The Hengwrt Canterbury Tales: Inadmissible Evidence?

61 The Hengwrt Canterbury Tales: Inadmissible Evidence? Michael Pidd, Estelle Stubbs and Claire E. Thomson One of the characteristics of Manly and Rickert’s work on The Canterbury Tales is their holistic approach to the manuscript tradition: they consider not only the text of The Canterbury Tales in 84 pre-1500 witnesses, but also the philology, codicology, palaeography, artwork and provenance of every one of these witnesses (Manly and Rickert 1940.) Whether or not an individual scholar possesses the knowledge and experience necessary to do justice to all these facets, Manly and Rickert do recognize that a manuscript tradition such as that of The Canterbury Tales is symbiotic, and every one of these manuscripts possesses particular physical characteristics and originates from a particular historical context. In order to understand the manuscript tradition of The Canterbury Tales we have to explore every element which has a bearing on the creation of the witnesses. We shall not arrive at any reliable answers through simply looking at the text alone. It is noticeable that witness descriptions written by scholars since Manly and Rickert repeat this holistic approach, considering all the manuscript versions on all these different levels in order to acquire as complete a picture as possible. However, our preliminary investigation into the presentday condition of the Hengwrt manuscript of The Canterbury Tales (National Library of Wales, MS. Peniarth 392 D) suggests that a further influencing factor to be taken into consideration is the manuscript’s history as recorded and preserved in facsimile. This is particularly relevant to scholarship of the late twentieth century, since photographic technology has advanced considerably and, as we shall now see, the history of the Hengwrt manuscript in facsimile is not necessarily one of progress towards a clearer, more informative image. Before we can take any holistic approach to the manuscript we first of all need to answer the question: how reliable is the manuscript’s testimony? One example of why the testimony of Hengwrt is open to doubt is the inscription Stokes which occurs in the footer of folio 85v. Manly and Rickert describe it as a name in ‘red crayon’ (1940, I: 282), whilst Doyle and Parkes speak of ‘a large red plummet inscription’ (1979, xlvii.) Mosser makes no mention of the inscription in his manuscript descriptions for The Wife of Bath’s Prologue on CD-ROM, although he emphasizes that these are a truncated and provisional version of his own wider project (see Mosser 1996.) However, whereas Manly and Rickert categorically state that the Stokes inscription is one of ‘only two legible clues’ to ‘the earlier history of Hengwrt’ (1940, I: 282), Doyle and Parkes are cautious, and merely note that the inscription had been ‘read by Manly and Rickert’ (1979, xlv); they acknowledge Manly and Rickert’s testimony, but are reluctant to confirm the reading themselves. Consequently, we decided to look for the inscription ourselves. It certainly is not present in the 1979 Variorum Chaucer facsimile, but this cannot be considered a reliable witness, since it is a facsimile in greyscale and is predominantly concerned with THE CANTERBURY TALES PROJECT OCCASIONAL PAPERS II, 1997, 61-68 62 Canterbury Tales Project Occasional Papers II reproducing the text of The Canterbury Tales. Many hours were spent in the National Library of Wales poring over every scratch and blemish on the vellum of folio 85v. Figure 1 is a recent photographic image of the page and the inscription does not appear to be present.1 The Library joined in the search and allowed folio 85v to be submitted to different light sources: ultra violet, infra red, fibre-optic light, a high-powered microscope and, of course, daylight. In spite of this, nothing except a few extremely faint and ambiguous scratches to the left of the two parallel lines could be found. It seems that Doyle and Parkes had good reason to be cautious of Manly and Rickert’s reading and this raises an interesting question: was the manuscript which Doyle and Parkes examined in 1979 in the same condition as when Manly and Rickert examined it in 1928, or did something happen in the interim which resulted in the disappearance of the inscription? Our investigations have suggested that the latter is correct. In 1930 Manly and Rickert negotiated with the National Library of Wales to have a photostatic copy made of Hengwrt, as they had done with all the pre1500 manuscript witnesses of The Canterbury Tales, and this copy was eventually made available on microfilm by the University of Chicago.2 Unlike the presentday manuscript, the Stokes inscription is clearly visible on the Manly and Rickert microfilm, despite the fact that it is photographic technology of the 1930’s, and the microfilm image is old and gradually deteriorating in quality. Furthermore, the British Library holds volumes of the original photostats which were commissioned by Manly and Rickert, used by McCormick and finally deposited with the Library in 1936 (British Library MS Facs 405[39].) Figure 2 is an image of folio 85v as it appears on Manly and Rickert’s photostatic copy.3 The quality of the image is close to that of a modern greyscale facsimile, and the reading which Manly and Rickert affirmed to be Stokes is unquestionable. Thus it would seem that, between Manly and Rickert’s testimony in 1928 and Doyle and Parkes’ cautious note in 1979, the Stokes inscription had disappeared. Since Hengwrt was rebound in 1956, it is possible that the manuscript also underwent conservation at this time. Unfortunately the Library does not have any records of manuscript conservation dating from this period. Therefore it is only possible to say that, around 1956, something may have been done to Hengwrt. Our investigations have shown that there are other inconsistencies between the Manly and Rickert photostats and the manuscript as it appears today. For example, inscriptions in ink, red plummet and dry point are apparent on the photostats but have never been described. However, the effects of conservation also appear to have uncovered text in the margins of the manuscript which, albeit illegible, is not visible on the photostats. All of this has repercussions throughout Canterbury Tales scholarship. Failure to examine the information content of older facsimiles and earlier witness descriptions in the light of the original can seriously limit the amount of potentially vital information available. This is perhaps symptomatic of our attitude to both scholarship and technology. There exists a complacency that scholarship is progressive, and that the methods developed to effect this scholarship are progressive. For example, in this present age we believe that our manuscripts are in better health than they were a hundred years ago, and 63 that high-resolution digital photography delivers images of manuscripts which are clearer, more faithful and more information-dense than black and white microfilms from the 1930s. By default, we believe that we probably know more about The Canterbury Tales than Manly and Rickert, because we can now see more information than they could, whether it be through new discoveries or new technology enabling those discoveries. Unfortunately, this is not necessarily true, as we have seen. The Hengwrt manuscript which Manly and Rickert examined in the late 1920s was different from the one which we can see today; a photostatic copy of the manuscript made in the early 1930s contains more information about the codicology of the manuscript than any attempt at a high resolution digital facsimile would produce. As a result, Manly and Rickert’s research into the provenance of this manuscript is more significant than has perhaps otherwise been assumed by their successors, because they had access to more information. The Stokes inscription is a good example, since Doyle and Parkes thought it could possibly be the name of a user or maker of the manuscript (1979, xlv) but they did not pursue this line of enquiry any further because there was no evidence on the manuscript itself that the inscription existed. As such, Manly and Rickert’s own research into the possible significance of this inscription has been overlooked by scholars ever since. Now that we have evidence that Manly and Rickert’s observation was correct, that the Stokes inscription really did exist, we are able to reconsider the results of their research into its possible significance and we have since focused on trying to establish the identity of Stokes and his relationship to Hengwrt. Manly and Rickert had identified a number of 15th century men named Stoke or Stokes who either had libraries or the money to purchase books, but concluded that there was no connection between these men and the later owners of Hengwrt (1940, I: 282.) We decided to search for men named Stokes who were alive in the late 14th century, and may have known Chaucer personally because if Ellesmere (Huntington, MS. 26.C.9) was produced around 1397-1405, as Scott has argued on the evidence of its illumination (1995, 87119), and if Ellesmere was written by the scribe of Hengwrt in his declining years (Doyle and Parkes 1979, xxxv), then it would follow that the production and early ownership of Hengwrt must have been contemporary with Chaucer, who may even have supervised its production, as Blake has suggested.4 During the course of our investigations, it became increasingly clear that there were several men of the name Stokes who could have had a personal acquaintance with Chaucer and we are at present trying to establish profiles for each candidate. Bearing in mind that Manly and Rickert considered the probable first owners of the Ellesmere manuscript to be the Earls of Oxford (1940, I: 155), our search for a suitable Stokes included not only looking for men who may have had connections with the London life of the late 14th century, but also for men who may have had some involvement with the Earls of Oxford. Men of the name de Stokes were employed as Keepers of the Wardrobe and in varying capacities in the Exchequer, the Law Courts, and the Royal Household for more than a hundred years before and after Chaucer’s death,5 and several men named de Stokes also worked as clerks for the Earls of Oxford during the period 1335-1435.6 The life of one man, a Richard de The Hengwrt Canterbury Tales 64 Canterbury Tales Project Occasional Papers II Stokes, could provide a link between Chaucer’s London and the possible regional associations of Ellesmere. He can be found in the company of some of Chaucer’s personal friends, such as Sir Simon Burley, Philip La Vache, Lewis Clifford and possibly the Stonor family,7 and also appears to have been acquainted with a number of important families from the area around Castle Hedingham, family seat of the de Veres, Earls of Oxford. Castle Hedingham is located on the borders of Suffolk and Essex, close to Cambridgeshire in an area which seems to have been a focus for Canterbury Tales manuscript activity in the 15th century. Richard de Stokes was born probably around 1330,8 and may have been a clerk in the service of John de Vere, 7th Earl of Oxford.9 From 1354 he was a clerk in the service of the Black Prince10 and in 1377 he became King’s clerk and a Baron of the Exchequer for Richard II (CPR 1909, 424.) His final payment in that capacity was at Easter 1399 (CPR 1909, 424.) No records for him appear to exist after 1401.11 In a short essay like this it is not possible to give all the associations of Richard de Stokes with specific people and places known to Chaucer and fully to explain their significance. Suffice it to say that during Richard de Stokes’ long life of service to royalty and the nobility, connections can be seen with people and prominent families with whom Chaucer was also acquainted. The noble households of the Ulsters, the Black Prince, John of Gaunt and the Earls of Oxford, were closely connected through much of this period both at war and at home, and it is conceivable that Chaucer and Stokes were known to each other from an early date. It may be coincidental that a Richard de Stokes also owned a London property in the Parish of St. Peter’s upon Cornhill, close by Chaucer’s residence in Aldgate, and next to Lime Street Ward, home of the Gregory family. The Gregory family were the probable first owners of the Hatton Hours and Psalter (Bodleian Library, MS. Hatton 4),12 which could be significant since Scott argues that Ellesmere and the Hatton Psalter shared illuminators (1995, 95-101.) The hand of the Hengwrt scribe is described by Doyle and Parkes as expert, possibly that of an ecclesiastic, administrator or lawyer (1979, xxxv, xxi.) Richard de Stokes can be associated with just such a literate community, within which the copying of Hengwrt could have taken place. If Richard de Stokes were an early owner of Hengwrt, then its date of production must be pre-1400. An earlier date for Hengwrt would have repercussions for other early witnesses, which may also have been produced during Chaucer’s lifetime. It is not possible, at present, to argue conclusively that the Richard de Stokes described above is the person whose signature appears on Hengwrt. However, our investigation into the name Stokes has had several important repercussions. It has led us to a group of manuscript-owning families from a very small area on the Suffolk/Essex/Cambridgeshire border. These families have Government and Court connections as well as links with religious houses in the region. This could be of significance when we consider that several manuscripts, identified by Robinson’s analysis as from the ‘O’ group (Robinson, in this volume), may have originated here. From the name Stokes another name has emerged, that of Fermer. The two names appear together and separately on documents relating to the Earls of Oxford for a period of more than 140 years,13 and are also to be 65 found separately on at least four Canterbury Tales manuscripts: Hengwrt, Additional 35286, New College, and Rawlinson Poetry 141 (Manly and Rickert 1940, I: 282, 47, 384 and 453.) In the 13th and 14th centuries, families with these names tenanted manors in Stoke Clare and Foxearth,14 two small villages close to both Castle Hedingham, and the Augustinian Clare Priory, possible place of copying of at least one Canterbury Tales manuscript. According to Manly and Rickert, ‘The Austin Canons were especially makers of books....’15 At present our evidence is only circumstantial, but the point of this exercise remains valid: because Hengwrt has suffered a loss of information since the 1950s, a great deal of Manly and Rickert’s research has been overlooked because the manuscript evidence which they took as their starting point is no longer visible. Instead of investigating the probability that this evidence had once existed, scholars have chosen to ignore it and, consequently, Manly and Rickert’s resulting research. It is a natural tendency to privilege the image of the real manuscript over facsimile representations. However, just as there is a need to recognize that The Canterbury Tales is a text recorded in many manuscript versions, likewise there is a need to recognise that each individual manuscript of The Canterbury Tales is now recorded in many representations. To privilege one particular image, whether it be the manuscript itself or the newest facsimile, is to increase the possibility of missing valuable information which may have been captured in other, older representations. Manly and Rickert had photostatic copies made of the 84 extant witnesses they investigated, but this paper has only explored one of them. Nevertheless we hope to have shown that a continuation of Manly and Rickert’s work into the biographical and regional connections between some manuscripts of The Canterbury Tales might be extremely productive. How many of the other 83 photostatic copies hold information which is now no longer visible upon the manuscripts themselves, and how much of Manly and Rickert’s research into the biographical and regional connections associated with these manuscripts would attain new relevance and clarity by our discovery that the evidence for their research really did and still does exist? The time has come for scholarship to re-examine not only the entire manuscript tradition of The Canterbury Tales as witnessed in primary material, but also the research conducted by our predecessors and, crucially, the technology and methodology which they harnessed to effect it, whether it be now dated photostatic images or subsidiary primary sources such as government and provincial records. All of this raises another issue: what status do the manuscript and the photostats have? It is generally believed that the authority of a manuscript is absolute, but if photostats preserve an image of a manuscript which is more ‘complete’ than the manuscript itself, one might be tempted to argue that the photostats are a more authoritative witness. This is ironic, for Manly and Rickert tell us in their Prolegomena that the photostats were deposited with the British Library primarily as an insurance against deterioration of the manuscripts themselves (1940 I: 6.) Having discovered inconsistencies between the Hengwrt manuscript and its photostatic copy, we have been working in association with the National Library of Wales and the British Library toward the establishment of a Hengwrt Project in Sheffield which seeks The Hengwrt Canterbury Tales 66 Canterbury Tales Project Occasional Papers II to fulfil three goals. The primary goal is to preserve: we shall compile a digital archive of every page of Hengwrt from both the photostats and the manuscript itself, accompanied by a forensic, folio-by-folio examination of the disparities between the two. The second goal is restorative: having established the archive, our intention is to record text in the margins of Hengwrt which, though currently illegible, may be recoverable from digitally enhanced images. The third goal is to re-examine the research which Manly and Rickert conducted as a result of manuscript evidence which is no longer available and, where possible, to build upon it with research of our own. Finally, in approximately two years time, subject to funding, we hope to make the entire Hengwrt archive available in electronic form for the wider scholarly community, complete with our analyses and findings. It is important that other scholars are made aware of the inconsistencies between the manuscript and the photostats, because if the Manly and Rickert photostats had never existed, potentially valuable information would now be lost, thereby reducing our ability to ascertain who made this manuscript, how, why, and for whom. Notes 1Reproduced with the permission of the National Library of Wales, 1996. 2This is on Reel 9 of the Manly Microfilm Collection: Canterbury Tales Manuscripts. University of Chicago Libraries. 3Reproduced with the permission of the British Library Board, 1996. 4N. F. Blake discussed this in ‘Different Versions of the Links in The Canterbury Tales,’ an unpublished paper delivered at the 8th York Manuscripts Conference, 1996. 5Men of the surname Stokes were employed in various roles. For Keepers of the Wardrobe, see: Crow and Olson 1966, 103 n.2. For the Exchequer, see: CPR 1916, 376. For the Barons of the Exchequer, see: CPR 1895b, 417 and 506; CPR 1902, 29; and CPR 1909, 424 and 518. 6Men of the surname Stokes were employed in the household of the Earls of Oxford. For example, see: CCR 1898a, 452; CCR 1898b, 201; CCR 1906, 201; CPR 1900, 254; and Woolgar 1993, 536. 7The name Richard de Stokes is found in several records associated with personal friends of Chaucer. For example, see: CCR 1920, 463; CPR 1909, 245; and BPR 1931, 398-9. 8Assuming that Stokes began service with Edward III in 1354 after a short period in the Oxford household, and given that he received a grand pay-off in 1399 after nearly 50 years royal service the late 1320s would probably be a realistic date for his birth. Further evidence to support this can be found in CIPM 1974, 19. 9For records referring to Richard de Stokes as clerk to John de Vere, 7th Earl of Oxford, see: CPR 1895a, 233; and CPR 1900, 254. Richard de Stokes is also found prosecuting a debt on behalf of John de Vere, 7th Earl of Oxford in CCR 1906, 201. 67 10 For Richard de Stokes’ service to the Black Prince, see: BPR 1931, 490; and BPR 1933, 477 and 490. A clerk named John de Stokes worked for the Black Prince in 1347, and Richard de Stokes could have been his son (for example, see: BPR 1930, 105.) 11 The last surviving record relating to Richard de Stokes is dated as 1401: CCR 1927, 474. 12 See Chew and Kellaway 1973, 130-1, no. 525; and Scott 1995, 102-4. 13 For example, see: Anderson 1993, 104; CPR 1900, 254-5; CPR 1905, 235; and CPR 1909, 64. 14 Harper-Bill and Mortimer 1983, 406, no.635; FFE ; and MRF. 15 Manly and Rickert 1940, vol. I, 342; although Doyle has suggested that it was Austin Friars rather than Canons who were producers of books. Bibliography Anderson, V. 1993. The De Veres of Castle Hedingham. Lavenham, Terence Dalton Ltd. BPR. 1930. Register of Edward the Black Prince Preserved in the Public Record Office: Part I, 1346-1348. London. —. 1931. Register of Edward the Black Prince Preserved in the Public Record Office: Part II, 13511365, (Duchy of Cornwall.) London. —. 1933. Register of Edward the Black Prince Preserved in the Public Record Office: Part IV, 1351-1365, (England.) London. CCR. 1898a. Calendar of Close Rolls Preserved in the Public Record Office: Edward III, 13301333. London. —. 1898b. Calendar of Close Rolls Preserved in the Public Record Office: Edward III, 13331337. London. —. 1906. Calendar of Close Rolls Preserved in the Public Record Office: Edward III, 1349-1354. Vol. ix. London. —. 1920. Calendar of Close Rolls Preserved in the Public Record Office: Richard II, 1381-1385. Vol. ii. London. —. 1927. Calendar of Close Rolls Preserved in the Public Record Office: Henry IV, 1399-1402. Vol. i. London. Chew, H. M. and W. Kellaway, Eds. 1973. London Assize of Nuisance 1301-1343; A Calendar. London Record Society Publications. Vol. 10. Leicester, London Record Society. CIPM. 1974. Calendar of Inquisitions Post Mortem and other Analogous Documents Preserved in the Public Records Office: 7-15 Richard II, 1384-1392. Vol. xvi. London. CPR. 1895a. Calendar of Patent Rolls Preserved in the Public Record Office: Edward III, 13341338. Part I. London. —. 1895b. Calendar of Patent Rolls Preserved in the Public Record Office: Richard II, 13771381. London. —. 1900. Calendar of Patent Rolls Preserved in the Public Record Office: Edward III, 1340-1343. London. —. 1902. Calendar of Patent Rolls Preserved in the Public Record Office: Richard II, 1388-1392. Vol. iv. London. The Hengwrt Canterbury Tales 68 Canterbury Tales Project Occasional Papers II —. 1905. Calendar of Patent Rolls Preserved in the Public Record Office: Richard II, 1391-1396. Vol. v. London. —. 1909. Calendar of Patent Rolls Preserved in the Public Record Office: Richard II, 1396-1399. Vol. vi. London. —. 1916. Calendar of Patent Rolls Preserved in the Public Record Office: Edward III, 1374-1377. Vol. xvi. London. Crow, M. M. and C. Olson, Eds. 1966. Chaucer Life Records. Oxford, Clarendon Press. Doyle, A. I. and M. B. Parkes. 1979. ‘Palaeographical Introduction.’ In The Canterbury Tales. A Facsimile and Transcription of the Hengwrt Manuscript with Variants from the Ellesmere Manuscript, ed. P. G. Ruggiers. Norman, Ok., University of Oklahoma Press: xix-xlix. FFE. Feet of Fines for Essex. Vol. 3. Colchester: Essex Archaeological Society. Furnivall, F. J., Ed. 1868-77. A Six-Text Print of Chaucer’s Canterbury Tales in Parallel Columns. Chaucer Society, 1st Ser. 8 vols. London, Trübner. Harper-Bill, C. and R. Mortimer, Eds. 1983. Stoke by Clare Cartulary: BL. Cotton Appx. xxi, Part II. Suffolk Records Society, Suffolk Charters. Vol. 5. Woodbridge, Boydell and Brewer. Manly, J. M. and E. Rickert, Eds. 1940. The Text of the Canterbury Tales: Studied on the Basis of All Known Manuscripts. 8 vols. Chicago, Chicago University Press. Mosser, D. W. 1996. ‘Witness Descriptions.’ In The Wife of Bath’s Prologue on CDROM, ed. P. M. W. Robinson. Cambridge, Cambridge University Press. MRF. Manorial Rolls for Foxearth, Chelmsford Record Office. Scott, K. 1995. ‘An Hours and Psalter by Two Ellesmere Illuminators.’ In The Ellesmere Chaucer: Essays in Interpretation, ed. M. Stevens and D. Woodward. California, Huntington Library: 87-119. Woolgar, C. M., Ed. 1993. Household Accounts from Medieval England. Part II: Diet Accounts (ii); Cash; Corn and Stock Accounts; Wardrobe Accounts Catalogue. Records of Social and Economic History, New Series xviii. Oxford, Oxford University Press.