"The peculiar limitations and advantages of EEBO over the printed versions of Greg's Bibliography of English Printed Drama, the printed Short Title Catalogue, and the Early English Books microfilms" by Gabriel Egan
When literary studies started to go electronic about 10 to 15 years ago, everybody got excited about hypertext. George Landow even famously argued that by enhancing our power to flit like readerly butterflies from one thing to another, hypertext magnified the textual jouissance that French theorists such as Roland Barthes had identified as the essence of literary pleasure. This, we now know, was nonsense. The greatest hypertext imaginable is the Worldwide Web, and I defy anyone to extract jouissance from flitting amongst the myriad objects that a Google search throws up in response to a simple enquiry about, say, W. H. Auden's prosody. The big deal is not hypertext but the capacity to search large textual corpora, which is what projects such as EEBO give us. Unlike the printed Short Title Catalogue and W. W. Greg's Bibliography of English Printed Drama, an electronic corpus frees you from the vagaries of an author's index: you can have the computer do a brute force search through all the records, whether or not the author indexed them.
Let me take a practical example, knowing that phrase 'title-page' was first use in Renaissance drama in the character King Simonides in the play by Shakespeare and George Wilkins's called Pericles, I found myself looking at the early printings of this play. Let's all do that (put 'pericles' into title and sort on date). I was struck by that word 'play' and realized that I wasn't used to seeing the word 'play' on a play title-page. When did publishers start doing that? EEBO should tell us, so let's try to get the answer . . .
We want to see how the word 'play' is used in play title-pages: a search for this word in 'title' is best, since everything on the title-page tends to go in there except what's in the imprint. Knowing that 'play' could also be spelt 'playe', we put 'play*' in a search on the title field, and get 713 records in return. Sorting them by date, and scan-reading them, it's pretty clear that we've hit lots of things that aren't plays. We need to select only things that are plays, and EEBO appears to offer this in its 'subject' field. Put 'drama' in there, and we get a manageable 71 records back: that sounds in the right ballpark for play title pages that use the word 'play'. (Remember that for most decades up to the 1640s, only a few dozens plays at most where printed -- it was a minority interest.) It sounds right, but it's dead wrong: this search has misled us entirely, because it depends on the records in EEBO using the 'subject' classification consistently. We know (or we should suspect) that the EEBO records come from disparate sources, and any of us with librarian instincts will probably suspect that heterogeneous records might well be inconsistent on this score.
Let's do a test to see. We'll use one in which the correct sequence of records is already known (well, fairly well known to me), which is the publishing history of Shakespeare's plays in the 1590s and 1600s. Let's search for Shakespeare in author, drama in subject, between 1590 and 1610, sorted on date. The results seem plausible: 25 records. Look at the bibliographical record for a particular play, doesn't matter which one, down there at the bottom is the 'subject' field and it's got 'drama' in there, which is why we hit it. But notice that there's no Hamlet in 1603 (Q1) and 1604 (Q2). Turns out that the records for Hamlet lack their subject-is-drama field, so in fact the 'subject' field is not a reliable way to cut down our original search by specifying genre: it'd just miss some records we'd want to see.
So, let's go back to our long-winded approach and see if we can improve it. Confining our attention to just the spellings 'play' and 'playe' (the likeliest for our modern sense of the word), we can use the 'title keywords' search to combine them (OR), limiting our attention to 1500 to 1650, and upon sorting these by date we do indeed get a sense that dramas stopped being called plays once the professional theatre took off and started again with Pericles. Repeat. Suddenly, around the time that the professional London stage took off with the opening of the first open-air amphitheatre playhouse in 1576, plays stopped being called plays on their title pages. There are several ways to explain this in relation to the market for drama, for example, that the professional stage distanced itself from the idea of 'playing' and focussed instead on named genres such as a 'tragedy'. Indeed if you make a study of these records you find that once the playhouses are open there's a shift in the marketing tone of play title-pages: browsers in the bookshops are encouraged to buy the text not in order to put the play on for yourself, as the earlier books had encouraged, but to recapture the rapture of seeing the plays in the professional theatres, which is what the title-pages now always stress. Moreover, this trend ends with Pericles, which IS called a play on its title-page, and is the first drama since the Tudor interludes to be described as a play on its title-page. You can't prove this claim with EEBO -- you need the STC or preferrably the Electronic STC to do this, and you really need to read all the title-pages -- but you can get a sense of the trend, and indeed it was by EEBO searching rather than ESTC searching that I got put onto this phenomenon.
And, of course, you actually can read all the title-pages on EEBO. But before moving on to that, let me remark that some of the bibliographical data in EEBO is in fact more reliable than that in the printed sources. SWITCH TO PPT ONLY. THESE ARE IMAGES I'VE PULL OFF EEBO Although no play was called a play on its title-page before Pericles, there had been a reference to plays, plural [SLIDE]. The Conspiracie, and Tragedie of Charles Duke of Byron was the title of a printing of both parts of George Chapman's two-parter about the recent trial and execution of the Marshall of France. Its title-page claimed that the contents were "Acted lately in two playes, at the Black-friers". The first word on the title-page of the 1608 printing of The Conspiracie, and Tragedie of Charles Duke of Byron is a xylographic (that is, woodcut) "THE" that Greg in his Bibliography of English Printed Drama labelled "block 1", and in his reference list entry of Notabilia he recorded [SLIDE]:
xylographs (lettering cut on wood): 'The' block 1 [203(a-c), 242(a)], block 2 [148(b), 243(a), 264(b), 308], block 2* [249(b, c), block 3 [256(a)]; 'Al Fooles' ; black-letter headings [412(c)]; and cf. woodcut in 
(Greg 1970, 1642)
Greg died before completing this volume of his bibliography, and the index entry does not fully reflect what he knew and recorded in the preceding descriptions. For example, his descriptions of items 142e, 204a, 204c, 222, 274, 275 record use of the blocks but are not referred to in this index entry. This I know from reading every entry, but in fact (and remarkably given its inconsistent use of the more fundamental criterion 'genre') EEBO contains seemingly complete information on xylographs, so let's search for them. Let's confine ourselves to 1600 to 1650, put 'xylographic' in the 'keyword' field, and we see that xylographs get used on such things as the Livy's Roman history, Montaigne's essays, and some plays. (Go into the first, Livy, and show why you need to search 'keyword' rather than 'title keyword'.)
By his designations "block 1", "block 2", "block 2*", and "block 3" Greg appears to have meant that each is a particular design of woodcut letter, not particular pieces of wood reused in different printings. But in fact "block 1" is a single piece of wood used in different printings by different printers. [SLIDE] In 1604 Valentine Simmes printed John Marston's The Malcontent for William Aspley with "block 1" on its title-page (Marston 1604, A1r) and to within fractions of a millimetre (attributable to paper expansion or contraction, and the vagaries of inking) the internal and external dimensions of the xylograph in British Library copy C.34.e.17 match those of the same block design [SLIDE] on the title-page of Simmes's 1604 printing of Thomas Dekker and Thomas Middleton's The Honest Whore for John Hodgets (Dekker & Middleton 1604, A1r), British Library copy at C.34.c.24. (Here I'd qualify what Justin said this morning about no needing to go to the British Library or the Bodleian. I agree that you don't if you're interested primarily in what the books say in words, but if you're interested in the book as a physical object then EEBO is a great way in but then you have to check the book itself.) There are no obvious defects in the xylograph that one might use to determine with certainty that these two Simmes printings were made using the same piece of wood, but that is the obvious inference from the virtually identical dimensions.
No earlier play printing had such a xylographic "THE" on its title-page, but four others soon followed. [SLIDE] In 1605 Thomas Creede printed The London Prodigall for Nathaniel Butter with the "block 1" design on its title-page (Anonymous 1605, A1r) and in the British Library copy at C.34.l.3 it has precisely the dimensions of the one on the Simmes title-pages for The Malcontent and The Honest Whore. Moreover, in this printing there are a number of defects visible in the woodcut that enable us to determine that the next three times it appeared on play title-pages the same piece of wood was used: the top right serif of the T has chipped, the right side of the central horizontal bar of the H has taken ink (become filled where it should be hollow), the outer edge of curve in the top left corner of the E has worn away (taken no ink) where it comes closest to the top-right corner of the H before it, and the bottom edge of the bottom horizontal bar of the E has a small chip that the serif of the central bar points to. [SLIDE] Those next three times this xylographic "THE" appears on a play title-page are: Chapman's The Gentleman Usher (Chapman 1606, A1r) printed in 1606 by Valentine Simmes for Thomas Thorpe, [SLIDE] Heywood's? The Fayre Mayde of the Exchange (Anonymous [possibly Thomas Heywood] 1607, A1r) printed by persons unknown in 1607 for Henry Rocket, and [SLIDE] Chapman's The Conspiracie, and Tragedie of Charles Duke of Byron (Chapman 1608, A1r) printed in 1608 by George Eld for Thomas Thorpe. The British Library copies of these (at C.12.g.4 (5), C.57.e.27 and C.30.e.2 respectively) show identical defects to the "THE" on the title-page of The London Prodigall, so these four plays' title-pages at least were printed using the same piece of wood. The two title-pages of 1604 by Simmes were made either with the same woodcut before it started to show wear, or from a woodcut that was virtually identical to the one subsequently used by Simmes, Creede, Rocket's unnamed printer, and Eld. The obvious inference, made by Akihiro Yamada and by Paul Edmondson, is that Simmes lent his decorative xylograph to men with whom he shared printing, as Creede and Eld were (Yamada 1994, 72-73; Edmondson 2001, 2, 22-23; Ferguson 1968, 88).
Whatever the explanation, putting a xylographic "THE" as the first word of a play title-page became quite a popular thing to do, as seen in Middleton's The Phoenix (Middleton 1607, A1r) in 1607, Dekker and John Webster's The Famous History of Sir Thomas Wyat (Dekker & Webster 1607, A1r) in 1607, Wilkins's The Miseries of Inforst Mariage (Wilkins 1611, A1r; Wilkins 1629, A1r) when reprinted in 1611 and 1629, the anonymous The Merry Devill of Edmonton (Anonymous 1612, A2r) when reprinted in 1612, Elizabeth Carew's The Tragedie of Mariam (Carew 1613, π2r) in 1613, and the anonymous The Famous Victories of Henry the Fifth (Anonymous 1617, A1r) when reprinted in 1617. The size and unusual design of the xylograph might, as Edmondson put it, "suggest that this play-text is the definitive article; the most longed for particular; the one and only" (Edmondson 2001, 22-23). On the other hand Simmes's use of the xylograph on half a dozen non-dramatic printings supports Ferguson's suggestion that it was simply a convenient means to express the printer's preference for the first line of a title-page being a large "THE" (Ferguson 1968, 79). Since over half the play titles in the period began with 'The', this decorative block was an economical way to embellish title-pages.
Anonymous [possibly Thomas Heywood]. 1607. The Fayre Mayde of the Exchange. STC 13317 BEPD 242a. London. [Valentine Simmes for] Henry Rocket.
Anonymous. 1605. The London Prodigall. As it Was Plaide By the Kings Majesties Servants. By W. Shakespeare. STC 22333 BEPD 222a. London. T[homas] C[reede] for Nathaniel Butter.
Anonymous. 1612. The Merry Devill of Edmonton. STC 7494 BEPD 264b. London. Thomas Creede for Arthur Johnson.
Anonymous. 1617. The Famous Victories of Henry the Fifth. STC 13073 BEPD 148b1. London. Bernard Alsop.
Carew, Elizabeth. 1613. The Tragedie of Mariam, the Faire Queene of Jewry. Written By That Noble Ladie, E. C.. STC 4613 BEPD 308A1, A2. London. Thomas Creede for Richard Hawkins.
Chapman, George. 1606. The Gentleman Usher. STC 4978 BEPD 226. London. V[alentine] S[immes] for Thomas Thorpe.
Chapman, George. 1608. The Conspiracie, and Tragedie of Charles Duke of Byron, Marshall of France. Acted Lately in Two Playes. STC 4968 BEPD 274a, 275a. London. G[eorge] Eld for Thomas Thorppe, sold [by Laurence Lisle].
Dekker, Thomas and John Webster. 1607. The Famous History of Sir Thomas Wyat. With the Coronation of Queen Mary. Written By Thomas Dickers, and John Webster. STC 6537 BEPD 256a. London. E[dward] A[llde] for Thomas Archer.
Dekker, Thomas and Thomas Middleton. 1604. The Honest Whore, With, the Humours of the Patient Man, and the Longing Wife. STC 6501 BEPD 204a. London. V[alentine] S[immes] for John Hodgets.
Edmondson, Paul. 2001. A Critical Edition of The London Prodigal. Unpublished PhD thesis. Birmingham. University of Birmingham.
Ferguson, W. Craig. 1968. Valentine Simmes: Printer to Drayton, Shakespeare, Chapman, Greene, Dekker, Middleton, Daniel, Jonson, Marlowe, Marston, Heywood, and Other Elizabethans. Charlottesville VA. Bibliographical Society of the University of Virginia.
Greg, W. W. 1970. A Bibliography of the English Printed Drama to the Restoration. Vol. 3 Collections, Appendix, Reference Lists. 4 vols. London. The Bibliographical Society.
Marston, John. 1604. The Malcontent. STC 17479 BEPD 203c. London. V[alentine] S[immes] for William Aspley.
Middleton, Thomas. 1607. The Phoenix, as it Hath Been Acted. STC 17892 BEPD 243a. London. E[dward] A[llde] for A[rthur] J[ohnson].
Wilkins, George. 1611. The Miseries of Inforst Mariage. STC 25636 BEPD 249b. London. [William White] for George Vincent.
Wilkins, George. 1629. The Miseries of Inforst Mariage. STC 25637 BEPD 249c. London. Aug[ustine] Mathewes for Richard Thrale.
Yamada, Akihiro. 1994. Thomas Creede: Printer to Shakespeare and His Contemporaries. Tokyo. Meisei University Press.