MS 17: SCRIBES AND SCRIPTS


1. Introduction
2. Scribe A
3. Scribe B
4. Scribe C
5. Scribe D
6. Scribe E
7. Scribe F
8. Gothic hands

1. Introduction

At least five scribes (A-E) worked as a team to produce MS 17; a slightly younger contemporary (F) made a few additions, and at least three scribes working after 1200 make substantial modifications. The original scribal team all wrote an English variety of late Caroline script or praegothica, but it is important to identify them as individuals. Detailed analysis of each scribe's style will confirm the date and place of MS 17's production established in the background essay "Location and Dating"; together with the codicological evidence presented in the background essay on "Materials and Structure", it will also establish that the book was planned as a whole and executed in a single operation.

In distinguishing individual scribes working at the same time and in the same place on a single manuscript, general impressions are insufficient, because the cooperative effort will favour homogenization. The quantitative formulae for dimensional analysis and the formal criteria for morphological analysis proposed by Leon Gilissen for distinguishing personal handwriting are more useful in this context.

Dimensional analysis reduces the proportions of script (the ratio of the height of the letter to its width, and the ratio of both to the unit of ruling) to a single modular ratio (rapport modulaire). Complementing this modular ratio is the quotient of weight, that is, a number expressing the ratio of pen width to height, angle, and incline of script. While Gilissen admits that the margin of error in establishing these statistics is wide, they can afford a fairly objective standard for verifying the identity of a scribe in two different passage or codices.1 The method is not without its critics,2 and indeed, the attempt to apply dimensional analysis to MS 17's scripts had to be abandoned, primarily due to the fact that there is not standard unit of ruling in this codex against which one can measure angle or proportion. Moreover, the extremely small scale of the gloss and annal hands make meaningful numbers difficult to obtain. Gilissen's second technique of morphological analysis is, however, both practical and sufficient. The principle of morphological analysis is that an individual scribe can be distinguished through two kinds of letterforms. The first comprises those complex signs, built up of a number of strokes, whose execution pushes the scribe's skill to its limit: examples are minuscule g, the ct and st ligatures, and the ampersand. The second class consists of forms where the chances that variation will confuse the reader are low, and therefore some rein can be given to personal style. The cedilla below an e in e caudata, for instance, can be rendered in a number of ways, because a reader will register almost any squiggle below an e as a cedilla. Besides such complex or variable forms, Gilissen signals minuscule a, r, and d as useful visual shibboleths.3

2. Scribe A

Scribe A, the directing scribe of the Thorney scriptorium for the MS 17 project, wrote, amongst other things, the very first pages of the manuscript, that is, the medical treatises on fols. 1v-2v; fol. 2r provides a good example of this hand. Scribe A's hand gives a global impression of height, compression and angularity. His writing has a perceptible forward slant, averaging about 5 degrees from perpendicular. The serifs at the top of his ascenders are sharply angled and occasionally split (cf. l and b in fleobotomare and variants), while the shoulders of letters like b, h, and r rise to a sharp peak (cf. col. A, line 22 brachio). His macrons are formed at a distinct acute angle. Majuscules like P, R, O, D, Q and M display a characteristic spikey s-curve on the second down-stroke (cf. col. A, line 20 De, col. A line 16 Medio).

Scribe A uses two forms of macron: a shallow cup with a pronounced upwards slant, and a horizontal line with short downward ticks at beginning and end, usually used to replace n (both forms are visible in indignationem, col. A line 29). His ampersand (cf. col. A, line 11) is finished in the lower half by a short elegant upsweep, and in the upper half by a transverse tick to the left. He also uses a Tironian et characterized by a wavy, upward-slanting top stroke, and a tick at the foot of the descender (cf. col. B line 4). A large, sharply-angled apostrophe, ending directly beneath its starting-point, is Scribe A's sign for -us (cf. col. B line 24), and the tail of his e caudata is a compressed wedge (cf. col. A line 25). His st ligature is distinctively tall, compressed and somewhat awkward (cf. col. A line 6 pestiferasque), and he never uses a ct ligature. He uses a variety of abbreviations for quod, including extending the hast upwards and crossing it (cf. col. B line 3) and has an usual manner of abbreviating id est by crossing the hast of the d in id (cf. col. A line 31)

The most striking and personal of Scribe A's characteristic letterforms is his minuscule g. Scribe A closes the upper bowl and forms the lower bowl of this letter in one continuous movement, proceding down from the upper bowl and swinging out to the right. The lower bowl is finished by a horizontal or downward-sloping bottom, which curves back in a hairline towards the initial down-stroke. In the typical English caroline fashion, the lower bowl is often not closed.4 His minuscule a shows an unusual variety of forms in the head stroke. When a is the initial letter, the head-stroke can be tall and prominent (cf. col. A line 2 anxietatem); otherwise, it can all but disappear, giving the appearance of a semi-uncial a (cf. col. A line 16 uitia), or it can conform to the usual caroline pattern (cf. col. A line 18 capitis).

These features of Scribe A's hand are discussed further in the essay on " Location and Dating", but for the moment they will serve primarily to distinguish his work from that of his major collaborator, Scribe B. An appreciation of the inner history of MS 17 depends to a large extent on recognizing the specific roles of these scribes. Scribe A directed the production of this manuscript. His leadership was primarily intellectual. Whereas Scribe B does the bulk of the writing, Scribe A writes the important and difficult materials such as glosses, labels, diagrams and tables, tasks which require understanding as well as skill and accuracy. Scribe A also does all the corrections. But he seems as well to have planned the structure of the codex. He writes the texts at the beginning and the end of the volume. Moreover, he is responsible for almost all the rubrics, and can be detected writing the opening lines of texts before handing them over to his colleagues to finish. This kind of leading-off is the office of the chief scribe.5 We may infer, then, that Scribe A was responsible for deciding what text or table went where, and there is ample evidence that he had a great concern for the exact order in which the apparently disparate items in MS 17 were to appear.

3. Scribe B

The contrast between the tall, compressed, angular hand of Scribe A and the shorter, rounder script of B can be seen on fol. 23r, where A writes the text within the roundels, the rubric ("Hanc sententiola..."), and the opening words of the text which follows ("Si quis igitur...") before handing over the task to B, who takes up the text at calculandi. B's smaller module gives his text a brighter, more open appearance than A's. His letterforms are rounder, he never inclines his ascenders more than a degree or two; the large feet on his minims, as well as his macron -- a straight line, abruptly hooked back at the end -- give a distinctively horizontal accent to his writing. Once again, minuscule g is the signature letter which enables us to identify the individual scribe. Scribe B makes his lower bowl like a flattened backwards c, the upper end of which projects noticeably from the lower bowl (line 2: ingenioli). Occasionally this lower bowl is finished with a short upward tick, but only rarely is it closed. The upper arm of B's ampersand ends with a short downwards stroke to the right, like the spout of a tea-pot (line 3, end). He does not use Tironian et. The cross-stroke of his minuscule x is sometimes shorter than the first strike, making the letter appear tilted; however, it usually descends very slightly below the line of ruling and curves up at the end, unlike Scribe A's straight cross-stroke (see numeral X in roundels). B ligatures both ct and st (line 3: distinctis). His e caudata is a loose trailing flash (line 11: que), or sometimes a loop (line 5: littere). A good way to distinguish him from Scribe A is his majuscule Q; A's always sits above the line (cf. Quatinus in rubric) while B's sits on the line, and has a distinctive tear-drop shape (Queris in line 16 of text). Other tell-tale differences between A and B include the majuscule S (line 2: Siquidem) which B begins with a little concave dip. He makes occasional use of the Insular autem (see fol. 90r, col. b line 13), but it is not easy to determine whether he was simply copying what he saw in his exemplar, or whether he had been trained to use Insular letter-forms and signs. The manner in which he renders the names of the Anglo-Saxon months in chapter 15 of De temporum ratione (fol. 76r-v) is also rather curious. In the first part of the chapter, his letter-forms are strictly caroline, e.g. using th for both thorn and eth. Gradually, he begins to restore thorn and eth, and even subsitutes Square Minuscule forms of r, g, and f. But this never assumes a consistent pattern.

Scribe B wrote a neat, regular and pleasing hand, but his accuracy and comprehension were not always up to his task. On the top line of fol. 23r, Scribe A, as was his wont, wrote the first line, ending in dies quam, before turning over the rest of the page to his colleague. Scribe B should have begun with queris. notataque littera quam, but omitted this phrase through homeoteleuton; Scribe A was later obliged to fill it in. Similar slips caused Scribe A to erase and re-do Scribe B's work on fols. 72rb and 81va. On fol. 63ra Scribe B had to re-write a whole paragraph, squeezing it into the available column space by ruling narrower lines. His mistake, this time, was homeoarchon: dropping about four lines of text between a sinistris aquilo (fol. 63ra, line 4 from bottom) and a sinistris eurus (fol. 63rb, line 1).

Scribe B's punctuation was not sophisticated, his repertory consisting of medial and terminal punctus and question mark. His puncti medii are sometimes transformed into puncti versi by Scribe A. Mathematics was not B's forte, and the superscript lines which indicate thousands are almost all supplied by A. His transcription of the Paschal tables abounds in inaccuracies, some forgiveable (e.g. writing VII instead of VIII), but some showing discouraging ignorance, e.g. dating Easter to XIX kl. mai. Scribe B. On ill. 2, line 4 reveals his difficulties in knowing where sentences begin and end, and Scribe A is frequently obliged to correct his word division. In short, Scribe B wrote a good hand, but needed constant supervision.

4. Scribe C

Scribe C wrote only a few columns of Gerlandus' treatise on the abacus, picking up from Scribe B about four lines from the top of fol. 51ra.

Comparing B and C at the point where C takes over helps to define the essential features of C's hand. B ends at Diffinitiones aliarum on lines 3-4. Despite a comparable ratio of height of letters to ruling space, Scribe C's writing looks smaller and tighter than Scribe B's. It inclines more to the right, producing a greater contrast of thicks and thins (compare the d in Scribe C's diuidenda in line 6 to Scribe B's duas in line 1). Scribe C accentuates the angularity of his writing by swinging his minims back to the left: a good example is furnished by the i and u in diuidi (line 7). He also splits his ascenders in a rather mannered way (diuidi line 7). The lower bowl of his g is small by comparison with that of Scribes A and B, and does not swing out to the right of the upper bowl at all (line 7: integros). His macron is very short and abruptly curled back; as well, his -er and -ur signs are characteristically tight. Scribe C's special eccentricity is his use of the ampersand to render et within words.

5. Scribe D

Scribe D wrote a large part of the anonymous treatise on calculation on fols. 52v-56r, as well as part of the chronicle at the end of Bede's De temporibus (fols. 61rb-61vb), part of the last column of Bede's De temporum ratione (fol. 65rb), the annal for 1085 in the Paschal tables concerning Gunter of Le Man's arrival at Thorney, and part of the legend in Gloss 59 to De temporum ratione. The following examples are from the top of the first column of fol. 52v.

Scribe D's hand is large. Not only it is tall (the proportions to the ruling space being the same as Scribe A's), but it is also perfectly upright and distinctly rounded. He not only does not split his ascenders, but scarcely even gives them serifs, preferring a clubbed d and l reminiscent of caroline script. His macron is a gentle curl, and feet of his majuscules have a characteristic wave (line 6: Articuli). His minuscule g closely resembles Scribe B's, but with no projection of the top of the lower bowl to the left; it is finished with a hairline stroke that almost closes the lower bowl. The cross-stroke of his x curls protectively under the preceding letter (line 4: flexione). His signatures are his Insular-type quod abbreviation (lines 3, 5, 9)6 and his affection for majuscule NS and NT ligatures (line 9), which he does not confine to the ends of words or lines.

6. Scribe E

Scribe E wrote most of the text of Bede's De temporibus. In addition, he wrote lines 3-11 of the first page of the Argumenta titulorum paschalium on fol. 156ra, and added the final annal of ch. 66 of De temporum ratione at the foot of fol. 120r, which Scribe B had omitted. His script is demonstrated by the first half of De temporibus ch. 10, in the lower half of fol. 59v.

In comparison with that of his fellow Thorney scribes, E's module is unusually small: in a 7 mm unit of ruling, his minims rise 2.5 mm., and his ascenders never more than 3.5 mm., or barely half the available space. In combination with his short ascenders and descenders, this makes his script look low, flat and rounded, and bestows a spacious quality to his page. This is emphasized by the perfect uprightness of his letters and by a shallow angle of writing, producing almost horizontal wedge-shaped tops to his minims and ascenders.

Though the overall aspect of his style is quite striking, E's individual letter-forms show few departures from the types used by A and B. His minuscule g (line 3: signum, egressus) resembles Scribe A's, with its single stroke closing the upper bowl and forming the lower. Occasionally he will make a g in three strokes, like Scribe B. His minuscule x is reminiscent of Scribe A's, but the cross-stroke dips in a shallow curve below the ruling (line 3: ex). Notice the thin projecting "tongue" of his minuscule e (line 4: parte), the gentle wave of his macron, and his characteristic closed superscript a (line 4: quarta).

7. Scribe F

Scribe F, who added the elements schema on fol. 13r, possibly wrote the annals for A.D. 1118-1139 in the Paschal table.

8. The Gothic hands

Of the later Thorney hands who made additions to MS 17, we will only discuss three who made substantial modifications to the manuscript. These are (1) the scribe who wrote gathering 17b; (2) the scribe who marked the signatures; and (3) the scribe who supplies the chapter numbers and titles for Helperic's De computo. We will designate them Gothic I, II and III.

As noted in the essay on "Materials and Structure", Gothic I adjusted the Paschal tables after 1200, and wrote completely new ones after 1421. Paschal tables do not afford a wide variety of letter-forms for study, but a rough dating to the thirteenth century is possible. Indeed, Gothic I also wrote the annals for 1279-1293 on Cotton fol. 83v. Compare, for example, the majuscule A of Anno in the Dominical letter column of the Paschal table with Alexander in the 1289 annal, or the xviii in the 1288 annal with the numerals in the Paschal table. Other features of Gothic I's hand are the small caroline a with its upper bowl open, the straight, acutely angled "tongue" of the minuscule e, and flash-stroke macron ( annal for 1279: denarius).

The signature marks were probably added at the time Gothic I's extra Paschal tables were inserted. Thus the work of Gothic II may be tentatively dated to the final decades of the 13th century. Under the ink signatures, red crayon drafts are sometimes visible. This same red crayon appears frequently in the margins of Bede's De temporum ratione and Helperic's De computo, sketching out chapter numbers and titles. What he notes in the margins is rather interesting: sources cited by Bede (e.g. the books of the Bible mentioned in De temporum ratione ch. 72) and Helperic (the Vergil citation on fol. 129r), names of significant figures in the great chronicle in De temporum ratione ch. 66 (e.g. Aristotle on fol. 109v), and the zodiac signs mentioned by Helperic in ch. 2 of his treatise (fols. 124r-125r). It would seem that a rather ambitious rubrication program was projected.

The only part which was carried out, apart from the signatures, was Gothic III's addition of chapter titles and numbers to Helperic. Gothic III also completed the statements on the calendar pages regarding the entry of the Sun into zodiac signs, left incomplete after May by the original 12th century scribes. His script is more fractured and angular than that of Gothic I, with a fair amount of biting of curves. His 8-shaped minuscule a, crossed x, snub-footed minims and very short lower bowl of his minuscule g, make his script look rather more recent. Indeed, Gothic III may be responsible for noting the year 1408 at the top of the key-letter table on fol. 30v. The forms of the numerals, compared with the chapter numbers in Helperic, reveal a similar slightly tilted x, a v with a steep, upward-curving first stroke, and a trailing terminal i. Plainly, in the thirteenth century and perhaps later, the monks of Thorney were still deeply interested in the contents of their computus manuscript, as well as in maintaining its annals.


1 Gilissen 1973, 31-32.

2 See particularly the remarks of Spunar 1976, 62-68.

3 Gilissen 1973, 46-48.

4 On the open g as a hallmark of English Caroline, see Vézin 1968, 286.

5 cf. Vézin 1973, 226; Gasparri 1976, 237.

6 Vézin 1977, 110.