Download as Adobe PDF - Edinburgh Research Explorer - University

Download as Adobe PDF - Edinburgh Research Explorer - University

Edinburgh Research Explorer Dialectal analysis and linguistically composite texts in middleEnglish Citation for published version: Laing, M 1988, 'Dia...

2MB Sizes 0 Downloads 0 Views

Recommend Documents

Download as Adobe PDF - Edinburgh Research Explorer
Calton with their aristocratic houses and impressive public buildings, and indeed notes how .... Yester, Thurston and Go

Download as Adobe PDF - Edinburgh Research Explorer
used to dress the memorial wax effigy of Charles II (1630-1685), made shortly after his death, which survives in Westmin

Download as Adobe PDF - Edinburgh Research Explorer
by Biermann in the Bundestag on Holocaust Remembrance Day in . ... See Yitzhak Katzenelson, e Song of the Murdered Jewis

Download as Adobe PDF - Edinburgh Research Explorer - University
Wasiu O. Popoola, Ioannis Papakonstantinou and Stanislav Zvanovec. Abstract—In this paper we experimentally ..... case

Download as Adobe PDF - Edinburgh Research Explorer - University
Feb 19, 2008 - renamed 'the Rover Group' and adopted a one-brand strategy. The Austin ... From 2000 onwards the decline

Download as Adobe PDF - Edinburgh Research Explorer - University
Aug 16, 2016 - tected at Old Grantown Wood (patch 64, 5 clones) and Anagach Wood 2. (patch 66, 5 clones) each, Birkhall

Download as Adobe PDF - Edinburgh Research Explorer - University
on the challenges of managing artisanal and small-scale mining. .... suggesting that strategies to regulate artisanal mi

Download as Adobe PDF - Edinburgh Research Explorer - University
Jan 15, 2014 - Jane Kaye*,1, Matthew Hurles2, Heather Griffin1, Jasote Grewal3, Martin ... Durbin2, Stephanie Dyke2, Dav

Download as Adobe PDF - Edinburgh Research Explorer - University
music of The Unthanks to the reality TV horror show Geordie Shore. 1 Constructed, culturally mediated versions of the No

Edinburgh Research Explorer Dialectal analysis and linguistically composite texts in middleEnglish Citation for published version: Laing, M 1988, 'Dialectal analysis and linguistically composite texts in middle-English' Speculum-A journal of medieval studies, vol 63, no. 1, pp. 83-103.

Link: Link to publication record in Edinburgh Research Explorer Document Version: Early version, also known as pre-print

Published In: Speculum-A journal of medieval studies

General rights Copyright for the publications made accessible via the Edinburgh Research Explorer is retained by the author(s) and / or other copyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associated with these rights. Take down policy The University of Edinburgh has made every reasonable effort to ensure that Edinburgh Research Explorer content complies with UK legislation. If you believe that the public display of this file breaches copyright please contact [email protected] providing details, and we will remove access to the work immediately and investigate your claim.

Download date: 08. Jun. 2018

Medieval Academy of America

Dialectal Analysis and Linguistically Composite Texts in Middle English Author(s): Margaret Laing Source: Speculum, Vol. 63, No. 1 (Jan., 1988), pp. 83-103 Published by: Medieval Academy of America Stable URL: http://www.jstor.org/stable/2854323 . Accessed: 11/06/2013 07:12 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]

.

Medieval Academy of America is collaborating with JSTOR to digitize, preserve and extend access to Speculum.

http://www.jstor.org

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

DialectalAnalysisand Linguistically CompositeTexts in MiddleEnglish By Margaret

Laing

In recent years students of medieval literatureand its historyhave begun increasinglyto appreciate the value of theirprimarysource materials- the manuscripts.Editors of Middle English textsare less apt nowadays, having found their"best text,"tojettison as worthlessall other survivingcopies and renderingsof it. It is recognized thata "corrupt"textmay reflectthe activity of a contemporaryeditor, critic,or adapter rather than that of a merely careless copyist.Medieval scribes,whetherprofessionalor amateur,clerical or lay,were producing worksof literaturefor theiroriginalconsumers;close examinationof scribalbehavior,whetherit be script,spelling,or choice and ordering of material, provides insightinto the way literarytexts were received, understood, and disseminated.The studies of paleographer,dialectologist,textual critic,and literaryhistoriancan and should be complementary. The work of the Middle English Dialect Project (now the Gayre Institute for Medieval English and ScottishDialectology) in Edinburgh has, since its inceptionin 1953, upheld the usefulness,and indeed necessity,in medieval studies of examining each scribal text separatelyas a valuable source in its own right. It is obvious perhaps that in order to place geographicallythe linguisticusage of a scribe his output must be subjected to dialectal analysis. It may be less obvious that such dialectal analysisoften provides the means to do far more than place a scribe on the map. By examining the language of scribal texts,it may be possible to identify and isolate archetypaland even authorial spellings. In 1978 Ian Doyle and Malcolm Parkes demonstratedthat two of the earliestcopyistsof Chaucer's Tales worked also on manuscriptsof Gower.' Comparison of the Canterbury linguisticusage of these and other London scribesenabled Michael Samuels to identifythe most likely forms of language used by Chaucer himself. Samuels judges Chaucer's own spellings to be almost identical with those I am indebted to Prof. Michael Benskin of the Universityof Oslo for help with an earlier version of this paper and to Prof. Angus McIntosh of the Universityof Edinburgh for many useful commentsand suggestions. 1 A. I. Doyle and M. B. Parkes, "The Production of Copies of the Canterbury Tales and the

and Libraries:Essays Amantisin the Early FifteenthCentury,"MedievalScribes, Manuscripts Confessio Presented to N. R. Ker,ed. M. B. Parkes and A. G. Watson (London, 1978), pp. 163-210. 83

SPECULUM63 (1988)

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

Dialectal Analysis preservedin Peterhouse, Cambridge, MS 75. I, the sole survivingtextof the EquatorieofthePlanetis.2 The localization of separate dialectal elementswithina single scribal text may itselfbecome a means by which to identifyauthorial language. In his paper on Langland's dialect3 Michael Samuels isolates a core of linguistic usage assignable to Langland himself.This core consistsof formsdemanded by the originalalliteration,togetherwith"relict"usage - thatis, formsalien to the language of the copyistsof the survivingLangland manuscriptsbut retained by them from their exemplars, thus reflectingthe language of an antecedentstage of copying. The identificationof two regionallydistinctstratain the language of the Fairfaxand Staffordmanuscriptsof Gower led Michael Samuels and Jeremy Smithto the inescapable conclusionthatGower'sown language was a mixture of the usage of North West Kent and South West Suffolk.4 The above studies depend on the recognitionof linguisticallycomposite texts and the abilityto sort out to some extent their constituentlayers of language. The recognitionof such compositesis not new. Early thiscentury J. W. H. Atkins,followinga previous study by Willi Breier, identifiedtwo This linguisticlayersin the Cotton manuscriptof TheOwl and theNightingale.5 is a comparativelysimple case, in which the Cotton scribe was evidently workingfrom an exemplar writtenin two differenthands. He reproduced enough of the distinctlinguisticfeatures of each to make it quite obvious where theirstintsin the exemplar began and ended. Robert Thornton, the copyistof Lincoln Cathedral MS A.5.2 and British Library MS Additional 31042, another more or less accurate transcriber, provides the control in Angus McIntosh's study of the alliterativeMorte Arthure.6 McIntosh reveals two furtherlayersof language behind Thornton's transcription,the oldest layer being localizable in Lincolnshire.There is no linguisticevidence in this one survivingcopy of the poem to support the previouslyreceived opinion of a West Midland provenance for the original although, of course, such negative evidence cannot be used to discount the possibility. Linguisticanalysis may sometimesconfirm,sometimesrefute,modify,or even confuse textual or historicalevidence, but it always adds somethingto our knowledge of a scribe and his work. As the investigationsin Edinburgh have progressed, our knowledge of the detail of Middle English dialect 84

2 M. L. Samuels, "Chaucer's toNormanDavis in Honour Spelling,"MiddleEnglishStudiesPresented ed. Douglas Gray and E. G. Stanley(Oxford, 1983), pp. 17-37. Birthday, ofHis Seventieth 3 M. L. Samuels, "Langland's Dialect," Medium/Evum54 (1985), 232-47. 4 82 M. L. Samuels and J. J. Smith, "The Language of Gower," Neuphilologische Mitteilungen (1981), 295-304. ed. J. W. H. Atkins (Cambridge, Eng., 1922), pp. xxvi, xxix5The Owl and theNightingale, xxxi; W. Breier,Eule und Nachtigall,Studien zur englischenPhilologie39 (Halle a. S., 1910), pp. 49-52. 6 Englishand Angus McIntosh, "The Textual Transmissionof the AlliterativeMorteArthure," ed. Norman MedievalStudiesPresentedtoJ. R. R. Tolkienon theOccasionofHis Seventieth Birthday, Davis and C. L. Wrenn (London, 1962), pp. 231-39.

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

85 Dialectal Analysis differentiations has increased- so much so thatAngus McIntoshand Martyn Wakelin were able to isolate nineteenvarietiesof West and Central Midland English in a single manuscript,sixteen of those typesbeing perpetuated by a single hand.7 The methodologybehind thissortof studyis mostfullyset out in an article by Michael Benskin and me in the Festschriftfor Angus McIntosh.8 This articleis now substantiallyreproduced in A LinguisticAtlasofLate Mediaeval English.9The present paper is designed to exemplifysome of the basic principlesoutlined in that article. It illustratesthe steps by which we may reach conclusions about scribal practicesand a scribe'sattitudestowardshis exemplars,and it provides some clues towardsthe unravelingof linguistically composite texts. Angus McIntosh firstdefined the three possible procedures followed by a medieval scribe copying an English manuscriptin a dialect other than his own.'0 A scribemay (a) copy the spellingsof his exemplar literatim, producing an exact transcription;(b) transformthe language of his exemplar into his own kind of language, producing a complete translation;(c) produce a mixture of the spellings of his exemplar and his own spellings,creating a socalled Mischsprache, a formof language not consistentwithany one regional variety.Texts of type (b), representingthe genuine individual usage of a particularwriterof Middle English,are valuable as primarysource material for dialect mapping. Texts of type (a) are also useful where the copy has been made froma dialectallyhomogeneous text.An exact copy does not, of course, tell us about the language of the copyist,but only about the usage of the scribe fromwhose work he is copying,whetherit be mixed or homogeneous. Indeed it is impossibleto tell froman isolated example of scribaltext whetheror not it is an exact copy of its exemplar. It is only in cases (such as that of Robert Thornton, noted above) where there survive examples of several differenttypes of language writtenin a single hand that we can deduce thatthe scribeis a literatim copyist.Texts of type(c) are not usable as for dialect source material maps. primary Linguisticallycomposite textsmay oftenseem to be of type (c) - random mixturesof two or more varietiesof language. In many such texts,however, 7 Angus McIntosh and MartynWakelin, "John Mirk's Festialand Bodleian MS Hatton 96," 83 (1982), 443-50. Mitteilungen Neuphilologische in Middle English 8 Michael Benskin and Margaret Laing, "Translations and Mischsprachen Manuscripts,"So menypeoplelongagesand tonges:PhilologicalEssaysin Scotsand MediaevalEnglish Presented toAngusMcIntosh,ed. Michael Benskin and M. L. Samuels (Edinburgh, 1981), pp. 55106; hereaftercited as Benskin and Laing. 9 Angus McIntosh, M. L, Samuels, and Michael Benskin,A LinguisticAtlas of Late Mediaeval English,4 vols. (Aberdeen, 1986); hereaftercited as LALME. In the general introductionto vol. 1, sections2.2.2-4 of chap. 2, sections 2-5 of chap. 3, and the appendices are substantiallyas in Benskin and Laing. 10Angus McIntosh, "Word Geography in the Lexicographyof Medieval English,"Annals of theNew YorkAcademyofSciences211 (1973), 55-66 (p. 61).

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

Dialectal Analysis turnsout to consistof separable layersof language, the apparent Mischsprache and very often it is possible to isolate stretchesof dialectallyhomogeneous usage. When dealing with linguisticallycomposite texts it is importantto realize that one scribe may produce examples of all three kinds of copying withina single long text. In the course of studyingthe dialect materialof medieval Lincolnshire" I had reason to make detailed dialectal analyses'2of two manuscriptscontaining copies of Richard Rolle's EnglishPsalter.'3They are Bodleian LibraryMS Bodley 467 and Sidney Sussex College, Cambridge, MS 89 (D.5.3). Rolle's EnglishPsalterconsistsof his prose translationof the book of Psalms together with a commentary.The work seems to have enjoyed great contemporary popularity, judging fromthe numberof copies whichhave survivedand their dialectal distributionacross the Midlands and into the south as well as in the north of England. Thirty-fivecomplete copies of the English Psalter are known,'4 sixteen of which contain interpolationsby other commentators. Eight of these interpolatedcopies contain materialwhichis distinctlyLollard in character.In spite of its admixture in these cases withhereticalmaterial, Hope Emily Allen considered that Rolle's Psalterwas probablythe orthodox English Psalter up to the Reformation.15 Bodley 467 and Sidney Sussex 89 are both examples of Rolle's original uninterpolatedcommentary,though Bodley 467 lacks the Canticles commonly present at the end of the text. Bodley 467 is writtenin threedifferenthands; Sidney Sussex 89 is in a single hand. The four hands happen to display very clearly the types of scribal procedure described above. The followingstudies, therefore,may together serve as an exemplum, illustratinghow some scribalpracticesby comparison illuminate others, and leading to conclusions about localization of manuscriptsand the use and distributionof exemplars. 86

" See Margaret Laing, "Studies in the Dialect Material of Mediaeval Lincolnshire,"2 vols. (Ph.D. diss. Edinburgh, 1978, unpub.). 12 These were analyses of the kind designed by Angus McIntosh and Michael Samuels for the Middle English Dialect Project. They are made by means of a questionnaire consistingof a predeterminedset of categories or items for which informationis elicited from a scribal text. For the most part the items are lexical, though some are morphological or phonological. A completed questionnaireis called a "LinguisticProfile"(LP). LPs representinggenuine regional usage form the data from which dialect maps are made. These data are printedboth as maps and in LP formin LALME. 13 See H. R. Bramley,ed., The Psalteror PsalmsofDavid TranslatedbyRichardRolle ofHampole (Oxford, 1884). 14 Bramley used only twelve manuscriptsin his edition. Many more were noted by Anna C. Paues in A Fourteenth Century EnglishBiblicalVersion(Uppsala, 1902), pp. xxiv,xliv (n. 2), and li (n. 4). (The revised 1904 edition does not include this material.) Dorothy Everett,expanding Paues's work, added two more manuscripts:see Dorothy Everett,"The Middle English Prose Psalterof Richard Rolle of Hampole," ModernLanguageReview17 (1922), 217-27. This material AscribedtoRichardRolle,Hermit is most convenientlysummarized in Hope Emily Allen, Writings ofHampoleand MaterialsforHis Biography(London, 1927), pp. 169-77. 15 See Hope Emily Allen, ed., English Writingsof RichardRolle, Hermitof Hampole (Oxford, 1931), p. 3.

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

Dialectal Analysis

87

BODLEIAN LIBRARY, MS BODLEY 467

This is a manuscriptof the firsthalf of the fifteenthcenturyand is the work of three hands. Hand A is a large, clear Anglicana script16and is quite cursive. A uses a single symbol for p and y, namely,y.17 His text runs from folio Ir to folio 92v (foot); this is from the beginning of the prologue up to Psalm 72, the middle of verse 1. The text breaks off in mid-sentence:"how gude god of Israel to yaimyat ar of ryghtherte ye prophet sekand...." The dialect of thistextbelongs to South East Lincolnshire. Hand B is a smaller,squarer script,also Anglicana. B also uses a single symbolfor b and y,namely,y. His textruns fromfolio 93r to folio 120r,line 11, stopping halfwaythroughthe line; thisis fromPsalm 72, middle of verse 1, continuingthe sentence from hand A: "endelese ioy and reprehendand hymselfyat lufed erthlythyng,"to Psalm 90, verse 13: "... with stynkand smele of hys ille ensampelle slaese menne yat come nere." The language of thistextis in general of a provenance more northerlythan Lincolnshire,but it contains near the beginninga few formsof a more southerlydistribution which may be assignable to South Lincolnshire. Hand C is an Anglicana scriptwithsome Secretaryfeatures.Unlike hands A and B, C does not confuse , and y. His textruns fromfolio 120r,line 11, to the end of the manuscript,folio 171r; this is from Psalm 90, verse 13: ". .. and whit euele hand 1tis with venemose word slaas be herere ... " to the end of Psalm 150. The dialect of this text belongs to South East Lincolnshire. Under the colophon, "Explicit psalteriumdauid," and in a later hand is "Iste est liber domini Hugonis Eyton suppriorismonasteriisanctiAlbani AnOn folio 171v is "W.S [?] ex do: fa: St. Non. 1639" glorumprothmartiris." and "Liber Thomae Barlow e coll. Reg: Oxon ex dono amicissimidomini Wheate de Glimton in agro Oxoniensi armigeri III Calend: Sept: M.DC. LVI."'8 In spite of these later associationswith St. Albans and Oxford, the linguisticevidence suggests that the manuscriptis the work of at least two South Lincolnshirescribes,and possiblyof three,and by implicationtherefore that it was produced in South Lincolnshire. The language of A's text is almost certainlya translationfromthe dialect of an exemplar into that of the scribe. In other words, it is an example of a (b) type text. Commonly the language of a scribe who translatesbecomes 16 For classifications of script,see Malcolm B. Parkes, EnglishCursiveBookHands 1250-1500 (Oxford, 1969). 17 Whether or not the scribe of a late Middle English textdistinguishesthe symbolsb and y, maintainingtheirseparate functions,or confusesthemas a single(usuallyy-like)symbolprovides a powerful dialectal discriminant.See p. 93 below and n. 23; see also the followingmaps in LALME, 2: 2 "THESE," 7 "THEY," 8 "THEM," 9 "THEIR," 31 "THAN," 32 "THOUGH," 54 "THROUGH," 188 "NEITHER + NOR," and 235 "THITHER." 18 For further Catalogueof Western discription,see F. Madan and H. H. E. Craster,A Summary at 392. the 2 in Bodleian 1895), (Oxford, p. Library Oxford, Manuscripts

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

Dialectal Analysis increasinglyself-consistentas his text proceeds, perhaps reflectinga shift fromcopying by eye to copying via the mind's ear, once the scribe has got into his stride;19over the firstfew folios he may well create a (c) type or mixed text,reproducing in his copy some formswhichdo not belong to his own dialect but rather to the dialect of his exemplar(s). Indeed such relict formsmay be verycommon, and even dominant,in the early part of a text, beforebeing displaced absolutelyby what is evidentlythe scribe'sown usage. The (c) type text graduallyevolves into a (b) type. In the firstten folios of A's text there appear a few relict formswhich suggestthat the dialect of the exemplar for A was of a more northerlykind than that of A himself;these relictformsare of veryinfrequentoccurrence (see Table 1). It is of course possible that the scribe of hand A copied from a textwhose more obviouslynorthernformshad already been modified,but in the absence of any positive evidence, it is here assumed that at no point are we dealing with a scribe who produced an exact literatim copy of his exemplar.20 B writesthroughoutin a verymuch more northerlydialect than A, and its provenance is fromfurthernorth than Lincolnshire.The language is internally consistent,except that near the beginningof the block of text written by B a few formsof a generallymore southerlydistributionappear, which are displaced by theirmore northerlyequivalentsas the text proceeds. In at noted (see Table 2), the displaced forms least eleven cases of the twenty-eight seem to have been added later,oftenwithan omissionmark,and it is possible thatsome of these are insertedby a differenthand. On the linguisticevidence of B's text alone, B's behavior seems identical with that of A; in other words, B was also a translator.If this were so, it would followthat B's own dialect was of a northerntype,and the displaced forms are relict ones from an exemplar in a dialect of a more southerly varietyof Middle English. Since Rolle was born in Yorkshireand apparently spent his entire life there,21his original version of the EnglishPsaltercan be assumed to have been writtenin a Yorkshiredialect. B would then have been translatingthe text fromthe dialect of a southernizedexemplar back into a dialect more like that of the original. Such dialectal progressionis, however, 88

19A scribewho habituallytranslatesthe language of his exemplars into his own dialect clearly will make no attemptto copy letterby letter.His smallestunit of copyingwill in general be the word; and as he works into his task, translatingwithmore and more fluency,he will begin to take in larger unitsat a timebefore referringback to his exemplar.The resultis thatthe copyist works to his own dictation,using his own familiarspellingsfor each phrase as he reads it. See Benskin and Laing, ?4.2.2 and ?7.1.6. 20 There are indicationsthat such literatim copyingis rare (see McIntosh, "Word Geography," n. 13). Moreover it is evident that cursive script does not lend itselfwell to letter-by-letter copying.If the flowof cursivewritingis to be maintained,thereis considerable pressure on the copyistto make his units of copying larger - at least no smaller than the word. Cursive script and the habit of translatingbetween dialectscan be seen to go togetherin the historyof Middle English,and even a word-by-word copyistcannot be assumed to have preservedthe orthographic details of his exemplar's language. 21 Apart from a brief sojourn in Oxford, and two possible visitsto the Sorbonne; see Allen, AscribedtoRichardRolle,pp. 430-526. Writings

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

89

Dialectal Analysis TABLE1

Relictsin theFirstTen Foliosof Hand A Item SHALL -OLD (cold, hold, etc.) -AND (hand, stand, etc.) -ONG (long, wrong, etc.) OE a/6

UsualForm inHandA

of Number Attestations

Suspected Relict

of Number Attestations

schal -old

80+ 15+

sal -ald

1 1

-ond

30+

-and

2

-ong

25+

-ang

2

o

40+

a ai

3 6

ay

2

by no means unprecedented in the course of several stages of copying a text:22it cannot be assumed that the dialect of the exemplar for any given copy of a text was more like that of the original compositionmerelyon the grounds that the exemplar is, by definition,prior to the copy in the textual stemma. The view that B was a translatorwould also require thatthe exemplar for the writerof northerndialect, B, was in a more southerlytype of language than B's own dialect: the formsappearing in the earlypart of the textwhich are displaced in the later part are of a more southerlydistributionthan the formsof the later homogeneous northerndialect and would thereforehave to be accounted for as relictfromB's exemplar. The exemplar for B could then not have been in the same dialect as the exemplar forA. For the relicts in A's textare of a northerntypeof language, and were A a translator,as is here assumed, then it is his own dialect that belongs to South East Lincolnshire, whereas that of his exemplar must have been northern. A shared northernexemplar would then fail to explain the southerlyrelictsin B.

22 One example is Corpus ChristiCollege, Oxford, MS 236: Richard Misyn's translationof Rolle's Incendiumamorisand Emendatiovitae.These textsseem to have gone througha stage of copying into a more northerlydialect before being retranslatedinto a language which cannot be far differentfrom Misyn'sown Lincoln usage.

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

90

Dialectal Analysis TABLE2 Dialectal Mixtureof Hand B Item

Usual Form in Hand B

Numberof Attestations

SHALL OE a/o

sal[l] a

3 sg. pres. ind. FROM

-es -ys -s fra

40+ 5+ 10+ 30+

has ded[e] yai when -aid wyld

30+ 25+ 25+ 20+ 15+ 8

HAS DEATH THEY WHEN -OLD WOULD sg.

aa

wald

100+ 50+

6

6

Displaced Form

Numberof Attestations

schall[e] o

8 6

-ith -eth

2 2

fram fro hath deth yei whan -old wold

1 1 1 1 1 1 2 2

Bracketscondenseinformation, twoformsof equal frequency. indicating

B takes over copying fromA in the middle of a sentence,albeit at the end of a folio; and C takes over copying fromB in mid-sentence,mid-line,and mid-folio.Clearly the scribesA, B, and C were workingin close contactwith one another,and their manuscriptwas evidentlyput togetherin one place. Since two of the scribes wrote in dialects of South Lincolnshire(the dialect of C as well as that of A belongs in South East Lincolnshire,though they differslightlyfromone another), it is reasonable to suppose that the manuscriptwas put togetherin South Lincolnshire.If, as has so farbeen assumed, the northerndialect of B's textis indeed B's own dialect,then he must have been a northernscribe workingin South Lincolnshire.B's access to an exemplar in a more southerlydialect than his own (albeitan exemplar different from that of his fellow copyists)would then be easily explained; indeed all the relictsin the early part of B's textare attestedin dialectsbelongingin or adjacent to South Lincolnshire. The forms that seem to have been added later,probablyin the course of subsequent correctionof the text,could have been added as formsdirectlycopied fromthe exemplar,eitherby B himself or by some other scribe. There is no reason to doubt that scribe A translatedfrom the dialect of his exemplar with increasing consistencyinto his own: this is the obvious interpretationof the familiarshiftfrom a somewhat mixed language to an internallyconsistentusage in whichmanyof thevariantsattestedin the earlier texthave been filteredout in favorof equivalentforms.The practiceof scribe C, however, throws considerable doubt on such an interpretationof B's apparentlysimilarbehavior. C begins by producing a textin a dialect which,over the firstfour or five folios,is ratherlike the dialect of B's text (whichhas so far been assumed to

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

91 Dialectal Analysis be B's own). Afterabout folio 124 the language of C's text begins to incorporate certain formsthat are not of northernorigin,and afterabout folio 135 it has become a firmlyestablishedand homogeneous language characteristicof South East Lincolnshire. At folio 154v (the beginning of Psalm 119) there is a change in ink color. Althoughneitherthe hand nor the script changes, the language once more becomes mixed,reintroducingsome of the more northerlyformscharacteristicin hand B. Withoutat this stage taking account of the language and possible copying practice of B, the best interpretation of C's behavior is probably as follows. C had an exemplar in a northerndialect, which he began by copyingprecisely.Aftera few folios he ceased to reproduce, more or less unchanged, the language of his exemplar and produced a section of text in a mixed language containingsome forms from the dialect of the exemplar and some from his own dialect: in other words, he began to translate,but at firstonly half-heartedly.Increasingly, however, this became a thoroughgoingtranslation,and the forms of the exemplar became filteredout altogether,so that afterfolio 135 C's text is a consistentrepresentationof his own dialect. The change in ink color at folio 154v indicates a break in the continuityof copying. A possiblyquite long intervalbefore its resumptionmightwell explain the break in the continuity of the translationand the reappearance of some of those formsassumed to be from the dialect of the exemplar. This interpretationof C's behavior is the most satisfactoryexplanation for the two sections of mixed usage. To engage in exact and sustained copyingis likelyto be the resultof a conscious decision so to do: it requires close visual attentionto the spelling,as well as to the textual content,of the exemplar. C gave such attentionover a mere five folios of his text and then driftedincreasinglyfrom the usage of his reflectsthe gradual development of the scribe's exemplar. The Mischsprache independence of the exemplar's spellings,here attained over some eleven and a half folios; once used to the language of his exemplar, he copied in unitsconceptual ratherthan orthographic.A thoroughgoingtranslationrepresents the culmination of this process; here the translationextends over twentyfolios,untilthe change of ink and evidentbreak in copying.It is most unlikelythat the firstfive folios representtranslation,which only gradually (over eleven and a half folios) was abandoned in favor of an exact copy: a scribecapable of copyingexactly,once he had decided to do so, would surely switchabruptlyfromone kind of language to the other- thereis no obvious motivationfor a driftfrom translationto exact copying,whereas with the progressionfromexact copy to translation,a driftis preciselywhatone would expect. The last section (over seventeen folios), begun after some interval since the translatedsection was completed, is a Mischsprache approximating more closely to the scribe's own dialect than to that of his exemplar,and it representsthe scribe's gettingback into his stride as a translator,without, however,there being sufficientlength of text for him to regain his former consistency.I shall thereforeassume that C, like A, was a scribebelongingto South East Lincolnshire,copyingfroman exemplar writtenin a more northerlydialect. The view that C startedby copyinghis exemplar exactlyand then drifted

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

Dialectal Analysis

92

TABLE3 Similaritiesbetween the Usages of B and the FirstFive Folios of C's Text, ContrastedwithLater Usage in C B's Usage

C's Usagein Folios120r-124r

C's Usagein Folios135v-154v

THESE THOSE THEM THEIR WHICH FROM THEN THAN

yere ya, yaa yaim,yam yair[e],yar[e] ye-whylk[e] fra yan yan

bere ba ((bha)) bam bair[e], bar[e] bh/he-whilk[e] fra ban ban

bere boo hem her[e] wheche be/be-whiche, fro ((fra)) banne banne

OE a/6 -OLD AMONG -ONG BY CHURCH DIE HOLY KNOW -LESS LORD SOUL

a -ald I-mang -ang by kyrke dye haly knaw[e] -lesse lard saule

a -ald I-mang -ang by kirke dye haly knaw[e] -lesse lard saule, sawle

oo, o -old, -aid among -ong be chirch[e] die holy know[e] -les lord soule, sowle

Item

SO

swa, sua, sa

swa, sua, sa

soo, so

Doubleparentheses condenseinformation, indicatethattheformis a minorvariant.Brackets twoformsof equal frequency. indicating to a translation into his own dialect casts a different light on the previous interpretation of B's behavior, namely, that B was a northern scribe translating from a more southerly exemplar. The first few folios of C's text are linguistically very similar to B's text: they are in a northern dialect of Middle English as opposed to the South East Lincolnshire usage of C's later text (see Table 3). Such similarities in usage indicate that the dialect of C's exemplar was very similar to the dialect written by B. (There are some differences between B's usage and the usage of the early folios of C's text, which will be discussed below.) If it is assumed that the main sections of text by A and B and the third section of C's text are translations, then it follows that A was a South East Lincolnshire scribe, working from an exemplar written in a northern dialect; B was a northern scribe, working from an exemplar in a more southerly dialect (possibly from South Lincolnshire); C was a South East Lincolnshire scribe, working from an exemplar in a northern dialect, possibly the same exemplar as that for A, which was in a language similar to B's dialect; and B was probably a sojourner in South Lincolnshire. This is not a very satisfactory reconstruction. It may be conjectured that a northern scribe, B, traveled from northern England to a religious house in

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

93 Dialectal Analysis South East Lincolnshireand broughtwithhim a copy, in northernEnglish, of Rolle's EnglishPsalter.With the help of A and C, he then made a copy for the Lincolnshire communityduring his stay there; that would reasonably explain the northernrelictsin A and C and his own contributionin a dialect of northernMiddle English that differedbut littlefromthatof the book he broughtto Lincolnshire.This, however,failswhollyto account forthe southerly relictsin B's own section of the copy: these presuppose a southerlyperhaps South Lincolnshire- exemplar for B's text. That the South Lincolnshire scribes, A and C, should be translatingfrom B's northerncopy, while B himselftranslatedfroma southerlycopy of the same book to fillin the textbetween A's contributionand that of C, is thoroughlyimplausible. In these circumstancesa simplerand more plausible interpretationis that all three scribesoriginatedfromand worked in South Lincolnshireand used the same northernexemplar. A translated,only lettingthroughsome of his exemplar's forms, and those near the beginning. B was a fairlyfaithful copyist,though to startwithhe veryoccasionallywrotea formbelonging to his own dialect. The other more southerlyforms,the words thatwere added later, many as insertions,may have been writtenby B or another South Lincolnshire scribe, possibly withoutrecourse to the exemplar. C, perhaps attemptingto continuethe language of B's preciselycopied northernEnglish, began by copying fairlyfaithfullybut after a few folios had drifted into translation:it is possible that C, having taken up B's text in mid-sentence and mid-folio,sought to achieve a smooth transitionratherthan an abrupt break betweenhis language and thatof B's text,perhaps witha viewto easing the task of reading aloud fromit. There are a few differencesbetween B's text,whichas it now appears was a fairlyfaithfulcopy from his exemplar,and the firstfivefolios of C's text, indicatingthat if theydid share an exemplar,at least one of the two scribes modifiedhis copy to some extent.Since C introducedconsiderablelinguistic modificationsafter only a few folios,it is not unreasonable to suppose that even at the beginning C's copy was not precise. In the absence of the exemplar, it is impossible to determine how close was B's copy; but the odd occurrencesof South LincolnshireformsindicatethatanythingB contributed himselfis probablyof South Lincolnshireorigin,and by thiscriterionhe left the language of his exemplar verymuch as he found it. The differencesbetween the usages are as follows: (1) B uses the symboly in both [6-0] and [i-i-j]contexts,whereas C distinguishesf fromy, using p and occasionallythand ph in [6-0] contexts.Note that in four places, all withinten lines of each other,C writespe for "ye." This is quite possiblya hypercorrectionof y to p from an exemplar which did not distinguishthe symbols,and insofaras the dialect of the exemplar can be assumed northern, it is highly unlikely that p and y were there P" is one of C's regular spellingsfor"thee,"and grammatical distinguished.23 23 In the northerndialects of English (including those of Scotland and parts of Lincolnshire and East Anglia), jb and y had fallen togetherby the late fourteenthcentury,and theirseparate functionsare realized by a single symbol which may be y-likeor (less commonly)j-like in

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

94

Dialectal Analysis function notwithstanding,the hypercorrectionmay have thus been compounded. (2) B tends to use y in vocalic contexts,whereas C tends to use i. (3) In [j-] contextsB uses 3 and occasional y,whereas C uses 3 and occasional 3h. (4) There is some variationbetween B and C in the use of final-e. (5) For "not" B has noth((not,noght,notht)),and C has noth(no3t)((nat,not)).24 Neither no3tnor nat appears at all in hand B, and both formsbecome less frequent in the translatedsection of hand C, nat almost disappearing, but reappearing afterfolio 154v. Since these formsdo not occur in B's text,the postulated common exemplar cannot be responsiblefor theirappearance in C's text: they must belong to C's own usage. No3t is a common South East Lincolnshireform,and nat is attestedin two Lincolnshirescribaldialectsvery close geographicallyto where C's dialect belongs.25 (6) For "but" B has bot,whereas C has but.Either the exemplar had both bot or one or other and but,and B and C each selected onlyone of these forms,26 of B and C forthisitemtranslatedinto his own usage. Since butis C's regular usage throughouthis text,it may be taken as the usage of his own dialect. (7) For "flesh"B has flesch[e],whereas C has flessh.Since flesshis C's regular usage throughouthis text,thisformis to be taken as his own usage. Flesch[e] is to be taken as B's own usage and/orthatof the common exemplar. of hand C (folios 124v-135r, and folios 154v-171r) The twoMischsprachen are characterizedby the presence of an assemblage of formsfound in folios 120r-124r, which are assumed to be from the exemplar, and other forms found in folios 135v-154v, whichare assumed to be C's own usage. As might the interval be expected aftera long stretchof translation,notwithstanding is more like C's own usage than that between them, the second Mischsprache of his exemplar.Apart fromthis,thereis onlyone obviousdifferencebetween the two Mischsprachen. In the first,and this only between folios 125r and 134r, bei 'they' appears beside pay and tai. In C's own usage, and in the second Mischsprache, only pay and bai appear. From the evidence of other South East Lincolnshiredialects,it can be assumed that bay,pai, and pei were all currentin C's own dialect; he wrote bei alongside bai at the beginningof the translatedsection, but later evidentlysettled for the formscommon to appearance. For an account of the distributionsof b, y,and thin the late fourteenthand early fifteenthcenturiessee Michael Benskin, "The Letters(1) and (y) in Later Middle English, and 7 (1982), 13-30. Some Related Matters,"JournaloftheSocietyofArchivists 24 Parentheses enclosinga formindicaterelativefrequency.Single parenthesesdenote reduced frequency,double parenthesesa minor variant. 25 These are Philadelphia, Universityof PennsylvaniaLibrary,MS Eng. 8, hand B, fols. 147r195v, and Oxford, Bodleian Library,MS Digby 99, fols. 18r-27r. See LALME, 3, LP nos. 551 and 4289. 26 Such selection is a common scribal practice. When two functionalequivalents for a given item are of equal statusin the dialect of the exemplar and a scribe'sdialect admits only one of the twovariants,the copyingscribemaywell balk at reproducingthe exoticvariantand substitute its familiarequivalent in all contexts,thus producing one example of "constrainedusage." See Benskin and Laing, ?5.

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

Dialectal Analysis

95

TABLE4 Structureof Bodley 467 Hand

Folios

Type of Text

Dialect

Suitable as Primary Dialectal Source Material?

A B C C

1r-92v 93r-120r 120r-124r 124v-135r

SE Lincs. Northern Northern Mixed

Yes Yes No No

C C

135v-154v 154v-171r

Translation Copy Copy Imperfect translation Translation Imperfect translation

SE Lincs. Mixed

Yes No

his own dialect and that of his exemplar (bai and pay,exemplar *yai,*yay). This interpretationis supported by the restrictionof "they"formsto kai and ,ay in the firstfour folios of C, which are evidentlycopied fairlyfaithfully fromC's exemplar.27 The dialectal structureof Bodley 467 is summarized diagrammaticallyin Table 4.28

SIDNEY SUSSEX COLLEGE, CAMBRIDGE,MS 89 (D.5.3)

This is a manuscriptof the late fourteenthor earlyfifteenth and century,29 one of the oldest survivingcopies of Rolle's EnglishPsalter.It is in a single hand, though the script and language are rather changeable throughout. The linguisticfluctuationsof the text are extremelycomplicated. However, the preceding account of the behavior of the three scribes of Bodley 467, particularlythat of hand C, can be used to throwsome lighton the copying practiceof the scribe of Sidney Sussex 89. Although the manuscriptis by a single hand, the language shiftsseveral times.There are, however,no clear-cutdialectalbreakssuch as thosebetween the three hands of Bodley 467. Sidney Sussex 89 is of particularinteresthere, because its language was used as representativeof medieval Lincolnshire dialect by Samuel Moore, 27 This representsanother very frequentlyobserved example of constrainedselection in the output of a "translating"scribe. Three functionalequivalentsfor "they"are known to him and since he is happy withthe variantsthathe meets in his exemplar thereis littleincentivefor him to employ the third variant- which may well belong to his own spontaneous usage - even though he is translatingthe language of his exemplar ratherthan copyingprecisely. 28 Hands A and C of this manuscriptappear in LALME with the respectiveLP numbers 75 and 62, and their dialects appear on the maps. The dialect of B's copy has not been precisely localized. 29 H. R. Bramley dates the manuscript"towards the end of the fourteenthcentury."M. R. James and Hope Emily Allen suggest merelythat it is fourteenthor fifteenthcentury.See H. R. Bramley,ed., ThePsalter,p. xxi; M. R. James,A Descriptive in Sidney CatalogueoftheManuscripts SussexCollege,Cambridge(Cambridge, Eng., 1895), p. 73; Allen, Writings Ascribed toRichardRolle, p. 172.

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

96

Dialectal Analysis

Sanford Meech, and Harold Whitehalland localized firmlyin Theddlethorpe on the coast of North East Lincolnshire.30On this location depends the easterlycourse of theirline delimitingthe domain of the verbal suffixof the plural presentindicativein -(e)s.31This placing also has considerablebearing on their line which divides the "dialect area" of the North East Midlands from that of the Central East Midlands.32Their use of the text and the localizationtheyaccepted for its language are both misleading. Their reason for placing the language of Sidney Sussex 89 in Theddlethorpe is that the manuscriptcontains,on the second of the flyleavesat its end, copies of three Latin deeds concerningpersons and propertyin Theddlethorpe and thatvicinity.The firstdeed, whichis dated 1311, concernsthe transferof certain lands from Philip the Vauntour (?) to GilbertWyles and his wifeElizabeth,all of whombelonged to Theddlethorpe. The second deed, dated 1320, is likewisea conveyance of lands, fromthe same GilbertWyles to Robert Agge of Mablethorpe; as in the firstdeed the land in question lies in Theddlethorpe. The third deed is an undated conveyance of land from Alan of Beesby to William of Witune (?), the land being in Wold Newton. Theddlethorpe, Mablethorpe, and Beesby lie withinsix miles of each other on or near the coast of Lindsey, North East Lincolnshire;Wold Newton is some sixteenmiles to the northwest. The originalsof at least the firsttwo deeds are likelyto have been drawn up in Theddlethorpe. However, the deeds are no solid evidence for the localization of the dialect of the scribe who wrote the literarytext to which they happen to have been attached. Scribes and manuscriptstraveled,and as the preceding studyof Bodley 467 shows,a textmay sometimesbe written in a dialect of a place verydifferentfromthe place in which the manuscript The presence of the Theddlethorpe deeds is at best itselfwas put together.33 30 Samuel Moore, Sanford B. Meech, and Harold Whitehall,"Middle English Dialect Characteristicsand Dialect Boundaries," Essaysand Studiesin Englishand Comparative Literature,13 (Ann Arbor, 1935). 31 The third-person plural presentindicativeinflectionhas a complex history,forsome account of whichsee Angus McIntosh, "Present IndicativePlural Forms in the Later Middle English of the North Midlands," MiddleEnglishStudiesPresentedtoNormanDavis, ed. Douglas Gray and E. G. Stanley (Oxford, 1983), pp. 235-44, and works there cited. For the present argument it sufficesto say that the -s endings for the present plural indicative,both as major and minor variants,extend a long way furthersouth than the line in the Moore, Meech, and Whitehall map suggests.They are to be found in the most southerlypart of Lincolnshirebeside Midland -n endings. See LALME 1:467, map nos. 652-53. 32 As is clear from the studyof dialects,both modern and medieval,during the last hundred years,the idea of "dialect areas" is itselfmisconceived.Dialect maps show a continuum,where differentformsof language have overlapping distributionsand each item (and thereforeeach map) displays a differentdistributionalpatterning.See Michael Benskin, "The Middle English Dialect Atlas," So menypeoplelongagesand tonges,pp. xxvii-xli,esp. pp. xxviii-xxix,and works therecited; and LALME 1, gen. intro., 1.2.2. 33For a discussion of the movementof scribesand texts,see McIntosh, "The Textual Transmission of the AlliterativeMorteArthure,"pp. 231-40; M. L. Samuels, "Some Applications of Middle English Dialectology,"EnglishStudies44 (1963), 81-94, esp. p. 94; and Michael Benskin, 5 (1977), 500"Local Archives and Middle English Dialects,"Journalof theSocietyofArchivists 514, esp. pp. 509-10.

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

97 Dialectal Analysis evidence for the manuscripthaving been in some place where there were in theconveyanceof propertyat Thedpersonsinterested,perhaps indirectly, dlethorpe.As the manuscriptcontainsa learned religioustextof a sortalmost certainlythe product of a religioushouse, it was perhaps owned in Theddlethorpe and copies of the deeds entered by a local owner on its previously blank flyleaves;there was no medieval religious house in Theddlethorpe, and it is unlikelythat the manuscriptwas put togetherthere. The prioryof Alvinghamand the abbey of Louth both owned land at Wold Newton, and Revesby Abbey (twentymiles to the southwest)owned Theddlethorpe rectory;it could have been writtenin any of these houses, and quite possibly the copies of Theddlethorpe deeds were there bound in as the original flyleaves. Wherever the manuscriptmay have been put together,there is no guarantee that the scribe who wrote it was originallynative to that part of the country.The linguisticcompositionof Sidney Sussex 89 is in factextremely complicated,as willappear fromthe followingaccount,and used uncritically, it is veryunsuitable as a primarysource text for dialect mapping. However, as withhand C of Bodley 467, it is possible to isolate at least one section of textas being in a homogeneous local dialect. Sidney Sussex 89 is a large manuscriptwith two columns of text to the page. My study has for the most part been from a complete microfilmin Edinburgh UniversityLibrary.The folios are not numbered; and the text, not counting the flyleaves,runs to 371 frames. References hereafterare counted from frames on the microfilm,by columns of text, and by Psalm numbers. The earlypart of the manuscriptis in a Textura script,whichlaterbecomes less formaland more cursive.The mostobvious single change is in the shape of the letterd, which is at firstTextura witha simple, straightascender, but later changes to Anglicana with a looped ascender. On other letters,too, ascenders become more prominentlylooped. In spite of thisdriftto cursiveness, the manuscriptis evidentlywrittenall in one hand, apart fromtwo very shortinterpolations,less than half a column long, by a singledifferenthand. These occur on frames 166 (Ps. 62.8-9) and 189 (Ps. 70.1). From frames1 to 25 (prologue up to Ps. 9.26) the language is of a northerly type,fromYorkshireor possiblyNorth Lincolnshire.I shall call thislanguage L 1. At frame 25 forms of a more southerlydistributionbegin to appear beside the northerlyforms.At frame 30 (Ps. 11.1) the scribebegins to write Anglicana d beside Textura d. By frame32 (Ps. 12.8) the transitionfromone lettershape to the other is complete and Textura d does not appear again. At the same time as the scribal mode is changing, the language is also changing. Over the next few frames,more and more nonnorthernforms appear, and by frame 36 (Ps. 14.1) the language has shiftedcompletelyand become a fairlyself-consistent language of a very differentcharacter from frames(see Table 5). I shall call this language L that of the firsttwenty-five 2a: it continues as far as frame 85 (Ps. 34.2). Within L 2a there is an internal shiftin the form for the first-person singularpersonal pronoun. In L 1 and earlyL 2a it is writteni, but at frame

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

Dialectal Analysis

98

TABLE5 betweenL 1 and L 2a Some Differences

L 2a

Item

L1

THEY THEIR SUCH WHICH MUCH SHALL sg./pl. TO prep. -AYin day,may, say,way PRAY HAS pres.part. CHURCH -LY

yai,yei yaire,yeire swilke,swyche ye-whylke mekil,mekille sal ((shal)) tille,to -ay

yei((yaionce)) yeire,yeir siche(and variantsin -ch-) swyche, ye-whiche (and variantsin -ch-) meche((mekil)) schalle,schal,shalle,shal to -ey

pray has,haues -and,-ande kirke -ly,-li

pray,prey hath,hathe -ond,-onde,-and,-ande((-end,-ende)) chirche -liche,-ly((-li))

denotea minor variant. Doubleparentheses 62 (Ps. 24.12) "I" begins to be writteniche; the -cheis usually scrubbed or crossed out, whetherby the original scribe or some other is uncertain.Otherwise,L 2a is internallyconsistent,except for the spellingof "shall," which willbe considered later. Matters of internal consistencyapart, there is a furtherusage which requires comment. The spellings for "are" in L 2a remain unchanged from L 1, namely,er[e].In the lightof other developmentsin the language of this manuscript,this continuityis difficultto account for and will be considered later. The next linguisticshiftoccurs at frame 85 (Ps. 34.3). I shall call this language, which continues to the end of the manuscript,L 2b, since, as will appear, it is not fundamentallydifferentfrom L 2a. Evidentlythere was a break in the continuityof copying at this point. In the left-handcolumn of frame85 is a small block of textin a neater,smallerwriting,whichon analysis of the script,34 however,proves to be the same hand as thatof the restof the text. It is as though these few lines had to be insertedlater into a gap that had been leftfor them,but which proved slightlytoo small to contain them. Moreover, while the inserted text has an average of nine words per line against the previous text's seven, the text followingthe insertionis in an appreciablybigger and more widelyspaced writing,having only fiveor six words per line. From the break at frame 85 "are" startsto be writtenar[e] instead of er[e]. Withinten framesar[e] has become the dominant form,and afterthat er[e] is extremelyrare. This shiftcoincides withthe completionof a change from 34 For an outline of some possible taxonomic procedures for the classificationof differences and similaritiesin scripts, see Angus McIntosh, "Towards an Inventoryof Middle English 75 (1974), 602-24. Scribes,"Neuphilologische Mitteilungen

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

Dialectal Analysis

99

shal[le] to schal[le] 'shall'. Shal[le] is the regular usage of early L 2a, but schal[le]appears sporadicallyfromframe42 (Ps. 17.4), increasinglydisplaces shal[le],and finallyousts it completelyat frame 83 (Ps. 33.9). Thereafter schal[le]is the regular usage of L 2b. From the break in copyingat frame 85 (the end of L 2a), the language graduallybecomes more mixed. Firstly-ay begins to appear beside -eyin "day," "may," "say," and "way." As the text 'which'; kyrke proceeds, occasional forms such as yai, yay 'they'; ye-whilke which are of all 'has' haues sal suilk 'shall'; 'such'; 'church'; swilk, appear, characteristicin L 1. Other forms,also of a northerlydistribution,and which do not occur in L 1, appear beside them in L 2b: yam 'them'; hali 'holy'; and saule 'soul'. Between about frames 185 and 200 iche'I'; -liche'-ly'; and -ondepresent participleending become rarer and almost disappear.35Moreover by the end of the manuscriptyai 'they'and -ayin "day", "may", "say", and "way" are as common or commoner than are yei and -eyrespectivelyin these contexts.L 2b has thus become much more like L 1. A broad viewof the nature of theselinguisticshiftssuggeststhatthe Sidney Sussex scribe'sbehavior was similarto thatof Bodley 467, hand C. He began by copying and then driftedinto translation;aftera break of an indeterminate period he resumed his copy and thereafterproduced a textin a mixed language, partlyhis own usage, and partlythat of his exemplar. However, on several counts this interpretationproves to be oversimplified.It fails to account for the followinganomalies: (1) For "are" the form in L 1 and L 2a is er[e]. In L 2b it is ar[e]. If er[e]is taken to be the usage of the exemplar (representedby L 1) and of the Sidney Sussex scribe himself(representedby L 2a), then ar[e] is inexplicable. Even if it is assumed that the Sidney Sussex scribe'susage admittedboth er[e]and ar[e], he would have no motive to shiftfrom a formcommon to his dialect and that of his exemplar to a contrastingform,when for other items his language approximates more closelyto thatof his exemplar. (2) The completionof the shiftfromshal[le]to schal[le]at the same point in the text as that of er[e] to ar[e] is inexplicable for the same reason as in (1) above. (3) In L 2b yam'them'; hali 'holy'; and saule 'soul' are assumed to belong to the exemplar's language since theydo not appear in L 2a. However, neither do theyoccur in L 1, in spite of abundant contextsin whichtheycould have appeared there. 35 Possiblythe earlier erasures of -chein "I" (L 2a) were an attemptto regularize the usage overall. This would assume thatboth i and ichewere known variantsin the dialect of the scribe of Sidney Sussex 89 but thatichewas his preferredform.He began to introduceiche,along with other formsfrom his own dialect, as his copy progressed. But the complete absence of ichein his northernexemplar later influenced him to select only the i variantshe found in frontof him. Another possible explanation of the scribe's confusion over the form of "I" remains completelyhypothetical.The dialect of his exemplar may have had ichfor "I" before vowelsbut i in other contexts.(Compare the early northernMiddle English use of ik before vowels and hand of i, I elsewhere.) Encounteringsuch a formin vocaliccontextsmayhave caused the present scribe's initial switchto his own preferrediche. But his own dialect did not make the formal contextual distinction,and as his copy progressed, he preferredthe consistencyof using the more frequentlyencountered i.

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

100

Dialectal Analysis

These details of the linguisticshiftfrom L 2a to L 2b suggest that there has been a change, albeit not great, in the dialect of the exemplar. The absence fromL 1 of yam,hali, and saule,and fromthe Sidney Sussex scribe's own usage as attestedby L 2a, argues such a change, and this view is supported by the evidence of the shiftfromer[e]to ar[e] and perhaps also by the completionof the shiftfromshal[le]to schal[le],which are otherwiseexceedinglydifficultto account for. It need not be supposed that a major change had occurred in the language of the exemplar. It maywellbe thattwoscribes, both of whom had acquired their habits of writtenlanguage in much the same area, cooperated to produce the exemplar,just as the scribes from South East Lincolnshirecollaborated on Bodley 467. Or it may be that the fromhis exemplar was by a single scribe,who copied more or less faithfully own exemplar the language of such a collaborativeeffort.(There is some appeal in the viewthatthe SidneySussex scribe'sexemplarwas thecomposite: the point at which one scribe took over from another would provide an obvious place to break offfor a copyistnearing the end of his day's work. A change of hand, and perhaps thereforea change in size of script,in his exemplar mightalso have influencedthe Sidney Sussex scribe'senlargement of his scriptin L 2b.) The linguisticevidence is adequately accounted forby the followingreconstruction.The Sidney Sussex scribemade his copy froman exemplar written in two very similar,but not identical,northerndialects; the firstunderlies Sidney Sussex 89 frames 1-85, the second, frames85 to the end. The Sidney Sussex scribe evidentlybegan by reproducing the language of the firstpart of his exemplar (L 1), but afterabout twelvefolios(frame 25) began to drift into translatingmode. He produced a linguisticallymixed text for two or three folios,but soon got into his strideas a consistenttranslator(frame30), the driftis producing language L 2a. Given the interveningMischsprache, to have been from to translation for the same reasons as unlikely copying those discussed in the account of Bodley 467 (above, pp. 90-92). The linguisticshiftfrom L 1 to L 2a accompanies a change in the mode of script. The very careful scriptof the earlier text graduallybecomes more cursive. This change of mode would be consistentwitha change fromword-by-word copyingto copyingby the mind's ear, thatis, fromcopyingto translation(see pp. 87-88 and 90-92 above). It is here assumed thatL 1 is copied language and representsthe language of the exemplar and thatL 2a is translatedand is by and large the copyist'sown usage: L 2a is a homogeneous dialect which is placeable in South West Lincolnshire. These assumptions are consistent withthe linguisticevidence of the interpolatingscribe'sshortcontributionof frame 166 (Ps. 62.8 and 9). The second interpolatedpassage on frame 189 (Ps. 70.1) is too short to provide any useful informationhere, but the first containsyai 'they'; sal[le] 'shall'; and tile (prep.) 'to', common in L 1 but in L 2b only found veryrarely,and even then rightnear the end of the manuscript.The scribe of the interpolationsevidentlyreproduced the formsof the later exemplar (that underlyingL 2b, here assumed to be common to both scribes), having no opportunityin such a short stretchof text to drift into translation,even supposing that was his habit; and the enclosing text,

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

Dialectal Analysis

101

by the main hand, contains the contrastingformsof the main scribe's own dialect,alongside those of the exemplar. The factthatL 2b becomes increasinglymixed cannot be explained whollyin the same way as was the second of hand C in Bodley 467 (see pp. 90-92 above), although there Mischsprache is a similarbreak in the continuityof copying. On the contrary,the Sidney Sussex scribe had eighty-fivefolios of text after his own break in which to regain his consistencyof translation,but the language becomes increasingly more rather than less like that of his exemplar. In these circumstancesthe best explanation is that spellingswhich had by now (afterfortyor fiftyfolios of copying) become very familiarto the copyistof Sidney Sussex 89 were intrudinginto his own active usage. This interpretationis the strongerbecause all the formsthat are in L 2b but not in L 2a are currentin dialects adjacent to the area of origin for L 2a, and there are no formsin L 2b that are not attestedin that area. If we assume that the language of the exemplar for Sidney Sussex 89 changes in some respects at the point in the text where L 2b begins, the change in usage fromer[e]'are' to ar[e] is now easy to explain. The exemplar underlyingL 1 and L 2a had er[e],whereas that underlyingL 2b had ar[e]. Both er[e] and ar[e] are current in dialects geographicallyadjacent to L 2a and probably also, therefore,in the Sidney Sussex scribe's own usage; he thus perpetuated in his copy whicheverhe found in his exemplar. The change of shal[le]to schal[le]may now be explained as follows.Sal is regular usage in L 1, though occasionallyshal appears also. Shal[le]is regular usage in L 2a, though schal[le]begins to be writtensoon after the change fromL 1 to L 2a and gradually displaces shal[le].Shal[le]does not appear at all in L 2b. If we assume that both the sch-and sh-variantswere familiarto the Sidney Sussex scribe- and thisassumptionis supported by the appearance of both formsin dialects geographicallyadjacent to L 2a - and that his exemplar had sh-but not sch-,the Sidney Sussex scribe'sdriftfromsh-to sch-is a driftfromtoleratedto preferredusage. Sal was also in his exemplar but not at all in his own dialect;36. and accordinglyit does not persistin his text afterthe closely copied section of L 1. The scribe for the later part of the exemplar may or may not have used sal: it is whollyabsent from L 2b. The absence of sh-spellingsin L 2b, giventhe SidneySussex scribe'stoleration of them in early L 2a, indicates that this second scribe did not use them at all; his formswere evidentlysch-variants,and possiblys + vowel variants also. The linguisticstructureof Sidney Sussex 89 is summarized diagrammatically in Table 6.37 To conclude, L 2a approximates to the Sidney Sussex scribe's own lan36 Althoughin Middle English dialectsoverall,s + vowel usuallyoccurs withsch-for [f] rather than withsh-,thereare some examples in Northand Central Lincolnshireof s + vowel occurring withsh-.See LALME 1:340-41, map nos. 144, 145, and 148; LALME 2:95, 101, and 107, sheet 3 of map 22, "SHALL sg and pl" and map 23, "SHOULD sg." 37 Language L 2a appears in LALME withthe LP number 46. This homogeneous section of Sidney Sussex 89 has been used to provide data for South West Lincolnshireon the maps.

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

Dialectal Analysis

102

TABLE6

Structure of SidneySussex89 Language L1 L 12 L 2a L 2b

Frames 1-25 25-36 36-85 85-end

TypeofText

Dialect

Suitable as Primary Dialectal SourceMaterial?

Copy Imperfect translation Translation Imperfect translation

Yorks.(?) Mixed

Probablynotpure enough No

SW Lincs. Mixed

Yes No

guage, but his spontaneous usage may include ar[e] 'are' and schal[le]'shall' in greater frequencythan theyappear in L 2a. Although L 2a is a homogeneous dialect, the relative frequencies of some of its formsmay not correspond preciselyto those of the Sidney Sussex scribe's spontaneous usage; some of these may approximate more closelyto those of L 2b.

The view that linguisticallythe majorityof Middle English texts present merelyMischsprachen, perhaps encouraged byJ. R. R. Tolkien's observations on early Middle English,38depends on the assumption that the distribution of linguisticforms in these texts is generally random. If, for instance,an index of formswas compiled for Sidney Sussex 89, or for hand C of Bodley What would not be clear 467, theywould indeed appear to be Mischsprachen. is thattheyeach contain two differentlayersof language (as well as mixtures of the two) and that withinthese the distributionis not random. In other words, not all later Middle English texts are hotchpotch; there are many whichmay be linguistically apparent Mischsprachen layered.One of the direct resultsobtainable from the analyses on which this paper is based is that we are able to isolate dialectallyhomogeneous componentsof much-copiedtexts and show that they are no less valid for dialectal studies than the most methodicallywrittenholograph. The corollaryto this,as illustratedby the Sidney Sussex scribe and hand C of Bodley 467, is that a single scribal text may give evidence for more than one Middle English dialect. In the case of Sidney Sussex 89 the benefitsto the dialectologistof this degree of analyticdetail are clear: the text is removed from its erroneous position in Theddlethorpe, and a section of its usage is confidentlyplaced some fiftymiles to the southwest.This sort of studymay sound a caveat to the studentwho wishes to use dialectal analysisfor the quick and easy localization of a manuscript; the process is not always simple. But it has the advantage of granting us insightsinto other matters.The Sidney Sussex scribe gives evidence, albeit through the veneer of his own usage, of an 38J. R. R. Tolkien, "Ancrene Wisse and Hali Mei6had," Essaysand Studies14 (1929), 104-26. But cf. Benskin and Laing, ?7.2.

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions

Dialectal Analysis

103

antecedent stage of copying involving two differingstrands of northern language. The three hands of Bodley 467 illustratethe way in which three scribesmay share the work of copyinga book and in the process affecteach other's copying practice. Hand B provides fairlyaccurate knowledge of the usage of their shared exemplar. These are small steps in the building of textual histories.But if similar analysis were given to the other seventeen uninterpolatedtextsof Rolle's EnglishPsalter,it would be surprisingif knowledge of theirtextual relationshipswere not considerablyaugmented. Investigationsof this kind may induce in some of us the exhilarationof a Sherlock Holmes, for whom the elucidation of the apparentlycomplex is its own reward. However, scholars tend, understandablyenough, to be more interestedin the textsand manuscriptstheythemselvesare workingon than in other people's. Now that A LinguisticAtlas of Late Mediaeval English is published, the data and methodologyin it are available to everyone.Those who care to will be able to carryout similarinvestigationson any manuscript or textthat may interestthem. Such studies need no longer be the preserve of the dialectologistor linguistichistorianalone; the integrationof different disciplinescan only help to increase our knowledge and understandingof manuscriptsand theircontents.The principlesillustratedhere are of service also to the editor and the textual critic,for it is the application of such principles to "unworthy"texts as well as to "good" ones that may lead to freshinsightsinto textualhistoriesor authorialspellings.In the words of Ian Doyle: "It is impossibleto pursue manuscriptstudies nowadays satisfactorily in individual isolation,for one cannot findall one ought to know by oneself and one ought not to keep all one knowsto oneself; thejigsaw puzzle we are all workingon is so big that it may need the help of everyeye to tryto fita piece in it."39

39A. I. and Readersin Fifteenth-Century England, Doyle, "Retrospectand Prospect,"Manuscripts ed. Derek Pearsall (Cambridge, Eng., 1983), pp. 142-46.

Dr. MargaretLaing is a memberof theResearchStaffof theGayreInstitute for Medieval ofEdinburgh, EdinburghEH8 9LN, U.K. University Englishand Scottish Dialectology,

This content downloaded from 129.215.52.164 on Tue, 11 Jun 2013 07:12:08 AM All use subject to JSTOR Terms and Conditions