SELIM08.doc Isabel de la Cruz, Selim 8 (1998): 147—156 STUDIES IN THE LANGUAGE OF SOME MANUSCRIPTS OF ROLLE’S EGO DORMIO1 INTRODUCTION Ego dormio was written by Richard Rolle of Hampole, a hermit in Yorkshire, who was born about the year 1300 at Thornton Dale, near Pickering. He lived, then, between c. 1300-1349 and is particularly known because of his various religious writings. No extant copy of his writings goes back farther than the last quarter of the fourteenth century. Ego dormio was allegedly written for the spiritual enhancement of a lady. In some of the manuscripts the text is addressed to a friend or a nun, but in others there is no addressee. Although we are ignorant of who the recipient of this treatise could be, recently it has been suggested it might have been written for a secular lady, possibly Margaret of Kirby (Ogilvie -Thomson, 1988: lxvii & Watson, 1991: 330). There are several extant manuscripts of Ego dormio of which the most relevant ones are kept in: 1. London, The British Library: Arundel 507 Additional 22283 (Simeon MS) Additional 37790 2. Cambridge: Cambridge University Library Dd V 64 Magdalene College, Pepysian 2125 3. Oxford, The Bodleian Library: 1 M y t hanks to Dr Jeremy Smith, from the University of Glasgow, who read a draft of this paper and improved it with useful suggestions. Any shortcomings, of course, remain my own. Isabel de la Cruz 144 Rawlinson A 389 which contains two different versions of the same text, usually known as Rawlinson 1 and Rawlinson 2 English Poet a 1 (Vernon MS) There are some others in: The Library of the Marquess of Bath: Longleat 29 Dublin, Trinity College Dublin: MS 155 London, Westminster School: MS 3 Paris, Bibliotheque Sainte Geniève: MS 3390 Tokyo: Takamiya 66. This manuscript, which is referred to by Allen (1931) as the Gurney MS, because it was owned by Hudson Gurney of Keswick Hall in the 19th century, was on extended loan to the Fitzwilliam Museum in Cambridge and was then known as Bradfer-Lawrence 10 (e.g. in Amassian 1979, Amassian & Lynch 1981 and Ogilvie-Thomson 1988 appears with this name). However, it is now in the possession of Professor Takamiya from University of Tokyo, so we will refer to it as Takamiya 66. All these texts are written in English. There is just one Latin translation of Ego dormio which is extant in manuscript Gonville and Caius College 140/180. Many of the English texts have much in common, as some of the most important works by Rolle can be found in the same manuscript. The text in Cambridge University Library Dd V 64 has often been considered the most authoritative. In fact, it is the most widely used for editions: - Allen (1931: 61-72) used it emended with reference to Rawlinson 1; - Horstmann (1895: 50-61), which used the same manuscripts as Allen, but also emended with Rawlinson 2, Bodleian English Poet a 1, and British Library Arundel 507; - Ogilvie-Thomson (1988: 26-33), transcribed the manuscript Longleat 29, though. Apart from the publications mentioned above, no editions have been made from the other manuscripts in the British Library, Magdalene College, Trinity College Dublin, Westminster School (London), Bibliotheque Sainte Geniève (Paris) or Tokyo (Takamiya 66). Studies in the language of some mss of Rolle’s E go Dormio 145 For the Latin text Amassian and Lynch (1981) used the only extant manuscript in Latin and compared it with the edition made by Allen based on Cambridge University Library Dd V 64. A modernized version of the text can be read for instance in Heseltine (1930: 89-100) and Colledge (1962: 143-154). In this article I will concentrate on three of the English manuscripts: Cambridge University Library Dd V 64, Longleat 29 and Trinity College Dublin 155. The last differs from the others in a very significant way. Although the question of source manuscript has not been completely solved, in most manuscripts the text is basically the same with minor textual varia- tions. The ones which modify the original text considerably are: Trinity Colleg Dublin 155, Takamiya 66 and Pepysian 2125 slightly. We cannot date Ego dormio exactly but, as Rolle seemed to have died in 1349, the text must have been written sometime before. Allen considers the approximate date could be 1343 when Rolle used alliteration for his Gastly Gladnesse (1931: 60). The manuscripts I am going to deal with were either copied in the late 14th century: Cambridge University Library Dd V 64; beginning of the 15th: Trinity College Dublin MS 155 or some time later in the 15th century (second quarter of the century): Longleat 29. Since the span of time from the first to the last one is not so wide, one should not expect the text to differ, but for the local varieties used by the dif- ferent scribes who copied them. In this way, linguistic variation between them is likely to reflect the choice of different regional forms by scribes coming from several parts of the country. As Laing (1992: 568) has suggested “texts surviving from the same period in more than one version can be of great help to the historical dialectologist”, since the comparison of different copies of a single text turns out to be an excellent way of identifying dialectal discriminants. The more parallel texts there are available, the better, as they are likely to help in supplying a full range of dialectal discriminants. However, in addition to the language, which I will be referring to later, the Trinity College Dublin MS 155 diverges from the others in a number of ways. Apart from some passages which are obviously deviant, like the omis sion of some lines appearing in the other manuscripts, the lyrics also show differences: the first lyric is divided into two, a new lyric is inserted almost at the end of the text and the second lyric is omitted altogether, so the text ends before the others. And what is unique about Trinity College Dublin MS 155 is Isabel de la Cruz 146 that the text departs in such a way from the others that it cannot be consid - ered just a copy, but a version of the original one. The meaning is preserved, but there are many omissions, interpolations and unique variants. The rea- sons adduced by Ogilvie -Thomson for this alteration is that it seems to be an attempt to adapt it for a male person. So, there are many masculine pronouns in the expansions, and the details of women´s clothing and the referencies to wooing, marriage or virginity have been either modified or eliminated (Ogilvie- Thomson 1988: lxx). That explains why syster, which appears in Cambridge University Library Dd V 64 and Longleat 29, is replaced with frend in Trinity College Dublin 155. METHODOLOGY The text in Trinity College Dublin MS 155 has been read from the original manuscript, photocopies and microfilm and the other two from enlarged mi - crofilm-prints. Reliable printed editions have also been used, such as the one made for the Early English Text Society by Ogilvie-Thomson from MS Long- leat 29 and Allen’s edition from Cambridge University Library DdV64. The methodology follows the original guidelines established by McIntosh and Samuels in conjunction with the production of A Linguistic Atlas of Late Mediaeval English (1986: I, 7). This way, several items are selected to collect the different forms. The chosen items comprise four classes of evidence: 1) purely graphological, like the use of thorn or th in words like thai/pai; 2) phonological, to examine some specific features like velarization versus palatalization as in whilk /which; 3) morphological, where the ending for the third person singular can be in thorn or -th/-s, as in hath/has and 4) lexical: poison/venym. (The latter feature is not so much represented in the text, as being such a short passage it shows not many different synonyms for the same concept). As mentioned in the introduction, Rolle was born in the North of England, so the typical features of the Northern dialect should be expected in his writings, such as the use of Scandinavian pronouns for the third person plural, ending -and in present participles, preservation of Old English long a, -s ending for the inflection of verbs in the third person singular and plural, etc. However, by reading, editing and analysing the language contained in the three manuscripts, it soon became clear the fact that the features in some Studies in the language of some mss of Rolle’s E go Dormio 147 of the texts were far from the Northern origin of the author. Thus, not taking anything for granted, I tried to find the answer to questions like: What kind of distributions are we going to find for the reflexes of Old English /y(:)/, u, i or e? or Do the h-forms occur for the third person pronoun? It is widely accepted now the existence of Mischsprache, which is charac- terised by the combinations of forms for the same item. According to the the group of Edinburgh and Glasgow Universities working on the Middle English Dialect project, who were the first to realize how medieval scribes treated the language of their originals, one of the three things may happen (Benskin & Laing 1981: 56, McIntosh & Samuels 1986: I, 13): A) The scribe makes a literatim copy, that is to say, he may leave the lan- guage as it appears in the original manuscript without making any alterations. B) He converts the language into his own dialect. C) He makes something between A and B. They pointed out how infrequent the first case was, so one can well imag- ine a scribe copying Ego Dormio and adding a few dialectal forms into the original, as scribes were used to translating between one variety into another. Subsequently, medieval scribes preserved some of the original forms of the manuscript, while they converted or translated others into the dialect to that part of the country they came from or they were working in. I seek to explain why the scribes of these manuscripts made the spelling choices that they did and to attain this purpose, it was useful to select some functional words, like: as, if, when, the, though, through, etc., but also other lexical units that can provide useful data from which to deduce which variety of Middle English the scribe is using. Thus, verb s in the third person singular, present participles and other words which can determine the dialect, such as the presence of velarization versus palatalization in words like mykel versus much, the reflexes of Old English long a, as in holy and other items which can be productive, such as sal versus shall, etc. are used as test items. By focusing on those features that are dialectally distinctive and that are essential to tell the difference between one Middle English variety from an- other, I tried to investigate the spelling practices of these three medieval copyists of Ego Dormio to discover the individual scribal practices of each of them and see whether any conclusions about the provenance of the scribe could be drawn. Isabel de la Cruz 148 FINDINGS It is a well-known fact that the demand for some books during the four- teenth century (and late thirteenth century) was such that the old scriptorium system was unable to cope with a wider readership. Outside workers had to be hired to increase the speed of copying texts and thes e scribes produced manuscripts with their own spelling systems, carrying out a scribal transla- tion from one dialect into another. There is evidence that such scribal translations became more and more common as the Middle English period progressed. The reason given by Smith (1992: 55) is that “literacy in the vernacular was becoming much more widespread in the thirteenth and fourteenth centuries. In such circumstances, the old monastic orthography and the discipline associated with it must have been impossible to sustain. New spelling systems, based upon practices de- veloped in individual schools and parishes rather than on those of a few monastic centres, came into use”. However, there is still no spelling norm to imitate, which explains the spelling variants found in the different areas. By reading the text it soon became evident that scribal translations in- creased, as the Middle English period progressed and the old scriptorium- system broke down (Benskin & Laing 1981: 88-91). Regional orthographies exerted influence because there was no fixed spelling-norm to imitate, and the standardisation of written English in the modern sense did not arrive until printers provide a normative model in the sixteenth century. Some of the scribal spelling practices can be perceived in the following table, where the modern English equivalent is given next to the form which appears in the manuscript. The first form reflects the most common spelling found while the frequency of the other forms varies from just one instance to several. The items were selected according to the criteria established before, so we could have evidence of every feature (graphological, phonological and grammatical): ITEM CUL Dd V 64 Longleat 29 TCD 155 self self self selfe as als as as if if if, jf ?if both bath bothe, both bope but bot bot but Studies in the language of some mss of Rolle’s E go Dormio 149 the pe pe, the pe it it hit hit (it) his/him hys/hym his/hym his/hym, him they pai pai, pay, pei pei them pam ham hem their peir, pair har her thy pi, py(n) pi, thi, py(n) pi some some sum some when when, whan when when, whan though tho though pou? through thorow progh, throgh pour?, poru? which whilk whiche, whoche whoche such swilk suche suche much mykel mich, mych(e) muche(l), michel, myche church kyrke chirch churche flesh flesch(e) flesche, fleishe flesche shall sal shal schal Present participles -and/ing byrnande, feland, havand, liftand, suffrand, praying, thynkyng, syng- yng, etc. brennynge, stan- dyng, sittynge, goynge, doynge, dremynge, etc. breininge /brennynge, sit tynge, goynge, doynge, etc. 3rd person sg. -s/th haldes holdeth aske°, knowe p (exception are: holdes, schewes) holy haly holy hol(l)y love lufe, lofe loue luf(e), loue evil evel, evyl euel yuel (euel) busy/busyness bisy/bisynesse, bysy besy/besyness, bisy, bisiness bisy, bysy, bysynesse mirth myrth myrth, mirth myrpe, mur pe lust lyst lust(e) lust give gif(e) gif, gyf, yif ?if, ?efe thing thing thyng(e), pynge ping, pyng CONCLUDING REMARKS Isabel de la Cruz 150 Obviously, all these characteristics demonstrate that the language of some of the texts is far from being that of Yorkshire. The team of the Linguistic Atlas of Late Mediaeval English analysed some folios belonging to these manuscripts: In the case of Trinity College Dublin 155 they reached the conclusion that because of the linguistic features of the text, it was copied in Staffordshire. Regarding Longleat 29, they did not analyse this piece, but many others by Rolle and decided it was copied according to the style in Ireland. Previously McIntosh and Samuels (1968) had suggested that the language was Anglo -Irish. Ogilvie-Thomson (1988: xxxiv), however, argues that the language shows some of the typical characteristics of this kind of language but many others are absent. For this reason, she concludes that “the sum of these features points out to the standard fifteenth-century literary language based on the East Midland dialect(s), with sprinkling of South-Eastern form. The scribe’s flexibility in all but a few Anglo -Irish criteria suggests that he had left his native country some time before, retaining only traces of his original linguistic habits” (1988: xxxv). In respect with Cambridge University Library Dd V 64 MS, which is in one hand but variable in lan- guage, they did not analyse this specific text, but just the translation by Richard Misyn of Rolle’s Incendium Amoris and Emendatio Vitae. This manuscript seems to preserve Northern forms better than the others. Regarding the other two manuscripts Jeremy Smith from the University of Glasgow, in personal communication, let me know a different possibility: According to him, these variable forms may indicate the emergence of a colourless language. Even if colourless language is a concept that has not been clearly defined so far, it refers to a usage that, although it is not a stan- dard, is not particularly distinctive in regional terms, that is, forms which are not regionally specific. To conclude we can well claim, as Ogilvie-Thomson (1988: lxxiv) had suggested, Trinity College Dublin 155 and Longleat 29 could have been copied from a common ancestor, because both show the same deviant ex- pression and because they seem to share more features, linguistically speak- ing, than Cambridge University Library Dd V 64 does, such as: 1) the palatalization of some groups (church, whoche, such, much) versus the velarization found in Cambridge University Library Dd V 64 (kyrke, whilk , swilk , mykel); Studies in the language of some mss of Rolle’s E go Dormio 151 2) the rounding of Old English long a into o (holy, hold) versus the preservation of in the other manuscript; 3) the use of Old English pronouns reflexes for the third person plural in the possessive and object form (ham, hem and har, her), while this feature is not present in the Cambridge University Library Dd V 64; 4) present participles always in -ing, rather than -and: breninge, sitting; although Cambridge University Library Dd V 64 shows both, the number of - and forms is superior to that of -ing forms; 5) verbs in the third person singular in -th, rather than -s: like in contenes (with two exceptions in Trinity College Dublin MS: holdes and schewes). Apart from the similarities found between Trinity College Dublin and Longleat manuscripts, there are also some differences between the two, both textually and linguistically, like the reflexes of Old English /y(:)/ in words like lust, mirth, evil and others. However, they seem to have more features in common, which are not shared by the other manuscript, as I have tried to demonstrate above. Isabel de la Cruz Cabanillas Universidad de Alcalá REFERENCES Allen, H. E. 1927: Writings ascribed to Richard Rolle, Hermit of Hampole and materials for his biography. Heath and Co., New York & Oxford University Press, London. Allen, H.E. 1931: English Writings of Richard Rolle. Alan Sutton, Gloucester. Amassian, M. G. 1979: The Rolle Material in Bradfer-Lawrence Ms 10 and its Relationship to other Rolle Manuscripts. Manuscripta, 23, 67- 78. Amassian, M. G. & Lynch, D. 1981: The Ego Dormio of Richard Rolle in Gonville and Caius MS. 140/80. Medieval Studies, 43, 218- 249. Isabel de la Cruz 152 Benskin, M. & Laing, M. 1981: Translations and Mischsprachen in Middle English manuscripts, Benskin, M. & Samuels, M. L. eds. So meny people longages and tongues. Philological essays in Scots and medi- aeval English presented to Angus McIntosh . Middle English Dialect Project, Edinburgh, 55-106. Colledge, E. 1962: The Mediaeval Mystics of England. John Murray, London. Gilmour, J. 1956: Notes on the Vocabulary of Richard Rolle. Notes and Queries, 201, 94-95. Heseltine, G. C. 1930: Selected works of Richard Rolle, hermit. Longmans, London. Horstman, C. 1895: Yorkshire Writers: Richard Rolle of Hampole and his Followers. Swan Sonnenschein & Co., London. 2 vol. Laing, M. 1992: A Linguistic Atlas of Early Middle English: the value of texts surviving in more than one version. Rissanen, M. et al. eds. History of Englishes. New Methods and Interpretations in Historical Linguistics. Mouton de Gruyter, Berlin, 566-581. McIntosh, A. & Samuels, M. L. 1968: Prolegomena to a study of medieval Anglo -Irish. Medium Aevum, xxxvii, 1-11. McIntosh, A.; M. L. Samuels et al. 1986: Linguistic Atlas of Late Mediaeval English (LALME) Aberdeen University Press, Aberdeen. 4 vol. Ogilvie-Thomson, S. J. 1988: Richard Rolle: Prose and Verse edited from MS Longleat 29 and related manuscripts. [EETS.], Oxford University Press, Oxford. Smith, J. J. 1992: The Use of English: Language Contact, Dialect Variation, and Written Standardisation During the Middle English Period. Machan, T. W. & Scott, Ch. T. eds. English in Its Social Contexts. Essays in Historical Sociolinguistics. Oxford University Press, Oxford, 47-68. Watson, N. 1991: Richard Rolle and the Invention of Authority. Cambridge University Press, Cambridge. * † *