346 American Archivist / Vol. 57 / Spring 1994

Perspective Downloaded from http://meridian.allenpress.com/american-archivist/article-pdf/57/2/346/2748663/aarc_57_2_y2340p4025584485.pdf by guest on 01 October 2021

On the Nature of Records Management Theory

MICHAEL BUCKLAND

Abstract: Theory is defined as a view or description of the nature of something. The nature of theory relating to records management is examined, including information re- trieval, the records life cycle, and information policy. Records management theory should not be seen in isolation and need not be unique to records management. Functional, pro- fessional, and educational contexts of records management theory are outlined, with ex- amples. Note: This paper is reprinted, with a few minor changes, from Association of Records Managers and Administrators, Proceedings of the AJRMA International 35th Annual Con- ference, 1990 (Prairie Village, Kans.: ARMA, 1990), pp. 801-13.

About the author: Michael Buckland is professor of library and information studies at the University of California at Berkeley. He has had experience in academic libraries and academic administration and has written about library planning and about the history and theory of information services. His most recent books are Information and Information Systems (1991) and Redesigning Library Services (1992). On the Nature of Records Management Theory 347

VARIOUS OPINIONS ARE POSSIBLE concern- especially since the Second World War. ing records management theory: Logic and physics are unusual in that if you have a view about the nature of some 1. That there is a well-defined and aspect of them, you can generally express established theory governing records your view (your theory) in such a way that

management practice Downloaded from http://meridian.allenpress.com/american-archivist/article-pdf/57/2/346/2748663/aarc_57_2_y2340p4025584485.pdf by guest on 01 October 2021 it can be experimentally tested and, per- 2. That there is no theory guiding re- haps, found to be inadequate. Finding ev- cords management practice idence to support theories is relatively easy 3. That there are a number of theories and of limited utility. It is good-faith effort supporting records management to refute theories that is basic to progress. practice, but they belong to allied Therefore, in those disciplines, it is gener- disciplines or sciences and are used ally expected that theories can be rigor- whenever the need for action ously stated and can be tested by seeking founded on principles manifests it- to refute them. (Typically a theory is a hy- self pothesis or combination of hypotheses that In this paper, I will argue that none of the has withstood some testing.) above is correct. Instead, I suggest that (i) Unfortunately, people have assumed that there is a body of records management the- activities that are not "hard" sciences ory that has not yet been well formulated; should also use the same restricted, spe- (ii) parts of this theory are and will increas- cialized sense of theory. When they look ingly be shared with other fields; and (iii) for this sort of rigorous sort of theory and much of it is, perhaps, not yet properly rec- refutation in the messy world of human ac- ognized as theory. But first we need to clar- tivities, such as the provision and use of ify our terms. In particular, before information services, they do not find addressing the question "Is there a records any—predictably!—and are deterred by management theory?" or, better, "What is this narrow preconception with rigorous, records management theory?" the prior formal theory from noticing theory that is question, "What is theory?" needs to be appropriate to the subject. (For an example addressed if we are not to waste our time. of an unsuccessful, misformulated search for theory, see Boyce and Kraft.2) Aristotle What Is "Theory"? knew better: "It is a mark of the educated man and a proof of his culture that in every The original meaning and underlying subject he looks for only so much precision sense of the word theory is a view of, or as its nature permits."3 perspective on, something. In its origins, There should be no rigid dichotomy of the word theory is related to the word the- "theory" versus "practice," since some ater.} More generally, theory is someone's view of what is involved—in effect the- view or description of the nature of some- ory-—underlies both. Nevertheless, the thing. In this general sense there is theory thoughtful practitioner in records manage- of anything that you can describe the na- ment, as in other useful fields, is generally ture of. One problem is that intelligent discus- sion of theory in most areas has been side- tracked by the high social prestige of the 2Bert R. Boyce and Donald H. Kraft, "Principles mathematical and experimental sciences, and Theories in Information Science," Annual Review of Information Science and Technology 20 (1985): 153-78. 'Aristotle, The Ethics of Aristotle: The Nichoma- ^Oxford English Dictionary, 2nd ed. (Oxford: Clar- chean Ethics Translated (Harmondsworth, U.K.: Pen- endon Press, 1989), vol. 7, p. 902. guin, 1955), 27-28. 348 American Archivist / Spring 1994

faced with choosing between formal, rig- ities that could be construed as evi- orous, "respectable" theory that seems di- dence of discrimination? How does a vorced from messy everyday realities and university find documents that might less formal "theory" that does reflect, show its contributions to the well- more or less, the untidy reality encountered being of local industry? in practice but does not seem much like Information retrieval can include Downloaded from http://meridian.allenpress.com/american-archivist/article-pdf/57/2/346/2748663/aarc_57_2_y2340p4025584485.pdf by guest on 01 October 2021 theory.4 The challenge for the thoughtful, up to three functions: (i) physically and especially for academics, is to seek to fetching a copy of a document; (ii) bridge the gap: to develop formal theories locating where some known docu- that are more realistic and to develop re- ment is currently housed; and (iii) alistic views that are framed in terms of identifying which documents, if any, general principles. fit some description. A well-defined records series is a start, but more is How Trivial Is Records Management? likely to be needed. One enters the large and complex area of indexing, Once we back off from the absurdity of classification, thesauri, data diction- assuming that records management theory aries, and so on. should have the characteristics of theory of 2. The records life cycle. A more effi- a formal science, the question changes cient and more cost- effective per- from whether there is a theory to what we formance of any kind of records are theorizing about. More bluntly: how management service is likely if the trivial or how complex is records manage- records manager becomes involved ment? in the full life cycle of records. But If we considered records management to involvement in the full cycle entails be only the numbering, shelving, fetching, involvement in a wide variety of is- and reshelving of pieces of paper, then our sues: form design and management, view of it—our theory—will also be rather human factors and workflow, reten- trivial. No matter how economically valu- tion schedules and appraisal for ar- able that activity may be, the theory is not chival retention, information likely to be very interesting. technology, and vital records protec- But if we view records management as tion. management of access to the working re- 3. Information policy. What (and who) cords of an organization, then some more determines who should and who complex aspects arise, of which I will men- should not have access to records? tion three examples: This involves legal and ethical issues 1. Information retrieval. Fetching Pur- arising from freedom of information chase Order no. 1234 is straightfor- and privacy legislation, trade secrets, ward, a matter of looking up the and more. These matters are compli- location and physically transporting cated, contradictory, and of great im- it. But how do you identify what re- portance, especially for records cords exist that would meet some ill- management in the public sector. For defined need? What about all example, should the public have ac- documents relevant to a firm's activ- cess to some or all of the personnel records of public officials, or do the privacy rights of such officials take precedence? When, and on what "Donald A. Schon, The Reflective Practitioner: How Professionals Think in Action (New York: Basic grounds, can working drafts of doc- Books, 1983), 42^5. uments be kept confidential? On the Nature of Records Management Theory 349

These three examples (information re- of theory is involved in studying ostriches. trieval, records life cycle design, and in- I doubt, however, that much is unique to formation policies) are sufficient to ostriches. Rather, we should expect variant indicate that if you view records manage- forms or special cases of theories also rel- ment broadly as having to do with access evant to other birds and or to other animals to the working records of an organization, of the same habitat. One can be an ostrich Downloaded from http://meridian.allenpress.com/american-archivist/article-pdf/57/2/346/2748663/aarc_57_2_y2340p4025584485.pdf by guest on 01 October 2021 then one can hardly avoid viewing the na- scientist but be suspicious of someone who ture of the field as practically and concep- insists that there is an Ostrich Science that tually complicated—as actually or is somehow separate from and independent potentially rich in theory of one kind or of other disciplines. Somebody's head may another. be in the sand.

Ostrich Science: Or, What Is Unique to Theoretical Contexts of Records Records Management? Management Theory Another problem of theory in profes- By "theoretical context" I mean a sional practice is the desire to identify a broader area of theory into which records theory or science that is unique to that oc- management theory fits. There can be cupation. This, too, is unwise. many such theoretical contexts. I will men- The issues and principles of who should tion just three—functional, professional, have access to records is both a legal issue and intellectual. and a records management issue. The life- Functional context. Records manage- cycle concept is common to both archives ment serves a firm or other organization. and records management. Indexing and Therefore, the mission of a records man- classification schemes are also concerns of agement service needs to be related to and librarianship, museology, database man- supportive of the mission of the organiza- agement, and other areas. Does this matter? tion as a whole. The role of the records Yes, it does matter in a positive way be- management program needs to be articu- cause it is beneficial that other occupational lated with the roles of other parts of the groups also have a vested interest in tack- organization: the roles, needs, and ling the same or related problems. What is activities of employees needing access to needed is communication and collabora- records; the roles of other information-re- tion, not self-imposed professional apart- lated programs, such as Management ness. Information Systems, data processing, the I would not regard the provision of ac- library, and the archives. In other words, cess to the records of an organization as records management theory is likely to be unimportant or easy. Status and respect sterile or incomplete unless it is related to come from doing a difficult and important a view of the organization as a whole. task well, not from insisting that one is not More than academic curiosity is involved. an archivist, not a librarian, not whatever. If we are to assert that the role of records Denning the scope of records management management is important, then we need to also defines by implication what sorts of have views on how a good records man- theory would be relevant. The more ideas agement program contributes to and sup- that can be derived from other fields, the ports the effectiveness of the organization. better the records management service will Professional context. Records man- be. agement can be seen as one member of a Consider a scientist studying an os- family of retrieval-based information sys- trich—an Ostrich Scientist. Doubtless a lot tems. An Italian is not the same as a Briton, 350 American Archivist / Spring 1994

but they are both Europeans and share mation resource management. In 1923 he some things in common as Europeans. Ar- wrote, somewhat prematurely, that admin- chives, libraries, records management pro- istrative documentation was a branch of grams, and corporate databases are not the "general documentation," which was born same, but they are all retrieval-based in- from the convergence and then the fusion formation services and so have some fea- of bibliography, archives administration, Downloaded from http://meridian.allenpress.com/american-archivist/article-pdf/57/2/346/2748663/aarc_57_2_y2340p4025584485.pdf by guest on 01 October 2021 tures in common as well as differences that and even museology.9 make them uniquely different.5 Otlet was a major figure in his time but Seeing records management as a mem- is now little known. His writings are only ber of a family of retrieval-based infor- now beginning to appear in English.10 mation systems is not new. This view was However, the idea that there might be the- adopted in the "Documentation" move- oretical and practical benefit in seeing re- ment early this century, especially in the cords management (and other sorts of writings of Paul Otlet (1868-1944).6 What information services) as part of a family of we now call records management, Otlet related information professions and in called "administrative documentation," adopting a comparative approach has re- and he saw it as an important specialty in ceived attention recently in North Amer- administration. He also included what we ica.11 now call Management Information Sys- Intellectual context. Another kind of tems as part of administrative documenta- context is what could be called the intel- tion. Otlet wrote a number of papers on the lectual context. Where did ideas in records subject,7 and his ideas were implemented management come from and where have to a limited extent in Europe.8 records management ideas gone to? Two Otlet viewed archives, bibliography, li- examples can illustrate this point. One braries, museums, and records manage- source of records management ideas is ar- ment as parts of the broad field of chival theory. Ernst Posner teased archi- "Documentation" or "General Documen- vists for their penchant for tracing their tation" or, as we might say today, infor- roots to ancient Assyria.12 Perhaps that an- cient archival activity is more accurately described as ancient records management.13 Luciana Duranti has recently summarized 5Michael K. Buckland, "Records Management in Its Intellectual Context: Experience at Berkeley," ARMA Quarterly 16 (1982): 26-28, 30; Michael Buckland, Information and Information Systems (New 'Otlet, Manuel, 13. York: Praeger, 1991). 10Paul Otlet, International Organization and Dis- 6W. Boyd Rayward, The Universe of Information: semination of Knowledge: Selected Essays, translated The Work of Paul Otlet for Documentation and In- and edited by W. B. Rayward, FID 684 (Amsterdam: ternational Organization, FID 520 (Moscow: VINITI, Elsevier, 1990). 1976); W. Boyd Rayward, "Otlet, Paul-Marie-Ghis- "C. S. Seibell, "Records Management in Its Intel- lain," in ALA World Encyclopedia of Library and In- lectual Context," Records Management Quarterly 21 formation Sciences, 2nd ed. (Chicago: American (1987): 3-14, 42, 49; J. Michael Pemberton and Ann Library Association, 1986), 626-28. Prentice, eds., Information Science: The Interdiscipli- 'Paul Otlet, Manuel de documentation administra- nary Context (New York: Neal-Schuman, 1990); tive, IIB publ. 137 (Brussels: Institut International de Buckland, Information and Information Systems. Bibliographie, 1923); Paul Otlet, Sur les possibility 12Ernst Posner, "Some Aspects of Archival Devel- pour les entites administratives d'avoir a tout moment opment Since the French Revolution," American Ar- lew situation presentee documentairement, IIB publ. chivist 3 (July 1940): 159-72; reprinted in M. F. 162 (Brussels: Institut International de Bibliographie, Daniels and T. Walch, eds., A Modern Archives 1930); Paul Otlet, Traite de documentation (Brussels: Reader (Washington, D.C.: National Archives and Editiones Mundaneum, 1934; reprinted Liege, Bel- Records Service, 1984), 3-14. gium: Centre de Lecture Publique de la Communaute "Bill Walker, "Records Managers and Archivists: Francaise, 1989), 350-55. A Survey of Roles," Records Management Quarterly 8Rayward, Universe of Information, 160, 184. 23 (1989): 18-20, 45, 49. On the Nature of Records Management Theory 351

that aspect of the intellectual ancestry of Comparison 14 records managers. One does not get very far trying to de- In the other direction, it is less widely fine and describe things in their own terms. known that fashionable current ideas in It is comparison that is the most basic in- computer science concerning hypertext and tellectual activity. Comparing the nature of Downloaded from http://meridian.allenpress.com/american-archivist/article-pdf/57/2/346/2748663/aarc_57_2_y2340p4025584485.pdf by guest on 01 October 2021 hypermedia have part of their roots in re- records management with other things is a cords management. A major inspiration in necessary condition for progress in devel- that field has been a visionary essay, "As 15 oping a view—a theory—of records man- We May Think," by Vannevar Bush. agement. To understand records manage- This fantasy, drafted in 1939, was based on ment better we need to devote more his experience with a project funded by attention to comparing records manage- Eastman Kodak and National Cash Regis- ment with other related activities. Only by ter to develop a faster method of retrieving means of a thorough understanding of sim- microfilmed business records using a ma- ilarities and dissimilarities can we expect chine known as a "rapid selector." Rapid to develop a deep understanding of the na- selectors were essentially a precomputer ture—a theory—of records management. version of Computer-Assisted Retrieval (CAR), in which the index entries were re- Summary corded as patterns on the microfilm itself 1. A theory is a view of the nature of next to the images of the documents. The something. index entries were searched using a pho- toelectric cell to detect matches between 2. Theory should not be more formal the index pattern sought and the index pat- than the nature of the phenomena to be de- terns on the microfilm. Rapid selector tech- scribed will permit. Overly formalized the- nology was initially developed in the late ories cease to describe reality in a useful 1920s by Zeiss Ikon in Germany and later, way. from the 1930s to the 1960s, in the United 3. Any view of records management as States, usually to assist in the locating and being more than the filing and fetching of retrieving of microfilmed corporate records documents will provide a basis for nontriv- or technical reports.16 ial theory. 4. Few parts of the theory in records management are likely to be unique to re- "Luciana Duranti, "The Odyssey of Records Man- cords management, though the detailed ap- agement," ARMA Quarterly 23 (July 1989): 3-11; plication and the overall blend will be. and (October 1989): 3-11. Shared theory is advantageous because it "Vannevar Bush, "As We May Think," Atlantic Monthly 176 (1945): 101-06. means that others share a motivation to "Emanuel Goldberg, "The Retrieval Problem in seek improved understanding. Photography (1932)," translated and notes by M. 5. Records management can be viewed Buckland, Journal of the American Society for Infor- mation Science 43 (May 1992): 295-98; Michael in relation to different contexts, e.g., func- Buckland, "Emanuel Goldberg, Electronic Document tional, professional, and intellectual. View- Retrieval, and Vannevar Bush's Memex," Journal of ing each such broader context yields a the American Society for Information Science 43 (May 1992): 284-94; S. N. Alexander and F. C. Rose, body of theory—or an opportunity to de- "The Current Status of Graphic Storage Techniques: velop one. Their Potential Application to Library Mechaniza- tion," in Libraries and Automation. Proceeding of the Conference on Libraries and Automation, Warrenton, Stevens, Microphotography: Photography and Pho- 1963, edited by B. E. Markuson (Washington, D.C.: tofabrication at Extreme Resolution, 2nd ed. (New Library of Congress, 1964), 111-40; and G.W.W. York: Wiley, 1968).