In her attempts to unravel the origins of the Towneley Plays manuscript and nineteenth-century receptions of it, Meg Twycross ‘points a (sympathetic) finger at the scholar and antiquary Francis Douce for the muddle we find ourselves in today’.Footnote ¹ Twycross presents a detailed reading of Douce’s correspondence and his shaping influence over the catalogue of the 1814 auction, which shows that many current presuppositions about the play can be traced to this origin. Twycross’s suggestion is that Douce’s misconceptions seem to have been born of a mixture of educated but misplaced speculation, and the commercial requirements of the auction house. Despite the sympathy that she expresses for this collector of manuscripts, rare books and other curios, the overall impression is that Douce’s scholarship was less than rigorous, and that this has caused difficulties for scholars following him.

In an earlier article, I explored Douce’s antiquarian practices with regard to the material that he pasted into his early and rare printed books.Footnote ² That examination revealed Douce to be a respectful owner of the books that he collected, adopting minimally invasive annotation practices such as the use of pencil or pasted leaves to avoid leaving permanent marks on the printed page. That examination also demonstrated that Douce was able to navigate his way without any visible cataloguing system through a complex collection of manuscripts, early books and curios, returning to and revising his earlier opinions. This article will investigate more closely the content of Douce’s practices of engagement with his text to see whether the impression formed by Twycross is borne out by other instances. By looking at the content of Douce’s annotations, the article will examine what they tell us about what he read and how he read it and what secondary and external sources he consulted to reach his conclusions. I will consider to what extent Douce was up-to-date with current scholarship and more generally what criteria he used to approach the evidence before him. Such questions are important because of the negative connotations often associated with the disciplinary label of antiquary, and because of the concern of modern professional scholars to define this approach and to separate it from – or associate it with – their own practices whose roots are said to have emerged around the period in which Douce flourished. According to this paradigm, antiquarian scholars were either the forerunners of today’s academic and scholarly professions, cultivating a breadth of knowledge and expertise that rivals that of modern specialists, or they represent over-enthusiastic eccentrics of the kind satirised in George Eliot’s 1872 novel Middlemarch. That novel, which is set in the years leading up to 1832, presents a stereotypical amateur scholar who, like many of the historical antiquarians studied by Philippa Levine and of Francis Douce’s learned correspondents, earned their living from a position in the church.Footnote ³ Interestingly, Levine identifies the era described by Eliot as key in creating this distinction between antiquarian and professional, singling out the importance of the government role in creating the full-time employees of the Public Records Office in 1838 in the move towards professionalisation. This move inevitably strengthened satirical currents that ridiculed the antiquarians, although, as Martin Myrone and Lucy Peltz point out, this satire is as old as the antiquarian practices that it lampoons.Footnote ⁴ Moreover, as Myrone and Peltz argue, satirical criticism frequently conceals the intellectual, social, imaginative and political links between antiquarian practices and the scholarly habits of the modern age. Douce himself is an example of these links because, although he is most usually described as an antiquary with a collection amassed in a private capacity whose eclecticism reflects the multidisciplinary approach of antiquarianism, he was also representative of the emerging professionalism of the era, having spent the period 1807–11 as keeper of manuscripts in the British Museum.Footnote ⁵ His habit of annotating printed pages in pencil, for instance, reflects the professional practice of the British Museum Library.Footnote ⁶ On the other hand, his private collection remained uncatalogued, despite the fact that his tenure at the British Museum coincided with a period of intense efforts towards cataloguing that collection.Footnote ⁷ In this, Douce embodies the sort of approach described by Rosemary Sweet, where the need for a catalogue was acknowledged but not respected by many antiquaries of the period, who often collected an excess of material.Footnote ⁸

Whether or not we would accept that Francis Douce’s collection was excessive, it was certainly extensive. The catalogue of books and manuscripts that were bequeathed to the Bodleian contains more 393 manuscripts and over 19,000 volumes of printed material.Footnote ⁹ Nor were books the only items that Douce collected. Antiquaries, as Benjamin Anderson and Felipe Rojas summarise, are ‘people who are interested in and knowledgeable about the material traces of the past’ and Douce’s acquisitions featured a variety of examples of these material traces.Footnote ¹⁰ A N L Munby cites an ‘ancient rolling pin’, ‘parts of the rood screen from Southwold’, a ‘gnostic crocodile’ and ‘a plaster cast of a hermaphrodite’.Footnote ¹¹ It is not possible within the scope of this article to give a complete overview of this collection, so I have chosen to focus on its early French printed material. This is, in part, because of my own research interests, but it is also a significant part of the collection. Of the 492 manuscripts from Douce’s collection held in the Bodleian, 140 contain French, making it the third language represented in the collection behind only Latin and English. The extensive collection of incunabula and early printed books is less well catalogued in respect to language, but examining the titles in the catalogue reveals that French also features prominently: of the nine per cent of Douce’s books printed before 1600, 658, or nineteen per cent, have titles in French. As this figure suggests, Douce was particularly interested in the language and literature of France: he had spent some of his education ‘at a French academy, kept by a pompous and ignorant Life-guardsman […] and he made no other acquirement there than a little French’.Footnote ¹² It was an interest he maintained throughout his life. From 1796 until 1833 Douce corresponded with Gervais de La Rue, whom he had met through the Royal Society of Antiquaries when the latter was exiled in England during the Revolution.Footnote ¹³ De La Rue, who had become president of the University of Caen, where he held the chair of history, was also a founding member of the Society of Antiquaries of Normandy, thus straddling, like Douce, the supposed divide between professional and antiquarian spheres of activity. Douce’s papers naturally only contain La Rue’s letters to Douce, and not the English scholar’s replies, but those letters are in large sheets of densely packed French. So too are the letters written to Douce by Auguste Le Prévost, another specialist in Norman history but also in archaeology and botany, who corresponded with Douce in the years 1817–22.Footnote ¹⁴

Douce did not simply collect French incunabula; he also read them, adding notes about the work and less frequently about the content to the preliminary pages of the volumes and to sheets pasted into his books. I was able to examine 293 of these items (mainly those produced before 1550), along with fifty further French items published before 1650. The majority of my sample of the earlier books contains annotations by Douce, as shown in fig 1.

Fig 1. Annotations in Pre-1600 printed books in Francis Douce’s collection. Image: author.

This is also true of the later sample, of which fifty-three per cent contain annotations by Douce, but the sample is much smaller, making it harder to determine how representative it is. Ten of the pre-1600 volumes that are not annotated are later volumes in multi-volume works where the first volume contains Douce’s bibliographical notes on the work as a whole, so the proportion of works that is annotated is greater than the proportion of volumes. The material traces of scholarship that Douce leaves in his collection give an insight into how an informed reader in this crucial period in English intellectual history understood and engaged with the works printed in French in the first century of print.

The Douce collection is indeed a rich field of investigation for anyone who wants to study antiquarian and scholarly habits of the early years of the nineteenth century, and in fact it presents evidence of such interests from earlier periods as well. The marks left on volumes in Douce’s collection provide insight not only into Douce’s own reading practices, but also into those of earlier readers, hence the four per cent of volumes that bear annotations but not necessarily by Douce. Unlike other collectors, Douce did not erase annotations of earlier readers, or seek to suppress them, and many of his volumes contain annotations by Douce and by other readers.Footnote ¹⁵ It is true that there are traces of erasure in the volumes in his collection: for instance, Bodleian, Douce S 218, a copy of La Chasse et le depart damours, contains a librarian’s note in pencil on a preliminary page, beneath Douce’s bibliographical notes in pen to the effect that ‘Inscriptions on last leaf indecipherable under u. violet lamp’ (fig 2).Footnote ¹⁶ This is an observation that I can confirm with some frustration, given that the inscriptions that occur on this last leaf and elsewhere in the book seem tantalizingly close to being legible. Other elements of reader engagement with the same copy can still be deciphered. The frequent underlining shows passages that the reader found significant while the final page also contains a sketch of the head of a man wearing a hat, suggesting that the illegible material may fall more either into the category of doodles or into that of considered reaction to the text.

Fig 2. Final page of La Chasse et le depart damours (Bodleian, Douce S 218) showing effaced reader inscriptions. Image: by kind permission of The Bodleian Libraries, University of Oxford.

The fading of the ink may not be a result of deliberate attempts to remove these marks, but it may well have been. Elsewhere in the collection, we frequently see close cropping of margins to remove marginal annotations.Footnote ¹⁷ In one scandalous instance, attempts have been made to bleach out Douce’s own marks of ownership by a thief who removed the book from the Bodleian and substituted a false provenance for the stolen copy (fig 3).Footnote ¹⁸ The story of theft and recovery in 1978 is told in pencil notes made by a librarian in the preliminary pages, which are also reproduced in the electronic catalogue of the library.Footnote ¹⁹

Fig 3. False book plate inserted into Bodleian, Douce CC 23, Gilles Corozet’s Hetacomgraphie, and annotation attesting to the volume’s theft and recovery: a) inside upper board; b) first flyleaf recto. Images: by kind permission of The Bodleian Libraries, University of Oxford.

However, although practices of cleaning and erasure are represented, so too are instances where marginal notes and other marks such as underlining have been preserved, and, a result, Douce’s collection in the Bodleian gives an insight into several generations of reading practices – and other patterns of engagement with the materiality of books – that books encountered between their printing in the years 1470–1600 and their arrival in the Oxford library following Douce’s death in 1834.Footnote ²⁰

The most consistent source of this engagement, however, is Francis Douce himself: the collection provides many opportunities for examining his practices. We see that these are minimally invasive, as we might expect from a professional librarian, that they show evidence that the scholar returned to his volumes and recorded fresh scholarship and new opinions on the works in his collection and that in so doing he considered the whole collection – print and manuscript. References in his notes allow us to see what works he was consulting, and in some cases what personal connections he was exploiting to form his opinions. The degree of accuracy of his references raises the question of whether these were intended to be personal notes or if they were aimed at other readers. A comparison with Douce’s published scholarship casts further light on this question as it allows us to judge his accuracy when he certainly was writing for others and compare it to his practices in his personal collection.

In terms of the minimally invasive practices, these take two forms: ink notes that are kept separate from the printed text and brief marginal pencil annotations. As we have seen, seventy-eight per cent of the sample is annotated and this most frequently takes the form of brief bibliographical notes in ink in Douce’s hand, usually in the blank preliminary pages of the copy and sometimes in paper tipped into the volume for the purpose. This practice of pasting sheets for annotation into the volume to be annotated can also be seen in the collection of Frederic Madden, who also served as keeper of manuscripts at the British Museum between 1837 and 1866. If anything, Douce’s practice was less invasive than Madden’s as Madden frequently added bibliographical references in ink onto the pages of the book to be annotated. Many of Douce’s volumes also contain notes or other marks of reading in the body of the text itself but these often take the form of bookmarks, frequently slips of paper cut off blank or nearly blank portions of letters addressed to Douce (some contain pieces of his name or address), to which Douce has occasionally added additional remarks. Sometimes longer notes have been inserted on leaves between the pages of the text, as was Madden’s standard practice, and more rarely Douce has made pencil (but not ink) annotations in the margins. That these are by Douce can be seen from the fact that they reproduce habits that we see in his more extensive ink notes and in particular his habit of marking points on which he is uncertain with a distinctive q (for question or query). Such notes are, however, rare and seem to indicate a reluctance to intervene on the printed page.

In one instance, Frederic Madden owned and annotated a book, of which Frances Douce also owned a copy. The book is a copy of Claude Fauchet’s Recueil de l’origine de la langue et poesie françoise, printed in Paris in 1581. Madden’s copy is held by the Houghton Library in Harvard,Footnote ²¹ while Douce’s copy is in the Bodleian.Footnote ²² Douce gives very few details in his annotations except to note that he bought the volume in 1818 ‘at the sale of Miss Humfries’ books’. Madden, on the other hand, makes extensive marginal annotations in ink, copying those in another copy in the British Museum collection,Footnote ²³ formerly in the library of George iii. These annotations, in French, in what Madden calls ‘King’s copy’, give information about manuscript readings from a medieval manuscript containing a fabliaux collection now held in the Bibliothèque nationale de France.Footnote ²⁴ The annotator of the King’s copy slightly modernises the French copied into marginal notes, and Madden does likewise, with the result that Madden’s annotations present a different orthography from the manuscript referenced. However, it is clear that Madden’s main concern was to preserve references to a manuscript preserved in a particular copy in the British Museum collection. Madden’s notes in this instance do not reference bibliographical literature and they appear to have been written at one time on the basis of a single reading of his source material.Footnote ²⁵

By contrast, evidence that Francis Douce did not produce his notes in ink on the preliminary pages at a single point in time can be seen in both the text and the handwriting of those notes. Many show that Douce did not simply write in his books at the point that they entered his collection but rather returned to them and reevaluated his thoughts, perhaps in the light of subsequent reading, perhaps as new scholarship emerged. Douce was not the only reader to make bibliographical notes and revisions to them in this way, and in one striking case – that of Le Tableau des riches inventions couvertes de voile des feintes amoureuses, qui sont representees dans le Songe de Poliphile Footnote ²⁶ – he uses his notes to record the deliberations of earlier readers (fig 4).

Fig 4. Francis Douce’s annotations on the flyleaf of Le Tableau des riches inventions couvertes de voile des feintes amoureuses, qui sont representees dans le Songe de Poliphile (Bodleian, Douce C 649). Image: by kind permission of The Bodleian Libraries, University of Oxford.

In his notes on the flyleaf of this copy, Douce reproduces the manuscript notes that are to be found in Alexander Pope’s personal copy, which he also notes is ‘now in M^r [James] Bindley’s and not M^r [William] Beckford’s, as stated by mistake in the Bibl.[iotheca] Spenceriana vol. IV, p. 165’.Footnote ²⁷ The first part of these notes indicate that the Hypnerotomachia Poliphilii was ‘Written in 1467 at Trevisa, first printed in Italian at Venice in 1498* by Aldus’. This asterisk corresponds to one at the foot of the leaf, ‘* NB. M^r Warton has thus corrected this mistake “1499. See Menag. V. 4.”’. Here one owner of a copy of the book has revised the notes recorded by a previous owner, and Francis Douce has felt that it is important to note both sets of annotations. The way in which he does this, with an asterisk indicating a note at the bottom of the page, suggests that it was not Douce’s initial intention to include both sets of notes, and this seems to be confirmed by the fact that the note recording Warton’s revision is in a darker ink from that showing Pope’s initial dating.

A similar set of annotations can be seen in an undated sixteenth-century edition of L’Histoire du noble preux & vaillant cheualier Guillaume de Palerne. Et de la belle Melior Footnote ²⁸(fig 5).

Fig 5. Douce’s annotations in Bodleian, Douce D 232: a) letter part 1; b) letter part 2; c) first flyleaf recto. Images: by kind permission of The Bodleian Libraries, University of Oxford.

Here again Douce takes issue with earlier scholarship that appears in the bibliographical notes in another copy ‘in the hand writing of M^r Kitson’. These notes had engaged with three generations of the Bonfons family: the grandfather Nicolas, the father Jean and the son, also named Nicolas, active 1550–90. The earlier opinion, which Douce cites, suggests that the volume in question ‘has certainly a more antique appearance than even the first of those printed and may possibly have been printed by a Nicholas [sic.] the father of Jean’, suggesting a date of publication around 1525.Footnote ²⁹ Again using an asterisk and an appended note, Douce writes ‘*No. This one of the latest of Nicholas’s public[ations] as the type and other internal evidence demonstrate F. D’. The fact that Douce signs this note with his initials demonstrates that he wishes to claim this opinion as his own. The notes in both Le Tableau des riches inventions and in Guillaume de Palerne transfer earlier manuscript annotations relating to bibliographical information from other copies of a work into the copy in Douce’s collection. In both these cases, the chronology of the added notes is clear from the way that they are laid out on the leaf and from the fact that, in writing them, Douce is consciously revising an earlier opinion. It is perhaps significant that in these two cases the opinion that he is revising is that of someone else, which may explain why Douce is so ready to draw attention to what he considers to be a previous error and to claim the correction in his own name. In the case of Guillaume de Palerne, it seems that Douce’s second thoughts are his better ones. Although the precise dating of this volume is still not established, subsequent scholarship seems to agree with Douce’s revised opinion, with the Houghton Library in Harvard cataloguing their copy to ‘cir.1550’ (ie around the beginning of the second Nicolas Bonfons’s activity), while the Universal Short Title Catalogue dates it to 1575.Footnote ³⁰

In other instances, revisions appear to reflect Francis Douce revisiting and revising his own previous thinking on a matter. In these cases, Douce may not signal his corrections so explicitly and notes added later can be detected principally through changes in the ink or the width of the pen used to make the note, so that the aspect of the handwriting is different. Changes in aspect can also be noted in the example quoted above dealing with the putative multiple Nicholas Bonfons, where the hand that has copied Kitson’s opinion is much rounder and more spaced-out than that which has recorded Douce’s evaluation of that opinion. Such changes in Douce’s notes are relatively common, particularly when the annotations contain quoted material, and especially when the text is in languages other than English. In the example of Guillaume de Palerne, it remains overwhelmingly likely that Douce wrote both parts of the note, despite the difference in the hands. Another example of a similar revision – this time of his own opinion – displays the same phenomenon. Douce 194 is a copy of an edition of the Roman de la Rose with neither date nor place, on whose endpaper Douce initially wrote ‘This book was undoubtedly printed by Guillaume Le Roy at Lyons about 1480’. He printed in that city from 1477 to 1488. This is probably the first edition of the Roman de la Rose. Later, Douce returned to the work and, on a slip of paper pasted into the volume, he added ‘I find reason to doubt my former note as to this book being printed at Lyon. I suspect it was p.d at Bruges or somewhere in Flanders’.Footnote ³¹ Unfortunately, Douce does not share his reason for revising his opinion and in this instance subsequent scholarship has supported his first conclusion regarding printer, date and place.Footnote ³² On the matter of its being the first edition, however, recent scholarship departs from Douce’s analysis, pointing to several earlier editions, including another very similar version by Jean Syber, also in Lyon, of which this is a near facsimile.Footnote ³³ It may be that Douce’s second thoughts about the place were prompted by his belief that this was a very early edition. This would be a logical assumption, given that the printer Le Roy was from the Low Countries and might have been presumed to have worked there in the early part of his career. If this was the reasoning, then Douce’s second thoughts are entirely out of keeping with modern scholarship, as well as being unmotivated in his note. The absence of justification means that the change of opinion appears to be the product of Douce’s own personal reflection, following a line of reasoning at which we can only guess. The fact that Douce does not say what has led him to change his thinking raises the question of whether the note was intended to be a purely personal reminder for the collector, rather than a note for a subsequent reader. If Francis Douce was the only audience for his notes, he had no need to justify his conclusions.

On the other hand, however, it is sometimes very clear to see why Douce has changed his opinion about items in his collection. A 1591 translation of Le Chevalier délibéré into Spanish, held at Bodleian, Douce M 664, contains a series of notes that are apparent by a change of aspect in Douce’s handwriting which are clearly the product of his subsequent reading, as he adds new opinions and sources in the light of what he has discovered (fig 6). The question is the original author of the French text, which is now attributed to Olivier de la Marche. Douce’s first note on the matter, visible halfway down his first page of bibliographical annotations, accepts this as a possibility.

Fig 6. Francis Douce’s notes in El cavallero determinado, trans by H. de Acuña, Bodleian, Douce M 664: a) first flyleaf recto; b) second flyleaf recto. Images: by kind permission of The Bodleian Libraries, University of Oxford.

In a subsequent note, with a different aspect, Douce adds ‘and “Collection universelle des Memoires particuliers relatifs a l’histoire de France. Tom. 8. p. vii.”’ together with details of a translation in Spanish printed in Antwerp. Another subsequent note, again signalled by a change in aspect, reads ‘There is sufficient proof that this work was written by Olivier de la Marche, in the description of N.^o 2862 of the Duc de la Valliere’s catalogue’. This note may or may not have been produced at the same time as the phrase that completes the line and runs onto the following one, ‘See also Goujet Bibliotheque Françoise Tom. IX. p. 375.’, which in turn is followed by another note in different ink, ‘The cuts I think were etched by Van der Borcht’. Over the space of at least four different annotative moments, Douce shows the results of his reading and reflection on the subject. He provides a wealth of detail that not only allows us to cross-check his conclusions but also see what he was reading and the sort of influence that it had upon him.

The first of the works cited in note by Douce here is a major work of scholarship from the sixteenth-century. Produced initially and separately by two scholars, François Grudé de La Croix du Maine and Antoine du Verdier in 1584 and 1585, the two works were frequently published in a combined edition, augmented and updated in the eighteenth century by Rigoley de Juvigny.Footnote ³⁴ Francis Douce himself owned all six volumes of an edition published in 1772–3, now catalogued as Bodleian, Douce I 278–83, so it seems that his first note referred to reading in his personal collection.Footnote ³⁵ His annotations frequently make reference to this work, which he also calls Juvigny, and it seems to have been a frequent initial point of consultation when considering the history of a work. The second annotation refers to an extensive series, the Collection universelle des mémoires particuliers relatifs à l’histoire de France, published between 1785 and 1791, and more especially to volume 8 which contained an edition of the Mémoires of Olivier de la Marche. This edition does not feature in Douce’s collection as that collection is conserved in the Bodleian, although we can suppose that Douce might have had easy access to it, either in libraries or reading rooms or through the collections of his associates. The publication date is a little after that of the combined La Croix du Maine/Verdier/Juvigny volume, which may explain the order of the annotations, though it is as plausible to assume that the move from Douce’s own collection to sources outside that collection may explain the order of the notes. The final reference, to the Duc de la Valliere’s catalogue, points to the 1783 catalogue of the sale of the duke’s books, which appears in Douce’s collection at BB 352–354. The reference to La Marche at 2862,Footnote ³⁶ is a discussion of a manuscript of the text, of which Louis César duc de Valllière had two, along with an early printed edition in French and a translation into Spanish. In the course of this discussion, the question of the author of the work is discussed in forceful terms, pointing out that, although the work has often been attributed to George Chastellain, the use of La Marche’s personal motto is proof that the text is by him. Incidentally, this analysis is shared by the modern editor of the work, who, while noting that one manuscript attributes it to Chastellain, recalls that six other manuscripts present La Marche’s name and that, as a consequence, ‘[t]here is no real dispute concerning the authorship of the poem’.Footnote ³⁷ In the la Vallière catalogue, this argument is followed by a brief biographical sketch of La Marche to provide some context to the work.Footnote ³⁸ Douce’s reference to this earlier work after that to Juvigny indicates that he was recording his impressions as he encountered the scholarship in his reading. Given that this reading does not follow the chronological dates of publication, we cannot presume that Douce was reading material as it came out. The accumulation of references with different dates does, however, show that he was not content with a single opinion on a work and was ready to revise his view in the light of fresh reading.

Douce’s reference to Van der Borcht in this connection is unrelated to his reference to Goujet, as is indicated by the change in ink between the two comments. Goujet, published 1745, presents an equally categorical case for La Marche’s authorship, based this time not on the comments made by the author within the text in question, but rather on the same author’s claims in another work, his treatise on duels, to have written Le Chevalier délibéré.Footnote ³⁹ Goujet’s account of the work is longer than Douce’s single page reference suggests and focuses on the content and development of the poem as much as on its author. As for Douce’s attribution of the illustrations of the 1591 edition to Van der Borcht, the collector does not give his reason for writing this. It may be motivated either by the style of the illustrations themselves or by the knowledge that the man to whom he attributes them worked for the printer of that copy. Whatever Douce’s motivation, the attribution is now shared by others, including a copy for sale in March 2024.Footnote ⁴⁰ This example shows something that is more generally the case throughout Douce’s collection: he seems willing to make what looks like unsubstantiated speculation on the origin of visual material that he owned, and this speculation nevertheless often seems well informed.

In the series of notes on El cavallero determinado we see some evidence of Douce’s use of bibliographical literature. It is to be noted that Douce provides himself and any subsequent reader with references that are precise enough for us to follow his reading. Such traces are found throughout his annotations to books in French. Many references to the same books recur throughout his collection, allowing us to see that Douce was returning to the same works on French bibliography that were the manuals most frequently used in this area. As in the case of his investigations of Le Chevalier délibéré, many of the sources cited by Douce are from the eighteenth century, though he does also refer to some more recent works, including the Bibliographical, Antiquarian and Picturesque Tour in France and Germany, produced by his correspondent Dibdin.Footnote ⁴¹ The three volumes of this work, published in 1821, are in Douce’s collection,Footnote ⁴² in an edition printed specifically for members of the bibliophile association, The Roxburghe Club, but without personal dedication to Douce. Among other nineteenth-century works cited by Douce is one that he calls ‘Bouterwek’, presumably Friedrich Bouterwek’s Geschichte der neuern Poesie und Beredsamkeit, published in Göttingen in twelve volumes, 1801–19.Footnote ⁴³ Of that work, Douce refers particularly to the history of Spanish literature, which was published separately in French, Spanish and English, and an abbreviation in a book from his collection, a French translation of the Spanish Celestina, published in 1542 and shelved at Bodleian, Douce C 394, suggests that he was using the English translation, published in 1823.Footnote ⁴⁴ Commenting on the authorship of the text, Douce comments that ‘Bouterwek in his hist. of Span. Literature p. 132 states that this romance was begun by Rodrigo de Cota & continued by De Rojas’. The use of the abbreviation of the English title is what suggests that Douce is referring to an English translation. Indeed, the reference does correspond to the 1823 English edition, but not to the page where the argument cited is to be found (which is p 134), but rather to the one where the beginning of the discussion of Celestina occurs. Once more, this is not a volume that is part of Douce’s collection in the Bodleian, but it seems to have been one to which he had access.

Some of the bibliographical notes in Douce’s copies show the range of bibliographical literature to which he referred. A frequent reference, as we saw above in the discussion of El caballero determinado, is to ‘Goujet’, as seen on several occasions in a series of notes composed in five phases on the flyleaf of a copy of Les Renards traversans (fig 7).Footnote ⁴⁵

Fig 7. Notes in Les Renards traversans, Bodleian, Douce B. Subt. 248 : a) first flyleaf insert; b) first flyleaf recto. Images: by kind permission of The Bodleian Libraries, University of Oxford.

The notes read:

Like Bouterwek’s work on Spanish and Portuguese literature, Goujet’s eighteen-volume work on French bibliography does not form part of the Douce collection. The fact that references to it occur in two different phases of the annotation (volume nine in phase two and volumes ten and eleven in phase four) suggests that Douce did not read it all at once. Given that the references to the volumes appear in sequence, it is possible that Douce was reading through the series in order, although the frequency with which he refers to Goujet in the volumes of his extensive collection makes it unlikely that this was a single reading of Goujet. It is far more plausible that his frequent references reflect several consultations of the work. In the case of the annotations to volume Les Renards traversans, the references to the different works do not seem to contribute to a developing scholarship on a particular question, as they did with the authorship of Le Chevalier délibéré. Instead Douce seems to have used this copy as a repository for his notes on the motif of the dance of the blind. This was a practice that was noted in my earlier article concerning the way that images are stored in related volumes in the same collection, to facilitate retrieval. The book is here used as a repository and an aide mémoire for a collector seeking to gather notes related to the subject.

It is interesting that, in doing this, Douce explicitly includes material from manuscript sources, in this case a manuscript not in his own collection, alongside bibliographical works. The reference to Senebier directs to an account of the work also contained within a manuscript (this time number 182 in the Geneva collection) and a discussion of its text and authorship, which Jean Senebier attributes to Pierre Michault.Footnote ⁴⁶ There are other cases in Douce’s collection where his notes contain references to manuscript material to expand or complement his understanding of the text. A very extensive example occurs at Bodleian, Douce 224, Les vertueux nobles et glorieux foiz du tresvaillant et renommé chevalier Tristan, published in Rouen by Jean Le Bourgeois in 1489.

This copy is unusual but not unique in Douce’s collection for containing annotations at several points in the book, and not just in bibliographical notes in the preliminary pages. Notes in the preliminary pages do exist in Bodleian, Douce 224, but Douce has also introduced several other notes at different points, many pasted into the book or inserted as loose leaves (fig 8). The volume also contains a number of notes by an earlier reader in a French sixteenth-century hand and underlining and other marks of reading, some of which look like they may have been made by this reader while others may have been made by Douce himself. What is more, one of the notes on the flyleaf complicates this question by signalling that – even though Douce’s handwriting is quite distinctive – it is not always easy to tell whether he produced a particular annotation. The flyleaf contains a lengthy quotation in French in a hand that looks quite different from Douce’s. An accompanying note, in Douce’s hand, informs the reader that:

Fig 8. Notes in Bodleian, Douce 224, Les vertueux nobles et glorieux foiz du tresvaillant et renommé chevalier Tristan. Two notes of a manuscript variant and a reference to La Ravalière’s Histoire des révolutions de la langue françoise: a) letter part 1; b) letter part 2; c) first flyleaf recto. Images: by kind permission of The Bodleian Libraries, University of Oxford.

This is further evidence of Douce comparing his printed texts to manuscripts, but it also suggests that the hand which transcribes the manuscript that Douce has sold on is also that of Francis Douce, despite its marked difference from his usual handwriting. The note in Douce’s hand in English tells us that the French material is a transcription from a manuscript that Douce used to own. It is probable that the transcriber is Douce himself, anxious to preserve a record of similar content even where he did not want to keep the physical copy of the manuscript (for which, in any case, he seems to have owned a duplicate manuscript). The divergence between the hand used to make the transcription and that which supplies its gloss could be presumed to be a consequence of the portion of Douce’s education spent in France and where he had maybe acquired a French handwriting that he used when he wrote that language. Some other volumes in Douce’s collection contain annotations where quotations have a markedly different aspect from his normal hand, even where the quotation is integrated into the annotation. However, this is not always the case, and the very same copy shows a contrary example. On another note inserted into the book, Douce quotes page 169 (but actually p 168) of La Ravalière’s Histoire des révolutions de la langue françoise, depuis Charlemagne jusqu’à saint Louis, dating the work to 1190.Footnote ⁴⁷ This note is in a version of Douce’s normal handwriting, albeit quite a tidy one. Another interesting feature of this footnote is that the work cited by Douce does not name Pierre-Alexandre, évêque de La Ravlière, as its author. He is, however, given as its presumed author in the catalogue of the Bibliothèque nationale de France, with the attribution being identified as that of Barbier.Footnote ⁴⁸ Douce’s attribution of the work to the same person indicates that he has a familiarity with bibliographical scholarship that goes beyond what is visible on the printed page. It is probable that his source for the attribution is also Barbier: a note that he made in Bodleian, Douce D 15, a copy of Les controversses des sexes masculin et feménin, remarks that Barbier has attributed that work to Gratien Dupont, but when quoting the Histoire des révolutions de la langue françoise, Douce does not cite Barbier. This again raises questions as to who the intended audience of Douce’s annotation is. Douce does not set out where he has sourced his attribution and so it is not apparent that the work cited is actually anonymous. This creates a potential difficulty for any reader attempting to follow his traces who does not know as much as Douce does. Even a reader who searched for the title rather than the author would have difficulties because Douce’s reference is sufficient to allow an approximate identification, but it is not precise. Presuming that Douce is reading the edition published in 1742 (which is the only one listed in the catalogue of the Bibliothèque nationale), his page reference is not correct: the text cited appears on that before the one to which he refers.

This is not the only approximate reference that we have noted in Douce’s annotations. If these annotations were intended as personal notes for the collector alone, they are probably sufficient to indicate approximately where a reference can be found. They do not, however, meet the standard of exact scholarly reference and might indicate that the annotations are intended for Douce alone. Douce’s obituary points out that the collector did not seek to publish his work even to the extent that he sealed his papers after his death, and this might indicate that his notes were intended for personal consumption only.Footnote ⁴⁹ This is not, however, entirely true, since Douce did publish a work on Holbein’s woodcut illustrations to the Dance of Death, and the annotations in his collected volumes show a level of engagement between him and other scholars, indicating that his scholarship was not an entirely private affair.Footnote ⁵⁰ Notes in his books, as for instance in R. Subt. 13, a copy of Giron le courtois printed by Vérard in 1504, show that Douce’s opinion was sought as a valuer and expert on rare books. In a note at the front of this copy, Douce tells how an associate of his, Henry Shaw, had brought a man named Skelton to him, with a manuscript containing the coats of arms of the knights of the Round Table for valuation, and that, on the same evening after he had said that it was of little value, another man, named Carlisle (presumably Nicholas Carlisle, secretary of the London Society of Antiquaries, with whom Douce also corresponded), sent him the same manuscript for valuation. The anecdote illustrates that Douce’s opinion was appreciated among his contemporaries. His engagement went beyond providing opinions for those who sought them and extended to lending out the books in his collection. This suggests that the annotations in them might not have been considered entirely personal, since Douce’s lending was systematic to the extent that he had a mechanism for recording it. He kept a notebook, now Bodleian, ms Douce E 75, in his papers, detailing the books that he had lent to others and in some cases his frustrations that these had not been returned. Any annotations in those books, therefore, might have been written with the wider audience of borrowers in mind, something that makes the approximate nature of these references harder to understand.

One possibility is that Douce deliberately produced slightly inaccurate references as a safeguard against copying by scholars who had not examined the sources themselves. This practice is not unknown, but there is no direct evidence that this was a deliberate policy on Douce’s part, as there would be, for instance, if he had drawn attention to the fact of an unexamined use of his work. There is, however, clear evidence that not only was Douce’s opinion valued by his contemporaries, but they expected his work to be accurate. We see this in comments that we find in books within the collection showing evidence that Douce was regarded as an authority on early manuscript and printed material and by things written elsewhere in the community to which he belonged. For instance, over a quarter of a century after Douce’s death, another of his correspondents, William Thoms, published an answer to a query in Notes and Queries, of which Thoms was then the editor, that draws on Douce’s work

Thoms’s explicit frustration with the deficiencies of Douce’s scholarship and his statement that Douce did not usually fail to provide references tells us that Francis Douce continued to be thought of as an authoritative scholar into the second half of the nineteenth century, even if his scholarship was acknowledged to have some deficiencies. Frederic Madden too has an enduring reputation as an accurate and exactly scholar, but his marginal annotations also do not exactly reproduce the text that he claimed to transcribe. Our survey of Douce’s engagement with bibliographical scholarship demonstrates that he read widely in relevant bibliographical literature, encompassing both established authorities and newer publications, and that he was able to draw plausible conclusions based on his reading and on his knowledge of texts and illustrations. Notwithstanding some inaccuracies in his references, and some erroneous conclusions that have not stood the test of time, his scholarship seems to be well informed and engaged. There is, it is true, some unreferenced notes that may fall into the category of what Meg Twycross considers educated but misplaced speculation, only some of which has been negated by subsequent scholarship. Most striking, however, is the evidence that Douce saw his work as an ongoing and dynamic process, even though his collection was not catalogued to facilitate this. Despite the absence of a catalogue, shelfmarks or other signs of organisation, Douce was able to locate volumes within the collection, to return to them and to record revised evaluations of the material. The absence of a catalogue is counter to the best practices of amateur scholarship and to the professional standards of the British Museum, whose practices Douce followed in other respects. This did not prevent him from intensive engagement with the items in the collection, producing notes in multiple phases that allow us to examine the practices of a scholar with an extensive collection in the first half of the nineteenth century.