KCL • CCH • Minor programme • AV1000 • Relational analysis
Because it is to a large degree a mechanical task, involving much that is mindlessly repetitive, managing a bibliography is a task well suited to the computer. Specialized software exists to help, but unfortunately the packages “worth the candle” tend to be more expensive than a beginner may be willing to contemplate. Because of the expense, this page outlines simple techniques that may be used to take maximum advantage of the machine without this software. It begins with a brief discussion of bibliographies in order to place the computational techniques in the context of good scholarly practice.
Bibliographies are of two kinds: either subordinate to a written argument, appearing usually at the end of the paper or book; or an independent publication that provides either a selective or comprehensive guide to a subject area. Managing the former kind is the subject of this page.
A subordinate bibliography is, as the etymology suggests, an account of the books you have used in producing written academic work. It is an explicit recognition of your sources which serves two purposes: first, to show your indebtedness to what others have done, and so to place your work in the the ongoing discussion that informs it; and second, to point your reader to sources he or she may wish to follow. At the early stages of academic training, where your audience is usually limited to your tutor, the immediate function of the bibliography is to identify quoted material, and so to help you differentiate between your own work and that of others.
Since a bibliography is intended to identify your sources unambiguously, so that your reader can find them with the minimum of fuss, each one should give all the publishing information necessary to do so. The best way to learn what this means is to follow the requirements of a specific assignment in your home discipline, but the usual minimum for printed works is the author's name, the title (both parts, if it has a subtitle), place of publication, publisher and date. You are well advised to write down this information the first time you have your source in hand or on screen, since this will save you the unnecessary labour of retrieving the item a second time—if in fact the item is still available.
For our purposes here, the important matter is that all the component parts are specified —author, title, place, publisher, date (in this example)—and that the entries in the bibliography are internally consistent.
For guidance in citing electronic sources see Citation rules in the Bibliography.
Here we examine the bibliography as a particular kind of textual data and look at the structure of typical bibliographic management software. This shows that formally the bibliography is quite simple and familiar.
Although the end in mind may be a listing at the end of a wordprocessed paper, the data you record in a bibliography is formally tabular. A bibliography consists, that is, of records (entries, individual books or articles) with well-defined fields (the parts of each entry), thus:
| Author | Title | Volume title | Series title | Editor | Place | Publisher | Date |
|---|---|---|---|---|---|---|---|
| Christopher S. Butler | Computers and Written Texts | Applied Language Studies | David Crystal and Keith Johnson | Oxford | Blackwell | 1992 | |
| M. B. Parkes | Scribes, Scripts and Readers: Studies in the Communication, Presentation and Dissemination of Medieval Texts | London | Hambledon Press | 1991 | |||
| Jane Roberts and Christian Kay, with Lynne Grundy | A Thesaurus of Old English | Vol. I: Introduction and Thesaurus | King's College London Medieval Studies | Janet L. Nelson | London | Centre for Late Antique and Medieval Studies, King's College London | 1995 |
| John Sinclair | Corpus, Concordance, Collocation | Describing English Language | John Sinclair, Ronald Carter | Oxford | Oxford University Press | 1991 | |
| Research in Humanities Computing 4: Selected Papers from the ALLC/ACH Conference, Christ Church, Oxford, April 1992 | Research in Humanities Computing | Susan Hockey and Nancy Ide | Oxford | Clarendon Press | 1996 |
You will note that for purposes of the tabular format, certain fields are left blank for certain items. A single-volume work, for example, cannot have a “volume title” distinct from its title; a book without an editor cannot name one; and so forth. In other words, the complexity of the most complex item determines the fields of the table; the less complex items leave empty fields.
The automatic management of bibliographic data—providing a mechanism for entering the items, allowing the collection to be searched and processing it in a particular format for output—is done by specialized bibliographic management software. In general, this software has the structure and function of a database management system and may be represented as in the diagram to the right. Note that between the user (from whom the input comes to to whom it goes) and the actual data is a mediating layer of software, and that attached to this software are user-modifiable input and output templates that specify the form of the data.
For current information about bibliographic management software see the Bibliography.
At minimum you should keep a wordprocessed file with all the bibliographic information recorded for each item in some internally consistent format, using the one most favoured in your discipline. Especially as you come more and more to specialise, you will find that you tend to cite some of the basic sources repeatedly. Keeping this information for the books you have used will obviously save you time later.
Annotating each item will also help. Following each item in your file, note down the gist of its contents, perhaps also its significance for your own work. Especially in projects involving numerous secondary sources, you may need reminding as to why you found a book or article particularly useful. If in describing the item you include relevant terms from your own “controlled vocabulary”, i.e. consistent set of keywords, you will then find it easy to locate items later by an automatic search. Formulating such a vocabulary will also help you think about your subject area.
If you anticipate acquiring bibliographic management software in the near- to mid-term future, you can save yourself significant time then by marking explicitly the identity of each field for each of your bibliographic entries. If you do this well, then you will be able automatically to import your entries into the software. Otherwise you will need to re-enter the data.
Following is a sample scheme, with no particular authority and no specific piece of software in mind. The important matter, again, is to be consistent. Putting each field on a line by itself is also a good idea.
{AU}Anne Carson
{TI}Eros the Bittersweet: An Essay
{PL}Princeton
{PU}Princeton University Press
{DA}1986
{AU}R. O. A. M. Lyne
{TI}The Latin Love Poets from Catullus to Horace
{PL}Oxford
{PU}Clarendon Press
{DA}1980
{AU}Macrobius
{TI}Commentary on the Dream of Scipio
{TR}William Harris Stahl
{PL}New York
{PU}Columbia University Press
{DA}1952
The only problem with recording information in this way is that it is not immediately formatted for use in a paper. By judicious use of search-and-replace or macros, however, the codes can be removed, even the italics or quotation marks put in automatically.
revised February 2008