Fingerprint: a sequence of characters derived from the text of an early printed book. A fingerprint can be used to detect variant settings of type in otherwise matching editions and to identify the reuse of the same setting in ostensibly different editions. It can also act as an identifier for any printed work, assisting identification of partial texts. Loosely, the fingerprint is an 'ISBN for older books'.
Much of what follows draws on Neil Harris's excellent survey, 'Tribal lays and the history of the fingerprint', in CERL papers VI (see also his discussion in Analytical bibliography : an alternative prospectus). This traces the principle of bibliographic fingerprinting to an article by Falconer Madan in 1893, but its current form dates from the 1970s and the advent of computing in librarianship. There is no standard as such, but Harris describes and compares three systems in use today: the LOC fingerprint used mainly in Italy and Germany; the STCN fingerprint used in the Netherlands and Belgium; and the Bibliographical Profile (not discussed here) developed by Douglas J. Osler for legal bibliography. The LOC fingerprint uses line endings while the other two use the position of signature-marks.
Project LOC was the London-Oxford-Cambridge collaboration of the British Museum, the Bodleian Library, and Cambridge University Library, which piloted a (computerized and still unrealized) union catalogue of books printed before 1801. The project was initiated in 1968 and issued a report, Computers and early books, in 1974, proposing a method for bibliographic fingerprinting. The detail of the method has evolved slightly since then and the current specification was published in 1984 as Fingerprints = Empreintes = Impronte [PDF].
A brief summary of the method is provided below. (The French text is available as part of the article 'Le système des empreintes'. Databases using this fingerprint formula include EDIT 16 and VD 17, and information on their use of fingerprinting is available in Italian and German, respectively.)
The LOC fingerprint is a 16-character code, divided into four groups of four characters. To construct or verify a fingerprint:
Two pieces of information complete the fingerprint. The source of the 3rd group of characters is denoted by a symbol in round brackets: 3 if from a page numbered 13 or XIII, 7 if from a page numbered 17 or XVII, and C if counted from β. This is followed by the date of the edition, either as presented in the text, or normalized together with a letter in round brackets to indicate its original presentation (e.g. R for roman numerals or A for arabic).
Naturally the full instructions include rules to deal with single sheets, blank leaves, illustrations, columns, marginalia and many other complications and exceptions which will not be discussed here: please consult Fingerprints (the manual and examples) and the Fingerprint newsletter.
An alternative approach is to record which characters appear above signature-marks. This has been used since the early 1970s in the short-title catalogues of the Netherlands and Belgium, the STCN and STCV.
The full specification was published in Studies in bibliography in 1986 as 'The STCN fingerprint'. The STCN also provides a quick introduction, in English, with images, and the Folger Shakespeare Library has blogged another clear illustrated example.
The MARC21 standard places the fingerprint identifier in field 026. Whether the LOC or STCN fingerprint is intended can be identified with $2 fei ("Fingerprints = Empreintes = Impronte") or $2 stcnf.
Another identifier that may be encountered is the notation used in Steele's A bibliography of royal proclamations. It consists of the last word of the first line of text; the word at the lower right-hand corner of the large initial; and the last word of the last complete line of text: see the explanation by Juliet McLaren of the ESTC.
Last updated May 2012Owen Massey McKnight <firstname.lastname@example.org>
Jesus College, Turl Street, Oxford OX1 3DW