A Book of Hours Owned by the Duke de Berry?

The first proper examination of British Library, Yates Thompson MS. 37 was Sydney Cockerell's description for Henry Yates Thompson's catalogue, in which he wrote:

“Provenance. In a miniature on f. 92 a man in a rich red robe kneels before the Virgin and Child. He somewhat resembles the portraits of Jean Duke of Berry, and the fact that the Hours are of the use of Bourges makes it not unreasonable to conjecture that the book was intended for that great collector, though there is no signature or coat of arms to make it certain.”

BL, Yates Thompson MS. 37, fol.92r (detail)

All subsequent writers seem to have ignored this idea, or rejected it implicitly: Meiss, for example, stated that “Everything points to the fact that the Book of Hours was painted between 1405 and 1410 for a member of Jean de Berry's court”. The British Library online description reports Cockerell's suggestion, without exploring its feasibility, as well as Meiss's.

Presumably unwilling to venture his next suggestion in the main text, Cockerell presented it instead in smaller type, as a footnote:

“In the catalogue of his library printed by Delisle in Le Cabinet des manuscrits (vol. III, pp. 170-194), no. 110 is described as follows: Unes petites heures, es quelles sont les heures de Nostre Dame, les sept psaumes, vigiles de mors et autres devocions, et au commencement a une oraison de saint Jehan Baptiste et le kalendrier, lesquelles monseigneur acheta à Paris en son hostel de Neelle, le 11 décembre 1415, 50 escus. f. 2 Quoniam. In order to apply this entry to the present manuscript it is necessary to assume that the cataloguer was capable of mistaking the sequence of St John's Gospel for a prayer to St John the Baptist and that he intended the word Quoniam to be taken as the opening word of the second leaf of the Hours of the Virgin (ignoring the preliminary matter). If these large assumptions are permissible, the description would fit the manuscript perfectly. It is at any rate a coincidence worth noting that the second leaf of the Hours of the Virgin in this MS. begins with Quoniam.”

Following the calendar at the beginning of the book there are the normal extracts from the four gospels, starting with that from John the Evangelist, not a prayer to John the Baptist as it is described in the inventory. In my view this is easily explainable as a simple scribal error (after all, in the French of the inventory, both names begin “Jehan” and end “-iste”, with just a few differing syllables in between) which could be due to a variety of factors such as misreading the manuscript, misremembering what one has correctly read in the manuscript, or mistranscribing from a rough draft. In fact, a prayer to John the Baptist as the first text after the calendar in a book of hours would be so odd that even if one were not attempting to associate this inventory item with an extant manuscript, one might very possibly consider this as a potential error in the inventory.

The second, apparently much more serious, problem for Cockerell's identification concerns the secundo folio. It is standard in medieval catalogues and inventories for secundo folio references to be taken from the second leaf of the volume or, as in the present case where the first text is a calendar, from the subsequent text. As the whole purpose of a secundo folio reference in inventories such as those of the Duke de Berry is to provide an unambiguous identification of a specific manuscript, the apparent mis-match seems to be a conclusive argument against Cockerell's identification. But as Cockerell observed, and as Séverine Lepape brought to my attention several years ago when she was describing the manuscript for the BL's online catalogue of illuminated manuscripts, “quoniam” is the first word of the second leaf of the Hours of the Virgin, on what is now numbered as fol.20r of the manuscript.

A comparison between Cockerell 's transcription of Delisle's transcription from the Duke de Berry's “1413” inventory and Delisle's transcription itself, however, shows that Cockerell made some silent alterations; in fact Delisle's text is as follows:

“Unes petites heures es quelles, es quelles sont les heures de Nostre Dame, les sept psaumes, vigiles des mors et autres devocions, et au commencement a une oroison de saint Jehan Baptiste et le kalendrier, lesquelles monseigneur acheta à Paris en son hostel de Neelle, le 11 décembre 1415, 50 escus. B 162, C 739. Quoniam. – 15 l.”

The alterations made by Cockerell are small and understandable, but are crucially significant. Where Delisle provides references to this item in the Duke's “1413” and “1416” inventories, using the sigla and item numbers “B 162, C. 739”, Cockerell instead inserts “f. 2”, to make clear that “Quoniam” is a secundo folio reference.

Even Delisle's transcription published in 1881, however, is not complete. Looking at the more precise transcription of the Duke de Berry's inventories published by Guiffrey in 1894, we see that Delisle not only omitted the detailed description of the binding, but also a few other details (bold emphasis added):

“1232. Item, unes petites Heures, esquelles sont les Heures de Nostre Dame, les sept Psaulmes, Vigilles de mors, et autres devocions; et au commancement a une oroison de saint Jehan Baptiste et le kalendrier; et a escript au commancement du second fueillet desdictes Heures de Nostre-Dame: quoniam; couvertes de drap d'or, fermans à deux fermours d'or esmaillez aux armes de Monseigneur, ouvré ledit drapt d'or à fleurs de liz, et par dessus une chemise de drap de damas bleu, doublé de tiercelin rouge; lesquelles Heures mondit Seigneur achapta à Paris en son hostel de Neelle, le XI^e jour de decembre quatre cens et quinze, pour le pris et somme de cinquante escuz d'or comptans de …
Redditus ut supra.
[S G, n^o 739; prisé xv liv. t.]”

The inventory explicitly states that the secundo folio reference is taken from the second leaf of the Hours of the Virgin, not the second leaf of the volume. Delisle was obviously aware of this, as becomes clear if you read his entire text, but to the more casual reader it is not obvious that this was normal practice for the books of hours in the Duke's inventories. The following are equivalent passages in other inventory entries:

“960. Item, unes belles Heures, très bien et richement historiées […] et au commancement du second fueillet desdictes Heures de Nostre Dame, a escript: audieritis; […]”

“961. Item, unes très grans moult belles et riches Heures, très notablement enluminées et richement historiées […] et au commancement du second fueillet des Heures Nostre Dame a escript: flamme; […]”

“997. Item, unes Heures esquelles sont les Heures Nostre Dame, les sept Pseaumes, […] et a escript au commencement du second fueillet desdictes Heures de Nostre Dame: stirpis; […]”

“1002. Item, unes Petites Heures esquelles sont les Heures de Nostre Dame, les sept Pseaumes, […] et au commencement du second fueillet des Heures de Nostre Dame a escript: sunt omnes fines; […]”

Thus of the two objections against the identification of Yates Thompson MS. 37 as the Hours bought by the Duke de Berry at the Hôtel de Nesle on 11 December 1415, one is in my opinion easily dismissed as a simple mistake, and the other is in fact very strong evidence for the attribution, instead of evidence against it.