Oxford English Dictionary (OED) Version 2.0 for Windows on CD-ROM (Oxford University Press). Price: £175+VAT for limited time, usually £250+VAT (http://www.oup.co.uk)

Ann Thompson, King's College London, ann.thompson@kcl.ac.uk

Ann Thompson . "Review of the Oxford English Dictionary (OED) on CD-ROM." Interactive Early Modern Literary Studies (January, 2002) 1-10: <URL: http://purl.oclc.org/emls/iemls/reviews/thompsonoed.htm>

In this review I report on my use of the Oxford English Dictionary (OED) on CD-ROM Version 2.0 for Microsoft Windows. There is no version of this product for the Apple Macintosh family of computers (which includes the iMac and iBook) but a comparable subscription service delivered via the Internet is available for any machine running a recent web browser. Version 2.0 of the CD-ROM product corresponds to the second edition of the paper-based OED as of 1999. At the time of writing, early January 2002, Version 3.0 of the CD-ROM product had just been released and existing owners of Version 2.0 are offered an upgrade for £51.03+VAT. It is claimed that Version 3.0 has "improved functionality, it will install and run from the hard drive and will include 'the Additions'". The 'run from hard drive' feature is acknowledgement from Oxford University Press (OUP) that many users have hard drives capables of holding the full contents of quite a few CD-ROMs (which was not the case when CD-ROMs first became widespread about 10 years ago), so they might prefer to have the software install itself in its entirety on the hard disk and so leave the CD-ROM drive available for other purposes, such as listening to music. Third-party software has been available for some time which 'virtualizes' any CD-ROM, copying its contents to the hard disk in such a way that the software is fooled into accepting this data image as though it were the original plastic disk. The OED is now a project in constant development, with 'the Additions' appearing quarterly. Since October 2001 Version 2.0 has been available direct from OUP for £175, and Version 3.0 is currently the same price, but reportedly will revert to £250 from June 2002.
I have been used to consulting the OED in print form, the monumental multi-volume 'proper' format rather than the tiny-type two volume affair (which I should think really is obsolete by now, unless you have developed a magnifying glass fetish). My current project (with Neil Taylor) is an edition of Shakespeare's Hamlet for Arden³, and naturally I have found myself looking things up in the OED frequently. I haven't yet had occasion to make myself familiar with the online version available through my university library, so this has been my first experience of the OED in electronic form. I propose to discuss this experience in itself, and here what I say will largely apply to online and CD-ROM versions equally; but I will conclude with a few words on the extent to which the CD-ROM version justifies its independent existence.
The CD-ROM installed without incident on my PC (which runs Microsoft Windows 98); once installed, to use the program you need to have the Data Disk in your CD Drive, and when you open the program you are asked to click to obtain 'authentication from the original CD', which comes through in second. This means, which is fair enough in commercial terms, that you will not be able to lend your copy of the electronic OED to friends and colleagues. Were you to change computers, you would simply uninstall the programme and reinstall on another machine, so this anti-piracy measure is not especially intrusive. An elegant, unfussy home page opens onto the resources of the dictionary through the Search button. The most basic use one can make of the CD-ROM is simply to go to the entry for the word you are interested in, via a 'Find Word' facility, just as one would in using the printed version by taking the appropriate volume off the shelves and leafing through. While the 'feel' of these two activities is necessarily radically different—the electronic OED cannot reproduce a sense of the sheer physical monumentality of the huge Victorian project—I was pleased to discover that the typographical design has been kept close to the printed original, while it has also been made visually pleasurable on the screen in that medium's own terms. It was really important to get this right, and full marks to OUP for making sure that the 'look' of the electronic OED remains both functional and attractive.
Of course, the 'added value' of the electronic OED lies in what you can't do with the printed version. With every word in the OED available for search (at least in theory, more on this later), it becomes possible to do things in a few seconds which would have taken my Arden predecessor, Harold Jenkins, years to complete. A straightforward example would be to assemble every quotation in the OED which is taken from Hamlet. Finding out how to do this does require a bit of experiment, if like me you are starting from scratch. Type 'Hamlet' into the 'Simple Search' dialogue box and select 'quotations' as the part of the dictionary you wish to search, and you will be given a Results List of 466 items, but they will not be quotations from Shakespeare's Hamlet, but quotations within which the 6-letter string 'H-a-m-l-e-t' occurs. It might just be useful to be pointed towards D. H. Lawrence's Twilight in Italy ("Hamlet is far more even than Orestes . . . a mental creature, anti-physical and anti-sensual"), but some titles are surprising: William Faulkner's Hamlet (a puzzle for a moment until you think The Hamlet) was clearly mined with some assiduity for OED-2, and there's also Bernard Kops's The Hamlet of Stepney Green, and references to Hamlet operas. A glance at the printed OED reveals the problem: the play Hamlet is known to the dictionary by the abbreviation 'Ham.' Clicking 'Help' gives you access to an Abbreviations List for Shakespearean Plays and Books of the Bible. (Now, why just those two?…) It also gives a list of abbreviations used in the apparatus of the OED such as 'lang.' for 'language' and 'Lap.' for 'Lapp[ish]'.
What it does not provide is a list of abbreviations used to shorten the names of authors. Access to 'Shakes.' for Shakespeare was provided by the concise and lucid manual that accompanies the CD-ROM, but scholars working on less sacred texts will find that they will want to get a feel for how the OED First Edition handled their authors' names from the printed edition. Oliver Goldsmith scholars will need to work in terms of 'Goldsm.', for instance. Indeed, in the course of finding this example, I came across on the same page a case of Shakespeare being not 'Shakes.' but 'Shaks.'. By using the 'wildcard' facility, that is by typing not 'Shakes.' but 'Shak*s' into the 'Simple Search' dialogue box and selecting to search the 'Full Text' of the dictionary, I increased the total number of quotations from Shakespeare available to me from 29956 to 32083. This isn't something that OUP can reasonably be expected to fix, but we will need to alert students to the need to explore what their target author names' abbreviation(s) might be, and familiarising them with the OED in its printed version may still be useful for this purpose. [1]
But to return to 'Ham.' Typing in 'Ham.' (or 'Ham' or 'ham') into the 'Simple Search' dialogue box and selecting to look in 'quotations' brings up 3096 results, most of which are from Shakespeare's play but a good few of which are medieval 'ham' ("Ant am in hare beddes so bisi ham a buten") or porcine 'ham' (including a quotation from the 25 August 1973 edition of Black Panther: "Picnic ham at 97 cents or salt pork at 99 cents a pound?"). To narrow the search to the Prince of Denmark, one needs to click on the 'More Options' button within the 'Simple Search' dialogue, which action makes available a second input box; here one can specify a further criterion by indicating who the quotation's author must be. Type in 'Shakes', and a more manageable 1367 results come up, rising only by three if 'Shak*s' is entered (and these three actually are references to notes in 'Shaks. Works' as edited by Steevens, Malone and Johnson). The list usefully allowed me to go through the entries that might concern an editor writing commentary for an edition.
One result of looking through this list may be worth reporting. A surprising number of words ending in '-ment' occur in it, many of which are first found or uniquely found in Shakespeare. Some examples are: 'annexment', 'blastment', 'definement', 'strewment'. Shakespeare's liking for '-ment's has in fact been noticed before (by Karl Elze in 1882), but having the Hamlet quotations all set out before one is a good way to discover it for oneself, so to speak. Indeed, the wildcard facility allows one to generate a list of all the Shakespearean '-ment's, which turn out to number 666. Mind you, this is out of an overall OED total of 29,297 Shakespearean quotations. By comparison, Jonson registers 104 '-ment's, but that is out of a total of only 4207 quotations. So Jonson's '-ment' score as a percentage of all words for whom his quotations are used is actually higher than Shakespeare's (2.47% against 2.27%). Of course, Jonson's '-ment's may well be less unusual '-ment's than Shakespeare's. (Incidentally, the politics of OED quotation rates is an interesting matter in itself: one wonders if, once the much-awaited Oxford Middleton edition under Gary Taylor's general editorship appears, the OED's current équipe will be moved to scan it again and raise the current 1871 quotations—mostly involving the playwright, but including such curiosities as 'Unfortunately, the Mayoress of Middleton was deafish'—to at least Jonsonian levels. For what it's worth, the raw Middleton '-ment' score is the significantly lower 1.12%.)
Thinking about comparing Shakespeare with Jonson and Middleton (and obviously if one were researching the '-ment' question seriously one would be working with concordances rather than with the subset of usages that get into the OED), I did find one thing impossible to do with the electronic OED that might have seemed likely to have been easy. Say that one wondered, 'How many times are Shakespeare and Marlowe featured together in a single OED entry?' For a start, two-term searches work only within a 'section', defined as 'an individual definition, etymology, or quotation', plus (in Full Text search mode), 'entry names and spellings sections'; so one would not expect to find two-term searches working over the entire range of a multi-definition entry. However, it is more surprising to discover that, even in full text mode, typing 'Shakes' in the top input box and 'Marlowe' in the bottom input box, only one result comes up, in the etymological note for 'massacre', noting how Shakespeare and Marlowe accent the first syllable of the word while Spenser accents the second. It seems unlikely that in no section of the OED are the two playwrights quoted together; and indeed it is easy to demonstrate that this kind of search doesn't work by using an actual entry and working backwards. There is a Hamlet quotation in the OED's I.1.a. definition of 'shadow' [2], and an earlier quotation from Chaucer in the same definition (i.e. in the same section). But a search will not reveal this; instead, we find seven items, each of which puts together the two authors within a stretch of text (for instance, in, again, an etymological note for 'fizgig'). So clearly the term 'full text', suggesting to the unwary something like 'every word in a section', conceals some kind of hierarchisation whereby authors' names are tied to their quotations in such a way as to make them inaccessible for search where another author's name is concerned. This seems a pity; when I was working on my book Shakespeare's Chaucer, it might actually have been very useful to look at the cases where a Chaucerian quotation is succeeded by a Shakespearean quotation in the same section of an OED entry. [3]
Copying text and printing it are reasonably easy once you have worked the system out. From the screen itself, although you can highlight text, there is no way to copy that directly into, say, a Word file. Nor can you print highlighted text directly (indeed, it is not clear to me what you can do directly with highlighted text). What you can print directly is an entire entry (not section), so if I wanted to 'print from screen' I could do so but I would end up with all of 'shadow' rather than definition I.1.a. On the other hand, you can print from screen directly the Results list from your search, but only one page at a time. The way around these limitations lies in a button called 'Save'. If I have the 'shadow' entry on screen and press 'Save', the entire entry is saved, in HTML format (the format in which web pages are written). If, now, I bring up the file I have named 'shadow.htm', I find I have a file which I can manipulate in the ordinary way: I can print highlighted text only, and I can save highlighted text and copy it into a Word document. Anyone with Windows 98 or later (that is, Windows Me, 2000, or XP) can work with HTML documents because Internet Explorer (Microsoft's web browser) is deeply integrated with the operating system. Users of Windows 95 and earlier will need to install Internet Explorer or Netscape Navigator (both available as free downloads) in order to use HTML files.
When I returned to my office after the Christmas break, I tried accessing the OED online via the university library. The screen that greeted me was familiar from the CD-ROM version, and it is clear that there is no major divergence, yet, between the two versions, the 'Additions' excepted. The odd failure to allow searching for two authors in a single section remains true of the online version. Online access allows for copy-and-paste operations more immediately than does the CD-ROM, presumably because the interface (being delivered via a standard web browser) is not so tightly controlled as the proprietary interface of the CD-ROM version, and also allows you to send definitions as emails to your friends and colleagues. So is it worth buying the CD-ROM? The case for doing so if you have free access to the online version at work is not all that clear, though it does feel nice to own one's very own copy psychologically; and, given the capacity for institutional network systems to develop a fault at just the wrong moment, it is good to know that you have the OED 'safe and sound' and accessible at all times. Indeed, this advantage is potentially more than psychological. Were a global corporation to succeed in having an offending word removed from the online database (as the McDonalds fast-food chain attempted with the word "McJob", meaning low-pay, low-skill employment), CD-ROMs, like books and unlike web sites, appear valuably resistant to external interference. If you don't have free and convenient access to the online version of the dictionary, the CD-ROM starts to look like pretty good value at £175+VAT, and until cheap wireless connection to the Internet becomes available via the third-generation mobile telephone network currently being built, the CD-ROM is the only way to use the dictionary while travelling.

Notes

1. Continually-updated databases such as the electronic OED are especially prone to such inconsistencies. The reviewer's totals for various searches with Version 2.0 of the CD-ROM are about 1% different from those achieved by the editor using Version 1.10 of the same poduct, which also is nominally based on the second edition of the paper version. The differences are not due only to the 'Additions'. Like Version 2.0, Version 1.10 usually abbreviates 'Shakespeare' to 'Shakes.' but just once its uses 'Shak.' instead; this the 'Shak*s' wildcard search would not find. [—Editor]

2. In Version 1.10 of the CD-ROM product, which is also supposed to represent the second edition of the OED, the illustrative quotation for 'shadow' comes not from Hamlet but from Measure for Measure. It would appear that OED-2 is not a stable referent: as well as 'Additions' there appear to be revisions of existing entries. [—Editor]

3. Such a search can be undertaken with Version 1.10 of the CD-ROM product, but with the arbitary and greatly limiting proviso that the two names 'Chaucer' and 'Shak*s' have to occur within 32 words of one another. [—Editor]