Saturday 21 October 2023

A.C. de la Mare and Neil Ker on Describing Script


Anyone who has ever attempted to describe a manuscript will have faced the issue of terminology for describing script. Over the course of the last 75 years numerous books and articles have been written, and conferences held [1], discussing the issue, and yet we have still not arrived at any real consensus.

I think that two main drivers lay behind these publications and conferences, especially the earlier ones. One was to try to make palaeography more "scientific" (with implications of reliability and accuracy), and the other was to compensate for a lack of reproductions. It seems to me that the former was somewhat misguided [2], and the latter is now outdated [3].

If a series of Books of Hours are described as being mid 15th-century French, Flemish, Spanish, Dutch, German, and Italian, then the knowledgeable reader will have a very good idea of what the script of each of them looks like, and how they differ from one another, without a formal description of their script. The same goes for an 11th-century biblical text, a 12th-century patristic text, a 13th-century academic text, a 14th-century legal text, a 15th-century Humanistic one, and so on. The date and place of origin, plus the type of text, is usually enough to indicate in general terms what the script looks like. No amount of description can ever convey its precise appearance  any attempt to do so is at best futile, and at worst misleading [4].

It seems to me that what is most valuable is to attempt to describe script in those cases where the reader cannot reliably imagine its general appearance, and in cases where the appearance is not what might be expected. Thus if a 'Humanistic' text is written in Gothic script, or a liturgical text is written in unexpectedly informal script, this certainly needs to be emphasised. At the top of this post, for example, is a 15th-century Sarum Breviary written in a script which is shockingly more cursive than expected.

I therefore personally usually use the most basic descriptors [5], including "Caroline", "Romanesque" [6], "Gothic", and "Humanistic". I may make additional comments about appearance or quality if I think they are noteworthy: "exceptionally fine", "rather irregular", and so on. The reader can see that these are my subjective opinions, and can treat them accordingly  e.g. by simply ignoring them, if they wish.

With this as preamble, I was very interested recently to notice that in the introduction to her catalogue of the Lyell manuscripts at the Bodleian Library [7], A.C. (Tilly) de la Mare, arguably the greatest palaeographer of the 20th century [8], states that she describes the "type of script, if distinctive" (my emphasis)

She then explains that her use of the terms "secretary" and "anglicana" follows that of Neil Ker, explaining that "his criteria for describing script have generally been followed here". For the latter, she refers us to Ker's Medieval Manuscripts in British Libraries, I (Oxford, 1969), pp. xixii.

So, what did (arguably) the 20th century's greatest cataloguer of manuscripts have to say on the matter? He wrote, 

"If the script is caroline minuscule or its successor from c. 1150, textura (text hand, littera textualis, gothic), it is not mentioned as a rule" (my emphasis)

In the same year that Ker wrote this, S. Harrison Thomson, in the Introduction to his Latin Bookhands of the Later Middle Ages, 1100–1500 (Cambridge, 1969), independently wrote:

"In matters of nomenclature I have been content to limit myself to a minimum. I sense a tendency among palaeographers to multiply descriptive adjectives for variations in script. This tendency seems to me confusing rather than enlightening."

I expect I will receive contrary opinions from people who know more about script than I do, so I should clarify one point: I think that if someone is writing about script per se, then it is incumbent upon them to use whatever words will best convey their message to their readers. But writing about script is a very different task from what concerns me here: writing a description of a manuscript.


[1] e.g. B. Bischoff, G. I. Lieftinck, and G. Battelii, Nomenclature des écritures livresques du IXe au XVIe siècle: Premier colloque international de paléographie Latine, Paris 28-30 avril 1953 (Paris: CNRS, 1954).

[2] I am strongly in favour of the use of accurate, consistent, and unambiguous terminology when describing manuscripts (as explained at length in this blogpost), but for script it seems that there simply is not enough agreement among specialists to make this feasible at present.

[3] In the past 30 years, and especially the last decade, it has become very rare for a catalogue of manuscripts to be published without a significant number of reproductions. Two striking recent exceptions are Richard H. Rouse and Mary A. Rouse, The Richard and Mary Rouse Collection, Medieval and Renaissance Manuscripts of the UCLA Library Special Collections, 1 (Tempe, Arizona: ACMRS, Arizona Center for Medieval and Renaissance Studies, 2017), which has no images at all, and David T. Gura, A Descriptive Catalogue of the Medieval and Renaissance Manuscripts of the University of Notre Dame and Saint Mary’s College (University of Notre Dame Press, 2016), which describes 69 codices and leaves/fragments from 86 more, yet has only 8 plates.

[4] Some methods seem to me to be overly cumbersome (and also subjective); for example, Michelle P. Brown, A Guide to Western Historical Scripts from Antiquity to 1600 (London: British Library, 1990), proposes the adoption of Julian Brown's terminology, which involves descriptions such as "littera minuscula protogotica textualis libraria media / formata" (p. 74); I am not aware of any catalogue that has adopted this system.

[5] In an otherwise complimentary review of my catalogue of the manuscripts at The Queen's College, Oxford, one reviewer wrote:

"The one area in which Kidd’s level of detail tapers off is in the description of scripts. These are described in very general and often vague terms, which, in contrast to the other fields, can create ambiguity for the reader in the classification of script type and its level of execution."

This is certainly a valid criticism in any cases where I described script incorrectly, but since every manuscript in the catalogue has at least two reproductions, I do not think that more detail in my "classification of script type" of script would have been necessary or helpful and, thanks to the images, there is little chance that my poor choice of terminology would create ambiguity.

[6] 12th-century script terminology is particularly problematic, because it attemps to label a non-style as if it were a style (it would arguably be more accurate to treat it as a transitional phase between two other styles, Caroline and Gothic, neither fish nor fowl), with terms such as protogothic, praegothica, pregothic, etc. But such terms all describe the script with reference to Gothic, a script style that did not yet exist; it would be more accurate to describe it in terms of the script from which it evolved, with a term such as "late Caroline". Hence my preference for "Romanesque", which (despite its own problems) is a well-established concept for the discussion of 12th-century styles.

[7] Albinia de la Mare, Catalogue of the Collection of Medieval Manuscripts Bequeathed to the Bodleian Library, Oxford, by James P. R. Lyell (Oxford: Clarendon Press, 1971).

[8] This may be an Anglo-centric opinion, but my impression is that Tilly's expertise covered the whole gamut of medieval and Renaissance script (she did not "just" work on Humanistic scripts), and she was also enormously knowledgeable about illumination, bindings, provenance, etc., something that cannot be said of many other palaeographers. She was also a great scholar, in my opinion, because of the way in which she was quick to publicly identify and correct her own mistakes.


  1. Thanks, Peter, for this interesting and slightly provocative explanation. I fear I don't agree with your assumption that an experienced codicologist/palaeographer will nearly always know from the contents, period and place of origin what kind of script is used. That is especially true for the later middle ages for which one can expect textualis, cursiva or hybrida for many of the kinds of text produced. I am currently studying a fragment of Ruusbroec's works. Size and layout suggest that it is linked to a group of early manuscripts, written in the second half or last quarter of the 14th century. The fragment I am looking at is written in 'cursiva' and probably in the 15th century, but I want to compare its script to the known early ones. Of the eleven early manuscripts I only find online reproductions of three of them and photographs in publications of another two. They are all written in 'textualis'. For the other six none of the descriptions in the online or printed catalogues specify what kind of script they are written in. Which means I need to visit six libraries (in Belgium, France, the Netherlands and Austria) to simply check this for myself. This could have been avoided if cataloguers had specified the script type. (Godfried Croenen)

  2. Dear Godfried,
    Thanks for this. You are of course right that I was being deliberately provocative! And I absolutely agree that there are some types of MS for which one cannot confidently guess the likely appearance, and for these the cataloguers in the past half-century would hopefully have specified the type of script. (Although it's not surprising if they didn't in older catalogues; and for a catalogue published in the last decade they would hopefully have provided a reproduction, which would render the written description unnecessary.)
    To take your example, though, I suspect that one cataloguer's textura another's gothic, etc., so if the appearance of the script really matters to you, then you're going to need to see an image, or the original MS, anyway.
    But I very doubt that you actually need to visit all six libraries if you just want to know what kind of script those MSS are written in: librarians are usually happy to answer such enquiries (and/or, these days, take digital snaps); and failing that, you should be able to find someone via social media who can go and look at/photograph each MS for you. Good luck!

  3. I should also add that -- as shown by my quotations above -- de la Mare and Ker would both use terms such as "secretary" and "anglicana" when appropriate.



I may ignore and delete anonymous comments