Chadwyck-Healey English Poetry (site licence 26,500), Early English Prose Fiction (site licence 5,750), and English Verse Drama (site licence 10,000) CD-ROM databases and their equivalents delived via Literature Online (LION, a range of subscription packages available). Chadwyck-Healey http://lion.chadwyck.co.uk and http://lion.chadwyck.com

Catherine Alexander
Research Fellow, University of Birmingham
Lecturer, Shakespeare Centre, Stratford on Avon, UK

Alexander, Catherine.  "Review of Chadwyck-Healey English Poetry, Early English Prose Fiction, and English Verse Drama CD-ROM databases." Interactive Early Modern Literary Studies (October, 1999) 1-16: <URL: http://purl.oclc.org/emls/iemls/reviews/alexlion.htm>

  1. As I settled to this review in August 1999 I was, probably like many readers of Early Modern Literary Studies, belatedly tying up the ends of last academic year, preparing for the next by writing course outlines, reading lists, lectures and papers, and trying to meet publishers' deadlines. Many of these tasks require library research time and as the best hard text comparison for the Chadwyck-Healey material is with a large, very well-stocked specialist library (an analogy they use in their promotional material), I chose to see how effectively it would fulfil these pressing needs. What follows is a report of this practical test of utility.

  2. I've been reworking a lecture on portraits of Queen Elizabeth I and looking again at what is traditionally called the Ermine Portrait, of 1585, attributed to William Segar. It makes use of a conventional representation of the queen's power--a sword--and introduces a new device, a crowned ermine with a raised right paw sitting on the queen's left forearm above a heavily embroidered cuff. This is usually read as a symbol of the monarch's chastity and Roy Strong has glossed it with a reference from Sidney's Arcadia, "rather dead than spotted". However, as the ermine in the portrait clearly is spotted, with the little dark flecks associated with the heraldic use of the creature or its fur, I needed to check out Sidney and search for other ermine references that predate the portrait and the text.

  3. The Early English Prose Fiction CD-ROM database, a collection of over two hundred works from 1500-1700, allows for a standard search using any or all of three basic fields--keyword, title, author--which can be refined by narrowing the year of publication and the gender of the author. A standard search for 'ermine' quickly generated five hits, presented as a brief summary of author and short title, that were easily investigated using the 'view text' and 'view context' facilities. It was a straightforward task first to highlight and then to print the two passages that suited my purpose. Too late I realised that this function does not print the full bibliographic details of the works, information that is available and easily retrievable, and had to return to this later. One useful reference, from Lyly's Euphues, I might have located by traditional means but the other, "the beautifull woman is like the Ermine, whose skinne is estimable, and carcas carion", from Henry Wotton's A Courtlie Controuersie of Cupids Cautels… (1578) I would not have located without the database. But then, without the pre-knowledge or the sense to vary my keyword search (to 'spotted'), I would not have located Sidney's ermine either. Spelled 'ermion' and accessed not through Arcadia or my second stab at a short title---Countess---but found in The Countesse of Pembroke's Arcadia, it was not one of my five initial hits.

  4. Should I choose to use either of these passages I am assisted by the provision of full bibliographical details and the inclusion in the displayed and printed text of page, folio and signature numbers where appropriate. I found too that a great strength of the collection is its use of first editions and the inclusion of prefatory material. There are two drawbacks for the inexperienced or occasional user: printing sections from the full text does not give the option of printing the bibliographical details which can make later identification of the passage difficult if one hasn't kept a manual record. The 'save' function is less transparent than the other facilities, requiring recourse to the accompanying handbook, and it is initially disconcerting when the save or print icons are deliberately inactive as some search screens are displayed.

  5. The English Poetry and English Verse Drama databases provide the full texts of every relevant item listed in the New Cambridge Bibliography of English Literature (C.U.P., 1969-72). They offer quantity but, thus confined, have been criticised for their under-representation of women and working-class authors. The aim of Early English Prose Fiction, on the other hand, is to offer a representative selection of texts, chosen by an editorial board (Holger Klein, David Margolies and Janet Todd) specifically to support academic teaching and research. There are 14 texts by women (Behn, Cavendish, Manley, Pix, Weamys and Wroth) and 143 by men (from Robert Anton to Henry Wotton). While the imbalance is marked, the ability to search by gender affords opportunities that many academics will welcome.

  6. English Poetry does not have a gender search facility but I didn't need it for hunting ermine. The first CD, of five, covers 600-1603, a useful end of era cut-off date that facilitates quick searches by eliminating the need to change discs or have a working knowledge of the possible generic classifications, such as Songbooks or Middle English Romances, that are one of the ways the collection can be manipulated. I was given seventeen hits that, like Prose Fiction, could be viewed as text (the full poem) or context (the 'ermine' quotation). The help function shows what can be done from each screen and supports navigation. English Verse Drama, on two CDs, is organised alphabetically by dramatist's last name, with one disk covering the alphabet from A-H and the other I-Z. It is thus always necessary to swap discs to complete a full search of the 1,800 works by over 500 named authors and over 300 anonymous ones. The anonymous texts are all on the first disc and are listed in the Table of Contents under 'A' and organized alphabetically by title in a series of subgroups: Elizabethan, Jacobean and Caroline, Medieval, Moralities, Tudor, and University Plays. The absence of prefatory material for many texts may be a drawback for some types of research and the exclusive use of the 1623 Folio for Shakespeare’s plays is a trap for the unwary or uninformed. Again, there isn't a gender search facility but looking through the accompanying hard text bibliography (and I certainly found this portable volume useful for comparison with 'real' library source texts) I found 23 women dramatists, and a very good representation of eighteenth-century ones. These potential inconveniences are more than compensated for by the sophistication and range of searches. A simple 'ermine' enquiry revealed 49 hits on disc one and 36 on disc two, each identified not only by play title but by act, scene and line references. With so many hits I found it most convenient to print a list of the context of each one (giving playwright, title, brief quotation, and act, scene, line location), a simple task achieved by highlighting the list. The search can be refined by the addition to the keyword (or, if appropriate, substitution of the keyword), of any or all of the refinements of play title, speaker, playwright, genre, and period. The ability to distinguish between dates of performance and publication is particularly helpful and when I confined by date only 2 hits proved relevant. Nevertheless, the complete list, providing references from the late thirteenth to the end of the nineteenth century, offers the possibility of exploring shifts in meaning and use.

  7. While I've focussed on a simple search facility for each of the three databases, that activity alone has revealed some of the potential that the collections have to support study. The potential is enhanced through the possibilities of searching by phrase--a facility I've used to explore textual appropriations of Shakespeare--and the use of proximity operators (finding work x near to word y), Boolean operators (finding word x and/or word y) and wildcard characters which help cope with spelling variants in early texts. This last facility allows one to find a word regardless of how part of it is spelled, so that lo?e will match lone, love, loue, or lose and lov* will match love, loving, or loveless (and Loughborough if the compositor used a v for the first u). It also needs to be stressed that each collection, regardless of how one manipulates it, offers access to full texts and thus makes available a remarkable range and quantity of material that many students, teachers and researchers would otherwise have little opportunity to view. As a teacher I can expect students in licensed institutions to look at out of print texts and, perhaps with occasional checks to ensure that the complete work is read, require them to consider word use and frequency. There is the potential for a focus on language and style that could cause a significant shift of direction--from cultural context back to text--in the study of literature.

  8. Literature Online (LION) brings together all the Chadwyck-Healey databases formerly only available on CD-ROM plus some new special collections (to which I didn't have access), Websters Third New International Dictionary, biographies and bibliographies, and web-site connections to electronic texts, and allows one to search the lot with a single request. The initial choices of 'Find authors', 'Find works', 'Browse authors and works', 'Search texts', 'Search secondary sources' and 'Individual literature collections' could be daunting in their range and choice but serve to overturn one frequently voiced concern about electronic collections. It has been suggested that they will never compete with the experience of using a hard text library and the possibility of making serendipitous discoveries but this is certainly not the case with LION: its very size and scope and the potential for making new connections sees to that. One could spend as long with LION as one does in a library and find the experience just as exciting and probably more productive.

  9. The 'Find authors' facility was particularly useful for autumn class preparation, offering 'Biography', 'Works by' and 'Works about' for each author hit. The biography file for an author is an essay of up to 3000 words "adapted from data developed by the H. W. Wilson Company, Inc" concerning the author's life. These essays print to a wide-screen format that loses the right hand side of the text, and there is no obvious way for the inexperienced user to save to file or disc. (Users familiar with the Windows clipboard feature will find that they can 'copy' the text and drop it into another application, but this is not apparent to novices.) Such problems are not uncommon in web-based resources and Chadwyck-Healey's product is better than many in this respect, but the problem ought to be eliminated entirely. Access to ABELL (Annual Bibliography of English Language and Literature) and LIFT (Literary Journals Index Full Text) is helpful for constructing reading lists and checking references, and for idle or purposeful browsing of the 'who has published what' variety. I found 236 works by Stanley Wells, for example, arranged chronologically from a Notes & Queries entry in 1959 to "'A funeral elegy': obstacles to belief" essay in Shakespeare Studies, 1997. However, if one chooses to browse the list of authors of secondary works one finds that Stanley Wells has been recorded variously as "Wells, S. W", "Wells, S", "WELLS, STANLEY", "Wells, Stanley", "Wells, Stanley (ed.)" WELLS, STANLEY W" and "Wells, Stanley W". One would expect that this makes no difference to the search engine, but when I selected all of these variant names in the index (and manually deselected "WELLS SLIGHTS, CAMILLE" and "Wells Slights, Camille" who had interposed themselves between the multiple personae of Stanley Wells) the total number of works found rose to 240.

  10. I encountered further anomalies. Completing a paper on Shakespeare and Race and looking for contemporary uses of the word 'negro' I had occasion to print off Francis Thynne's "Art cannot take awaye the vice of nature" from the English Poetry Database:

    The healthfull bathe which daielie wee doe see
    To cure the sores and fleshe of lothsome skinn,
    Cann never make the negro white to bee,
    Or clense the harlot from her loathed sinne…

    Locating the same piece on LION took some finding. LION's search screen divides the collection into 4 main categories: 'All', 'Poetry', 'Drama', and 'Prose'. If 'All' is selected the results are tabulated into hits for 'Poetry', 'Drama', and 'Prose'. A 'negro' word search clearly needs refining but clicking on 'enter additional terms', or even simply switching from 'Poetry' to 'Drama', deletes anything already entered in keyword or author boxes. This becomes quite an annoyance when one is trying to slide effortlessly between the genres, as the databases otherwise encourage. The kind of 'additional terms' one can enter varies according to genre. Naturally, only within 'Drama' can one decide to limit one's search to stage directions and for 'Prose' one can choose to search just the narrative text or just the apparatus. However, it is not obvious why the choices of literary period vary as they do between genres: Tudor should be Tudor for poets and dramatists alike, and most disturbingly for 'Poetry' the list of periods skips from Tudor (1500-1580) to Jacobean and Caroline (1603-1660) without the Elizabethan (1580-1603) period offered for 'Drama' and for 'All'. Equally confusing, the appearance of 'Moralities 1405-1603' in the literary period field for the 'Drama' search screen clashes with the choice of 'Moralities' in the sub-genre field on the same screen and one is left wondering which, if either, to activate. Having chosen to search 'All' texts, the date refinement 'Elizabethan 1580-1603' gave no hits in poetry, presumably because that literary period is not on offered on the 'Poetry' search screen. Accessing through 'Browse authors and works' generated a show of temperament (from me and the system): I was informed of the presence of 136 authors whose last names begin with T but only allowed to look at 5 of them at a time. I discovered that at the bottom of the screen one may select how many one would like to see at a time, and I suspect that other users will follow me in muttering "all of them, please" as they change this setting, only to discover that 99 is the maximum.[1] Eventually finding the piece I was interested to note that the printed LION text gives fewer bibliographical details (no publisher or editor or volume length), but provides line numbers for the verse, and highlights the hit with an icon that looks like a cricket ball. (A dispiriting reminder in England during the last test match). It's helpful too to be given the memory size needed for printing and saving.

  11. Exploring the gender balance in LION (530 female authors from Diane Ackerman to Zephenia and 2666 males in total, or 'Living in the years 600-1655' 19 women to 685 men) and browsing through some of the results I found another anomaly. Using the 'Find author' search generates 'Newcastle, Margaret Cavendish, Duchess of' yet 'Browse authors and works', which is arranged alphabetically, does not give an entry under N. One understands that LION has been put together from a number of separate products, but they all were made by Chadwyck-Healey and we may hope that sooner rather than later the inconsistencies across the various indices will be removed.

  12. Having used Early English Prose Fiction to print a list of all the hits, in context, of 'time' in Robert Greene's Pandosto (45), I tried the obvious search on LION and asked for the frequency of the same word in The Winter's Tale. Somewhat surprised to find no hits I tried more numbers games and looked for 'love' in Othello (0), 'magic' in Shakespeare (0), and so on, only eventually getting a response with 'blood' in Macbeth (24). Remembering that Chadwyck-Healey rely upon the 1623 Folio text of Shakespeare's play, I altered the search term to loue for Othello and the missing hits turned up, and likewise the spellings magick and magicke elicited the expected hits for the entire Shakespeare corpus. However, the Winter's Tale mystery remained until I browsed the index of titles and chose The winters tale (1623), whose lack of an apostophe appears to have been the problem.

  13. None of the difficulties diminishes the value of this remarkable resource that continues to expand and develop. While its size may daunt, that's a problem for the user, not the provider and the discrepancies are certainly no greater, and in most cases far fewer, than would be found in conventional cataloguing. Equally, time spent converting saved texts or scrolling through the quantity of secondary material is as nothing compared to the common library experiences of missing books, incomplete catalogues, restricted access, photocopying difficulties, queues and delays. I wonder, too, how common was my experience of assuming that search failures on LION were somehow my own fault (lack of technical expertise) whereas any difficulties encountered in traditional libraries are the fault of others. I anticipate a reduction in the higher education equivalents of the dog-ate-my-essay excuse; the book's not in the library or the library is closed.

  14. I return the review material and my LION password with great reluctance and make a strong recommendation to all lecturers, students and researchers to use this fine resource. It's a healthy reminder of one's own ignorance and an exciting opportunity to remedy the situation.


  15. I ran these programs an IBM compatible PC running the Microsoft Windows 95 operating system using a Pentium processor of 166MHz with 32MB RAM. For the Internet-delivered LION product my machine was connected via a 33.6kbps modem to a dialup Internet Service Provider.


1. Editor's note: I traced Dr Alexander's footsteps and also was initially unable to find the poem by Francis Thynne. The biography essay indicated that he is known to the database, but the 'works by' link for him led to an empty document. This glitch was not confined to Thynne: Aesop, Tasso, and the notoriously prolific Robert Greene had produced not a line between them, according to the 'browse authors' facility. However, the 'works by' link from within the individual biographies for these authors led to the expected lists of their output and from one of these I was able to find Thynne's poem "Art cannot take awaye the vice of nature".

1998-, Lisa Hopkins(Editor, EMLS).