Barrow's Scribes and MoA's hands: Scribal Identification in G.W.S. Barrow's Regesta Regum Scottorum II

Feature Article 8: Alice Taylor (Co-Investigator), in consultation with Teresa Webber (Co-Investigator)

The first major attempt to identify scribes of Scottish charters (of any kind) was in G.W.S. Barrow's Regesta Regum Scottorum II: The Acts of William I, 1165–1214 (Edinburgh University Press, 1971, with the collaboration of W.W. Scott). Although Barrow had previously edited the charters of Mael Coluim IV (published in 1960), he had felt that the twenty-seven surviving originals of Mael Coluim's charters were not enough to draw any firm conclusions.[1] The charters of William the Lion, however, were another matter. Barrow found 159 surviving original charters, a number which included 'a handful of spurious and purported originals'. He went on to identify twenty-six or twenty-seven hands within the corpus.[2] Within these, Barrow identified five professional scribes: Richard of Lincoln, Gervase, Hugh de sigillo, William de Boscho, and Gilbert of Stirling (links to the entries for these people in the PoMS database are given below).[3] Although Barrow was aware that, 'strictly speaking', his list of twenty-six was a list of hands, not scribes, he nonetheless called them 'scribes' in the critical apparatus of the edited texts in RRS, ii.[4] What follows explains Barrow's referencing, lists the scribes he identified as well as the charters they are supposed to have written (given according to RRS-ii number and PoMS H-number). These documents can be found in MoA's database by searching with the PoMS H-number.

Barrow's scribal identifications have not been used in the database for Models of Authority.  In part, this is because our conceptual classification differs from Barrow's own. In Models of Authority, we have catalogued only 'hands', not scribes, and our understanding of a 'hand' is rather more limited than Barrow's. The difference between a 'scribe' and a 'hand' is important. A scribe refers to a person, someone who wrote and copied documents. A hand, however, is understood to mean a coherent style of writing which has internally consistent and identifiable features and ductus which allow it to be identifiable as a single unit. To put it more simply, it is conceivable that a single scribe (i.e. the person) could write in many hands.  The number of hands, therefore, in our database should in no way be interpreted as mapping neatly on to the number of scribes identifiable as working in twelfth- and thirteenth-century charter production. The number of scribes will not number more than the number of identifiable hands; it is more than possible, however, that the number of scribes was substantially less than the number of hands. 

The category of 'hand' in the Models of Authority database is conceptualised as document-specific. That is, the a priori aim of the palaeographical database in MoA is not to identify the distinctive features of a hand in order to identify scribes but, instead, common features of style and ductus which, by isolating allographs, have the potential to be studied as part of broader palaeographical developments. In short, a 'scribal' approach would encourage the identification of distinctive features of ductus and style of individual hands, whereas MoA's is to identify and track commonality and change over time. Of course, the database will also provide data for scholars and students to be able to identify and categorise distinctive features, so questions about scribes and scribal identification will inevitably emerge from MoA's data once it has been collected. MoA's palaeographical methodology will be set out more fully in future Features of the Month.

The point of providing Barrow's scribal identifications in this FoM is two-fold. First, it will allow for MoA's identifications to be tested against Barrow's and vice-versa. These have wider implications. Barrow posited that the tendency of royal acta from the last two decades of William's reign (c.1195–1214) to be written in 'cursive 'chancery' or 'chapel' hands', 'suggests that the scribe's office was made more specialized and more of a full-time occupation as the reign progressed'.[5] How far cursivity in ductus and/or in style (for the distinction, see Tessa Webber's November 2015 FoM) was a feature associated with royal acta, as opposed to a palaeographical development more broadly is, of course, a key question which MoA hopes to answer. Second, providing Barrow's 'list of scribes' will promote communication between different media—between database digital editions and more traditional print editions—to allow for scholars and students familiar with Barrow's identifications to navigate new media from an existing starting point. In this way, it will be possible to demonstrate both the debt to and developments away from earlier scholarship in a clear and straightforward way.

Understanding the scribal reference system in RRS, ii.

Barrow grouped his scribes according to the chancellorship in which they first appeared, and assigned them a majuscule letter (A–G).

A = Nicholas of Roxburgh, 1165–71

B = Walter de Bidun, 1171–8

C = Roger de Beaumont, c.1188–9

D = Hugh of Roxburgh, 1189–99

E = William Malveisin, 1199–1202

F = Florence of Holland, 1203(?)–10(?)

G = William de Boscho, 1211–14

He then added a minuscule alphabet letter (ranging from a–l) afterwards. Thus, for example, scribe Df refers to a scribe whose hand appears during the chancellorship of Hugh of Roxburgh (1189–99). Rather interestingly, no further scribes of royal charters could be identified in any chancellorship later than Hugh's (that is, there was no new scribe identifiable for the fifteen years 1199–1214).  No scribe in RRS, ii, therefore was assigned a majuscule later than 'D'

As stated above, Barrow also identified five scribes to whom he added a recorded identity. Richard of Lincoln (Aj); Gervase (Da, perhaps Db as well), Hugh de sigillo (Dd), William de Boscho (De) and Gilbert of Stirling (Df). Four out of the five 'professional chapel scribes' identified by Barrow started to work during the chancellorship of Hugh of Roxburgh (1189–99). If this is borne out by the work on Models of Authority, its significance would need to be examined further, because it may well be related to the changing capacities of royal government developing in the same period, most notably in probative accounting.[6]

But Barrow's methodology must also be investigated. The basis of his identifications was, in general, the witness lists to royal charters. Barrow noted that, for example, Ricardus clericus meus (or some similar variant) attested many of the charters which he had identified as written by the same hand (Aj), often in the penultimate or last position in the witness-list. As it is not unusual for self-identified scribes to appear in last position, Barrow posited that Aj 'may be identified with the king's clerk Richard of Lincoln'.[4] Only Scribe De ever identified himself. RRS, ii, no. 434 contains an unusual dating subscription: 'datum per manum Willelmi de Boscho tempore Florentii Cancellarii anno ab incarnatione domini mciiio' ('dated by the hand of William de Boscho in the time of Chancellor Florence in the year from the Incarnation of the Lord, 1203').[7] Thus, four out of Barrow's five named scribes who wrote royal charters were identified on the basis of their attestation patterns. The extent to which this methodology can be justified will be examined in MoA. In particular, did these scribes exclusively write royal charters, or are their hands identifiable in other, non-royal documents?

Appendix: A list of Barrow's scribes

G.W.S. Barrow provided a full list of scribal identifications at pp. 85–91 of RRS, ii. However, a small number of his identifications did not correspond to those provided in the critical apparatus to each edited charter text. Thus, in his full list in the introduction, RRS, ii, no. 59 (a charter of King William to Geoffrey de Melville, see the image: here) was assigned to Scribe Al but, in the critical apparatus of no. 59, Barrow lists it as belonging to Scribe Aj. Teresa Webber has compared Barrow's full list with each critical apparatus and identified these discrepancies, which have been incorporated into the list which follows. A list of these discrepancies can be found at the end of this Feature of the Month.

In general, readers should note the following categories in the references. The RRS, ii, number refers to the edited text of the charter; the H-number refers to the entry made from that charter in the PoMS database. You can search MoA's own database with either the RRS-number or the H-number (giving only the numerical values: for example, 1/6/61 not H1/6/61).

Barrow's list of scribes was as follows:

The chancellorship of Nicholas of Roxburgh (1165–71).


RRS, ii, no. 76 (H1/6/61).


RRS, ii, no. 81 (H1/6/65), 82 (H1/6/66), 83 (H1/6/67), 85(A) (H1/6/69), 175 (and: 175 duplicate) (H1/6/153), 214 (H1/6/190), 236 (H1/6/212).

also identified in GD 55/1, 55/2, 55/6.


RRS, ii, no. 10 (H1/6/10).


RRS, ii, no. 39 (H1/6/34).


RRS, ii, no. 28 (H1/6/23).


RRS, ii, no. 46 (H1/6/41), 62 (H1/6/47), 66 (A) (H1/6/51), 105 (B) (H1/6/89). 


RRS, ii, no. 26 (H1/6/21).


RRS, ii, no. 25 (H1/6/20), 27 (H1/6/22), 40 (A) (H1/6/35), (41: H1/6/36).


RRS, ii, no. 45 (H1/6/40), 47 (H1/6/42), 48 (H1/6/43), 66 (B) (H1/6/51), 67 (A) (H1/6/52), 75 (H1/6/60), 86 (H1/6/70), 113 (H1/6/97), 135 (H1/6/118), 136 (H1/6/119), 146 (not in PoMS).

Aj (Richard of Lincoln)

RRS, ii, no. 42 (H1/6/37), 43 (H1/6/38), 44 (H1/6/39), 59 (H1/6/44), 73 (H1/6/58), 79 (not in PoMS), 80 (H1/6/64), 101A (H1/6/85), 105 (A) (not in PoMS)), 114 (H1/6/98), 125 (H1/6/108), 133 (H1/6/116), 137 (H1/6/120), 144 (not in PoMS), 152 (H1/6/132), 204 (H1/6/180).


RRS, ii, no. 14 (H1/6/10), 145 (H1/6/126).


RRS, ii, no. 76 (H1/6/154), 206 (H1/6/182).

The chancellorship of Walter de Bidun (1171–8)


RRS, ii, no. 153 (H1/6/133), 165 (H1/6/144), 171 (H1/6/150), 200 (H1/6/176), 201 (H1/6/177), 213 (H1/6/189, but only a facsimile), 263 (not in PoMS).


RRS, ii, no. 46 (B) (H1/6/41), 167 (H1/6/146), 178 (H1/6/156), 181 (H1/6/159), 266 (H1/6/240), 267 (H1/6/241), 294 (H1/6/266), 302 (H1/6/273), 338 (H1/6/309).


RRS, ii, no. 78 (H1/6/63), 199 (H1/6/175).

The chancellorship of Roger de Beaumont (c.1188–9)


RRS, ii, no. 252 (H1/6/228), 268 (H1/6/242), 269 (H1/6/243), 367 (H1/6/338), 368 (H1/6/339).


RRS, ii, no. 264 (H1/6/238), 265 (H1/6/239).


RRS, ii, no. 195 (H1/6/172), 293 (H1/6/265), 301 (H1/6/272), 364 (H1/6/335).


RRS, ii, no. 211 (H1/6/187).


RRS, ii, no. 235 (H1/6/211). 


RRS, ii, no. 322 (H1/6/293).

The chancellorship of Hugh of Roxburgh, 1189–99.

Da (Gervase the clerk)

RRS, ii, no. 286 (H1/6/259), 303 (H/1/6/274), 309 (H1/6/280), 313 (H1/6/284), 340 (H1/6/311), 341 (H1/6/312), 343 (H1/6/314), 350 (H1/6/321), 365 (H1/6/336), 380 (H1/6/351), 414 (H1/6/382), 422 (H1/6/390).

Db (Gervase the clerk?)

RRS, ii, no. 67 (B) (H1/6/52), 314 (H1/6/285), 320 (H1/6/291), 342 (H1/6/313), 369 (H1/6/341), 410 (H1/6/378), 418 (H1/6/386), 429 (H1/6/396).


RRS, ii, no. 85 (B) (H1/6/69), 292 (H1/6/264), 402 (H1/6/370), 403 (H1/6/371).

Dd (Hugh de sigillo)

RRS, ii, no. 375 (H1/6/346), 401 (H1/6/369), 405 (H1/6/373), 427 (H1/6/394), 428 (H1/6/395), 429 (H1/6/396), 432 (H1/6/399), 439 (H1/6/406), 444 (H1/6/411), 446 (H1/6/413), 451(H1/6/418).

De (William de Boscho)

RRS, ii, no. 358 (H1/6/329), 377 (H1/6/348), 383 (H1/6/354), 386 (H1/6/357), (404?, but only a facsimile: H1/6/372), 434 (H1/6/401).

Df (Gilbert of Stirling)

RRS, ii, no. 389 (H1/6/360), 445 (H1/6/412), 459 (H1/6/426), 460 (H1/6/427), 461 (H/1/6/428), 462 (H1/6/429), 463 (H1/6/430), 464 (H1/6/431), 469 (H1/6/435), 470 (H1/6/436), 471 (H1/6/437), 474 (H1/6/440), 479 (H1/6/445), 483 (H1/6/449), 484 (H1/6/450), 485 (H1/6/451), 486 (H1/6/452), 489 (H1/6/455), 493 (H1/6/459), 496 (H1/6/463), 504 (H1/6/471), 508 (H1/6/474), 514 (H1/6/480), 516 (H1/6/482), 517 (H1/6/483), 520 (H1/6/486), 521 (H1/6/487), 524 (H1/6/490).

Discrepancies between critical apparatus and Barrow's full list of scribes

This list has been assembled by Teresa Webber

1) RRS, ii, no. 85 (A) was assigned to Scribe Ai in Barrow's list (p. 86) but to Scribe Ab in the critical apparatus (p. 183).

2) RRS, ii, no. 59 was assigned to Scribe Al (p. 87) in Barrow's list but to Scribe Aj in the critical apparatus (p. 160).

3) RRS, ii, no. 201 was omitted from Barrow's list but assigned to Scribe Ba in the critical apparatus (p. 255).

4) RRS, ii, no. 292 was omitted from Barrow's list but was assigned to Scribe Dc in his edition (p. 313).

5) RRS, ii, no. 295 and 495 (assigned to Scribes Dc and Df respectively) do not survive as originals (instead, facsimile copies) and so have not been included here. It is probable that the inclusion of nos. 295 and 495 were typographical errors; the correct entry should have been no. 292 and 492. Our thanks to Dauvit Broun for pointing this out.


[1] Regesta Regum Scottorum Volume 2: the Acts of William I, 1165–1214, ed. G.W.S. Barrow with the collaboration of W.W. Scott (Edinburgh, 1971), p. 84. Barrow identified '7 or 8 written by the same hand', which he labelled Scribe A; see Regesta Regum Scottorum Volume 1: the Acts of Malcolm IV, 1153–65, ed. G.W.S. Barrow (Edinburgh, 1960), pp. 85–6.

[2] RRS, ii, p. 84.

[3] RRS, ii, p. 85. Barrow also used the BLitt thesis of Norman Shead to identify Gervase the clerk as once belonging to the household of Joscelin, bishop of Glasgow (ibid. ,p. 89).

[4] RRS, ii, pp. 85.

[5] RRS, ii, p. 84.

[6] Alice Taylor, The Shape of the State in Medieval Scotland, 1124–1290 (Oxford, 2016), pp. 361–6.

[7] This is unusual in another way. A.D. dates were, in general, not included in the dating clauses of royal acta until 1221 (and then they were quickly replaced by regnal year by April 1222); for dating clauses in general, see Dauvit Broun, Scottish Independence and the Idea of Britain from the Picts to Alexander III (Edinburgh, 2007), pp. 191–206.


Posts by Date


RSS / Atom