[x-posted on the 18th-century BookTracker Facebook discussion page]
As I was working today on BookTracker, I realized that there are quite a few catalogs etc. in Googlebooks, and I was wondering whether there might be a special section of the BookTracker devoted to such material. The rationale would be that there would be more information there than elsewhere about difficult to find titles. This is a type of information that to my knowledge is not getting aggregated elsewhere, but could be assembled pretty painlessly by the folks here, if the works contained in a catalog could be linked to a full catalog record. Thoughts?
DM
Early Modern Online Bibliography 
August 16, 2010 at 1:42 am |
Ben can reply to this question much better than I can do, but I would like to clarify the type of catalogs to which you are referring. Are you speaking of booksellers’ catalogs? auction records? private library holdings? or something else?
Some of these sources often lack the detailed information that could positiviely identify what edition of a particular title is being referenced. That said, these sources can be a treasure trove of information. See, as an example, Norbert Schurer’s “Four Catalogues of the Lowndes Circulating Library, 1755-66.” PBSA 101:3 (September 2007).
If I am understanding you correctly, Dave, linking each title found in a given catalog would be fairly labor intensive–especially if the catalog details are fuzzy, incomplete, or the like.
LikeLike
August 16, 2010 at 2:17 am |
Well, bookseller’s catalogs in Google books were one of the things I was thinking about, but it seems to me that things like advertisements and other kinds of notices might get collected and tagged somehow as part of a bibliographic database. I understand it’s labor-intensive, but I was wondering whether such a collective tagging operation would turn up useful information. (This does seem like a project that machine reading could do very well, incidentally.) But it just seems like Book Tracker has the potential to be not just a text-locator for our won purposes, but a kind of historical repository of earlier finding aids.
LikeLike
August 16, 2010 at 2:33 am |
Good ideas in many ways, and you are aware of how I view Google’s usefulness as a finding aid. Jim Tierney has been working for years indexing the contents of 18th-century periodicals and is still hard at work on this database. Corky McGinness (Royal Holloway) spent 25 years or more identifying all the advertisements and notices of music and musical performances in 18th-century newspapers–and did so before the advent of digital tools. Her project is awaiting transformation into an online searchable database.
Machine reading in many ways seems ideal for such a task, but I would add the caveat that such projects beenfit immensely from human expertise in behind-the-scenes tagging and coding. Part of Tierney’s trouble in getting financing for his project is that too many think that if pages are searchable, then this labor-intensive work is not necessary–but it is.
LikeLike
August 16, 2010 at 10:40 am |
I think the Tierney and McGinness projects you describe are incredibly heroic and valuable, but I wonder if a more “Web 2.0” approach to collaboration would help things happen faster. One of the thoughts I had was that if BookTracker took off, it might be helpful to organize usergroups that would start assembling and tagging texts with particular interest uniting them: I was immediately thinking of a book history group interested in things like catalogs, and maybe groups like a Scottish or Empire studies groups (doing various colonial regions) that could go in their own direction. Best of all, texts could end up with multiple tags and in multiple categories, so that scholars could recognize the connections.
All of this is predicated, though, on the popularity of BookTracker, and getting enough people involved so that it takes on its own momentum.
LikeLike
August 16, 2010 at 12:45 pm |
I’d be interested in hearing more about the kind of tagging Dave has in mind. Right now, the site is set up pretty much as just a finding aid, though it would certainly be possible to extend its functions. If nothing else, it would be possible to create a category of “Reference books” or “Books about books” or something like that which could be used to flag records. Steve Karian had added a couple of later reference books to the site (Arber’s Term Catalogues [1903] and Plomer’s Dictionary of Printers and Booksellers [1922]). Those aren’t eighteenth-century books, but they’re certainly of interest.
Dave, I see that you added a link for William Bent’s The London catalogue of books in all languages: arts and sciences, that have been printed in Great Britain, since the year M.DCC. … With their sizes and prices. … (1773). Could we take that as a concrete example for talking about the sorts of things that one could do? As Eleanor notes, that text has only pretty minimal descriptions of books (4to, 8mo, etc.), but the fact that it covers just a 23-year period would probably mean that identifying which edition was being recorded wouldn’t be too bad in many cases.
What could using catalogues like this help us do that a modern resource like the ESTC doesn’t allow us to do? That’s a real question—not one intended to close off possibilities, but rather to open them up.
Bent does offer prices for the books he records, for instance. If you could find enough similar catalogues and get all of that pricing information into a well-structured database, it would become possible to start taking a hard quantitative look at book pricing patterns. (Maybe that kind of study has already been done without the benefit of databases? Eleanor, this is the sort of thing you know much more about than I do.) The ESTC does record prices when the price is noted on the title page, but of course that wasn’t the case for every text. (A search for “price” in the notes field of the ESTC yields 26,788 results, which is only a fraction of the 460,000+ records in the ESTC.)
Then, too, the fact that prices are in the notes field at all (the MARC field 500, showing the ESTC’s roots in library cataloguing) means that it wouldn’t be altogether easy to explore pricing trends using the ESTC in its current form, even if that information was available for many more texts. Steve Karian’s talk at the last ASECS noted some of the ways that the ESTC’s use of the MARC record limits or at least complicates the kinds of bibliometric work you can do with it, since information isn’t always recorded in consistent ways (there’s nothing to show that an entry in MARC field 500 is a price, for example, and so should be treated as a number). I’m pretty sure there’s interest in creating a more database-y tool that would accommodate that sort of work, so it would be great if we could discuss the sorts of information that scholars would like to be able to mine.
Getting the information that’s contained in books like Bent’s (linked above) into a usable form is, I suspect, going to be a tall order. It does seem like the sort of thing that would be a good candidate for machine reading (in that machines aren’t subject to boredom or vexation). But then one would need a machine-readable version of the text, and we don’t really have that yet. (To see what I mean, have a look at the plain text that Google offers for Bent’s book: there are places where the OCR is pretty good, but it absolutely chokes on the tabular form of the entries. There are large blocks where it simply gives up and shows you the page image with no interpreted plain text; even when it is able to render plain text, it fails to interpret the prices at the right of each line, after the two or more em-dashes.)
I know that Laura Mandell and company are making progress on customizing the Gamera OCR engine for eighteenth-century print. Their work may make it possible to get more out of these kinds of texts. (And things would really start cooking if one could combine better plain text with the web-based version of Juxta that Andrew Stauffer and the folks at NINES will be developing with support from a Google grant.) For the time being, though, it’s going to be a lot of tedious work—though not an impossible task for even a relatively small group of people who were really interested in a particular topic, as Dave suggests.
Dave’s right to note that the effectiveness of any kind of tagging project is going to depend on the momentum that any site can generate. It seems like every week I learn about a new digital project that looks so great it makes me wonder why I haven’t been using it—and also makes me wonder how many more projects are out there doing potentially really useful work that I haven’t heard about yet. This is why I’m really keeping my fingers crossed for the success of 18thConnect: we need a way not just to keep apprised of what different projects are available, but also to coordinate and cooperate on big, exciting projects like the one Dave is beginning to imagine here.
LikeLike
August 16, 2010 at 1:14 pm |
Hi Ben and Eleanor,
To some extent I’m using your site and this blog to think aloud, but it does seem tantalizing to me to have a group of users in BookTracker who could potentially be aggregating the “stuff” in Google books to make it more usable. For one thing, I’m wondering if some kind of “genre” or “category” tag could be introduced, so that we’re not just looking for authors and titles. Even the “books on books” category would be a start.
As for Bent’s catalog, I snagged it because I’m doing my “snapshot of a year” approach, and then I realized that a group of such finding aid texts might be a useful resource for lots of different projects. But I think the key is mobilizing the users of such a site to begin to essentially replicate what you’ve already done, by organizing the material at a new level.
That’s what I was thinking last night, anyway. I’d love to hear if others are interested in this kind of use of Book Tracker, and if it would help it with its mission.
LikeLike
August 16, 2010 at 5:58 pm |
In terms of a database involving pricing and economic info for eighteenth-century books, Shef Rogers (University of Otago) has taken up William St. Clair’s call and is building a database that
will offer the prices paid for copyrights and other cost and pricing
information that will enable us to “value publications” in economic terms.
The main fields of interest for the database are the author, title,
place and date of publication, genre, who paid for the copyright and
how much, and then the price of the work, in order to calculate values
for price/sheet (a formula by which many payments were calculated),
and payment/price ratios (how many copies of his or her own book an
author could buy with the payment received, at the retail price), and
their averages, which are calculated for each set of data searched, so
one can find averages for poetry, for a decade, for gender, and for
genre, and further break those down chronologically if one wishes.
There’s quite a bit one can do with book catalogs–but I had at first thought that Dave was speaking about Booksellers’, Circulating Library, reading club catalogs and the like. I’m actually out of the country and thus pressed for time, but I will detail more upon my return.
Yes, both McGinness’s and Tierney’s projects would benefit from Web 2.0 tools, but both projects were started years before the Internet was even around. Tierney inherited the files that had been in prepartation through collaboration between British and American scholars–some were lost due to German u-boat activities during World War II. Jim, in fact, has received some funding from Mellon–and he now has some help. The basis for his project, however, absolutely has required extensive knowledge of 18th-century newspapers.
LikeLike
August 16, 2010 at 9:06 pm |
As it happens, Bent’s Catalogues have recently been the focus of a very interesting article that is sure to stir up discussion among eighteenth-century book historians:
J. E. Elliott, “The Cost of Reading in Eighteenth-Century Britain Auction Sale
Catalogues and the Cheap Literature Hypothesis,” ELH 77 (2010): 353-84.
Elliott shows how these catalogues can be used to test beliefs about the economics of culture. The article is well worth a read and is available digitally on Project Muse.
Steve
LikeLike
August 17, 2010 at 12:22 am |
Hmm, interesting. Here’s the abstract
Eighteenth-century cultural historians have long asserted that the British book market was flooded with sharply discounted classics after the 1774 Donaldson copyright verdict, giving birth to a middle-class reading public and radically altering the use of literature as a social and political steering mechanism. Despite the work of William St. Clair and others in recent years, the empirical basis for this claim has never been systematically examined. In particular, the vibrant and extensive resale market for literary works in the eighteenth century has been almost wholly ignored. A preliminary review of this market through its period auction catalogues and Bent’s retail pricing indexes suggests that the cost of literature for those buying to read varied minimally between 1750 and 1790. Price data for major figures and works is given in both real and nominal terms, correcting for purchasing power differentials, opportunity costs, and alternative access to reading material. Signal attention is granted selected editions of Milton and Shakespeare as the dynasty-making authors and editions of the era. What emerges is a late-century book market identifiable not so much by price movements as by market complexity and product differentiation. Contrary to conventional wisdom, therefore, the cheap literature hypothesis fails in all relevant particulars. Its inability to underpin widely held assumptions about literary canon formation and common reading habits should encourage us to rethink the complex relationship between culture and society at the time. Above all, we should consider more seriously the importance of source work to effective theory building.
The question of when theory building is appropriate is something that recurred throughout the Swift discussion, and would be a excellent question to take up on its own. But I agree that it would be interesting to see whether source materials like Bent could be systematically aggregated in the ways I’ve been suggesting.
LikeLike
August 17, 2010 at 1:05 am |
Thanks, Steve, for this citation for Elliott. His findings support some of my work on Harrison as a micro-test case of St. Clair’s macro-narrative. St. Clair has presented rich work that has spurred important work in the economics of the trade especially in terms of reading, but I do think that more and more micro-studies will offer challenges to his broad theory. Jan Fergus’s work discussed some of the differences she found between her micro-examination and St. Clair’s. Rick Sher has also questioned St. Clair’s broader claims.
LikeLike
August 17, 2010 at 9:12 am |
In addition to Fergus and Sher, Tom Bonnell has an important article that reveals an astonishing number of errors in St. Clair’s discussion:
“When Book History Neglects Bibliography: Trouble with the ‘Old Canon’ in The Reading Nation,” Studies in Bibliography 57 (2005-2006): 243-61 (online in Project Muse).
And Bonnell briefly discusses St. Clair in his book The Most Disreputable Trade: Publishing the Classics of English Poetry 1765-1810 (Oxford, 2008), pp. 33-34.
LikeLike
August 17, 2010 at 2:24 pm |
Steve,
Special thanks for bringing up Bonnell’s work–the article, in my mind, is a must read.
LikeLike
April 30, 2011 at 8:46 am |
Eleanor, I am interested in Charles Cooke, Harrison’s neighbour on Paternoster Row, and am currently reading Bonnell on his ‘cheap and elegant pocket library’. My particular interest is in the British Theatre component of the library.
That aside, on the usefulness of bookseller’s catalogues/auction records etc I have found the detailed analysis of one bookseller’s/publisher’s advertisements in William Cameron’s article on John Bell very helpful. While not concerned specifically with pricing it demonstrates their use as a dating tool and a provides a different perspective to Dilly’s frequently quoted assertion against Bell after the House of Lord’s decision.
Cameron, William J., ‘ John Bell (1745-1831) : A Case Study of the Use of Advertisement Lists as Evidence of Publishing History’, Humanities Association Review, 26 (1975), 196-216
It focuses on the use of Bell’s stock lists published between 1776 and 1778 in Bell’s British Theatre and then looks at ‘newly published’ advertisements in the same series before seeking lists in his other publications of 1774 – 1777.
LikeLike