Library Philosophy and Practice 2011

ISSN 1522-0222

Records Management Theory's Dilemma: What is a Record?

Joshua Finnell
Humanities Librarian
Denison University
Granville, Ohio, USA


In the Euthyphro, Socrates encounters Euthyphro outside the courts where both are enroute to trials pertaining to piety. Euthyphro is prosecuting his father for murder, while Socrates is on trial for engaging in impious acts. In the course of their conversation Socrates questions whether prosecuting one's own father is impious, especially after Euthyphro boldly claims his act is not (Plato, 2005, p. 4). This inevitably leads Socrates to ascertain the correct definition of piety, so that he can accurately separate pious acts from non-pious acts. In his dialogue with Euthyphro, Socrates is essentially searching for a universal definition of piety so he may recognize them pious activities where they occur.

A familiar dialogue is taking place in the professional literature of records and information management. Since the early 1990s, the theoretical foundation of a records management theory has been constructed on convergence (Pemberton & Nugent, 1995; Walters, 1995; Zawiyah M Yusof & Robert W Chell, 2002). Michael Buckland speaks to this collaborative approach in pointing out the lack of uniqueness in records management theory when he writes, "The issues and principles of who should have access to records is both a legal issue and a records management issue. The life-cycle concept is common to both archives and records management. Indexing and classification schemes are also concerns of librarianship, museology, database management, and other areas" (Buckland, 1994, p. 349). However, while certain concepts are shared across disciplines, arguably the most foundational definition is the most divergent: a record. Each discipline (Archival Science, Library Science, Computer Science) defines the term record in its own way. Unfortunately, much like Euthyphro, records managers at all levels have difficulty espousing a universal definition of the term, while claiming sole responsibility for the authority, organization, authenticity, and sustainability of records.

In this paper, I wish to explore Euthyphro's definitional responses to Socrates query as a metric for exploring the problematic nature of defining a record. Within each response, I will examine parallels problems faced by records manager theorists in positing a definition. This is purely an intellectual exercise, but one that ultimately wishes to contribute to the development of a records management theory. In the words of Michael Buckland, "One does not get very far trying to define and describe things in their own terms. It is comparison that is the most basic intellectual activity. Comparing the nature of records management with other things is a necessary condition for progress in developing a view - a theory - of records management" (Buckland, 1994, p. 351).

"Piety is doing what I'm doing here today, namely, prosecuting my father for murder"

The first definition that Euthyphro posits is that, "piety is doing what I'm doing here today, namely, prosecuting my father for murder"(Plato, 2005, p. 5). This statement provides an example of piety, but does little in the way of providing a working definition. Socrates denies this first definition because it circles the definition of the piety without providing a definition of the concept. Records are often defined in this way. Records managers in an academic setting might say a transcript is a record. In a corporate setting, a records manager would say that a tax return is a record. While these statements provide the variability of records in different settings, they fail to provide an adequate definition of a record. What differentiates a receipt from a record? Is an email a record? Are electronic records different than physical records? It would be fair to say that receipts and emails are types of records, but this simply begs the question. Definition by example is not a sufficient definition of a record.

"Pious acts are loved by the gods"

In his second definition, Euthyphro appeals to authority in his definition, "pious acts are loved by the gods" (Plato, 2005, p. 7). As opposed to the first definition, this effort provides a working definition. However, in order to determine which acts are pious one must have knowledge of what the gods love. If we were to apply this definition to records, we might state this definition as "records are defined by professional organizations." While Socrates may not have had access to the knowledge of the gods, we certainly have access to the glossaries of professional organizations. If I wish to know if a receipt is a record, I can simply check the Association of Records Managers and Administrators (ARMA). According to ARMA:

Records are the evidence of what the organization does. They capture its business activities and transactions, such as contract negotiations, business correspondence, personnel files, and financial statements. Moreover, Records come in many formats:

  • Physical paper in our files, such as memos, contracts, marketing materials, and reports
  • Electronic messages, such as e-mail content and their attachments and instant messages
  • Content on the website, as well as the documents that reside on PDAs, flash drives, desktops, servers, and document management systems
  • Information captured in the organization's various databases (ARMA, 2010, para. 1).

This definition certainly answers the critiques posed in response to the first definition, namely whether electronic records are different than physical records. Moreover, it also provides a working definition of records as evidence in a business setting. However, as pointed out earlier, the records management theory is built upon the concept of convergence among many fields, computer science, library science, and archival science to name just a few. These fields exist in many settings, both corporate and academic. So, if we are to accept that records are defined by professional organizations, coupled with the concept that records management theory is one of convergence, then it behooves us to explore how allied fields define the concept.

In the field of Computer Science, a record is defined as "a collection of related data items or fields. A record usually forms part of a file. Records may be of either fixed or variable length; variable records require a separator at the end of the field, in order that the end of the record can be detected by the computer" (ebrary, Inc, 2005, p. 443). This definition doesn't say anything about evidence. Rather, it defines a record as a part of another entity, a file. Computer Science, though an allied field, can be quite technical in its definitions. Therefore, it would be fair to examine Computer Science in its blended environment of Library and Information Science. According to the Dictionary for Library and Information Science, a record is:

An account of something, put down in writing, usually as a means of documenting facts for legal or historical purposes. Also, to make such an account. In a narrower sense, a formal document in which the content is presented in a named set of standardized data elements treated as a single unit, for example, a certificate, deed, lease, etc. In archives, a document created or received, and subsequently maintained, by an institution, organization, or individual in the transaction of official or personal business or in fulfillment of a legal obligation (Reitz, 2004, p. 722).

This definition has more overlap with ARMA, in that records are referred to as facts, and stand for an account of "something." Moreover, as a more general definition it points to similarities in convergent fields, such as archives. A cross-reference, however, points to narrower definitions of the term in the form of bibliographic records, catalog records, and sound recordings. These additional terms point to definitions that are specific to the field of Library Science, leaving one to ponder if the general definition really provides the essence of the concept. Curiously, in the exact same dictionary, the term receipt is defined as, "A written document stating that something has been received, usually in exchange for payment of an amount noted in the acknowledgment. Sometimes found among personal and family papers, receipts may contain information of historical importance" (Reitz, 2004, p. 720). This begs the question posed in the first definition, "what differentiates a receipt from a record?" Archives often contain historical ledgers, reflecting the accounting practices and transactional items of an era. However, is a receipt a record in the same way that an academic transcript is a record? As a convergent field in the theory of records management, it may be helpful to see how archivists define a record.

The Society for American Archivists provides a robust definition consisting of seven elements:

  1. A written or printed work of a legal or official nature that may be used as evidence or proof; a document.
  2. Data or information that has been fixed on some medium; that has content, context, and structure; and that is used as an extension of human memory or to demonstrate accountability.
  3. Data or information in a fixed form that is created or received in the course of individual or institutional activity and set aside (preserved) as evidence of that activity for future reference.
  4. An instrument filed for public notice (constructive notice); see recordation.
  5. Audio · A phonograph record.
  6. Computing · A collection of related data elements treated as a unit, such as the fields in a row in a database table.
  7. Description · An entry describing a work in a catalog; a catalog record (SAA, 2010).

This definition encompasses all the previous definitions in that archivists define a record as a data element, evidence, and an institutional activity. While this definition underscores the convergent elements of a records management theory it also introduces a profound philosophical statement that may or may not be shared among the other fields – a record as an extension of human memory. While it is beyond the scope of this paper to explore the merits of the definition, suffice it to say Socrates would have been delighted at this statement.

As can be seen amongst these professional organizations, overlap exists in defining a record. At the same time, each organization adds a variation to the definition. In Computer Science, a record is not thought of in terms of evidence. While, librarians may view a record as a statement of fact, records management professional and archivists lean towards the interpretative mode of evidence. Moreover, archivists include an element of intentionality to a record, that it be preserved for future use. If we accept this second definition, "records are defined by professional organizations," then we find ourselves with the same concerns Socrates faced with piety being defined as that which the gods love: "What if the gods don't agree" (Plato, 2005, p. 10). The different definitions that exist among these organizations make the case that an archivists might not agree that a cataloging record is the same as a historical diary. To an archivist, the cataloging record lacks intentional preservation, and is more of a finding aid. This definition results in an object of information being considered a record by one organization (librarians) and not by another (archivists). Definition by appeal to authority is not a sufficient definition of a record.

"Pious acts are loved by all the gods"

In an attempt to overcome disagreement among the gods as a disqualifier, Euthyphro amends the second definition with the addition of group consensus. His third attempt defines pious acts as "those acts that are loved by all the gods" (Plato, 2005, p. 11). Of course, within the context of his current situation, Euthyphro believes that all the gods love the idea that he is prosecuting his father for murder. In terms of records management, the varying definitions between professional organizations do not give us a clear picture of consensus. However, professional organizations are often more concerned with the practice as opposed to the theory of records management. Practice, often a derivative of theory, is sometimes confined in its ability to give a representational view of the field and its approach to records management. Scholars, on the other hand, can provide a meta-analysis of the field without being confined to a specific trade or practice. To rephrase Euthyphro's definition in terms of records management, records are those documents that are identified by all the scholars in the field. Unlike gods, access to the thoughts of academics is slightly more accessible. Therefore, finding consensus among scholars should provide us with a working definition of a record.

David Bearman is a recognized scholar as an information strategist in the fields of archives and museums. He served as the Deputy Director of the Office of Information Resources Management at the Smithsonian Institution and founding partner of Archives & Museum Informatics in Toronto. He has developed national information systems, policies, and standards for several countries. His scholarship in the field of electronic records is of particular note. Having spent the last thirty years in the field of records management, his writing provides insight into the historical view of records within the scholarly domain. In a 1997 working meeting of the Electronic Records Research Committee, Bearman summarized the archival view of the record,

Archivists distinguish between records and information or data; not all information or data is a record. Records are that which was created in the conduct of business and communicated between parties to that business; some archivists believe records must be "set aside" in the course of business to be considered a record. In any case, the fact of being transacted in a particular business context is crucial to record, thus an adequate record will contain evidence of the context of its creation. The consensus, largely developed since 1990, is that…Records are evidence of transactions (relationships of acts), means of action and information about acts (Bearman & Trant, 1997, para. 9).

Bearman stresses the evidence of transaction as an accepted definition of records.

Gregory S. Hunter is a professor in the Palmer School of Library and Information Science at Long Island University and president of Hunter Information Management Services, Inc. He is both a certified records manager and a certified archivist. He served as principle archivist and records manager on the grant to build an electronic records archive for the National Archives and Records Administration and is co-inventor of four patents in the area of digital preservation. His research is largely focused in the area of archives and digital preservation, and reflected in his work as series editor of The Archivist's and Records Manager's Bookshelf. Within this publication, Hunter defines a record through its characteristics,

In both electronic and paper environments, a record has three characteristics:

  • Content: that which conveys information (text, data, symbols, numerals, images, sound, and vision)
  • Structure: the appearance and arrangement of the content (relationships between fields, entities, language, style, fonts, page and paragraph breaks, links, and other editorial devices)
  • Context: the background information that enhances understanding of technical and business environments to which the records relate (application software, link to function or activity, provenance information) (Hunter, 1997, p. 242).

While providing a more robust definition, with a rubric for identification, Hunter's definition aligns with Bearmans' in that a record conveys information. In this particular chapter, Hunter is speaking to the profusion of records in an electronic as opposed to a paper environment. Moreover, he is concerned with archival records in particular. While he senses a shift in roles for archivist in terms of the record life cycle, he concurs with Bearman that records are defined as evidence of a transaction.

Luciani Duranti is Chair and Professor of Archival Studies in the School of Library, Archival and Information Studies at the University of British Columbia. She has published extensively in the field of records management theory, specializing in accuracy, reliability, authenticity, authentication, and preservation. She also serves as director of the International Research on Permanent Authentic Records in Electronic Systems (InterPares) project. This research is focused on the long-term preservation and authentication of born-digital records. Duranti defines a record as "documents (i.e., recorded information) created (i.e., made or received and set aside for action or reference) by a physical or juridical person (i.e., an individual or an entity capable of rights and duties) in the course of a practical activity, as a means and instrument for it (Duranti, 2001, p. 272). With the added addition of authority, Duranti, like Bearman and Hunter, also espouses the definition of a record as transactional.

Geoffrey Yeo has worked in the field of archive and records management for the last 25 years as a consultant, archivist, and scholar. From 2002 to 2005, he served as director for the Masters Programme in Archives and Records Management in the School of Library, Archive and Information studies at University College London. Currently, he serves on the board of ICARUS, the International Centre for Archives and Records Management Research and User Studies. His research is focused on the relationships between individuals, records, and organizations. With his focus on records within the context of relationships, Yeo defines records as "persistent representations of activities, created by participants or observers or their authorized proxies" (Yeo, 2007, p. 342). While this definition says nothing of transaction, evidence, or authenticity, this representational definition of records is flexible enough to include those criteria without making them necessary to the definition. Yeo summarizes this point of view when he writes,

…[to] emphasize only evidence or information is limiting…undervaluing the complexity of records. Emphasis on evidence is often intended to link recordkeeping to the worlds of law and corporate governance; emphasis on information suggests an alignment to librarianship or computing. A focus on memory perhaps implies an association with history or cultural identity. All these perspectives are valid, but none is comprehensive. The representational view of records is multidisciplinary and embraces a wide spectrum of understanding (Yeo, 2007, p. 343).

Unlike Bearman, Hunter, and Duranti, Yeo's definition of records reflects the convergence of disciplines upon which records management theory is built. By not privileging one criteria over another, Yeo provides the most flexible definition. However, can it be safely assumed that all scholars in the field of records and information management accept this definition? Without consensus among the scholars, this definition fails. A more robust study would need to be completed to definitively answer this question, but a citation analysis through Google Scholar shows Shepherd and Yeo's Managing Records: A Handbook of Principles and Practice has been cited by over 41 scholarly articles. While Yeo himself makes no claim that his definition will be universally acceptable, the representational definition of records is clearly influencing future research. As a corollary, defining records as "persistent representations of activities, created by participants or observers or their authorized proxies" approximated Euthyphro's criteria of being "loved by all the gods."


Given the complex nature of records within the domain of records management theory, simple and limiting definitions of a record only serve the dominant paradigm from which the definition is espoused. This intellectual exercise ultimately attempted to reveal the hierarchy of thought that exists in approaching the record. Those involved with working with records everyday (administrative assistance, office workers, student assistants) may only define the record by way of example. As expected, when asked to define a record records managers may default to example, "The expense report I just filed is a record." While these definitions give us context, they fail to provide an adequate definition. The definitions provided by professional organization, within which everyday records managers functions, tend to define the record in the context of how they process the records. However, archivists, librarians, and computer scientists may define the record to narrowly, ignoring the multidisciplinary nature of the record and its multiple representations to different users. Disagreement among professional organizations leads one to explore the scholarly community, within which professional organizations align standards and best practices, to find provide a flexible definition. Defining a record as a persistent representation provides both professional organizations and records managers a shared foundation to build a convergent theory of records management.

Ultimately, Socrates accepts Euthyphro's third definition but questions its circular logic, articulating Euthyphro's Dilemma. Socrates asks whether the pious is loved by the gods because it is the pious, or whether the pious is the pious because it is loved by the gods (Plato, 2005, p. 13). The question gets down to the essence of the concept itself. In terms of records management theory, we may rephrase this question as are "persistent representations of activities, created by participants or observers or their authorized proxies" a record because that is the true definition of a record, or is a record a persistent representations of activities, created by participants or observers or their authorized proxies because that is how the scholarly community defines it. It is beyond the scope of this paper to answer Euthyphro's dilemma, but hopefully this analysis has raised questions for further discussion.

Ultimately, if one disagrees that there is a universal definition of a record, then they are bound to a view of records as contextual. If records are contextual, so then is the theory of management applied to them. Hence, records management must be contextual to the organization. Since organizations have different opinions on what constitutes a record, a records management theory by its nature will vary by organization. As a corollary, a theoretical field of records will always exist in silos (archival vs. librarian vs. information science). To answer the most fundamental question is to build a convergent theory. How do you define a record?


ARMA. (2010, July 20). RIM Fundamentals, Retention Schedules, Electronic Records - ARMA International. ARMA International. Retrieved July 20, 2010, from http://www.arma.org/rim/fundamentals/index.cfm

Bearman, D., & Trant, J. (1997). Electronic records research: working meeting: May 28-30, 1997: a report from the archives community. D-Lib Magazine, 3(7-8).

Buckland, M. (1994). On the Nature of Records Management Theory. The American Archivist, 57(2), 346-351.

Duranti, L. (2001). Concepts, Principles, and Methods for the Management of Electronic Records. The Information Society: An International Journal, 17(4), 271. doi:10.1080/019722401753330869

Ebrary, Inc. (2005). The Hutchinson Dictionary of Computing and the Internet. Abingdon: Helicon Pub.

Hunter, G. S. (1997). Developing and Maintaining Practical Archives: A How-to-Do-It Manual. New York: Neal-Schuman Publishers.

Pemberton, J. M., & Nugent, C. R. (1995). Information Studies: Emergent Field, Convergent Curriculum. Journal of Education for Library and Information Science, 36(2), 126-138.

Plato. (2005). Plato's Euthyphro, Apology, Crito, Phaedo: Benjamin Jowett's Translation. Millis, MA: Agora Publications, Inc.

Reitz, J. M. (2004). Dictionary for Library and Information Science. Westport, Conn: Libraries Unlimited.

SAA. (2010, July 20). Record | Society of American Archivists. Society of American Archivists. Retrieved July 20, 2010, from http://www2.archivists.org/glossary/terms/r/record

Walters, T. O. (1995). Rediscovering the Theoretical Base of Records Management and Its Implications for Graduate Education: Searching for the New School of Information Studies. Journal of Education for Library and Information Science, 36(2), 139-154.

Yeo, G. (2007). Concepts of Record (1): Evidence, Information, and Persistent Representations. The American Archivist, 70(2), 315.

Zawiyah M Yusof, & Robert W Chell. (2002). Towards a theoretical construct for records management. Records Management Journal, 12(2), 55-64.