John Northwood’s Miscellany Revisited
John Northwood’s Miscellany Revisited
Abstract and Keywords
London, British Library, MS Additional 37787, a volume of prayers and other devotions and related material, was part-edited by Nita S. Baugh as A Worcestershire Miscellany Compiled by John Northwood c. 1400 (1956). Baugh’s title was based on ownership inscriptions of John Northwood, monk at Bordesley Abbey, Worcestershire, and members of the Throckmorton family also of Worcestershire. These associations have made the manuscript an important witness in narratives about Cistercian participation in the production and circulation of Middle English verse manuscripts in the West Midlands and the role of monasteries in fostering vernacular writing and book production, including the Vernon and Simeon manuscripts. This chapter proposes that this view is called into question by careful codicological examination of the volume. Through challenging these propositions it suggests alternative ways to explore and explain the production of books containing vernacular prayers and devotions in late medieval England.
THIS CHAPTER EXPLORES LONDON, British Library, Additional MS 37787, part-edited by Nita S. Baugh as A Worcestershire Miscellany Compiled by John Northwood c. 1400 (Philadelphia, privately published, 1956). Additional MS 37787 is predominantly composed of prayers and other devotions and related material. The volume contains several marks of ownership. The earliest owner recorded is John Northwood, who entered the Cistercian Abbey of Bordesley (Worcestershire) as a novice in 1386, according to an inscription in the volume (fol. 182v). The volume also includes an ownership inscription that states that it belongs to Northwood (fol. 183r). Other inscriptions show that the volume was in lay hands, though still in the West Midlands, in the later 15th century. Lady Goditha Throckmorton Peyto (d. 1502)1 is recorded as one of the owners of the manuscript on fol. 2r. A record of the gift of the volume to a Goody Throckmorton also occurs on this page.2 Sixteenth-century owners were the wife of John Rudhall (possibly of Worcester) and Susanna Willescotta who bought the volume in 1571 and who is otherwise unknown.3 Presented by Baugh as a miscellany compiled by a Cistercian monk in the West Midlands c. 1400,4 Additional MS 37787 has played a small but crucial part in histories of Middle English devotional writing. The evident connection with Bordesley and later with the Throckmorton family has given rise to narratives about Cistercian participation in the production and circulation of Middle English verse manuscripts in the West Midlands and the role of monasteries in fostering vernacular writing and books, including the (p.102) Vernon and Simeon manuscripts.5 In this chapter I propose that the analyses of Additional MS 37787 on which these narratives depend are called into question by careful codicological examination of the volume. Additional MS 37787 is arguably a prime example of where application of the term miscellany associated with a focus on ownership rather than manuscript materiality may have misled us into accepting certain propositions about the production (p.103) and provenance of a manuscript.6 Through challenging these propositions I shall offer a new account of the manuscript and explore the implications for the wider narratives concerning West Midlands book production and the production of books containing vernacular prayers and devotions in late medieval England.
1 Codicology and contents
Baugh was unable to carry out a full palaeographical or codicological examination of Additional MS 37787.7 Doyle has noted that the manuscript is in ‘one or more expert text-hands’.8 In her unpublished doctoral thesis, Rebecca Farnham analysed in detail the codicology, decoration, and scribal work. Farnham proposed that the volume was the work of four scribes and four decorators and that it was produced in ‘five separate sections’.9 While I do not agree with some of the detail, I concur broadly with Farnham that consideration of ruling, hand and script, and decoration provides evidence that the manuscript is composed of several sections. I propose to call these sections units.
1.1 Unit 1: quires 1–2, fols 3r–17v
Unit 1 comprises two quires of eight leaves ruled for 19 lines. The first four leaves of quire 2 have bifolium signatures I, II, III, IIII. Quires 1 and 2 are written continuously though there is no catchword. There are red rubrics and red initials but no decoration. Red brackets denote quatrains. The script is Textura Semi-Quadrata in a hand that appears nowhere else in the manuscript. See Figure 6.1. Letter y is distinguished from the thorn with a diamond shape dot and a straight hairline descender that makes a 45 degree angle with the line. The first and middle stroke of w are as high as the other ascenders. The final page has 20 lines of writing, line 20 being written below the last (p.104) ruled line, suggesting that no further quire was anticipated and an effort has been made to cram the material (Prayer to the Trinity, fols 16r–17v)10 into two quires. This suggestion is also supported by the omission of the final two lines of the last eight-line stanza and possibly by the tactical omission of stanzas to fit the space (the Vernon text has five additional stanzas).11
The material in Unit 1 is in Middle English and the unit has the appearance of having been produced as a discrete little vernacular devotional resource. There are six items, one in prose, the rest in verse. The verse items are all prayers and many texts are associated with guidance on their use and efficacy. The Hours of the Cross (fols 12v–14r) is preceded by a verse indulgence followed by verses and a prayer to be repeated after the text for each hour. The Prayer at the Elevation (fol. 12r–v) is preceded by a Latin rubric stating that it should be recited at the levation. The prose Form of Confession is preceded by a vernacular verse rubric that states that it ‘teches man to saluaciun’ (fol. 3r). A Latin colophon explains that ‘Hec confessio prescripta compilatur non ut quilibet eam totam dicat sed ut ea in quibus se reum esse cognoscit confiteatur’ (fol. 11v), clarifying that the text is for instruction rather than recitation in full.
1.2 Unit 2: quires 3–6, 25–7, fols 18r–48v and fols 161r–183v
Codicological, palaeographical, and art-historical examination reveal these quires to have been, like unit 1, the outcome of a discrete production campaign—though this one was of two scribal stages or stints.12 Quires 3–5 (fols 18r–36v) are of eight leaves (five leaves are missing at the beginning of quire 3); quire 6 (fols 37r–48v) is of 12 leaves; quires 25–7 are of eight leaves and a singleton (fols 161r–182v and fol. 183, the Northwood ownership inscription). Quires 3–5, 25–7 and the singleton are ruled for 20 lines and quire 6 for 23 lines. The first page of quire 5 (fol. 29r) carries the bifolium signature dI. The quires are linked with catchwords (fols 20v, 168v, 175v). Rubrication is in red and the text is divided by blue initials decorated with red penwork (e.g. fols 173v, 183r). Lines are filled with red and blue decorative fillers in both stints and on fol. 183r. I discern two main stages or stints because quire 6 is treated as a final quire of a stint. It has 12 leaves rather than the usual eight, is ruled for 23 lines rather than 20, has an additional line on fol. 48v to finish the text, and below this an inscription by the scribe requesting that he not be blamed for his copying.13 The first page of quire 25 has a text space of (p.105)
These quires are distinguished as a production unit principally by their decoration and scribal hand. Two pages carry decoration. In quire 5 the beginning of the Debate between the Body and Soul is marked by a painted initial A and short bar-frame border (fol. 34r). See Figure 6.2. The serifs and (p.106)
One scribal hand copied the quires of this unit with the exception, possibly, of the novitiate inscription on fol. 182v. This stint is signed by the scribe on fol. 182r: ‘Finito libro sit laus et gloria christo. Corpus scriptoris seruat deus omnibus horis. Hic pennam fixi penitet me si male scripsi.’ The main unit 2 scribe, like the unit 1 scribe, used Textura Semi-Quadrata. Features that distinguish this scribe from that of unit 1 are y with a dot that has a curved tail and a descender that has a flick to the right at the bottom; w of x-height; and ascenders with forked tops. The tail of g curls up to form a two-compartment letter. Farnham attributes the novitiate inscription to this scribe but I am not so sure.14 It is difficult to tell as the inscription is in Textura Quadrata with calligraphic qualities rather than Semi-Quadrata and the language Latin, therefore lacking many examples of the diagnostic y and w. However, the y in ‘Bordesley’ is formed in a very similar way to the thorn in ‘Norþewode’ and lacks the rightward flick on the descender and the w is above x-height, the middle stroke being higher than the others. It has not previously been noticed that the novitiate inscription scribe inserted the text after the page was decorated (see below) whereas the main scribe worked before decoration was applied. Whether or not the scribe of the novitiate inscription also wrote the rest of the text in unit 2, it is clear that fol. 182v was written at a later stage of the production process and out of the usual sequence of production.
MWM lists 27 items in this unit (many of them composite). Most are charms, prayers, and indulgences. Five items are in Middle English, though one is only a rubric: the Stations of Rome (fol. 18r, very fragmentary, end only); the Debate between the Body and Soul (fols 34r–45r); a prose Vision of St John on the Sorrows of the Virgin which calls for devotions (fols 161r–162v); a fragmentary verse Complaint of the Virgin which calls for devotions (fol. 170r); and a rubric following the Latin Names of God (fols 174r–178r) recommending (p.108)
1.3 Unit 3: quires 7–24, fols 49r–160v
The portion of the manuscript that I identify as unit 3 runs from fol. 49r to fol.160v. Quires 7–14 and 16–24 comprise eight leaves; quire 15 is of 12 leaves. Pages are ruled for 20 lines. Bifolium signatures are frequent and in a different form from that in unit 2.15 Catchwords are found on fols 62v, 65v, 70v, 83v, 92v, 99v, 107v, 115v, 128v, and 136v. The text is divided with alternating red and blue initials with penwork flourishing. Line spaces are filled with red and blue line fillers though they are of a design different from those in unit 2. The rhymes of quatrains are bracketed (fol. 144r–v).
The hands and decoration of this unit are distinctive. The text is divided by gold initials on rose and blue grounds. The beginnings of some texts are signalled by sumptuous border work. For example, Anima Christi sanctifica me (fol. 75r–v) begins with a very large initial A that is incorporated in a full-page bar-frame border. See Figure 6.4. The palette is rose, blue and orange. Sections of the vine that runs parallel with the gold of the bar-frame terminate in pairs of highlighted leaf forms that are topped with gold balls and pen-work squiggles tinted with green. Throughout are alternating red and blue initials decorated with penwork in a contrasting colour, either red or black.16
Up to fol. 142r the scribe of unit 3 uses a large Textura Semi-Quadrata which is very regular, an effect gained from very short ascenders and descenders and the treatment of tails: the letter g is single-compartment (contrast the scribe of unit 2); its tail and the ascender of d run almost parallel with the line on some pages. Two-thirds of the way down fol. 142r the script changes to Textura Quadrata, continuing to fol. 160v. This is the first point at which this unit includes vernacular material (apart from a rubric on fol. 81v) and where prose gives way to verse. Farnham detects a new hand at this point.17 It is (p.110)
MWM lists 73 items in these quires, some of them composite. The unit formerly included the Arma Christi (fol. 62v with lost illustration; only a rubric survives) along with other prayers, indulgences, and meditations.18 There is overlap with the other units—Stations of Rome occurs here and in unit 2 (fol. 60v; cf. fol. 18r); the Hours of the Cross here and in unit 1 (fols 96v–99r (Latin); fols 12v–14r (Middle English)). Several items also occur in later medieval English books of hours. The prayers to Mary Obsecro te (fol. 111r) and O intemerata (fol. 86v) were particularly popular after 1400.19 The Latin Prayer to St Apollonia (fol. 131r–v) is paralleled in the Bolton Hours and Latin anthems for St Mary Magdalene and St Anne are also found in books of hours.20 The lost Sayings of St Bernard (fol. 81v) is also found in BL, Additional MS 33381, fol. 161. Apart from the rubric for this text, nine Middle English items are included, all of them at the end of the stint, in quires 22–4 (22 also includes Latin material). They are all in verse and mostly prayers: Prayer for Three Boons; Prayer to the Trinity; Song of Love-longing to Jesus; Prayer to the Virgin; Seven Deeds of Spiritual Mercy; Confession for Negligence of the Deeds of Mercy; Prayer for Saving the Five Wits; Confession of Negligence of the Commandments; and Prayer at the Elevation.
2 The Bordesley provenance revisited
Baugh carried out an examination of the language of the texts, concluding that it is mixed. However, she asserts that ‘In spite of these dialectally inconsistent (p.112) features, in all of these pieces (and in every text in the MS.) there are present also forms peculiar to the W. Midl. dialect’, and concludes, ‘We have in this MS. a miscellany of texts copied from originals of various sorts and, in the copying, brought more or less into conformity with the W. Midl. dialect of the district in which Bordesley was situated.’ Baugh’s dialectal analysis therefore appeared to confirm her hypothesis concerning the Bordesley/Northwood provenance of the volume. An Electronic Version of A Linguistic Atlas of Late Mediaeval English (E-LALME) locates Additional MS 37787 in Worcestershire, noting that the novitiate inscription associates the volume with Bordesley but not specifying which texts were analysed.21 Although Farnham identified that the manuscript was composed of separate production units, and before her Doyle identified that it might be in more than one hand, Baugh’s claims about the dialect of the vernacular items have not been reviewed on a unit-by-unit basis. Having distinguished the hands of several scribes at work and discrete production campaigns the question arises as to whether Baugh’s claims and E-LALME’s localisation hold good for each of the different units and to what extent they may differ in their display of West Midlands forms. My diagnostic items are those from the list of forms recorded for this Linguistic Profile in E-LALME that map exclusively or nearly exclusively to the West Midlands.
2.1.1 Unit 1
A Form of Confession has very few West Midlands forms; heor(e three times only (111, 115, 306) as against þer (107); mon once (81) as against many instances of man and womman (e.g. 74); vuel (238); cussyng (‘kissing’, 212, 276).22 Sin never has a rounded vowel. In the verse, Prayer at the Elevation has wonne (‘wan’), interestingly rhymed with manne (2, 4) perhaps suggesting the West Midlands, and mon and ron (lines 5, 7); mon and won also occur in the Hours of the Cross (lines 10, 14, 25, 26). A Prayer to the Trinity has stud-fast (3) and vuul (11, ‘evil’). However, this unit does not, I suggest, bear out Baugh’s assertion that the texts have been ‘brought more or less into conformity with the W. Midl. dialect’. It is very obviously a mixed text. Notable non-West Midlands forms in the Form of Confession include senned (‘sinned’, 60); wershep (105); and the very rare yew (‘you’, 294; cf. Confession to Jesus Christ, 72).
The Stations of Rome fragment includes monkynde (5). The 406-line Debate between the Body and Soul shows no forms diagnostic of the West Midlands. I find no diagnostic items in the remaining three vernacular texts in this unit. This unit does not, I suggest, bear out Baugh’s assertion that the texts have been ‘brought more or less into conformity with the W. Midl. dialect’.
2.1.3 Unit 3
Of the three units, unit 3 has the highest density of West Midlands forms. The rubric for the lost Sayings of St Bernard (fol. 81v) has mon and deouele. A Prayer for the Three Boons (fols 142r–143v) has whon (79) and mon (87), neither in rhyme position, þeose (88), and studefast (73). Prayer to the Trinity (fols 143v–146r), like its other rendition in unit 1, has studfast (3) and vuel (‘evil’, 11). The text with most West Midlands colouring is Song of Love-longing (fols 146v–156r). Here ‘sin’ regularly has a rounded vowel (e.g. sunnus, 24, 30, 116, 232); mon is found in and out of rhyme position (185, 321, 338); other diagnostic forms are luytel (80); uchi (‘each-a’, 94); al-þau (141, 333, 345); kussyng (159); muchel (237); and vnkuynde (323). Prayer to the Virgin (fols 156v–157v) has sunne out of rhyme position (7) but synne to rhyme with þer-inne (19–20). Deeds of Spiritual Mercy (fols 157v–158r) has pruide (14, not particularly localised to the West Midlands however) but synne (8, 17, 24). Deeds of Mercy (fols 158r–159r) has vchon (14) and mon (17), neither in rhyme position. Prayer at the Elevation (fol. 160r–v) has mon (11, 26). Judging from the rhymes mon, gon, on, Ion (185–9) however, this probably was originally a West Midlands text so we should not be surprised if it exhibits a high density of characteristic forms. Nonetheless the scribe of this unit clearly tolerates and perhaps introduces West Midlands spellings. He also seems to have access to material circulating—perhaps rather narrowly—in the West Midlands. Baugh stated that 14 of the 20 pieces in English in Additional MS 37787 are also to be found in the Vernon manuscript and that Vernon and Additional MS 37787 ‘share the same textual tradition’.23 Baugh’s sense of the significance of the numbers is weakened when the composite production structure of Additional MS 37787 is taken into account (not to mention that Vernon, with its hundreds of texts, shares material with many manuscripts).24 (p.114) Nonetheless that four items in this unit occur elsewhere only in Vernon may indicate access to rare local texts.25
In conclusion, the three units exhibit different characteristics with respect to dialect. The scribe of unit 3 tolerates West Midlands forms and may indeed introduce them. The scribe of unit 2 does not use West Midlands forms, with one stray exception. In the middle of the spectrum is the scribe of unit 1 who is comfortable with a range of practice. This examination of dialect suggests that the three units derive from different scribal practices. It also suggests that the scribe of unit 2 may not be of the West Midlands.
The affiliations of the manuscript decoration have not been much studied. Farnham inferred that the volume was decorated at Bordesley, though she also noticed similarities between the decoration of unit 3 and a manuscript thought to have been decorated in London.26 The presence of two quite different styles of decoration in the manuscript in itself suggests that the work was done not only by different hands but also by hands from different workshops, and the fact that the two styles are associated with different scribal stints with differing dialect profiles also means that we need not assume that decoration was carried out in one place, let alone in a single West Midlands location. A further search for comparators suggests that we should be looking at one or more of the main centres of book decoration for the origins of the work in Additional MS 37787.
The two decorated pages in unit 2 have on the face of it the best claim to be Bordesley decoration because one of them is the novitiate page. But, as mentioned above, this page was decorated before the novitiate inscription was entered; the scribe has scratched away part of the top border to allow space for the decorative N in ‘Anno’, and e in ‘deuotissime’ is one of several parts of the inscription to cover the border.27 This suggests that the identity of the owner and destination of the volume may not have been known when the volume was decorated. This weakens the hypothesis of a Bordesley provenance for the decoration. Some of the motifs characteristic of the unit 2 border artist are identified by Scott as new in the second decade of the 15th century: spiky acanthus leaves and green tint on pen squiggles. These characteristics (p.115) may be suggestive of provenance too. Scott claims that the acanthus leaf was ‘a favourite theme in at least one London shop of c. 1410–1425’.28 The decoration would therefore seem to precede the writing of the inscription but postdate, stylistically, its content (the year 1386 is given for Northwood’s entry into the novitiate of the Cistercians). Possibly the inscription was copied into the space from another source, into a blank frame that had been decorated elsewhere.
The decoration of unit 3 recalls that of prayer books and books of hours made in the late 14th and early 15th centuries. The ratio of initial size to the frame and text space is paralleled for example by Cambridge, CUL, MS Ee.1.14, fol. 44r, a London manuscript made c. 1405 (the historiated initials in this manuscript, not paralleled in Additional MS 37787, are attributed to the German artist Herman Scheere).29 The Middle English primer CUL, MS Dd. 11.82 has 20 lines to a page and full or partial borders in a similar palette.30 London, BL, Harley MS 2445 is a smaller volume (130×85mm compared with 1953140mm for Additional MS 37787) but, like Additional MS 37787, it contains prayers, includes some vernacular texts, and has indulgences in red,31 and initials and borders in a similar palette (including orange) though with different motifs; it includes some English material.32 Also comparable in size, ruling, and palette, though considerably more sumptuous, is London, Lambeth Palace Library, MS 474, a prayer book made for a priest that later passed into royal hands.33 Cambridge, Emmanuel College, MS 246 has a similar palette and large initials filled with leafy vines that emerge from their serifs.34 We noted above that unit 3 is the unit most strongly associated with the West Midlands on grounds of dialect. Possibly it was copied by, and for, those (p.116) comfortable with West Midlands orthography. The numerous analogues for the decoration, however, suggest a metropolitan provenance for the artists. Further comparative work is required to test this hypothesis.
3 Binding and organisation of Additional MS 37787
Baugh saw Additional MS 37787 as a miscellany driven by the tastes and predilections of its compiler. She noted that some of the items shared with Vernon appear in a different order and in two sections in Additional MS 37787, using this as evidence that John Northwood made his own selection of material from a common source: Northwood ‘chose only those that interested him’ (p. 38). Farnham rejected Baugh’s label ‘miscellany’ for the volume, but argued for even more purpose—both in the production design and in text selection and sequencing. She proposed that production in sections was deliberate ‘to allow a greater freedom of compilation in the final gathering of the quires’, producing a ‘compilation with its own dynamic’ that included two ‘vernacular reading programmes’ which were the ‘result of deliberate choice’ related to ‘Cistercian spirituality’.35 As we have seen, on the basis of codicology, dialect, and decoration, the materials bound together in Additional MS 37787 might not even all have been produced at Bordesley or even in the West Midlands. Close examination of the binding opens up the possibility that the components of the volume may not all have been at Bordesley, for they may have been bound together much later in the 15th century.
The British Library catalogue entry for Additional MS 37787 describes the binding as ‘original binding of oak boards covered with red vellum’, ‘English’, and dates it as 15th century.36 Farnham too sees the binding as ‘original’ and dates it c. 1400, contemporary with the date she provides, following Baugh, for the contents of the codex.37 The characteristics of the binding themselves do not determine a date. After c. 1450 a cover of red-dyed tawed skin gradually became less common than a cover of tanned, often stamped, leather. But tawed skin continued to be used so this does not determine the date.38 However, (p.117) close examination of the binding suggests that it could postdate the writing on the vellum endleaves (fols 184r–187v) of English prayers in a late, unpractised hand.39 Mary Erler identifies the hand as that of Lady Goditha Throckmorton Peyto.40 Farnham argues that this is the hand of Goody Throckmorton.41 Whatever the identity of the person who wrote the prayers on the endleaves, however, it is clear that they were written much later than the rest of the manuscript.
The endleaves appear to have been bound into the codex at the time the binding was applied. This by itself would not indicate the date of binding, of course, because the endleaves could have been inscribed after the book was bound. However, the final page of the endleaves (fol. 187v) and the back board provide evidence that the writing of the late material preceded the binding. The wood of the inner face of the back board with its channelled sewing supports and the overlaps of its red vellum cover are clearly visible today. The inner face would have been originally covered by a pastedown and fol. 187v could have been this pastedown: it bears about eight lines of mainly illegible unpractised writing in the hand of the endleaves scribe and carries the impression of the sewing supports from the back board. There is further evidence that this leaf was the pastedown. The inner face of the board and the red skin overlaps carry the impression of writing. This writing is illegible but its angle gives the impression of mirror writing indicating that it constitutes offsets of script (e.g. the partially illegible word ‘..yt..’ (fol. 187v), appears to be reversed on the red vellum).42 Arguably this is the effect of the application of glue to stick fol. 187v down on the back board. We should consider the possibility therefore that the components of the codex may not have been brought together until the later 15th or early 16th century and that the assembly of the materials could be associated with the volume’s Throckmorton Peyto owners.
Close examination of the way the components have been arranged suggests that perhaps several prayer books were assembled to make a composite volume. We have seen that throughout the codex leaves are missing. It is not clear if material has been lost after binding or before, or both. The person who (p.118) prepared the units for binding could have been patching up incomplete books. There are signs of cobbling together of discontinuous material even within the production units I have identified. For example, quire 12 (fols 71r–78v) in unit 3 is highly suspect: the final four leaves (fols 75–78) appear to be singletons sewn on to stubs. Folio 74r is blank and the text on fol. 75r is fragmentary, lacking its beginning. The text on fol. 79r starts with another acephelous text and does not follow that on fol. 78v. There is evidence that some at least of this interference is associated with the removal of devotional images. The text and images of the Arma Christi are missing after fol. 62v as is an image of Christ after an indulgence on fol. 64v. We can only speculate on the reasons for the losses. Given that the codex is replete with indulgences, it is perhaps unlikely that the images were removed because of disapproval or censorship. More likely, perhaps, they were removed for devotional purposes. Although some pages show signs of rubbing, those at the ends of the units are not especially discoloured with the exception of fol. 3r, the first page of unit 1. This reinforces the impression that the binder/compiler was patching together portions from other books rather than compiling a volume from booklets. There is no sign of vacant sewing holes but earlier holes could have been reused. It is difficult to explain the presence of badly rubbed pages but it is notable that many occur where leaves have been removed (e.g. fols 18r, 51r, 57r, 63r, 66r, 115v). Perhaps the leaves were removed because they were too badly damaged to be worth retaining, in which case the abrasion might be the result of tracing words of favoured devotions with the fingers.43 Alternatively, perhaps the damage occurred after the removal of the adjacent leaves, and when the materials were unbound, leaving the rubbed leaves vulnerable to damage. Other rubbed pages occur at quire ends, e.g. fols 100r, 107v, 175v, again perhaps because damage was suffered when the quires were unbound.
Aside from the fact that all of the components (including the endleaves) include many prayers and indulgenced prayers, is there any explanation for the way these rather scrappy fragments were organised? We can only speculate. The placing of quires 1–2 first suggests some privileging of the vernacular by the binder or the person who prepared the material for binding. Possibly, too, this material was held to resemble a ‘primer’ since it covers basic catechetical material. Such material often came first in manuscripts (cf. London, BL, Arundel 2, fol. 3). When we examine the order in which the discrete production units have been arranged in the codex the most striking question is why unit 2 has been interrupted by unit 3. We cannot of course assume that we have all of unit 2; some material may have been lost, discarded, or bound (p.119) elsewhere. Possibly quires 3–5 were placed immediately after quires 1–2 with a view to bringing together the vernacular booklet (unit 1) with the Stations of Rome. However, the first five leaves of quire 3 are missing. Stations has up to 667 lines in other manuscripts, with a prologue in addition in two, in some cases most of the text being copied as prose (DIMEV, 1909), so it could easily have filled the missing 10 pages if not 10 pages and a missing preceding quire (in Lambeth Palace Library, MS 306 the text runs to 27 pages). Quires 25–7 may have been placed at the end of the codex rather than being left in association with the other material from this unit because their final leaf includes a colophon that announces that the scribe has finished copying the book (‘finito libro sit laus et gloria christo’, fol. 182r) and bears the novitiate inscription (fol. 182v). The Northwood ownership inscription is on a singleton (fol. 183r) that would have been logically bound next to the novitiate inscription since both refer to Northwood.
We are now in a position to pull together the outcomes of the unit-by-unit analysis and consider their implications for the view that Additional MS 37787 was copied and decorated at Bordesley, and was compiled or even copied by the Bordesley monk John Northwood and associated with West Midlands vernacular book production. We have seen that Additional MS 37787 appears to be a composite manuscript and the outcome of several production campaigns rather than the planned volume proposed by Baugh and accepted by others, the miscellany title and the weight given to the Northwood ownership evidence perhaps encouraging uncritical acceptance of this view. Unit 1 is a vernacular ‘primer’ with little claim to having been produced in the West Midlands and need not have been in the West Midlands until the later 15th century. The other two units are more extensive codices of English and Latin devotional material each with its own decorative programme. The evidence for Northwood’s ownership of the volume and its being at Bordesley is strongest for unit 2, for the novitiate and Northwood ownership inscriptions belong with this unit on decoration and codicological grounds. The vernacular material in it, however, cannot safely be used on grounds of dialect as evidence for the collection and copying of vernacular devotional materials at Bordesley. Nor can we rule it out, of course. But Northwood is unlikely to have been this unit’s scribe. He does not name himself in the two scribal colophons as one might expect given that he is named in the inscriptions. Nor can unit 2 be used as evidence for book decoration at Bordesley. The novitiate inscription was added after the page was decorated. The decoration may have been applied before a particular owner had been identified. At any rate it need (p.120) not have been applied at Bordesley and the London analogues suggest an alternative provenance for this decoration campaign.44 The vernacular material in unit 3 has most claim to being from regional sources; this unit includes rare West Midlands verse and is most coloured by West Midlands dialect. The decoration, however, was not necessarily local. Later readers—possibly Throckmorton women—provided another unit of material that became fly-leaves when the components were bound. The binding up of the units was a final act of compilation of what is perhaps best described as a process of collection of materials that stretched from Northwood to the Throckmortons and may have had intervening sources, stages, and participants we cannot now recover. In view of the composite nature of the volume we can, therefore, no longer imagine Northwood or another single compiler selecting from a source common to Additional MS 37787 and Vernon, and if this is accepted then a linchpin has been removed from the narrative of the Cistercian role in West Midlands vernacular book production.
Recent work on books of hours and looser compilations of prayers has added a great deal to our understanding, but the model of the miscellany produced for a particular patron, exemplified by Baugh’s interpretation of Additional MS 37787, still holds sway. In the view of Charity Scott-Stokes, for example, ‘A medieval book of hours was in essence a miscellany of prayers made for an individual, a family or a community.’45 Several recent studies have drawn attention to the interface between the destination of a prayer book—the individual, family, or community it was made for—and its production.46 We still know relatively little, however, about how the contents of prayer books were sourced and transmitted47 and it may be that the miscellany–patron model leads us to neglect material evidence—codicological, artistic, and textual—for the complex of processes involved. By deconstructing Additional MS 37787 using such methods I have tried to bring a sharper focus on materiality to the study of manuscript provenance and compilation. Further identification and analysis of composite codices such as Additional MS 37787 may contribute to a material history of the late medieval prayer book that has yet to be written.
(1) For Goditha’s date of death, see Jan Broadway, ‘Peyto family (per. 1487–1658)’, Oxford Dictionary of National Biography (Oxford, Oxford University Press, online edn, January 2008), http://www.oxforddnb.com/view/article/72342?docPos=2.
(2) Mary Erler, Women, Reading and Piety in Late Medieval England (Cambridge, Cambridge University Press, 2002), p. 114. The gift inscription distinguishes ‘Goodyth Peyto’ and ‘Goody Throkmarton’. The ownership inscription relating to Lady Peyto is in a different hand from that relating to the gift to Goody.
(3) Rebecca Farnham, ‘The Producers and Readers of London, British Library, Additional MS 37787’, Ph.D. thesis (Birmingham, 2002), pp. 221–2.
(4) A reference to the death of Robert Tresilian (d. 1388, fol. 175v) has been taken as a terminus a quo for the volume.
(5) Baugh (ed.), A Worcestershire Miscellany established the main lines of interpretation in the preface to her partial edition. Although she kept an open mind as to whether John Northwood was the scribe and compiler, she inferred that ‘the MS was clearly written at Bordesley abbey for John Northwood’s use’ (p. 37). She states ‘There is nothing in the MS or its subsequent history to contradict the belief that it was written in the Cistercian abbey of St Mary’s Bordesley’ (p. 15). Don C. Skemer, Binding Words: Textual Amulets in the Middle Ages (University Park, Pennsylvania State University, 2006), p. 19 assumes that John Northwood is the scribe of the volume as does Simon Horobin, ‘The Scribes of the Vernon Manuscript’, in Wendy Scase (ed.), The Making of the Vernon Manuscript: The Production and Contexts of Oxford, Bodleian Library, MS Eng. poet. a. 1 (Turnhout, Brepols, 2013), pp. 27–47, at p. 27, n. 1. Erler, Women, Reading and Piety, p. 113 accepts Baugh’s assertion that the manuscript was written at Bordesley Abbey. Eamon Duffy, The Stripping of the Altars: Traditional Religion in England 1400–1580, 2nd edn (New Haven, CT, Yale University Press, 2005) accepts that the manuscript was compiled ‘by the Worcestershire cleric John Northwood’ (p. 278) giving the manuscript as an example of clerical compilation. Chris Cannon describes the manuscript as a miscellany and accepts Bordesley as the place of compilation, using this as an example of ‘the degree to which the textual culture of British monasteries recognised only the most fragile boundaries between available languages and textual kinds’, going on to make an argument about vernacular traditions in monasteries: ‘Monastic Productions’, in David Wallace (ed.), The Cambridge History of Medieval English Literature (Cambridge, Cambridge University Press, 1999), pp. 316–48, at p. 326. Baugh notes that Additional MS 37787 is of ‘special interest’ (p. 39) because 14 of the Middle English items are shared with the Vernon manuscript (Oxford, BodL, MS Eng. poet. a. 1). She tentatively suggests that this points to a Cistercian origin for the Vernon manuscript. Kari Sajavaara, ‘The Relationship of the Vernon and Simeon Manuscripts’, Neuphilologische Mitteilungen, 68 (1967), 428–39 accepts Baugh’s analysis and interpretation of the textual affiliations and dialect of Additional MS 37787, using the Bordesley connection to speculate that the Simeon manuscript (London, BL, Additional MS 22283) and Vernon were copied in one of the Worcestershire or Warwickshire Cistercian houses. A. I. Doyle, ‘The Shaping of the Vernon and Simeon Manuscripts’, in Derek Pearsall (ed.) Studies in the Vernon Manuscript (Cambridge, D. S. Brewer, 1990), pp. 1–13 accepts the claim that the manuscript was a ‘collection … by John Northwood, a Cistercian of Bordesley’ and that sharing 14 of 20 texts with Vernon suggests that Additional MS 37787 was ‘multiply-related’ to Vernon (p. 8); cf. ‘Introduction’, in The Vernon Manuscript: A Facsimile of Bodleian Library, Oxford MS Eng. poet. a. 1, with an introduction by A. I. Doyle (Cambridge, D. S. Brewer, 1987), pp. 1–16, at pp. 14–15. Ryan Perry, ‘The Clopton Manuscript and the Beauchamp Affinity’, in Wendy Scase (ed.), Essays in Manuscript Geography: Vernacular Manuscripts of the English West Midlands from the Conquest to the Sixteenth Century (Turnhout, Brepols, 2007), pp. 131–59 accepts that Northwood was the compiler and Bordesley the place of production, stating that it ‘firmly link[s]’ Vernon to Bordesley (p. 143). Jeremy J. Smith, ‘Mapping the Language of the Vernon Manuscript’, in Scase (ed.) The Making of the Vernon Manuscript, pp. 49–67 argues on the basis of the shared texts and language that Vernon’s uniqueness ‘can be somewhat overstated’ (p. 52). In the earlier 14th century Bordesley was bequeathed books by Guy Beauchamp, earl of Warwick: M. Blaess, ‘L’abbaye de Bordesley et les livres de Guy de Beauchamp’, Romania, 78 (1957), 511–18.
(6) In a discussion of analogues for the Vernon manuscript, ‘Some Vernon Analogues and their Patrons’, in Scase (ed.), The Making of the Vernon Manuscript, pp. 247–68, at p. 261, I drew attention briefly to some problems with the current account of Additional MS 37787. In the present chapter I develop these remarks.
(7) She worked from photostats of selected pages and with the palaeographical guidance of Robin Flower, at that time Deputy Keeper of Manuscripts at the British Museum: Baugh (ed.), A Worcestershire Miscellany, pp. 35–7.
(8) Doyle, ‘Introduction’, p. 14.
(9) Farnham, ‘Producers and Readers’, p. 43. Her findings are summarised in the entry in Manuscripts of the West Midlands: A Catalogue of Vernacular Manuscript Books of the English West Midlands, c. 1300–c. 1475 (Birmingham, University of Birmingham, 2006), http://www.hrionline.ac.uk/mwm (hereafter MWM).
(10) All titles in this chapter are based on those in the list of contents in MWM, where IMEV numbers (originally Carleton Brown and Rossell Hope Robbins (eds), Index of Middle English Verse (New York, Index Society, 1943)) and other bibliographical numbers are also given.
(11) Oxford, BodL, MS Eng. poet. a. 1, fol. 114r–v.
(13) ‘Non me set copiam reprehendat lingua loquentis.’
(15) Quires 17–22 are signed with assorted symbols for the quire and with roman numerals for the bifolia, for example quire 18, fol. 108r has eI; quire 21, fol. 129r has +I and fol. 130r has +II; quire 22, fol. 137r has VI in red ink. From quire 23 an alphabetical system starts from letter b: bI (fol. 145r), bII (fol. 146r), etc.; cI (fol. 153r), cII (fol. 154r).
(18) The surviving rubric is in Latin but this does not necessarily indicate the language of the text. For examples of manuscripts that include Latin verse on the Arma Christi, see R. H. Robbins, ‘The Arma Christi Rolls’, MLN, 54 (1939), 415–21, at 415, n. 2.
(19) Kathryn A. Smith, Art, Identity and Devotion in Fourteenth-century England: Three Women and Their Books of Hours (London, British Library, 2003), p. 194.
(20) Charity Scott-Stokes, Women’s Books of Hours in Medieval England (Cambridge, D. S. Brewer, 2006), pp. 114–15, 119, 124.
(21) A. McIntosh, M. L. Samuels, and M. Benskin with M. Laing and K. Williamson (eds), An Electronic Version of A Linguistic Atlas of Late Mediaeval English, rev. edn (Edinburgh, University of Edinburgh, 2003), http://www.lel.ed.ac.uk/ihd/elalme/elalme.html, Linguistic Profile 7640. Baugh (ed.), A Worcestershire Miscellany, p. 86 considers the main diagnostic features to be OE a + nasal spelt o and OE ȳ̆ spelt u.
(23) Baugh (ed.), A Worcestershire Miscellany, pp. 37, 39. Mary, Mother, Well Thee Be (fols 156v–157v, IMEV, 2119) is listed by Baugh as having a parallel in Vernon but this is an error. The Vernon text is IMEV, 2110.
(24) Baugh’s argument about textual affiliations is also questionable but I have no space to pursue this point here.
(25) Seven Deeds of Spiritual Mercy (fols 157v–158r); Deeds of Mercy (fols 158r–159r); Five Wits (fol. 159r); and Ten Commandments (fols 159r–160r); cf. BodL, MS Eng. poet. a. 1, fols 115v–116r.
(27) See Figure 6.3. The alternative possibility, that the decorator has carefully painted around the inscription, does not align with the work of decorators. Furthermore, close examination reveals ink on top of the paint.
(28) Kathleen L. Scott, Dated and Datable English Manuscript Borders, c. 1395–1499 (London, British Library and the Bibliographical Society, 2002), p. 42; her example is London, BL, Arundel MS 38, a manuscript of Hoccleve’s Regiment of Princes that she dates (on internal grounds) 1410–13.
(29) Paul Binski and Patrick Zutshi, Western Illuminated Manuscripts: A Catalogue of the Collection in Cambridge (Cambridge, Cambridge University Press, 2011), p. 181; cf. London, BL, Stowe MS 16, also attributed to Herman Scheere.
(30) Binski and Zutshi, Western Illuminated Manuscripts, p. 186, where it is dated first quarter of the 15th century.
(32) Dated as the last quarter of the 14th or the first quarter of the 15th century in British Library, Digital Catalogue of Illuminated Manuscripts, http://www.bl.uk/catalogues/illuminatedmanuscripts/record.asp?MSID=3687&CollID=8&NStart=2445.
(33) Eamon Duffy, Marking the Hours: English People and Their Prayers (New Haven, CT, Yale University Press, 2006), p. 100 describes this volume as made c. 1415, ‘probably in London’.
(34) Illustrated in Eleanor G. McCullough, ‘Praying the Passion: Laypeople’s Participation in Medieval Liturgy and Devotion’, Ph.D. thesis (York, 2011), pp. 96, 101–2. McCullough dates the volume later 14th century.
(36) As Baugh did not have an opportunity to examine the codex for herself she was able to make no comment on its binding.
(37) See the entry in MWM and cf. Baugh (ed.), A Worcestershire Miscellany, pp. 13–17.
(38) Alexandra Gillespie, ‘Bookbinding’, in Daniel Wakelin and Alexandra Gillespie (eds), The Production of Books in England 1350–1500 (Cambridge, Cambridge University Press, 2011), p. 170. The descriptions of bindings in R. M. Thompson (ed.) with Michael Gullick, A Descriptive Catalogue of Manuscripts in Worcester Cathedral Library (Cambridge, D. S. Brewer, 2001) show that tawed skin was used in the rebinding of the earlier 16th century and examples are dated all through the 15th century.
(39) A modern paper endleaf occurs after fol. 187.
(42) Unfortunately using a mirror does not make the writing on the end board any more legible.
(43) Soiling can also reveal patterns of devotional use: see Kathryn M. Rudy, ‘Dirty Books: Quantifying Patterns of Use in Medieval Manuscripts Using a Densitometer’, Journal of Historians of Netherlandish Art, 2 (2010), 1–26.
(46) For example, Michael Camille, Mirror in Parchment: The Luttrell Psalter and the Making of Medieval England (London, Reaktion Books, 1998); Duffy, Marking the Hours; Smith, Art, Identity and Devotion; Virginia Reinburg, French Books of Hours: Making an Archive of Prayer, c. 1400–1600 (Cambridge, Cambridge University Press, 2012), pp. 53–83.
(47) Cf. Nigel Morgan, ‘Books for the Liturgy and Private Prayer’, in Nigel J. Morgan and R. M. Thompson (eds), The Cambridge History of the Book in Britain, Volume II, 1100–1400 (Cambridge, Cambridge University Press, 2008), pp. 291–316, at p. 316: ‘although ownership of individuals and institutions is known for many Psalters and Books of Hours, evidence as to who made them and where they were made is for the most part lacking’.