthelgifu was a late tenth-century Englishwoman of considerable wealth. Her will survives, dated to 980-90, and, in it, she disposes of large estates, gold, slaves, horses and garments. She leaves to Beornwynn her dark blue (blæwen) dress (cyrtel) which is untrimmed at the bottom (neaþene unrenod).70 Whitelock translates blæwen as ‘blue’, but I have suggested there is sufficient evidence to prefer ‘dark blue’.71

67

Judy Pearsall (ed.), The Concise Oxford Dictionary, 10th edition, Oxford: Oxford University Press, 1999. Henceforth referred to as the COD. 68 Owen-Crocker, op. cit., p. 217. 69 Gale R. Owen, “Wynflæd’s wardrobe”, Anglo-Saxon England, 8 (1979), 195-222, 203-04, 211. 70 Dorothy Whitelock (trans. and examined), The Will of Æthelgifu: a Tenth Century Anglo-Saxon Manuscript, with a note on the document by Neil Ker and analyses of the properties, livestock and chattels concerned by Lord Rennell, Oxford: Roxburghe Club, 1968, p. 13. 71 Biggam (1997), op. cit., pp. 91-104. The DOE prefers ‘some shade of blue’.

18

C. P. Biggam

5.6.2 (cyrtel; dunn) As explained above (Section 2), this research is not concerned with textiles described as ‘dark’. The colour word dunn has a general definition of ‘dark, dusky’ in the DOE, and the specific meaning of ‘dark’ in the contexts of animals, the onyx stone, and clothing. In the context of landscape features, however, it is defined as ‘dull brown’, and I suggest that it could have had an additional sense of ‘brown’ in the contexts of animals,72 and the onyx.73 If this suggestion is accepted, there seems no reason to suppose that clothing would be the one context in which a meaning of ‘brown’ is excluded, so examples of textiles with dunn have been included in this research. In Æthelgifu’s will (see Section 5.6.1), she bequeaths to Wulfgifu one or more of her dunn dresses (sing. cyrtel).74 If this colour is possibly to be understood as brown, there is a further difficulty in that brown clothing can be produced without the use of dyes, the principal consideration of this research. Brown wool can be spun from naturally brown fleeces, and Lord Rennell expresses confidence that this is exactly the meaning of Æthelgifu’s dunn dress/es.75 If Rennell is correct that the colour of these dresses relates to a natural fleece colour, then this example is not relevant in the present 72

I make this suggestion because OE dunn is used to gloss Latin baius (= badius) ‘bay (horse)’, meaning a brown horse with black points (principally, the mane, tail, and lower limbs). Badius is also glossed by brun ‘brown, dark’ in, for example, the First Cleopatra Glossary (William Garlington Stryker, The Latin-Old English Glossary in MS Cotton Cleopatra AIII, unpublished dissertation, University of Stanford, 1951, 76). In addition, dunn glosses Latin balidus ‘dun, dark brown’. This evidence appears to indicate at least a distinct possibility that dunn could be defined as ‘brown’ as well as ‘dark’ in most contexts (DMLBS, s.v. badius, cross-reference from baius, and balidus). 73 The only example of dunn glossing a word cognate with onyx, in this case, onichinos, occurs in the Leiden Glossary. The OLD defines onychinus as ‘made of onyx marble’, and onyx marble is yellowish green, white and brown (Walter Schumann, Gemstones of the World, New York: Sterling, and London: N.A.G. Press, 1977, pp. 210-11). Kitson suggests that glosses of onichinus with OE dunn and brun may have originated in a confusion of Latin flavus ‘yellow’ with fulvus ‘brown, tawny’ or furvus ‘sombre’. He points out that furvus is used more or less as a synonym for fulvus in Late Latin, and translates OE brun and dunn as well as deorc (Peter Kitson, “Lapidary traditions in Anglo-Saxon England: part I, the background; the Old English lapidary”, Anglo-Saxon England, 7 (1978), 38-39). 74 Whitelock (1968), op. cit., p. 12 n. 12, explains that the Old English is ambiguous as regards the number of gowns involved. 75 Ibid., p. 82.

Old English colour lexemes

19

investigation into dye colours. I have retained examples of dunn textiles in this research, however, as there remains a possibility that they were dyed. Owen makes the point that dull-coloured garments were needed by women when attending church,76 so the dunn dresses, which were sufficiently valuable to be listed in the will of a very wealthy woman, may have been of a finer textile than wool, or of fine white wool, and dyed a shade of brown considered appropriate for religious contexts. Since the principal sense of dunn was ‘dark’, such a colour is likely to have been a dark brown. 5.6.3 (tunece; dunn) In the will of Wynflæd a dunn tunic (tunece) is bequeathed.77 Little is known about Wynflæd, although it can be said that she had an interest in religious houses, and may even have been a lay-abbess.78 The text of the will dates to the mid tenth century, and survives in an eleventhcentury copy. As discussed above (in Section 5.6.2), dunn may refer to a natural, undyed wool colour, or it may refer to a dyed fabric, most likely dark brown in colour. 5.6.4 (hæwen, wæden, weolocread) See Sections 5.7.1 and 5.7.2 under Tunics. The context is Aldhelm’s criticism of the vivid and luxurious clothes worn by certain churchmen and nuns. These include the Latin garment term tunica, which is not glossed in Old English. The OLD describes the tunica as the standard Roman garment for women, reaching to the feet. The word habit, normally used of nuns’ dresses, seems inappropriate in this case, considering the colours involved. 5.7 Tunics Modern English tunic is defined as ‘a loose sleeveless garment reaching to the thigh or knees’ (COD). This seems to be the best term for the standard male garment designated either tunece in Old English,

76

Owen, op. cit., p. 204. Whitelock (1930), op. cit., p. 14. 78 Ibid., p. 109. 77

20

C. P. Biggam

or tunica in Latin, although the Anglo-Saxon garment could extend to a little below the knees, and often had sleeves.79 5.7.1 (hæwen, wæden) Towards the end of Aldhelm’s prose De virginitate, he proceeds to thunder against the vanity and insolence of men and women of the Church, including monks and nuns, who wear adornments and fine clothes. He specifically mentions red and blue tunics (tonica coccinea sive iacintina).80 The Latin word used here for ‘blue’ (hyacinthinus) is glossed, in various manuscripts of this text, with OE hæwen ‘blue’,81 and, in other manuscripts, with OE wæden ‘blue’.82 The latter word is cognate with OE wad ‘woad’, and means literally ‘woaded’. 5.7.2 (weolocread) The red (coccineus) tunics in the same passage as in Section 5.7.1 above, are glossed with OE weolocread, meaning literally ‘whelkred’.83 This compound term is appropriate for the extremely expensive and prestigious Tyrian Purple dye, derived from specific Mediterranean shellfish, but also for the local British shellfish dye, praised by Bede for its beauty, and described by him as coccineus ‘red’ and having a rubor pulcherrimus ‘most beautiful redness’.84 6. Conclusion It is clear from the historical record that Anglo-Saxon England was rich in textiles, both in quantity and quality, and much of this information is found in Latin texts. Dodwell refers to “the vast textile wealth of churches and monasteries”, and provides a review of the evidence in his chapter on textiles which provides an often stunning 79

Owen-Crocker, op. cit., pp. 245-51. Ehwald, op. cit., p. 318. 81 Goossens, op. cit., p. 479; Napier (1900), op. cit., pp. 134, 147; J. J. Quinn, The Minor Latin-Old English Glossaries in MS Cotton Cleopatra A.III, unpublished dissertation, University of Stanford, 1956, p. 164. 82 Napier (1900), op. cit., pp. 163, 170. 83 Goossens, op. cit., p. 479; Napier (1900), op. cit., p. 134. 84 Charles Plummer (ed.), Venerabilis Baedae Historiam ecclesiasticam gentis Anglorum, Historiam abbatum, Epistolam ad Ecgberctum una cum Historia abbatum auctore anonymo, Oxford: Clarendon Press, 1896, p. 10. 80

Old English colour lexemes

21

impression of Anglo-Saxon wealth, and love of luxury and colour.85 It is understandable that it was the exotic and expensive fabrics and colours which were the subjects of written descriptions or comments, and this review of textiles described by colour from the Old English records shows a similar picture. A large proportion of the references come from the wills of wealthy people, and Aldhelm’s criticisms of the inappropriate clothes of men and women of the Church who could afford a degree of ostentation. The ordinary world of natural wool colours and home-dyed textiles is almost invisible, with the probable exception of the medical references. The use of blue straining-cloths and red binding-strips is, at least sometimes, connected with domestically produced textiles (linen and wool), and probably homegrown dye-plants (woad and wrætt), and may involve colour symbolism. This paper, and its brief conclusion, should be regarded as an interim report pending a survey of the dye-names and dye-plant names which will follow.

85

Dodwell, op. cit., pp. 129-69.

This page intentionally left blank

Slang terms for money: a historical thesaurus Julie Coleman A project as extensive and long-lived as the Historical Thesaurus of English (HTE) requires large amounts of external finance, and Christian has often been successful in the tiresome activity of funding application. This paper combines these two closely related fields - the HTE and MONEY - by providing a classification of terms for money listed in cant and slang dictionaries published between 1567 and 1874.1 This is an exploration of the potential usefulness of a historical thesaurus of slang to supplement the vocabulary provided by the Oxford English Dictionary (OED), and hence by the HTE. The following classification includes terms for specific amounts of money, many of which are names for notes or coins. Some of these are also used with reference to money in general. Because many cant and slang lexicographers from this period reproduced material found in earlier dictionaries, only the date of the first dictionary appearance is noted. Variant spellings follow the original form, and are preceded by ‘>’. Some of these are clearly the result of erroneous copying, but a few may reflect genuine variation or change in the spoken language. When the first glossary of cant appeared in 1567, attached to Harman’s Caveat, there were no notes and no coins, as we understand them.2 The economy was based on the value of silver. Changes in the 1

These are the dictionaries discussed in J. Coleman, A History of Cant and Slang Dictionaries. Volume I: 1567-1784, Oxford: Oxford University Press, 2004 and Volume II: 1785-1858, Oxford: Oxford University Press, 2004. I have added material from the various editions of J.C. Hotten, A Dictionary of Modern Slang, Cant, and Vulgar Words, London: John Camden Hotten, 1859, and from J. Greenwood, The Seven Curses of London, London: Stanley Rivers and Co., 1869. 2 Thomas Harman, Caveat or Warening for Common Cursetors, London: William Griffith, 1567. The historical material in this paper is combined from a number of sources: J. Giuseppi, The Bank of England: A History from its Foundation in 1694, London: Evans, 1966, J. Craig, The Mint: A History of the London Mint from A.D. 287 to 1948, Cambridge: Cambridge University Press, 1953, and the excellent websites of:

24

Julie Coleman

price of precious metals affected the value of coins, which were exchanged according to weight rather than at their face value, so names were needed both for coins and for amounts of money. For example, the pound sterling was an amount of money (twenty shillings), whereas the old sovereign (issued 1489-a1625) and guinea (issued 1663-1813) were gold coins originally issued at twenty shillings, but exchanged according to the fluctuating value of the gold they contained, sometimes for up to thirty shillings. Especially for larger coins, distinctions between names for amounts of money and designations of specific coins are not always preserved in slang terminology, which explains some of the muddiness of the classification. The chronology of this classification is based on each term’s earliest appearance in slang dictionaries. Dates in square brackets represent the OED citation range. OED dates for some of the slang dictionaries consulted here are not always the same as mine. For example, I have used c1698 for B. E.’s New Dictionary of the Terms Ancient and Modern of the Canting Crew; the OED uses 1690, c1690, and a1700. I have used the publication date of Vaux’s Memoirs (1819)3; the OED uses the date noted in the dedicatory letter (1812). I have used 1797 for Potter’s New Dictionary of all the Cant and Flash Languages4; the OED uses 1795, the date of a first edition that I have not been able to examine. This means that although the slang dictionary and OED dates do not always tally, the citations are sometimes identical. In addition, it is not always possible to achieve a perfect match between OED and slang dictionary definitions. For example, the OED defines pony as “£25”, but not as “money” or “£50”, the definitions found in the slang dictionaries I have examined. I have not commented separately on each of these near misses, which accounts for apparent omissions of OED material. Chard Ltd (http://www.24carat.co.uk), the Bank of England (http://www.bankofengland.co.uk), Tony Clayton (http://www.tclayton.demon.co.uk/coins.html#Index), and Predecimal.com (http://www.predecimal.com/index.html), which includes a history of British coinage by Ken Elks. 3 J.H. Vaux, Memoirs of James Hardy Vaux. Written by Himself, London: W. Clowes, 1819. 4 H.T. Potter, A New Dictionary of all the Cant and Flash Languages..., 3rd edition, London: B. Crosby, 1797.

Slang terms for money

25

Money; a (large) sum of money; a coin lowre 1567>lower 1610>loure 1665>lour 1673 [lour, lower n21567-1889], shells 1591 [1592 + 1611], coale 1688>cole c1698 [1673-1870]>coal 1820, the ready 1688 [1688-1977], balsom c1698, chink c1698 [1573-1611], cly c1698 [c1690 + 1834 + 1858], cog c1698 [1532-1729], coliander seeds c1698 [c1690 + 1725]>coriander seeds 1785 [1737], crap c1698 [a1700-1787]>crop 1725, darby c1698 [1682-1785]>derby 1811, dust c1698 [(1526 +) 1607a1845], gelt c1698 [a1529-1968]>gilt 1725 [1598-1885]>gelter 1797, gingerbread c1698 [a1700-1864], goree c1698 [a1700 + 1725]>gory 1741, king’s pictures c1698, lurries c1698 [1673-a1700], muck c1698 [?a13251864], quids c1698, recruits c1698 [1662-1818], rhino c1698 [1688-1851], ribbin c1698 [a1700-1812]>ribben 1788>ribbon 1797>ribband 1819, round sum c1698 [1579-1887], Tom fool’s token c1698, blunt 1703 [1812-a1845], spanks 1725 [1725], bit 1753 [1894-1928], iron 1785 [1785-1966], the mopusses 1785 [1699-1980], plate 1785, spankers 1785 [1663-1785], the Spanish 1788 [1788-1869], glanthem 1789, kelter 1789 [1807-1828], rhino of rag 1797, rubbish 1797, stephen 1797 [1812-1834], rag 1809 [1590-1846], rust 1809 [1858], pony 1811, brads 1819 [1812 + 1841], bustle 1819 [1812 + 1830], dibs 1821 [1812-1883], brass 1822 [1597/8-1871], dimmock 1822, the needful 1822 [a1777-1981], rent 1822 [1828-1977], monish 1835 [17811866], pewter 1835 [1829-1888], rivits 1835 [1846-1937], tip 1835, posh 1839 [1830-1905]>pash 1839, meg 1845, denaly c1855, okre c1855 [1854 + 1890], soskin5 c1855, tin c1855 [1836-1854], chinkers 1859 [1834], feathers 1859, greed 1859, rowdy 1859 [1841-1856], stuff 1859 [1775-1896], clink 1861 [1729-c1817] Saved or hoarded money grannam-gold c1698 [a1700]>grannan gold 1779>grannon gold 1780>grannum(’s) gold 1785 [grandam gold [(1598 +) 1700], nest-egg c1698 [1686-1990] Stolen money, or money as the object of theft caravan c1698 [1688 + 1690], cargo c1698 [1690], prey c1698, ring c1698 [a1700 + 1796] Money paid as a bribe oil of palm 1821 [1819 -1935], grease 1835 [1823] A gold coin6 old Mr. Gory 1673 [a1700 + 1725], goldfinch c1698 [1602-1896], yellowboy c1698 [1662-1957], queer ridge 1848, ridge blunt 1848 5

It seems unlikely, given the citation gap, that this is related to the fifteenth and sixteenth-century terms seskyn and suskin “a Dutch coin of the value of six mites” (OED). 6 Terms for ‘gold’ and ‘silver’ in general, such as ridge and wedge, are excluded here unless they are specifically defined with reference to money in the slang dictionaries.

26

Julie Coleman

A silver coin smash 1797 [1795], white 1816 [c1374 + 1390 + c1676 + 1823-1960], queer wedge 1848 A guinea; a sovereign; a pound; twenty shillings heart’s ease 1665 [a1700 + 1785/96], husky lour 1674, job c1698, meg c1698 [1688-c1742], piece c1698 [1616-1727/41], strike 1703, ned 1753 [17531846], canary bird 1785, ridge 1785 [1665-1955 “gold”], yellow George 1785 [1784 + 1785], stranger 1785 [1785], (rum) quid 1789 [1688-1977], good-looking picture 1797, noge 1797, shino 1797, bean 1811 [1811-1928], shiner 1821 [1760-1887], sufferer 1835, cooter 1839>cuta 1845>couter 1851 [couter, cooter 1846-1880], foont 1839, deanee 1845, James 1865 [15671893] Ten shillings; half-sovereign; half a guinea smelt c1698 [1635 + 1688 + (1822)]>smell 1848 (erron), half a slat 1747, half a ned 1789, regent 1821, half-couter 1851, net-gen 1851 Seven shillings spangle 1811 [1811 + 1823] Six shillings and eightpence noble c1698 [1350-1996] Six shillings six bob bit 1821 A crown; five shillings bull’s eye c1698, (hind) coach-wheel c1698 [c1690 + 1812 + 1834], decus c1698 [1688 + 1822], ounce 1725, bord 1741, whore’s curse 1785, bull 1789 [1812 + 1852], dews 1835, ewif-gen 1851, cart-wheel 1859 [1867 + 1885], thick ’un 1865 [1848-1968] Half a crown; two shillings and sixpence fore coach-wheel c1698, george c1698 [1659-1785], slate c1698 [a1700]>slat 1703, trooper c1698 [a1700], half an ounce 1725, half a bull 1789, half bull 1822 [1789-1906], alderman 1839, bull 1848, flatch-ynork 1851, half case 1857, halftusherroon 1859 Thirteen and a half pence loon-slatt c1698 [a1700], hangman’s wages 1785 A shilling; twelve pence bord 1567 [1567-1688], hog 1673 [1673-1875], twelver c1698 [a1700-1732], traveller 1747, mejoge 1753, boar 1754, button 1785, grunter 1785 [1785 + 1858], she-lion 1785 [1785], thirteener 1785 [1762-1836], bob-stick 1789, peg 1797, bender 1809, peg-stick 1809, bob 1811 [1789-1915], breaky leg 1839, deaner 1839 [1839-1946], twelve 1839, tevis 1845, gen 1851, rogue and villain 1857, stag 1857 [1857 + 1887], Abraham’s willing 1859, chinda 1871 Elevenpence leven 1851

Slang terms for money

27

Tenpence jumper 1821, net-yenep 1851 Ninepence ill-fortune c1698, the picture of ill-luck 1785, enine-yenep 1851 Eightpence teaich-yenep 1851 Sevenpence neves-yenep 1851 Sixpence half a bord 1567, half-bord 1665>half-boad 1828>half-head 1828, half a hog 1674, pig c1698 [1622 + a1700], sice c1698>size 1797 [sice, size 1660-1709 + (1830)], simon c1698 [a1700]>smon 1708, kick 1725 [c1700-1871], sibuxom 1747, griff-metoll 1753, cripple 1785 [1785 + 1885], crook-back 1785, syebuck 1785>syeboord 1871, tester 1785>teaster 1821>teaser 1848, crook 1788, bender 1789 [1836-1855], Tilbury 1796 [1796-1812], tizzy 1796 [1804-1946], bob 1797, half a grunter 1809, tanner 1811 [1811-1908], bandy 1821, fiddler 1835 [1846-1885], tinker 1835, downer 1839, lord of the manor 1839 [1839-1972], snid 1839 [1839], sprat 1839 [1839-1902], scab 1845, tawne 1845, hop 1848, exis-yenep 1851, suchera 1871 Fivepence five win 1789, kid’s eye 1821, ewif-yenep 1851 A groat; fourpence flag 1567 [1567 + a1700 + 1851], croker c1698, groat7 1835 [(1351) + 13621885], rouf-yenep 1851, bit 1859, joey 1859 [1865-1884] Three and a half pence yenep-flatch 1851 Threepence treswins 1665, threpps c1698, thrums c1698>thrum 1741 [1699 + 18441933], erth-yenep 1851

As the value of silver rose, it became an uneconomical material for small change. The last silver farthing was minted in 1553, after which production was suspended until the introduction of copper farthings and halfpennies in 1672. Throughout most of the seventeenth and eighteenth centuries there was a shortage of small change. Locally issued tokens were customarily used instead, and accepted as legal tender up to the value of sixpence. Pennies and twopence-pieces, containing their own value in copper, were issued from 1797 onwards.

7

It is not clear why this standard English term was included in a slang glossary.

28

Julie Coleman

After 1860, farthings, halfpennies, and pennies were all produced in bronze. A copper coin mag 1797, brown 1816 [1812-1865] Twopence deuswins 1665, deuce c1698 [a1700 + 1851]>dace c1698>duce 1703, jemmy 1835, owl-yenep 1851 A penny wyn 1567>win 1608 [1567-1900]>wind 1703>wing 1845, brown 1820, georgy 1820, yenep 1851, brown bread c1855, copper 1859 A halfpenny make 1567 [?1536-1618 + 1826-1946]?>meg 1753 [?1747-1988]?>mag 1797 [?1775-1918], baubee c1698 [bawbee 1542-1862], mopus c1698, rap 1785, scuddick 1821, tannie 1821, tonic 1822, copper 1835, flatch 1851, curdie 1871 A farthing; a quarter of a penny grig c1698 [1656/7-1839], jack c1698 [a1700-1873], mopus c1698, rag c1698, token c1698, scrope 1703, meg 1753, fadge 1789 [1789-1873], queer rag 1809, Covent Garden 1859 Half a farthing doit c1698 [1594-1850] A sixth of a penny bodle c1698 [1650-1834]

Among the terms for small change are several that the slang dictionaries define more specifically than the OED, which describes them as general terms for coins of small value. These include mopus [1699-1980], rap,8 scuddick [1823-1901], all of which frequently occur in the construction ‘not a …’. Similarly, the OED defines copper as “copper money … a copper coin” [(1588) + 1712-Mod], while the slang dictionaries define it, variously, as “a penny” or “a halfpenny”. Either the slang lexicographers are closer to these terms’ usage and are thus able to define them more precisely, or else they are unaware of or insensitive to their wider use. Token, which the OED defines as “A stamped piece of metal, often having the general appearance of a coin, issued as a medium of exchange by a private 8

OED 1a: “A counterfeit coin, worth about half a farthing, which passed current for a halfpenny in Ireland in the 18th c., owing to the scarcity of genuine money. Now only Hist.” [1724-1827], also 1b “taken as a type of the smallest coin” [1823-1881].

Slang terms for money

29

person or company, who engage to take it back at its nominal value, giving goods or legal currency for it”, indicates the complex legal status of all small change in this period. Because gold and silver are both relatively soft metals, they quickly became worn in circulation. This made it possible to pass poor quality counterfeits and even blank discs, which accounted for about a sixth of silver coinage by the end of the seventeenth century when they were withdrawn for recoinage. Another method of coin-fraud saw the gold and silver content of coins reduced, sometimes to as little as half of their legal weight, by clipping off small amounts of metal or burning it off with acid. Since small change also contained its value in metal, less ambitious fraudsters specialized in copper coinage. Counterfeit coin(s) swimmer c1698, queer 1741 [1812-1981], button 1785, dump 1788 [18211908], frying-pans 1797, smash 1797 [1795-1860], prince’s money 1813, (half a) case 1839, schofel 1839 [shoful 1828-1856], sheen 1839 [18391890], sinker 1839 [1839-1900], the things 1839, sham 1845, gammy 1857 Counterfeit queer c.1698, gammy 1839, bogus 1848 [1839-19429] A maker of counterfeit coins queer cole maker c1698, queer bit maker 1785, smasher 1797 [1795-1895], bit-smasher 1809, bit cull 1821, face-maker c1855, bit-faker 1869 Counterfeiting money making browns/whites 1816 To pass counterfeit money ring the changes 1811 [1786 + 1874], smash 1811 [1801-1905], work the bulls 1839 One who passes counterfeit money queer cole fencer c1698, smasher 1809, dealer in queer 1822, schofel-pitcher 1839 [1859] Passing counterfeit money smashing 1799 [1812 + 1891], slumming 1839 [1839 + 1888], schofelpitching 1859 To clip money niggle c1698 Clipping money nigging c1698 [1699 + 1796]

9

Only the first citation refers to money.

30

Julie Coleman

The metal clipped from coins spangle 1674, end c1698, nig c1698 [1699 + 1709 + (1725-c1825)], shavings c1698 [a1700], parings 1725, curle 1785 One who clips or otherwise reduces coins niggler c1698, curler 1809, nigger 1835

The Bank of England began issuing banknotes soon after its establishment in 1694, initially for precise sums, but increasingly, as the eighteenth century progressed, with fixed denominations. Until 1855, cashiers still had to fill in the name of the payee and sign each banknote individually. These banknotes could be cashed in for gold and silver coins (i.e. real money), as required. A bank-note; bank-notes (rum/queer) screen 1789 [1789-1864], flimsey 1811 [1824-1845], rag 1811, screave 1821 [1801], leel 1871 Large (of a bank-note) long-tailed 1839 A hundred thousand pounds plumb 1785 [plum 1689/1702-1898] Fifty thousand pounds half a plum 1826 Fifty pounds pony 1848 Ten pound note double finnip 1839>double finnuff 1859 Five pound note finnip 1839 [1839-1966]>finnuff 1859

Slang and cant terms for money have a variety of origins. Terms for coins sometimes refer to the colour (e.g. okre (for ochre), goldfinch, yellow-boy, and canary-bird for gold coins; white for silver; brown for small change) or lustre of the metal (e.g. shiner “a guinea”). Others terms refer to the size (e.g. coach-wheel, cart-wheel, and thick ’un, all ‘a crown’; fiddler, sprat, and scab, all ‘sixpence’), weight (e.g. ounce ‘a crown’; sinker and swimmer both ‘a counterfeit coin’), sound (e.g. chink, chinkers, and clink, all for ‘money’), or quality (e.g. cripple, crook-back, and bender, all for ‘sixpence’ “that

Slang terms for money

31

piece being commonly much bent and distorted”10). Many refer to the value, either directly (e.g. twelver and thirteener ‘a shilling’; sice ‘sixpence’; deuce ‘twopence’), or by mathematical formulas (usually half (a) …). A few appear to refer to the method of production (e.g. smash ‘a silver coin’; strike ‘a guinea’; smelt ‘half-guinea’). Images and inscriptions supply some nicknames for coins. General examples include good-looking picture ‘a guinea’ and king’s pictures ‘coins’. The use of george for both ‘a guinea’ and ‘half a crown’ and of georgy for ‘penny’ may result from St George’s presence on the reverse of many coins. Bord ‘a shilling’ may refer to the shield on the reverse of Tudor and Stuart shillings (Partridge)11. Partridge also notes that hog ‘a shilling’ is “probably ex. the figure of a hog on a small silver coin”, but I have not been able to find any examples. The name decus for ‘a crown’ is from the inscription decus et tutamen ‘ornament and safeguard’. Presumably because of discontinuities in coinage, some denominations were designated by the name of the monarch depicted on them (e.g. James, ?George, ?ned). Regent, for a half-sovereign, is a punning reference to the position of the Prince Regent, waiting for the death of his father before he could become a full sovereign. It appears that the development of a slang sense can have a dragchain effect on related terms. For example, hog is listed with the sense ‘a shilling’ in B.E.’s dictionary in c169812, to be joined by boar in 1754 and grunter in 1785. Half that amount, sixpence, is designated a pig. It is possible that kid’s eye ‘fivepence’ is similarly related to bull’s eye ‘a crown’. Similarly, once iron came to be used for money, it opened the way for other base metals, including brass, pewter, and tin to acquire the same meaning. Some of these terms allude to the effect that the money has on the possessor (e.g. balsom [sic] ‘money’; heart’s ease ‘a guinea’; tonic ‘a halfpence’), or to the speaker’s lack of familiarity with the coin (e.g. stranger ‘a guinea’). Many imply that money is not important, either by reference to refuse (e.g. dust, muck, rubbish, rust all for ‘money’) 10

F. Grose, A Classical Dictionary of the Vulgar Tongue, London: S. Hooper, 1785, s.v. cripple. 11 E. Partridge, A Dictionary of Slang and Unconventional English, 5th edition, London: Routledge, 1967. 12 B.E., A New Dictionary of the Terms Ancient and Modern of the Canting Crew, London: W. Hawes, c1698.

32

Julie Coleman

or to worthless objects of a similar shape and size (e.g. rivits [sic] and brads both for ‘money’). Dibs, also for ‘money’, are sheep’s knucklebones or small rounded pebbles used in a children’s game. A few of these terms are derived from the names of foreign coins, many of which were in circulation in Britain to make good shortages of small change and precious metals. These include doit ‘half a farthing’ from eMnDu duit; flag ‘fourpence’ from MLG vleger; and soskin ‘money’ possibly from MDu seskijn ‘a coin worth six mites’ (all OED). More speculative etymologies include couter ‘a sovereign’, from Romany kotor (Partridge); deaner ‘a shilling’, from French denier or Lingua Franca dinarly (Partridge); mejoge ‘a shilling’, from Shelta midgic (Partridge); posh ‘money’, from Romany posh ‘half’ (OED); schofel ‘counterfeit money’, from Yiddish schofel ‘worthless stuff’ (Partridge); and tanner ‘sixpence’, from Romany tawno ‘small’ (Hotten, from Ainsworth13). Also suggesting financial exchanges between excluded social groups is monish ‘money’, which may represent a pronunciation influenced by Yiddish (OED). A few disguised terms employ word-play (e.g. she-lion for ‘shilling’), rhyming slang (e.g. rogue and villain for ‘shilling’; lord of the manor for tanner ‘sixpence’; Covent Garden for ‘farthing’) and back slang (e.g. flatch for ‘half(pence)’; yenep for ‘penny’, etc.) may have been used to refer to coins on the brink of liberation from their unsuspecting possessor. Terms like threpps and thrums, both for ‘threepence’ represent non-standard pronunciations of the standard term, although they do not appear to have been created for the sake of deception. Slang lexicographers are particularly fond of anecdotal and narrative etymologies. In 1785 Grose (op.cit.) listed: WHORE’S CURSE, a piece of gold coin value five shillings and threepence, frequently given to women of the town by such as professed always to give gold, and who before the introduction of those pieces, always gave half a guinea.

13

W.H. Ainsworth, Rookwood: A Romance in Three Volumes, London: Richard Bentley, 1834.

Slang terms for money

33

The Lexicon Balatronicum14 has: TILBURY. Sixpence: so called from its formerly being the fare for crossing over from Gravesend to Tilbury Fort.

Narrative etymologies survive into modern slang dictionaries. Beale’s edition of Partridge’s dictionary15 gives the following explanation for rhino ‘money’: Origin problematic. In Malaya, long ago, the rhinoceros was almost ‘worth its weight in gold’ to those opportunists who converted every part of a slain rhinoceros into aphrodisiacs and sold packets at very high prices, to Chinese mandarins, who placed great faith in them.

The trouble with such accounts is that they are entirely beyond proof. Slang terms cannot be subject to the same etymological investigation as more fully documented standard English terms, and many of the etymologies listed here may owe more to lexicographers’ wishful thinking than to solid fact. Most of the terms listed here are also to be found in the OED, and hence in the Historical Thesaurus, often with earlier citation dates, and sometimes much earlier. A few predate existing OED first citations, however, as shown above. The most striking are bender ‘a shilling’ (27 years), blunt ‘money’ (109 years), bull ‘a crown’ (23 years) and rust ‘money’ (49 years), which are joined by many minor antedatings. While this thesaurus is by no means either an exhaustive or authoritative account of slang usage in this field, it does document the available contemporary dictionary evidence. This, in itself, reveals interesting details in the history of individual terms. Why, for example, are brass ‘money’ and James ‘a guinea’ first listed in nineteenth-century slang dictionaries when they are both documented from the sixteenth century? The obvious and most probable explanation is that earlier slang lexicographers, not aiming to provide comprehensive coverage, had overlooked them. However, a few intriguing possibilities suggest themselves, which require further 14

Lexicon Balatronicum, London: Printed for C. Chapel, Pall-Mall, 1811. P. Beale, (ed), A Dictionary of Slang and Unconventional English by Eric Partridge, 8th edition, London: Routledge, 2002.

15

34

Julie Coleman

exploration. Had once standard or colloquial terms drifted into slang? Had the terms moved down the social spectrum? Had slang users revived terms drifting into obsolescence? The re-presentation of OED material in thesaurus form opens up intriguing new research possibilities, previously impossible or impractical. Scholars in many fields will soon find that the HTE is worth its weight in gold. A reliable historical thesaurus of slang is much further off.

‘Huv a wee seat, hen’: evaluative terms in Scots Fiona Douglas and John Corbett

Introduction During her distinguished academic career, Christian Kay has combined a record of research into lexicography and lexical semantics with a longstanding interest in teaching pragmatics and stylistics. As a token of our respect and gratitude for the insights she has afforded us in these fields, we offer the beginnings of an exploration into the semantics, pragmatics and stylistic force of evaluative terms in Scottish speech and writing. Our data come mainly from the Scots Thesaurus and the growing archives of the Scots Corpus of Texts and Speech (SCOTS), a project that Professor Kay initiated and to which she continues to contribute. At the time of writing, the SCOTS project is in its early stages, with almost 400,000 lexical items in its archives, a sizeable proportion of which is literary Scots. As the archive grows, the nature and distribution of evaluative terms in Scottish speech and writing will be easier to explore; this chapter aims only to present a preliminary sketch of possible directions for future research. The Scots Thesaurus is a sample of the vocabulary of modern Broad Scots, arranged in lexical sets.1 Like all thesauri, it provides a fascinating ‘map’ of the linguistic culture from which it springs. The SCOTS archive is a body of electronically-searchable texts, compiled by the University of Glasgow and made available on the Web.2 The SCOTS texts range from Broad Scots to standard Scottish English.3 1

Caroline Macafee, “The Scots Thesaurus: an index to the dictionary record”, Review of Scottish Culture, 6 (1990), 85-86; I. Macleod, (ed.), The Scots Thesaurus, Aberdeen: Aberdeen University Press, 1990. [Reprinted 1999, Edinburgh: Polygon] 2 W. J. Anderson (forthcoming), “The SCOTS Corpus: a resource for language contact study”, in the proceedings of a conference held in St Andrews, Scotland (1113 June 2004) on “Language Contact and Minority Languages on the Littorals of Western Europe”, to be published in the Symposium Logos Series Studies in

36

Douglas & Corbett

Kay in a comparison of the Old English Thesaurus and the Scots Thesaurus (ST),4 observes that [ST] presents a lexicon which is to some extent incomplete. This is partly a matter of editorial selection [...] and of the exigencies of space, but also indicates the extent to which Standard English has taken over in certain areas of Scottish life. The impression given by the ST lexis is that it is largely nominal and descriptive: a bedrock of names for natural phenomena and everyday objects, leavened by a highly descriptive vocabulary for individuals and their characteristics, words such as fushionless, glaikit, peeliewallie, or phrases such as a bonnie fechter or a knotless threid. This passage identifies a neglected area of research into the use of Scots terms: in a lexical economy in which both Scots and Standard English terms have currency, what is the function and value of those ‘highly descriptive’ Scots evaluative terms? Evaluative terms have recently become a focus of interest in pragmatics, discourse analysis and corpus linguistics.5 In their introduction to a ground-breaking anthology of articles on evaluation, Thompson and Hunston argue for the importance of evaluative language.6 For example, evaluative language functions to: x Express opinion x Maintain relations between speakers or writer/readers

Eurolinguistics, Vol. 4, edited by S. Ureland and S. Pugh; F. M. Douglas, “The Scottish Corpus of Texts and Speech: Problems of Corpus Design”, Literary and Linguistic Computing, 17:2, 2003, pp. 259-61. 3 For definitions and discussion of the Broad Scots–Scottish English continuum, see J. Corbett, J. D. McClure and J. Stuart-Smith, eds. The Edinburgh Companion to Scots, Edinburgh: Edinburgh University Press, 2003. 4 C. Kay, “A lexical view of two societies: A comparison of the Scots Thesaurus and a Thesaurus of Old English”, in A. Fenton and D. A. MacDonald (eds.), Studies in Scots and Gaelic, Edinburgh: Canongate, 1994, pp. 41-47. 5 For example, S. Hunston and G. Thompson (eds.), Evaluation in Text: Authorial Stance and the Construction of Discourse, Oxford: Oxford University Press, 2000; P. Leistyna, and C. F. Meyer (eds.), Corpus Analysis: Language Structure and Language Use, Amsterdam: Rodopi, 2003. 6 Hunston and Thompson, op. cit., pp. 6-13.

Evaluative terms in Scots

37

x Manage discourse organisation (e.g., by signalling boundary points in sequences of discourse). The second of these functions promises to be particularly relevant to the use of Scots, particularly in situations where a sense of the individual’s relationship to the community is negotiated through texts that contain markers of Scottish identity. Such situations extend from casual conversation to newspaper texts.7 This chapter sketches some possible approaches to combining thesaurus and corpus resources in order to investigate the cultural function of evaluative lexis. Defining Evaluative Lexis The kind of terms that fall into the category of evaluation are various. J. R. Martin attempts to formulate the possibilities in a set of related systems such as AFFECT (emotions such as happiness or fear), JUDGEMENT (ethical assessments such as morality), and APPRECIATION (aesthetic judgments such as beauty).8 Each system has positive or negative polarity. A random selection of terms from the Scots Thesaurus that would fit neatly into Martin’s model of APPRAISAL are: AFFECT JUDGEMENT APPRECIATION

7

POSITIVE canty, joco, mirrie, (15.5.4) dacent, douce, leal (15.3.1) bonny, guidly, wallie (15.6.1)

NEGATIVE crabbit, disjaskit, dour (15.3.5) coof, notour, waster (15.3.2) ill-farrant, peelie-wersh, scabbit (15.3.2)

Cf. F. M. Douglas, The Role of Lexis in Scottish Newspapers, unpublished thesis submitted to the University of Glasgow in fulfilment of requirements for degree of PhD in the Department of English Language, 2000. 8 J. R. Martin, “Beyond exchange: APPRAISAL systems in English”, paper given to Sydney Linguistics Circle, NCELTR, Macquarie University, 3 June, 1994; J. R. Martin, “Beyond exchange: APPRAISAL systems in English”, in S. Hunston and G. Thompson (eds.) op. cit., pp. 142-75; J. Rothery and M. Stenglin, “Interpreting Literature: the role of APPRAISAL”, in L. Unsworth (ed.), Researching Language in Schools and Communities: Functional Linguistic Perspectives, London: Cassell, 2000.

Douglas & Corbett

38

However, as Thompson and Hunston’s introduction and the succeeding chapters in their anthology show, defining evaluative terms–even agreeing a name for them–is a tricky matter. Thompson and Hunston identify affect,9 attitude,10 appraisal,11 connotation,12 modality,13 and stance14 as possible terms, before finally opting for evaluation themselves.15 They define evaluation as “the broad cover term for the expression of the speaker or writer’s attitude or stance towards, viewpoint on, or feelings about the entities or propositions that he or she is talking about”.16 With two qualifications, this is the definition that we shall adopt here. The first qualification goes beyond issues of terminology to address the nature of ‘expressing an attitude or stance’ on something.17 Cruse observes that evaluative lexical items sometimes describe emotional states such as surprise:18 I am dumfounert. – and sometimes they simply express them: Jings! Martin’s distinction between ‘construed’ and ‘implicated’ affect touches on the difference to which Cruse alludes; however, Martin

9

N. Besnier, “Reported speech and affect on Nukulaelae Atoll”, in J. H. Hill and J. T. Irvine (eds.), Responsibility and Evidence in Oral Discourse, Cambridge: Cambridge University Press, 1993, pp. 161-81. 10 M. A. K. Halliday, An Introduction to Functional Grammar (2nd ed.), London: Edward Arnold, 1994. 11 Martin, 2000, op.cit. 12 J. Lyons, Semantics, Cambridge: Cambridge University Press, 1977. 13 Halliday, op. cit. and M. R. Perkins, Modal Expressions in English, Norwood, NJ: Ablex, 1983. 14 D. Biber, S. Johansson, G. Leech, S. Conrad and E. Finegan, The Longman Grammar of Spoken and Written English, London: Longman, 1999, and Hunston and Thompson, op. cit. 15 Hunston and Thompson, op. cit., p. 2. 16 Hunston and Thompson, op. cit., p. 5. 17 A. Cruse, Meaning in Language. Oxford: Oxford University Press, 2000, pp. 5859. 18 A. Cruse, op.cit., pp. 58-59.

Evaluative terms in Scots

39

rightly blurs the apparently clear distinction.19 The use of even an apparently neutral term in context can load it with evaluative meaning. The term ‘literary’, for example, can refer descriptively to a certain kind of text. However, in examples from the SCOTS archive, the term can be used in either a positive or, at least, a more ambivalent way: …the excellent literary quality of the Lanarkshire dialect material… …ere’s maybe something aifter aa in fit es literary kinna lads say aboot e ‘hard’ Northeast… As Sinclair, Louw and Channell have indicated, continued use of terms in positive or negative contexts shapes their evaluative force, a point we shall return to in due course.20 In the following extract from an abstract by McClure, the authorial stance is clear from the distribution of negative and positive terms that contrast attitudes to the Scots language over time.21 ‘Literary’ is embedded again amongst more explicitly positive terms. (All terms that could be considered evaluative have been highlighted.) The Scots language, traditionally neglected or actively suppressed in the education system, has in recent years enjoyed a quite dramatic reversal in its fortunes. In chronological order, the following related developments have taken place: a remarkable literary efflorescence, including a corpus of brilliant, inventive, and strongly politically-motivated poetry…

19

Martin, 2000, p. 155. J. Sinclair, Corpus, Concordance, Collocation, Oxford: Oxford University Press, 1991; B. Louw, “Irony in the text or insincerity in the writer? The diagnostic potential of semantic prosodies”, in M. Baker, G. Francis and E. Tognini-Bognelli (eds.), Text and Technology: In Honour of John Sinclair, Amsterdam: Benjamins, 1993, pp. 15776; and J. Channell, “Corpus analysis of evaluative lexis”, in Hunston and Thompson, op. cit., pp. 38-55. 21 J. D. McClure cited in I. McGugan, Inquiry into the role of educational and cultural policy in supporting and developing Gaelic, Scots and minority languages in Scotland, Edinburgh: The Scottish Parliament Education, Culture and Sport Committee, 2003. 20

40

Douglas & Corbett

Apparently non-judgemental or technical descriptive terms, then, can nevertheless be used with evaluative force, both to describe and even to express opinions and emotions.22 Our second qualification is that some apparently technical descriptive terms can also be used nonliterally to manage interaction, as in the use of wee to create a sense of intimacy or familiarity amongst speakers: Get a wee drink o watter doon ye. You huv a wee seat, hen. We shall look in more detail at wee shortly. To summarise thus far: given the symbolic use of Scots to establish and maintain social relations, an investigation of the nature and function of evaluative vocabulary in Scots and Scottish English promises to be fruitful. The Scots Thesaurus offers an initial way into evaluative lexis in Scots; resources like the SCOTS archive promise an increasingly powerful means of exploring the function of evaluative terms across a range of genres. Some lexical items express evaluation; others describe evaluation; and yet other terms that are not in themselves evaluative can be used pragmatically to communicate evaluations in context. Again, the Scots Thesaurus and the SCOTS archive offer resources for the preliminary investigation of this topic. Evaluative Lexis in the Scots Thesaurus The key sections of the Scots Thesaurus that deal most explicitly with evaluation are: x Section 8 (‘Life cycle, family’) x Section 9 (‘Physical States’) x Section 15 (‘Characters, emotions, social behaviour’) We focus here on Section 15, which is the largest and, since it deals explicitly with descriptions of character, emotion and behaviour, its lexis is most obviously evaluative. As noted above, Martin’s system of affect has positive and negative polarities. Cruse supports such a dichotomy, observing that 22

Cf. Hunston and Thompson, op. cit., p. 15.

Evaluative terms in Scots

41

oppositeness is ‘in some way cognitively primitive’.23 Cruse’s turn from the study of lexis to cognition recalls van Dijk’s assertion that ideologies are sociocognitive phenomena that construct the in-group in a positive light and the out-group in a negative light.24 One line of investigation would be to identify the resources for evaluation in Scottish discourse, monitor the use of evaluative language in context, and explore the ideological or socio-cultural function of evaluation in establishing and maintaining in-group/out-group identities. As Table 1 shows, many of the categories in Section 15 fall easily into positive or negative categories, though others are more ambivalent. Table 1: Initial categorisation of section 15 Scots Thesaurus categories according to positive/negative polarity. Good/Positive 15.2.3 Intelligence

Bad/Negative 15.2.4 Stupidity

15.3.1 Misc. positive types 15.3.3 Warm, friendly 15.5.2 Compassion 15.5.4 Joy, merrymaking 15.5.15 Skill

15.3.2 Misc. negative types 15.3.4 Aggressive 15.3.5 Pessimistic 15.3.6 Touchy

15.3.9 Bold

15.5.20 Obedience 15.6.1 Misc. sources contentment 15.6.5 Elation

15.3.7 Weakness of character 15.3.8 Treacherous 15.3.10 Timid 15.3.11 Wild, eccentric

15.6.6 Go

15.3.12 Villainous

15.6.11 Liking

15.4.2 Spendthrifts, spongers 15.4.3 Affectation 15.4.5 Slovenliness, laziness 15.5.3 Mockery 15.5.5 Uproar

15.4 Personal, values and behaviour 15.4.1 Thrift, miserliness, avarice 15.4.4 Dress sense

15.6.12 Gratitude 15.6.17 Laughter 15.7.6 Endearments

23

Ambivalent 15.1 General & neutral terms 15.2 Social status & inherent qualities 15.2.1 High class 15.2.2 Low class 15.3 Character types

15.3.13 Earnest 15.3.14 Independent

15.4.6 Eating habits 15.5 Social behaviour 15.5.1 Persuasion 15.5.6 Drink

Cruse, op. cit., p. 167. e.g. Teun A. van Dijk, Ideology: A Multidisciplinary Approach, London: Sage Publications, 1999, pp. 8-10. 24

Douglas & Corbett

42 Good/Positive

Bad/Negative Ambivalent 15.5.8 Nosiness, slander 15.5.7 Chitchat 15.5.9 Quarrels 15.5.13 Mastery 15.5.10 Opposition 15.5.19 Indifference, restraint 15.5.11 Abusiveness 15.6 Emotions & states 15.5.12 Revenge 15.6.3 Miscellaneous sensations 15.5.14 Scolding 15.6.4 Strong emotions 15.5.16 Bad 15.6.10 Longing workmanship 15.5.17 Shambles 15.6.18 Grimaces, gestures 15.5.18 Being hampered 15.7 Interjections 15.6.2 Misc. sources 15.7.1 Miscellaneous discontent 15.6.7 Anxiety, care 15.6.8 Madness 15.6.9 Jealousy 15.6.13 Anger 15.6.14 Shame 15.6.15 Fear, disgust 15.6.16 Sorrow, tears 15.7.2 Disgust 15.7.3 Impatience 15.7.4 Discontent 15.7.5 Gratuitous abuse

At first glance, this table seems to confirm the stereotypical dourness of the Scot. That this trait is evident to the casual reader of the Scots Thesaurus is supported by Fenton’s observation in the introduction that ‘criticism rather than praise is a kenmark of the Scots language’,25 and by the editor Iseabail Macleod’s query:26 Is the [Scots] national reputation confirmed by the long section (15.4.1) on ‘Thrift, miserliness, avarice’? Grippie, meeserable, meesart, nairra-beagun, peengin, pegral, scrubby, scruntie, wheetie all mean ‘miserly, mean’; they are treated together with cannie ‘frugal’, fendie ‘thrifty’, etc., on the other side of the same coin. Scottish grudgingness with praise seems to be confirmed by 25 A. Fenton,“Introduction”, in I. Macleod (ed.), The Scots Thesaurus, Aberdeen: Aberdeen University Press, 1990, p. xii. 26 I. Macleod, “Research in progress: Some problems of Scottish lexicography”, English World-Wide, 14:1 (1993), 115-28.

Evaluative terms in Scots

43

the fact that the section on ‘Miscellaneous negative types’ (15.3.2) is several times as long as that on ‘Miscellaneous positive types’ (15.3.1). As Macleod suggests, the apparent preference for negative terms is evident both in the categories as a whole, and in the terms within each category. There are more terms listed for ‘Low Class’ (15.2.2) than ‘High Class’ (15.2.1), for example. Furthermore, while the great majority of ‘Low Class’ terms are negative (165 of 182, or 91%), only about half of the ‘High Class’ terms are positive (63 of 127 or 49.6%). Indeed, 53 of the ‘High Class’ terms are negative, including cockapentie ‘a snob’, crouse ‘conceited, arrogant, proud’, hingthegither ‘clannish’ and sneist ‘behave in a contemptuous, arrogant way, be scornful or supercilious’. Some classifications, however, seem to determine whether the polarity of their members will be positive or negative. Not surprisingly, 59 of the 67 lexemes within ‘Intelligence’ (15.2.3) are positive, only one being obviously negative (souple, which can mean ‘devious’), while, at least out of context, the others appear neutral or ambivalent e.g. ‘sanshach’ or ‘pawkie’. The lexemes in the category of ‘Stupidity’ (15.2.4) are overwhelmingly negative, 159 of the 161, with none apparently positive, again out of context.27 It is again noticeable that there are considerably more terms available in Scots to express the negative characteristic ‘Stupidity’ than there are to express its positive counterpart. Happily for Scottish self-esteem, a disproportionate weighting towards negative evaluative terms seems to be a general feature of lexicographical reference works. Channell notes that During the writing of CCED [Collins Cobuild English Dictionary], compilers noted more than double the number of negatively loaded words to positively loaded ones. It is too early to know why that might be, or even if it is a substantive observation. It might be, for example, that compilers were more 27

Cf. K. L. Allan, “An examination of metaphor from Old English to present day English focusing on notions of intelligence/cleverness and stupidity”, Ph.D. thesis submitted to the Department of English Language, University of Glasgow, 2003, for a discussion of the concepts of intelligence and stupidity in the Historical Thesaurus of English.

44

Douglas & Corbett

sensitive to negative items because the social consequences of an error with a negative term are much greater than those arising from a misuse of a positive item. The whole area of evaluative language seems to require tying up with the notion of ‘facework’ employed by Brown and Levinson (1987) in their explanation of ‘politeness’.28 To relate evaluative language to ‘facework’, and to explore other pragmatic issues, it is necessary to turn to speech and writing in context, as evidenced, for example, by the SCOTS archive. We shall turn to language in context in the next section. Before doing that, it is illuminating to consider the distribution of evaluative lexis in the Scots Thesaurus in greater detail, with reference to Martin’s model of APPRAISAL, mentioned briefly above.29 To summarise, the model of APPRAISAL assumes that the evaluative resources of a language fall into three general categories, APPRECIATION, AFFECT and JUDGEMENT. Together, these form a network as shown in Table 2. The general system of APPRECIATION involves the speaker’s evaluation of the worth of something. It can be subdivided into reaction (i.e. an expression of the impact of something on the speaker), composition (i.e. an expression of its aesthetic characteristics) and valuation (i.e. an expression of its worth). The system of AFFECT indicates the evaluation of mental states such as happiness, security and satisfaction. The final system, JUDGEMENT, involves an initial choice between social sanction and social esteem. Social sanction comprises of the categories ethics (good/bad) and truth (true/false). Social esteem involves resolve (brave/cowardly, reliable/unreliable, etc), capacity (intelligent/stupid, gifted/incapable, etc) and normality (normal/peculiar, lucky/unlucky, etc).

28

Channell, op. cit., p. 55; she refers to P. Brown and S. Levinson, Politeness, Cambridge: Cambridge University Press, 1987. 29 Martin (1994, 2000), op. cit.

Evaluative terms in Scots

45

Table 2: The APPRAISAL System (cf Martin, 2000; Rothery and Stenglin, 2000 )

APPRAISAL

APPRECIATION

reaction composition valuation

AFFECT

un/happiness in/security dis/satisfaction

ethics social sanction JUDGEMENT

truth

social esteem

Resolve capacity normality

The question is: to what extent do the resources of Scots realise the forms of evaluation thus detailed? Looking again at Section 15 of the Scots Thesaurus, we can begin to see the distribution of terms according to types of evaluation, or, in Martin’s terms, APPRAISAL. Table 3 shows how part of Martin’s APPRAISAL system, the subsystem of APPRECIATION, might be mapped onto Scots Thesaurus categories and examples. Table 3: APPRECIATION and the Scots Thesaurus categories APPRECIATION

ST CATEGORIES

EXAMPLES

REACTION

Grimaces, gestures Laughter Mockery Uproar Opposition

bo ‘make a face’, girn ‘snarl’, ‘grimace’ heffer ‘laugh heartily’, snicher ‘snigger’ dunt ‘insult’, taisle ‘tease’, ‘irritate’, ‘vex’ brangle ‘state of confusion’, too hoo ‘fuss’ conter ‘oppose’, disassent ‘refuse to agree’

COMPOSITION

Dress sense Affectation

barrie ‘smart’, perjink ‘trim’, ‘neat’ dink ‘prim’, ‘precise’, yaup ‘speak affectedly’ gabbit ‘fastidious about food’, vorax ‘voracious’ bauchle ‘spoil’, hairy ‘rough’, ‘untidy’ boorach ‘muddle’, ‘mess’, moger ‘mess’

Eating habits Bad workmanship Shambles

Douglas & Corbett

46 VALUATION

High class Low class Liking Chitchat Nosiness, slander

digne ‘worthy’, far kent ‘widely known’ coof ‘lout’, landwart ‘rustic’, ‘uncouth’ belufit ‘beloved’, kittle ‘make excited’ blether ‘talk foolishly’, clype ‘gossip’ blaud ‘defame’, clash ‘tell tales’

It is always difficult to assign members to single, broad semantic categories in a clear-cut, absolute manner. However, it is evident from the above that Scots evaluative lexis can be distributed across Martin’s categories of APPRECIATION, AFFECT and JUDGEMENT; in other words, there are terms aplenty in Broad Scots for expressing impact, aesthetic appreciation, emotion, ethical judgements, and so on. In the following section, we shall examine how these systems are realised in the dynamic context of a dramatic text. Evaluative Language in the SCOTS archive ‘Facework’ is the designation given to a host of strategies, some of them linguistic, designed to negotiate the individual’s relationship with the community.30 For example, an individual will wish to protect his or her autonomy and wish to be respected by the community for who he or she is (negative face). The same individual will also wish to be considered a respectful member of the community, and to be seen to harmonise with its values (positive face). When our face is threatened, we may attempt to save face by switching topics, hedging, disclaiming, paraphrasing, joking, apologising or justifying. Obviously, some of this activity will involve evaluation. For example, the term dour is defined in the Concise Scots Dictionary as having four possible senses, all of them evaluative: 1. 2. 3. 4.

30

determined, hard, stern, severe obstinate, stubborn, unyielding sullen, humourless, dull slow, sluggish, reluctant (to do something)

Cf. Brown and Levinson, op.cit. and S. Ting-Toomey, “Intercultural conflict styles: a face-negotiation theory”, in Y. Kim and W. Gudykunst (eds.), Theory in Intercultural Communication, Newbury Park, Calif.: Sage, 1988, pp. 213-35.

Evaluative terms in Scots

47

In the SCOTS archive, the term occurs several times in a short story entitled ‘A Gey-Dour Bitch’ by Sheena Blackhall.31 The story is a monologue by someone whose character is evaluated by others, probably as a combination of the first three senses, ‘severe’, ‘stubborn’ and possibly ‘sullen’: She’s a dour bitch, that’s fit I think o her A damned dour bitch, that’s fit I think o her In response to this evaluation, the main character negotiates face, not by rejecting the description, but by asserting that her dourness is in part a result of circumstances, that it is an inevitable part of her character, that it is only occasional, and that it is justifiable: Weel, if I was dour afore, I wis ten times waur eftir thon. ye’ll hae tae pit up wi me, dour bitch or no. I can be dour fyles. It’s a peer pianie that plays nae mair nor eichtsome reels an cheerie ballants. Evaluative terms, then, can be related to facework: individuals or groups anticipate, make and respond to descriptions, whether they are negative, like dour, or positive, like douce (1. sweet, pleasant, loveable; 2. sedate, sober, respectable; 3. neat, tidy, comfortable). The few examples of douce so far collected in the SCOTS archive tentatively suggest that in Scottish English the second sense is more prominent (with the added hint of hypocrisy at times), whilst in Broad Scots the first sense is also still current, particularly in relation to plants, animals, women and children: Spittin wild cat, douce blue-bell Fellow-traivellers like yersel rub yer niv on eir braid foreheids. Douce kinna beasts a swack an douce-like quine whiles, the wee ones that’s douce Feet together, backs straight white socks and navy bloomers, Douce quines frae Broomhill the universal misery of beggardom. Douce, sober citizens 31

S. Blackhall, S., “A Gey Dour Bitch” in The Fower Quarters: Tales by Sheena Blackhall. Edinburgh: GKB Books, 2002, pp. 10-18. [See also http://www.scottishcorpus.ac.uk/]

Douglas & Corbett

48

that was her mother’s way. Keeping up a douce front. Gin we leive mair douce lik an hae a mair hummil style The availability of corpus evidence for the use of lexical items allows us to explore in depth some of the issues touched on above. The discussion of polarity in relation to categories such as ‘Intelligence’ and ‘Stupidity’ assumes that the Scots Thesaurus is an at least partial index of the expressive resources of the language. Channell argues that assumptions such as this are inadequate, because dictionaries and other reference books compiled before the coming of corpora did not have access to information about lexical meaning that was not open to introspection.32 Only an analysis of corpus evidence, she argues, shows that expressions like par for the course, roam the streets and regime imply negative polarity, while other expressions, like off the beaten track imply positive polarity. These implications are not open to introspection because the terms themselves do not express negative or positive meanings; their evaluative force comes from their being used consistently in either negative or positive ‘frames’.33 An example from the Scots lexicon is wee, which is defined thus in the Concise Scots Dictionary: 1. small, tiny, little, restricted in use 2. as intensifier with nouns signifying a small amount 3. football used to describe the reserve team A list of combinations is then given, the foremost amongst them wee ane ‘a young child, a little one’, an expression that also appears in the Scots Thesaurus, under ‘Childhood, Infancy’ (8.5). Douglas demonstrates that wee is amongst the most widely used Scotticisms in newspaper texts, and it appears in both the written and spoken materials in the SCOTS archive.34 At first glance, wee simply refers to size, although evaluation is also implied in the sense ‘restricted in use’. However, it is clear from its use in context that wee has a number

32

Channell, op. cit. Sinclair, op. cit. 34 Douglas (2000), op. cit. 33

Evaluative terms in Scots

49

of evaluative functions. It is used 29 times in one of the SCOTS texts, Refuge, a play by Janet Paisley about victims of domestic violence.35 CHILDHOOD it minds me ae bein wee, an bein looked efter a wee boy calls mummy, mummy what’s the wee boy’s name? the wee boy’s two She’s got a wee boy and a bairn and wee Timothy’s demolishin the playroom CAROLANNE: She’s got children. AGNES: Wee yins? INTENSIFICATION a wee bit ae practice a wee bit ae rejuvenation a wee bit upset DRINKS ah huv a wee can get a wee drink ae watter doon ye a wee kick-starter, just DESCRIPTION OF PEOPLE (NEGATIVE) Right wee chatterboax, isn’t she? Racket dis ma heid in. Only wee men need tae cut their wimmin doon DESCRIPTION OF PEOPLE (POSITIVE) That ma cocoa? Yer a wee lamb Might just pick up a wee pick-me-up Just a couplae wee yins. Tae practise oan. DESCRIPTION OF BEHAVIOUR gie a wee cough couple o gills and a wee shufty at the talent dae ye think ye could see yer wey tae a wee warnin furst

35

Janet Paisley, Refuge, 1997: http://www.scottishcorpus.ac.uk/.

50

Douglas & Corbett

DESCRIPTION OF THINGS Wee curer the night, but Ah was that prood ae ma wee hoose You huv a wee seat, hen don’t bring oot oney wee things a wee treat this last wee while if ye end up wi a bagfu, bring a wee yin back for me

Only in a few instances here is it clear that wee is a neutral description for ‘small’ – the most obvious is don’t bring oot oney wee things, whereby one character warns another character that small toys will be unsuitable for a child. The use of wee to refer to children is partly literal (they are indeed small), but it also carries with it a sense of vulnerability and intimacy. The senses of intimacy and vulnerability transfer to some positive descriptions of adults (wee lamb), but here and in the negative descriptions, there may be an additional connotation of insignificance (wee pick me up, and wee yin tae practise oan refer to casual relationships – enjoyable but transient – while wee chatterboax and wee men are contemptuous dismissals). Although the use of wee with different kinds of drink tends towards the formulaic (in expressions like a wee dram), it is nevertheless interesting: in this play, expressions like wee can (beer) and wee kickstarter (coffee) downplay, or possibly deny, the importance of alcohol and caffeine to the speaker by presenting the beverage as insignificant. The offer of a drink that does not contain stimulants (Get a wee drink ae watter doon ye) focuses more on the connotations of intimacy and reduces the social distance between speakers. The same effect is achieved by the offer of a chair (You huv a wee seat, hen) – presumably not a small chair! The use of wee to describe ‘things’ in this play hints at a developing semantic prosody of the kind described by Sinclair and Channell.36 Here, wee tends to describe things that are desirable: curer, hoose, seat, treat, yin. We can appeal to the concept of facework to explain this apparent pattern: the speaker expresses a desire for or pride in something but does not wish to appear greedy or 36

Sinclair, op. cit. and Channell, op. cit.

Evaluative terms in Scots

51

selfish. The speaker therefore downplays its significance. Furthermore, the use of wee with behaviours refers to something the speaker intends to do (shufty) or wants from others (cough, warnin). When asking for something the speaker desires but the hearer may not wish to give, it saves the hearer’s face to have the act of giving described as insignificant. A further issue is how Scots evaluative terms in the systems of APPRECIATION, AFFECT and JUDGEMENT function in an actual text. To explore this issue, we shall focus on an exchange in Refuge in which one of the women in the safe house, middle-class Carolanne, rises and enters the kitchen after a night on the town. She is met by another of the residents, working-class Sadie. SADIE: Oh, great, up in time fur lunch, then? No, dinnae tell me, let me guess. Eh, a generous wedge of ripe camembert, deep fried, served with cranberry sauce and crispy salad. Or, the freshest of fresh mussels cooked to perfection in a creamy white wine and garlic sauce. Or - would that be toast? (BEFORE CAROLANNE CAN THROW UP, BETH ENTERS FROM KITCHEN WITH A BOWL OF FOOD WHICH SHE PUTS DOWN TO COLLECT TOYS THE CHILDREN HAVE LEFT BEHIND. SADIE AND CAROLANNE FREEZE AS SHE ENTERS, WATCHING BETH’S PROGRESS AS SHE JUGGLES TOYS, COLLECTS THE BOWL AND HEADS FOR UPSTAIRS.) SADIE: Nae bletherin noo, there’s work to be done. (BETH EXITS. SADIE SHUTS THE DOOR) SADIE: Right wee chatterboax, isn’t she? Racket dis ma heid in. CAROLANNE: She’s working though. SADIE: At hauntin the place. A fortnight an she’s still creepin aboot wi nae shoes oan. We should hire her oot. Rentaghost. (CAROLANNE MAKES A FACE AND TURNS TO HER COFFEE) SADIE: Must be catchin. You look gey peely wally. CAROLANNE: Headache. SADIE: The mornin efter, huh? CAROLANNE: Don’t know. I never had one before.

52

Douglas & Corbett

SADIE: Bet you never hud a bit ae rough before either. CAROLANNE: (LOOKS UP TOO QUICKLY) I never. Oww. SADIE: I don’t know sae much. You couldnae huv slid a razor blade between yeese withoot cutting aff something vital. (SADIE GOES ON INTO KITCHEN) CAROLANNE: We danced. That’s all. SADIE: (FROM OFF) Hey, never heard it cawed that before. CAROLANNE: Oh, come on. He had (WINCING) tattoos. SADIE: (FROM OFF) On his teeth. CAROLANNE: Did he have teeth? SADIE: (FROM OFF) Mostly. (COMING BACK IN WITH COFFEE) I think he left them in yer neck. CAROLANNE: You’re having me on. Is that lunch? SADIE: A wee kick-starter, just. Good night, but. Dae ye mind chattin up the lamppost oan the wey hame? CAROLANNE: Now I know you’re exaggerating. Because I remember that lamppost. SADIE: So ye should. A sudden but intense attraction. Ye made a big impression there. CAROLANNE: I was holding it up. SADIE: Oh aye. CAROLANNE: Aye. I mean yes. It was leaning over. SADIE: Nope. Straight as a die. Giving an absolute and unequivocal identification of the semantic nature of the lexis used evaluatively here is problematical, not least because the meaning of any text and its constituents is always open to interpretation. Ever since Stanley Fish’s counterblasts against stylistics in the 1970s, there has been an issue about whether meanings are determined by linguistic constituents or whether they are imposed by a community of readers and listeners.37 There is a particular danger in arguing that because a lexical item has been categorised in a particular way in a thesaurus, its meaning in context will therefore be determined for all contexts. This is clearly not the case. For example, the above dialogue contains tattoo, which (if it qualified for entry as a Scots term) would most likely be categorised in the Scots Thesaurus 37 S. Fish, Is There a Text in this Class? The Authority of Interpretive Communities Cambridge, Mass.: Harvard University Press, 1980.

Evaluative terms in Scots

53

under Skin (9.8) alongside various terms for epidermal tones, such as measlet ‘blotched’, plouk ‘pimple’ and sleekit ‘having glossy skin’. In context, though, tattoo is taken as a valuation of social class, and the literally absurd allegation that he has tattoos on his teeth functions to intensify the valuation. In sum, meanings cannot be ‘read off’ a thesaurus or dictionary. Nevertheless, the dictionary/thesaurus meanings are relevant to any given use in context. What follows, then, is not an unambiguous or ‘fixed’ account of the meanings of this dialogue, but a record of our engagement with a text as readers, drawing upon our interpretation of thesaurus categories in relation to one possible system for modelling evaluative lexis. The italicised terms from the above dialogue are shown in Table 4 with their polarity, their value in the system of APPRAISAL, and their likely Scots Thesaurus categorisation (since this is a mixture of Scottish and English terms, only a few of the terms here are actually in the Scots Thesaurus). Table 4: Evaluative lexis in a scripted dialogue (extract from Refuge by Janet Paisley) (a) Sadie Term great generous freshest cooked to perfection creamy bletherin right wee chatterboax racket dis ma heid in hauntin creepin rentaghost gey peely wally a bit ae rough something vital on his teeth wee kick starter just good chattin up intense attraction big impression straight as a die

Polarity + + + + + + + + + + + +

APPRAISAL value valuation composition composition composition composition valuation valuation reaction reaction reaction reaction reaction capacity valuation valuation composition valuation valuation valuation reaction reaction valuation

ST Category Misc. positive Misc. food Misc. food Misc. food Misc. food Chitchat Chitchat Uproar Uproar Fear, disgust Fear, disgust Fear, disgust Bad health Low class Parts of body Parts of body Misc. food Misc. positive Chitchat Strong emotions Strong emotions Skill

54 (b) Carolanne Term tattoos having me on exaggerating leaning over

Douglas & Corbett

Polarity -

APPRAISAL value composition truth truth valuation

ST Category Skin Mockery Weakness of character Bad workmanship

This activity highlights the dialogic ‘flow’ of this exchange. The character of Sadie is given much more evaluative vocabulary than Carolanne; she is the one who dominates this excerpt using a variety of linguistic strategies – commenting interrogating and contradicting – while it is Carolanne’s role largely to respond and justify. If we concentrate on the evaluative lexis, we see Sadie’s initial turn sarcastically using the culinary language of the middle-class to taunt hungover Carolanne. Sadie then directs her sarcasm against the silent Beth, whom she characterises using Scots as an empty chatterbox (right wee chatterboax…racket dis ma heid in). When Carolanne defends Beth, Sadie expresses the fear and disgust that Beth’s behaviour arouse in her. She then directs her attention towards Carolanne again, linking her appearance with Beth’s (both are pale) but evaluating Carolanne as in bad health, again using Scots (peely wally). Her next evaluation concerns Carolanne’s recent choice of a sexual partner, whom Sadie characterises again in class terms (a bit ae rough). Carolanne and Sadie then engage in bantering around how far Carolanne’s relationship has gone with someone whom both characterise as being of a lower social class, symbolised by tattoos…on his teeth. Carolanne accuses Sadie of exaggerating, and attempts to redirect the conversation towards the inadequacy of Sadie’s lunch, a cup of coffee. Sadie responds by evaluating this as an insignificant stimulus (a wee kick-starter, discussed above) and she counter-attacks by directing the exchange towards Carolanne’s inebriation the previous night. The excerpt ends with further banter about the nature of Carolanne’s behaviour the previous night. Each contradicts the other about the valuation given to the lamp-post that Carolanne was holding onto – was it leaning over (an example of bad workmanship) or straight as a die (a product of skilled workmanship)? By this point Carolanne is more of an equal partner in the exchange, entering into the fantasy that Sadie is constructing on her own terms.

Evaluative terms in Scots

55

This description of the ‘flow’ of the dialogue shows the importance of evaluation in the process of bonding – what we called earlier the establishment and maintaining of social relations. Evaluative language is used in conversational gambits to orient participants in respect of in-groups and out-groups. Sadie here is the key player: her initial evaluations (in English) mock Carolanne by appropriating the language of her class to nauseate her. Then her evaluations (in Scots) invite Carolanne to share her negative perceptions of Beth; when Carolanne refuses the invitation, she evaluates Carolanne’s health, a common expression of empathy. In the ensuing banter, Sadie constructs a fantasy in which Carolanne has enjoyed sexual relations with a person of low class, and got drunk enough to attempt to seduce the lamp-post that was holding her up; Carolanne denies the accusation and constructs a counter-fantasy in which she is supporting the badly-constructed lamp-post. The exchange works to construct the character of Sadie as someone who needs to challenge others – middle-class Carolanne and silent Beth – and orient them towards her own working-class garrulity. That her strategy is at least partly successful with Carolanne is signified by the latter temporarily adopting markers of Sadie’s speech community: Aye. I mean yes.

Conclusions It has been our intention in this chapter to raise more questions than we attempt to answer. We have argued that the evaluative nature of much Scots lexis has been often commented on; yet it remains under-researched as a pragmatic resource for in-grouping and outgrouping. There remains the difficult issue of how we define evaluative lexis, across the Broad Scots – Scottish-English continuum, bearing in mind that, in specific contexts, technically descriptive lexis can be given evaluative force. We have attempted to manage this issue by relating semantic categories used in the Scots Thesaurus to pragmatic functions offered by Martin’s model of APPRAISAL, and using these combined resources to interpret the way characters in a scripted dialogue use evaluative text to establish and maintain social relations. The example is offered as a case study for a series of much broader and deeper analyses of the nature and function of evaluative texts across a range of genres of Scottish speech and writing.

56

Douglas & Corbett

Given the space at our disposal, this sketch can only begin to show the kinds of applications that can be made of resources such as dictionaries, thesauri and corpora – the kind of resources to which Christian Kay has devoted much of her professional career. Kay argues for a larger Scots Thesaurus, one that incorporates all the data in the Dictionary of the Older Scottish Tongue and the Scottish National Dictionary within a framework supplied by the Historical Thesaurus of English.38 Ten years after this suggestion, DOST and SND have been combined in an electronic version the Dictionary of the Scots Language, and HTE nears completion. An amalgam of available resources, plus updated lexical material, would make the kinds of analysis attempted here much subtler and more revealing. As she heads towards what we hope is a long and productive ‘retirement’, we continue to look to Professor Kay for inspiration and leadership in the lexicography of English and Scots, and in its wider applications.

38

Kay, op. cit.

Lexical splits and mergers: some difficult cases for the OED Philip Durkin Christian Kay and Irené Wotherspoon have in a recent article looked at some of the complicated issues surrounding orthography and homophony that can exist within a lexical set.1 Here I will examine the related issue of lexical splits and mergers, i.e. cases where two or more words of distinct etymology converge, and also cases where what is etymologically a single word diverges, giving two or more lexical items distinct in spelling or pronunciation or both. I will look in particular at the challenges that dealing with such words can pose for a historical dictionary, where complex changes must be accommodated within a diachronic account of individual word histories. I will take examples from entries which have been revised for the third edition of the Oxford English Dictionary (on OED3 see the online Preface to the Third Edition of OED, www.oed.com /about/oed3-preface).2 Words which have split will require two distinct dictionary entries under the usual modern forms, but splits which have occurred in historical times pose some problems in assigning data to the respective entries. At mantle and mantel, or ordnance and ordinance, from a modern synchronic perspective the split is (largely) complete, but the earlier evidence requires some difficult decision-making. Complete mergers, such as mare ‘female horse’, are largely unproblematic, but examples of partial merger, as is 1

Christian Kay and Irené Wotherspoon, “Wreak, wrack, rack, and (w)ruin: The history of some confused spellings”, in Teresa Fanego, B. Méndez-Naya and E. Seoane (eds.), Sounds, Words, Texts and Change. Selected Papers from 11 ICEHL, Santiago de Compostela, 7-11 September 2000, Amsterdam: Benjamins, 2002, pp. 129-43. 2 James A. H. Murray, Henry Bradley, Sir William A. Craigie and Charles T. Onions (eds.), The Oxford English Dictionary, 1884-1933; Robert W. Burchfield (ed.), Supplement, 1972-86; John A. Simpson and Edmund S. C. Weiner (eds.), 2nd edition, 1989; John A. Simpson, Edmund S. C. Weiner and Michael Proffitt (eds.), Additions Series, 1993-97; John A. Simpson (ed.), 3rd edition (in progress) OED Online, March 2000- , Oxford: Oxford University Press.

58

Philip Durkin

probably the case with certain of the senses of mean ‘ungenerous, ignoble’ (of Germanic origin) and mean ‘intermediate, average’ (of Romance origin), are much harder to deal with, and require explicit comment from the lexicographer. The modern English words mantle (‘loose sleeveless cloak’ and related senses) and mantel (‘ornamental structure of wood, marble, etc., above and around a fireplace’ and related senses) show a process of split in the written form of the word.3 In Old English, classical and post-classical Latin mantellum is borrowed as mentel in the sense ‘long sleeveless cloak’ (with i-mutation probably resulting from suffix substitution). In Middle English this persists, coalescing with borrowing of mantel, the Anglo-Norman and Old French reflex of the Latin word (and perhaps also reborrowing of the Latin word itself) to give Middle English mantel, mantle (and other related spellings). From the thirteenth century in British sources the Latin word also shows the sense ‘piece of timber or stone supporting the masonry above a fireplace’, and this is shown also by the Latin word’s AngloNorman and Central French reflexes from the fourteenth century, as also by the English word. Middle English thus has, not particularly unusually, a borrowed word showing polysemy which broadly reflects polysemy in the donor languages. For a dictionary on historical principles such as the OED the methodology for such a word is straightforward: an identical word-form which shows a range of senses of a common origin will be treated under a single dictionary entry. (This is not to disregard the problems that polysemy can pose for many models of both diachronic and synchronic linguistics, but simply to note that the traditional methodology of historical lexicography demands that such material should be treated under a single dictionary headword, thus ensuring an approach which is consistent, even if from the synchronic standpoint it may sometimes 3

It is doubtful whether there is any consistent corresponding distinction in the spoken form. To take a sample from pronouncing dictionaries, Clive Upton, William A. Kretzschmar, Jr., and Rafal Konopka, The Oxford Dictionary of Pronunciation for Current English, Oxford: Oxford University Press, 2001, here followed by OED3, gives identical realisations in British English for both words, but for U.S. English gives 'mæn(t)ԥl for mantle but 'mæn(t)l for mantel, while Daniel Jones, English Pronouncing Dictionary, 15th edition, ed. Peter Roach and James Hartman, Cambridge: Cambridge University Press, 1997, gives (regardless of variety) 'mæn.tƺ for mantle and 'mæn.tԥl for mantel.

Lexical splits and mergers

59

seem somewhat arbitrary.) The modern reflexes of Middle English mantel, mantle, however, pose a problem for the historical dictionary, as in the modern written language ‘loose sleeveless cloak’ and ‘ornamental structure of wood, marble, etc., above and around a fireplace’ and their related senses and compounds (e.g. mantelpiece, mantelshelf) are now usually distinguished in spelling, as mantle and mantel respectively. (A very few examples of the spelling mantle are still found in modern use for the ‘fireplace’ word, but they are hugely outnumbered by the spelling mantel, which must be regarded as the usual modern spelling in these senses.) They are thus treated as two distinct words by general dictionaries of the modern language (regardless of the extent to which their methodology is otherwise influenced by historical lexicography), albeit with cross-references or notes in some cases acknowledging that either spelling can still more rarely be found for each word.4 The situation is more difficult for a historical dictionary, especially for one with the time depth of the OED: a way must be found of representing Old English mentel, Middle English mentel, mantel, mantle, and the distinct (late) Modern English words mantel and mantle, together with all of the intermediate points and varied actual spellings represented by these abstractions. The solution for OED is determined by the policy of placing material under headword forms which are decided by the majority usage of modern standard written English (in so far as the materials available to us enable us to determine what this is). This suggests that the reflex of Middle English mentel, mantel, mantle yields two distinct modern word forms, mantel and mantle. The divergence runs along semantic lines, with the groups of senses related to or developed from respectively ‘loose sleeveless cloak’ and ‘piece of timber or stone supporting the masonry above a fireplace’ gradually coming to be distinguished in (written) form, and this divergence is therefore reflected in the structure of the two OED entries, with material being divided between the two entries on semantic grounds, regardless of the form of the word. Each entry shows historical examples with both -le and -el types, and each entry has a separate list of historical 4 Thus both Judy Pearsall (ed.), The Concise Oxford English Dictionary, 10th edition, revised, Oxford: Oxford University Press, 2002, and Judy Pearsall (ed.), The Oxford Dictionary of English, 2nd edition, Oxford: Oxford University Press, 2003, note that mantle also occurs as a spelling of mantel ‘mantelpiece’.

60

Philip Durkin

spellings, compiled from the OED’s slip collections (including the fruits of targeted reading of historical sources for spelling variation as well as for semantic data), from the evidence of period dictionaries, and from an extensive diachronic sampling of electronic resources.5 The results may be illustrated by the variant form lists, extracts from the etymology, and the definitions of the first senses from each entry (for reasons of space, the illustrative quotations are omitted): mantle n. Forms: OE-ME mentel, ME mantal, mantille, manttell, mantyl, mantyle, mauntel, mauntell, mayntell, mayntelle, mentell, mentil, mentile, (perh. transmission error) manntel, ME-15 mantell, mantelle, ME16 mantel, mantil, mantill, mantyll, mantylle, mauntil, ME- mantle, 15 mauntelle, 16 mantoll; Eng. regional 18- mentle, 19- mantel; Sc. pre-17 mantale, mantel, mantell, mantill, mantyl, mantyll, pre-17 17- mantil, mantle. [Partly < classical Latin mantellum cloak (see below), and partly a reborrowing in early Middle English < its reflex Anglo-Norman mantel, mantelle, mantle cloak, Old French mantel (c980; Middle French mantel, manteau, French manteau). In Old English (as also in Old Frisian) showing imutation, prob. arising from suffix substitution (see -EL¹). With classical Latin mantellum cf. mantica (see MANTICULATE v.); further etymology uncertain (see below). Classical Latin mantellum is attested only in Plautus; in post-classical Latin a 4th-cent. grammarian has the form mantelum (citing the Plautus passage) by confusion with classical Latin mantƝle hand-towel: the two words are no longer thought to be etymologically related. Post-classical Latin mantellum is attested more freq. from 9th cent.; freq. in British sources from 12th cent. ... Cf. also MANTEL n. and MANTEEL n.1] I. A protective garment or blanket, and related fig. uses.

... 1. a. A loose sleeveless cloak. mantel n. Forms: ME- mantel, 15 mantalle, mantell, 15- mantle. [Orig. a variant of MANTLE n., now usu. distinguished in form in the senses below. Cf. Anglo-Norman mantle (1370 or earlier in sense 1a), Old French, Middle French mantel (1332 in sense 1a); post-classical Latin mantellum (from 13th cent. in British sources in sense 1a): see further discussion s.v. MANTLE n.] 5

For more detail on this area of the OED’s work see Philip Durkin, “Root and branch: revising the etymological component of the OED”, Transactions of the Philological Society, 97 (1999) 1-50, and again the online Preface to the Third Edition of OED, op. cit.

Lexical splits and mergers

61

I. Simple uses. 1. †a. A piece of timber or stone supporting the masonry above a fireplace; = MANTELTREE n. 1. Obs. Recorded earliest in mantel-stone, sense 3. b. An ornamental structure of wood, marble, etc., above and around a fireplace; the manteltree of a fireplace together with its supports. Cf. MANTELPIECE n. 1. c. A shelf formed by the projecting surface of a mantelpiece; = MANTELSHELF n. 1.

A similar split, in this case showing a distinction in both the written and the spoken form in modern English, is shown by the OED entries ORDINANCE n. and ORDNANCE n., where a Middle English borrowing of the semantically complex Anglo-Norman and Middle French ordenance shows a subsequent split, with the form ordnance coming to be associated solely with senses related to and developed from ‘military materials’, ‘artillery for discharging missiles’, ‘the government department responsible for this’, etc. A similar approach is thus adopted for OED to that shown at MANTLE n. and MANTEL n., with the historical material being divided on semantic grounds, with an etymology at ORDNANCE n. making clear that the two are originally the same word: “Orig. a variant of ORDINANCE n., now distinguished in form in the senses below”. (In a different category, and outside the scope of this article, are cases where the division of material between entries is on a formal rather than a semantic basis. This is commonly the case where developments of a word form in different varieties of English are treated in separate entries, normally where the form in another variety has come to have some currency (perhaps as a marked regionalism or colloquialism, or perhaps with a specific range of meanings) in standard (written) English, for instance in OED3 MA’AM n. and MARM n., or MAYBE adv. and MEBBE adv.) Turning now to mergers, the OED’s policy of listing material under its present-day reflex (if one exists) again determines the approach taken, and with mergers which have been completed in the past the procedure is thus relatively simple, even if the detail may be complex. For instance, the modern English word mare ‘female horse’ shows a semantic and formal blend of two originally distinct Old English words, mearh ‘horse’, which gives rise to the modern standard English form, and (West Saxon) mƯre, mýre, (non-West Saxon) mƝre ‘female horse’ (the reflex of a Germanic feminine

62

Philip Durkin

derivative of the Germanic base of mearh), from which the modern sense has arisen. As there is a single modern output word, the full history of both words can be treated under a single dictionary headword, with a breakdown of the forms into their historical types, a full etymological account of the origin and subsequent history of each type, and (not shown here for reasons of space) a selection of illustrative quotations for each form type under each sense:6 Forms: Į. OE mearh, (rare) mearg, (Anglian) merh. ȕ. OE (West Saxon) mire, myre, (non-West Saxon) mere, (Anglian, rare) mære. Ȗ. ME maar, maire, ME-15 maare, ME, (18- Eng. regional) mar, ME- mare, 15 marre, mayre, 15-16 mayer, 16 maer; Sc. pre-17 maer, mair, mar, mayr, pre-17 17- mare. į. ME-15 mere, ME meere, mer, meyre, mure, ME-15 meare, ME, (18- Irish English (north.)) meer, 15 (Eng. regional (north.)) myer, 18 (Eng. regional (south-west.)) meäre; Sc. pre-17 meire, mer, meyr, meyre, miere, pre-17 17- mear, meare, meir, mere, 18- meer, 19- meere, mehr. [A merging of two distinct words: Old English mearh horse and Old English mƯre, mýre mare. (i) (Represented by the Į forms) Old English mearh (strong masculine, inflected forms mƝar- or, by analogical replacement, mear-) ‘horse’, whose surviving instances occur chiefly in poetry, is cognate with ... < a Germanic base, cognate with ...; further etymology uncertain: perh. ult. cognate with Sanskrit marya- young man, stallion (cf. discussion s.v. MARRY v.). (ii) (Represented by the ȕ forms) Old English (West Saxon) mƯre, mýre (weak feminine) ‘mare’, is cognate with ... < a Germanic feminine derivative of the base of Old English mearh. Whereas the masculine word has all but died out in the various Germanic languages, its feminine derivative retains its vigour. The Ȗ forms represent later reflexes of Old English mearh and (in inflected forms, with loss of -h and compensatory lengthening) mear- (giving ă and Ɨ respectively in Middle English); the į forms represent a mixture of reflexes from different sources: Old English mƝar- (the regular inflected form of mearh; giving long open Ɲ in Middle English); Old English (Anglian) merh and (in inflected forms) mƝr-, mer- (smoothed forms of mearh; giving Ɵ, long close Ɲ, and long open Ɲ respectively in Middle English); Old English (Anglian) mƝre (giving long close Ɲ in Middle English); and Old English (West Saxon) mýre (giving occasional spellings with -u- in Middle English). 6 An early draft version of this same material was included in Durkin (1999), op. cit., and the reader may be interested to compare the two to see an illustration of how OED3’s style and procedures have developed between the early stages of work for the new edition and the actual publication of revised material online.

Lexical splits and mergers

63

In current British regional use pronunciations with base vowel /(Ú/ are general in southern and central England (Surv. Eng. Dial. also records a few isolated examples with diphthongized /D,/ in Hampshire: perh. a reflex of Old English (West Saxon) mƯre), while in the north of England, Scotland, and Northern Ireland pronunciations with base vowel /LÚ/ are common. In Middle English, from at least the 13th cent., Ȗ forms occur in the sense ‘female horse’, while į forms occur in the generic sense ‘horse’. The latter sense (irrespective of form) died out at the end of the Middle English period (cf. note at sense 1 below). By the end of the 16th cent. the Ȗ form mare had wholly supplanted the į forms in standard English.]

Partial mergers are more difficult to reflect in a historical dictionary like the OED. Such a case is probably shown by OED’s MEAN a.¹ and MEAN a.² The first of these is the reflex of Old English m¼ne, an aphetic variant of gem¼ne, having senses grouped in OED under the branches “I. Held commonly or jointly”, “II. Inferior in rank or quality; unpleasant”, and “III. With approbative connotation”. MEAN a.² is a Middle English borrowing from AngloNorman (and hence ultimately from classical Latin mediƗnus “that is in the middle”), with senses grouped in OED under the branches “I. Intermediate, intermediary” and “II. Moderate, middling; average”. It is at branch II in each entry where difficulties arise. The senses at each, with dates of first attestation, are: mean a.¹, OED3, branch II, definitions and first dates: II. Inferior in rank or quality; unpleasant. 2. Of a person or body of people, a person's condition, etc. (In early use freq. in the comparative.) a. Of low social status; spec. not of the nobility or gentry. Cf. COMMON a. 12. Now rare exc. in the superlative (chiefly hyperbolically), as the meanest ʊ. a1375 b. Inferior in ability, learning, perception, etc. Now chiefly in the superlative and as in sense 4. a1387 †c. Poor, badly off. Obs. c1400 d. Of a political body, authority, etc.: weak; comparatively powerless. Now rare. c1524 †e. Abject or debased. Obs. rare. c1680 f. U.S. colloq. In low spirits; unwell or in a poor state of health. 1845 b. Petty, insignificant, unimportant; inconsiderable. Now rare. ?c1430

64

Philip Durkin

c. Undignified, low. Of literary style, etc.: lacking in elevation or adornment; unambitious (not always with depreciative connotations). Now rare. c1450 d. Unimposing or shabby; characterized by poverty; humble. Used esp. of a building, place of habitation, etc. 1600 4. no mean ʊ: denoting something very good or noteworthy of its kind. Now used chiefly to express approval or admiration. 1580 5. a. Of a person, a person's character, etc.: lacking moral dignity, ignoble; small-minded. Now rare or passing into sense 6. 1665 b. U.S. colloq. Of a horse, etc.: vicious or hard to control. Sometimes also in extended use, of a person when drunk: uncontrollable, violent. 1835 c. U.S. colloq. to feel mean: to feel ashamed of one's conduct; to feel guilty of unfairness or unkindness. 1839 d. colloq. (orig. U.S.). Of a person, a person's actions, etc.: disobliging, uncooperative; unpleasant, unkind; vicious, cruel. 1841 6. Niggardly, miserly, stingy; not generous or liberal. 1840

mean a.², OED3, branch II, definitions and first dates: II. Moderate, middling; average. †7. Not much above or below the average; moderate, mediocre, middling. a. Moderate or middling in size, stature, or age. Obs. a1387 b. Moderate or middling in quality or strength. Obs. ?1440 c. Moderate in amount, or in degree of excellence; tolerable, mediocre. Later used only disparagingly (and so in some cases not readily distinguished from MEAN a.1). Obs. a1500 d. Of soil or land: moderately fertile. Obs. 1523 8. Math. a. Of a value: so related to a given set of values that the algebraic sum of their differences from it is zero; that is the arithmetical mean (MEAN n.3 9a) of a set of values; average. Hence used before the name of a variable quantity to express the mean average value of that quantity (as in mean diameter, distance, motion, temperature, etc.). Cf. MEAN PROPORTIONAL n. a1450 b. Mil. mean point of impact, the centre of a cluster or pattern of impact points made by bombs or bullets, calculated as an average of their coordinates (see quot. 1973). 1945 †9. Using moderation; temperate. Obs. rare. a1500

The semantic overlap is thus readily apparent, and both the nature of the problem and the solution arrived at are explained in the etymologies of the two entries:

Lexical splits and mergers

65

mean a.¹, OED3, etymology: [Aphetic < I-MENE a. Cf. Old Frisian mƝne shared, common (West Frisian mien). In Old English (and in the earlier stages of other Germanic languages) substantially the only sense of I-MENE a. and its cognates was ‘possessed jointly’, ‘belonging equally to a number of persons’; however, already in Old English there existed a spec. sense ‘of ecclesiastical orders: minor, inferior in degree’, which, although it did not survive into Middle English, may have informed the development of mean. The semantic development shown by the Old English spec. sense of IMENE a. was carried further with Middle English mene, mean (as with Dutch gemeen and German gemein; cf. COMMON a.), so that the word acquired the general senses of ‘ordinary’, ‘not exceptionally good’, ‘inferior’. In English this development was aided by the fact that the native word coincided in form with MEAN a.2, which was often used in a disparaging or reproachful sense. The uses in branch II might be referred almost equally well to the native or to the foreign adjective; the truth is probably that the meanings of two originally quite distinct words have merged. It has sometimes been supposed that the sense development of the word has been influenced by Old English m¼ne false, wicked (cognate with MAN n.2 and MAN a.); but this seems unlikely, as this adjective did not survive into Middle English, while the moral senses of mean only appear in modern English.]

mean a.², OED3, etymology: [< Anglo-Norman mene, men, meen intermediate, middle, middle-sized (cf. Old French meien (first half of the 12th cent.), moien (c1260), Middle French, French moyen (c1330)) < classical Latin mediƗnus (see MEDIAN a.2). Cf. Spanish mediano (1070), Occitan mejan (13th cent.), Italian mezzano (14th cent.), Portuguese mediano (17th cent.). Cf. MESNE a. and MOYEN a. Among parallel senses of the word in Old and Middle French are: ‘situated in the middle’ (first half of the 12th cent.; cf. sense 1a), ‘moderate or middling in size or age’ (c1260; cf. sense 7a), ‘ordinary, mediocre’ (1273; cf. senses 7b, 7c), ‘of a number in a proportion: that is a mean’ (1377; cf. sense 8a), ‘of a verb: middle’ (1530; cf. sense 5).]

The solution adopted is etymological, placing at MEAN a.² those senses which closely parallel those of the Anglo-Norman and Old and Middle French words, while placing at MEAN a.¹ the main group of senses which can be put under the general heading ‘inferior in rank or

66

Philip Durkin

quality; unpleasant’, citing as support for this the similar pejoration shown by cognates in other West Germanic languages, and also the apparent precursor sense ‘of ecclesiastical orders: minor, inferior in degree’ shown by gem¼ne in Old English. That this solution is a pragmatic one is however acknowledged by the explicit statement “The uses in branch II might be referred almost equally well to the native or to the foreign adjective; the truth is probably that the meanings of two originally quite distinct words have merged”. Thus far, the question has been where to assign in the larger structure what are nonetheless fairly coherent individual senses. The situation becomes more difficult still where the assignment of individual quotations to a particular sense is doubtful, and here recourse must be made to a note at the sense in question drawing attention to the difficulty: mean a.², OED3, sense 7c: c. Moderate in amount, or in degree of excellence; tolerable, mediocre. Later used only disparagingly (and so in some cases not readily distinguished from MEAN a.1). Obs. a1500 (a1460) Towneley Plays 12 My wynnyngis ar bot meyn, No wonder if that I be leyn. 1516 R. FABYAN New Chron. Eng. (1811) VI. cxciv. 197 She was..but of meane fayrenesse as other women were. 1546 Certificates Commissioners County of York (1895) II. 213 Of honest qualities and condicions, and meane lerenyng. 1551 R. ROBINSON tr. T. More Vtopia II. sig. Kivv, The resydewe they sell at reasonable and meane price. 1580 J. LYLY Euphues & his Eng. in Euphues (new ed.) f. 43v, Let thy apparell be but meane, neyther too braue.., nor to base. 1600 P. HOLLAND tr. Livy Rom. Hist. XLII. lxvi. 1155 The Consull contenting himselfe with a meane good hand..retired with his forces into the campe. 1604 E. GRIMESTON tr. J. de Acosta Nat. & Morall Hist. Indies IV. xxxiii. 299 In that countrie it is but a meane wealth. a1628 J. PRESTON New Covenant (1634) 24 It is better for thee..to have meane gifts, than to have high gifts. 1719 D. DEFOE Farther Adventures Robinson Crusoe 41 My own House..where I should see there had been but mean Improvements.

This is perhaps the most frustrating situation of all for the lexicographer, but is also perhaps the most interesting indicator of an area for further study by the specialist in historical semantics.

Of fæderan and eamas: avuncularity in Old English Andreas Fischer We begin with two aspects of Old English kinship terminology and Anglo-Saxon family structure. It is a well-known fact that the kinship terminology of Old English, like that of other early Germanic dialects, is of the so-called “bifurcate-collateral” type, which tends to have different words for maternal and paternal relatives: thus we have fadu ‘father’s sister’ and modrige ‘mother’s sister’, fædera ‘father’s brother’ and eam ‘mother’s brother’.1 It is equally well known that in early Germanic society families are held to have been of the “patrilineal extended” type, defined by Anderson as a “residential unit composed of the males of a paternal lineage, their wives, and the unmarried females begotten by the members”.2 When women married, they left this “residential unit” and joined that of their husbands. In families of this type, maternal uncles (in contrast to paternal ones), had a very special role: a woman who married “out” was potentially isolated, and if her husband’s family for some reason turned against her or her children she depended on the help of her nearest male relative, i.e. her brother. This ‘mother’s brother’ was thus expected to be a protector of his sister and her children, and there was a special bond of mutual obligation and affection between a maternal uncle (eam, avunculus) and his nieces and – especially – nephews. The name, in modern scholarship, for this phenomenon is “avuncularity”.3 1

The same goes for Latin, where we find amita ‘father’s sister’, matertera ‘mother’s sister’, patruus ‘father’s brother’ and avunculus ‘mother’s brother’. The Old English words are often, but not always, used as translations of the Latin ones. 2 R. T. Anderson, “Changing kinship in Europe”, Kroeber Anthropological Society Papers, 28 (1963), 1-48, 4. 3 Alexander Callander Murray, Germanic Kinship Structure: Studies in Law and Society in Antiquity and the Early Middle Ages, Studies and Texts 65, Toronto: Pontifical Institute of Mediaeval Studies, 1983, pp. 60-64; S. Graf von Pfeil, “Avunkulat”, Reallexikon der Germanischen Altertumskunde, Vol. 1, Berlin and New York: Walter de Gruyter, 1973, pp. 525-27.

68

Andreas Fischer

Avuncularity, like so many features attributed to early Germanic culture and society, is based on a passage in Tacitus’s Germania,4 but its actual importance is a matter of doubt: there are literary examples showing the special position of the maternal uncle, but none of the surviving laws document it. Pfeil thus concludes his discussion by stating that the special position of the mother’s brother was probably a social reality, but had no legal basis.5 The two phenomena just sketched, bifurcate-collateral kinship terminology on the one hand and avuncularity on the other, are linked, if only loosely. Different terms for paternal and maternal relatives may, but need not, mean that these relatives play different roles in the extended family. As a concept, avuncularity is of course more tangible when it goes together with a special word for the maternal uncle, but it does not depend on it. This loose connection between kinship terminology and avuncularity could be re-phrased in the form of two questions: Do societies that have avuncularity always (or: usually) have a special word for the maternal uncle, and is the word for maternal uncle more prominent in these societies than the word for paternal uncle? The first question can only be answered through a comparative study of kinship systems and kinship terminology in different societies.6 It is, therefore, the second question which I shall address in this paper, by investigating the way in which the two words fædera ‘paternal uncle’ and eam ‘maternal uncle’ are used in Old English texts.7

4

Josef Lindauer (trans. and ed.), Cornelius Tacitus: Germania / Bericht über Germanien: lateinisch und deutsch, München: Deutscher Taschenbuch Verlag, 1975. 5 Pfeil, op. cit., p. 526. 6 See, for example, William A. Foley, “Kinship”, Anthropological Linguistics: An Introduction, Language in Society 24, Oxford and Malden, Massachusetts: Blackwell, 1997, pp. 131-49. 7 On kinship in Anglo-Saxon England and kinship terminology in Old English see, for example, Andreas Fischer, “Notes on kinship terminology in the history of English”, in Katja Lenz and Ruth Möhlig (eds.), Of dyuersitie & chaunge of language: Essays Presented to Manfred Görlach on the Occasion of his 65th Birthday, Anglistische Forschungen 308, Heidelberg: C. Winter, 2002, pp. 115-28; Lorraine Lancaster, “Kinship in Anglo-Saxon society”, British Journal of Sociology, 9 (1958), 230-50, 359-77; H. R. Loyn, “Kinship in Anglo-Saxon England”, Anglo-Saxon England, 3 (1974), 197-209.

Of fæderan and eamas

69

Healey and Venezky’s Microfiche Concordance8 and the entries for the two words in the Dictionary of Old English (henceforth DOE)9 reveal that fædera and eam appear to have been used with about equal frequency: according to the DOE the corpus of surviving Old English texts contains “ca. 40” occurrences of the former and “ca. 45” of the latter. Numbers are not everything, however, and it is necessary to examine the evidence in more detail. I shall do this by looking at some representative texts, moving from simple cases (seen from the point of view of the question to be studied) to more complex ones. A very simple case is constituted by glossaries, which are essentially lists of Latin words with Old English translation equivalents. Such glossaries document Old English lexis, but they do not reveal much, if anything, about life in Anglo-Saxon England. Two of these glossaries, Antwerp Glossary 610 and Ælfric’s Glossary11 contain lists of kinship terms with fædera as a translation of patruus and eam as a translation of avunculus. Unlike the glossaries, the Old English Orosius12 is a coherent text, but, the Latin original being a history of the world written by a Spaniard in the early 5th century, it is not concerned with AngloSaxon England either. If we look at its kinship terminology, then we do so mainly to check how successfully the translator has rendered the wording and the meaning of the Latin text. The Orosius contains two examples of fædera and eleven examples of eam. While the two examples of fædera mean ‘father’s brother’, eam does not consistenly render Latin avunculus ‘maternal uncle’, but also occurs in places “where the Latin has either no equivalent or the term avus ‘grandfather’; the immediate source [of these problematic instances] may be either a gloss or commentary, or a Latin MS with a scribal error, though carelessness or deliberate emendation on the part of the 8

Antonette diPaolo Healey and Richard L. Venezky, A Microfiche Concordance to Old English, Toronto: Pontifical Institute of Mediaeval Studies, 1980. 9 Angus Cameron et al., Dictionary of Old English, Toronto: Pontifical Institute of Mediaeval Studies, 1986- . 10 L. Kindschi, The Latin-Old English Glossaries in Plantin-Moretus MS. 32 and British Museum MS. Additional 32246, Stanford diss., 1955. 11 Julius Zupitza (ed.), Ælfrics Grammatik und Glossar, Sammlung englischer Denkmäler in kritischen Ausgaben 1, Berlin: Weidmannsche Buchhandlung, 1880. 12 Janet Bately (ed.), The Old English Orosius, Early English Text Society ss 6, London: Oxford University Press, 1982.

70

Andreas Fischer

translator cannot be ruled out” (DOE s.v. eam 2). It is risky to draw conclusions from so little evidence, but the inconsistent use of eam in the Old English Orosius may indicate that it was a semantically less stable word than fædera. This difference in semantic stability may be connected with the etymology of the two words: Old English fædera is derived from fæder (compare Latin patruus < pater), i.e. a family member of the same generation, whereas eam (< West Germanic *awa-heim-) contains *awa ‘grandfather’ (compare Latin avunculus < avus ‘grandfather’), i.e. a member of the generation above.13 I will come back to this point at the end of my paper. We now turn to two texts concerned with matters Anglo-Saxon, namely Bede’s Historia Ecclesiastica and the Anglo-Saxon Chronicle. Authors of fact-based, historical texts like these are, of course, constrained by the reality they record. If a person’s paternal, rather than maternal, uncle dies, for example, they will record this fact irrespective of the lesser role the fædera may have played in AngloSaxon society in general. At the same time it must be remembered that a chronicler or historian is free, within certain limits at least, to refer to a person, variously, as A’s son, B’s father, C’s husband, D’s fædera, E’s eam, F’s nephew, G’s cousin, and so on, depending on the relationship the writer wants to foreground. Differences in the use of words like fædera and eam, therefore, may be attributed to cultural or textual rather than strictly historical factors. Bede’s Historia Ecclesiastica14 only contains one instance of eam, translating avunculus (hyre eames sunu < filius auunculi sui) and two instances of fædera translating patruus (his fæderan sunu < filius patrui eius, wið his fædran < contra [...] patruum suum).15 In the Anglo-Saxon Chronicle, however, eleven annals contain references to an uncle. I quote the respective passages from Swanton’s translation,16 but I have 13

Avunculus literally means ‘little grandfather’. Thomas Miller (ed.), The Old English Version of Bede’s Ecclesiastical History of the English People, 4 vols., Early English Text Society os 95 and 96, 110 and 11, London: Oxford University Press, 1890-98. 15 The three passages are to be found, respectively, in book V, ch. 17 (Miller, op. cit., p. 452, l. 22), book III, ch. 1 (ibid., p. 152, l. 6) and book III, ch. 18 (ibid., p. 236, ll. 9-10). 16 M. J. Swanton (trans. and ed.), The Anglo-Saxon Chronicle, London: J. M. Dent, 1996. Swanton’s translation follows the layout of the comparative edition by Earle and Plummer (p. xxx). I have adopted his method of giving the scribal manuscript 14

Of fæderan and eamas

71

added the actual words used in the Old English annals from Earle and Plummer’s edition:17 E 737 Here [...] King Ceolwulf received Peter’s tonsure; and granted his kingdom to Eadberht, the son of his paternal uncle [fæderan sunu]; [...].18 A 901 [899] Then Æthelwold, his father’s [Alfred’s] brother’s son [fæderan sunu] rode and seized the manor at Wimborne and at Twinham without leave of the king [Edward] and his councillors.19 D 1049 [1048] Here [...] Harald, the paternal uncle [fædera] of Magnus, went to Norway after Magnus was dead [...].20 C 1046 [1049] But Harold, his relative [mæg], fetched [Earl Beorn] from there and led [him] to Winchester and there buried [him] with King Cnut, his uncle [eam].21 D 1050 [1049] [...] then Earl Swein came with treachery, asked Earl Beorn, who was his uncle’s son [eames sunu], [...]22 D 1050 [1049] He [Beorn] was [...] interred with King Cnut, his uncle [eam].23 E 1066 [T]he king gave to St Peter and him [Leofric, abbot of Peterborough] the abbacy in Burton, and that of Coventry which the earl Leofric, who was his uncle [eam], had made earlier, and that of Crowland and that of Thorney.24 dating first and, where necessary, the adjusted dating afterwards, in square brackets (p. xvi). 17 John Earle and Charles Plummer (eds.), Two of the Saxon Chronicles Parallel. A revised text, edited, with introduction, notes, appendices and glossary by Charles Plummer on the basis of an edition by John Earle, 2 vols., Oxford: Clarendon Press, 1892-99. I have normalised spelling and morphology. 18 Swanton, op. cit., p. 45. 19 Ibid., p. 92, also p. 93 [D]. Reference is to a conflict between the two cousins Æthelwold and Edward, Edward being the son of King Alfred and Æthelwold the son of Alfred’s elder brother and predecessor Æthelred I. 20 Ibid., p. 167. 21 Ibid., p.168, also p. 171 [E]. 22 Ibid., p. 169. 23 Ibid., p. 170. 24 Ibid., p. 198.

72

Andreas Fischer

D 1079 Here Robert, the son of king William [I], ran from his father to his uncle [eam] Robert in Flanders, because his father would not let him govern his earldom in Normandy [...].25 E 1094 Also in this year the Scots trapped and killed Duncan [II], their king, and afterwards for a second time took Donald [Donald Bán], his paternal uncle [fædera], as their king, through whose instruction and instigation he was betrayed to death.26 E 1126 In this same year the king [Henry I] had his brother Robert taken from the bishop Roger of Salisbury, and committed him to his son Robert earl of Gloucester, and had him led to Bristol and there put in the castle. That was all done through the advice of his daughter [Maud] and through her uncle [eam], David, the king of Scots [brother-in-law of Henry I].27 E 1137 This year the king Stephen went across the sea to Normandy, and was received there because they imagined that he would be just like the uncle [eam, i.e. Henry I] was, and because he still had his treasury; but he distributed and scattered it stupidly.28

All instances, four of fædera and seven of eam, are motivated by historical facts and not, it seems, by cultural preference or textual factors.29 When, for example, earl Beorn’s eam is mentioned in the annals for the year 1049 (D 1050), this is not because there is special emphasis on the relationship between nephew and maternal uncle, but because that uncle, Cnut, had been king. Nevertheless, two passages quoted may perhaps be seen in the light of avuncularity and its “opposite”, i.e. a hostile relationship between paternal uncle and nephew. In the first (D 1079) Robert, son of king William I, runs away from his father and seeks refuge with his eam (note that uncle and 25

Ibid., pp. 213-14. Ibid., p. 230. 27 Ibid., p. 256. 28 Ibid., p. 263. 29 The explicit complex expressions fæderan sunu (E 737, A 901) and eames sunu (D 1050) may have been chosen because single-word alternatives such as nefa may have been semantically ambiguous. On the meaning of nefa in Old English charters see Kathryn A. Lowe, “Never say nefa again: problems of translation in Old English charters”, Neuphilologische Mitteilungen, 94 (1993), 27-35. 26

Of fæderan and eamas

73

nephew here even have the same name!), while in the second (E 1094) Duncan II of Scotland is betrayed by his fædera Donald Bán. It should be remembered, however, that both passages refer to events in the 11th century, whereas avuncularity has been postulated for early Germanic society. I will devote the rest of this paper to the poetry, which yields two examples of fædera (in Genesis) and three of eam (two in Beowulf, one in Riddle 46).30 The two occurrences of fædera in Genesis (ll. 1900 and 2080) are unremarkable. Reference in both cases is to Abraham and to Lot, who according to Genesis 12.27 and 14.12 are (paternal) uncle and nephew. In ll. 1900-01 Abraham says, stressing the mutuality of their blood-relationship: “Ic eom fædera þin / sibgebyrdum, þu min suhterga” [I am your paternal uncle / through blood-relationship, you are my nephew]31 and in l. 2089 Abraham is referred to as “fædera Lothes”. The two instances of eam in Beowulf (ll. 889 and 1117) are much more interesting with regard to the way family relationships are referred to. I begin with the second, from the so-called Finn episode, which the scop recites in Heorot during the festivities following Beowulf’s victory over Grendel (Beowulf ll. 1063/68-1159). Hildeburh, a woman from the tribe of the (Half-)Danes has been given in marriage to Finn, leader of another tribe, the Frisians. On the occasion of a visit by her brother Hnæf and his men, enmity breaks out and both Hildeburh’s son and her brother, the boy’s eam, are killed. Later in the story, the Danes, now led by a man called Hengest, take revenge, kill Finn and finally take Hildeburh, who has lost her husband, her son and her brother, back to her own tribe. The passage in question describes the burning of Hildeburh’s son’s and Hnæf’s bodies:

30

Genesis is quoted from George Philip Krapp (ed.), The Junius Manuscript, The Anglo-Saxon Poetic Records: A Collective Edition I, New York: Columbia University Press, 1936, Beowulf from Fr. Klaeber (ed.), Beowulf and the Fight at Finnsburg, 3rd edition with First and Second Supplements, Lexington, Massachusetts: D. C. Heath and Company, 1950, Riddle 46 from George Philip Krapp and Elliott van Kirk Dobbie (eds.), The Exeter Book, The Anglo-Saxon Poetic Records: A Collective Edition III, New York: Columbia University Press, 1936. 31 My translation.

74

Andreas Fischer Het ða Hildeburh æt Hnæfes ade selfre sunu sweolode befæstan, banfatu bærnan, ond on bæl don eame [emended, the MS. has earme] on eaxle. (ll. 1114-1117) [Then Hildeburh commanded at Hnæf’s pyre that her own son be consigned to the flames to be burnt, flesh and bone, placed on the pyre at his uncle’s shoulder;]32

Hildeburh’s situation is a prototypical example of a young woman in a society which has a family organisation of the “patrilineal extended” type. She has been married into a tribe which is not her own, and both she and her son might have to rely on her brother’s (his maternal uncle’s) support. Part of the tragedy of the Finn story lies in the fact that it is the visit of that very uncle which causes, or contributes to, the boy’s, his uncle’s and his father’s death. Note that eam in this passage is emended from earm,33 but it is undoubtedly the word intended by the Beowulf poet: the mentioning of eam here evokes what might be called “avuncularity ex negativo”: maternal uncle and nephew are united, but only in death, and the sister / mother is left without her brother and her son. The other passage in Beowulf where the word eam occurs is entirely different in nature. It is also part of a so-called digression, a story told by one of Hrothgar’s men during the ride back from Grendel’s mere (Beowulf ll. 875-900). Reference is made to Sigemund and his deeds, “his distant voyages / obscure, unknown to all the sons of men, / his feuds and crimes – except for Fitela, / when of such things he wished to speak to him, / uncle to nephew – for always they were, / in every combat, companions at need;”:34 [...] buton Fitela mid hine, þonne he swulces hwæt secgan wolde, eam his nefan, swa hie a wæron æt niða gehwam nydgesteallan; (ll. 879-882) 32

R. M. Liuzza (trans.), Beowulf: A New Verse Translation, Peterborough, Ontario: Broadview Press, 2000, p. 87. 33 The obviously erroneous earme ‘arm’ in the manuscript may have been prompted by eaxle ‘shoulder’ in the same line. 34 Liuzza, op. cit., p. 80.

Of fæderan and eamas

75

In this passage the close relationship between Sigemund (l. 875) and Fitela (l. 879) is emphasized twice: they are called nydgesteallan ‘friends in need’ and they are explicity referred to as eam and nefa.35 This may not be the whole story, however. In his note, Klaeber points out that according to the version of the story in the VËlsungasaga, SinfjËtli (Fitela) is not only Sigmundr’s (Sigemund’s) nephew, but through incest also his son: “The fact that Fitela is referred to as Sigemund’s nefa only (881), might perhaps be held to betoken Sigemund’s own ignorance of their true relation, or it may be attributed to the Christian author’s desire to suppress that morally revolting motive.”36 There is even a third possibility, namely that the poet did not mean to suppress anything, but used eam and nefa as darkly ironic references to incest. This interpretation is not as farfetched as one might think when one looks at the third and last instance of eam in poetry, in Riddle 46: Wer sæt æt wine mid his wifum twam ond his twegen suno ond his twa dohtor, swase gesweostor, and hyra suno twegen, freolico frumbearn; fæder wæs þær inne þara æþelinga æghwæðres mid, eam ond nefa. Ealra wæron fife eorla ond idesa insittendra.37 [A man was sitting at wine with his two wives and his two sons and his two daughters, gracious sisters, and their two sons, freeborn and firstborn children. The father of each of these noble youths was in there with them, uncle and nephew. There were in all five men and women sitting within.38]

The solution of this riddle, by general consent, is Lot’s family, Lot’s two ‘wives’ being his own daughters. According to the story told in 35

Pfeil, op. cit., p. 526, mentions this passage as one of the literary examples which document avuncularity. 36 Klaeber, op. cit., p. 159 (Klaeber’s emphasis). 37 Krapp and Dobbie, op. cit., p. 205. 38 S. A. J. Bradley, Anglo-Saxon Poetry: An Anthology of Old English Poems in Prose Translation, London: Dent, 1982, p. 380.

76

Andreas Fischer

Genesis 19:30-38, Lot, having left Sodom before its destruction, fled into the mountains with his daughters, who subsequently had a son each by him through incest.39 This short text being a riddle, the phrase eam ond nefa is clearly used as a means to mystify the reader and may be decoded in two ways. As Williamson points out in his commentary, both Lot’s daughters and sons are “all by definition siblings. Taking one son as Ego 1, his mother’s brother (eam) is the other son or Ego 2. If each son is eam to the other, then each must also be nefa, as that is the reciprocal relation (son of ego’s sister)”.40 However, since Lot’s sons, as sons of his daughters, are also his grandsons, the phrase eam ond nefa may also conjure up eam meaning ‘grandfather’ and nefa meaning ‘grandson’.41 However eam and nefa may be understood in this passage, it is clear that the expression does not evoke the positive relation of avuncularity, but something darker and altogether less wholesome. The question asked at the beginning of this paper was whether in societies that have avuncularity the word for maternal uncle is more prominent than the word for paternal uncle, or, more specifically, whether in the surviving Old English text eam is used more frequently and/or more prominently than fæder. The answer, to a large extent, is no: Healey and Venezky’s Microfiche Concordance and the respective entries in the DOE show that fædera and eam are about equally frequent in the surviving corpus of Old English and my study of some selected, representative texts has revealed hardly any really unusual occurrences of the two terms in question, fædera and, especially, eam. The big exception are the three examples from Beowulf and Riddle 46: not only are they the only three passages in the whole corpus that emphasize the role of the eam and the special relationship between eam and nefa, but they do so in a peculiar, inverted way: in the Finn episode in Beowulf, Hnæf and his nefa are united (only) in death and in the other two examples there are intimations of an unnatural form of avuncularity. The evidence, admittedly, is scanty, but it suggests 39

Lot’s wife, it will be recalled, had looked back upon the destruction of Sodom and had been turned into a pillar of salt (Genesis 19:26). Note that the incest in Genesis 19 happens between father and daughters, whereas the incest in the VËlsungasaga happens between brother and sister. 40 Craig Williamson (ed.), The Old English Riddles of the Exeter Book, Chapel Hill: University of North Carolina Press, 1977, pp. 283-84. 41 See my remark, above, on the etymology of eam and note 29 on nefa.

Of fæderan and eamas

77

that in the cultural memory of the Anglo-Saxons represented by Old English poetry avuncularity lingered on as a negative rather than a positive concept.

This page intentionally left blank

$ho:fian{*}/vK2: a LAEME-based lexical study1 Roger Lass & Margaret Laing As well as its primary function as the basis for a linguistic atlas, the LAEME database can serve as a quarry for many other kinds of historical linguistic research. In a volume honouring Christian Kay, a scholar who appreciates more than most both the tribulations and the rewards of long-term and large-scale scholarly enterprises, we offer this extract from ours. In recognition of her thesaurean interests and expertise, we have chosen to illustrate LAEME’s utility as a source for historical lexicography and semantics. The paper is designed to show, in a modest way, how the Protean nature of the LAEME database enables interdisciplinary investigations: our sources are the LAEME Corpus of Tagged Texts and the Etymological Corpus and the Corpus of Changes2. 1. The data 1.1. The following quotations from the LAEME corpus contain a verb which we have lemmatised as $ho:fian{*}/vK23 equivalent to MED 1

LAEME = Margaret Laing and Roger Lass, in preparation, A Linguistic Atlas of Early Middle English. This is one of the two major linguistic atlas projects being undertaken at the Institute for Historical Dialectology, University of Edinburgh. The other is Keith Williamson, in preparation, A Linguistic Atlas of Older Scots (cf. also the closely related Corpus of Scottish Correspondence, Anneli Meurman-Solin, University of Helsinki). The work of IHD is being supported by funding from AHRB for which we are very grateful. We are also grateful to Keith Williamson for very useful comments on this paper. 2 For descriptions of the different corpora (still in preparation) and the links between them see M. Laing and R. Lass, forthcoming, “Early middle English dialectology: problems and prospects”, in A. van Kemenade and B. Los (eds.), Handbook of the History of English. Oxford: Blackwell. 3 Our lemmatisations in the present LAEME-internal format have the following patterns. $ introduces the item being identified. The identifier may belong to a number of types. In this case the unattested OE item is qualified by the usual asterisk; though in this format, for internal computational reasons, it follows the item and is placed in braces. The / divides the lexical element of the label from any identifying

80

Lass & Laing

hǀven v. The semantic range covered by this word is complex: it appears to be a verb with both locative and motional components referring to air, or sea or land. MED divides the senses of hǀven v. into three main clusters. The first involves suspension, hovering, being poised and other stationary or locational properties. The second, not entirely distinct, involves waiting, expectation, residing and standing still. The third is motional, consisting of the senses ‘move onward, proceed, go, ride’. This paper attempts to reassess these clusters at a finer level of semantic resolution. Our lemma choice, an unattested Old English Class II weak verb, is also suggested as a possible etymon by OED and MED. This raises some interesting historical questions. We will attempt first to establish the shades of meaning in this verb’s usage within the LAEME corpus; then we will examine its possible etymology, and relation to an interesting class of Old English deverbal formations. We will present as examples all instances of this verb occurring in the LAEME corpus so far. Our first three appear in London, British Library, Arundel 292: The Bestiary,4 which dates probably from the last quarter of the thirteenth century and has been assigned to West Norfolk. We will translate each quoted passage leaving our verb in its original form for reference. In each case we will discuss its possible meaning in relation to MED’s definitions for these particular passages and to what our independent readings elicit. 1.2. Our first quotation (fol. 4v) comes from a passage that describes the eagle seeking out and circling above a well before ascending through the planetary spheres to the heaven. A Ïelle he sekeÇ Çat springeÇ ai . boÇe bi nigt a bi dai . ±er-ouer he flegeÇ . a up he teÇ . til-Çat he Çe heue ne seÇ . Çurg skies sexe a seuene . til he cumeÇ to heuene . So rigt so he cunne . houeÇ in Çe sunne . Çe sunne sÏi

grammatical information: here, v = verb, K2 = class II weak. 4 See C. Brown and R.H. Robbins (eds.), Index of Middle English Verse, New York: Columbia University Press: 1943, 3413. For editions see J. Hall (ed.), Selections from Early Middle English 1130-1250, 2 vols., Oxford: Clarendon Press, 1920: 1, XXI, 2, pp. 579-626, J.A.W. Bennett, G.V. Smithers and N. Davis (eds.), Early Middle English Verse and Prose, 2nd. edn., Oxford: Clarendon Press, 1968: XII, R. Morris, An Old English Miscellany, EETS OS 49, London: TrÝbner, 1872, pp. 1-25.

A LAEME-based lexical study

81

deÇ al his fligt . a oc it makeð his egen brigt . Hise feÇres fallen for Çe hete . a he dun mide to Çe Ïete . FalleÇ in Çat Ïelle grund . Çer he ÏurdeÇ heil a sund . ‘A well he seeks that always springs both night and day. He flies over it and up he rises until he sees the heaven, through the sixth and seventh spheres5 until he comes to heaven. As straight as he can [he] houeÇ in(to) the sun. The sun burns up his wings and also it makes his eyes bright. His feathers fall because of the heat and he [falls] down with them to the wet. He falls into the bottom of the well where he becomes whole and sound.’

The citation of this passage in MED comes under the following list of definitions: “hǀven v. 1. (a) To remain suspended in the air; of birds: hover; be poised high in the air, soar” (cf. also OED s.v. Hove, v.1, sense 1). This choice of possible meanings for houeÇ does not suggest directional motion; we are forced into a translation of the type ‘soars/hovers in the sun’. Once the sun has burnt the eagle’s ‘flight’ (i.e. wings) his feathers fall, causing him also to fall to the earth. The obvious picture given by this contrast is two contrary directional motions — like Icarus, the eagle flies too high (albeit deliberately) and suffers a consequent fall. The question is, does the verb houeÇ indicate part of the upward journey? It is true that the eagle’s ascent through the “sixth and seventh skies” to the heaven has already occurred. He could be in a position “to remain suspended in the air” in the burning heat of the sun. But what are we to do with the introductory clause of our passage So rigt so he cunne? The simplest translation for this is ‘as straight/directly as he can’ which would

5

Here we follow MED’s definition s.v. skÄ (e n. 1, (b) a sphere of the celestial realm. This interpretation is similar to that of Hall (op.cit., p. 597) whose note explains that “he flies through seven to the highest eighth, heaven”, following an astronomical definition from the South English Legendary. Bennett (in Bennett, Smithers and Davis op. cit., p. 356) takes a different view. He notes that cosmologically speaking the sun was supposed to be in the fourth sphere. This leads him to gloss skies as ‘clouds’ and he then takes sexe a seuene as “probably a mere tag”, which he takes to mean “whatever clouds may chance to come his way”. This explanation takes no account of the exegesis in the Significatio following the allegory. Here it states quite clearly that the “sun” in the context of the story represents God and therefore His rightful location would be above the highest celestial sphere.

82

Lass & Laing

imply continued linear motion6; in other words "in" may be illative not locative in this context and houeÇ would then be a verb of motion implying rising toward a goal. 1.3. The next two passages (fol. 8r) concern the whale. The first describes the whale tricking fish into becoming his prey. Çis fis Çat is vn ride . Çanne him7 hungreÇ he gapeÇ Ïide . Vt of his Çrote it smit an onde . Çe sÏetteste Çing Çat is o londe . Çer-fore oÇre fisses to him dragen . Ïan he it felen he aren fagen . he cumen a houen in his muÇ . of his sÏike he arn uncuð . §is cete Çanne his chaueles lu keÇ . Çise fisses alle in sukeÇ . ‘This fish that is monstrous, when he is hungry he gapes wide. Out of his throat there issues a breath, the sweetest thing that is on land. Therefore other fishes are attracted to him. When they sense it they are glad. They come and houen in(to) his mouth and of his trickery are unaware. This whale then locks his jaws, and sucks in all these fishes.’

MED places this citation in the following category: “hÉven v. 2. (b) to remain, stay, linger”. Although the context suggests that these definitions are indeed apposite8, the verb here is arguably also ambiguous with respect to illative and locative senses. Part of the interpretation depends on how the sequence “come and houen in” is to be bracketed. The question is whether the motional component inheres entirely in the verb “come” with “houen” providing a subsequent locative element, or whether “come” and “houen” together constitute an illative semantic unit. Which of the following forms should our interpretation take: [come and [houen in]] or [[come and houen] in(to)]? The MED classification appears to be semantically too tightly specified, though evidently the fish have 6

Note that the glossary to Bennett et al. s.v. rilt translates rigt here as “straight in front”. This appears to take no note of so...so and denies the possibility of forward motion which the construction most naturally implies. 7 MS him him 8 Though there seems no reason in principle why it could not equally well have been listed in MED s.v. “hÉven v. 1. (c) to float”; see the next quotation.

A LAEME-based lexical study

83

stopped moving of their own volition by the time they are sucked from the whale’s mouth into his gullet. However, not all contexts are ambiguous. Our verb is used in connection with the whale a little later in a passage describing how he behaves when the sea is upset by storms. This usage strongly reinforces the claim for a locative component. Çis fis ÏuneÇ ÏiÇ Çe se-grund . a liueÇ Çer eure heil a sund . til it cumeÇ Çe time . Çat storm stireÇ al Çe se . Çanne sumer a Ïinter Ïinnen . ne mai it Ïunen Çer-inne . So droui is te sees grund . ne mai he Ïunen Çer Çat stund . oc stireÇ up a houeÇ stille . Ïiles Çar Ïeder is so ille . ‘This fish dwells on the seabed and lives there ever whole and sound, till the time comes that a storm stirs up the sea. When summer and winter struggle it may not remain there. So troubled is the seabed that he cannot stay there at that time, but moves up and houeÇ still, while the weather there is so ill.’

In MED, this citation is listed s.v. hÉven v. 1. (c) to float; also, rise to the surface of the water. The following stille makes it clear that ‘float’ is the correct interpretation. The deliberate contrast of stireÇ up and houeÇ stille accentuates the distinction in type between the two verbs here. 1.4. Our next set of quotations comes from Cambridge, Corpus Christi College 145 (C), a copy of the South English Legendary9, dating probably from the first quarter of the fourteenth century and placed in NW Berks. 1.4.1. The first passage (fol. 72v) is from the life of St Brendan10. The story describes how seafarers mistake the whale for an island and go “ashore”. They light a fire and begin to heat a cauldron whereupon the whale starts to move and they quickly depart. Our passage tells of their return visit to the whale. 9

C. d’Evelyn and A.J. Mill (eds.), The South English Legendary, EETS OS 235, London: Oxford University Press, 1956. 10 See IMEV 2868, M. Görlach, The Textual Tradition of the South English Legendary, Leeds Texts and Monographs 6, Leeds: The University of Leeds School

84

Lass & Laing ¶is holymen wende uorÖ . and godes grace nome So Öat to Öe grete viss . wel sone suÖÖe hicome . And alond Öar hi houede . hor caudron hy fonde Öere As hi leuede uppon is rug . of Öan oÖer lere Louerd crist Öt such abest . houi ssolde so stille And soffri men Öer-uppe gon . and don al hore wille . ‘These holy men went forth and had god’s grace, so that to the great fish they came very soon afterwards. And on land where they houede, their cauldron they found there as they had left [it] upon his back the previous year. Lord Christ that such a beast should houi so still and suffer men to go upon him and do all their will.’

MED does not cite this quotation from the C text. Instead it uses the version of St Brendan in Oxford Bodleian Library, Laud Misc 108 (L)11. For LAEME the sample that has been tagged from L does not include the Brendan story. The two versions differ slightly, affecting the sense of the first example of our verb. The L text is quoted in MED as follows: “SLeg.Brendan (Ld) 373: To Öis grete fischse .. huy come, ¶at houede ase it were a lond. Ibid. 375: Louerd crist, Öat swch a best scholde houi so stille!”12 Both these contexts indicate a meaning that matches the locative sense (‘float’) in The Bestiary. The first instance in the C text above is, however, rather different. What does it mean? The context, alond, is something we have not met before. Whatever it means here, it cannot be either ‘soar’ or ‘float’. The sequence of events is that first the holy men approach the whaleisland, then they land on it and then they find their cauldron: hi come ... hi houede ... hy fonde. houede appears to mean ‘landed’ or ‘came of English, 1974, 167-68, d’Evelyn and Mill, op.cit., pp. 180-204. 11 C. Horstmann (ed.), The Early South English Legendary, EETS OS 87, London: TrÝbner, 1887, pp. 220-40. 12 Note that this passage is on fol. 106r of L. There are three other early 14th-century versions of SEL samples from which will in due course also be added to the LAEME corpus: Oxford Bodleian Library, Ashmole 43 (A), London, British Library, Egerton 2891 (E) and Harley 2277 (H). For the first citation in this passage, A, E and H all have: As alond Öt houede. For the second citation A has: houy scholde so stille but E and H lack our verb here. E has: hym scholde holde so stille; H has: scholde beo so stille. All these readings clearly carry the locative sense as does a further example (in L only) a few lines further on: And eueuere [sic] houede Ö is muche fichs = stille so eny ston. A, C, E and H all have simply was in this line.

A LAEME-based lexical study

85

ashore’ and thus contains both motional and locative components, i.e. it is allative13. The second occurrence (as in the L version) is quite clearly consonant with the established meaning ‘float’. 1.4.2. Our next example (C fols. 92r-v) occurs in the Life of St John the Baptist14. The Baptist’s head has been hidden and the secret of its location lost. The Baptist wishes his head to be found and he comes to the Abbot Marcel in a dream and tells him where to find it.

¶

¶is godeman sone aros him up . a ne abod nolt forte day He openede is celle dore . a biheld aboute A sterre he fond biuore Öe dore. houi Öere-wiÖ-oute . ¶e cleroste Öt milte be[o] . [...] ForÖ wende Öe sterre forte he com . ouer Öulke place . ¶er Öis holy heued lay . Öoru oure louerdes grace . ¶o he com ouer Öulke place . he houede Öere astonde ‘This good man soon arose and did not wait until day. He opened his cell door and looked about. He saw a star houi there outside the door, the brightest that could be. [...] Forth went the star until it came over that same place where this holy head lay, through our Lord’s grace. When it come over that same place it houede there a while.’

Another version of the same text is found in L, fols. 33v-34r15: we include this because it gives a grammatical form of the verb not found in the other examples16. ¶is guode man a-ros up sone Ÿ and ne a-bod noult forto day Ake he openede is celle-dore Ÿ and bi-heold a-boute 13 Though it is arguable that houede here refers back to the previous sojourn of the holy men on the back of the fish. We have translated the verb leuede as a pluperfect because it unambiguously refers back to the abandonment of the cauldron the previous year. The verb houede could also have pluperfect sense: ‘and on land where they had stayed they found their cauldron as they had left it’. In this case houede would have to be categorised with the senses in MED s.v. “hÉven v. 2. (b) to remain, stay, linger”. 14 IMEV 2945, Görlach, op. cit., p. 176, d’Evelyn and Mill, op.cit., pp. 241-46. 15 Horstmann, op.cit., pp. 29-33. 16 Note that versions A, E and H match C’s text in this example.

86

Lass & Laing Ane steorre he fond bi-fore Öe dore Ÿ houinde Öare-with-oute ¶e clereste steorre Öat milhte beoŸ [...] Forth wende Öe steorre forto huy comen Ÿ ouer Öat-ilke place ¶are his holie heue lay Ÿ ÖoruÑ ore louerdes grace ¶o heo cam ouer Öulke place Ÿ heo houede Öare ane stounde

MED lists only the second citation from the L version and places it under sense 1. (a) To remain suspended in the air. These usages relating to the star seem unproblematic. It is floating in the air, apparently at eye level (bi-fore Öe dore). First it is still, ‘waiting’ for the abbot, and then in motion (wende), as a guide to the location of the head, then still again (houede), marking the spot. This type of usage might be seen to form the basis for assigning the related senses of soaring in air or floating in water. 1.4.3. Our next example (Corpus fol. 87r) occurs in a miracle story of the Virgin Mary (the Miracle of Sir Emmery)17, told as part of the life of St Theophilus. The tale concerns a rich man fallen on hard times. The Devil tempts him with a promise of a return to his former wealth, and in exchange requires the knight to bring his wife to meet him. The knight, not realising it is the Devil, is delighted with the riches he has regained and duly brings his wife to meet his benefactor. On the way the wife (who is in ignorance of the agreement) asks to stop at Our Lady’s Chapel to pray. Here, unknown to the knight, Our Lady changes places with his wife and in her likeness goes with him to meet the Devil. Ac Öo Öe deuel oure leuedi sei . he gan to grede heie False trichour he sede to Öe knilt . wy bi-traistou me so . Ssel ich habbe Öis for me godhede . Öt ich habbe Öe ido Ne holde ich forward quaÖ Öe knilt . war-of destou mene ¶u luxt loude quaÖ Öe deuel . Öou brecst forward al clene ¶ou bringst mid Öe mi meste uo . a ssost me Öi wif lede 17 IMEV 59, Görlach, op.cit., pp. 173-74, d’Evelyn and Mill, op.cit., pp. 231-34. Note that the A version of SEL lacks the Life of Theophilus altogether and that the L version lacks the miracle story. The E and H versions of the relevant line match the C version except that H has al witles instead of as witles.

A LAEME-based lexical study

¶

87

¶is knilt houede as witles . he nuste hou he it sede ¶ou luÖer Öing quaÖ oure leuedi . wy wostou so fawe ¶t he Öe hadde is wif ibrolt . Öou wost it nere nolt lawe ‘But when the devil saw Our Lady he began to shout loudly. “False traitor”, he said to the knight “why do you betray me so? Is this what I get for the goodness that I have done for you?” “Do I not keep to the agreement?” said the knight, “What are you talking about?” “You’re clearly lying”, said the Devil, “You are breaking the agreement completely. You are bringing with you my greatest foe and you should be bringing me your wife”. This knight houede as if he were witless; he didn’t know how he [the Devil] could say it. “You wicked thing” said Our Lady, “Why did you think so readily that he would have brought his wife? You knew it would not be lawful”.’

MED places this citation in the same category as the fishes in the whale’s mouth, namely sense 2. (b) to remain, stay, linger. Here, however, there seems to be implied a behaviour associated with a specific state of mind. The knight is characterised as witless and unknowing. Given the previous usages we assume that this one is metaphorical: there is no indication that the knight’s feet are off the ground. Like the fishes with the whale or the whale in the storm, his situation has deprived him of the possibility of voluntary motion. ‘Remained’, ‘stayed’ or ‘lingered’ all seem a little colourless here. The sense needed is clearly ‘stood motionless’, ‘was paralysed’, ‘was unable to move’. If a metaphorical translation is preferred, perhaps the best would in fact be ‘floated’ or ‘hung there’. 1.4.4. Here it is worth mentioning another citation of our verb in MED s.v. hÉven v. 2. (c) ~ stilleliche, remain motionless, come to a stop. MED quotes from the L version of SEL’s Life of St Mary of Egypt18. This text has not yet been tagged for the LAEME corpus from any manuscript version. Here, meantime, we give a fuller context than MED, taking the text from the L version (fols. 119v-120r)19: A-non so heo bi-gan hire oresun Ÿ wonder men milten i-seo 18

IMEV 2990, Görlach, op. cit., 158-59, d’Evelyn and Mill, op. cit., pp. 136-48, Horstmann, op. cit., pp. 260-71 19 Horstmann, op.cit., p. 266.

88

Lass & Laing Fram Öe eorÖe heo was op i-houe Ÿ Öe heilte of fet Öreo ¶are heo houede stilleliche Ÿ as Öei heo were with-oute breÖe Heo ne wawede leome non Ÿ bote hire lippene vnneÖe [...] ¶is Monek stod and bi-Öoulte him Ÿ gret wonder he hadde and fere He ne heold hire no womman Ÿ ake treouwede Öat hit sum gost were For heo houede so a-boue Ÿ ake after hire oresun ¶is holi Öing wel stilleliche Ÿ to Öe eorÖe alilte a-doun ‘As soon as she began her orison, men could see a miracle. She was lifted up from the earth the height of three feet. There she houede motionless as though she were without breath. She did not move a limb; except only her lips [...] This monk stood and pondered — he had great wonder and fear. He considered her not to be a woman but that she was some kind of ghost because she houede so above [the ground]. But after her orison this holy thing very quietly alighted down to the earth.’

The verbal collocations in this example are quite explicit: St Mary is “lifted up”20 to “the height of three feet” and remains “motionless”. Unlike the knight in the previous example, her feet are certainly off the ground — exactly three feet off. This example clearly belongs more properly under MED’s sense 1. (a), along with St John the Baptist’s star, and may be translated ‘hover, remain suspended in the air’. Given the commentary above, MED’s three sense clusters appear to have a different set-theoretical structure than the three separate definitions would suggest. There are two major independent sense clusters and also a significant overlap between the two. Confining ourselves here to the LAEME examples, we give to the locative sense the mnemonic “float” (whether in air, water or land), we assign to the motional sense the mnemonic “land”, and to the intersection of the motional and locational sets the mnemonic “soar”. In summary: MOTION ‘land’ ‘soar’ ‘float’ LOCATION

20 i-houe is a the strong past participle from OE hebban directly related to our verb – see the following discussion.

A LAEME-based lexical study

89

Investigation of the etymology may give some indication as to why these apparently disparate senses co-occur. Indeed, OED lemmatises the two main sets of meanings separately under two homophonous but apparently historically unconnected head words, neither of which is provided with a firm etymology. The OED’s “Hove, v.1”, which is described as “derivation unknown” covers mainly the locational senses but also overlaps with the motional ones. Its “Hove, v.2” entry lacks any derivation and is given the primary sense “raise, lift”. In the following section we will attempt to connect the two senses and suggest a common etymology. 2. Etymology We can find no evidence for our verb in the standard Old English dictionaries or the Microfiche Concordance To Old English (R.L.Venezky and A di P. Healey, 1980). The only attempt to connect it with an attested Old English word is a suggestion in MED: “?OE *hÉfian; ?a new formation in ME based on hÉf(on, p. of OE hebban”. This etymology seems very likely to be right, but the supporting argumentation needs to be made explicit. Given the lack of attestation, why are we so sure (a) that there was an ancestral Old English verb at all and (b) that if there were it would have been a Class II weak verb? We need to ask question (a) because there are not only no pre-Conquest attestations, but there appear to be none before the early 13th century. The reason we assume an actual Old English verb rather than a Middle English secondary formation is the geographical distribution of the forms. Its occurrence in texts originating from places right across the country – Robert of Gloucester’s Chronicle21, The South English Legendary and The Bestiary – is less likely to be the result of convergent invention than of inheritance. This is essentially a methodological rather than a purely empirical argument. Assuming then that there was an ancestor why should we characterise it as Class II weak? Firstly, the ancestral verb must have been weak since there is no evidence of any strong forms but there are multiple instances of the weak past tense. That being the case, the verb must have belonged to Class II for the following reasons: 21

See the citations in MED s.v. hǀven v.

90

Lass & Laing

(1) It has a back root vocalism; this excludes weak Class I whose vowels are always front because of the ancestral *-jan suffix. (2) It does not have a geminate versus non-geminate alternation anywhere in the paradigm, which excludes the characteristic Class III type (libban, habban). (3) At least one of the Middle English forms show the characteristic marker of Class II: the -i- formative in the infinitive, houi. Let us assume OE *hǀfian to have existed. As a Class II verb, it was almost certainly derived, as Class II consists mostly of denominal, deadjectival and deverbal secondary formations. It is in fact historically the same class as that to which Latin denominal -Ɨre verbs belong, e.g. am-or > am-Ɨre22. The source suggested by MED is a reasonable one for a verb of this shape. Since Class II formations leave the root untouched, suffixation of the past root of hebban (strong Class VII) would give *hǀf-ǀj-an > *hǀf-jan > *hǀfian. The remaining problem is semantic: how do we get from the past tense of ‘heave’ to the rich collection of senses we have been discussing? As a Class VII strong verb the expected ablaut series for hebban would be a / ǀ / ǀ / a. The unexpected e-vocalism in the present is due to an original *i formative like that seen in its almost certain cognate Latin cap-i-ǀ. The best story of the complex semantic development, which is still incomplete, is outlined in the OED’s entry s.v. Heave v. Here it is with expansions where necessary: the IE root must be of the shape *kap-, which would by Grimm’s Law have given Gmc *xaf-; this, with following standard changes, is clearly the shape underlying hebban. According to the OED, the original sense “as evidenced by ... derivatives, as well as by L capƟre, was ‘take’, whence, through ‘take up’, came that of ‘lift, raise’, already developed in Com. Teut.” A similar sense development could very well take us from ‘lift’ to ‘float’ and thence to ‘float’ in the motional sense. The history and the attested senses would then account for why $ho:fian{*}/vK2 is both locative and motional. 22

For detailed discussion see R. Lass, “Old English -ian: Inflectional or derivational?”, Vienna English Working Papers 2.1, 1993.

The rhyme potential of Scots Caroline Macafee Some of my happiest memories of teaching at the University of Glasgow are of the course I shared with Christian, entitled “Meaning, Form and Style”. I particularly enjoyed, and learnt a great deal from, her handouts on formal poetics. This paper harks back to those days, and also picks up her interest in ripping out dictionaries and knitting them up again. It has always seemed to me a pity that there is no rhyming dictionary of Scots. A dictionary of rhymes is one of the standard reference tools available to the writer in English, but no such tool exists for Scots, despite writers more and more turning to reference books to remind them of the riches of the language, and to confirm and broaden their personal knowledge, as the language becomes less used in speech. A rhyming dictionary would be useful to poets and song-writers in Scots, and to teachers and students using Scots for creative writing in the classroom. It would also promote the appreciation and understanding of Scottish literature, by recovering poets’ intentions with regard to rhyme, and might even improve accuracy in the recitation and performance of Scots verse and song. The core material for such a dictionary already exists in the pronunciation entries1 in The Concise Scots Dictionary (CSD).2 At present, this material, which constitutes a considerable enlargement, revision and systematisation of the pronunciations in The Scottish National Dictionary (SND), is underused, because it demands an acquaintance with phonetic symbols.3 There is further relevant material in The Linguistic Atlas of Scotland, vol. 3 (LAS3),4 in Ulster 1

By A. J. Aitken. Mairi Robinson (ed.), The Concise Scots Dictionary, Aberdeen: Aberdeen University Press, 1985; now published Edinburgh: Polygon. 3 William Grant et al. (eds.), The Scottish National Dictionary, Edinburgh: Scottish National Dictionary Association, 1931-1975; now published Edinburgh: Polygon. 4 James Y. Mather and H. H. Speitel (eds.), The Linguistic Atlas of Scotland, vol. 3, London: Croom Helm, 1986. 2

92

Caroline Macafee

sources5 and in sources for Orkney and Shetland.6 Standard English overlaps with Scots; such material (words like cat, the, bus, narrative and ode) would also have to be included. Even with the addition of the shared material, a pocket dictionary would suffice to cover the active, current Scots vocabulary of most speakers. But the more comprehensive the dictionary was, the more useful it would be to writers of dialects whose rhyme potential is radically different from mainstream literary Scots. It would also be satisfying to give writers better access to the recessive and less wellknown vocabulary. Much of the impetus for a thoroughly inclusive treatment would really be scholarly, however, rather than utilitarian. A fuller rhyming dictionary would incorporate vocabulary back to 1700 (the starting date of SND), and would benefit scholars of Scottish literature by making it possible for the first time to see the relationship between a poet’s rhyme choices and the full rhyme potential of the language at a particular point in time. A rhyming dictionary could also be presented in such a way as to give it a valuable secondary function as a pronouncing dictionary. An analysis of the vocabulary by rhyme sets is, after all, precisely an analysis by phonemes. The general user (and the literature student) could be spared an engagement with the International Phonetic Alphabet (IPA), by identifying the pronunciation of a given word in terms of what it rhymes with. A full index (after the fashion of The Scots Thesaurus7 or the Penguin edition of Roget’s Thesaurus8) would be space-consuming, but very direct and reliable in use – were it not for the vagaries of Scots spelling. Alternatively, the user might be directed to work down a list of keywords, arranged in some quasi-

5

James Fenton, The Hamely Tongue. A Personal Record of Ulster-Scots in County Antrim, 2nd edition [no place of publication]: Ullans Press, 2000; Caroline Macafee (ed.), A Concise Ulster Dictionary, Oxford: Oxford University Press, 1996. 6 Such as Gregor Lamb, Orkney Wordbook. A Dictionary of the Dialect of Orkney, Birsay: Byrgisey, 1988; and John Graham, The Shetland Dictionary, Stornoway: Thule, 1979. Material unique to the Northern Isles was omitted from CSD as a spacesaving measure. 7 Iseabail Macleod et al., (eds.), The Scots Thesaurus, Aberdeen: Aberdeen University Press, 1990; now published Edinburgh: Polygon. 8 Robert Dutch (rev.), Roget’s Thesaurus, Harmondsworth: Penguin, 1966.

The rhyme potential of Scots

93

alphabetical way by stressed vowel (echoing the arrangement of the rhyme lists themselves) in search of the right category of rhyme. It would be possible to extend the coverage backwards in time, because CSD does give pronunciations for obsolete vocabulary. The beginning of Modern Scots at 17009 is probably as far back as it would be reasonable to go. For one thing, Older Scots is too far back for the modern writer to go in search of rhymes: the literature is not well enough known to modern readers, whereas with the eighteenth century, we are into a Burnsian and ballad vocabulary (though I am perhaps being wildly optimistic in regarding these as common cultural reference points). For another thing, there are very few people who have occasion to read aloud in carefully reconstructed Older Scots, and those who would attempt it can no doubt handle IPA, and can therefore refer directly to CSD, and to other, more specialised, resources.10 The question arises of how accurate and complete our knowledge is of the pronunciation of words that did not survive into the era of modern dialectology. In fact, we can reconstruct the pronunciation of most words with some confidence, within a robust and detailed reconstruction, based on spellings (including dialect variants), etymologies, and knowledge of rhymes in Older Scots. The biggest source of uncertainty is the break in the written tradition at the end of the Older Scots period. On one level, there is the discontinuity in orthography; on another, there is the much narrower range of genres and registers being written in early Modern Scots. These changes make it look as if much of the variation that existed in Older Scots had disappeared by the eighteenth century. But many alternative forms, the results of minor sound-changes whose geographical spread is not always clear, may in fact have lingered on in speech. The study of 9

Also the dividing date between SND and A Dictionary of the Older Scottish Tongue. In particular A. J. Aitken, “How to pronounce Older Scots” in A. J. Aitken et al. (eds.), Bards and Makars: Scottish Language and Literature, Medieval and Renaissance, Glasgow: Glasgow University Press, 1977, pp. 1-21; revised Caroline Macafee, incorporating material by the late A. J. Aitken, “The phonology of Older Scots”, in John Corbett et al. (eds.), The Edinburgh Companion to Scots, Edinburgh: Edinburgh University Press, 2003, pp. 138–169; and †A. J. Aitken, The Older Scots Vowels: A History of the Stressed Vowels of Older Scots from the Beginnings to the Eighteenth Century, ed. Caroline Macafee, Scottish Text Society, 2002, with its substantial index of Scots words with their stressed vowels indicated. 10

94

Caroline Macafee

early Modern Scots rhymes might even bring some of these to light. First of all, though, we need a set of expectations, a first approximation, such as a rhyming dictionary would provide. Any discrepancies from the expected patterns of rhyme warrant attention. To give a small example, it seems to have gone unremarked that the rhyming of joy as /ÙA;e/ is an anglicised rhyme (/ÙA;e/ being an alternative, now obsolete, form in English). The first instance mentioned in SND (s.v. joy) is Burns in “The Brigs of Ayr” (ll.219, 220), but he is in any case writing English at this point: “Then, crown’d with flow’ry hay, came Rural Joy, And Summer, with his fervid-beaming eye.” (The Scots for eye is, of course, ee.) The two subsequent nineteenth-century examples mentioned by SND may or may not derive from Burns. Burns also has an internal rhyme of joy with cry in “Song, composed in August”. But at least once elsewhere, he rhymes joy with expected /@i/: The best laid schemes o’ Mice an’ Men, Gang aft agley, An’ lea’e us nought but grief and pain, For promis’d joy! (from “To a Mouse”11)

That /@i/ rather than /A;e/ is the expected vowel in joy is not noticed by SND, nor, surprisingly, by CSD. In current speech, it has been replaced by /oi/;12 but with its original vowel, Vowel 10 (to use Aitken’s numbering system), it should be /@i/ and indeed rhyme with word-final Vowel 8 as in agley.13 It is possible that /@i/ and /A;e/ were still phonetically close enough at this time to make a decent nearrhyme. A study of rhymes amongst Vowels 1, 8 and 10 in Burns and other eighteenth-century poets would be interesting.14 11

Quotations from Burns are from James Kinsley (ed.), Burns Poems and Songs, Oxford: Oxford University Press, 1969. 12 Perhaps itself a partial anglicisation, as being the Scots realisation of /OI/. 13 The diphthongal reflex of ai and ei word-finally. 14 It is remotely possible that Vowel 10 shared with Vowel 1 (as in Fife versus five, etc.) the split into /@i/ and /A;e/ according to the environments of the Scottish VowelLength Rule. There are so few examples in Scottish Vowel-Length Rule long environments that it is hard to draw any conclusions. However, poison has /@i/, suggesting that there was no such split in Vowel 10. That the split had already occurred in Vowel 1 is shown by the merger of the short alternative with Vowel 8.

The rhyme potential of Scots

95

The Scots in eighteenth- and nineteenth-century texts presents the problem for the reader that it is often disguised by Standard English spellings. There are several examples in the early Burns piece, “Song, composed in August”. The spelling is almost entirely English, but swallow (read swalla) rhymes scottice with yellow (read yalla), and Nature (read naitur) with creature (read craiter); and there are good internal rhymes of winds (read wuns) with guns, and join (read jine) with combine; and a near-rhyme of bush (read buss) with thrush. (Recordings by the Voice Squad15 and by the Old Blind Dogs16 both closely follow the misleading spellings of the published text.) It would be pleasing to think that a rhyming dictionary could help scholars and performers of verse and song to recover the Scots rhymes in Burns, though there have been two attempts to enlighten the public with full transcriptions of the poems,17 and yet the horse will not drink. Perhaps accuracy in delivering rhymes is something that only matters to hopeless pedants. I have needlessly allowed my enjoyment of folksongs to be spoiled by noticing, for instance, that a singer of “My Harry was a gallant gay” rhymed gay and awa – why not sing away, perfectly good Central Scots variant? Or, in “True Thomas”, ahint : wind – if not the ballad Scots behind : wind (with /@i/) then surely ahin : win? And what has got into a singer of “Logie o’ Buchan” who normally gives awa the North-Eastern /A/ vowel but adopts a Central Scots /O/ every time she renders the chorus: “Think nae lang, lassie, though I’m far awa, An’ I’ll come back an’ see ye, in spite o’ them a’”? And I won’t even mention the effect it has on me to hear shoon rhymed as it is spelled, with moon or whatever. Perhaps it is too much to expect from the tradition bearers that they should direct their attention to pronunciation. The resulting selfconsciousness might be counter-productive. In an interview filmed by Doc Rowe,18 the late Willie Scott talked about giving words their ‘right expression’, by which he meant pronouncing them as spelled, and indeed in the same interview he sang “Hame fareweel” with disown us as /dIsVun Vz/. 15

Many’s the Foolish Youth, Tara Music Co, Dublin, 1987. New Tricks, Klub Records, Glasgow, 1992. 17 James Wilson, Scottish Poems of Robert Burns in his Native Dialect, London: Oxford University Press, 1925; Robert Pate, 120 Scotch Poems of Robert Burns in his own Dialect, Wigtown: G. C. Book Publishers, 1997. 18 Shown at the Songs and Singing event in Auchtermuchty, 17 October, 1998. 16

96

Caroline Macafee

Folk performers may feel comfortable with an unexamined compromise amongst their own Scots (if any), the Scots of the singer(s) they learned the song from, and the pronunciations suggested to them, naively, by any written versions that they have seen. Perhaps it does not matter that John Strachan, for instance, in his traditional rendering of “Clyde’s Water”19 sang: The horse that I’m to ride upon Cost me twice thirty poun’ An I’ll pit trust in ma ain horse heels An he’ll cairry me safe an soon … … Oh, Clyde, ye Clyde, ye rollin Clyde, Your waves are wond’rous strong; Mak me a wreck as I come back, But spare me as I gyang.

/pVun/ /sun/

/stroN/ /gjaN/

… … Willie’s brother stands on the bank: “Oh, how can Willie droon? Oh, turn ye tae your high horse heid An he’ll learn ye how tae sweem.” … … “Oh why could I turn to my high horse head An learn how tae swim? It’s the deepest pot in aw the Clyde, An here that I maun droon.”

whereas he could have rhymed poon : soon, strang : g(y)ang and, as a near-rhyme, droon : soom. In the ordinary way of things, a performer cannot be expected to revive an archaic pronunciation in order to mend a rhyme, and even moribund rhymes are an obstacle to comprehension, so we do not necessarily expect to hear, for instance, the Central Scots /I/ vowel in croon : afternoon : shoon : tune in a Burns Supper rendition of “The Holy Fair”. But as Aitken said about the pronunciation of Older Scots,20 we do expect scholars and students to prepare their readings carefully. 19

Scottish Tradition 5: The Muckle Sangs: Classic Scots Ballads, School of Scottish Studies, University of Edinburgh and Tangent Records, London, 1975. 20 Aitkin (1977), op. cit.

The rhyme potential of Scots

97

The intended pronunciations in early Modern Scots works are knowable: the less readers need specialised training to get at that knowledge, the more chance, one hopes, of its being applied. As with Older Scots, there are decisions that have to be made, and problems that have to be thought out in advance of attempting a reading. So, for instance, in Burns’ “Song: Tibby I hae seen the day”, there is a rhyme moor : stoor : poor, which can only be made to work by reading moor and poor as English: Yestreen I met you on the Moor Ye spak’na but gaed by like stoor Ye geck at me because I’m poor But fien’ a hair care I. –

In “To the Rev. John McMath, Inclosing a copy of Holy Willie’s Prayer, which he had requested”, mouth and skouth must be given their Scots pronunciations, but truth and ruth, which would take /I/, must be given their English pronunciations: They take religion in their mouth; They talk o’ mercy, grace an’ truth, For what? - to gie their malice skouth On some puir wight, An’ hunt him down, o’er right an’ ruth, To ruin streight.21

There are times, however, when we must admit ourselves defeated by an eye-rhyme, as in: In thae auld times, they thought the Moon, Just like a sark, or pair o’ shoon, Woor by degrees, till her last roon Gaed past their viewin, Ah’ shortly after she was done They gat a new ane. (from Burns, “To W. S*****n, Ochiltree”)

21

The spelling <streight> suggests that a near-rhyme /wIxt : strExt/ was intended.

98

Caroline Macafee

Moon, shoon, done will all rhyme, in /I/, but roon is an eye rhyme with the spellings. There is no combination of Scots and English pronunciations that will make all four of these words genuinely rhyme with each other. One of my own main interests in a rhyming dictionary is that it would make clear, for the first time, the different rhyme potentials of different dialects. Numerous sound-changes separate the Central, Southern, Northern and Insular dialect groups, some of which go back at least six centuries. The vocabulary may be shared, but the rhymesets have been extensively reshuffled, presenting poets and songwriters with quite different linguistic resources. Some dialects, including an eastern bloc that extends northwards from Fife, and parts of the South-West, have /e/ as the normal reflex of Vowel 3, as in meat, head, etc. For these dialects, as for the Fife makar, David Lyndsay, Vowel 3 rhymes with Vowel 4, as in mate, rather than Vowel 2, as in meet. Only the most conservative dialects, now mainly Orkney and Shetland, keep Vowel 7, as in spune ‘spoon’, use, puir ‘poor’, separate as a front rounded vowel. In most it has unrounded to merge with /I/ in the short enviroments of the SVLR, /e/ in the long ones, giving rhymes such as spune and tin, puir and hair. In Angus, the traditional reflex of Vowel 7 is /e/ even in SVLR-short environments – though this may now be moribund – thus spune ‘spoon’ would rhyme with stane ‘stone’. In Southern Scots, the diphthongisation of final /u/ leads in some localities to a merger with /Vu/, making you and pu ‘pull’ rhyme with growe ‘grow’, how ‘hoe’ etc. It is noticeable, however, that the word zoo, one of the test words in LAS3, was never given with a diphthong: evidently *zow would not be acceptable. There are other instances where English word-forms have, in effect, been borrowed into Scots, one of the earliest being boat, which existed alongside the expected bate (now obsolete) from the late fifteenth century (spelled ). Door is another very ordinary word that has been present in its English form since the sixteenth century (and dure is also now obsolete, without apparently having yielded a /der/ form to parallel flair, or flure, ‘floor’). The recent large-scale obsolescence of Scots word-forms in favour of their English cognates, is, of course, a phenomenon of language death, so there is a case for presenting the dictionary user with the historically ‘correct’ forms, giving the

The rhyme potential of Scots

99

necessary indication, of course, that they are, or are believed to be, obsolete. But, despite the detailed coverage in CSD, there are items where the predictable dialect form might never have been attested, and might not, in practice, be acceptable to speakers, e.g. grue ‘melting ice on water’ is recorded as a lexical item, from Southern Scots, but with no record of a dipthongal form, in contrast to grue ‘shudder’, where the expected Southern form is attested. The North-Eastern dialect has had its rhyme potential enlarged by the falling together, as /i/, of Vowel 7 (as in puir ‘poor’, spune ‘spoon’) with the already merged Vowels 2 and 3 (like English as in meet and meat, and also, unlike English, as in heid ‘head’, pear ‘pear’) and Vowel 4 before /n/ (as in stane ‘stone’). This feature of the dialect is the subject of an anecdote told by Dieth: A rather fantastic story, associated with the name of a well-known Scottish judge of a former generation, is told of an Aberdeen woman who made a verbal will bequeathing her money to “the peer o’ Aiberdeen”. A legal decision was wanted as to what she had meant. The possible claimants were (I) the town authorities for the relief of the poor, (II) the Earl of Aberdeen, (III) Aberdeen Harbour authorities, then constructing a pier. It was wittily suggested that there was a fourth possibility, that she meant the money to be spent in encouraging the cultivation of a new kind of pear-tree to be called the “peer o’ Aiberdeen”. The four words, poor, peer, pier, pear are all pronounced alike.22

The North-Eastern poet, J. C. Milne, showed great facility with the rhyme potential of the dialect. In “Dominie Dandie (Two)”,23 for instance, he produced twelve different rhymes for skweel ‘school’.24 As a basis for comparison, The Penguin Rhyming Dictionary offers the following rhymes for ‘school’ in English:25

22 Eugen Dieth, A Grammar of the Buchan Dialect, Cambridge: Cambridge University Press, 1932, p. 19. 23 Quoted from Leslie Wheeler (ed.), Ten Northeast Poets, Aberdeen: Aberdeen University Press, 1985, pp. 101–02. 24 With a /w/ glide after the velar. 25 Rosalind Fergusson (ed.), The Penguin Rhyming Dictionary, Harmondsworth: Penguin, 1985.

100

Caroline Macafee boulle (denoting type of marquetry) … Boole … cool, fool, ghoul, Goole …, joule, who’ll,26 pool, Poole, pul (Afghan coin), spol, rule, drool, shul (synagogue), tool, stool, yule, you’ll, mule, mewl, pule …, tulle, babul (type of acacia), Stambul … cagoule … ampoule … ferrule … spherule

and some others, mainly compounds of the words above, and many more ending in -ule. So we turn to Milne: I’ve been te skweel and college and hae ta’en a gweed degree, And noo I think I’m thinkin it’s a teacher I wid be! And in twa-three years I’se warrant, gin the warld’s waggin weel, They’ll mak me Dominie Dandie wi’ a couthy country skweel! … … And ilka Sunday mornin te the kirk I’d walk in style, Wi’ a muckle black umberella and a swagger and a smile, And I’d nivver miss a sermon for widder, sark or peel, Gin I were Dominie Dandie wi’ a couthy country skweel.

With no feeling of strain or recherché ingenuity, he goes on for sixteen stanzas in this vein, repeating only weel (twice) and chiel (once). To list his rhymes, they are: weel ‘well’, chiel ‘fellow’, peel ‘pill’, creel, tattie-dreels ‘potato drills’, reel (fishing), feel ‘fool’, Deil, reel (the dance), Eile ‘Yule’, queel ‘cool’, spiel ‘climb’, heel. Of these, only feel ‘fool’, Eile ‘Yule’, and queel ‘cool’ also figure in the English list. As with many aspects of the Scots language, its rhyme potential is in danger of being lost without ever having been fully appreciated. Sadly, a rhyming dictionary is a tool that might have been of more use a couple of generations ago, when a wider range of people knew a broader canon of Scots. Hugh MacDiarmid might have found inspiration in it, as he did Wilson’s The Dialects of Central Scotland.27 But at the end of the day, no lexicographer could anticipate

26 This would not work in Scottish English, as the vowel is long before a morpheme boundary. 27 James Wilson, The Dialects of Central Scotland, London: Oxford University Press, 1926.

The rhyme potential of Scots

101

the flights of someone like George Bruce Thomson,28 steeped in the genuine idiom of a vibrant dialect, and with his own original genius for rhyme: He said that he was able for tae play at coup-the-ladle, Wi’ a laidder ower a tricle cask, an’ ca the churn forby; Anidder o his win’ers wis that sawdust mixed wi’ cin’ers Wis their spice for feedin’ hens at Birnieboosie. An educatit ostrich fae the wilds o’ Timbuctoo, He hid for scrattin’ up his neeps an’ hidna them tae pu’; I never heard the like o’ that come oot o’ ony mou’ But Macfarlan’s o’ the Sprotts o’ Birnieboosie. (from “Macfarlan o’ the Sprotts”29)

28

Thomson died young, and left only a few pieces, including the two marvellous comic songs, “Macfarlan o’ the Sprotts” and “McGinty’s Meal an Ale”. 29 Quoted from James Alison (ed.), Poetry of Northeast Scotland, London and Edinburgh: Heinemann Educational Books, 1976, pp. 98–99.

This page intentionally left blank

Of politeness and people Terttu Nevalainen & Heli Tissari Introduction English speakers have not always been polite. The adjective was first attested in English around 1400 with reference to smoothed or polished objects, and it continued to be associated with materials like glass and stone until the eighteenth century. What interests us in this paper is the transition of polite from the material sphere to the social, and the ways in which this process can be contextualized. We shall begin by considering some corpus linguistic evidence in order to trace the process across time – how and when being polite became a human property. The information provided by the Historical Thesaurus of English shows the lexical company “polite” words keep, and gives us access to the metaphors associated with “good behaviour” and “courtesy”. We will conclude by discussing how this Thesaurus information could feed into scripts for typical polite behaviour in English between 1500 and 1800.1 1. Sources and uses of “polite” adjectives The present-day use of the adjective polite is restricted to humans, their actions, or the results of their actions. In the British National Corpus (BNC), polite typically collocates with nouns such as conversation, society, smile, applause, way, letter(s), interest, words and company, and takes human complements with the verb to be.2 Examples (1) to (4) illustrate these uses: 1

Our research for this paper was supported by the Academy of Finland Centre of Excellence funding for the Research Unit for Variation and Change in English at the Department of English, University of Helsinki. We would like to extend our heartfelt thanks to Professor Christian Kay for the advice and support she has given VARIENG over the last six years. 2 The corpora referred to in this study include the following four collections: (1) the British National Corpus (BNC web online created at the University of Zurich, see ); (2) the Early Modern English section of the

104

Nevalainen & Tissari (1) He could dance as well as fight, and he could make polite conversation. (BNC, EVF 354) (2) I have now received a polite letter from the B.T. together with a phone card worth 30 p. (BNC, CH5 3047) (3) I always had to be polite and caring. (BNC, AJK 131) (4) When they visit I expect my husband to be polite to them. (BNC, AC3 1592)

The adjective polite is more frequent than its near-synonym courteous but less frequent than civil. The most common noun collocates of courteous in the BNC are manner, letters, man, staff and way, partly overlapping with the principal collocates of polite. The polysemy of civil accounts for its overall frequency: its primary senses relate to the citizens or people of a country in contrast to military or religious institutions. The principal collocates of civil in the BNC are war, servants, service, rights, aviation, liberties, society, engineering, law and disobedience, none of which have the “polite” sense of the word. The three “polite” adjectives date from different periods and have different source domains. Courteous goes back to Middle English and is associated with courtly behaviour: “manners such as befit the court of a prince; polite, kind and considerate”. Civil, by contrast, pertains to citizens; dictionaries suggest its ‘civilized’, ‘educated’, and ‘wellbred’ senses appeared in the sixteenth century, giving rise to ‘polite’, ‘obliging’ and ‘uneffusively courteous’ in the early seventeenth century. Polite is the latest of the three; besides its literal sense, it had two extended senses at the turn of the sixteenth century: ‘cleansed’, ‘neat’, ‘orderly’ used with reference to material entities, and ‘refined’, ‘elegant’, ‘cultivated’ used about intellectual pursuits such as the arts. Helsinki Corpus of English Texts (for HCE compilers and lists of texts, see Matti Rissanen, Merja Kytö and Minna Palander-Collin (eds.), Early English in the Computer Age: Explorations through the Helsinki Corpus, Berlin and New York: Mouton de Gruyter, 1993); (3) the Corpus of Early English Correspondence (CEEC) and its 18th-century Extension (CEECE) (for compilers and lists of texts, see Terttu Nevalainen and Helena Raumolin-Brunberg, Historical Sociolinguistics: Language Change in Tudor and Stuart England, London: Longman, 2003, and Mikko Laitinen, “Extending the Corpus of Early English Correspondence to the 18th Century”, Helsinki English Studies Vol. 2, 2002, ); and (4) the English Drama collection in the Chadwyck-Healey Literature Online database (for LION, see [accessed January 2004]).

Of politeness and people

105

The latter sense is attested with humans from the seventeenth century onwards. It was not until the mid-eighteenth century that it acquired its current senses ‘courteous’, ‘treating others with consideration’, ‘having or displaying good manners’. The noun politeness appears in the sense of ‘polished manners’ and ‘courtesy’ in the early eighteenth century, as in example (5) cited in the OED (politeness n3).3 (5) 1702 Eng. Theophrast. 108 Politeness may be defined a dextrous management of our Words and Actions whereby we make other people have better Opinion of us and them~selves.

2. Politeness in corpora 2.1 Early Modern politeness Electronic corpora largely confirm the dictionary information on polite: no instances of its ‘polite’ sense proper occur in the three Early Modern English corpora we examined. The Corpus of Early English Correspondence (CEEC, 1410–1680) has no instances of either polite or politeness. The noun does not appear in the Early Modern English section of the Helsinki Corpus (HCE, 1500–1710), and there is only one instance of the adjective, cited in (6). It comes from Thomas Elyot’s Gouernour (1531), which sets requirements for the pronunciation of nurses and other women attending on a nobleman’s son. The sense is ‘refined’ or ‘cultivated’. (6) Semblably the nourises and other women aboute hym, if it be possible, to do the same: or, at the leste way, that they speke none englisshe but that which is cleane, polite, perfectly and articulately pronounced, omittinge no lettre or sillable, as folisshe women often times do of a wantonnesse, (HCE, Thomas Elyot, The Boke Named the Gouernour, 1531, pp. 22–3)

Before 1700, the use of polite in the large LION English drama database is limited to the ‘elegant’ and ‘refined’ senses of the adjective. Apart from a single reference to a most neate fine street in 3

These dates are provided by The New Shorter Oxford English Dictionary, edited by Lesley Brown (1993), and the online version of the new edition of the Oxford English Dictionary (http://dictionary.oed.com/).

106

Nevalainen & Tissari

Rome in the literal sense of polite, it is typically applied to nouns such as method or rules of courtship, oratory, conversation, phrase(s), way of expression, garniture and parts in the sense of ‘qualities’. Except the occasional polite companion and (most compleat) gentleman, it is the behaviour of people, and verbal behaviour in particular, rather than the people themselves that are characterized as polite. Philip Massinger’s comedy The Picture (1630) makes polite into an adverb in “She speakes well Polite, and courtly” (III.ii). Polished occasionally takes similar referents including thoughts, speech and poesie, but is mostly used in concrete contexts with reference to materials such as steel, stones and ebony. There are no instances of the adjective courteous in the HCE, but the sense appears in the noun courtesy, which is more frequent than civility. Courtesy and courteous lose ground with time in the CEEC, while civil and civility become more frequent. The early seventeenthcentury illustration in (7) describes the Japanese as “curteous people”. (7) They wear theyre hare longe, bownd upp like the Chines, with a bodkin thrust through, but it is made up on the right side of theyre heades, & are a very gentle & curteous people. (CEEC Supplement, Richard Wickham, 1614, p. 274)

The various modern uses of civil were already common in the Renaissance, but the polite sense was also prominent in all the three databases examined. The HCE has some thirty instances of civil, one third of the “polite” kind, as in (8). As an aside, the case in (9) is a sixteenth-century view of the civil service. (8) in some places beyond Seas, 2500. are taught in one Schoole) without any noise, in a pleasing & profiting manner, & in their playing years; not onely the English, Latine, and Greek Tongues, (together with the Duties of Piety, and civil behaviour) but also the Easterne, and other needful forreign Languages, (HCE, Charles Hoole, A New Discovery of the Old Art of Teaching Schoole, 1660, pp. 218–19) (9) he left soch a companie of fellowes and scholers in S. Iohnes Colledge, as can scarse be found now in some whole vniuersitie: which, either for diuinitie, on the one side or other, or for Ciuill seruice to their Prince and contrie, haue bene, and are yet to this

Of politeness and people

107

day, notable ornaments to this whole Realme: (HCE, Roger Ascham, The Scholemaster, 1570, p. 280)

2.2 Post-Restoration politeness The diversification of the notion of politeness appears from our eighteenth-century material, the Corpus of Early English Correspondence Extension (CEECE, 1680–1800). The “polite” senses of civil still outnumber those of polite in the corpus, but the proportion of polite and its derivatives increases noticeably. Personal correspondence closely reflects the cultural focus on politeness in post-Restoration Britain. “Polite sociability” became the hallmark of the Age of Enlightenment, raising self-consciousness about manners. This polite urbane lifestyle could be taught by education and learnt through practice, as shown by the personal letters of the upwardly mobile Verney family. Susan Whyman’s analysis4 of the forms of social interaction they engaged in includes financial and social favours, giftgiving and patronage, visiting and dining; visiting in London specifically meant social calls in a coach, which was interpreted as a symbol of power. The notion of politeness as refinement and elegance became common with human referents in the eighteenth century. In example (10), Eliza Pierce contrasts polite people and the beau monde with rustics, and Edward Gibbon regrets the degradation of the once polite Royal Court in (11). (10) when I want something pour passer le tems & beleive I shall find more entertainment in them then I even did in Mr. Fieldings Amelia. tho’ you probably and the rest of the polite people will laugh at the Rusticity of my Tast, but how can you expect a better from A Country Girl that has seen nothing of the beau Monde for allmost 7 Year ... (CEECE, Eliza Pierce, 1752, p. 65) (11) Every thing follows the example of the Court which from one of the most polite in Europe is become bigotted, gloomy and covetous. (CEECE, Edward Gibbon, 1764, p. 172)

4 Susan E. Whyman, Sociability and Power in Late-Stuart England: The Cultural Worlds of the Verneys 1660-1720, Oxford: Oxford University Press, 2002.

108

Nevalainen & Tissari

This idea of refined politeness was often satirized in comedies of manners. David Garrick writes about his role in The School for Lovers (12): (12) I have Spoke the Prologue, & don’t appear in ye Play till the beginning of the 2d when Sr John Dorilant (that’s ye Name) makes his Entrance – and a fine, polite, Sentimental, Windling Son of a Bitch it is – a great favourite of ye Ladies … (CEECE, David Garrick, 1762?, p. 354)

Mary Wollstonecraft is more serious in her criticism of false politeness (13): (13) Vanity in one shape or other reigns triumphant – and has banished love in all its modifications – and without it what is society? A false kind of politeness throws a varnish over every character – neither the heart nor sentiments appear in their true colours. (CEECE, Mary Wollstonecraft, 1786?, p. 117)

As suggested by the seventeenth-century collocations of polite listed above, language made up an integral part of the refined sense of politeness. Lawrence Klein notes that “[i]n the ‘polite’ idiom, the standard for language was appropriateness to a certain social practice, the urbane conversation of gentlemen and ladies”5. Language is also included in the definition of politeness dating from the beginning of the eighteenth century in (5), above. A shift in emphasis can, however, be detected in the description of politeness William Pitt gives to his nephew Thomas in 1754: (14) Now as to politeness; many have attempted definitions of it: I believe it is best to be known by description; definition not being able to comprise it. I would however venture to call it, benevolence in trifles, or the preference of others to ourselves in little daily, hourly, occurrence in the commerce of life. A better place, a more commodious seat, priority in being helped at table, 5

Lawrence Klein, “‘Politeness’ as linguistic ideology in late seventeenth- and early eighteenth century England”, in Dieter Stein and Ingrid Tieken-Boon van Ostade (eds.), Towards a Standard English, 1600-1800, Berlin and New York: Mouton de Gruyter, 1994, pp. 31-50.

Of politeness and people

109

&c. what is it, but sacrificing ourselves in such trifles to the convenience and pleasure of others? And this constitutes true politeness. It is a perpetual attention, (by habit it grows easy and natural to us), to the little wants of those we are with, by which we either prevent, or remove them. Bowing, ceremonious, formal compliments, stiff civilities, will never be politeness: (CEECE, William Pitt, 1754, pp. 35–6)

This description downplays purely formal “stiff civilities”, and stresses the idea of positive acts of politeness being performed for others. We would hence like to suggest that this orientation towards others marks the semantic extension of polite to ‘considerate’ and ‘courteous’. In the CEECE data, polite begins to be associated with grammatical complements (prepositional and infinitive constructions) in the second half of the eighteenth century. There are also cases where the beneficiary of the polite action can be deduced from the context. These various uses of the ‘courteous’ sense of polite are illustrated by the letter excerpts (15) to (17). (15) I should have spoken louder, & concluded him to be deaf: but finding him very amiable & very elegant, & very polite to me ; & very unlike an old Man ; I never thought about his being deaf ... (CEECE, Hester Piozzi, 1779, p. 354) (16) I never saw Mr. Ferguson but twice, and know little of him, but as he was polite enough to ask my commands I recommended your Letter to his care. (CEECE, Mary Wortley Montagu, 1758, p. 152) (17) in respect to my last indeed, desiring you to sign your Consent to the Navigation, your fulfilling my Request was the politest answer you could give me. (CEECE, Erasmus Darwin, 1766, p. 38)

As a synonym of polite, civil can similarly take a prepositional complement indicating the beneficiary, as in (18). (18) I already understand almost all that is said and can ask for any common things I want. with regard to other things the people here are extremely civil to strangers and endeavour to make this town as agreable as possible. (CEECE, Edward Gibbon, 1753, p. 2)

We may conclude this brief survey of corpus evidence by noting that polite continued as a polysemous adjective throughout the eighteenth

110

Nevalainen & Tissari

century. It lost its concrete sense ‘polished’, ‘smoothed’, but retained its abstract ‘elegant’ and ‘refined’ sense, which came to share the meaning potential of the word with the more other-oriented senses of civil and courteous in the course of the eighteenth century. But the source domain of politeness remained transparent to writers such as Anthony Ashley Cooper, the 3rd Earl of Shaftesbury, who summed it up by saying: “We polish one another, and rub off our corners and rough sides by a sort of amicable collision” (cited in Porter6). 3. Politeness in the Historical Thesaurus How do the adjectives polite, civil, and courteous and their corresponding nouns relate to other words in the lexical field of politeness? In the following we sketch a model for this field in the vein of Kövecses7, suggesting conceptual metaphors which help to construct the concepts of “good behaviour” and “courtesy”, as represented by data in the Historical Thesaurus of English.8 We provide a prototype for both “good behaviour” and “courtesy”, and a general script for “meeting people”, based on the same data, but more in the vein of Abelson.9 Our sample from the data is restricted to vocabulary dated between 1500 and 1800, including both new lexical material and older expressions still in use. 3.1 Good behaviour: metaphors and prototype Beginning with “good behaviour”, we suggest a central metaphor SOCIETY IS A STRUCTURED COMPOSITION, which implies that EVERY PERSON HAS THEIR OWN PLACE IN IT. Correspondingly, GOOD BEHAVIOUR IS STAYING IN ONE’S PLACE or, to provide more mobility, GOOD BEHAVIOUR IS MAKING THE 6 Roy Porter, Enlightenment: Britain and the Creation of the Modern World, London: Allen Lane, Penguin Books, 2000. 7 Zoltán Kövecses, Metaphors of Anger, Pride and Love: A Lexical Approach to the Structure of Concepts, Amsterdam: John Benjamins, 1986. 8 This data was kindly provided to us by Flora Edmonds on 16 February 2004. The data in the thesaurus was still incomplete in regard to both sections. 9 Robert P. Abelson, “Psychological status of the script concept”, American Psychologist Vol. 36, No. 7, 1981, 715-29. The two prototypes modelled on Kövecses (1986, p. 59, pp. 95–6, pp. 103–4) are also scripts, but of a slightly different kind.

Of politeness and people

111

RIGHT MOVEMENTS. The latter idea can be further illustrated through a dance, such as the seventeenth-century favourite, minuet. One needed some patience to learn the right movements, including the small steps. Since models of behaviour spread from the court (in the case of the minuet, the French court10), it was no wonder that “good behaviour” had such near synonyms as punctuality and exactness. Consider also the phrases comme il faut and just so. Any government of or co-operation among people requires, at least to some extent, the idea that SOCIETY IS A STRUCTURED COMPOSITION. A good example is the Early Modern court, where everyone knew their place. One of the earlier senses of the noun civility, “civil organisation and government” (OED civility n5†), testifies to the close metonymic relationship between politics and “good behaviour”. There was a tension between the medieval belief that everyone was born to their right place and occupation, and the Early Modern opportunities for social aspirers. This tension is apparent in the concept of “good behaviour” as well. While the word manner(s) originally implies “distinct types or kinds” (OED, manner n1), such as the four estates, learning decorum suggested a way of resembling the nobility. Similarly, children could be brought up to behave like their own kind, or to achieve more than their parents. Such near synonyms of “good behaviour” as (good) breeding reflect the metonymy CAUSE FOR EFFECT.11 Metonymy is also at work if one’s poise or good looks are interpreted as signs of wealth, and consequently of nobility and good behaviour. In terms of conceptual metaphors, one might say that THE BODY IS THE INSTRUMENT OF GOOD BEHAVIOUR (which accords very well with dancing), and that GOOD BEHAVIOUR IS GOOD LOOKS. Consider such near synonyms for “good behaviour” as good abearance, carriage, comeliness, and handsomeness, or the phrase set/put a good/better face (up)on. The adjective polite and the noun politeness fall most naturally into this set of vocabulary, having their origins in shiny, smooth, good-looking objects. 10

See StreetSwing.com/Dance History Archives, 1999, accessed 30 June 2004, at http://www.streetswing.com/histmain/z3minuet.htm. 11 Günter Radden and Zoltán Kövecses, “Towards a theory of metonymy”, in KlausUwe Panther and Günter Radden (eds.), Metonymy in Language and Thought, Amsterdam/Philadelphia: John Benjamins, 1999, 38.

112

Nevalainen & Tissari

The first sense of the adjective proper in the OED is “Belonging to oneself or itself; (one’s or its) own; owned as property; that is the, or a, property or quality of the thing itself, intrinsic, inherent.” When proper is applied to behaviour, it seems to confirm that GOOD BEHAVIOUR IS STAYING IN ONE’S PLACE. It even seems to suggest that “good behaviour” may exist in someone as an innate, “permanent attribute or quality” (OED inherent a3), especially if they do not let themselves fall below their estate or attempt to rise above it. However, there is another tension between claims about people’s innate faculty of goodness (and good behaviour) and the necessity to regulate people’s actions evidenced by such words as civility (one cannot help thinking about Rousseau in this context, although many English names are relevant as well). This tension can be seen either in terms of humankind’s possessing or not possessing a free will (Erasmus, Luther), or in terms of form as against matter – consider the OED definition of ceremony (n1), another near synonym for “good behaviour” at the time (italics added): “An outward rite or observance … the performance of some solemn act according to prescribed form …” The metaphors THE BODY IS AN INSTRUMENT FOR GOOD BEHAVIOUR and GOOD BEHAVIOUR IS GOOD LOOKS allow for a difference between a person’s “real self” and the outward manifestations of the self. We suggest the following prototype for “good behaviour”:12 1. A person is born, or raised through either education or his/her own achievement (including marriage), to a certain place in society. This person knows what place s/he occupies in relation to other people. 2. S/he behaves and dresses according to his/her position. It looks good (everything is in order) and, in this way, s/he helps to construct a good society (an orderly composition). Caveat: People would often like to occupy a more important place in society.

12

We have freely adapted Kövecses’s idea of prototypes for emotion concepts (1986, p. 59, pp. 95–6, pp. 103–4), omitting physiological effects and adding the caveat. The same applies to the prototype of “courtesy” offered later in this study.

Of politeness and people

113

3.2. Courtesy: metaphors and prototype Let us then move on to “courtesy”, paying special attention to distinguishing it from “good behaviour”. The interaction between these two concepts should be self-evident from the beginning, among other things because many ideas concerning “good behaviour” spread from the place that inspired the word “courtesy”. Our basic answer to the question of what the concept of “courtesy” adds to the model above is nevertheless that “courtesy” is something extra. Although this something may be a virtue, it also serves the caveat in the prototype for “good behaviour”. “Courtesy” is thus more on the value scale than “good behaviour” but it can also reach a critical point when it becomes too much (overcivility). The word civility is included in both sections of the Thesaurus, but in the context of “courtesy”, we might emphasize that it is not only associated with good order, but also with “freedom from barbarity”, “culture”, and “refinement” (OED civility n10–11). It is possible to interpret the near synonym of “courtesy”, humanity, in similar terms. The adjective urbane in its turn draws a line between cultural refinement and uncivilized vulgarity. Language is an important element of the cultural refinement inherent in “courtesy”. A near synonym of “good behaviour”, honesty, suggests that “good behaviour” requires that one speak the truth. “Courtesy” makes one speak well (fair-spoken, soft-spoken). In terms of conceptual metaphors, we suggest that whereas GOOD BEHAVIOUR IS KNOWING ONE’S PLACE, COURTESY IS MAKING OTHER PEOPLE MORE COMFORTABLE IN THEIR PLACES. A quote from Hobbes’s Leviathan touches the meeting point of the two concepts (OED complaisance n): “Compleasance: that is to say, That every man strive to accommodate himselfe to the rest.” One cannot know one’s own place without relating to other people. However, it is under “courtesy” that we find such expressions as management and accommodation for “conciliatory behaviour”, and moreover, agreeability, pleasance, douceur, plausibility, geniality, and affability. The OED defines comity as “kindly and considerate behaviour towards others” (comity n1). That such behaviour is associated with nobility is evidenced by the older nouns gentrice and hendeness/hendiness (also see MED entries for hend*). One’s high position allows one to stoop without losing one’s dignity and

114

Nevalainen & Tissari

authority. The nouns easiness and ease underline the comfort provided by “courtesy”, while a person who makes other people comfortable is winsome and clever. A special case of “courtesy” which deserves to be mentioned here is gallantry, making women feel good about themselves. A more minor metaphorical aspect of the data could be phrased in the following way: while GOOD BEHAVIOUR IS THE NATIVE HABITAT, COURTESY IS THE CLIMATE. An example is make fair weather. We suggest the following prototype for “courtesy”: 1. A person A knows another person or certain people’s B (,C…) social status in relation to himself/herself. Motivated by either love or ambition, A wants to make B (,C…) feel more comfortable in their place(s) and, optionally, respond to A’s needs. 2. Consequently, A speaks very nicely to B (,C…) and accommodates himself/herself to B (,C…) more than required by the norms of good behaviour. Caveat: devaluation of courtesy: If courtesy is the norm, courtesy becomes (a form of) good behaviour. 3.3 A script for “meeting people” The Thesaurus data also allows us to offer a very general script for “meeting people”. The relevance of this script is that one needs to be familiar with it in order to behave in a polite, civil, or courteous manner. Abelson (op. cit., p. 715) defines “script” as “one type of schema”: “Scripts embody most of the conceptual issues raised by other types of schemata, but are simple and well-structured enough to permit more focused analysis and experimentation.” The following script, “meeting people”, is divided into three major stages. In the first stage, one acknowledges the presence of the other. In the second, one exchanges information, complements, and/or services, while in the third, one signals the approaching departure. The terms “equifinal” and “free behaviours” are borrowed from Abelson. The term “equifinal” “indicates that several different actions may accomplish the same result”, whereas “free behaviours” are “those activities that may plausibly and commonly intermix with the ongoing

Of politeness and people

115

script” (op. cit., pp. 723–24). In a strict interpretation, free behaviours probably cannot form a stage in a script, but as regards the one below, we also have to take into account Abelson’s own remark that it is “probably naive” to “expect … a linear … process” (op. cit., p. 718). In other words, people meeting each other might also omit any stage in the script or change their order. Here is the script for “meeting people”: 1. Equifinal (/complementary) actions: greeting with words, by slight movement of cap, nod, bow, or handshake, welcoming. Addressing each other in “Courteous forms of address/title”. 2. Free behaviours (/optional?): remembrances/greetings sent, introducing someone else, (exchanging) compliments, kisses, (exchanging) courteous actions, services, information. 3. Equifinal (/complementary) actions: good wishes, leave-taking / bidding farewell. We have added the explanatory term “complementary” to suggest that the actions listed often go together rather than alone, and the term/question mark “optional?” to suggest that “good behaviour” or at least “courtesy” may require some of the actions listed as “free behaviours”. English has both words to be used in and words for (describing) the events and actions in the script.13 The former include how-d’ye do and au revoir, not to mention such courteous titles as signor, sir, and ma’am. The latter include reversion “return of a courtesy”, and wring, shrug, and hand-gripings for “hand-shake”. Polite, civil or courteous behaviour in Early Modern England and especially the eighteenth century required that one master not only this general script but also more specific scripts for all the important and even further situations. In a more focused analysis, which would require more data from the period than is possible to treat here, one could, for example, divide the more specific situations according to time (time of day, season, a person’s age), place (street, home, court,

13 We do not attempt to distinguish between “events” and “actions”. As can be seen in Abelson’s table 1 (op. cit., p. 716), this may get quite complicated.

116

Nevalainen & Tissari

club, theatre etc), and even form (letter-writing as against meeting in person).14 4. Conclusion The corpus data on polite(ness) discussed in this article agreed with historical dictionaries in first-dating reference to humans in the seventeenth century. Corpora also provided more detailed information on the semantic diversification of polite. In the Corpus of Early English Correspondence Extension (CEECE, 1680–1800), the adjective polite gained ground at the cost of the “polite” senses of civil, showing that the notion of politeness as refinement and elegance in people and their behaviour predominated in the eighteenth century. The syntactic patterning of polite in the corpus data however also suggested that, from the mid-eighteenth century onwards, politeness was increasingly motivated by consideration for others. Returning to our examples (13) and (14), it is now possible to note a correspondence between the prototypes for “good behaviour” and “courtesy” and the corpus data: people tend to differentiate between outward forms of good behaviour (“bowing, ceremonious, formal compliments, stiff civilities”, “false kind of politeness”) and the sources of “true politeness” (“the heart”, “sentiments”, “sacrificing ourselves”). Our study suggests that the Historical Thesaurus of English could be used to model many other concepts as well, and even behavioural patterns, such as “meeting people”, while the scripts presented above could be checked against more data, such as contemporary conduct books, novels and diaries.

14 For conventions of letter writing, see Terttu Nevalainen, “Continental conventions in early English correspondence”, in Hans-Jürgen Diller and Manfred Görlach (eds.), Towards A history of English as a History of Genres, Heidelberg: C. Winter, 2001, pp. 203-224.

ME douten and dreden Michiko Ogura The Oxford English Dictionary cites two major senses of doubt (v.), i.e. ‘to doubt’ and ‘to fear’. For the latter sense, part of the etymological explanation reads: “Branch II ‘to fear, to be in fear’, a development of the verb in OF, was an early and very prominent sense of the vb. and its derivatives in ME”.1 We find the same sense in redoubt (v.) and †adoubt (v.). Dread (v.), on the other hand, shows a sense ‘to doubt’, though obsolete and rare.2 The noun dread also has a sense ‘doubt, risk of the thing proving otherwise’.3 Phrasal expressions are also ambiguous: e.g. withouten dout ‘undoubtedly’ or ‘without fear’; without dread means ‘without doubt, doubtless’ but no dread ‘no fear’ or ‘no doubt’. This paper aims at investigating the semantic and syntactic overlapping of a native verb dreden and a loan verb douten, together with their prefixed cognates and their synonyms and Old English counterparts. OE ondrædan is the most ordinary verb denoting ‘to fear’, often occurring with a coreferential dative pronoun. In the Gospels it translates Latin timere, though Ru1 chooses forhtigan,4 e.g.: (1) Mt 10.31 [nolite ergo timere multis passeribus meliores estis uos] Li: nellaÇ ge forÇon ondréde of monigum Çrowungum Çy betro 7 Çy sellra gebiÇon iuh Ru1: ne forþon forhtigaþ mongum ge sindun bettra þonne þas spearwas

1

OED, s.v. doubt v., II. 5 OED, s.v. dread v., 2 †c. 3 OED, s.v. dread sb., †3 4 Editions used for this survey are listed with abbreviated titles in the references. For Old English poetry see under Krapp and Dobbie (eds.), ASPR. Examples of those works with editions not given in the references are cited from OED, MED and MC. 2

118

Michiko Ogura

WSCp: Ne ondræde ge ge synt selran þonne manega spearuan WycEV: Therfore nyle e drede; e ben better than many sparwis. (AV: Feare yee not therefore, ye are of more value then many Sparrowes.) When used as renderings of other Latin verbs of fear, other words are chosen from among the Old English synonyms. In the next example an infinitive timere and two passive infinitives conturbati and conterriti are rendered by different lexemes, i.e.: (2)

Lk 24.36-37 [pax uobis ego sum nolite timere conturbati uero et conterriti existimabant sé spiritum uidere] Li: sibb iuh ic am nallað ondrede efne-gestyredo woeron uutedlice 7 gefyrhtedo weron wóendon á hine gást ote hia gesego Ru2: sib iowih mið ic am nallað geondreda efne-gistyrede werun wutudlice 7 fyrhtede werun woendun hine gast o gisege WSCp: sib sy eow ic hit eom ne on-dræde ge eow Ða wǙron hig gedrefede 7 afærede 7 hig wéndon o hig gast gesawon WycEV: Pees to ou; I am, nyle e drede. Sothli thei troublid and agast, gessiden hem to se a spirit. WycLV: Pees to ou; Y am, nyle e drede. But thei weren affraied and agast, and gessiden hem to se a spirit. (AV: Peace be vnto you. But they were terrified, and afrighted, and supposed that they had seene a spirit.)

In the Psalter we find a similar choice in rendering timere, as in (3) where PsG1E has ondrædan with a coreferential pronoun, and in (4) where a noun timor is translated ege.

ME douten and dreden

(3)

119

Ps 48.17 [Ne timueris cum diues factus fuerit homo. et cum multiplicata fuerit. gloria domus eius] A: ne ondred ðu ðonne weolig geworðen bið mon 7 ðonne gemonigfaldad bið wuldur huses his D: ne andræd þu þonne weli eworden bið mann þonne emænifyld bið wuldur huses. E: Ne ondred þu þe þonne welyg geworden bieð mon 7 þonne gemonifalded bið wuldor hus his. WycEV: Ne thou shalt dreden, whan riche a man shal be maad; and whan shal be multiplied the glorie of his hous. WycLV: Drede thou not, whanne a man is maad riche; and the glorie of his hows is multiplied. (AV: Be not thou afraid when one is made rich, when the glory of his house is increased.)

(4)

Ps 118.120 [Infige timore tuo carnes meas. a iudiciis enim tuis timuí] B: efæstna mid ðinum ee mine flæsc from domum soðlice þinum ic ondred D: onfæstna ee þinum flæsc mine fram domum soþlice þinum ic adræd F: onfæstna ege þinum flæsc mine fram dome soðlice ðinum ic adræde WycEV: Pricke with thi drede my flesh; forsothe fro thi domes I dradde. WycLV: Naile thou my fleischis with thi drede; for Y dredde of thi domes. (AV: My flesh trembleth for feare of thee: and I am afraide of thy Iudgements.)

Ondrædan frequently occurs with a coreferential pronoun and often in a negative construction. Typical are such poetic formulas as

Michiko Ogura

120

Jul 210 Ne ondræde ic me domas þine ‘I do not fear thy judgments’5 and GenA 1037 Ne þearft ðu þe ondrædan deaðes brogan ‘Thou needest not fear the terror of death’.6 Forhtian and afæran, on the other hand, are often used in the past participle, as in And 1339-40 wurdon hie ða acle on þam ofenge, forhte, afærde, ond on fleam numen ‘then they became dismayed by the defence, terrified, afraid, and put to flight’7 and JDayII 125-6 stent hergea mæst heortleas and earh, amasod and amarod, mihtleas, æfæred ‘the greatest of troops will stand disheartened and timid, confused and confounded, powerless and afraid’.8 No conflict or rivalry seems to be found among these synonyms in the Old English period. Another prefixed synonym ofdrædan is used distinctively, often in the past participle, showing no obvious conflict with ondædan: e.g. CP(H) 35 239.6-7 Ĉæt is ðonne ðæt hie eallneg ræswað & ondrædað ðæt hi mon tælan wille, & beoð eallneg mid ðæm ymbeðoncan abisgode & ofdrædde ‘That is, then, that they always suspect and fear that one should intend to blame them, and are always occupied with the thought and afraid’. OE tweo(ga)n is used with or without ge-, showing no dialectal conflict with other verbs, as seen in the Gospel, i.e.: (5)

Mt 28.17 [quidam autem dubitauerunt] Li: sume ðonne getwiedon Ru1: sume þonne tweodun WSCp: Witodlice sume hig tweonedon WycEV: sothli summe of hem doutiden (AV: but some doubted)

Boethius, Gregory’s Dialogues and some homilies use this verb rather often, especially in negative constructions. Here are some examples:

5

Also in Jul 134 Næfre ic me ondræde domas þine, but cf. JDayII 15 Ic ondræde me eac dom þone miclan. 6 Also in ChristB 779 Ne þearf him ondrædan deofla strælas. 7 Also in ChristC 892a forhte afærde. 8 The translation is quoted from Graham D. Caie (ed.), The Old English Poem Judgement Day II, Cambridge: D. S. Brewer, 2000, p. 91.

ME douten and dreden

121

(6)

Bo 16.38.3, 5 Hwæt, we gen[og georne] witon ðæt nanne mon þæs ne tweoð o se sie strong on his maegene þe mon gesihð o stronglic weorc wyrcð; ne þon ma, gif he hwæt bið, ne tweoð nænne mon o he hwæt ne sie. ‘Lo, we understand very well that no one will doubt this, that he is strong in his might who is seen to do a strong deed; no more than that, if he is able, no one will doubt that he is able.’

(7)

Bo 34.93.25 Ne þearft ðu no tweogan [ymbe o] þe þu ær tweodest, o is be [Ö¸ ge]sceaftum þe nane sawle n[abbað]; ‘Thou needest not doubt about what thou had doubted, that is about these creatures which have no soul.’

(8)

Bo 37.113.21 Forðæm ne þearf nænne wisne mon tweogan o ða yflan næbben eac ecu edlean hiora yfles; o bið ece wite. ‘Therefore no wise men need to doubt that the evil also have the eternal reward of their evil, that is, eternal punishment.’

(9)

HomS14 (BlHom4) 58 Forþon ne þearf þæs nanne tweogean, þæt seo forlætene cyrice ne hycgge ymb þa þe on hire neawiste lifgeaþ. ‘Therefore no one needs to doubt this, that the forsaken church will not think of those who live in the neighbourhood.’

Michiko Ogura

122 (10)

HomU11 (VercHom7) 72 Nu sio idelnes swa swiðe þam lichoman dereð, ne tweoge þe na þæt hio þære sawle ne sceððe, ‘When the idleness will harm the body so greatly, do not doubt that it will not injure the soul’

(11)

HomU3 (Belf12) 94 Ne þearf us na tweogean þæt he us næle eft þare læna munegiæn þæs þe he us her on weorlde to forlǙt. ‘We need not doubt that he will remind us again of the loans of what he allowed us here in this world.’

As seen in the examples, tweo(ga)n often takes a þæt-clause, as well as the genitive þæs and a coreferential pronoun þe in the imperative construction. Still the negative particle ne in (6) (in the o-clause following the second tweoð), næbben in (8) and næle in (11) are not necessary. A verb of negative import often takes a þæt-clause which contains an expletive ne, as explained by Mitchell (OES, § 2044): e.g. ÆHom 9.182 and him swiðe forbudon þæt hi nan ðing ne bodedon be þam Hælende nahwær.9 ME tweonen occurs in the same syntactic environment of negative constructions, and at times an ‘expletive’ negative appears in the subordinate clause. (12)

9

Vesp.D.Hom.Fest.Virg. 137.27 (Warner) Ne tweonige nane mæn, o seo gedrefde moder nolde beon gescild wið þære (ge)drefednysse

Bo 37.113.21 is quoted in Mitchell (OES, § 2039) as an example of the ‘expletive’ negative. The noun tweo also occurs in a similar context as he quotes: e.g., GD 213.20 7 ne Öuhte nanum men Öæs tweo, Öæt gif Öæt stanclif feolle, Öæt hit ne ofsloge Öæt scræt.... He asks whether “the idiomatic piling up of negatives” can be a factor of the occurrence of the ‘expletive’ negative. I have no doubt that the excessive use of negatives may trigger the occurrence of an ‘expletive’ ne, but we should also be mindful of the fact that this happens at times only with nouns or verbs of the negative import in the subordinate clause following and appears only too unpredictably.

ME douten and dreden

123

‘No one doubts that the troubled mother would be protected from the trouble’ (13)

Bod.Hom.Dom.2 Quadr. 52.21 (Belfour) ne sceal he him tweonigæn ðæt he ne mage Godes mildheortnesse bigitæn ‘he shall not doubt that he can obtain God’s mercy’

A coreferential pronoun occurs in (13), which was not an Old English feature. Another new trend of the verb is an ‘impersonal’ construction, which is also found in Vespasian D.XIV and Bodley 343 homilies. (14)

Vesp.D.Hom. 76.31 ac us tweoneð hweðer ge mugen mare deopnysse þær on þearlice tocnawen ‘but we doubt whether ye can recognize much profundity therein strictly’

(15)

Bod.Hom.Evang. 16.2 7 gyf him na ne tweonæÖ o he þæs tyðe béo, ac ilyfð on heortan, swá hwæt swa he cwæð hit bicymeð 7 iwurð ‘and if he has no doubt that he is granted of this but believes in his heart, whatever he says will come true’

Tweogan means ‘to doubt’ and ondredan ‘to fear’, and the two verbs show no conflict in sense.10 But one may fear because one doubts. Doubt bears fear, hence Satan’s speech: (16)

10

Vesp.D.Hom.Nicod. 84.12 Sathanas þa, þære hellen ealder, andswerede 7 cwæð, “Hwæt tweonest þu of þe? Hwæt ondrædest þu þe þone Hælend to onfone, minne wiðerwinne?”

In TOE, ondrædan appears mostly under 8. Emotion and tweo(ga)n under 6. Mental Faculties.

Michiko Ogura

124

‘Then Satan, the prince of hell, answered and said, “Why doubtest thou about thyself? Why fearest thou for taking hold of the Lord, my adversary?”‘ Ondrædan, ofdrædan and adrædan do not simply give way to the non-prefixed dreden. Through the transitional period to early Middle English, ME adreden takes two syntactic environments, i.e., constructions with a coreferential pronoun (e.g. PMor(Lamb) 6 ful sare ic me adrede) and the ‘be + past participle’ (e.g., Lag Brut(C) 27962 for heo weoren adradde). ME dreden shares the first feature (e.g. Orm 151 Ne dred te, Zacarige, nohht, Noff me, noff mine wordess!), but the second feature is shared with ofdreden, which is found in the variant reading of the following context: PMor(Jes-O) 94 Hwat schulle seggen oþer don, þer engles heom drede [vr. beð ofdrade]. In the course of Middle English, the non-prefixed dreden keeps taking a coreferential pronoun, especially in verse (e.g., Cursor (C) 3665 I dred me sare and Gen & Ex 767 He dredde him to leten is lif), but at the same time comes to be used as an alternative of or in parallel with be(n) dredful or be(n) afered (e.g. Wycl.Lantern 16/6 I quake, I drede, & vgli I am aferde). Adjectival forms and nouns can also be found as manuscript variants, as in the following: (17)

Cursor 17665 C: And war we for þe dredand sare G: And war we for þe dred-ful sare F: And for the dred we alle in care

ME aferen (< OE af¼ran) is used in two major constructions, i.e., with direct object (e.g., Lag Brut (C) 25553 feorlic wes þat sweouen; þene king hit auerde ‘marvellous was the dream; it frightened the king’) and in the past participle (e.g., Lag Brut (C) 26084 and neoðeles næs ic nauere; of Ardure afæred sære ‘and nevertheless I was never afraid of Arthur sorely’). ME affraien (
ME douten and dreden

125

similarities make the two verbs used in similar contexts by the same author, e.g., (18)

Chaucer CT.Kn.A. 1518 Soore afered of his deeth was he Chaucer CT.NP.B. 4475 Be ye affrayed of me that am youre freend?

(19)

Gower CA 4.600 Were thou afered of hire yhe? Gower CA 4.3400 So that withinne his herte affraied, ... he wissheth after deth.

ME douten (
11

Cursor 3000 C: Parfai for i me vm-bithoght Yee war men þat godd duted noght;

For editions of the texts I have used see the references.

Michiko Ogura

126

G: Parfay for i me vmbi-thoght, þat goddes au ge douted noght. F: parfay for I. me wele be-þogt ge ar men doutes god rigt nogt. T: Sir he seide I me biþougt þat goddes awe dredde ge nougt where a later MS. T has dreden with a slight alteration of the context. Douten is often used in the sense ‘to fear or be afraid’, especially when co-occurring with dreden or the noun drede: e.g., þe grace of ihu 24 þe children wiþ in þe moder wome Wel sore sul dute and drede þer for and Chaucer CT.Mel.B. 2517: I sey nat thow shalt be so coward, that thow doute ther wher as is no drede. Douten means ‘to fear’ as a rendering of L timere, especially in the midfourteenth century MPPsalter. e.g., 3.6: Ich ne schal nougt doute [L. timebo] þousaundes of folk þat bysetten me and 144.20 He shal do þe wil þe doutand hym [L. timentium se]. MED cites two obvious examples of douten and dreden as manuscript variants: (21)

SLeg.Becket (Hrl) p. 73 Ac ich douti [Ld: drede], for mi wrecche gult, that wors schal beo the ende

(22)

PP1.C (Hnt) 21.314 That is soþ .. bote ich me sore doute [B: drede]

Because both douten and dreden take a subordinating Öat-clause or occur with a coreferential pronoun, it is quite likely that they may share some syntactic environments. When one’s knowledge is limited, one may doubt and then fear, and so the two verbs can be synonymous and at times can be alternative. Dreden is, therefore, used in the sense ‘to be in doubt’: e.g., MPPsalter 76.15 þe wicked ben trubled, dredand wheþer Öou be God oÖer non. (See the next paragraph for manuscript variants.) Dredeles, as an adverb, may mean either ‘fearlessly’ (e.g., Chaucer Bo.3.m.l2.11: The poete of Trace,

ME douten and dreden

127

Orpheus, ... hadde maked the hertes and the hyndes to joynen dreedles here sydes to cruel lyouns for to herknen his song) or ‘without doubt’ (e.g., Chaucer TC 1.1048: And dredelees, if that my lyf may laste, ... som of hem shal smerte). As has been seen, douten and dreden appear as manuscript variants. The EETS edition of The Earliest Complete English Prose Psalter, based on BM. Additional MS. 17376 and Trinity College, Dublin A.4.4, shows a remarkable contrast in the lexical choice Table 1 shows the number of some words of emotion which occurred as manuscript variants. Doute(n) (noun and verb forms) in Add. corresponds to drede(n) in Dublin twelve times, while the reverse occurs only once. Examples are: (23)

Ps 111.1 Blisced be þe man þat douteÖ [D: dredeÖ] our Lord; he shal wil greteliche in his comaundementg.

(24)

Ps 30.23 Ha Lord, ful michel his þe multitude of þy swetnes, þatou hidest to þe doutand [D: men dredyng] þe.

(25)

Ps 76.15 Ha God, men segen þe, and dreden; and þe wicked ben trubled, dredand wheþer þou be God oþer non [D: douted wheþer þou wer Godd or nogt and þe w. beþ sturbled].

Michiko Ogura

128

Table 1: MS Variants in Add. / Dublin12 drede(n) doute(n) doute(n) dreden glorie(n) ioye(n) gladen mild(nes) welelikand pite reuful ire

/ / / / / / / / / / / /

drede(n) doute(n) drede(n) douten ioye(n) ioye(n) ioyen meke(nes) (wel)plesyng(es) mercy peteful wraÖ

50 20 12 1 38 32 5 10 4 2 2 2

ME douten and dreden showed rivalry during a very limited time and space. To sum up, they became synonymous because (i) they were used in similar syntactic structures, (ii) in religious contexts they shared the senses of fear and doubt which come from lack of knowledge, (iii) they could be renderings of L timere, (iv) they both started with d- and were phonetically suitable to be used in pairs or as alternatives, and (v) douten was derived from an Old French verb of twofold meanings. Some manuscripts made the most of these features or some authors used them alternatively, but this rivalry did not last as long as that between afered and affraied.

Editions and Abbreviations: Belfour, A. 0. (ed.) 1909, Twelfth-Century Homilies in MS. Bodley 343. EETS, o.s. 137. rpt. London, 1962. [Bod.Hom.]

12

Other correspondences are: blisful/ mekefull (Ps 68.20), de-boner / meke (Ps 85.4), deliten and gladen / glade and ioie (Ps 66.4), wele-quemand / plesyng (Ps 88.17), wraþen / ben wroþ (Ps4.5), wreke / auengyng (Ps 98.9), etc.

ME douten and dreden

129

Benson, L. D. (ed.) 1987, The Riverside Chaucer. 3rd ed. Oxford: Oxford University Press. [Chaucer Bo, CT, TC] Brenner, E. (ed.) 1908, Der altenglische Junius-Psalter. Anglistische Forschungen 23. Heidelberg: Carl Winter. [PsGlB] Bülbring, K. D. (ed.) 1891, The Earliest English Prose Psalter. EETS, o.s. 97. rpt. New York: Kraus Reprint, 1987. Forshall, J. and F. Madden (eds.) 1850,The Holy Bible, containing The Old and New Testaments ... by John Wycliffe and his Followers. Oxford: Oxford University Press. [WycEV, WycLV] Harsley, F. (ed.) 1889, Eadwine’s Canterbury Psalter. EETS, o.s. 92. London. [PsG1E] Hecht, Hans (ed.) 1900, Bischof Waerferths von Worcester Übersetzung der Dialoge Gregors des Grossen. Bib. ags. Prosa 5. Leipzig: Georg H. Wigand. [GD] Holt, R. (ed.) 1878, The Ormulum, with the Notes and Glossary of R. M. White. 2 vols. Oxford: Clarendon Press. [Orm] The Holy Bible, 1611 Edition, The First Edition of the Authorized Version, rpt. Nashville: Thomas Nelson Publishers, 1993. [AV] Kimmens, A. C. (ed.) 1979, The Stowe Psalter. Toronto Old English Series 3.Toronto: University of Toronto Press. [PsG1F] Krapp, G. P. and E. V. K. Dobbie (eds.) 1931-53, The Anglo-Saxon Poetic Records I-VI. London: Routledge and Kegan Paul, New York: Columbia University Press. [ASPR, GenA, And, Jul, ChristB, ChristC, JDayII] Kuhn, S. M. (ed.) 1965, The Vespasian Psalter. Ann Arbor: University of Michigan Press. [PsG1A] Kurath, Hans, Sherman Kuhn, et al. 1956-2000, A Middle English Dictionary. Ann Arbor: University of Michigan Press. [MED] Macaulay, G. C. (ed.) 1900, The English Works of John Gower. 2 vols. EETS, e.s.81, 82. London: Oxford University Press. [Gower CA] Madden. Sir F. (ed.) 1847, Lagamon’s Brut, or Chronicle of Britain. 3 vols. rpt. New York: AMS, 1970. [LaƧ] Mitchell, Bruce. 1985, Old English Syntax. 2 vols. Oxford: Clarendon Press. [OES] Morris, Richard (ed.) 1865, The Story of Genesis and Exodus. EETS, o.s. 7. rpt. New York: Greenwood Press, 1969. [Gen & Ex]

130

Michiko Ogura

——. 1874, Cursor Mundi. EETS, o.s. 57, 59, 62, 66, 68, 99, 101. rpt. London, 1961.[Cursor] ——. 1874-80, The Blickling Homilies of the Tenth Century. EETS, o.s. 58, 63, 73. rpt. London, 1967. [BlHom] Pope, J.C. (ed.) 1967, Homilies of Ælfric. EETS, o.s. 259, 260. London. [ÆHom] Reid, T. B. W. (introd., notes and glossary to Wendelin Foerster’s critical edition) 1942. Yvain. Manchester: Manchester University Press. Roberts, Jane, and Christian Kay with Lynne Grundy (comp.) 1995, A Thesaurus of Old English. 2 vols. King’s College London Medieval Studies XI. [TOE] Roeder, F. (ed.) 1904. Der altenglische Regius-Psalter, rpt. Tübingen: Max Niemeyer, 1973.[PsG1D] Scragg, D. G. (ed.) 1992, The Vercelli Homilies and Related Texts. EETS, o.s. 300. Oxford: Oxford University Press. [Verc.Hom.] Sedgefield, W. J. (ed.) 1899, King Alfred’s Old English Version of Boethius. rpt. Darmstadt: Wissenschaftliche Buchgesellschaft, 1968. [Bo] Simpson, J. A. and E. S. C. Weiner. (eds.) 1989. The Oxford English Dictionary. 2nd ed. Oxford: Clarendon Press. [OED] Skeat, W. W. (ed.) 1887, 1871, 1874, 1878, The Gospel according to St. Matthew, St. Mark, St. Luke and St. John. rpt. Darmstadt: Wissenschaftliche Buchgesellschaft, 1970. [Li, Ru1, Ru2, WSCp; Mt, Mk, Lk, Jn] Sweet, Henry (ed.) 1871-72. King Alfred’s West-Saxon Version of Gregory’s Pastoral Care. EETS, o.s. 45, 50. London. [CP2] Venezky, R. L. and A. diPaolo Healey (eds.) 1980, A Microfiche Concordance to Old English. Newark: University of Delaware. [MC] Warner, Rubie D.-N. (ed.) 1917, Early English Homilies from the Twelfth Century MS. Vesp.D.XIV, EETS, o.s. 152. rpt. New York, 1971. [ Vesp.D.Hom.] Whitehead, F. (ed.) 1980, La Chanson de Roland, Oxford: Blackwell. [Roland]

What did Anglo-Saxon seals seal when? Jane Roberts The Old English word insegel ties historians in knots, which is hardly surprising for the concept SEAL, like DOOR, WINDOW, or LOCK, is inherently complex.1 In this note I should like to begin by looking at what meanings the word holds in Anglo-Saxon historical and legal writings before widening my sights to other examples of its use. I begin with the historical writings because it is in them that I should expect to find insegel in its prototypical meaning. In origin its stem is cognate with Latin sigillum,2 which, Chaplais points out, “in its normal usage could only mean either the matrix of a seal (in the shape of a signet-ring or of any type of seal-die) or the impression made from it on wax, lead or some other soft material”, though with the caveat that it “is also sometimes used as a synonym of signum, the cross of a subscribing witness in a charter”.3 What I find puzzling is how discussion of insegel has to a large extent focused on AngloSaxon historical and legal writings in Old English without much by way of reference to the word insegel in other contexts. Refreshingly, Heslop does draw attention to the more general use of seals, “[t]here are indications that seals were used for closing boxes, and even for securing the bandages used after an ordeal by hot iron”, but he adds “it 1

Christian and I have often mulled over the difficulties presented by just what such words mean, and I hope therefore to amuse her with my attempts to get to grips with the meanings held by inseil in its brief history within the English language. 2 No late or medieval *insigillum is to be found. The stem represents sigill-um ‘little sign, figure, or token, seal’, a diminutive of signum. Cf. insigne ‘mark, sign’, also in medL. as ‘seal’. The medL. verb insigillare does exist (OED “though sometimes as a rendering of OE inseglian”); and OF enseel is common (>enseal). Also OHG insigilen and ON innsigla. Noun cognates: OFris insigel, -il (MDu inzegel), OHG insigili (MHG insigele, gil, Ger insiegel), ON innsigli (Da indsegl). 3 Pierre Chaplais, “The origin and authenticity of the royal Anglo-Saxon diploma”, in Felicity Ranger (ed.), Prisca Munimenta. Studies in Archival & Administrative History Presented to Dr A. E. J. Hollaender, London: University of London Press, 1973, pp. 28-42, p. 51 and n. 62 [repr. from Journal of the Society of Archivists 3, 2, October 1965]. For signum in this sense he points to the Old English hondseten.

132

Jane Roberts

is hard to imagine that the latter were more than occasional supplementary functions of a class of object made primarily for closing letters, whether formal or informal”.4 For the most part the historians seem transfixed by the “normal usage” of an object, a seal matrix. That is entirely understandable, for if they looked up the word inseil, they would find that according to the Oxford English Dictionary (OED) it was in use from a.1000-a.1225 with the meanings ‘A seal; app. orig. the impression made in wax with which a letter, etc. is sealed; also the signet or engraved instrument with which the impression is made.’ The OED, normally so reliable a guide, stays firmly with concrete senses (as does the Anglo-Saxon Dictionary of Bosworth and Toller5), and does not volunteer any ‘also fig.’ note to point to some more general meaning such as ‘sign’.6 There is agreement that some Anglo-Saxon seals remain in existence even from before Edward the Confessor’s introduction of the patent writ. The earliest matrix extant is the 860s ring of Æthilwald, an East Anglian bishop. Two other early seal-matrices are Edith of Wilton’s ring (975 x 984), and Ælfric’s seal (c.980, if the Hampshire ealdorman), which could have been worn as a pin or brooch.7 Heslop, both in his 1980 article and in his entry on seals in the Blackwell Encyclopaedia, gives an excellent overview of these and other Anglo-Saxon seals.8 However, quite what purposes such seals were put to in Anglo-Saxon England is hotly debated. Thus, Alfred’s 4

T. A. Heslop, “Seals”, in Michael Lapidge, John Blair, Simon Keynes and Donald Scragg (eds.), The Blackwell Encyclopaedia of Anglo-Saxon England, Oxford: Blackwell, 1999, pp. 413-14. 5 T. Northcote Toller (ed.), An Anglo-Saxon Dictionary Based on the Manuscript Collections of the Late Joseph Bosworth, Oxford: Oxford University Press, 1898, Supplement, 1921, Revised and Enlarged Addenda by Alistair Campbell, 1972. 6 James A. H. Murray, Henry Bradley, Sir William A. Craigie and Charles T. Onions (eds.), The Oxford English Dictionary, 1884-1933; Robert W. Burchfield (ed.), Supplement, 1972-86; John A. Simpson and Edmund S. C. Weiner (eds.), 2nd edition, 1989; John A. Simpson, Edmund S. C. Weiner and Michael Proffitt (eds.), Additions Series, 1993-97; John A. Simpson (ed.), 3rd edition (in progress) OED Online, March 2000- . Oxford: Oxford University Press. www.oed.com. 7 Note that fibula is sometimes glossed by sigil, e.g. Erfurt 408 (J. D. Pheifer (ed.), Old English Glosses in the Épinal-Erfurt Glossary, Oxford: Clarendon Press, 1974, p. 22), but OE sigil occurs as a word for ‘seal’ only in glosses. Cf. also unseglian (‘þa locan unsegloden’) alongside the more usual uninseglian. 8 T. A. Heslop, “English seals from the mid ninth century to 1100”, Journal of the British Archaeological Association, 133 (1980), 1-16; and op. cit.

What did Anglo-Saxon seals seal when?

133

reference in his version of Augustine’s Soliloquies has played an ambiguous role: geþenc nu gyf ðines hlafordes ærendgewrit and hys insegel to ðe cymð, hwæðer þu mæge cweðan þæt ðu hine be ðam ongytan ne mægæ, ne hys willan þær-on gecnawan ne mæge.9 Now consider whether you can say, if your lord’s letter and his seal comes to you, that you can’t recognize him by it or you can’t understand his bidding in it.10

Although the linking of ærendgewrit and insegel looks very like brevis et sigillum, the question often debated is whether or not the letter and seal (whatever seal may denote) in the passage need be understood as physically joined.11 Harmer argues that Alfred is writing about a sealed letter,12 as does Whitelock, even though the first examples of surviving sealed letters are from over a century later.13 Chaplais is less certain, pointing out that “we cannot tell whether the seal was carried loose or whether it was impressed on a loose tie or actually fastened to the letter itself.”14 For Keynes in his monograph on King Æthelred’s diplomas,15 the placing of hys implies that the

9

Thomas A. Carnicelli (ed.), King Alfred’s Version of St Augustine’s Soliloquies, Cambridge, Mass: Harvard University Press, 1969. 10 Simon Keynes and Michael Lapidge, Alfred the Great. Asser’s Life of King Alfred and other Contemporary Sources, Harmondsworth: Penguin, 1983, p. 141, translate: “Consider now, if your lord’s letter and his seal came to you, whether you could say that you could not recognize him by this means, and could not thereby know his intention”. 11 Carnicelli (1969), op. cit., p. 100 points out that analogically the phrase refers “to the scriptures”; see further Milton McG. Gatch, “King Alfred’s Version of Augustine’s Soliloquia: Some Suggestions on its Rationale and Unity”, in Paul E. Szarmach (ed.), Studies in Earlier Old English Prose, Albany: State University of New York Press, 1986, pp. 17-45, pp. 30-31. 12 Florence E. Harmer, Anglo-Saxon Writs, Manchester: Manchester University Press, 1952, p. 3. 13 Dorothy Whitelock, English Historical Documents c. 500-1042, 2nd edition, London and New York: Eyre Methuen, 1979, p. 917. 14 Chaplais, op. cit., p. 53 15 Simon Keynes, The Diplomas of King Æthelred ‘the Unready’, Cambridge: Cambridge University Press, 1980, p. 137; and see n. 185, where he draws attention to arguments put forward by Pierre Chaplais, “The Anglo-Saxon chancery: from the

134

Jane Roberts

letter and seal are separate, a position implicit also within his statement a decade later that: “King Alfred refers to a lord’s ‘written message and his seal’ . . . as if it were commonplace in his reign for the king (and perhaps any other lord) to make known his will by means of a written document associated with the impression of a seal”.16 That a seal could serve as some form of identification is undisputed. The Fonthill Letter (Sawyer no. 1445), addressed to Edward the Elder, Alfred’s son, supplies a reference to another early Anglo-Saxon seal, again possibly to a sealed document: Ða gesahte [? for gesohte] `he´ ðines fæder lic a brohte insigle to me, a ic wæs æt Cippanhomme mit te. Ða ageaf ic ðæt insigle ðe. a ðu him forgeafe his eard a ða are ðe he get on gebogen hæfð.17 Then he visited your father’s body and brought a ‘seal’ to me. And I was at Chippenham with you. Then I gave you the ‘seal’ and you allowed him his land and the possessions in which he has hitherto lived.18

Just what the trouble-prone Helmstan’s ‘insigle’ was remains unclear, as indeed do the circumstances of his visit to Alfred’s grave. The letter is dated to 899 x 924 (i.e. the reign of Edward the Elder) in Sawyer, and Whitelock thinks it “probably early in the reign”.19 Keynes, although “uncertain how much faith can be placed in the text as it diploma to the writ”, in Ranger, op. cit., pp. 43-62, pp. 51-54 [repr. from Journal of the Society of Archivists 3, 4, October 1966]. 16 Simon Keynes, “Royal government and the written word in late Anglo-Saxon England”, in Rosamund McKitterick (ed.), The Uses of Literacy in early Mediaeval Europe, Cambridge: Cambridge University Press, 1990, pp. 226-57, p. 244. 17 Simon Keynes, “The Fonthill Letter”, in Michael Korhammer with the assistance of Karl Reichl and Hans Sauer (eds.), Words, Texts and Manuscripts: Studies in AngloSaxon Culture Presented to Helmut Gneuss on the Occasion of his Sixty-Fifth Birthday, Cambridge: D.S. Brewer, 1992, pp. 53-97, p. 88. 18 Keynes (1992), op. cit., p. 88, translates: “Then he sought your father’s body, and brought a seal to me, and I was with you at Chippenham. Then I gave the seal to you, and you removed his outlawry and gave him the estate to which he still has withdrawn”. Harmer, op. cit., p. 12, translates: “Then he made his way to thy father’s (i.e. King Alfred’s) body, and brought an insegel to me, and I was at Chippenham with thee. Then I gave the insegel to thee. And thou didst give him back his home and the estates to which he has now returned”. 19 Whitelock (1979), op. cit., p. 544.

What did Anglo-Saxon seals seal when?

135

stands”,20 nevertheless concludes “that the letter was written nearer c.920 than c.900”,21 a conclusion Gretsch supports from her reading of the document’s wording.22 Even so, should insegel in the Fonthill Letter be understood as referring to some form of documentation accompanied by a seal, it may also be claimed as adding to the evidence for use of sealed letters before Edward’s reign. It is clear from Domesday Book that sigillum could on its own mean ‘sealed writ’ in 1086,23 but that is some way in the future. Two references in charters from Æthelred’s reign help little, viewed on one side as further evidence for the joining of a seal physically to a document, on the other as inconclusive. In the dispute over land in Berkshire, described in a chirograph dated 990 x 992 by Sawyer (no. 1454), Æthelred’s ‘insegel’, whatever it was, is given into the hands of abbot Ælfhere to take to a meeting at Cuckamsley: þa sende se cyning be Æluere abbude his insegel to þam gemote æt Cwicelmeshlæwe a grette ealle þa witan þe þær gesomnode wæron24 The king sent his seal to the meeting at Cuckamsley by Abbot Ælfhere, and greeted all the councillors who were assembled there

For Chaplais in the 1960s a loose seal was carried at Cuckamsley, and “the royal greeeting to the assembly was conveyed orally by the abbot”.25 However, as Keynes points out: “It is possible that this was a message conveyed orally by the abbot on the king’s behalf, supported by an impression of the king’s seal, but it is also possible that insegel 20

Keynes (1992), op. cit., p. 90 and fn. 152. Ibid., p. 95. 22 Mechthild Gretsch, “The language of the ‘Fonthill Letter’”, Anglo-Saxon England, 23 (1994), 57-102, 94. 23 Chaplais (1973), op. cit., p. 51, n. 67. Pierre Chaplais, English Diplomatic Practice in the Middle Ages, London and New York: Hambledon and London, 2003, p. 38 and n. 38, discussing Pope John XV’s intervention in the 990 quarrel between Æthelred II and Duke Richard I of Normandy as described in a contemporary or early contemporary manuscript, remarks of the phrase “sine sigillo eorum” that the seal is “likely to have been a loose impression of a seal-matrix”, referring back to 1973, p. 52 and n. 73. 24 A. J. Robertson (ed.), Anglo-Saxon Charters, 2nd edition, Cambridge: Cambridge University Press, 1956, no. 66, p. 136, ll. 9-11. 25 Chaplais (1973), op. cit., p. 56. 21

136

Jane Roberts

in this instance refers to a sealed writ”.26 According to another charter, its drafting dated 995 x 1005 by Sawyer (no. 1446), Archbishop Ælfric had the king’s ‘gewrit a his insegl’ at the hearing of a dispute over land ownership in Kent: Þa ða him seo talu cuð wæs · þa sende he gewrit a his insegl to þam arcebisceope Ælfrice a bead him o he a hys þegenas on East Cent · a on West Cent · hy onriht gesemdon · be ontale · a be oftale.27 When the claim was known to him, he sent a letter and seal to Archbishop Ælfric, and gave orders that he and his thegns in East Kent and West Kent should settle the dispute between them justly, weighing both claim and counterclaim.

This charter, preserved in the Textus Roffensis, records that the king sent letter and seal, but need these have been joined? For Keynes, unwilling to rule either way, this passage “is as clear an example of a reference to an administrative communication in writing as one could reasonably hope to get”.28 More recently, Wormald very sensibly keeps his eye on the importance of such documents as implying central supervision of litigation in shire courts, and does not speculate on how we are to visualize seal and/or letter. 29 Words for writ and seal occur linked once more in Old English, in the Peterborough Chronicle. Here Sparrowhawk, abbot of Abingdon, meets the newly installed archbishop of Canterbury. The events relate to 1051: 1048 Ða com Sparhafoc abb(od) be weg[e] to him mid þæs cynges ge write a insegle. | to þan þet he hine hadian sceolde to b(iscop) into Lundene. þa wið cweð se arce(biscop). a cwæð þet se papa hit him forboden hæfde.30

26

Keynes (1990), op. cit., p. 246; cf. Keynes (1980), op. cit., pp. 136-38. Robertson (1956), op. cit., no. 69, p. 140, ll. 15-18. 28 Keynes (1990), op. cit., p. 246. 29 Patrick Wormald, The Making of English Law: King Alfred to the Twelfth Century, Vol. 1 Legislation and its Limits, Oxford: Blackwell, 1999, pp. 153, 544. 30 C. Plummer, Two of the Saxon Chronicles Parallel, 2 vols., London: Oxford University Press, 1892-99, repr. with additions by D. Whitelock 1952, I, 172. 27

What did Anglo-Saxon seals seal when?

137

Then Abbot Sparrowhawk met him on the way with the king’s writ and seal to the effect that he was to be consecrated bishop of London by the archbishop. But the archbishop refused and said the pope had forbidden it him.31

Nevertheless Sparrowhawk, apparently with the king’s “full permission”, held the bishopric of London for the summer and autumn before being expelled. Stenton points out that this, the first time a pope rejected a candidate put forward by a king for a see in England, was “a most remarkable act of papal authority” on the part of Leo IX.32 But as actual sealed writs are extant from the reign of Edward the Confessor,33 this linking of gewrit and insegel does not arouse controversy. I mention it, however, to point to what looks like a further example of an established English collocation, however the elements be interpreted. Clearly, the issue of whether or not writs were sealed before Edward’s reign is unresolved. By contrast, there seems some measure of agreement that “the diploma was not sealed” in Anglo-Saxon England,34 or at least that royal charters were “always unsealed”.35 In being without clear marks of validation, “neither seal nor autograph subscription or signa”,36 they were oddly unlike their continental counterparts. Heslop, discussing the seal-matrices extant from AngloSaxon England, points out that seals were “intended originally for closing correspondence” and that “this remained their primary function until at least 1100”;37 and in his later entry on seals in the Blackwell Encyclopaedia he states circumspectly that Anglo-Saxon practice was unusual in “refusing to use seals on charters”.38 One 31

Dorothy Whitelock with David C. Douglas and Susie I. Tucker (eds.), The AngloSaxon Chronicle: A Revised Translation, London: Eyre and Spottiswoode, 1961, p. 117. 32 F. M. Stenton, Anglo-Saxon England, The Oxford History of England II, 3rd edition, Oxford: Clarendon Press, 1971, p. 467. 33 Chaplais (1973), op. cit., p. 45. 34 Harmer, op. cit., p. 35. 35 Chaplais (1973), op. cit., p. 45. 36 Pierre Chaplais, “The royal Anglo-Saxon ‘chancery’ of the tenth century revisited”, in H. Mayr-Harting and R. I. Moore (eds.), Studies in Medieval History Presented to R. H. C. Davis, London, 1985, pp. 41-51, p. 46. 37 Heslop (1980), op. cit., p. 2. 38 Heslop (1999), op. cit., p. 414.

138

Jane Roberts

piece of evidence for the Anglo-Saxon understanding of sealed charters has however been overlooked, a passage in the Vercelli Book tenth homily. Here God speaks to a rich man, asking why he does not share his wealth with the poor: To hwan [geagnedest þu þe anum þæt ic inc bam sealde. To hwan] feddest ðu þe ænne of ðam þe ic inc bæm gesceop, to welan a to feorhnere? To hwan heold ðu hit þe sylfum a þinum bearnum, þæt meahte manegum mannum genihtsumian? Unyðe þe wæs þæt ðu hit eal ne meahtest gefæstnigan, ne mid insigelum eal beclysan. Wenst ðu ðæt hit þin sie þæt sio eorðe forðbringeð, hio þe groweð a bloweð a [sæd lædeð a] onlif[an] bringeð? Eall ic nu afyrre minne fultum fram ðe. Hafa æt þinum gewinne þæt ðu mæge a on þinum geswince. Ic ofteo mine renas þæt hie þine eorðan ne onhrinað, a ic afyrre fram þe mine mildheortnesse, a þonne bið sona gecyðed a ætiewed þinra yrmða dæl. Gif ðu wene þæt hit þin bocland sie a on agene æht geseald, hit þonne wæron mine wæter þa ðe on heofonum wæron, þanon ic mine gife dæle eorðwærum. 39 Why [did you keep for yourself alone what I gave to both of you. Why] did you feed yourself alone from those things which I made for the support and well being of you both. Why did you keep for yourself and your children what could have been enough for many? You found it hard that you could not secure it all and protect it all with seals. Do you think that it is yours, what the earth produces, that for you the earth grows and gives breath and [makes seed grow and] yields food? I shall now take away my support from you completely. Get what you can from your labour and in your wretchedness. I shall withdraw my rain showers so that they make no contact with your land and I shall take away my mercy from you and then straightaway your share of hardships will be known and seen. If you think that it is your freehold and made over into your own keeping, then they were my showers of water that were in the heavens, whence I share out my gifts to the world’s inhabitants.40

39

D. G. Scragg (ed.), The Vercelli Homilies and Related Texts, Early English Text Society os 300, Oxford: Oxford University Press, 1992, pp. 205-06, ll. 160-72. 40 MacArthur in L. E. Nicholson (ed.), The Vercelli Book Homilies. Translations from the Anglo-Saxon, Lanham, New York, London: University Press of America, 1991, p. 77, translates the two sentences in question by “It was not easy for you that you might secure it all, nor with seals enclose (it) all” and “If you fancy that it is your legal property [lit. land held by written title] and granted into (your) own possession, they were then my waters which were in the heavens, whence I dispense My gifts among earth-dwellers.”

What did Anglo-Saxon seals seal when?

139

I have underlined both in the Old English and its translation two sentences that involve imagery from the granting of charters, present in all five extant variants of this passage,41 for which the Vercelli Book, dated to the second half of the tenth century, provides the earliest witness in English. There is nothing so explicitly concerned with land tenure in the nearest witness for the putative source of Vercelli homily X, a pseudoAugustine homily, Sermo CCCX, as it appears in Salisbury, Cathedral Library 9,42 unless ‘adsignas’ be thought of as triggering the sentence that contains the phrase ‘mid insigelum’. Because Scragg in his edition of the Vercelli homilies presents from Salisbury 9 a selective Latin text in order to indicate words likely to have been available to the redactor of Vercelli X, he omits sentences not reflected in his English homily and he has, moreover, made silent emendations where the Salisbury 9 text “has errors or omissions, or is illegible”.43 The text cited below is taken from Cross’s edition of the Sermo CCCX text in Salisbury, Cathedral Library 179, with the variant readings recorded by Cross from Salisbury 9 and from Vienna, Österreichische Nationalbibliothek 994:

50 Quid tibi soli uendicas quod ambobus dedi? Quod solus comedis quod ambobus creaui? Quid te fingis filiis tuis seruare quod potest omnibus sufficere? Quia quod ego do non potest deficere aut si tuo labori hoc quod habes adsignas, aut si tuos esse putas ipsos fructus quos terra producit, ecce, aufero auxilium meum et habeto laborem tuum. Subduce pluuiam et sterilem

41 Five altogether of the nine variants of this homily. The other four are in Cambridge, Corpus Christi College 419 + 421 and Cambridge, Corpus Christi College 302 (both more or less complete homilies); and ll. 122-end in Cotton Faustina A. ix and Cambridge, Corpus Christi College 302 (the latter repeated material within the manuscript). 42 Scragg, op. cit., p. 203. 43 Italics as in Scragg (op. cit., p. 203). The Latin he supplies for comparison (pp. 20506) signals omissions: ‘Quid tibi soli uindicas quod ambobus dedi? quid tu solus comedes quod ambobus creaui? Quid te filiis tuis fingis seruare quod potest omnibus sufficere? . . . Aut si tuo labore hoc quod habes adsignas, aut si tuos esse putas ipsos fructos quos terra produxit. Ecce aufero auxilium meum, habeto laborem tuum. Subduco pluuiam meam et arefaciam terram tuam. Aufero misericordiam meam, et tunc apparebit miseria tua. Si terra tua est, mea est pluuia quae super terram tuam discendit.’

140

Jane Roberts

faciam terram tuam. [fol. 46v] Auferam misericordiam meam et apparebit miseria tua. Si terra tua est, mea est pluuia. 50 uendicas] V, S9 uindicas. 50-51 Quod solus comedis quod ambobus] S9 Quid tu solus comedis quod ambob . . . [illegible]; V omits. 51 creaui] V dedi uel creaui; S9 illegible. 52 Quia quod ego . . . deficere] V omits. 53 labori] S9 labore. 54 producit] S9 produxit. 55 pluuiam] V, S9 pluuiam meam. 55 sterilem faciam terram tuam] V afficiam terram tuam; S9 arefaciam terr . . . [illegible]. 56 apparebit] S9 parebit. 57 pluuia] V pluuia quae super terram tuam discendit. 44

Cross shows that these three closely related manuscript witnesses to Sermo CCCX, Salisbury 9 (for Scragg, nearest to Vercelli X), Salisbury 179 (the version cited by Godden as nearest the Ælfric homily noted below) and Vienna 994, all have affinities with what must have been the text type available in Anglo-Saxon England. None of these Latin texts contains any explicit reference to charters or seals. Yet, the maker of Vercelli homily X, well aware of the use of seals as a mode of validating charters, introduces into his text the legal concept bocland where Old English land on its own would have sufficed to translate terra. Even though ‘bocland’ may be thought of here as ‘figurative, referring to the earth (and its fruits)’,45 its presence in the passage reinforces how the earlier phrase ‘mid insigelum’ is to be understood. The rich man of Vercelli homily X is clearly concerned to safeguard his property with seals attached to documentation.46 Thus it would seem that the securing of charters with seals was not a concept foreign to Anglo-Saxon audiences at least as early as the 960s to 970s, the period to which the Vercelli Book is generally dated. The audience for this homily must have been aware of the greater security of land tenure afforded by protection mid insigelum. 44

J. E. Cross, “A Sermo de Misericordia in Old English prose”, Anglia, 108 (1990), 429-40 (text, pp. 431-33). 45 Dictionary of Old English in Electronic Form A - F, Toronto: Pontifical Institute of Mediaeval Studies, 2003, s.v. boc-land 1.c. 46 Scragg does not discuss these sentences in his commentary, nor are they discussed in most recent essay I have seen on homily X, Jonathan Wilcox, “Variant texts of an Old English homily: Vercelli X and stylistic readers”, in Paul Szarmach and Joel Rosenthal (eds.), The Preservation and Transmission of Anglo-Saxon Culture: Selected Papers from the 1991 Meeting of the International Society of AngloSaxonists, Stud. in Med. Culture 40, Kalamazoo, MI: Medieval Institute Publications, Western Michigan University, 1997, pp. 335-51.

What did Anglo-Saxon seals seal when?

141

It is curious that this Old English passage has not played a part in discussion of Anglo-Saxon seals and charters, but hardly surprising, for printed versions of the passage as it appeared in other manuscripts were available only in scattered publications.47 The whole Vercelli Book could be consulted in Förster’s much used reduced facsimile48 and his edition of Vercelli homilies I to VIII was well known,49 but his long-awaited edition of Vercelli homilies IX to XXIII never appeared (already in print, it was lost during Second World war bombing).50 Even so, that material reflecting the pseudo-Augustinian Sermo CCCX lay behind the passage was first pointed out by Becker.51 This was the source text that Jost had earlier identified as lying behind a section of homily VII in Ælfric’s second series of Catholic Homilies (it was noted also in relation to another Ælfrician homily).52 Part of this Ælfrician homily corresponds quite closely in content to the passage from Vercelli X discussed above. Ælfric is discussing how God tests the rich by giving them more riches than they need: hwí sceal he ðonne him anum geágnian þæt him bám is forgifen? Gif ðu talast to ðinum geswince þæt þæt ðu hæfst. oððe gif ðu wénst þæt ðære eorðan wæstmas ðine sind. ðonne cweð se ælmihtiga wealdend to ðe; Efne nu ic ðe oftéo minne fultum. and hafa ðe þín geswinc; Ic ofteo mine rénscuras. and ic wyrce ðin lánd unwǙstmbære; Gif þæt land ðin is. se rén is min;53 Why then must he own by himself alone what is given to them both? If you ascribe to your labour what you own or if you think that the fruits of the earth are yours, then the ruler almighty will say unto you: “Now behold, I shall 47

A. S. Napier (ed.), Wulfstan, Sammlung englischer Denkmäler 4, Berlin: Weidmann, 1883 [repr. with appendix by K. Ostheeren, 1967], homily XLIX, is probably the best known of the versions available. 48 Max Förster, Il Codice Vercellese . . . riprodotto in Fototipia . . . con Introduzione, Rome: Vatican, 1913. 49 Max Förster (ed.), Die Vercelli-Homilien: I-VIII. Homilie, Bibliothek der angelsächsischen Prosa 12, Hamburg: H. Grand, 1932. 50 Scragg, op. cit., p. xxii. 51 Wolfgang Becker, “The Latin manuscript sources of the Old English translations of the sermon Remedia Peccatorum”, Medium Ævum, 45 (1976), 145-52. 52 K. Jost, Wulfstanstudien, Bern: A. Francke, 1950, pp. 246-47; J. C. Pope (ed.), Homilies of Ælfric: A Supplementary Collection, 2 vols., Early English Text Society os 259, 260, London: Oxford University Press, 1967-68, II, 509. 53 Malcolm Godden (ed.), Ælfric’s Catholic Homilies: The Second Series, Text, Early English Text Society ss 5, London: Oxford University Press, 1979, p. 62, ll. 71-78.

142

Jane Roberts

withdraw my help, and you, keep your labour. I shall withdraw my rain showers and I shall make your land infertile. If the land is yours, the rain is mine.”

The Salisbury 179 version of Sermo CCCX is deemed the nearest identified source for this section of the homily, and it is interesting therefore to look at the extracts drawn from it in Godden and Clemoes’s edition of the Catholic Homilies as evidence for Ælfric’s source: Quid tibi soli v[i]ndicas quod ambobus dedi? Quod solus comedis quod ambobus creavi? . . . Si tuo labori hoc quod habes adsignas, aut si tuos esse putas ipsos fructus quos terra producit, ecce, aufero auxilium meum et habeto laborem tuum. Subduco pluviam et sterilem faciam terram tuam . . . Si terra tua est, mea est pluvia.54

The omissions made by Godden, which can be gauged from Cross’s edition of the relevant passage as cited above, are of materials not reflected in Ælfric’s homily. It is possible that the gaps Godden signals between Ælfric’s text and the nearest identified source reflect absences in Ælfric’s actual source. The reference to keeping for one’s self and one’s children what God gives for all, probably not a part of whatever source text was used here by Ælfric, is a sentence that could have helped point the maker of the English homily illustrated from Vercelli X towards his more specifically legalistic interpretation. By contrast, there is no charter imagery in Ælfric’s passage, although it stems ultimately from some similar source. As Cross has pointed out, the Vercelli homilist and Ælfric were working from divergent Latin texts, and we cannot be certain of what actually lay before each. Whereas in the case of Vercelli X the Latin wording available could conceivably have prompted its author to think in terms of seals and charters, Ælfric might have had a redaction more tightly focused on the pairing of contrasted rich and poor man and without reference to heirs and the passing on of property. Yet, no matter the degree of divergence between the two English homilies, let alone the impossibility of knowing the wording of their actual source texts, it 54

Malcolm Godden, Ælfric’s Catholic Homilies. Introduction, Commentary and Glossary, Early English Text Society ss 18, Oxford: Oxford University Press, 2000, pp. 396-97. The omissions signalled are made by Godden.

What did Anglo-Saxon seals seal when?

143

can be assumed from Vercelli X that tenth-century Anglo-Saxons were well aware of the use of seals for the authentication of charters. Similarly, there can be no doubt of the functionality of the seven seals in a passage in Wærferth’s translation of Gregory’s Dialogues: a eac Iohannes stefn heþwæreþ þam ylcan aÑyte in þam ilcan wene, se sæde, þæt he hesawe boc mid seofon insehlum ÑeinseÑlode, a þæt næniÑ man wære Ñemeted wyrðe ne on heofonum ne on eorðan ne under eorðan, þe þa boc moste untynan a hire inseÑlu tobrecan . . . . . þa boc swa þehhweþre Iohannes sæde, þæt heo wære æfter þon untyned þurh þone leon of Iudan cynne.55 and John’s testimony also accords with that same understanding in that same expectation, who said that he saw a book sealed with seven seals, and that that no man of sufficient worth might be found in heaven or earth or under the earth of sufficient worth who could open the book and break apart its seals. Nevertheless John said that the book would thereafter be opened through the lion of the race of Judah.

This is the book in Revelation, the ‘librum, scriptus intus et foris’ (‘book written within and without’).56 Where once these seals would have been understood as applied to a scroll,57 with writing not just on the inside but on the outside also, the shift in the early Christian world to the use of the codex for recording Christ’s teachings and to the codex’s eventual dominance led to changing visualization of the ‘book’. It is impossible to say how Wærferth visualized this ‘boc’: van der Meer’s Apocalypse collects together many splendid images from the medieval world, not just of rolls with seven seals but of codexes 55

H. Hecht (ed.), Bischof Waerferths von Worcester Uebersetzung der Dialoge Gregors des Grossen, Bibliothek der angelsächsischen Prosa 5, Leipzig: G.H. Wigland, and Hamburg: H. Grand, 1872 [repr. Darmstadt, 1965], p. 332, ll. 21-26. 56 Rev 5: 1; compare also Is 29: 11-12 and Ezek 2: 9-10. (Vulgate citations from Weber, R. (ed.), Biblia sacra iuxta vulgatam versionem, Stuttgart, 1969.) Translations The Holy Bible translated from the Latin Vulgate diligently compared with the Hebrew, Greek, and Other Editions in divers languages. The Old Testament first published by the English College at Douay, A.D. 1609 and the New Testament first published by the English College at Rheims, A.D. 1582, London, 1899. 57 G. S. Ivy, “The bibliography of the manuscript-book”, in Francis Wormald and C. E. Wright (eds.), The English Library Before 1700, London: University of London, Athlone Press, 1958, pp. 32-65, 32, points out that liber “originally denoted a roll, as did volumen or volume”.

144

Jane Roberts

with seven bands also.58 Wærferth’s emphasis is upon ‘þa boc’, and he shifts tenses when reporting what John said, by contrast with Gregory’s wording, which continues in the present tense (in the edition of the Dialogues cited, biblical passages are helpfully italicized): Et Iohannis uox in ea aestimatione concordat. Qui cum signatum librum septem sigillis uidisse se diceret, quia nemo inuentus est dignus neque in caelo, neque in terra, neque subtus terra aperire librum et soluere signacula eius, adiunxit : Et ego flebam multum. Quem tamen postmodum librum per leonem de tribu Iuda dicit apereri.59 And John’s words accord with this assessment; who when he says that he saw a book sealed with seven seals, adds, because no-one, neither in heaven nor on earth nor under the earth, has been found worthy of opening the book and loosing its seals, And I wept copiously. He says that nevertheless the book will be opened by a lion of the tribe of Judah.

Where Hecht, the editor of Wærferth’s translation, signals a gap, suggesting that he had compared the English with the Latin and found it wanting, it is possible to argue that Wærferth decided to omit “adiunxit : Et ego flebam multum” as extraneous to his purpose. The book of the Apocalypse and its seals appear also in the gloss to the tenth-century Durham Ritual: Dignus es domine deus accipere librum et aperire signacula eius. quoniam occisus es et redemisti nos deo in sanguine tuo. wyrðe arð driht' god onfoa bóc a vntyne insiglae his f'ðon ofslægen arð a gilesdes vsig gode in blode ðinvm60

Christ, the lion of the race of Judah, is the redemptor who will open the seals of the book at the end of the world. For Orm, late in the twelfth century, insegel remains the word to use when writing of ‘an

58

F. van der Meer, Apocalypse: Visions from the Book of Revelation in Western Art, Antwerp: Mercatorfonds, 1978. 59 Adalbert de Vogüé (ed.), Grégoire le Grand. Dialogues, 2 vols., Sources chrétiennes 251, 260, 265, Paris: les éditions du cerf, 1978-80, III, 158, ll. 10-15. 60 A. H. Thompson and U. Lindelöf (eds.), Rituale ecclesiae Dunelmensis, Surtees Society 140, Durham, 1927, p. 29, l. 9.

What did Anglo-Saxon seals seal when?

145

boc | Bisett wiþþ seffne innseÑÑless’ (l. 260) whose ‘seffne inseÑÑless’ (l. 265) will be opened by ‘Godess Lamb’.61 How we are to visualize the sealing devices used in some other contexts is as foxing as teasing out the meaning of insegel in relation to early Anglo-Saxon charters. Sometimes we gain a clear image of the form the seal takes as well as of its purpose. A 1081 diploma of William I to supporters of Baldwin, abbot of Bury St Edmunds, notes the validation of a charter seal-impression: 62 & we gefæstnodan eac þas kartan mid uran agenum hand gewrite. & mid ures insigeles onðryste to þam þat þara foresprecenan stowe freodom on ecnysse þurhwunie. and we also confirmed the charter with our own writing and with the thrust of our seal in order that the aforesaid foundations immunity may continue forever.

Although the word ‘onðryste’ is not recorded elsewhere,63 it translates ‘impressione’ and clearly refers to the impression made by the king’s seal. So too, when Ælfric describes the sealing of Christ’s tomb, it is likely that his words indicate the making of an impression: Ge habbað weardas. farað to & healdað; hi ða ferdon to; a mearcodon ða þruh mid insegle. a besæton þa birgine;64

61

Orm uses inseil four times in his dedication, all in reference to the Revelation book with seven seals (Robert Holt (ed.), The Ormulum, with the Notes and Glossary of Dr. R.M. White, 2 vols, Oxford: Clarendon Press, 1878). 62 D. C. Douglas (ed.), Feudal Documents from the Abbey of Bury St. Edmunds, British Academy: Records of the Social and Economic History of England and Wales, 8 (1932), 53, ll. 40-42 from no. 7 (fold-out reduced facsimile at end of volume); no. 8 is a brief writ in Latin noting the decision given at trial for Baldwin. For both these documents and others about the same incident see David A. E. Pelteret, Catalogue of English Post-Conquest Vernacular Documents, Woodbridge: Boydell Press, 1990, no. 28. 63 It is probably related to thrust forms, the verb recorded first in the twelfth century and the noun from the sixteenth century (T. F. Hoad (ed.), The Concise Oxford Dictionary of English Etymology, Oxford: Oxford University Press, 1986, p. 492), and cognate with the native word þræstan. 64 Peter Clemoes (ed.), Ælfric’s Catholic Homilies:The First Series, Text, Early English Text Society ss 17, London: Oxford University Press, 1997, p. 299, ll. 12-14.

146

Jane Roberts

You have watches [to keep]. Go to [your positions] and hold [them]. Then they went to [their positions], and marked the tomb with a seal and guarded the grave.

This at any rate is what is to be understood by ‘signing the stone’ in the closest parallel verse in the gospels, in Matthew 27.66: illi autem abeuntes munierunt sepulchrum signantes lapidem cum custodibus. And they departing, made the sepulchre sure, sealing the stone, and setting guards.

Figurally however, the marking of the stone might have recalled the king’s marking of the stone that closes the lions’ den ‘obsignavit rex annulo suo et anulo optimatum suorum’ (‘which the king sealed with his own ring, and with the ring of his nobles’) in the book of Daniel,65 where the specification of agency of rings helps confirm this interpretation of Ælfric’s phrasing. But is ‘insegle’ to be understood as seal impression or seal ring? Again, an insegel is used by St Benedict in the Old English Dialogues, to secure a loc: a þa þa he Ñeseah, þæt þa þearfan Ñenoh hæfdon, he het þone cniht stiÑan of þære wintreddan a beleac þæt winern a asette his aÑen inseÑl on þæt loc. he forlet hit swa belocen a sona him hecyrde eft to cyrcan.66 and when he saw that the poor had had enough, he ordered the boy to come up out of the wine press and closed the wine-pantry and placed his own seal on the lock. He left it closed in this way and straightaway went back to church.

Here the function of the seal is made explicit: it makes the wine-store doubly secure. But is this ‘loc’ a bolt or bar or lock in our terms? How we interpret loc inevitably affects out interpretation of insegel. Gregory’s words indicate that it is the seal’s impress here:

65 66

Dan 6: 17, and cf. (in the story of Bel) 14: 10, 13. Hecht, op. cit., p. 59, ll. 1-8.

What did Anglo-Saxon seals seal when?

147

Quibus cum se idonee satisfecisse pauperum conspiceret, ex calcatorio puerum iussit ascendere, apothecam clausit, atque inpresso sigillo proprio munitam reliquit, moxque ad ecclesiam rediit.67 When he saw the poor appropriately satisfied, he ordered the boy to climb up from the winepress, closed the store-room, and left it protected by the impress of his own seal, and immediately he went back to the church.

Is Wærferth unwittingly communicating to us his difficulty in explaining how Boniface secured the wine-store? Or is he carefully explaining the securing of the door in words he expects will be understood? There are quite a few instances of the sealing of a loc in Old English where the verb used is geinseglian, a verb used also of the sealing of doors, but there is only one further such context that includes the noun insegel. In the Old English Gospel of Nicodemus, great emphasis is placed on the securing of Joseph of Arimathea, and the Latin source is followed closely. Three passages focus on the sealing of Joseph’s prison. Here the texts and translations all come from the recent edition.68 First, the door is closed ‘fæste’:69 And Anna and Caiphas þæt loc geinseglodon and þærto hyrdas setton. and Anna and Caiphas set a seal over the lock and placed guards there. Signauerunt ostium cubili super clauem Annas and Caiphas, et custodes posuerunt. Annas and Caiphas set a seal over the door of the chamber and placed guards.

It is worth noting that the modern English translation of the Latin gives no direct translation of the phrase ‘super clauem’, whereas the Old English has loc, a word that could more readily be understood as ‘lever, bar’ than the Latin clavis. In the second passage, when Joseph’s prison is opened, the Old English text does not translate 67

de Vogüé, op. cit., II, 78, 41-44. J. E. Cross (ed.), Two Old English Apocrypha and their Manuscript Source. ‘The Gospel of Nichodemus’ and ‘The Avenging of the Saviour’, Cambridge Studies in Anglo-Saxon England 19, Cambridge: Cambridge University Press, 1996. 69 Ibid., XII.1 68

148

Jane Roberts

‘signa’ directly, but makes it clear that logically both lock and key needed to be unsealed and door as well: and hig uninseglodon þæt loc and þa cægan and, þa duru geopenigende and unsealed the lock and the key and, when the door opened Et aperientes clauem et signa And opening the [door with the] key and the seals of the door70

Here too the translation of the Latin demonstrates just how tricky such words as door, lock and seal are: oddly the rationalizing phrase ‘[door with the]’ is given square brackets, but not the added ‘of the door’. The third passage, in which insegel is used, comes from questions that soldiers put to Joseph’s captors, as to why Joseph had vanished despite being locked up securely: ge hyne on fæstre clusan beclysdon and þæt loc myd insegle geinseglodon you locked him up in a very secure prison and sealed the lock with a seal inclusistis in cubiculum super clauem signantes you shut him up in a chamber sealing the entrance above the lock71

The additional explanatory phrase ‘myd insegle’ gives the impression of clarificatory detail, but insegel is open to interpretation either as the sealing implement (by means of ring or whatever) or the applied seal (with seal, in whatever form). Interestingly, in the alternate witness to the Old English text the second of these clauses runs ‘a o loc myd insægle macodon’,72 which can be translated very literally as ‘and made the closure with a seal’. Even though the Cambridge manuscript, on which the recent edition is based, is shown to be nearer generally to

70

Ibid., pp. 174-75, XIII.2. Ibid., XIII.1. 72 W. H. Hulme, “The Old English version of the Gospel of Nicodemus”, Publications of the Modern Language Association of America, 13 (1898), 457-542, 487, ll. 5-6. 71

What did Anglo-Saxon seals seal when?

149

the Latin original,73 perhaps this clumsier reading, with its unusual correlation of macian with loc,74 reflects the original translation, for which the simpler ‘myd insegle geinseglodon’ is a substitution. On balance, it is likely that ‘myd insegle’ is to be understood as ‘with a[n impressed] seal’, given that ‘signa’ occurs in the second passage translated. When the translator of the Seven Sleepers encountered the first description of the sealed casket, ‘scribentes litteras, et sigillantes deposuerunt secrete’ [‘they wrote the document and sealed it and deposited it secretly’],75 he included at this point a more explicit reference to the box in which the document is placed: ‘and hi ðæt gewrit mid twam sylfrenan inseglum on anre teage geinsegledon’.76 The translation in Skeat’s edition, in suggesting that silver seals are attached to a document, is misleading, ‘and they sealed up the writing with two silver seals in a casket’, where the elements in the clause are wrongly ordered. They should instead be read in the order ‘and they sealed the written statement in a casket with two silver seals’, as is made clear from the later finding of the casket: ‘ane teage, seo wæs geinsægled mid twam sylfrenan insæglan’ (‘a casket, which was sealed with two silver seals’) [‘loculum sigillatum duobus sigillis argenteis’ 88-89, ll. 324-25 ‘a box sealed with two silver seals’].77 With the addition of a new detail in cross-reference to the earlier sealing of the casket, the seals are given emphasis: ‘þæt þa insægla wæron eft to swutelunge hwæt man þærinne funde, þonne se tima gewurðe eall swa God wolde þæt þa gewurðan sceolde’ [‘that the seals were to reveal afterwards what should be found inside when the time should come just as God wished’].78 It is possible that insegel here is used as a more general term for ‘lock, bolt, etc.’, but more likely that the silver seals on the outside of the casket bore the stamp 73

Cross (1996), op. cit., p. 97. The words correlate otherwise of the making of an agreement, in Peterborough Chronicle 1094 (Plummer, op. cit., I, 229). 75 Hugh Magennis (ed.), The Anonymous Old English Legend of the Seven Sleepers, Durham: Durham Medieval Texts, 1994, 80-81, l. 159. 76 Walter W. Skeat, Ælfric’s Lives of Saints. Early English Text Society, OS 76, 82, 94 and 114, London, New York, Toronto: Oxford University Press, 1881-1900; repr. 1966, 508-09, ll. 343-44 = Magennis 43, ll. 313-14. 77 Skeat, op. cit., 534-35, ll. 755-56 = Magennis, op. cit., 55, ll. 694-95. 78 Skeat, op. cit., I. 534-35, l. 758-60 = Magennis, op. cit., 55, ll. 697-99. 74

150

Jane Roberts

of the king’s ring. Similarly, it is likely that some form of seal impression may have been involved in sealing a wounded hand for the length of the ordeal period: a inseglige man þa hand, a sé[c]e man ófer þæne þriddan dǙg, swa hwæðer swa heo béo ful swa clæne binnan þam insegle.79 and the hand is to be sealed, and after the third day it should be established whether it is festering or clean under the seal.

Although insegel could have extended its field of reference to include ‘bolt, bar, lock, etc.’, the exact nature of the sealing provided is not any more transparent in these contexts than in the passages more closely related to the putative sealing of writs and documents. There are grounds, however, for recognizing the greater referentiality of the word insegel in abstract areas, although the dictionaries do not note its figurative extension for Old English.80 The word occurs twice in a phrase used of the making of the sign of the cross. One of these instances is in the late West Saxon life of St Machutus, first printed as recently as 1984: ‘Þa se halga machu hine geinseglade · mid insegle þære halgan rode’ (Then the holy Machutus sealed himself with the seal of the holy cross).81 And perhaps it was because the life of Machutus was as yet undeciphered that the dictionary makers were slow to pick up a comparable phrase in the Vespasian life of Guthlac: ‘Mid þy þære nihte þystro gewiton, and hit dæg wæs, þa aras he and hine sylfne getacnode insegle Cristes rode’ [‘When the night’s dark clouds dispersed and it was day, he got up and signed himself with the seal of the cross of Christ’].82 In both these passages insegel is wider in extension than Chaplais’s “normal 79

Felix Liebermann (ed.), Die Gesetze der Angelsachsen, 3 vols, Halle: Max Niemeyer, 1903-1916, I, 387 [5.2]. 80 For fuller discussion of these figurative uses of insegel see “Guthlac of Crowland and the seals of the cross”, in Karen Jolly, Catherine Karkov and Sarah Keefer (eds.), Cross and Culture in Anglo-Saxon England (forthcoming). See also Jane Roberts, “The case of the miraculous hand in the Old English prose life of Guthlac”, American Notes and Queries, 15 (2002), 17-22. 81 David Yerkes, The Old English Life of Machutus, Toronto, Buffalo, London: University of Toronto Press, 1984, pp. 40-41. 82 Paul Gonser (ed.), Das angelsächsische Prosa-Leben des hl. Guthlac, Anglistische Forschungen 27, Heidelberg: Carl Winter, 1909, 2.61.

What did Anglo-Saxon seals seal when?

151

usage” for the Latin sigillum, which is the word translated in Felix’s life of Guthlac, whereas Bili’s life of Machutus has signaculum,83 so the collocation of insegel and rod may well have been established in the language. The Old English prose life of Guthlac has another instance of insegel used figuratively, in the first miracle connected with the saint. The boy is marked out at birth by a heavenly sign (‘tacn of heofenum’), which clearly enclosed him with redemptive seals (‘swytelice mid inseglum beclysde’); and what many people saw was a hand of the loveliest red colour holding a golden cross, coming from heaven and pointing to the door of the house in with Guthlac was born. The word insegel continued in use into the early Middle English period, overlapping with segl (never frequent) and with the Frenchderived seal, which was to slip into its semantic space. Orm uses insegel of the seven seals of the Apocalpyse, just as had the translator of Gregory’s Dialogues. In a Lambeth Homily the sign of the cross on ‘ure forheafod and þa .vii. Ñeade ures lichomes’ is ‘þet inseil þe þe deofel ne mei nefre to breocan’.84 In the Bodley 34 life of Margaret the saint defiantly tells Olibrius she has no fear because the Lord ‘haueð his merke on me iseiled wið his in-seil’.85 In these last two instances it is possible to argue for the continued use of insegel as for a ‘sign’ that seals,86 evidence for this extension not only in Old English but in early Middle English too. Yet, the Middle English Dictionary editors point out that in-seil in St Marherete might be interpreted as ‘the instrument with which the impression is made’, 83

For the meaning of signaculum, compare the phrase ‘signaculum apostolatus’ (‘seal of apostleship’) of I Corinthians 9.2. 84 Lambeth Palace Library 487, XII (Richard Morris, Old English Homilies and Homiletic Treatises (Sawles Warde, and þe Wohunge of Ure Lauerd : Ureisuns of Ure Louerd and of Ure Lefdi, &c.) of the Twelfth and Thirteenth Centuries. First Series. 2 vols., Early English Text Society, os 29, 34, London: N. Trübner & Co., 1868; repr. 1969, 126-27). 85 Frances M. Mack (ed.), Seinte Marherete. Þe Meiden ant Martyr re-edited from MS. Bodley 34, Oxford and MS. Royal 17A xxvii, British Museum, Early English Text Society, os 193, London: Oxford University Press, 1934, p. 12, ll. 12-13. 86 An abstract sense is implied also by the gloss insegle, mercelse for signaculo (gloss no. 4058 in L. Goossens, The Old English Glosses of MS. Brussels, Royal Library, 1650 (Aldhelm's De laudibus uirginitatis), Brussels, 1974, and compare Arthur S. Napier, Old English Glosses: Chiefly Unpublished, Oxford: Clarendon Press, 1900, gloss 4177.

152

Jane Roberts

explaining the Middle English forms otherwise as ‘[a]n impression made in wax, lead, or the like; a seal’.87 By a satisfying coincidence, it is a Bodley 34 text too that attests to the first instance of the new loan word seal, for which the OED first citation is dated c.1230. The actual passage, from Hali Meiðhad, is to be found in the OED stashed under the verb seal, ‘Ant tu . ._þet art iloten to him wið mei[ð]hades merke, ne brec þu nawt þet seil þet seileð inc togederes’.88 Tucked away as if anomalous, the passage falls under subdivision 1.b, characterized as ‘fig.’, with the meaning ‘A token or symbol of a covenant; something that authenticates or confirms; a final addition which completes and secures.’ The most recent editor of Hali Meiðhad points out that the ‘image of the seal of virginity’ is based on the Song of Solomon 4: 12.89 The next citation under 1.b is virtually three centuries later, from Tyndale’s gospels. It is noteworthy also that the figurative use of seal appears a generation or more in advance of what might be thought of as its prototypical use in a 1258 charter of Henry III, ‘We senden Ñew þis writ open, sened wiþ vre seel’. This 1258 citation is placed at the head of OED sense 1.a: ‘A device (e.g. a heraldic or emblematic design, a letter, word, or sentence) impressed on a piece of wax or other plastic material adhering or attached by cords or parchment slips to a document as evidence of authenticity or attestation; also, the piece of wax, etc. bearing this impressed device.’ In conclusion, it is typically the function both of the insegel and the seal to authenticate or to attest, and these words therefore easily extend to such senses as ‘symbol, token’. Because closing is often a factor in completing and securing, seal and insegel can overlap in meaning with such words bolt, bar, lock; and from the middle of the nineteenth century seal has held the technical meanings (see under 7.b in the OED) that support the definition ‘Any means of preventing the passage of gas or liquid into or out of something, esp. at a place where two surfaces meet’. Opening can of course be effected by the unsealing or breaking of insegel or seal. By contrast, such words as door and window, which allow the temporary closure of space, have 87

Hans Kurath, Sherman M. Kuhn and Robert Lewis (eds.), Middle English Dictionary, Ann Arbor: University of Michigan Press, 1952–2001. 88 Bella Millett (ed.), Hali Meiðhad, Early English Text Society, OS 284 (London, New York, Toronto: Oxford University Press, 1982), p. 5, ll. 12-13.. 89 Ibid., p. 30.

What did Anglo-Saxon seals seal when?

153

more to do with affording openings than with closure. We can prattle about doors to the future or windows of opportunity, but seals of approval or confession are concerned with authentication and closure respectively. The metaphoric ranges do not match. And that is because insegel, every bit as much as its successor in English, seal, is essentially a slippery word, not just potentially but actually huge in extension. Appendix NB: These citations for inseg(e)l are from the DOE database.90 Branch III draws together those instances open to grouping under I or II. inseg(e)l I. a seal .(adjoined in some way to a document as evidence of authenticity) HomS 40.3 (ScraggVerc 10) B3.2.40.6 [(163)] Unyðe þe wæs þæt ðu hit eal ne meahtest gefæstnigan, ne mid insigelum eal beclysan. HomS 7 B3.2.7 [(62)] Tohwon heolde þu hit þe anum and þinum bearnum þæt þæt mihte manegum genihtsumian uneaðe wæs þe þæt þu hit eall ne mihtest gefæstnian ne mid inseglum beclysan? HomS 40.1 (Nap 49) B3.2.40.2 [(194)] Unyðe þe wæs, þæt þu eall ne mihtest gefæstnian ne mid inseglum beclysan. GDPref and 4 (C) B9.5.6 [(44.332.21)] & eac Iohannes stefn geþwæreþ þam ylcan andgyte þam ilcan <wene>, se sæde, þæt he gesawe boc mid seofon inseglum geinseglode, & þæt nænig man wære gemeted wyrðe ne on heofonum ne on eorðan ne under eorðan, þe þa boc moste untynan & hire inseglu tobrecan þa boc swa þehhweþre Iohannes sæde, þæt heo wære æfter þon untyned þurh þone leon of Iudan cynne.

90 I wish to thank Toni Healey, Joan Holland and their colleagues in Toronto for allowing me to present their materials in this Appendix. I hope I have collected all relevant entries. The gloss CorpGl 2 (Hessels) D4.2 [(17.327)] Sigillum signum anuli is omitted because it does not contain any Old English words.

154

Jane Roberts

..(relating to period after Edward’s reign, so allowed to be adjoined) ChronE (Plummer) B17.9 [(1048.8)] Ða com Sparhafoc abbod be <wege> to him mid þæs cynges gewrite & insegle. to þan þet he hine hadian sceolde to biscop into Lundene. þa wiðcweð se arcebiscop. & cwæð þet se papa hit him forboden hæfde. .( used to authenticate that what is sealed has remained sealed) ..(marking a box) LS 34 (SevenSleepers) B3.3.34 [(310)] And hi ða twegen getreowfæste wæron: dydon þærrihte eallswa hi ær gemynton: eodon into ðam scræfe dearnunga onsundran, and þas halgan martyrrace eallswa heo gewearð on anum leadenum tabulan ealle mid stafon agrofon, and hi ðæt gewrit mid twam sylfrenan inseglum on anre teage geinsegledon and wið þa halgan ðærinne swiðe digollice ledon, and ðæs scræfes locstan hi wel fæste beclysdon, and him ðanon syððan hamweard gewendon. LS 34 (SevenSleepers) B3.3.34 [(693)] And mid þy þe hi in becomen, þa gemetton hi on þa swiðran hand ane teage, seo wæs geinsæglod mid twam sylfrenan insæglan þe þa twægen getreowfæste menn ðærinne ledon þa Decius se casere het þæt scræf forwyrcan, swa we ær beforan rehton, þæt þa insægla wæron eft to swutelunge hwæt man þærinne funde, þonne se tima gewurðe eallswa God wolde þæt þa gewurðan sceolde. ..(marking a ‘loc’) GD 1 (C) B9.5.2 [(9.59.1)] & þa þa he geseah, þæt þa þearfan genoh hæfdon, he het þone cniht stigan of þære wintreddan & beleac þæt winern & asette his agen insegl on þæt loc. GD 1 (H) B9.5.8.2 [(9.59.1)] & þa þa he geseah, þæt þa þearfan genoh hæfdon, þa het he þone cnapan stigan nyðer of þære treddan & beleac þæt winern & asette his agen insegl on þæt loc & forlet hit swa belocen & him sona to cyrcean gehwearf. ..(marking the bandaging over an arm put to torture) LawOrdal B14.33 [(5.2)] & inseglige man þa hand, & <sece> man ofer þæne þriddan dæg, swa hwæðer swa heo beo ful swa clæne binnan þam insegle.

What did Anglo-Saxon seals seal when?

155

II a seal, an engraved stamp of hard material to make an impression upon wax, &c.91 GDPref and 4 (C) B9.5.6 [(16.283.20)] Ac swa swa hit fulloft gelimpeð, þæt þa þe eallinga men wenað, þæt hi fullmedome syn, hi þonne gyt hwæthugu fullfremedra wisena nabbað beforan þam eagum þæs hean cræftigan, swa swa hit eac fulloft gelimpeþ, þæt we unlæredan, þonne we sceawiað þa inseglu & onlicnessa, þe þonne gyt fullfremedlice ne beoð agrafene, we þonne eac heriað þa, swylce hi syn fullmedomlice geworhte. Ch IWm (Douglas 7) B15.1.191 [(60)] And mid ures insigeles onðryste to þam þat þara foresprecenan stowe freodom on ecnysse þuyrhwunie. III ambiguous between I and II ..(relating to period before Edward's reign, so disputed) Solil 1 B9.4.2 [(24.1)] Geþenc nu gyf ðines hlafordes ærendgewrit and hys insegel to ðe cymð, hwæðer þu mæge cweðan þæt ðu hine be ðam ongytan ne mægæ, ne hys willan þæron gecnawan ne mæge. Ch 1454 (Rob 66) B15.5.14 [(9)] Þa sende se cyning be æluere abbude his insegel to þam gemote æt Cwicelmeshlæwe & grette ealle þa witan þe þær gesomnode wæron. Ch 1445 (HarmD 18) B15.5.7 [(60)] Ða he ðines fæder lic & brohte insigle to me. Ch 1445 (HarmD 18) B15.5.7 [(61)] Ða ageaf ic ðæt insigle ðe. Ch 1456 (Rob 69) B15.5.16 [(9)] Þa ða him seo talu cuð wæs þa sende he gewrit & his insegl to þam arcebisceope ælfrice & bead him þæt he & hys þegenas on East Cent & on West Cent hy onriht gesemdon be ontale & be oftale.

..(marking Christ’s tomb)

91

PrudGl 1 (Meritt) C94.1 [(371)] sigillorum i, statuarum anlicnyssa. is not relevant.

156

Jane Roberts

ÆCHom I, 15 B1.1.17 [(299.13)] Hi ða ferdon to: & mearcodon ða þruh mid insegle. & besæton þa birgine. ..(marking the lock of a prison door) Nic (A) B8.5.2.1 [(13.2.14)] And we gehyrdon þæt secgan þæt Ioseph þe ðæs hælendes lychaman bebyrigde, þæt ge hyne on fæstre clusan beclysdon and þæt loc myd insegle geinseglodon and þa ge þær to comon þa ne fundon ge hyne na. ..(in glosses)92 ÆGl B1.9.2 [(316.11)] sigillum insegel. AntGl 2 (Kindschi) D1.2 [(736)] Sigillum bulla insegel. IV fig. a seal that protects, sign, token LS 10.1 (Guth) B3.3.10.1 [(1.11)] Ða se tima com, þæt heo þæt bearn cennan scolde, þa sæmninga com tacn of heofenum, and þæt tacn swytelice mid inseglum beclysde. .(referring specifically to the protecting sign/seal of the cross) LS 10.1 (Guth) B3.3.10.1 [(2.61)] Mid þy þære nihte þystro gewiton, and hit dæg wæs, þa aras he and hine sylfne getacnode insegle Cristes rode. LS 13 (Machutus) B3.3.13 [(28r.1)] Þa se halga Machu hine geinseglade mid insegle þære halgan rode & ætforan his leorningcnihtum eode ongean þa nædran & he aþænede & alegde his stæf þe he him on handa hæfde ofer þæs dracon sweoran & him to cwæþ: Far nu þu manna ne þu n fre eft to þysum iglande ne gecer, ne þu nanum cristenum men ne derige. .(in glosses) AldV 1 (Goossens) C31.1 [(4058)] signaculo insegle mercelse.

92 Does the inclusion of ‘hring’ mean that ClGl 1 (Stryker) D8.1 [(2473)] Fibula sigel áhringe, fifele are comparable? Or CorpGl 2 (Hessels) D4.2 [(6.170)] Fibula hringe sigl. But CorpGl 2 (Hessels) D4.2 [(17.327)] Sigillum signum anuli is irrelevant. See also Heslop (1980), op. cit., p. 5 and fn. 26.

What did Anglo-Saxon seals seal when?

157

AldV 13.1 (Nap) C31.13.1 [(4177)] signaculo mercel, insegle.

Abbreviations Sawyer = P. H. Sawyer, Anglo-Saxon Charters: An Annotated List and Bibliography, Royal Historical Society Guides and Handbooks 8, London: Royal Historical Society, 1968.

This page intentionally left blank

Notes on the medical vocabulary of John Keats Jeremy J. Smith Reconstructing the registers and domains of the past One of the great tasks of English historical linguistics has been the reconstruction not just of the forms of past states of the language – its phonology and writing-systems, its grammar and its vocabulary – but also of how these forms functioned in terms of meaning, i.e. semantics. For many years this task was comparatively neglected; that it is now being seriously addressed again is largely due to the labours of the team working on the Historical Thesaurus of English, the great project to which Christian Kay has devoted most of her academic working life. These notes are offered as a small response to these labours. It has long been known that the meanings of words change over time, and cultural historians have taken advantage of this knowledge in offering new interpretations of past texts; a classic example of such an endeavour is C.S.Lewis’s Studies in Words, which traces the changing meaning of individual words to show how literary writers from the past used these words differently from those of the present day.1 However, one of the underpinning principles of the Thesaurus has been that not only the meanings of words but also the configuration of semantic categories have changed over time. Thus the languages of religion, or of pain, or of farming, or of science, differ from generation to generation. The development of the Thesaurus has made it possible to trace these shifts in a much more coherent and non-impressionistic way than has hitherto been achieved. As a consequence, it is now possible to show how complete registers of language, not just individual words, operate in writings from the past. In this paper, an attempt is 1 C. S. Lewis, Studies in Words, Cambridge: Cambridge University Press, 2nd edition, 1967.

160

Jeremy J. Smith

made, using the working files of the Thesaurus, to demonstrate how this knowledge enhances our interpretation of the poetry of a major literary figure from the beginning of the nineteenth century: John Keats (1795-1821). Keats and medicine: a biographical sketch That the life and writings of John Keats are profoundly intertwined has long been assumed by critics; as Robert Gittings has put it, Keats “could, and did, transmute almost any experience into poetry. This is why no part of Keats’s life should be neglected, and every incident, once truly recorded, may have immense value in interpreting his poetry”.2 Keats’s biography has always played an important role in interpretations of his verse. The facts of Keats’s short life may be briefly rehearsed. The child of ‘rising’ parents in the harsh, competitive and uncertain world of Regency London, Keats was early an orphan. His guardians, on the basis of small legacies, supported Keats in a medical training which resulted in his being awarded the Licentiate of the Society of Apothecaries in 1816. For many years it was fashionable for literary historians to decry this qualification, but Gittings and others have shown that this was a real professional attainment, the result of hard application and considerable commitment in a highly competitive environment.3 However, Keats, in the face of considerable opposition from his guardian Richard Abbey, abandoned the earning potential of a medical career for poetry, funded by small sums of money derived from a minor inheritance. His poetic career was brief, for he died of tuberculosis, in Rome, in 1821. This change from medicine to poetry has the potential to be seen as the comprehensive abandonment of one career for another; certainly Abbey, who accused Keats of being a “Silly Boy” for his decision, saw it as such.4 However, it is now generally accepted that Keats saw poetry and medicine – and, for that matter, politics – as closely related

2

R. Gittings, John Keats, London: Heinemann, 1968, pp. 627-28. Op. cit., pp. 116-17. 4 Op cit., p. 153. 3

The medical vocabulary of John Keats

161

disciplines.5 Given the current fashion for ‘holistic’ medicine it is perhaps easy for modern readers to appreciate the way in which distinct parts of life can be connected. Belief in the interconnectedness between the soul, the body and human social organisation is nothing new; it was a commonplace of Enlightenment and Romantic medical thought. The dominant figures in this kind of medicine were German-speaking – including the poet and polymath Goethe, or the doctor-dramatist Schiller – and Germanic Naturphilosophie insisted that “nature embodied a transcendental unity of plan”.6 As Roy Porter has comprehensively traced, throughout human history it has been usual for doctors to consider the “whole being” in making a diagnosis; the “specialisation” of scientific medicine, which “meant to get its magic bullets on target,”7 can in some ways be seen as a Victorian/twentieth-century aberration, deriving from notions which arose in post-revolutionary Paris. Keats’s teachers were on the cusp between Enlightenment and “scientific” thinking: a compromise-position which may be seen as a peculiarly English contribution to the evolution of medicine. The dominant figure in the poet’s medical education, Astley Cooper of Guy’s Hospital, had worked in revolutionary Paris, but had – in a not uncommon trajectory – subsequently redirected his earlier “democratic” opinions in favour of a distinctive, radical approach to medicine, which may be characterised as “humanitarianism”; the word was first used with its modern sense of compassion for the unfortunate in the period following the battle of Waterloo.8 Cooper and his associates emphasised that “the practitioner of ‘first rank’ must know ‘more than Medicine’”.9 Keats seems to have taken this advice to heart. After all, Apollo, who plays a key role in Hyperion, his poem about transition, was – according to Keats’s beloved Lempriere – the god of both medicine and poetry, and there is much evidence that the poet saw a close 5

Cf., e.g., N. Roe, John Keats and the Culture of Dissent, Cambridge: Cambridge University Press, 1997. 6 Roy Porter, The Greatest Benefit to Mankind, Cambridge: Cambridge University Press, 1997, p. 24. 7 Op. cit., p. 305. 8 Hermione de Almeida, Romantic Medicine and John Keats, New York: Oxford University Press, 1991, p. 35. 9 Op cit., p. 36.

162

Jeremy J. Smith

connexion between the two disciplines. As Hermione de Almeida has shown in detail, Romantic medicine informed every aspect of Keats’s writing: the poet, she writes, “used his education to exemplary artistic effect, the perceptual advances of Romantic medicine reflected in his poetry”.10 Andrew Motion writes similarly: “While we read his work and notice its consoling luxuries, its healthy and unhealthy airs, its medicinal flowers, its systems of nervous sensibilities, its working brains, its ethereal flights, its emphasis on “sensation”, its fluctuating temperatures, its marvellous chemical transformations, and its restorative sleeps, we realise that these things are not just incidental details, but the components of a selfless and moral imagination.”11 Keats never gave away his medical books, he kept open the option of returning to a medical career, and in May 1819 he even considered signing on as a ship’s surgeon in an East Indiaman, thus combining medicine and exploration.12 The language of Romantic medicine The integration of medical thinking in Keats’s poetry is, therefore, everywhere evident, but it is sometimes disguised for modern readers by the fact that medical language – like all forms of living language – is subject to change; the “scientific currency” of Keats’s day is different from our own.13 As a result, many words which we no longer consider to be primarily medical in denotation or connotation had medical meanings in Keats’s time. Although numerous scholars – notably de Almeida – have traced many of these meanings, the development of the Historical Thesaurus project offers the possibility of a comprehensive survey of specialised vocabularies at various stages in the history of English, and the assembly of terms in notional 10

Op. cit., p. 7. Andrew Motion, Keats, London: Faber, 1997, p. 131 12 Op. cit., p. 132. Keats was, of course, not the only Romantic poet to be interested in the artistic possibilities of health and sickness. Coleridge, for instance, was much concerned with questions of sickness, and the figure of the nurse is an important part of his poetic vocabulary. See R. Holmes, Coleridge: Early Visions, London: Harper Collins, 1998, p.15, footnote, and references there cited. A classic discussion of the literary uses of illness is Susan Sontag, Illness as Metaphor, New York: Farrar, Straus and Giroux, 1978. 13 De Almeida, op. cit., p. 13. 11

The medical vocabulary of John Keats

163

categories makes it possible for the development of new and enhanced interpretations of well-known texts. The files in the Historical Thesaurus database for the fields CURE and HEALTH, which can be time-limited, yield a fairly sizeable listing of the medical vocabulary in Keats’s day. It is noticeable that many words are included in these fields which would not be considered primarily “medical” in Present-Day English. A selection follows: CURE: balm, bland, broach, cot, discuss, elegant, explore, gentle, rally, repel, resolve, revel, simple, solicit, steel, stuff, subdue, tap, touch, translate, union, watch with HEALTH: burning, fanatic, fine, fond, foul, gasping, gazeless, indifferent, indolent, melancholic, panting, passion, rage, rude, shadow, spin, stifle, swirl, thing, touch, unkind, vex, visionless, wandering, wither, wrong Even a cursory survey of Keats’s verse demonstrates how common these words are in his writing, and how a grasp of their medical meanings can inform our reading of his verse. To demonstrate the point, I propose to examine the presence of a selection of these forms in two poems from the beginning and end of his brief poetic career: Endymion, which he published in 1818, and The Fall of Hyperion, which was abandoned in the autumn of 1819 and not published in the poet’s lifetime. The later poem tells how the sun-titan, Hyperion, is to be replaced by the god Apollo. The Fall is a substantial reworking of an earlier, more substantial version of the myth, Hyperion, which appeared in the major volume of poems which Keats published in 1820. Major additions in The Fall in comparison with the published version include an initial dream-vision, in which the first-person narrator describes an encounter with the priestess Moneta. Explicitly medical situations are to be found throughout The Fall. The dreamer collapses in a “cloudy swoon” (canto I, line 5; all citations from Keats’s verse are from Barnard 1973). The East wind is “sickening” (I.7); “flesh, near cousin to the common dust,/Will parch for lack of nutriment” (I.109-110); “a palsied chill” afflicts the dreamer (I.122); Moneta seems to him to suffer from “ferments”

Jeremy J. Smith

164

(I.290) in the brain; she in turn accuses the dreamer of being a “fever of thyself” (I.169) as opposed to a poet, who “is a sage,/A humanist, physician to all men” (I.189-190); the poet, she says, “pours out a balm upon the world” (I.201). Moneta’s face is “wan” (I.256), “Not pined by human sorrows, but bright-blanched/By an immortal sickness which kills not” (I.257-258). Saturn describes how the fallen titan has been changed “into a shaking palsy” (I.426), suffering from “the pain of feebleness” (I.429); Hyperion himself, though still “snuff[ing] the incense teeming up/From man to the sun’s God” (II.16-17), suffers from “horrors, portioned to a giant nerve” (II.23) which make him “ache” (II.24); his “ample palate” (II.32) tastes “poisonous brass and metals sick” (II.33). More subtly, Moneta’s reference to “an electral changing misery” (I.246), with its reference to the role of electricity in the constitution of the body, relates to a contemporary debate about the basis of neurophysiology; another Romantic text inspired by this notion was, of course, Mary Shelley’s Frankenstein (1818).14 The medical underpinning of the poem, therefore, is fairly clear, but our grasp of its comprehensiveness is enhanced still further if we note how apparently “non-medical” vocabulary has a medical meaning in Keats’s day. “Burning”, for instance, is perhaps of particular biographical interest. Keats uses the word a good deal in his poetry; in The Fall it appears as follows, in the Dreamer’s narrative: For by my burning brain I measured sure Her silver seasons shedded on the night, And every day by day methought I grew More gaunt and ghostly. (I.393-396) One of the meanings of the word, current from the late medieval period to the middle of the nineteenth century, is medical, as follows: “Heat attendant upon disease or a serpent’s bite; the disease itself; esp. erysipelas or St. Anthony’s fire, and venereal disease” (OED). The latest citation in this sense given by OED is from a medical glossary of 1860: “Burning, an old English name for Gonorrhoea”. In its OED 14

See R. Porter, op. cit., p. 252.

The medical vocabulary of John Keats

165

citations it is quite frequently collocated with “rage”, which falls into the same Thesaurus category, and it is therefore perhaps no surprise that, within a few lines, we read how the Dreamer “Gasping with despair/Of change, hour after hour I cursed myself” (I.398-399). The word must have had very definite medical connotations for Keats, since – as is now generally accepted by scholars – he suffered from gonorrhoea, contracted most probably during his visit to Oxford in 1817.15 Contemporary medical opinion required, as a remedy for this condition, the sparing use of mercury, usually administered either by rubbing on the skin or in pill form. The reference to how Hyperion’s “ample palate” (II.32) tasted “poisonous brass and metals sick” (II.33) may be connected to this experience. Keats seems – in accordance with contemporary, erroneous medical views on the connexion between the diseases – to have been worried that an ulcerated throat, contracted during his visit to Scotland in the summer of 1818, may have been a sign that his gonorrhoea was developing into the much more serious disease of syphilis. As a result, he took fairly heavy doses of mercury which may have made him especially susceptible to the tuberculosis of which he eventually died.16 A more subtle example is the word “fanatic”, with which the poem opens: Fanatics have their dreams, wherewith they weave A paradise for a sect, the savage too From forth the loftiest fashion of his sleep Guesses at Heaven; pity these have not Traced upon vellum or wild Indian leaf The shadows of melodious utterance. (I.1-6) The word “fanatic” is of course still current in Present-Day English, with the meaning “a person whose enthusiasm or zeal for something is extreme or beyond normal limits”; the word derives from Latin fanaticus “belonging to a temple, hence, inspired by a god, frenzied, from fanum temple” (CED). The word, first recorded in English in the early sixteenth century, developed religious connotations in the 15 16

See Gittings, op. cit., p. 236 and Motion, op. cit., pp. 196-97. Gittings, op. cit., p. 343.

Jeremy J. Smith

166

seventeenth century and took on a primarily religious denotation in the eighteenth century; this is the overt meaning which Keats offers here, supported by the reference to “A paradise for a sect”. However, there was a distinct, medical meaning still current in Keats’s time: ‘a mad person’ (not necessarily religious). OED cites the following reference from the Medical Journal of 1806: “Dr. G[all] gave .. hints how to treat fanatics, by using topical remedies and poultices.” According to Keats, fanatics, like savages, have not the consolations available to the literate: “bare of laurel they live, dream, and die” (I.7). “Laurel”, of course, is a reference to the traditional laurel-crown of the poet, both an image and an actual object current in the house of the poet’s friend and early patron, Leigh Hunt.17 The consolation, then, is “Poesy [which] alone can tell her dreams” (I.8). The poet is explicitly contrasted with the fanatic (I.17), and this comparison is developed throughout Moneta’s speech: “The poet and the dreamer are distinct, Diverse, sheer opposite, antipodes. The one pours out a balm upon the world, The other vexes it.” (I.199-202) In other words, the poet offers the “topical remedy” or cure – “balm” – whereas the fanatic simply “vexes”. The Historical Thesaurus classifies both “balm” and “vex” as medical terms in Keats’s day; poets and fanatics may therefore be seen as opposed in a quite specific, medical sense. That the word “fanatic” has a medical sense in Keats’s day clarifies why the poem begins with the reference. At first sight, medical situations are rather less common in Keats’s first major poem, Endymion. Sporadic metaphors and similes hint at a medical context, e.g., “nature’s wonders pulsed tenfold” (book I, line 105); Peona, Endymion’s sister and confidante, has healing powers “like some midnight spirit-nurse” (I.413) which allow Endymion’s “brain” to become “healthier” (I.465); “atomies” “buzz about our slumbers, like brain-flies,/Leaving us fancy-sick” (I.851853; see also “brain-sick” II.43). The speech of the aged Glaucus in Book III plays with images of mortality (“remorseless as an infant’s 17

Gittings, op. cit., p. 145.

The medical vocabulary of John Keats

167

bier” III.520), describing Circe as a kind of anti-physician inflicting sickness: “soon these limbs became/Gaunt, withered, sapless, feeble, cramped, and lame” (III.637-638); the subsequent resuscitation of the nymph Scylla clearly has a medical parallel. It has been noted, too, that Keats used “ethereal” in the poem in a quite definite scientific sense.18 Moreover, in a discussion of the poem in his letters Keats refers, perhaps significantly, to “Pleasure’s Thermometer”,19 although the medical use of the thermometer was not current in his own day, possibly because the instrument had not become appropriately refined for clinical use.20 However, the following words, which the files of the Thesaurus classify as having medical meanings in Keats’s day but have a much more general currency in Present-Day English, are all to be found in the poem: bland, explore, gentle, resolve, simple, steel, stuff, touch; burning, fine, fond, gasp, gaze, panting, passion, rage, shadow, stifle. One of the most commonly employed of these words in Endymion is “touch”, which occurs no fewer than twenty-one times and undoubtedly contributes to perception that the poem is particularly concerned with sensuality; as John Locke put it, touch is “a sense spread over the whole body, tho’ it be most eminently placed in the ends of the fingers” (Elements of Natural Philosophy 1754 edition, xi.50; cited OED sv. TOUCH n.). According to the OED citations, the word “touch” had primary medical meanings during the eighteenth and nineteenth centuries. Both noun and verb were used in particular with reference to gynaecological examination, e.g., “Upon touching I found the os uteri a little more dilated” (1754-1764), “[Touch:] Term for the examination of the womb, or mouth and neck of the womb” (1860). Intriguingly, the word seems also to have used in the phrase “to touch the gums, to induce salivation, as by the use of mercury”; by 1893, this phrase seems to have become proverbial: “The patient should be brought slightly .. under its [i.e. mercury’s] influence, so as just ‘to touch the gums’ as the phrase is.” Keats was working on Endymion at the time he seems to have caught

18 J. Barnard, ed, John Keats: The Complete Poems, Harmondsworth: Penguin, 1973, p. 564, discussing I. 298; cf. also II. 672. 19 Op cit., pp. 567-68. 20 Porter, op. cit., pp. 344-46.

168

Jeremy J. Smith

gonorrhoea, though it should be pointed out that the poem was well under way before he visited Oxford. One of the most important uses of the word occurs in Book I of the poem, when Endymion, in conversation with his sister Peona, develops an argument about the human aspiration for the spiritual life: Behold The clear religion of heaven! Fold A rose leaf round thy finger’s taperness, And soothe thy lips; hist, when the airy stress Of music’s kiss impregnates the free winds, And with a sympathetic touch unbinds Aeolian magic from their lucid wombs; Then old songs waken from enclouded tombs; Old ditties sigh above their father’s grave; Ghosts of melodious prophesyings rave Round every spot where trod Apollo’s foot … (I.780-790) The phrase “sympathetic touch” has a quite specific medical meaning in Keats’s time. We have already noted the medical meanings of “touch”, and the gynaecological reference point is obviously relevant to the image Keats is developing (cf. “impregnates”, “wombs”). “Sympathetic” is given both physiological/pathological and anatomical meanings by OED, with citations dating between 1728 and 1888. The definitions offered by OED are respectively as follows: “applied to a condition, action, or disorder induced in a person, or in an organ or part of the body, by a similar or corresponding one in another’; “designating one of the two great nerve-systems in vertebrates, consisting of a double chain of ganglia, with connecting fibres, along the vertebral column, giving off branches and plexuses which supply the viscera and blood-vessels and maintain relations between their various activities; belonging to or forming part of this system.” Music becomes the midwife, whereby human invention harnesses natural sounds. That “touch” also has, according to OED, a contemporary musical meaning adds to the density of reference which Keats is trying to achieve; Keats himself collocates "touch" and "sound" in:

The medical vocabulary of John Keats

169

or ye, whose precious charge Nibble their fill at ocean's very marge, Whose mellow reeds are touched with sounds forlorn By the dim echoes of old Triton's horn (I.203-206) and the culminating reference to Apollo (god of music, as well as god of poetry and medicine) is significant. The reference to an Aeolian harp is a characteristically Romantic gesture, which Keats had already used in his early Ode to (of course) Apollo of 1815 (see also, for instance, Coleridge’s Effusion XXXV: The Eolian Harp). Conclusion The examples just given are all examples of a traditional artistic activity: inventio, which according to the classical rhetoricians meant harnessing materials to hand for creative purposes (a parallel may be drawn with “found art”, as demonstrated inter alia by Picasso’s famous – notorious? – linking of a bicycle saddle and handlebars to create a bull’s head). However, in order to grasp how these inventive processes worked, it is necessary to have a clear idea of the materials available from which authors could make their choices. Although the analyses offered above are hardly radical in offering new readings of the poems, it may be hoped that they provide sufficient evidence to show how the Historical Thesaurus project will allow literary as well as linguistic historians to formulate new research questions, and to develop new insights and an enhanced appreciation of authors from earlier periods. Abbreviations CED= Collins English Dictionary (2003 edition) OED = Oxford English Dictionary (http://www.oed.com)

This page intentionally left blank

‘Tell her to shut her moof’: the role of the lexicon in TH-fronting in Glaswegian Jane Stuart-Smith & Claire Timmins Introduction The discovery of TH-fronting, or the use of [f] for /th/ in words such as think and tooth, as a feature in the speech of working-class Glaswegian adolescents provoked both academic and general interest when it was reported in 1998.1 The finding causes an apparent difficulty for models of language change which hinge on dialect contact associated with geographical mobility, since the Glaswegian speakers who use [f] most in the 1997 sample are also those with the lowest geographical mobility. In fact, our results seem to reflect those of other, earlier studies elsewhere in the UK, which also show relatively less-mobile working-class speakers using TH-fronting.2 Identifying the sociolinguistic mechanisms underlying the diffusion of TH-fronting have been the primary points of interest for this variable, and in particular, whether the broadcast media might or might not play a role.3 The role of television as a factor in this and other changes spreading rapidly through urban accents in the UK is now being investigated systematically in a three-year project based in Glasgow. But there are other aspects of TH-fronting which also deserve discussion, and which have received much less attention. Here we focus briefly on the role of the lexicon in the spread of TH-fronting in Glaswegian and in so doing, we uncover the dynamic processes 1

Rob Corbidge, “It’s the way you tell ‘em, me old Jock sparrer”, Sunday Times, 29/03/98; J. Stuart-Smith, “Glasgow: Accent and voice quality”, in Paul Foulkes and Gerard Docherty (eds.), Urban Voices, London: Arnold, 1999, pp. 203-22. 2 For a recent review, see, e.g., P. Kerswill, “Dialect levelling and geographical diffusion in British English’, in David Britain and Jenny Cheshire (eds), Social Dialectology: In honour of Peter Trudgill, Amsterdam: Benjamins, 2003, pp. 223-43. 3 E.g., P. Trudgill, Dialects in Contact, Oxford: Blackwell, pp. 54-55.

172

Stuart-Smith & Timmins

involved in the diffusion of an innovatory variant into a traditional dialect, and at the same time we also discover more about the fate of traditional Scots variants in modern Urban Scots. The term TH-fronting was coined by Wells to refer to the ‘replacement of the dental fricatives [th, dh] by labiodentals [f] and [v] respectively’ in Cockney and other London accents.4 Wells also notes that /dh/ may not show [v] in word-initial position, and this constraint, together with other factors particular to the change in Glaswegian, has led us to use TH-fronting to refer specifically to the use of [f] for /th/. The first published reports of TH-fronting in Glaswegian date to the early 1980s and are given by Macafee.5 The expected local pronunciation for /th/ in Scots in Glasgow and more generally across the Central Belt is [h], in word-initial and intervocalic position, and before /r/, as in e.g. thing, nothing, three.6 This feature is, according to Johnston, ‘becoming more common over time there, though still variable’; in word-final position, /th/ is expected. The existence of local [h] complicates the diffusion of [f]. Unlike in other accents of English, where [f] competes solely with [th], in Glasgow adding [f] creates a three-way system of phonetic variants, [th]/[f]/[h]. These in turn divide into two categories indexing social functions. [th] is typical of supralocal and local educated norms (the regional standard), while [h] and [f] mark local non-standard, nonestablishment, ‘non-posh’ norms. [h] is well-known in Glasgow as a vernacular variant, while [f] has only recently risen above the level of social consciousness.7 Lexical diffusion, or the gradual spread of sound changes across the lexicon of a language, is now a well-established theoretical component in understanding sound change both by historical linguists

4

John Wells, “Scotland”, in: Accents of English, Cambridge: Cambridge University Press, 1982, p. 328. 5 C. Macafee, Varieties of English around the World: Glasgow, Amsterdam: Benjamins, 1983, p. 34, n. 26; and C. Macafee, Traditional dialect in the Modern World: A Glasgow Case Study, Frankfurt: Peter Lang, 1994, p. 29. 6 P. Johnston, “Regional Variation”, in Charles Jones (ed.), The Edinburgh History of Scots, Edinburgh: Edinburgh University Press, 1997, pp. 433-513; also Macafee (1983), op. cit., p. 33, n. 23. Stopping of /th/ is also found word-initially; see Johnston, op. cit., p. 506. This variant, and other rare variants will not be discussed here. 7 See, e.g., Hugh Reilly, “In need of some speech ferapy”, The Scotsman, 25/09/02.

'Tell her to shut her moof'

173

and sociolinguists working on language variation and change.8 Given that TH-fronting seems to be a relatively recent innovation in Glasgow, we might expect to find evidence for lexical diffusion for [f], and our results confirm this. However, what is perhaps more interesting is the way that the existing and particular lexical distribution of [h] imposes special constraints on the progress of THfronting in Glaswegian. The data: The Glasgow Speech Project Our discussion is based on evidence for TH-fronting derived from the analysis of two speech corpora collected in 1997 and 2003, which together and with other studies, comprise a long-term research programme into aspects of language variation and change in Glaswegian which we call The Glasgow Speech Project. The 1997 corpus comprises read and spontaneous speech from 32 individuals stratified according to age, gender and social class.9 The 2003 corpus contains read and spontaneous speech recorded from 36 working-class adolescents divided into three age groups (1: 10/11 years; 2: 12/13 years; 3: 14/15 years). Here we concentrate on the results from spontaneous speech alone, since /h/ - like other stigmatized features of Glaswegian vernacular, such as oot (/ut/) for out – did not occur in read speech. All instances of /th/ for each speaker were identified and auditorily transcribed from the 1997 corpus. These data were then statistically analysed for the effects of AGE, CLASS and GENDER using logratio linear modelling and then Bonferroni-corrected t-tests. The first 35 instances of /th/ were transcribed from the 2003 corpus and then tested for the effects of AGE and GENDER using chi-square tests, with the level of significance set to p <0.05. The majority of the 1997 corpus and the whole of the 2003 corpus were collected by the same fieldworker, Claire Timmins. The nature of the spontaneous speech recording was the same for both 8

E.g., W. Wang, “Competing changes as a cause of residue”, Language 45 (1969), 925; A. McMahon, Understanding Language Change, Cambridge: Cambridge University Press, 1994, 50-56; W. Labov, Principles of Linguistic Change: Internal Factors, Oxford: Blackwell, 1994. 9 More details are given in J. Stuart-Smith, “Glasgow: Accent and voice quality” in: Paul Foulkes and Gerard Docherty (eds.), Urban Voices, London: Arnold, 203-222.

Stuart-Smith & Timmins

174

corpora, as was the area, and even the school, of the working-class adolescent sample. Thus comparison between speech of the 8 working-class adolescents of 1997 with that of the 36 speakers of the same age and profile in 2003, though numerically uneven, in other respects provides good evidence for a real time study of TH-fronting in these speakers. Sound change in progress: TH-fronting The results for 1997 show clear evidence for TH-fronting as a feature of one group of speakers, working-class adolescents (Table 1), and the significant effect of AGE confirms this as an apparent time change. Compared to Macafee’s mention of occasional TH-fronting this also constitutes evidence for real time change. Note however, that [h] is still numerically predominant. The variation for /th/ is spread across [th], [h] and [f]. Table 1. Distribution of phonetic variation for (th) for all speakers in spontaneous speech in 1997 corpus.

MCOF MCOM MCYF MCYM WCOF WCOM WCYF WCYM

[th]

[f]

[thf]

[t]

[s]

[h]

98.40 97.60 93.33 88.89 81.68 74.39 21.21 33.62

0 0 0 0 0 0 32.90 22.41

0 0 0 0 0.26 0.81 0.00 0.86

0.80 1.20 0.89 4.58 0.26 1.63 0.43 0.86

0 0 0 0 0 0 0.43 0

0.27 0.60 5.78 6.21 16.54 12.60 44.59 41.38

[m] 0.53 0.60 0 0.33 1.27 10.57 0.43 0.86

Note n = 2021. All factors and all their interactions were significant. Across variants working-class adolescents are significantly different from their middle-class counterparts. Working-class girls are polarized from all middleclass speakers in their high use of [f] and low use of [th].

'Tell her to shut her moof'

175

Table 2. Main phonetic variants for (th) for all speakers in spontaneous speech in 2003 corpus age/group

gender

[th]

[f]

1

f m f m f m

12.23 4.76 27.34 20.71 20.69 21.40

47.97 58.86 26.08 43.62 39.31 36.45

2 3

[h] 39.80 36.38 46.58 35.67 40.00 42.15

Note n = 1013 There is a significant effect of AGE, but not GENDER, with the youngest group using more [f] and less [th].

Consider now the results for 2003 (Table 2), which are only from working-class adolescents.10 The youngest group uses [f] the most, but at the expense of standard [th], not [h]. This reflects the overall pattern for all speakers. Adding [f] increases their potential repertoire for expressing a specific kind of local identity (‘us’ as ‘normal’, ‘Possil’/’Ruchill’ and so on) which is generated with an array of specific social and linguistic practices, and by rejecting features associated with perceived middle-class, establishment norms (‘them’ as e.g. ‘pure gay’, ‘Bearsden’, and so on).11 Statistical comparison of the two sets of results is highly significant, providing evidence for a real time change in progress: [f] is increasing, [h] is increasing slightly, and [th] is reducing. The role of position in the word Table 3 gives the distribution of main variants for /th/ according to position in the word in the two corpora.

10

Only results for the most frequent variants are presented for the 2003 data. J. Stuart-Smith, C. Timmins and F. Tweedie, “Conservation and innovation in a traditional dialect: L-vocalization in Glaswegian”, English World Wide (forthcoming). 11

Stuart-Smith & Timmins

176

Table 3. Distribution of [th], [f] and [h] variants for (th) in spontaneous speech in working-class adolescents according to position in the word in 1997 (n = 324) and in 2003 (n = 1013). 1997

2003

[th]

[f]

[h]

[th]

[f]

[h]

word-initial word-internal word-final

31.38 17.05 43.23

32.65 9.08 56.77

35.96 73.86 0

20.64 6.30 28.92

29.18 22.19 71.18

50.18 71.50 0

overall

30.28

32.62

35.79

18.62

40.85

40.56

Several findings are consistent from 1997 to 2003. [th] and [f] may occur in all positions in the word. [h] is predominantly found wordinternally, and is absent from word-final position. Statistical evidence for variation and change is found for word-initial position and wordinternal position, and in each case the change involves reduction of standard [th]. Word-initially, [h] was more common in 2003. Wordinternally [f] shows the main increase, with [h] being more or less stable. The dramatic rise in [f] in word-final position is statistically only a tendency, but still striking, and is clearly linked to the absence of [h]. It is clear from these results that position in the word is an important factor which both constrains and facilitates the diffusion of [f], and this is bound up with the behaviour of [h] and [th]. But position in the word results from a generalization of lexical distribution, and the following section shows the importance of individual words in this change, and particularly for the survival of [h]. The role of the lexicon The distribution of the main phonetic variants for /th/ in 1997 and 2003 are displayed in three pairs of tables, for word-initial, wordinternal and word-final position respectively; see Tables 4-6. The total number of words containing /th/ is always more in 2003 which reflects the larger sample of 36 speakers as opposed to 8 speakers in 1997.

'Tell her to shut her moof'

177

Table 4. Lexical distribution of main variants for /th/ in word-initial position in working-class adolescents in 1997 and 2003 in descending order of frequency.

2003

1997 word think thing thought three through thirty thinks thingmy thank third thirteen thematic thieving thinking throat thrown Thursday

[th] [f] 13 4 4 6 9 6 1 3 2

6 5 22 11 5 2

Total

45 19

64 30 26 17 14 8 7 5 2 2 2 1 1 1 1 1 1

6 2 2

2 1 1 1 1 1 1

[h]

word

[th] [f]

think thing three thought things through thingy thousand thingwy thinking thirty third Thursday thingmy thinks thank thingummy

23 11 12 8 5 10 6 6 2 4 7 2 3 5

thistle threw thin thirteen thingd thingwys throwing thanks theft therapist thingamajiggy thingy'd thinked thinness thong throat throw

5 1

1 3

[h] Total

22 155 200 8 59 78 32 1 45 33 41 1 30 36 21 31 9 15 8 14 11 13 3 3 10 2 9 6 8 5 8 2 7 1 5 6 5 1 2 4 4 4 2 2 2 2 1 1 1

1 1 1 1 1 1 1

4 4 3 3 2 2 2 1 1 1 1 1 1 1 1 1 1

Stuart-Smith & Timmins

178

thumb thumbs

1 1

1 1

Table 5. Lexical distribution of main variants for /th/ in word-internal position in working-class adolescents in 1997 and 2003 in descending order of frequency. 1997 word something nothing anything Catholics everything without Catherine menthol Strathmore

[th]

[f]

[h] Total

word

2 3 4 6

2 7 3 1 1

31 21 15

3 2 1 1

2003

2

4

36 31 23 7 6 3 2 1 1

[th]

[f]

[h] Total

something anything nothing healthy everything birthday without maths Samantha

5 3 3 3

3 1 1 39

153 46 40

3

10 9 5 4

Jonathon Dorothy athletics birthdays Blackthorn Catherine enthusiasm Grath’s months Southpark

1

1

3 2 1 1

1 1 1 1 1 1

21 1

161 50 44 42 21 14 9 5 5 4 2 1 1 1 1 1 1 1 1

'Tell her to shut her moof'

179

Table 6. Lexical distribution of main variants for /th/ in word-final position in working-class adolescents in 1997 and 2003 in descending order 1997 word maths month mouth fourth seventh underneath fifth

2003

[th] [f] Total 5 3 1 1 1

6 2 4 2 1 2

11 5 4 3 2 2 1

word teeth month fourth Smith twentieth eighteenth bath breath McGrath underneath fifth ninth sixth twelfth worth fifteenth Goth Grath length loudmouth Portsmouth sixteenth strength

[th] [f] Total 2 3 2 3 2 3 1 2

10 8 6 4 7 4 2

12 11 8 7 7 6 5

3 2 1 2 2

3 3 3 2 2 2 2 2 1 1 1 1 1 1 1 1

2 2 2 1 1 1 1 1 1 1 1

/th/ is most frequent in word-initial position, occurring in 17 words in 1997 and 36 words in 2003. The most striking result is the lexical restriction of [h] to think or thing, and forms derived from these stems. In 1997 this gives 4 words, think and far less frequently thinks, and the most frequent item thing and its less common derivative thingmy. In 2003, the think/thing set extends across 13 words, but again think and thing are the most common, and their derivatives, such as thinking and thingy respectively, are far less frequent. However this lexical restriction is countered by the relative frequency of thing and think which together and alone constitute around half of the total instances

180

Stuart-Smith & Timmins

of /th/ in both corpora. [f] is creeping into the think/thing group but is found far more often in other words, of which thought and three are the most frequent. The proportion of words in which [f] is used increases from 53% to 72% from 1997 to 2003, indicating a gradual spread through the lexicon, though from a substantial base. Few words retain [th]. Across both corpora only thank and thirteen resist a nonstandard variant; in 1997 thieving and thematic occur only once and with [th]. Overall we can see that in this position the prevalence of [h] in the think/thing set blocks [f] from being numerically frequent, but [f] can still achieve a decent quantity by occurring in the fairly wide base of other, less frequent, words which cannot take [h]. A similar, but stronger, pattern emerges in word-internal position. Here in both corpora [h] is even further restricted to four words, something, anything, nothing, everything, all of which contain – thing.12 But in this position these words are even more frequent, thus boosting the relative frequency of [h], and restricting the possibility of occurrence of [f]. [f] is found sporadically in the –thing set, but is more commonly observed in other words where [h] is not possible (Catholics is most common in 1997, healthy in 2003). This time there is no apparent change in lexical spread in this position. [f] occurs in slightly fewer words in 2003 (74%) than in 1997 (78%). It seems that word-internally the potential base of words which may not take [h], and so which could contain [f], is smaller, and this too acts to restrict the spread of [f]. Words that take only [th] across both corpora are rare and include two proper names (Strathmore, Blackthorn). In word-final position [f] is at last without competition. [h] does not and cannot occur in any words, a restriction which accords with the general phonotactic prohibition on /h/ word-finally in most accents of English. [f] appears to be spreading unchecked. In terms of lexical distribution [f] occurs in all but two words found in 1997. This high proportion (75%) increases further to 83% in 2003. It is difficult to see any pattern underlying words which here retain [th]. sixth and sixteenth both occur only once, and with [th], though the incidence of [f] in other ordinals suggest that this may well be conincidental. Otherwise only worth and the proper name Grath show [th]. In general it seems unlikely that there is any real lexical restriction on any word taking [f] word-finally, and [f] is diffusing freely in this 12

[h] in birthday is unusual.

'Tell her to shut her moof'

181

position. We note with interest how the availability of word-final [f] as a variant for /th/ permits our Glaswegian speakers to use nonstandard variants in all positions in the word. By 2003, [th] has been eroded to less than 20% of the total variation for /th/. These data confirm a process of lexical diffusion for TH-fronting in real time in word initial position. The good spread across the possible sites for /th/ in other positions in the word suggests that lexical diffusion is already quite advanced in these positions. In general, [h] occurs in few lexemes, but these tend to be very frequent. [f], on the other hand, occurs in far more words, but these tend to be less common. The effect of the spread of [f] at the moment seems to be a dramatic reduction in [th]; [h] is still prevalent. Concluding Remarks The findings of this brief analysis may be summarized as follows: 1. TH-fronting is evidenced both as an apparent time and a real time change in the speech of working-class adolescents in Glasgow. 2. TH-fronting is entering a system with a well-established local nonstandard variant, [h], which means that the process is not binary but involves three main elements. 3. The spread of TH-fronting is constrained by position in the word and the distribution of the existing variant [h] 4. TH-fronting is taking place via a process of lexical diffusion, which is now relatively far advanced 5. The local Scots variant [h] is lexically restricted but maintains a dominant position because the small number of words that take [h] remain very frequent. There can be no doubt that TH-fronting is now a well-established feature of Glaswegian vernacular, and our results can only give a glimpse of the full extent of this sound change. That [f] has to compete with existing local vernacular [h] makes TH-fronting in Glasgow more complex than elsewhere in the UK, though the situation is reminiscent of the incursion of T-glottalling into Tyneside

182

Stuart-Smith & Timmins

English where ‘glottal reinforcement’ in, e.g., water represents the local non-standard variant.13 The persistent frequency of Scots [h] for /th/ provides an effective brake on TH-fronting spreading unchecked throughout the lexicon. And this is despite the small amount of words that show [h], which in turn boil down to derivatives of think and thing. Analysis of the OUT vowel and Scots L-vocalization in Macafee’s 1984/5 dataset and the 1997 corpus revealed similar results. Scots forms occur in small numbers of words, but these continue to be very frequent, to the extent that there appears to be no quantitative reduction in the Scots variants of OUT or Scots L-vocalization over the past few decades.14 Thus Scots variants in these samples of Glaswegian vernacular appear to be remarkably vigorous despite continual lexical erosion of less frequently used items. However, there is a fundamental difference between the maintenance of Scots OUT/Scots L-vocalization and that of [h] for /th/. In the former, the alternation is essentially between a clearlyrecognized Scots/vernacular/non-standard variant and an alternative pronunciation closer to Scottish Standard English, albeit with a characteristic phonetic realization. In the latter case, the introduction of [f] now leaves [h] existing as one of two possible non-standard alternatives to [th], and whilst [h] is undoubtedly strong, [f] is now found occasionally in words formerly only occupied by [h]. It remains to be seen whether [h] will be able to withstand the onslaught of THfronting in the future. Acknowledgements It has been a pleasure to write this paper for Christian Kay for whom we have enormous affection, respect, and admiration. The work presented here was largely made possible because of her continual support and encouragement, especially at the early stages. 13 E.g., J. Milroy, L. Milroy, S. Hartley, and D. Walshaw, “Glottal stops and Tyneside glottalization: competing patterns of variation and change in British English”, Language Variation and Change, 6 (1994), pp. 327-57. 14 J. Stuart-Smith, “The phonology of Modern Urban Scots”, in John Corbett, J. Derrick McClure, and Jane Stuart-Smith (eds.), The Edinburgh Companion to Scots, Edinburgh: Edinburgh University Press, pp. 110-37.

'Tell her to shut her moof'

183

We would like to acknowledge financial support from the Leverhulme Trust (Grant no F/179/AX) for analysis of the 1997 corpus, and from the ESRC (grant no R000235797) for the collection and analysis of the 2003 data. We are also grateful to Graham Caie, John Corbett, Robert Lawson, Suzy Orr and Jeremy Smith for comments on an earlier draft of this paper.

This page intentionally left blank

Forces of change: are social and moral attitudes legible in this Historical Thesaurus classification? Louise Sylvester 1. Introduction In an extremely thought-provoking essay, Christian Kay reflects on the cognitive salience of the procedures employed by those classifying material for the Glasgow Historical Thesaurus of English (HTE).1 As Christian’s essay indicates, it is becoming increasingly clear that cognitive linguistics has thrown up a number of issues which are pertinent to diachronic semantic studies. Among these is Rastier’s suggestion that the project of diachronic semantics leads us to look at social evaluations to find the forces which shape and distort the lexicon.2 It seems to me that the HTE, a project which is at the interface between historical semantics and lexicography, may present a rich resource with which such questions may be addressed. Before the HTE is considered in this light, however, certain theories that have arisen within cognitive linguistics need to be taken into account. One of the important suggestions, for example, is that whereas the traditional view understands the role of language as mapping elements of the external world onto linguistic forms, cognitive linguists argue that a particular situation can be construed in different ways, and different ways of encoding a situation constitute different conceptualizations (Lee; Taylor; Croft and Cruse).3 Included within 1

C.J. Kay, “Historical semantics and historical lexicography: will the twain ever meet”, in J. Coleman and C.J. Kay (eds.), Lexicology and Semantics in English Historical Linguistics: Selected papers from the 4th G.L. Brook Symposium, Manchester, August 1998, Amsterdam: John Benjamins, 2000. 2 F. Rastier, “Cognitive semantics and diachronic semantics: the values and evolution of classes”, in A. Blank and P. Kock, (eds.), Historical Semantics and Cognition, Berlin: Mouton de Gruyter, 1999, p. 126. 3 See, for example, D. Lee, Cognitive Linguistics: an Introduction, Oxford: Oxford University Press, 2001; J.R. Taylor, “Cognitive semantics and structural semantics”,

186

Louise Sylvester

this notion of construal are the ideas that meaning, particularly the meaning of complex concepts, presupposes the existence of knowledge systems.4 This paper will consider these theories about meaning and conceptualization, and their implications for investigations of the data and definitions in lexicographical projects, focusing particularly on the HTE. I shall consider also whether it is possible to bring together different discourses which seek to understand and account for cultural and semantic shift. This will enable an investigation of how far changing conceptualizations are visible through the lens of the HTE classification, using data concerned with consent, coercion and resistance in connection with conceptualizations of sexual contracts across time. If the HTE is published electronically, as seems increasingly probable, it will be possible to search the thesaurus for all the occurrences of an individual word within the entire classification, perhaps through the kinds of searches that are found in the WordNet electronic lexical database, and thus to address questions about polysemy and diachronic change. From an onomasiological point of view, beginning, that is, with a concept rather than a word, it is still somewhat cumbersome to trace all the relevant parts of the data. Indeed, until the project reaches completion, it will not be possible to do so. I have chosen, therefore, to focus on a particular category, 02.05. Will/faculty of will, an extremely large category, containing 6531 records. I shall quote sufficiently extensively to offer a sense of the category, and will include larger sections of the category in two appendices.5 I have reformatted this material so that it matches the presentation in the Thesaurus of Old English (TOE), and I have left in a skeleton of the numbering system of the original classification, again matching the presentation in the TOE, in order to make the task of

in A. Blank and P. Kock, (eds.), Historical Semantics and Cognition, Berlin: Mouton de Gruyter, 1999, pp. 17-48. W. Croft and D.A. Cruse, Cognitive Linguistics, Cambridge: Cambridge University Press, 2004. 4 R.W. Langacker, Foundations of Cognitive Grammar. Vol. 1: Theoretical Prerequisites, Stanford: Stanford University Press, 1987, pp. 184-85. 5 I am grateful to Christian Kay and the staff at the Glasgow Historical Thesaurus of English project for giving me access to the data via a CD-ROM of the materials which had been classified at March 2004.

Forces of change: social and moral attitudes

187

reading the extracts as easy as possible.6 Bearing in mind Lee’s suggestion (op. cit., p.116) that semantic changes are the result of changes affecting the degree of salience of the various elements of the relevant frame, this paper will examine the evidence provided by the HTE classification in order to investigate how far shifts in social and moral attitudes are legible within the metalanguage and the vocabulary groupings of the HTE. 2. Structural and cognitive views on the lexicon If we are to discover the ways in which categories change over time we need to combine different kinds of knowledge about how meaning is structured. In order to accomplish this, encyclopaedic knowledge encompassing conceptual domains needs to be brought together with linguistic knowledge. Discussing cognitive semantics and structural semantics, Taylor affirms that changes in word meaning are likely to have as much to do with changes in background assumptions, that is domain-based knowledge configurations, as with designation (op. cit., p. 40). In linguistic terms, rather than through the lens of cultural and social history, the question is of particular relevance to the HTE. As is now quite widely known, the HTE will, when complete, offer a conceptual classification, in a hierarchical series of categories, of the entire lexis of English from the Anglo-Saxon period to the present day, including obsolete senses.7 In establishing the project Samuels’s intention was that it would provide a resource for the diachronic study of English and, in particular, the investigation of the ways in which the English language, as a system, regulates change. The HTE was established at a time when the thinking about language was predominantly structuralist and so language was conceptualized as a system within the foundational principles of the project. For scholars working now, within the frameworks of cognitive linguistics, making use of the HTE classification raises the question of how far it is possible to extract from it information about semantic shift which 6

J. Roberts and C.J. Kay with L. Grundy, A Thesaurus of Old English, London: King’s College London Medieval Studies, 1995. The style and status labels which appear in the HTE may be found at: http://www2.arts.gla.ac.uk/SESLL/EngLang/thesaur/abbrev1.htm 7 As proposed in M.L. Samuels, Linguistic Evolution with Special Reference to English, Cambridge: Cambridge University Press, 1972, p. 180.

188

Louise Sylvester

takes account of real-world knowledge. In other words, can we verify an understanding of semantic shift and its relation to changes in cultural understanding from the HTE records? Alternatively, it may be argued that the two aspects, cultural understanding and lexicalization of concepts, are not separable; if this is so, the HTE can only illustrate the ways in which semantic shift is culturally determined, and our interpretations are cultural constructions. Given the beginnings of the HTE project, we should continue to be aware that, as Taylor observes, structuralist semantics has proved an invaluable tool in explicating semantic change, one which points to changes in the ‘system’ (op. cit., p.37). I have argued elsewhere8 that discussions of semantic shift tend to focus on the change in meaning of a single word, traditionally looking at such processes as narrowing and widening, e.g. OE mete “food” > PDE meat; OE bridd “young bird” > PDE bird; or pejoration and amelioration, e.g. OE scitol “purgative” > PDE shit, or OE prætig “sly” > PDE pretty. The HTE is, of course, designed precisely to illustrate such changes to the lexicon, although it is perhaps most accessible through the headings and so tracking these shifts is most easily done from an onomasiological point of view. When I began to think about the idea of willing or unwilling consent, and the ways in which legal definitions from the medieval period onwards have sought to interpret this distinction.9 Christian directed me towards the category of terms 02.05. Will/faculty of will in the HTE classification. Looking at the nouns in the subcategory 02.05.03 Will/wish/inclination within 02.05 Will/faculty of will, we find evidence that some terms persist in the language with a particular meaning, while others drop out fairly rapidly, though they may, of course, resurface at another point in the classification: 02.05.03 Will/wish/inclination n: dom OE; gemede OE; myne OE; gewilnes OE; gewyrd OE; i-will
L.M. Sylvester, “Evidence for diachronic semantic change in the Historical Thesaurus of English: A cognitive linguistic approach”, in I. Moskowich-Spiegel Fandino and B. Crespo Garcia (eds.), English Historical Linguistics: An Atlantic View, Universidade da CoruÒa, 2004, pp. 189-221. 9 L.M. Sylvester, “The vocabulary of consent: classification and use”, in J. Coleman and C.J. Kay (eds.), op. cit., pp. 157-78.

Forces of change: social and moral attitudes

189

ywil(l)e c1200-a1275; talent a1300-1530; device a1300-1648; (one's) list a1300-1867; courage c1320-a1626; affection 1330-1877 ai; volunty c1330-1652; gree c1330-a1734; pleasance c1340-1530; one's pleasure c1368–; affect 1374-1626; fantasy c1374-a1618; disposing c1380-1611; wish 1390–; disposition 1393–; mind a1400/50–; plight c1400-1726; volente/volante 1418/20-1525; pleasing c1430-1527; fancy 1465–; hest 1500/20-1845; cue 1565-1851; month's mind 1580-1826 ob ex dl; disposedness 1583-1876; humo(u)r 1590–; wouldings a1640-a1758; beneplacit 1643-1658; inclination 1647–; wouldingness a1660; beneplaciture 1662; good-liking 1690; notion 1746–; draught 1758-a1775; tid a1774 sc–; inkling 1787 dl–

This subcategory contains the term heart
Here we see some, perhaps predictable, extensions of the usage of the term heart as it shifts across various word-classes and, arguably, gains an intensive sense. The HTE also offers evidence for processes of grammaticalization: we find the term will recurring throughout the classification of the category 02.05. Will/faculty of will, both alone and in numerous compounds. It also occurs with meanings which shade towards an indication of future events. For instance, the

190

Louise Sylvester

classification of the category begins with nouns with the sense “Will/faculty of will”: 02.05 Will/faculty of will n: will<willa OE–; wilhede 1340; volition 1738–; voluntary faculty 1867 .will-power: volition 1844; will-power 1874– .will as natural instinct/drive: will to 1823– ..spec.: vitativeness 1843–; will-to-live 1903–; will to power 1907–

Later we find that the category also includes terms signifying necessity, and the verb form of the term will appears to look both ways: towards the meaning of desire, and also towards a hardening of the sense of driving what will happen: 02.05.02 Must of necessity vi: needs must 1390–; must needs a1400-1782 .be necessary condition: shall<sculan OE-1605+1818 .must inevitably: will c1611– ..be an inevitable consequence: will 1387–; it is (of) force 1483-1802

It is easy to fall into the trap of assuming that because a sub-category appears further down in the classification, the sense comes into use later in time, as in the example of will “be an inevitable consequence” and will “must inevitably” as well as will “act of own free will” which appears in the classification at 02.05.01 with the date 1393–. We need to be careful not to make arguments for trends in semantic shift that are not supported by the evidence of the dates of usage which are derived from the Oxford English Dictionary (OED). As the category numbers indicate, many of the groupings are co-hyponyms, and the dates make it clear that many of the senses co-exist and overlap in temporal distribution. When we look at a small group, quite low down in the classification, we can see that a sub-category may be made up of terms which remain in the language from the medieval or early modern periods to the present day bearing the same senses (as hungrily and cravingly in the example below), and of terms which appear only fleetingly, and which for syntactic or other reasons, do not remain in the language either with this particular sense, or, perhaps, at all: 02.05.03.03.03.01 .craving av: hungrily 1377–; cravingly 1621–; esuriently 1883 al fg; hungeringly 1884

Forces of change: social and moral attitudes

191

In some cases, survival rates clearly vary according to the form of the word, as we see in the cases of hungrily and hungeringly. Sometimes word class may be a deciding factor; while esuriently av is only attested once, the OED, and therefore the HTE editors, suggest that esurient aj is in use from a1672 up to the present day. The kinds of changes we have just been looking at are of the type which have received most attention within historical semantics. Perhaps, however, there are ways in which the structuring principles of the HTE may find a rapprochement with some of the tenets of cognitive linguistics. Much has been written about the cognitive salience of categories, beginning with Rosch’s work10 and extending the findings made within psychology to research within linguistics. The suggestion has been made that as linguists, as well as anthropologists, continue to seek to explain the link between language and cultural construction, the theoretical nexus of language, categorization and culture is likely to tighten.11 Even though ideas derived from cognitive psychology were not part of the design of the HTE project, questions about categories underlay the design of the taxonomy which was devised for it, and the organizational principle of the HTE is that of a hierarchical arrangement of semantic categories based on relationships of hyponymy.12 The HTE consists, therefore, of a sequence of categories and subcategories, organized on the basis of the semantic relationship “is a kind of”. Thus, 02.05.01. Free will “is a kind of” 02.05. Will/faculty of will, and 02.05.01.01 .of actions represents “a kind of” exercise of free will. Taylor (op. cit.) notes that three kinds of relations between linguistic units need to be recognised. One is the “is-a” relation in which one unit instantiates, or may be 10

E.H. Rosch, “On the internal structure of perceptual and semantic categories”, in T.E. Moore (ed.), Cognitive Development and the Acquisition of Language, New York: Academic Press, 1973, pp. 111-14; “Cognitive representations of semantic categories”, Journal of Experimental Psychology: General 104, 1975, 192-233; and with C.B. Mervis, W.D. Gray, D.M. Johnson and P. Boyes-Braem, “Basic objects in natural categories”, Cognitive Psychology 8, 1976, 382-439. 11 R.E. MacLaury, “Preface: Linguistic and anthropological approaches to cognition”, in J.R. Taylor and R.E. MacLaury (eds.), Language and the Cognitive Construal of the World, Berlin: Mouton de Gruyter, 1995, pp. vii-xiii, vii. 12 C.J. Kay, I. Wotherspoon, and L. Sylvester, “One thesaurus leads to another: TOE, HTE, TME”, in C.J. Kay and L. Sylvester (eds.), Lexis and Texts in Early English: Studies Presented to Jane Roberts, Amsterdam: Rodopi, 2001, pp. 173-86.

192

Louise Sylvester

regarded as, a more fully specified instance of a more schematically characterised unit. A second relation is the “is part-of” relation: one unit is part of a larger, more complex unit. A third relation is the “islike” relation in which one unit resembles another unit, in some aspects. Of this final example, Taylor suggests that the simulacrum can thus be assimilated to the exemplar as a marginal instance to a prototype (op. cit., pp. 19-20). As we work down the HTE classificatory headings, the “is a kind of” relation is clearly evident, and the classification also includes the second, “is part-of” relation. The HTE may thus be said to include examples from both the first and the third of Taylor’s list of relations (so, for example, a viola is-like a violin/is-a-kind-of stringed instrument, eschewing the abstract for the concrete, for a moment.). These sense relations may be seen as fitting into the ideas of dynamic construal that have become so important in a cognitive view of meaning. In this connection we may note Croft and Cruse’s suggestion that one of the main aims of the dynamic construal approach to word meaning is to achieve a unified account of both hard and soft aspects of word meaning, both flexibility and rigidity, and to locate the origins of these at first sight contradictory properties. The ‘hard’ properties include sense relations such as hyponymy, incompatibility, meronymy and antonymy, and the existence of structured lexical sets (word fields) (op. cit., p.106). This formulation would seem to offer some validity to the kind of semantic information that is encoded in the HTE classification, built as it is on the relations of hyponomy and meronymy. It seems, then, that there is some theoretical justification, with regard both to the categorial arrangement, and the semantic relationships on which it is built, to assimilate some of the insights we may gain from the HTE classification into a cognitive framework, as Kay’s paper indicates ought to be possible without inordinate difficulty. What is less easy to determine, is how to make use of this kind of information to shed light on diachronic semantic shift which must have been initiated by speakers, and which must therefore have arisen within a context of dynamic construal. One remaining question concerns how these shifts, and the forces which shaped them, can be read off the HTE classification.

Forces of change: social and moral attitudes

3. Construal and concepts

193

Cultural shift

My focus in this paper is on the relationship of language to an understanding of concepts within our culture, and the ways in which these may be seen to change, and, in doing so, produce changes in the lexicon. The observation by Rastier with which we began concerns diachronic semantics and social evaluations; cultural historians, however, have not always considered that we need to track shifts in social evaluations, in particular where these are concerned with abstract ideas such as romantic love and desire. The HTE category under consideration, 02.05. Will/faculty of will contains the subcategory 02.05.03.03. Desire and this is part of the reason that this category seems such a useful one with which to investigate whether the HTE classification offers evidence for changes in social attitudes alongside the information it contains about semantic shifts. Discussing the cultural history of romantic love and desire, Belsey observes that however much sexual practices and preferences are known to differ culturally, the desire that motivates them is tacitly understood to be the same throughout history.13 Both linguists and scholars of popular culture now suggest, however, that emotions should not be regarded as pre-social essences, but as socially ordered and linguistically mediated;14 they argue that the “reality” of sex does not pre-exist the language in which it is expressed; rather, language produces the categories through which we organize our sexual desires, identities and practices.15 We are faced here with the Saussurean paradox: is language a system inherited by speakers, or is it created by speakers. Saussure emphasised that the signs that make up a language are not labels for an independently given list of concepts; he argued instead that it is the language itself that structures cognition, thereby creating the concepts through the very process of symbolising them. Saussure, however, has been seen as shying away from the full implications of his theory. Thus, he observes that if, of the three “synonyms” redouter, craindre, and avoir peur, redouter did not exist, its meaning 13

C. Belsey, Desire: Love Stories in Western Culture, Oxford: Blackwell, 1994, p. 7. S. Jackson, “Women and heterosexual love: complicity, resistance and change”, in L. Pearce and J. Stacey (eds.), Romance Revisited, London: Lawrence and Wishart, 1995, pp. 49-62, p.51. 15 D. Cameron and D. Kulick, Language and Sexuality, Cambridge University Press, 2003, p.19. 14

194

Louise Sylvester

would be shared out amongst its competitors.16 Here, therefore, Saussure appears to presuppose the existence of a conceptual content, which is independent of language, and which has to be lexicalised in some way or other.17 For an understanding of the framing of the idea of consent, the subcategory 02.05.03.02. Unwillingness would seem to be an important one. The noun category 02.05.03.02. Unwillingness contains 46 items. It also has two small subcategories hanging off it: 02.05.03.02. instance of contains just two items, nill 1526-1677 and stay 1550-1567. The second, 02.05.03.02. one who is unwilling has only three items: boggler 1606, drawback 1618, and hanger-back 1923+1962. This relative paucity of tokens may look like reluctance to name those who resist doing what is expected of them. The adjectival category, 02.05.03.02. Unwilling, however, contains 75 items. It is perhaps easier to characterize those who are unwilling than to name the characteristic. The subcategory 02.05.03.02. Unwilling is perhaps one which reflects a variety of nuances of disapproval, and perhaps requires the replacement of euphemisms. Certainly, the modern connotations of some of the terms we find here seem very different from one another; for example, dangerous, indisposition and grudging: 02.05.03.02. Unwillingness n: uniwill c1175; danger c1290-1526; loathness a1300-1709; nilling c1374-1710+1865 ai; unlust 1390-1535; throness/thraness a1400; loathiness c1449; unlustiness a1470-1583; difficulty 1513-1769; unreadiness 1526-a1761; scruple 1526–; sticking 1528-1725; remorse a1529; sweerness 1533-1659 sc & nd; untowardness c1547-1579; dangerousness 1548; loathsomeness 15561807; envy 1557; retractation 1563; unwillingness 1593–; indisposition 1594–; loathfulness 1596; backwardness1597–; offwardness 1600 (2); grudging 1601–; reluctation 1605-1674; hink c1614-1709 sc; reluctancy 1634–; disinclination 1647–; indisposedness 1651-a1691; shyness 1651–; nolence/nolency 1652 (2); nolition 1653–; costiveness 1654 ai; scrupling 1665–; reluctance 1667–; queerness 1687; uneasiness a17151737; tarditude 1794; inalacrity 1813-1855; grudgingness 1820–; tarrowing 1832 sc–; stickling 1848–; grudgery 1889; balkiness 1894–; stickiness 1933– 16

F. de Saussure, Cours de Linguistique GÊnÊrale, ed. C Bally and A Sechehaye, Paris: Payot, 1943, p. 160. ( First edition 1916). 17 See Taylor, op. cit., p. 22.

Forces of change: social and moral attitudes

195

The connotations of many of the terms in this subcategory (see Appendix 1 for the whole subcategory), in particular those in use in earlier periods, look very different from those with which we now associate those words, such as revere and bitch, that are still in the language. It is interesting, for example, that the verb force, which appears at 02.05.03.02.0 Be unwilling (see Appendix 1) had a brief existence with the sense ‘be unwilling’ between 1546 and 1591. The first problem is, of course, how to get at the connotations of these terms, and for that we need recourse to large historical corpora which will show us the items in use. It may be that the citations in the OED will have to serve as a starting-point, though it seems a pity that the OED offers no style labels for many of these terms as clues to their register. Simply looking at the items, we may wonder if we can we read off this list any sense of social attitudes towards those who respond with doubt to the actions asked of them, and if it is possible to detect any temporal shifts in such attitudes. We may wonder, further, whether this category includes those who wish to refuse offers of sex. A second problem is that this line of inquiry returns us to the problems of individual words, and takes us away from the information afforded by the conceptual grouping of the lexical items. Meaning as construal Cognitive linguistics suggests that it is an error to suppose that people use language in order to refer directly to “things in the world”, that is, to things outside of the mind; rather, language is used to refer to mental “projections” of the world.18 From its very inception, cognitive grammar is said to have emphasised the role of “construal” in semantics: linguistic expressions do not refer “directly” to states of affairs in the world, but to speakers’ conceptualisations of these states of affairs; thus, cognitive semantics defends itself against the charge that it is excessively concerned with real-world, “universal” categories, for which a language merely supplies a list of names. Moreover, cognitive semantics fully accepts that different languages may make available to their speakers different sets of

18 R. S. Jackendoff, Semantics and Cognition, Cambridge, Ma.: MIT Press, 1983, p. 29.

196

Louise Sylvester

“conventionalised” modes of construal.19 In this connection, we may note that the taxonomy which provides the underlying structure for the HTE was deliberately designed as a folk taxonomy, in an attempt to reflect what it is that people know. The editors’ understanding was that this information consists in general of a smattering of scientific knowledge mixed with the way in which the world is constructed in the popular imagination. Addressing precisely the question of what speakers know and how this is reflected in the categorial hierarchies they construct, Croft and Cruse (op. cit., pp. 96-97) ask about difference between terms which may appear at more than one level in a taxonomy: what is the difference, for example, between the basic level thrush and the subordinate level thrush. They suggest that an important factor is richness of content, in terms of knowledge, memories, connections and so forth. While a competent naturalist will have a relatively rich representation of items such as blue-tit, swallow, thrush and so on, town-dwellers may well know the names, but very little else about the different birds, so these names will not form satisfactory labels for basic level items: the names will be little more than place-holders for potential knowledge. Such people may be able to form an image of a generic (garden) bird, but not have enough experience or knowledge to be able to visualise individual species. Insofar as they have both a superordinate level and a basic level construal of bird, these will not be different level construals of the same category, but two different categories that fit at two levels. Croft and Cruse go on to question what happens in the case of speakers who can operate with two different systems, the technical and the social; this is precisely the sort of question that is addressed within the HTE taxonomy. An example which illustrates this may be found in the category 01.02.08.01.09 Fruit & vegetables where we find a subcategory 01.02.08.01.09.03.07 Fruits as vegetables. This heading seems to indicate that two types of knowledge may be present at the same time: the knowledge that the data, including types of capsicum such as red pepper and a number of names for different kinds of tomatoes, are, indeed, fruits, and the information that most people eat them as though they were vegetables. Croft and Cruse assume that speakers who can operate with two different systems will adjust themselves cognitively to each situation, 19

Taylor, op. cit., p. 31.

Forces of change: social and moral attitudes

197

the technical and the non-technical, and revert to the societal default construal of the terms. They ask whether such speakers actually change their categories or simply construe a new level for them. If they actually restructure the categories, for instance by backgrounding aspects of knowledge that are highly relevant in a professional setting, then they are effectively creating new conceptual categories (op. cit., p.97). A clear parallel may be found in terms of my discussion about the concepts of will and consent. Legal definitions of coercive sex include the ideas of trespass and rape; personal definitions, however, may not include these concepts. In a personal definition, coercive sex may form part of fantasies about sex, fantasies which may be derived from models in romantic fiction. Croft and Cruse conclude that it would seem reasonable to assume that level cannot be construed independently of content, that is to say, any observed movement up or down a taxonomic hierarchy will be a consequence of different construals of the category denoted by a lexical item. What is difficult, is to see how it is possible to include, or to infer, a range of construals over time from a fixed taxonomy. How might it be possible to reconcile the diachronic nature of the data contained within the HTE classification with its defining headings which are in part those of the OED, and in part the creations of the HTE’s editors? In this respect, we may compare the HTE with another lexicographical resource, the Middle English Dictionary (MED); here, the definitions are intended to be specific to a historical understanding of the particular period of language use. Knowledge of concepts The question of how concepts are understood is crucial to the investigation of this paper: it is not possible to trace changing conceptualizations without some sense of how concepts are perceived and how meanings are attached to them. On the other hand, we need to discover if we can discuss concepts that go beyond concrete nouns, and the basic domains delineated by Langacker and others, without, finally, writing cultural history that ignores questions of linguistic or mental structures. The issue of meaning, and of how we understand what we perceive, has been much discussed recently. The root of the question concerns how far mental perceptions may be said to be stable, how far they form a background of information that comes into

198

Louise Sylvester

play when we are faced with new objects, and how far all understanding of meaning, including conceptual meaning, is “online”, a dynamic construal created in and by linguistic usage. To address this issue we need to refine our understanding of the set of possible relationships between a conceptual classification and what we understand by knowledge of concepts. Taylor (op. cit., p. 38) observes that although Saussure used the word concept to designate the semantic pole of the linguistic sign, many semanticists have been reluctant to appeal to concepts at all. Concepts, by definition, are private, mental entities; a person can only have access to another person’s concepts through the medium of language. If language is defined as a means of symbolising concepts, however, there is no methodology for independently establishing the nature of another person’s concepts. Taylor notes that a common view amongst psychologists is that a concept is a principle of categorisation. To “have” a concept is to have the means to categorise entities as examples of that concept; thus, to have the concept tree is to have the ability to recognise a tree when one sees one. Most discussions about the nature of categories, then, propose a constant underlying mental representation of some kind for each category. Recent research, however, suggests that mental categories are variable as well as stable, and that people appear to be able to create categories on the spot: see, for example, Barsalou on ad hoc categories,20 Aitchison on the mental lexicon,21 and Smith & Samuelson on category instability.22 Indeed, Smith and Samuelson suggest that category variability. and in-task category creation, are facts that are not well explained by the idea of a concept. Cognitive processes, they argue, depend broadly on the immediate input and its larger context; they are temporally extended with real rise times and decay times, such that activity at any moment depends on, and emerges out of, preceding activity and they change as a direct consequence of their own activity. Smith and Samuelson state that these facts about perceiving and remembering have profound 20

L.W. Barsalou, “Ad hoc categories”, Memory and Cognition 11, 1983, 211-27. J. Aitchison, Words in the Mind: An Introduction to the Mental Lexicon, Oxford: Blackwell, 1994. 22 L.B. Smith and L.K. Samuelson, “Perceiving and remembering: category instability, variability and development”, in K Lamberts and D. Shanks (eds.), Knowledge, Concepts and Categories, Hove: Psychology Press, 1997, pp. 161-95. 21

Forces of change: social and moral attitudes

199

implications for theories of categories; they conclude that objects cannot map stably into a set taxonomy because the objects of mental life are not themselves stable entities (op. cit., p.173). 4. Changing conceptualizations as indicated by legal and social historians. When we consider the diachronic conceptualization of the idea of consent, a concept which appears to be part of the meaning of will and to be included in notions of free will and coercion, we may wish to ask how it is understood and interpreted by legal and social historians. Scholars examining the law of rape in the Middle Ages are agreed on the difficulty of teasing out definitions from the statutes, eyre rolls and legal treatises of the period.23 The issue of consent is, however, crucial to legal definitions of rape and to the manner of its prosecution in the medieval period.24 Blackstone observes that Roman law supposes that women never go astray without the seduction and arts of the other sex and therefore by restraining and making so highly penal the solicitations of the men, the Roman legislators meant effectually to secure the honour of the women. Blackstone goes on to state that English law does not entertain quite such sublime ideas of the honour of either sex: it does not lay the blame of a mutual fault upon one of the transgressors only and therefore makes it a necessary ingredient in the crime of rape, that it must be against the woman’s will.25 Brundage, however, suggests that both medieval and modern notions of rape are grounded in the raptus of Roman law and that in ancient Roman law raptus consisted in the abduction and sequestration of a woman against the will of the person under whose authority she lived. In contrast to legal definitions of modern day rape, sexual intercourse

23

B. Hanawalt, Crime and Conflict in English Communities 1300-1348, Cambridge, Ma.:Harvard University Press, 1979, p. 104; J. M. Carter, Rape in Medieval England: An Historical and Sociological Study, New York: University Press of America, 1985, p. 35. 24 Carter, op. cit., pp. 4-5. 25 W. Blackstone, Commentaries on the Laws of England, (4 vols.), ed. J. Stewart, London: V. & R. Stevens and G.S. Norton, 1854, IV, p. 261.

200

Louise Sylvester

was not a necessary element of ancient raptus.26 He goes on to argue that sexual advances of any kind, no matter how amiable, involve pressure of some sort by one person on another, observing that society may choose to draw the line between permissable and nonpermissable pressures at any point along a very wide spectrum. The consent issue that medieval jurists addressed in rape cases was not “Did she consent?” but rather, “Was her consent procured by force, either physical or mental?”(op.cit. pp. 68-8). Discussing erotic love in the law of rape, Naffine states that if we look more closely at the meaning given to consent in the law of rape, we can observe the continuing objectification of one of the parties (the person to whom sex is proposed) in the contractual form of sex. She argues that what we discover is that consent to sex (for the purposes of the law of rape) does not mean free agreement; it does not entail a mutual transaction or negotiation on equal terms. Legal interpretation of consent for the purposes of the law of rape (of both a judicial and academic nature) reveals that the sexuality presupposed is still a traditional compelling, coercive sexuality.27 Interestingly, the notion of the imposition of sexual desire on another appears only obliquely in the HTE classification of the category 02.02.22. Love: one of the subcategories hanging off the heading 02.02.22.07. Action of caressing is 02.02.22.07..act of influencing/persuading by caresses which contains two items: coying 1580-1603+1887 and coaxing 16721683+1870. Turning to the HTE classification of 02.05.Will/faculty of will, it is interesting to note that that the subcategory 02.05.03.03. Desire immediately succeeds the subcategory 02.05.03.02. Unwillingness examined above (see Appendix 2 for the subcategory 02.05.03.03. Desire in full). This arrangement prompts consideration of the question of how far these ideas are linked only in the minds of the editors of the HTE, and how far it is possible to trace a diachronic link between them. The subcategory 02.05.03.03. Desire ends with the grouping of the transitive verbs, and finally with the reflexive verbs, 26

J.A. Brundage, “Rape and marriage in the medieval canon law”, Revue de droit canonique 28, 1978, 62-75, reprinted in J.A. Brundage, Sex, Law and Marriage in the Middle Ages, Aldershot: Variorum, 1993, p. 63. 27 N. Naffine, “Possession: erotic love in the law of rape”, Moderm Law Review 57, 1994, 10-37, p. 10.

Forces of change: social and moral attitudes

201

and both the headings and the items they contain seem suggestive in a context of thinking about consent and coercion: Desire vt: awilnian OE; cepan OE; facian OE; friclan OE; frymdig beo OE; recanOE; (ge)willian OE; wilne<wilniaOE-c1540; will<willan OEa1677+1875; desire c1230; appetite c1385-1652; appete c1385-1685; wait after1393-1533; set one's mind on 1509-1827; list 1545-1587; exopt 1548; have a mind for 1616–; desiderate 1645–; lust 1648; have a mind to 1674-1726; want 1706–; choose 1766 vu–; have eyes for 1810– ; like 1822–; be stuck on 1886 us sl– .a person/his presence: want c1760– .a person to do something: want 1845– .something to be done: will<willan OE–; love c1380-1682 .pursue object of desire: follow a1300 fg– ; court 1571– .cry for object of desire: wail for 1573/80-a1771 .excite (desire): tickle 1547/64 Desire vr: lust oneself a1586-1599

This subcategory is immediately succeeded by the set of headings relating to wishing beginning with 02.05.03.03.01.02. .a wish, and the next subcategory in the classification is 02.05.03.03.02. Longing/yearning. As a reader interested in the construction of heterosexual desire in literature, this raises the question of how far I am imposing my own interests in interpreting these categories and their data in a way which includes the relationship of sexuality and desire and their enactments. When we look again at the classification and data of the notion of desire, do we find any hint of this interpretation of sexuality, an interpretation which is, arguably, found in conventional romance from the medieval to the modern period? Thus we return to the main topic of investigation raised in this paper: how far it is possible to learn something about changing social attitudes in the area of will, consent and coercion from the HTE classification; and if the various modes of understanding, linguistic, encyclopaedic, cultural and social, can be placed in productive dialogue with one another. As noted above, the notion of construal in cognitive linguistics includes the idea that meaning presupposes knowledge systems28 and that there is no need to make a principled distinction between “encyclopaedic knowledge” and “linguistic 28

R.W. Langacker, op. cit., pp. 184-5.

202

Louise Sylvester

meaning”.29 These tenets of cognitive linguistics suggest that both cultural understanding and semantics must form part of our understanding of meaning. 5. Conclusion The insights relating to construal proferred by cognitive linguistics suggest a need to address issues concerned with how meaning in language is understood and the relationship of our understanding in this area to the evidence that lexicographical projects offer. If meaning is no longer considered to be a property of words, what are the implications of the various theories of construal of meaning for lexicographical projects such as the HTE, a project that was envisaged not only in terms of semantic relations, but as a “conceptual classification”?30 This is not an issue that is unique to the HTE: the editors of the electronic lexical database WordNet have been described as vacillating between a position that stays entirely in the plane of words, and deals with conceptual relationships in terms of the relationships between words, and one which separates the conceptual plane from the terminological plane.31 However, one of the editors of that project asserts that the lexicon is constrained by the kinds of concepts that are available to us by virtue of our perception of, and interaction with, the world around us; she claims that the WordNet project deals in concepts.32 Some light may be shed on the question of the theory of construal and the understanding of concepts by various caveats that have been put forward in relation to the notion of construal. In particular, these include the idea that the construal of interpretations is not unconstrained. The constraints are many and varied: they vary in strength and may reinforce one another or cancel one another out. These constraints can also be overcome by cognitive effort, but the stronger the constraint, the greater the cognitive effort required to impose a construal that defies the constraint. They also vary in their 29

J. R. Taylor, op. cit., p. 20. M.L. Samuels, op. cit., p. 180. 31 D. Soergel, Review of C. Fellbaum, “WordNet: an electronic lexical database”, in D-Lib Magazine at http://wwwdlib.org/dlib/october98/10contents.html 32 C. Fellbaum, Introduction to WordNet: An Electronic Lexical Database, Cambridge, Ma.: MIT, 1998, pp. 1-19. 30

Forces of change: social and moral attitudes

203

stability under change of context and they include the nature of reality: some aspects of experience lend themselves more readily to construal in certain respects and less readily to construal in other respects.33 Geeraerts argues that while the dynamic flexibility of prototypical categories does indeed demonstrate the interpretive potential of linguistic categories, it also constrains the set of possible interpretations, as these are required to link up with existing concepts. Interpretation does not go wild because speaker and hearer share a system of preferential interpretations that also constrains the form deviant readings may take. He concludes that the theoretically important point is that this restrictive preferential pattern shows up in the structure of language itself, in prototypicality, and that it would therefore be wrong to stress only the potentially infinite set of interpretations that languages allow: the interpretive constraints that lie at the basis of mutual understanding are also part of the structure of language.34 This last point suggests that the HTE, which is structured as a series of hierarchical categories organized by conceptual meaning, does have something to tell us about language and meaning, since linguistic meaning is not as wide open as the notion of construal may lead us to expect. It seems clear, then, that meaning is culturally constructed, based on the input speakers receive about words within the contexts in which concepts are encountered. Pragmatics and discourse analysis, disciplines which grew out of semantics, offer information about the processes of interpretation that take place within the structure of conversations. Equally, however, there are “hard” properties of word meaning and constraints on the interpretation of words. Lexicographical projects such as the HTE may be mined for the information they offer about the lexicalization of concepts in language. The work of Aitchison (op. cit.) is suggestive about the role of lexical fields in the storage of words in the mental lexicon, indicating that the basis on which the words and concepts are grouped in the HTE is cognitively salient, in line with Christian’s suggestions 33

W. Croft and D.A. Cruse, op. cit., p. 101. D. Geeeraerts, “Cognitive semantics and the history of philosophical epistemology”, in R. A. Geiger and B. Rudzka-Ostyn (eds.), Conceptualizations and Mental Processing in Language, Berlin: Mouton de Gruyter, 1993, pp. 53-79, p. 73. 34

204

Louise Sylvester

in her paper on historical semantics and historical lexicography. Nevertheless, our knowledge that interpretation is culturally constructed should remind us of the modest words of the last editor of the MED: the MED will stand as a monument of historical lexicographical practice of a particular time and place.35 The HTE can tell us many things; it may also be able to tell us something about the constructions we place on the concepts we encounter in the literature of the medieval period, and of our own time. Appendix 1 Extract from 02.05. Will/faculty of will in the HTE 02.05.03.02. Unwillingness n: uniwill c1175; danger c1290-1526; loathness a1300-1709; nilling c1374-1710+1865 ai; unlust 1390-1535; throness/thraness a1400; loathiness c1449; unlustiness a1470-1583; difficulty 1513-1769; unreadiness 1526-a1761; scruple 1526–; sticking 1528-1725; remorse a1529; sweerness 1533-1659 sc & nd; untowardness c1547-1579; dangerousness 1548; loathsomeness 15561807; envy 1557; retractation 1563; unwillingness 1593–; indisposition 1594–; loathfulness 1596; backwardness1597–; offwardness 1600 (2); grudging 1601–; reluctation 1605-1674; hink c1614-1709 sc; reluctancy 1634–; disinclination 1647–; indisposedness 1651-a1691; shyness 1651–; nolence/nolency 1652 (2); nolition 1653–; costiveness 1654 ai; scrupling 1665–; reluctance 1667–; queerness 1687; uneasiness a17151737; tarditude 1794; inalacrity 1813-1855; grudgingness 1820–; tarrowing 1832 sc–; stickling 1848–; grudgery 1889; balkiness 1894–; stickiness 1933– .instance of: nill 1526-1677; stay 1550-1567 .one who is unwilling: boggler 1606; drawback 1618; hanger-back 1923+1962 Unwilling aj: unwillende OE+c1330; loath/loth
R.E. Lewis, “The Middle English Dictionary at 71”, Dictionaries 23, 2002, 76-94, p. 88.

Forces of change: social and moral attitudes

205

c1598-1632 sc; backward 1599–; unpregnant 1602-1603; scrupulous 1608-1754; refractory c1610; behindhand 1611; unprone1611+1883; nilling 1620-1710; backwards 1627-1683; shy a1628–; retractable 1632; uninclinable 1640-1656; unapt 1640–; unbeteaming 1642; averse 1646– ; indisposed 1646–; disinclined 1647–; tender of/in 1651-1840; relucting 1655-1659; aversant 1657; scrupulous of/in 1658-1662; incomposed 1660; unready 1672-1857; boggling a1683; unconsenting a1693–; retrograde 1695-1760; lought a1700; reluctant 1706–; uninclined 1729-1740; tenacious 1766-1802 er; disinclinable 1769; illdisposed 1771; unwishful 1775-1876 di+1894; unaffectioned 1788; back-handed 1817; sweert 1824 sc–; tharf 1828 dl & fg –; backward in coming forward 1830–; unvoluntary 1834; misinclined 1837; squeamy 1838 og us–; ba(u)lky 1847–; retractive 1869; grudging 1874–; tharfish 1876 dl–; -shy 1884– .characterised by unwillingness (of actions etc.): grudged 1549c1636+1853; unwilling 1613–; reluctant1667–; uncheerful a1684+1858; restive1806/7– .more unwilling: worse-willing 1549-1584 .most unwilling: the last 1535-1588 Unwillingly av: ungeorne OE; unwillum OE; unwilsumlice OE; unwilles/unwillan
206

Louise Sylvester

make dainty 1579-1649; hang back 1581–; rue 1583-1630; make doubt 1586; make (a) scruple 1589–; make it scrupulous 1593; yearn 1597; hang the wing a1601-a1624; make squeamish 1611-a1617; smay 16321667+1841 dl & dl di–; boggle a1638–; hang off 1641–; reluct 16481899; shy 1650–; scruple 1660–; make boggle 1667-1768 ai; revere 1689; begrudge 1690; have scruples 1719–; stop c1738; bitch 1777; reprobate 1779; stickle 1819–; reluctate 1835; disincline 1885 Hesitate/scruple at vt: find bones in 1459-a1529; stick at 1525–; make bones of/about 1548–; scotch at 1601-1627+1887 dl; strain at 1609– .be unwilling to grant/allow: grutch a1300-1719 nn dl & ai; begrudge 1362–; grudge c1500–; repine 1615 .render unwilling: indispose 1692– Be unwilling vr: sonyie c1500 sc Unwilling ph: invita Minerva 1584-1626+1848– .against the will of: unwilling to (unto) 1555-1654

Appendix 2 Extract from 02.05. Will/faculty of will in the HTE 02.05.03.03. Desire n: (ge)cneordnes OE; friclo OE; myn(e)le OE; willa OE; i-will
Forces of change: social and moral attitudes

207

.desire including thought: feeling c1810 py– .desire excluding thought: feeling c1810 py–; appetency 1836/7 mt .instinctive/natural desire: appetite 1366– .secret desire: grudging 1625-1694 .contrary desire: counter-lusting 1656-1666 .trivial desire: curiosity 1605-a1718 .temporary desire: frenzy 1632-1761; mania 1689–; furor 1704–; influenza 1774–; rage 1785–; furore 1790–; craze 1887–; enthusiasm 1916– Pertaining to/characterized by desire aj: desirous a1420-1652; appetitive 1577–; appetitual 1616; epithymetic 1631–; epithymetical 1646; appetitious 1653-1668; desiderative 1655/60+1816; orectic 1779 pl–; appetent 1837 mt; .desirous: lustbære OE; lustful OE; gelysted OE; geþancol OE; willende OE; willesful a1225-c1290; desired a1300-1598; desirous c1300–; wilful 1340-1573/80; willing 1382–; desiring c1386–; talentive a1400-c1450; lusty c1400-1657; desirant c1450; desireful c1500 nn rr–; fond of 15521779; lusting 1559-1844; desirable 1759 .capable of desire: gewilniendlic OE .desired: gelustful OE; sought a1300–; desired 1382–; sought after/for 1605– ; envied 1631–; desiderated a1743–; courted 1793; ..much: well(-)desired 1604 ..equally: efenleof OE ..long: long-wished 1570-a1649; long-desired 1570–; long-wished-for 1748 .desirable: geornlic OE; (ge)giernendlic OE; mynelic OE; gewilniendlic OE; wilsum OE; gewyscendlic OE; yearnandlike a1300 (2); desiderable a1340-1675; fair c1380-1676; desirable 1382–; desireful 1382 ai–; dainteous a1400; desirous1430-1728; expetible 1569-1721; lustful 1610-a1667; appetible 1622-1660+1847; desiderate 1640; deligible

1680; wantable 1970– ..exceedingly desirable: covetable c1340–; over-desirous 1483 .stimulating desire: orectic 1892 me Desirously av: wilfully c1350-c1611; desirously c1400–; desiderantly c1450; appetently a1479; desiringly 1552–; wantingly 1894 .in a desired manner: desiredly 1625-1666 .desirably: desiderably 1635; optably 1657; desirably 1823– Desire vt: awilnian OE; cepan OE; facian OE; friclan OE; frymdig OE; beo OE; recan OE; (ge)willian OE; wilne<wilnia OE-c1540; will<willan OE-a1677+1875; desire c1230; appetite c1385-1652; appete c13851685; wait after 1393-1533; set one's mind on 1509-1827; list 15451587; exopt 1548; have a mind for 1616–; desiderate 1645–; lust 1648; have a mind to 1674-1726; want 1706–; choose 1766 vu–; have eyes for 1810–; like 1822–; be stuck on 1886 us sl–

208

Louise Sylvester

.a person/his presence: want c1760– .a person to do something: want 1845– .something to be done: will<willan OE–; love c1380-1682 .pursue object of desire: follow a1300 fg–; court 1571– .cry for object of desire: wail for 1573/80-a1771 .excite (desire): tickle 1547/64 Desire vr: lust oneself a1586-1599

Key word in context: semantic and pragmatic meaning of humour Irma Taavitsainen … honesty, integrity, courage, trust, gratitude, humour, creativity, wisdom or love (CA5 2542 BNC) … þrow sum corrupte humour. And þis sikenes bynemet a mannes wit … (Gilbertus Anglicus, p. 27; emphases mine)

What can the context of a word tell us about its meaning? In Presentday English humour is a positive notion, collocated with pleasant ideas, whereas its common collocations in Middle English are negative adjectives ill, evil, corrupt. My aim is to discuss the semantic and pragmatic meanings of this lexeme. First, I shall outline the semantic meaning and then assess how the pragmatic meaning is negotiated. My study focuses on the occurrences of the word humour (with its spelling variants) in the corpus of Middle English Medical Texts (MEMT, forthcoming).1 The corpus is used to locate collocations for the semantic meaning, and to find illustrative examples for qualitative analysis of the pragmatic meaning. Scientific Thought-styles and Middle English Medical Texts Beside communication and representation, language is used to construct ideology. The work by the Scientific Thought-styles project at the University of Helsinki shows that linguistic features typical of scholasticism and empiricism are very different from one another, reflecting the underlying methodology and philosophy of science.2 1

The spelling variants can easily be found in word lists (a regular function in MEMT Presenter software). 2 The present project members are Irma Taavitsainen, Päivi Pahta, Martti Mäkinen, Turo Hiltunen, Maura Ratia, Carla Suhr, and Jukka Tyrkkö. The project is part of the Research Unit for Variation and Change in English at the Department of English (VARIENG), University of Helsinki, with a centre of excellence status and funding by the Academy of Finland 1999-2005; the research reported here was also supported

210

Irma Taavitsainen

The kinds of evidence for scientific knowledge varies: scholasticism is logocentric and texts accordingly rely on quotative evidence seen in their frequent use of speech act verbs of saying; in contrast, empiricism is based on observation, with sensory verbs of perception predominating.3 To assess thought-styles, the project is compiling an electronic corpus, the Corpus of Early English Medical Writing (CEEM 1375-1750). Its first part, MEMT, which will appear in 2005, contains vernacular texts of the scholastic period from c.1375 to c.1500. It is mainly compiled of edited medical treatises, but provides a great deal of new material for researchers as several unpublished theses are included. It also contains transcripts of some new texts. Shorter texts are included in toto while longer texts are represented by extracts. Its main branches have been devised according to the tradition of writing as “Surgical texts”, “Specialized texts”, and “Remedies and materia medica”. Texts in the first and second categories are learned, and have their origins mostly in the academic tradition. The third group includes remedy books, recipes, prognostications, and materia medica, represented by herbals. This category contains learned elements as well, but the transmissions are extremely complicated.4 Verse texts are recorded as a separate group (see MEMT Introduction for details).

by that funding. Christian Kay visited VARIENG several times during the period as she served on the Advisory Board. We are grateful for her input, constructive discussions and inspiration. 3 See Irma Taavitsainen, Päivi Pahta, Noora Leskinen, Maura Ratia, and Carla Suhr, “Analysing scientific thought-styles: What can linguistic research reveal about the history of science?”, in Helena Raumolin-Brunberg et al. (eds.), Variation Past and Present: VARIENG Studies on English for Terttu Nevalainen, Mémoires de la Société Néophilologique de Helsinki 61, Helsinki: Société Néophilologique, 2002, 251-70; and Irma Taavitsainen, “Evidentiality and scientific thought-styles: English medical writing in Late Middle English and Early Modern English”, in Maurizio Gotti and Marina Dossena (eds.), Modality in Specialized Texts: Selected Papers of the 1st CERLIS Conference, New York, Oxford, Wien: Peter Lang, 2001, pp. 21-52. 4 Martti Mäkinen, “Herbal Recipes and Recipes in Herbals – Intertextuality in Early English Medical Writing”, in Irma Taavitsainen and Päivi Pahta (eds.), Medical and Scientific Writing in Late Medieval English, Cambridge: Cambridge University Press, 2004, pp. 144-73; and Martti Mäkinen, English Herbals in the Field of Medieval Medical Intertexts. Diss. University of Helsinki, forthcoming.

Semantic and pragmatic meaning of humour

211

Semantics and pragmatics: definitions of meaning Semantics and pragmatics are closely connected and partly overlapping fields of study. Recent textbooks in semantics cover areas that have previously been treated in the pragmatic domain only, such as speech acts and implicatures. To make a distinction, semantics has been defined as pertaining to the meaning of words and sentences; pragmatics to the meaning of utterances. Semantics focuses on linguistic expressions, while pragmatics pays attention to the interlocutors, speakers and hearers. Negotiation of meaning is a central issue as pragmatics focuses on meaning in interaction, how hearers add contextual information to the semantic structure and how they draw inferences from what is said.5 Semantic and pragmatic meanings are complimentary, both making use of context to a greater or lesser degree but focusing on different aspects of language.6 Historical semantics deals with semasiological and onomasiological changes, i.e., what changes a lexical meaning underwent or what developments and restructuring of coding took place in a particular domain.7 Historical pragmatics focuses on language in interaction, taking the contributions of both speaker and hearer into account; the units of study are utterances, and the context, both linguistic and sociohistorical, provides a clue to what speakers mean and how readers construct the meaning.8 Humoral medicine The word humour was a key ideological concept of medicine in the period when the Ptolemaic world view prevailed. The idea of the macrocosm of heaven being projected onto the microcosm of man was 5

See, e.g., Jenny Thomas, Meaning in Interaction: An Introduction to Pragmatics, London and New York: Longman, 1995, and Jef Verschueren, Understanding Pragmatics, London: Arnold, 1999. 6 Cf. Katarzyna M. Jaszczolt, Semantics and Pragmatics: Meaning in Language and Discourse, London: Pearson Education, 2002, p. 1. 7 Elizabeth Traugott and Richard B. Dasher, Regularity in Semantic Change, Cambridge: Cambridge University Press, 2002, p. 25. 8 Susan Fitzmaurice and Irma Taavitsainen, “Approaches to negotiated meaning in historical contexts”, in Susan Fitzmaurice and Irma Taavitsainen (eds.), Methods in Historical Pragmatics, Mouton de Gruyter, forthcoming.

212

Irma Taavitsainen

the cornerstone of the medieval world view. According to the humoral theory, the human body consisted of four bodily humours or fluids: blood, phlegm, choler and bile. Health was understood as a balance between the humours, while sickness was due to an excess or lack of one of them.9 In the holistic system the planets and the signs of the zodiac had an influence on human life and the surrounding environment: the system of physical and physiological fours related to the four elements (fire, air, earth and water), the four qualities (hot, cold, moist and dry), the seasons of the year, the ages of man, the points of the compass, and the signs of the zodiac (in groups of four according to the elements). This idea underlies all medieval medical and medico-astrological texts as a standard component of learning used to explain the existing order and the structure of the universe.10 Semantic meaning: the company a word keeps The outline of semantic development is demonstrated in the Oxford English Dictionary (OED) on-line. The earliest meanings of humour are concrete and physiological and develop through multiple meanings into abstract senses in Present-day English (for the range of meaning and examples, see OED). The direction of change from concrete to abstract is claimed to be universal,11 and there is a tendency towards an increase in ‘expressiveness’ or subjectivity in semasiological changes.12 Present-day humour has been ascribed at least two prototypical centres with an overlap with wit, MOOD and AMUSEMENT, but this word has not been observed to have 9

See Faye Getz, Medicine in the English Middle Ages, Princeton, New Jersey: Princeton University Press, 1998, pp. 55-56. 10 Nancy Siraisi, Medieval and Early Renaissance Medicine: An Introduction to Knowledge and Practice, Chicago and London: University of Chicago Press, 1990, pp. 104-6; John. A. Burrow, The Ages of Man: A Study in Medieval Writing and Thought, Oxford: Clarendon Press, 1986, pp. 12-13; Andrew Wear, Health and Healing in Early Modern England, Aldershot: Ashgate, 1998, pp. 37-39. 11 Elizabeth Traugott, “On regularity of semantic change”, Journal of Literary Semantics 14 (1985), pp. 155-73; Elizabeth Traugott, “On the rise of epistemic meanings in English: an example of subjectification in semantic change”, Language 65 (1989), pp. 31-55; Eve E. Sweetser, From Etymology to Pragmatics: Metaphorical and Cultural Aspects of Semantic Structure, Cambridge: Cambridge University Press, 1990. 12 Traugott and Dasher, op. cit., p. 94.

Semantic and pragmatic meaning of humour

213

developed negative or markedly subjective meanings.13 In addition, the lexeme has come to mean the faculty of perceiving what is ludicrous or amusing; the collocation sense of humour occurring frequently in the British National Corpus. The meanings in the late medieval period are directly related to medieval medicine (see OED): (1) moisture; damp exhalation; vapour; or any fluid or juice of an animal or plant, either natural or morbid, connected with mediæval physiology; (2) spec. In ancient and mediæval physiology, one of the four chief fluids (cardinal humours) of the body … by the relative proportions of which a person’s physical and mental qualities and disposition were held to be determined.

In semasiological studies, the form of the lexeme is typically kept constant, though there may be phonological changes. The Historical Thesaurus of English will be indispensable in showing affiliations of the word with various semantic fields, and more evidence from corpora of various genres and registers, including both literary and non-literary texts, will be helpful for tracing the path from concrete to abstract. As an overall line of development based on OED entries, constitutional physiological senses seem to have given way to temporary states of mind, a particular disposition or liking for some action without apparent ground or reason, and a quality of action which excites amusement. The Middle English Compendium (MED) on-line lists typical collocations of humour. Collocates pertaining to moisture are radical and seminal, and to the anatomical meaning kindeli, natural, god, innatural, unkinde, unnatural, gleimi (viscous) and roten. Figurative senses are also mentioned with ille (ivel, qued, wikked) and collocations of the anatomical structure of the eye are crystalline, cristalline (cristallinus), glasen (glasi), vitreus, whitish (whitli), albugineus, and airi.

13 Päivi Koivisto-Alanko, Abstract Words in Abstract Worlds: Directionality and Prototypical Structure in the Semantic Change in English Nouns of Cognition. Diss. Mémoires de la Société Néophilologique de Helsinki 58, Helsinki: Société Néophilologique, 2000, pp. 240 and 247.

214

Irma Taavitsainen

The collocates in MEMT do not follow these patterns. The words radical and seminal do not occur in MEMT at all. In both surgical and specialized treatises, the collocations pertaining to the structure of the eye are frequent as some of the treatises in the first category are sophisticated descriptions of human anatomy, and an ophthalmological text is included in the second category. Other texts containg several passages discussing humoral issues are a learned embryology de Spermate,14 and the physiological part of the learned encyclopaedia Of the Properties of Things (see below). Of the adjectives mentioned in MED, kyndely occurs frequently in passages dealing with humoral doctrine and natural is even more frequent as a common collocate of hete. Vnkynde also occurs in theoretical passages dealing with physiology, e.g., in Trevisa’s text. The adjective euel is found once in collocation with humour in a highly learned treatise, and wikked humours occurs in a herbal treatise. A KWIC concordance of humour in MEMT shows that hot, fiery, watery, moist, dry, black, and kind, are frequent; unkynd, corrupt, rawe, and malencolyous occur as well. All examples are connected with the physiological senses and pertain to the doctrine of humoral medicine. There are differences between individual texts depending on their field of writing, but the genres and layers of writing are otherwise fairly similar, as the examples below show. The first is from a highly learned text on technical phlebotomy from the end of the fourteenth century, the second from Chauliac’s anatomical text which has university origins, the third from one of the best-known herbals of the fifteenth century, and the last is a quotation praising the medicinal properties of aqua vite for corrupt humours in a remedy book (all emphases mine): Alexandire forsoþe commandiþ a pacient leucoflamcie to be fleubotomyed; euel humourus forsoþe if þey be in veynes ar competenly brou3t out be flebotomye. (Of Phlebotomy, p. 51) þise humours forsoþ, which bene engendred of chile in þe lyuer, as it is seid, bene double. Som bene naturale, seid of naturalitee of nutricion, Som no3t naturale. Naturale3 with þe blode bene send to al 14

Päivi Pahta, Medieval Embryology in the Vernacular: The Case of De spermate. Diss. Mémoires de la Société Néophilologique de Helsinki 53, Helsinki: Société Néophilologique, 1998.

Semantic and pragmatic meaning of humour

215

þe body to be engendred & noreshed. No3t naturale3 bene sequestrate & sent to place3 ordeyned for som helpynge3, Or þai bene sent & put out fro þe body as colre to þe chest of þe galle3, Melancolie to þe splene, fleume to þe iuncture3, þe watery superfluitee to þe reyne3 & to þe uesic. (Chauliac, Anatomy, p. 126) þis herbe drawith oute al wikked humours of a man þis herbe is hote and drye. (Agnus castus, p. 153) The nurisshyng and generacion of lys and of ycche and of hete in man is body is of humours putrified, for roten humours by nature been put oute and putrified by þe porys of þe skynne and þe mouthe of þe porys. And with corrupte humours, wormes and lis been gendrid and ycche and hete of the skynne, þe whiche vexen men and suffre not to slepe. And by cause hit is ingendrid of corrupcion, þere is noo thing in this worlde medicinal þat better and more naturally may cure hit and kepe hit fro putrefaction and consume humours corrupte þenne may oure quynte essence, yf þou haue hit redy at hande; or elles take þe beste aqua ardent þat þou may gete. (Remedies, Ferguson 205, p. 231)

Pragmatic meaning: Humour as a commonplace in various genres In cultural history, appropriation has been defined as the process by which meaning is produced and the ways in which discourses affect the reader and lead to a new form of comprehension of oneself and the world.15 I have modified the theory of appropriation for linguistic stylistics and to historical pragmatics, showing how shared cultural sets, in this case scientific doctrines, are appropriated, understood and acted upon in different ways by different audiences.16 Meanings of

15

Roger Chartier, The Order of Books: Readers, authors and libraries in Europe between the fourteenth and eighteenth centuries, Stanford, CA: Stanford University Press, 1994, pp. ix-x, and Roger Chartier, Forms and meanings: Texts, performances and audiences from codex to computer, Philadelphia: University of Pennsylvania Press, 1995, p. 89. 16 See Irma Taavitsainen, “Genres and the appropriation of science: Loci communes in English literature in late medieval and early modern periods”, in Janne Skaffari, Matti Peikola, Ruth Carroll, Risto Hiltunen, and Brita Wårvik (eds.), Opening Windows on Texts and Discourses of the Past, Amsterdam and Philadelphia: Benjamins, 2005, pp. 179-96; and Irma Taavitsainen, “Genres of secular instruction: a linguistic history of

216

Irma Taavitsainen

texts are historically constructed and sociohistorically anchored, created in negotiation between text participants in interaction. Texts have multiple, and even contrasted uses as the following examples show. Meanings and appropriations are elusive and difficult to capture, but context provides clues to the various ways people understood science and scientific ideas Matters of health are of general interest, and knowledge of the basic doctrines must have penetrated all layers of society, forming the common ground. Linguistic realisations are always based on choices made between various ways of communicating ideas to different audiences, and there are differences between genres and registers of writing. Before the development of strong vernacular and national intellectual traditions in late medieval and early modern England scientific ideas spread through loci communes or commonplaces. In my earlier studies, I have argued for different appropriations in different genres for different audiences. Humoral theory was derived from learned science and originated in academic settings, but became modified for broader audiences: details became less specific, everyday applications were added, and the underlying text type often changed from expository to instructive.17 The following examples illustrate my point. An explicit formulation of the basis of humoral medicine is found in Bartholomaeus Anglicus’ learned encyclopaedia De proprietatibus rerum, translated into English by John Trevisa in 1398. The passage below is expository: An humour is a substaunce fletinge in dede, and is ibred and comeþ of gederinge of þe element qualitees, ... For humour is þe firste principal material of bodies þat haueþ felinge and chief help in here worchinge, and þat bycause of norischinge and of fedinge. Constantinus seiþ þat þe humoures beþ iclepid þe children of þe elementis, for eueriche of þe humours comeþ of qualite of elementis. And þere beþ foure humours: blood, flewme, colera, and melencolia. (Trevisa, p. 147)

Definitions are given by two NPs joined by be or come (cf. Presentday scientific writing), and enumeration of details follows. Likewise, useful entertainment”, Miscelanea: A Journal of English and American Studies (University of Zaragoza). Special issue of ESSE7, forthcoming. 17 See Egon Werlich, A Text Grammar of English, Heidelberg: Quelle & Meyer, 1982.

Semantic and pragmatic meaning of humour

217

the example from Chauliac’s surgical treatise quoted earlier represents academic learning and displays a high degree of specificity. The following extract is from The Wyse Boke of Maystyr Peers of Salerne, a text in the Salernitan tradition of health instruction, which shows another way of dealing with the same doctrine. This extract explains the system of correspondences first, but soon reverts to advice and instruction. Applications were useful: it was good to know how to recognize the complexions by the looks or how to time taking medicine for the best possible effects. … euery man is made of iiij humours that is for to wytte of Blode, of red colorye, off blak colorye, and of flewme. Eche of thes iiij humours pursueth the bodye yat is in hys own stede … The kyndes of these iiij humours be suche. The bloode is hote and moyste and swete … The 4 humoures ben departyd be nygth and be daye/… And so in these iiij humours be conteynyd xxiiij howrys devidyd as yei ben before rehersyd. And thus these iiij humours have iiij yssuyes goyng owte of the body … And for ye tyme that any of these iiij humours before seyde ouyr konnyth evyr soo it tucyth man or woman than to grete evyll. But the so grete and comown maystres schwld weche hwmowr yt is ffor why ys the blode ouercome the humoure ... Than geffe hym drynkyes and metyes goode yn ayenste yat is byttyr, dreye, and colde.. (Peers, p. 307)

The simplest versions of the doctrine are found in verse. Applications are more prominent and rhymes make the contents easy to remember: God made all mankynd that lyues on the erthe Of iiij elementys, als we in bokys rede: Of fyre & of ayre, of watir & of erthe, That gendirs in vs humors, als Arystotille vs lers. … Of thes iiij humors ilk man is made, Bot all is nogt in lyke of qualite & state: Ffor iocund & amerous & laykand the rede is & mery, Ffleschely enowhe & synghand, myld & full hardy. The whit is splepand & slaw & redy to spit, Als fat is his face & dull is his wit. The gul is a gyloure, crabid & hardy; Quaynte & full large he is, dry & wily. The blak is a chynche, drery, & full of gyle,

Irma Taavitsainen

218

Proude & full couetus & couert he is in wyll. (Compendium of Astrological Medicine, pp. 411-2)

Conclusion Humour is a particularly good word to study from both the semantic and the pragmatic point of view, as humoral theory was a cornerstone of medieval physiology. All examples in MEMT pertain to the concrete, physiological meaning of the word. In contrast, the presentday meanings of the lexeme are abstract and positive, and a great deal remains to be done in tracing the semasiological path. The pragmatic meanings can be accessed through the context of genres and layers of writing. Learned writings deal with the doctrine in an expository way, giving definitions, whereas texts for more heterogeneous and more popular audiences revert to applications, giving practical advice and instruction.18

Works cited Agnus castus = Brodin, Gösta ed. 1950. Agnus Castus, A Middle English Herbal. Uppsala: Uppsala University. 119-164. Chauliac, Anatomy = Wallner, Björn ed. 1964. The Middle English Translation of Guy de Chauliac’s Anatomy, With Guy’s Essay on the History of Medicine. Lunds Universitets Årsskrift, N.F., Avd. 1, Bd 56, Nr. 5, Lund: CWK Gleerup. 114-142. Compendium = Mooney, Linne R. 1984. “A Middle English Verse Compendium of Astrological Medicine”. Medical History vol. 28 411-416. Gilbertus Anglicus = Getz, Faye Marie ed. 1991. Healing and Society in Medieval England: A Middle English Translation of the Pharmaceutical Writings of Gilbertus Anglicus. Madison: University of Wisconsin Press. 1-54. Of Phlebotomy = Voigts, Linda Ehrsam & Michael R. McVaugh. eds. 1984. A Latin Technical Phlebotomy and its Middle English 18

I am grateful to Turo Hiltunen and Jukka Tyrkkö for discussions about the topic. I hope to continue studies on humour with assessments of Early Modern contexts.

Semantic and pragmatic meaning of humour

219

Translation. Transactions of the American Philosophical Society 74, Part 2, Philadelphia: American Philosophical Society. 37-53. Peers = Heffernan, Carol F. 1993. “ The Wyse Boke of Maystyr Peers of Salerne: Edition and Study of a Fourteenth-Century Treatise of Popular Medicine”, Manuscripta vol. 37 No. 3. 307-317. Remedies, Ferguson 205 = Halversen, Marguerite A. 1998. The Consideration of Quintessence: An Edition of a Middle English Translation of John of Rupescissa’s Liber de consideratione de quintae essentiae omnium rerum with Introduction, Notes and Commentary. Unpublished Doctoral Thesis, Michigan State University. 207-254. Trevisa = Seymour, Michael C. et al. 1975. On the Properties of Things: John Trevisa’s Translation of Bartholomaeus Anglicus, De Proprietatibus Rerum, vol. 1 A Critical Text. Oxford: Oxford University Press, 129-162.

This page intentionally left blank

Lexicographical Lyrics (from ‘Poems Written for Translation into an Abandoned Language’)

James McGonigal The Blank Page Coming across a blank page in a book of writing like a sudden mist a clean bedsheet or a loch reflecting November I wondered was it deliberate tactics on the part of the writer like the point in a conversation when one of us conceals our thinking and we turn the page of the dialogue and overleaf there are hundreds of words to get lost in again almost forgetting the threat and promise of that moment and whether the absence was error or style.

The Upsidedown Room I have forgotten the word for a painting on the ceiling but that’s what this room was like as we entered

222

James McGonigal

a chandelier standing erect as a fountain the sole focal point on a barley white floor armchairs and sofa afloat like red clouds with cloudlets of cushions asleep in their arms back in one carpeted corner of heaven two signs of disturbance a television and a stereo hung down like bats close in to the wall.

Praise I often praise God for the glittering giftwrap of another language – often but not enough it seems for just as He lets me remember the machair and lichen word perfect just as often He casts me up on a desolate shoreline having lost all of the words but one for the smell of the waves on my jacket which we call the ‘praise’ on the cloth.

Lexicographical Lyrics

Neighbours from Heaven The smokestained sandstone of neighbouring houses like the pelt of old tabbies with black glassy eyes behind those black windows the dark lives of neighbours become clearer as night falls and windows light up and they contemplate dinner and hunting excursions through the bushes of telly and video undergrowth above us the swallows type emails to heaven about life and death matters bright trivial swift.

223

This page intentionally left blank

Notes on Contributors Carole Biggam is an Honorary Research Fellow in the Dept of English Language, University of Glasgow, and a Life Member of Clare Hall, Cambridge. She researches in the historical semantics of early mediaeval languages, in particular plant-names, colour and architectural terms. Graham D. Caie has been the Professor of English Language, University of Glasgow, since 1990 and followed Christian Kay as Head of Department. He researches in Old and Middle English Language and Literature and leads the Digitisation of Middle English Manuscripts in the Hunterian project. Julie Coleman is a Reader in the English Department at the University of Leicester. Her main area of research is the history of cant and slang dictionaries from the earliest period to the present day, but she has a wider interest in lexicology, lexicography and dictionary research in general. She founded the International Society for Historical Lexicography and Lexicology. John Corbett is Senior Lecturer in English Language, University of Glasgow. He works primarily in modern English and Scots language studies, although he also has long-established interests in Scottish literary studies. Christian Kay supervised his PhD thesis. Fiona Douglas is a Lecturer in English Language at the University of Leeds. She works primarily on modern Scots and Scottish English, corpora and newspaper language. Whilst studying for her PhD at Glasgow University, she worked part-time on The Historical Thesaurus of English. Philip Durkin is Principal Etymologist at the Oxford English Dictionary, where he leads a team of specialist editors preparing etymologies for the third edition of the OED. He has a particular interest in loanwords in Middle English and modern English, and particularly in borrowing of polysemous words.

226

Notes on Contributors

Andreas Fischer is Professor of English Philology at the University of Zurich. His research interests include historical lexicology and semantics, the language of Shakespeare and Joyce, as well as varieties of English world-wide. He is a member of the board of the Institute for the Historical Study of Language at the Department of English Language, University of Glasgow. Carole Hough is Reader in English Language, at the University of Glasgow. She specialises in Old English, onomastics and semantics, and is editor of the journal Nomina. Margaret Laing is Research Fellow in the Institute for Historical Dialectology, English Language, University of Edinburgh. She contributed to the production of A Linguistic Atlas of Late Mediaeval English (AUP/Mercat Press, 1986). She is now engaged in the creation of A Linguistic Atlas of Early Middle English in collaboration with Roger Lass. She has published extensively on early Middle English dialects and scribal systems. Roger Lass is Professor Emeritus of Linguistics and Honorary Research Fellow in English at the University of Cape Town. He has published numerous books and technical papers on theoretical and historical linguistics and the history of English. Most of his current research is dedicated to A Linguistic Atlas of Early Middle English for which, among other things, he is responsible for the Etymological Corpus. Caroline Macafee is Honorary Fellow at the University of Aberdeen. She has written extensively on Scots language, including "The history of Scots to 1700" (incorporating material by the late A. J. Aitken) in volume XII of A Dictionary of the Older Scottish Tongue. James McGonigal is Professor of Curriculum Studies at Glasgow University. He specialises in the teaching of literature in primary and secondary schools, the teaching and learning of language forms and functions in schools, Scottish cultural and creative studies and literary

Notes on Contributors

227

modernism. As can be seen from his present contribution, he is an accomplished poet. Terttu Nevalainen is Professor of English Philology and the Director of the Research Unit for Variation and Change in English at the Department of English, University of Helsinki. Her research interests include corpus linguistics and the social history of the English language. Michiko Ogura is Professor of English at Chiba University, Japan. She has published extensively on Old and Middle English philology, including Verbs of Motion in Medieval English, Verbs in Medieval English: Differences in Verb Choice in Verse and Prose and Old English ‘Impersonal' Verbs and Expressions. Jeremy Smith is Professor of English Philology at Glasgow University and Head of the Department of English Language. He researches on and teaches in Old and Middle English language and literature, the history of English and of Scots, and codicology. The focus of his current research is on Middle English linguistic studies. Jane Stuart-Smith is Reader in English Language at Glasgow University. Her research is concentrated on variation and change in Glaswegian accent, and most recently, on the role of the broadcast media in accent change. Louise Sylvester is senior lecturer in English at the University of Central England. Her research is focused on semantics , lexicology and lexicography. Her work on the Glasgow Historical Thesaurus project resulted in her first book Studies in the Field of Expectation and she coedited with Christian Kay Lexis and Texts in Early English. Irma Taavitsainen is Professor of English Philology, Director of the Department of English and Deputy Director of the Research Unit for Variation and Change in English at the University of Helsinki. Her fields of interest are medical and scientific writing, historical pragmatics and corpus linguistics.

228

Notes on Contributors

Claire Timmins is Research Assistant at Glasgow University on the project “Accent Change in Glaswegian: A sociophonetic investigation” with Jane Stuart-Smith and specialises in phonetics and sociolinguistics. She has published a number of articles on accent change with Jane Stuart- Smith. Heli Tissari is currently a Fellow of the Helsinki Collegium for Advanced Studies at the University of Helsinki, Finland. Her research concerns English emotion words, and she specializes in metaphors and cognitive linguistics. Irené Wotherspoon is Senior Research Assistant at the Historical Thesaurus. Her areas of research are lexicography, lexicology and etymology. She has worked with Christian since Christian’s arrival at Glasgow University and has papers published in the fields of linguistic computing, etymology and lexicology, some as co-author with Christian Kay.

Tabula Congratularia Sylvia Adamson William Aitken Kathryn Allan R.E. Allen Jean Anderson Wendy Anderson Jamal Ardehali Janet Bately Dave Beavan Alison Bennett Michael Benskin Carole. P. Biggam Michael Bilynsky Derek Britten Elisabeth Burr Graham Caie Gerard Carruthers Pauline Cairns Susan Castillo Thomas Chase Julie Coleman John Corbett John G.Coyle Richard Cronin Robert M.Cummings Marace Dareau Florence Davies Marilyn Deegan Hans-Juergen Diller Fiona Douglas Philip Durkin Flora Edmonds Louise Edmonds

Cathy Emmott Gwen Enstam Anthony Esposito Theresa Fanego Eileen Finlayson Andreas Fischer Caroline Gevaert Heinz Giegerich Douglas Gifford William Gillies Stuart F.Gillespie Robert A.D.Grant Manfred Görlach Lesley Haughton Antonette Healey Simon Horobin Carole Hough Werner Huellen Freda Hughes Lorna Hughes Rosemary Huisman Alice Jenkins Richard Johnstone Leena Kahlas-Tarkka Dieter Kastovsky Päivi Kilpinen Matti Kilpiö Grzegorz Kleparski Vassiliki Kolocotroni Erik Kooper Merja Kytö Margaret Laing Roger Lass

230

Tabula Congratularia

Eleanor Lawson Nigel Leask James Lonie Kathryn Lowe Angelika Lutz Paddy Lyons William Mack Margaret A. Mackay Willy Maley Pauline Maridor Robert W.Maslen Lister Matheson Caroline Macafee Donald Meek Anneli Meurman-Solin Robert Millar Lilo Moessner Elizabeth Moignard Annabelle Mooney Kath Mulvenna Jane McCawley Derrick McClure Kirsteen McCue Margery Palmer McCulloch James McGonigal Donald Mackenzie Iseabail McLeod Mike MacMahon Dorothy McMillan Frances McSparren Terttu Nevalainen David J.Newell Des O’Brien Michiko Ogura Cerwyss O'Hare

Lisa-Lena Opas-Haninen Päivi Pahta Minna Palander-Collin David A.Pascoe Carol Percy Hans Peter Adam Piette Rhiannon Purdie Scott Rae Helena Raumolin-Brunberg E. Reay Alan Riach Matti Rissanen Jane Roberts Christine Robinson Amel Salman M. L. Samuels Hans Sauer Maggie Scott Nick Selby Harold Short John Simpson Jennifer Smith Jeremy Smith Angus Somerville Daniel Soule Pauline Cairns Speitel Eric Stanley Merja Stenroos Jane Stuart-Smith Louise Sylvester Irma Taavitsainen Harumi Tanabe Freda Thornton Claire Timmins

Tabula Congratularia

Heli Tissari Nicola Trott Noriko Unebe Theo van Heijnsbergen Theo Venneman Doreen Waugh Christopher Whatley Christopher Whyte Rosie Williams Ian Williamson Keith Williamson Irené Wotherspoon Alec Yearling Judith Youngson

231