On August 24th at 18.30, at the Ferme du Manoir, there was a talk by...


Anthony Pincott,

Secretary of the Bookplate Society, London


Find out about the Bookplate Society... go to http://www.bookplatesociety.org/



Anthony Pincott has been collecting British bookplates for over 35 years. He is building a comprehensive database that will grow to perhaps 100,000 entries. It will cover British ex-libris down to the year 1900 and American bookplates to about 1800. In this talk he discusses the aims and scope of his project and examines the difficulties and eventual benefits.


-   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -  -   -   -   -   -   -   -   -





My chief interest is in ex-libris of the British Isles prior to about 1820. For me, part of the attraction in collecting and studying bookplates lies in coming across ex-libris that I have not seen before. In any collection of reasonable size there will be a few bookplates of special interest, and occasionally one finds material that is quite scarce or even unique. Long ago I accepted that I could never hope to possess everything – there are just too many ex-libris, some of which will never be found outside the institutional collections in which they are now held. So I am content to expand my collection where I can and, for the rest - rather like a bird-watcher who records sightings -, to become a notional or virtual collector, and be someone who gathers information about interesting and rare specimens and where to find them.


Librarians, booksellers and bibliographers are today far more aware of the need to record the provenance of books and to know something of their past owners. David Pearson’s excellent book entitled Provenance Research in Book History, which I recommend to you, republished by the British Library in 1998, has a chapter on bookplates and highlights the need for libraries to record  provenance information as they update their book catalogues. Some libraries are beginning to do this, but at present their databases are not easily searched – for example there are 40,000 references to bookplates in the book catalogue of the Bodleian Library, Oxford, but it would take weeks to extract useful information about these ex-libris. I know of a few British and North American institutions which are cataloguing their bookplate collections, including Exeter Public Library, the John Johnson Collection (which is part of the Bodleian), and both the Societies of Antiquaries at London and Newcastle Upon Tyne. The University of British Columbia also has a bookplate database project in hand, but like the others it is concerned just with recording material in its own collection. These databases would be impossible to unify, because each has a different approach to the information recorded. None is as complete as I would wish to see. So a separate, unified approach has been needed. 


Some years ago I decided that I wanted to undertake a lifetime’s project to build a comprehensive database listing all British bookplates made prior to 1900. This was not a rigid date, because the database ought also to include all the post-1900 work of artists and engravers who were actively making ex-libris at my 1900 cut-off date. American examples of the 18th century, up to about 1820, also deserved inclusion.


My purpose in compiling such a database was to:

·         Include as many as possible of the bookplates which seem to be unrecorded elsewhere;

·         Get together in one place the references to information published in bookplate literature plus biographical and other information which earlier collectors may have gathered in their individual collections, but which has never been published; and

·         Create the ability to search through this database in a variety of ways that help collectors, bibliographers, heralds, family historians, graphic artists, and any other potential users.

The golden days of collecting British plates were more than a century ago. The Ex Libris Society had been born in 1890 and began publishing its Ex Libris Journal. Soon there appeared two important classic reference works: Walter Hamilton’s extensive record of Dated Book-Plates in 1895 and Henry Fincham’s valuable directory of Artists and Engravers of British and American Book Plates two years later (SLIDE 1).

That same year 1897 saw the death of Sir Augustus Wollaston Franks (Geneva 1826 – London 1897), pre-eminent among bookplate collectors (SLIDE 2).


He bequeathed 35,000 British bookplates to the British Museum, where he had worked all his life. Franks was born near here, in Geneva, in 1826, and he also gathered a great number of old French and German ex-libris, including many early examples. Thanks to the great efforts of Edward Gambier Howe over a period of 5 years, the British items were sorted into alphabetical order by family name and pasted into large albums bound in leather. A slip of paper was written out for each bookplate, and these slips of paper were the basis for a 3-volume printed catalogue (SLIDE 3) which appeared in 1903-4. Collectors received this with great approval, and at the time Franks’s collection was though to be fairly comprehensive.

However, just a couple of years later, in 1906, the marvellous collection of Julian Marshall was auctioned. It was evident from the Sotheby’s catalogue, also compiled by Gambier Howe, that a great number of uncommon and fine bookplates had escaped Sir Wollaston’s net. Marshall’s treasures were dispersed worldwide. Some went directly into the collection of W.E. Baillie, an American whose bookplates are now in the Metropolitan Museum of Art in New York. Others found their way directly, or via dealers, into the hands of George Viner, Franks’s foremost successor, who died in 1952. Amongst his many other gifts, Viner donated to the British Museum’s Department of Prints and Drawings some 9,000 ex-libris which were “Not in Franks”.


In the 1970’s, when I was a beginner-collector, I started to catalogue the Viner Collection. In the days when the British Library was still in the British Museum, long before the move to the new building at St Pancras, you would have seen me, some weekday evenings and Saturday mornings, in the typing room, with a portable typewriter and a Viner album in front of me. But I made slow progress. My list never got beyond beginning the letter D!  The task was too great to tackle in that way. I gave up.


Thirty years later, computers have changed everything. As you all know, they make it possible to enter vast quantities of data. The record for each ex-libris can be organised into sub-sections, or “fields”. Addition, modification and output of the information are easy, but a computer’s most valuable power is to search through data and to sort it by words or by fields. Thus it became possible for me to make a fresh start on the task of creating a union catalogue of bookplates. I began this by transcribing the Franks Catalogue into an Excel spreadsheet, which is a grid of boxes, or cells, in rows and columns.


I did consider scanning the three volumes using Optical Character Recognition software, but unfortunately I quickly found that the typeface and layout of the text made this impossible. So every entry had to be typed in, with the Franks number of each bookplate at the start of each row down the page, and with the data fields across the columns from left to right. I decided not to use database software because in Excel:

·         I could insert new lines and columns with ease;

·         I could use macros for speeding up repetitive tasks like inserting lines;

·         I could see on screen the entries above and below the one I was working on. This has been very helpful when entering plates which have only slight differences; and

·         I could avoid having to commit myself, at the start of this project, to choosing a specific database software. I expect that by the time I’m ready to make the data available, technology will have moved on, and improved databases will be available. Of course, it will be troublesome, one day, to migrate the data from Excel into a new database, but I don’t view this as an overwhelming problem.


Once I had transcribed the Franks catalogue into three Excel files I was already far ahead of anything attempted previously. However, merely copying out the Franks Catalogue would not have met today’s needs because the printed catalogue does not contain enough information. In it we find (SLIDE 4) the following:


Text Box: ·         The Franks number;
·         The family name;
·         The inscription (not always complete);
·         Any date that appears on the bookplate;
·         The engraver’s inscription;
·         The style of the bookplate; and
·        Brief  biographical details, but these were only mentioned by Gambier Howe when they help us to know the approximate date of the plate ie when the owner acquired a title, who he married or when he died.










A modern database needs to split up such data if it is to be fully searchable. Dates of birth, marriage, succession, and death all need separate fields. The location of the owner’s library needs to be recorded by house name, town and county or country. Heraldic quarterings and impalements must all be searchable by family name. This results in a more than two dozen fields. And there is still much further information that each database record needs to include (SLIDE 5):

Text Box: ·         Reference to biographical directories such as the Oxford Dictionary of National Biography and Burke’s Landed Gentry;
·         How similar ex-libris may be distinguished one from the other; 
·         Dimensions of the engraved area;
·         Method of production (perhaps using FISAE technical symbols);
·         References to text and illustrations in bookplate literature;
·         Mottoes;
·         Queries; and
·        Which collections hold examples of this particular ex-libris.











This all results in about 40 fields in total, and one may find that others could be useful: for example I haven’t included a field for the owner’s occupation. If the nature of these fields is of interest to you, we can discuss them further during questions at the end.


You will see (SLIDE 6) that compiling a complete catalogue is a multi-stage process:

Text Box: 1. TRANSCRIBE the 3 volume printed catalogue
2. Look at MICROFILM for mottoes, differences between plates etc.
3. Examine the ORIGINAL ex-libris for dimensions, manuscript notes etc




Transcription of the printed catalogue was only the first step because Gambier Howe had neither the time nor the resources to be able to give anything more than brief biographical and heraldic details. I own a copy of the full microfilm of the Franks Collection, and the second step is for me to work through the images, checking the entries and recording such information as differences between similar bookplates, the exact inscription, mottoes, and points worth mentioning. For example, one has to look out for any design that is of a similar pattern to another bookplate elsewhere in the collection, because the two may be by the same engraver. At some future time the third phase will be to go through each Franks album, looking at the originals, and noting down sizes of ex-libris. Here, by the way, I mean the engraved area, not the size of paper or the plate mark, which I don’t believe to be worth recording.


Beyond the printed inscription, a few bookplates may carry valuable additional notes in manuscript, but not on every copy - this is why it is helpful to gather information from many collections. For example, my copy of Charles Thompson’s Chippendale bookplate of about 1760 (SLIDE 7)



carries a manuscript note at the top that its owner was “at Lisbon during the earthquake”. This was the famous catastrophe on All Saints Day, 1st November 1755, which claimed 60,000 lives and which had wide impact across Europe through the writings of Voltaire in Candide. A further note below states that Thompson was “buried in Sherwood Forest”. So this one print of the bookplate is the key to determining who its owner was. Another example of a useful manuscript addition is the bookpile of William Jackson, where you will see  (SLIDE 8) that my copy of this plate bears both a place and date: “Durham 1747”.  It is quite possible that these manuscript additions are unique to my particular copies of these ex-libris, and so you can understand why this extra evidence of ownership needs to be entered into a central database.


Just a few British collectors have in the past taken the trouble to search for biographical information. Their notes have been scattered into public and private collections, but it would be most worthwhile to gather such data. Look for example at the mid-18th century bookplate of James Wickins. My copy came from the collection of Tom Owen, and you see here (SLIDE 9)


his handwritten note which tells us that this ex-libris came from a book in which the owner wrote his address as The Close in Salisbury, Hampshire. So Tom Owen was able to identify him as the same person whose wall monument in Salisbury Cathedral carries the inscription: “James Wickins Esqr. made his terrestrial sojourn eighty-two years and quitted it in the hope of a better world on the 19th day of May 1827. Anne his widow died 20th January 1850 aged 97 years”. For most British bookplates we lack details of the owner. Sometimes the International Genealogical Index, available online, is of help, but it’s slow work to find people with certainty, and impossible in the case of common names like John Smith. One day there will be an online index of all memorial inscriptions in British churches, but who will have the time to check and cross-reference them to a bookplate database?


To expand on what I mean about differences between plates, I should explain that copper plates were not just engraved and forgotten. Very often one finds that the inscription and heraldry were changed several times as errors by the engraver were corrected, or the owner of the bookplate acquired new titles or moved address or got married, or the copper plate might have been re-used by a son or other relative. As the copper plate got worn, it might have its life extended by being re-engraved. Let’s look at these two Scottish bookplates for Bannatyne. (SLIDE 10). 



At first sight they appear to be exactly the same, but look again. Careful examination shows many places where the righthand ex-libris has been re-engraved. The problem with the Franks Catalogue is that it states only that the second ex-libris is “the previous plate reworked”, which is thoroughly unhelpful because most users of the catalogue do not have the Franks originals or the microfilm in front of them.


Look also at my next slide. (SLIDE 11). Both ex-libris for James Yonge carry the same inscription. The heraldry is identical. The layout is similar, but they were engraved on different copper plates, at different times. Gambier Howe mentions only that the second one is “a different plate”, so users of the Franks Catalogue have been unable to know which is which. Gambier Howe also fails to record the signature of the engraver, a Mr E Stammers of 99 Strand, London. In other cases the dimensions of the engraved area may be different, or there is some other obvious distinction, but often a careful explanation of the differences is needed. This would, of course, be made far easier if the database were also to include images, and I’ll come back to this aspect.


Having transcribed Gambier Howe’s brief entries, my next step has been to refer to a very special copy of the Franks Catalogue. It had originally belonged to Dr John Pearson, whose huge collection of British bookplates was purchased by the American, Warren Loewenhaupt, prior to Pearson’s death in 1952. Loewenhaupt was an alumnus of Yale University, and the Pearson collection is now in the basement of the Seeley-Mudd library at Yale, together with the collections of Irene Pace-Andrews, Baron Gunnar Trönnberg and Harry Scammell. I fear that these collections receive very few visitors. However, Pearson’s special interleaved copy of the Franks Catalogue did not go to Yale. It remains in the Print Room of the British Museum, where I was permitted to photocopy it. Here is a typical page (SLIDE 12). On one side are the Franks entries, while on the other, in Pearson’s tiny handwriting, are his additions. He found these extra plates in his own collection, the Marshall catalogue, Allen’s American Book-Plates, Oliver’s West Indian Bookplates, the Ex Libris Journal, in the collections of George Viner and Horace Jones, and from several other sources.


Reading Pearson’s handwriting was not easy. I started, strangely, with volume H to R, then S to Z and had commenced A to G when a disaster intervened. My laptop computer was stolen from me at Brussels airport. At the time I was madly angry, both with the thief and with myself for not being 100% reliably) backed up. It still makes me cross to think of it,




and I remain ready to inflict barbaric tortures on the culprit, if ever I catch him!  It taught me a very hard lesson about the essential need for regular backups when undertaking a task of this nature. The section H to R has now been completed, and I have got as far as C in the file A to G, so I’m not yet halfway through entering the Pearson additions. His notes were far from complete, they were sometimes difficult to decipher, and the inscriptions were not necessarily accurate, so I often had to write something in the Queries field, to be resolved at a later date. It will be necessary to find and examine originals of all the ex-libris he recorded, but at least I know that these exist somewhere.


As regards the Viner Collection that I started so long ago to catalogue in the British Library’s reading room, I have transcribed my old notes into a separate Excel file and have added to them by taking my computer to the Print Room at the British Museum. So far, I have catalogued only about 30% of the Viner Collection. The Print Room opens only on weekdays, and then only for 4¾ hours each day, which means that fewer than 100 ex-libris can be recorded per day. It’s a miserably slow process, and I wonder if a better approach may be to arrange for the digital photography of the whole collection. Leaving aside the question of cost, it would then be possible to do the cataloguing at home. There would then be the long job of transferring each entry into the right place in the Franks sequence.


Here, then (SLIDE 13) is my overall methodology. Even with the cataloguing of the Franks and Viner Collections and the inclusion of bookplates known to Dr Pearson, there is so much more to be done before my database could truly be called a union catalogue. The main reference books in bookplate literature need to be checked, and many collections remain to be visited, notably those in Liverpool, Oxford, the National Libraries of Scotland and of Wales, two or three in London, plus various private collections.















































Turning to the American examples, Allen’s American Book-Plates, published in 1894 and never updated, probably lists fewer than one-third of all American bookplates pre-1820. The Henry C. Eno auction catalogue of 1916 contains much useful information and is almost the American equivalent of the Julian Marshall sale catalogue. One could spend months and months going through the collections at Yale, the American Antiquarian Association, the New York Public Library, the Metropolitan Museum of Art, plus those in the hands of private collectors such as Lew Jaffé and Bill Butler. 


I think I shall need some help!  It may well be preferable to make a first edition of my union catalogue generally available, even if it is far from complete. It will then be possible for other people to contribute some of the many additions and corrections needed. Perhaps they will be able to do this by entering data on a website set up just for this purpose.


How will this union catalogue be of value, whether consulted on CD-ROM or from a webpage? On the next slide (SLIDE 14) are some of its possible uses:

Text Box:  
For collectors it will be a means to check that a print is recorded. They can then draw attention to any omissions;
It will provide biographical and heraldic information of the owner;
The database will help with identifying anonymous plates and in distinguishing between similar plates;
Sorting the database by the DATE field will generate an up-to-date version of Walter Hamilton’s Dated Bookplates;
Sorting the database by the ARTIST field will generate the equivalent of Henry Fincham’s directory of artists and engravers;
Sorting the database by LOCATION will create a geographical index, for example by county or by town; 
Family historians will be able to search for the ex-libris of their ancestors; and
Each individual or institution owning a bookplate collection will be able to mark up the database to show which items they possess, so this will make it easy to catalogue each collection.





















One important question has been how to number each record in the database. Early in this project I decided to keep the Franks numbers and to insert extra entries by adding decimal places. To understand this, let’s look at the next slide (SLIDE 15) which takes as its example the family name Welch. On the left are the existing Franks entries and on the righthand side are some newly included plates. In Gambier Howe’s system the Lords come first, and so Lord Welch is given the decimal number F.31250.1, even though F.31250 is for the surname Welby.  We have no first name for Mr Welch, so he is put in the M’s before Mrs Welch. Then there are four new records, with a series of decimals. You will see that S. Welch and Simon Welch are not together, even though both come from the same copper and may be the same person. It’s one of the many drawbacks to this method.


F.21250  (Welby)
                                                F.31250.1  Lord Welch of Nella
F.31251  Welch (Armorial Spade Shield)
F.31252  Welch (Armorial)
F.31253  A. D. Welch
F.31254  Arthur D. Welch
                                                F.31254.1  B. T. Welch
F.31255  George Asser White Welch Esqr.
                                                F.31255.1  Mr. Welch, Lincoln’s Inn
F.31256  Mrs. Welch
F.31257  R. H. Welch
                                                F.31257.1  Richard Welch Esq. Oxford
                                                F.31257.2  S. Welch
                                                F.31257.3  Samuel Welch
                                                F.31257.4  Simon Welch  (F.31257.2 reworked)
F.31258  William Welch
F.31259  Colonel Welchman C.B.
F.32763  B. W.
                                                F.32763.1  B. J. W.































So the problems are these:



There are some major gaps in the Franks Collection. In particular, Sir Wollaston was not much interested in printed labels, however early, and at the end of his collection the series of institutional and circulating library bookplates is woefully incomplete, so the many new additions therefore have to be numbered using two places of decimals.


However, I believe the advantages outweigh the disadvantages, because this method of numbering is flexible and enables me to keep the Franks number as the main key to finding an entry. It’s better than using as a key the accession number generated by a database program. Particularly important is knowing whether or not I have already recorded a plate that is “Not in Franks”.


This idea of gathering images to include in the database is greatly appealing. I purchased a scanner about two years ago and I got a lot of practice using it during the first few months of 2005. I was helping the auctioneers Bonhams of New Bond Street, London to sort out and catalogue the 23,000 bookplates in the collection of the late John Simpson, and with the agreement of Bonhams and John’s sisters I was able to scan some 10,000 bookplates. These images will be immensely useful for future publications of The Bookplate Society. It was a huge task, and I have yet to finish it, because as the day of the auction in May 2005 got closer and closer I found myself running out of time. So I had to scan six or eight bookplates to a page, and now I need to go back through each image file and re-save it as separate images.


Working with an image on screen enables you to look at the fine details of the engraving far more clearly than with a magnifying glass. It’s also possible to bring different states of an ex-libris onto the screen for careful comparison. The result is that I have become interested in collecting not only information for the fields in my union catalogue but also in acquiring images of a high quality sufficient for later publication. Both the Society of Antiquaries and the Royal College of Surgeons have kindly given me permission to scan their bookplate collections one day.


You may suggest that I should keep two images of each bookplate – firstly in a high resolution TIFF format needed for printing and secondly a low resolution JPEG file for inclusion in the database for making available online. However, working with images is VERY time-consuming, and the files occupy a great deal of computer memory. If I tried to gather 100,000 images in the two formats I would be overwhelmed by the task and would also very quickly run out of space on my computer.



When will my database be finished? The answer is that I don’t yet know. I’ll need to issue a first edition CD-ROM in order to start getting help from other people, but doing this too early will be greatly counter-productive because I would lose too much time answering other people’s queries. It’s the work of a lifetime, and will always be incomplete.  After the death of Sir Augustus Wollaston Franks a bookplate was engraved by Sherborn to record bequests to the Society of Antiquaries (SLIDE 16). We see here Franks’ motto: NUNC MIHI – now to me – MOX ALIIS – soon to others, and this is how I feel about my union catalogue. Someone else will carry on the task when I’m no longer able to do it, but I trust I’ll have many years yet, assuming my wife does not murder me for spending too much time on bookplates!


© 2006 Anthony Pincott     All Rights Reserved