concepts concerning evidence and recordness with creating institutions' own
self-interest in protecting themselves legally and ethically. It thus sanctions a
potentially powerful strategy to get archival issues addressed by record creators at
the front end of the records continuum, which is essential if an archival record
is to survive in the electronic era. Yet with its heavy focus on institutional and
official records in its formulation and examples, the accountability approach
also carries with it, as some Australian advocates are now beginning to
recognize,70 a danger of rendering into two camps the administrative and
cultural roles of archivists, and thus of devaluing archives' role as a bastion of
national culture and societal memory in favour of narrower, strictly legal
accountabilities. The same threat is implicit in the emphases of Canada's neo-
Jenkinsonians and, as will be seen, in the formulations of some electronic
records theorists.
"Reinventing archives": electronic records and archival theory
The revitalization or rediscovery of provenance has also been motivated by the
many challenges posed to archivists by electronic records. Discussion about such
records is increasingly dominating the professional discourse, and is leading to
exciting new conceptual insights, as well as to new strategies and practices.71
Despite significant contributions by Canadians and Australians, the leadership
in the electronic records discourse belongs to the United States, especially to
David Bearman.72
The early impact of electronic records, or machine-readable records as they
were called, was not quite so promising, however. In panic over the then
relatively new technology, some commentators in the 1970s and early 1980s
advocated that archivists should stop being archivists, and instead become com
puter specialists or information managers in order to cope with this challenging
new medium. In what I have called the "first generation" of electronic records
archives, there was also a strong emphasis on information content over
provenancial context, on library cataloguing over archival description, on one
time, one-shot statistical datafiles over continuing and continually altering rela
tional databases and office systems, and on treating electronic datafiles as
discrete and isolated items rather than as part of the comprehensive,
multi-media information universe of the record creator.73 Such approaches by
the [41] pioneering, first-generation of electronic records archivists are perfectly
understandable: the only working models available to them had been created by
data librarians dealing with social science datafiles bearing the above characte
ristics. This changed by the mid-1980s when new information technology featu
ring relational databases became the norm in business, universities, and
government. The archivally valuable computerized data in such large social and
economic programmes' relational systems are often added, revised, or deleted
almost every second. Outside the world of such databases, wherein information
is at least structured logically, there is the automated office, where text, data,
graphics, images, and voice are converted into electronic formats, and even com
bined into "compound" or "smart" multimedia documents. All these new and
complex computerized formats, until controlled, standardized, and linked to
business processes, threaten decision-making accountability and the long-term
corporate memory of record creators, especially when joined with a telecommu
nications revolution affecting the transmission and interconnectivity of this
electronic information. Even more, these new formats threaten the very possibi
lity that archives can continue as vibrant institutions able to maintain such
records in their full context or functionality over decades and centuries. If an
electronic document has only a transient existence as a "virtual" composite or
fleeting "view" on the computer screen of randomly stored information created
by the different commands of different users in different organizational
structures for different purposes, how does any one accountable institution pre
serve reliable evidence of specific transactions? What is the functional context
of such transient and disjointed data? Whither provenance? Electronic records,
much like the earlier thinking of Peter Scott, bring archivists to the era of virtual
archives and virtual records, where the physical record and its arrangement, so
central to much traditional archival discourse in this century, is now of rather
secondary importance compared to the functional context in which the record is
created, described by its creator, and used by its contemporaries. Such revolutio
nary changes suggested by the electronic record have led archival theorists, such
as Sue McKemmish of Australia, to ask, "Are records ever actual?"74
Answers to these fundamental challenges are beginning to come. Archivists
are now perceiving that a world of relational databases, of complex software
linkages, of electronic office systems, of hypermedia documents, of multi-layered
geographical information systems, is, when all the high-technology rhetoric is
ARCHIEFWETENSCHAP
70 Upward and McKemmish, "Somewhere Beyond Custody," pp. 145-46, and Frank Upward in Archival
Documents, p. 43. For an interesting attempt to break out of this mode, see Sue McKemmish, "Evidence of
Me," Archives and Manuscripts 24 (May 1996), pp. 28-45.
71 Regarding the latter, a great number of strategies and practices have evolved, or at least are being recom
mended to archivists, to deal with electronic records, although there is no space to discuss them in this
essay devoted to conceptual discourse rather than practical methodologies—which is not to say that those
methodologies do not generate their own controversies, such as whether archives need acquire physically
all electronic records in order to ensure their authenticity or the appropriate linkage of creator metadata
and archival contextualized authority files. The best single source for strategic approaches to electronic
records remains Margaret Hedstrom, ed., Electronic Records Management Program Strategies (Pittsburgh,
1993), which offers case studies, with analyses of critical factors of success and failure, of electronic
records programmes at international (2), national (4), state (4), and university (1) levels, with an overall
assessment, and an extensive (59 pages) annotated bibliography compiled by Richard Cox for readers to
continue their explorations. See also David Bearman, "Archival Strategies," paper discussed at the SAA
1994 conference, and forthcoming in the American Archivist.
72 For a sample, see David Bearman's works cited throughout these notes; ten of his essays are now collected
into David Bearman, Electronic Evidence: Strategies for Managing Records in Contemporary Organizations
58
TERRY COOK WHAT IS PAST IS PROLOGUE
(Pittsburgh, 1994); and a wide range of his commentary and analysis appears throughout all the issues
of Archives and Museum Informatics, which he edits. The other principal American voice has been Margaret
Hedstrom: see her ground-breaking SAA Manual, Archives and Manuscripts: Machine-Readable Records
(Chicago, 1984); and more recently "Understanding Electronic Incunabula: A Framework for Research on
Electronic Records," American Archivist 54 (Summer 1991), pp. 334-54; "Descriptive Practices for
Electronic Records: Deciding What is Essential and Imagining What is Possible," Archivaria 36 (Autumn
1993), 53-62; and with David Bearman, "Reinventing Archives for Electronic Records: Alternative Service
Delivery Options," in Hedstrom, Electronic Records Management, pp. 82-98. An early pioneer for electronic
archiving was also American: Charles M. Dollar; see his "Appraising Machine-Readable Records," (origin
ally 1978), in Daniels and Walch, Modern Archives Reader, pp. 71-79; and, more recently, Archival Theory
and Information Technologies: The Impact of Information Technologies on Archival Principles and Methods
(Macerata, Italy, 1992); and "Archival Theory and Practices and Informatics. Some Considerations,"
in Bucci, Archival Science on the Threshold, pp. 311-28. An early Canadian voice was Harold Naugler,
The Archival Appraisal of Machine-Readable Records: A RAMP Study With Guidelines (Paris, 1984).
73 Cook, "Easy to Byte, Harder to Chew: The Second Generation of Electronic Records Archives," pp. 203-8.
74 For a stimulating discussion, see Sue McKemmish, "Are Records Ever Actual?," in McKemmish and Piggott,
The Records Continuum, pp. 187-203.
59