The Ontology of Documents, Revisited

Three contributions are made to understanding the nature of documents. A survey of definitions of "document" from the last century shows that those definitions which most accurately reflect the ways in which the term "document" is used in practice are typically compound definitions, consisting of two or three elements that each refer to a different function of documents: medium, message, and meaning. Locating documents in E. J. Lowe's four-category ontology results in consideration of documents as universals rather than as particulars. Analysis of B. Smith's theory of document acts suggest that all documents, not just the ones that are involved in declarations, are creative in the special sense that they are generative of quasi-abstract entities of the kind that collectively comprise social reality.

The goal of all of this though is not to draw definite conclusions, but to contribute to a solidification of connections between LIS and literatures that might not previously have loomed large on our collective radar, and to spark conversations about this material, so I shall be more than happy if I manage to do that, and I hope that you shall be somewhat satisfied too.

I
Definitions of "document" The initial touchstone when it comes to any discussion of definitions of "document" is Michael Buckland's seminal and highly-cited 1997 paper (Buckland, 1997).This will be so familiar to everyone, more than twenty years later, that I hesitate to give more than the most cursory review, but I do think it is worth reminding ourselves what Buckland's goal was with this article and what he consequently chose to leave out.
Buckland explicitly poses the question "What is a document?" in the context of a historical discussion of the limits of documentation, as that activity was pursued and developed in the first half of the twentieth century.He asserts (p.804) that such discussion is still "relevant to the clarification of the nature and scope of information systems," given the way documentation has developed in the second half of the twentieth century, but (apart from brief consideration of "contemporary" definitions drawing from semiotics) restricts his survey of definitions of "document" to those emerging from the documentation movement prior to the mid-1960s.
So Buckland begins by considering the oeuvre of Paul Otlet (1868Otlet ( -1944)), the Belgian visionary well-known to all of us, who with Henri La Fontaine founded the Institut international de bibliographie in 1895-which in 1931 became the Institut, and in 1937 the Fédération, internationale de documentation-and who wrote the Traité de documentation, published in 1934.For Otlet, says Buckland, the category of "document" includes not just graphic and written records (i.e., representations of ideas or of objects) but also the objects themselves-"if you are informed by observation of them" (Buckland,p. 805).For example: "natural objects, artifacts, objects bearing traces of human activity (such as archaeological finds), explanatory models, educational games, and works of art" (Buckland,p. 805)-that is, objects "not intended as communication" (p.807).Comparing Otlet's ideas to some of those promoted by modern-day cultural anthropologists and museologists, Buckland further quotes Otlet: "Collections of objects brought together for purposes of preservation, science and education are essentially documentary in character (Museums and Cabinets, collections of models, specimens and samples).These collections are created from items occurring in nature rather than being delineated or described in words; they are three dimensional documents."(Otlet, 1920;translated in Otlet, 1990; cited by Buckland, 1997, p. 807)

Year
Author Definition 1935 Schürmeyer (trans.Buckland, 1997) "any material basis for extending our knowledge which is available for study or comparison" 1937 Institut international de coopération intellectuelle / Union française des organismes de documentation "Any source of information, in material form, capable of being used for reference or study or as an authority.Examples: manuscripts, printed matter, illustrations, diagrams, museum specimens, etc." 1942 Donker Duyvis (trans.Voorhoeve, 1964) "the repository of an expressed thought" 1951 Briet (trans.Buckland, 1997) "evidence in support of a fact"; i.e., "any physical or symbolic sign, preserved or recorded, intended to represent, to reconstruct, or to demonstrate a physical or conceptual phenomenon" Table 1.Some pre-1966 definitions of "document" cited by Buckland (1997).
Some further definitions of "document" cited by Buckland, all except the last focusing on the supposed materiality of documents, are listed in Table 1 (emphases added).The last is from Suzanne Briet (1894Briet ( -1989))-the French librarian who co-founded the Union française des organismes de documentation in 1931 and published Qu'est-ce que la documentation? in 1951, and who is celebrated in library schools around the world for her recognition that a photo of a star, a stone in a museum, an antelope in a zoo-all have "become physical evidence being used by those who study" them (Buckland, 1997, p. 806) and therefore can be considered to be documents.
Especially since the publication of Ron Day's masterful translation and analysis of Qu'est-ce que la documentation?(Briet, 2006), much has been written about Briet's supposedly structuralist and proto-semiotic approach to the definition of "document."Buckland infers from Briet four "rules for determining when an object has become a document" (p.806): 1. the object must be material (this despite the translation of Briet: "any physical or symbolic sign ...," emphasis added); 2. it must have been someone's intention that the object is to be treated as evidence; 3. the object must have been processed in some way; and 4. the object must be perceived as a document.
Buckland then cites Day as specifying that it is "indexicality-the quality of having been placed in an organized, meaningful relationship with other evidencethat gives an object its documentary status" (Buckland, 1997, p. 806).
In contrast to the emphasis on the materiality of documents that pervades his prior discussion, Buckland's conclusions include (p.808) the identification of an "evolving" notion of "document" that has "increasingly emphasized whatever function[s] as a document rather than traditional physical forms of documents."Buckland sees (p.804) a move from a traditional concern with "text and text-like records (e.g., names, numbers, and alphanumeric codes)" to "any phenomena that someone may wish to observe: Events, processes, images, and objects as well as texts," and remarks (p.808) that "The shift to digital technology would seem to make this distinction even more important." Buckland never intended to provide comprehensive coverage, even of the time period to which he limited himself.A few of the definitions that did not make it into his survey, including one from Ranganathan, are listed in Table 2 (emphases added).(Buckland did cite the famous Indian librarian, but the work he chose was from a slightly later date than that of the one given here.)

Year
Author Definition 1907 Institut international de bibliographie (trans.Weitenkampf, 1908) "anything which represents or expresses, by the aid of any signs whatever (writing, image, diagram, symbols) (1) all written material, whether handwritten, printed, or typed; (2) all painted, drawn or engraved material; (3) all sound or voice recordings; (4) all printed photographs and exposed or printed film, still and motion picture; and (5) all reproductions of the foregoing, by whatever process reproduced." Table 2. Some further pre-1966 definitions of "document" not cited by Buckland (1997).
Over the last century, various suppliers of definitions of "document" have chosen to emphasize the supposed physical, material, or concrete nature of documents.The first of Buckland's (1997, p. 806) four "rules for determining when an object has become a document," inferred from analysis of Briet's (1951) discussion, points to materiality as a necessary condition.(Buckland's translation of Briet's definition-which begins "any physical or symbolic sign ..." (emphasis added)-might seem to contradict this inference, but other authorities cited by Buckland certainly lie in the explicitly materialist camp; see Table 1.)We might call definitions of this kind definitions of document-as-medium, since the idea they promote is of documents as media, vehicles, or channels, for the storage and/or carrying of messages.
At the same time, other definitions (including Briet's) have been constructed so as to emphasize a different kind of essence-not documents' materiality, but instead their informative, evidentiary, or signifying quality.We might call definitions of this kind definitions of document-as-message, since they represent documents as aggregations of signs (i.e., messages or texts), for the expression and/or transmission of meanings.
Thirdly, yet other definitions focus neither on documents' materiality nor on their signhood-i.e., not on documents as bearers of meaning, physical or otherwise-but on their status as meanings in themselves.These are definitions of document-as-meaning.
Compound definitions are those that simultaneously assign to documents two or three of the essential qualities of materiality, signhood, and meaninghood (where those latter terms stand for "being a sign" and "being meaning," respectively).For example, Wersig & Neveling (1976) define "document" as "A unit consisting of a data medium, the data recorded on it, and the meaning assigned to the data." In Table 3 (emphases added), the highlights are presented, in chronological order, from a list of more than twenty-five definitions extracted from glossaries, dictionaries, standards, and other literature dating from 1966 through 2019."A record that contains information content.In common usage it still normally means a piece of paper with words or graphics on it.In library and information work, the term is however used to mean any information-carrying medium, regardless of format.Thus books, manuscripts, videotapes and computer files and databases are all regarded as documents."

2004
Reitz "A generic term for a physical entity consisting of any substance on which is recorded all or a portion of one or more works for the purpose of conveying or preserving knowledge.In the words of the communication theorist Marshall McLuhan, a document is the 'medium' in which a 'message' (information) is communicated.Document formats include manuscripts, print publications (books, pamphlets, periodicals, reports, maps, prints, etc.), microforms, nonprint media, electronic resources, etc. ... Also, any form printed on paper, once it has been filled in, especially one that has legal significance or is supplied by a government agency, for example, an application for copyright protection...." Table 3.Some post-1966 definitions of "document." No particular synchronic trend is apparent in this data, but what is apparent is (a) the wide variety of definitions, and of combinations of category memberships, and (b) the fact that such combinations-that is, compound definitions-are more common than single ones.I take this data as evidence for concluding that documents are (or, at least, are typically considered to be) complex objects rather than simple ones.
The three-part compound definition corresponds partially to the four-entity data model that lies at the heart of IFLA's Functional Requirements for Bibliographic Records (FRBR;1998).This model distinguishes between works, expressions of those works, manifestations of those expressions, and items (i.e., copies of those manifestations).Items are the only physical entities recognized in this model; each item can be viewed as the medium for (or carrier of) a given expression or aggregation of expressions, which in turn can be viewed as the message(s) representing a given work or aggregation of works (see Figure 1).On the face of it, at least two separate conceptions of "document" can be derived from the WEMI (works, expressions, manifestations, items) model: one that treats documents simply as items, i.e., as physical media; and a second that conceives of documents as complex entities that exist simultaneously as material, signifying, and meaningful things, i.e., as messages (expressions) and meanings (works) as well as as media (items).As a possible way of deciding between these two alternatives, we might consider that, in the course of a discussion of the scope of FRBR (p. 8 of the final report, 1998, emphases in original), the word "document(s)" is used eight times as a synonym for "information resource": "... [U]sers may make use of bibliographic records for a variety of purposes ...: to determine what information resources exist ...; to verify the existence and/or availability of a particular document ...; to identify a source ... from which a document can be obtained ...; to select a document or group of documents that will serve the information needs of the user; ....

"... [T]
he functional requirements for bibliographic records are defined in relation to the following generic tasks that are performed by users when ... making use of ... library catalogues: • using the data to find materials that correspond to the user's stated search criteria (e.g., in the context of a search for all documents on a given subject ...); • using the data retrieved to identify an entity (e.g., to confirm that the document described in a record corresponds to the document sought by the user ...); • using the data to select an entity that is appropriate to the user's needs ...; • using the data in order to acquire or obtain access to the entity described (e.g., ... to access online an electronic document ...)." Even though "document" is not subsequently used in the rest of the final FRBR report, the clear implication is that, in the FRBR world-view, documents are the sorts of things that are sought, found, selected, and acquired, as a result of judgments made by catalog users as to the relevance of those things, given users' needs and wants.Such judgments are made on the basis of assessments of documents' formats and contents-i.e., on the basis of evaluation of the qualities of documents as media, as messages, and as (aggregations of) meanings.On this reading, documents are not to be conceived primarily as physical items that have the properties of instantiating given manifestations, given expressions, and given works; rather, any given document is, simultaneously, a medium, a message, and a meaning.

II
Lowe's category theory So much for an empirical survey of the various kinds of definitions of "document" that have been suggested over the years.The results raise a question which demands an ontological approach: in other words, it requires some input from the philosophical subfield of ontology, the study of the nature of being.
One of the tasks of ontology that has been deemed more or less important since at least the time of Aristotle is the identification of the "highest," "topmost," or most general categories or kinds of things that exist in the world.Some ontologists have established systems of top-level categories that are hierarchical in structure, with one category containing all things at the very top, divided into a small number of sub-categories, each of which is subdivided into a small number of sub-sub-categories, and so on.Typically in such structures, the sub-categories at any given level are both exhaustive and exclusive, so that any individual thing is a member of one and only one sub-category at that level.Figure 2, for example, depicts the top-level structure proposed by the American philosopher Roderick Chisholm (1916-99) in his A Realistic Theory of Categories from 1996.One of the most well-known contemporary top-level ontologies is that devised by the British philosopher Jonathan Lowe (1950Lowe ( -2014)), who wrote as E. J. Lowe and who was Professor of Philosophy for many years at Durham University in England.Lowe's system, which he promoted as a means of understanding the foundations of natural science, rests on three basic binary distinctions.Lowe distinguishes between universals and particulars, between substances and properties, and between abstracta and concreta, in arriving at the structure depicted in Figure 3. (The base diagram as presented here is lifted straight from one of Lowe's earlier publications.He would subsequently make some changes in his use of terminology, and a few slight amendments to the base diagram are made in Figure 3: "Properties" is merged with "Relations" to form a category of "Attributes"; the place of "Relations" is taken by "Kinds"; and "Tropes" is replaced by "Modes.")Questions about whether or not these distinctions may be sustained, and if they can, how that may be done, have been among the most hotly debated in metaphysics for more than two thousand years and it is certainly not my goal to attempt to survey answers to those questions today.What I am going to do instead is very briefly to characterize the distinctions that Lowe makes.Firstly, universals vs. particulars."Even in this matter," Lowe says (2003, p. 8), "there is controversy."Lowe conceives of universals as things that are repeatable, that is, as things that may be "borne" or possessed by many different particulars, at different times and places; whereas particulars are each "wholly confined to a unique space-time location and cannot 'recur' elsewhere and elsewhen" (p.8).In other words, universals are instantiable (by particulars), and particulars are not.Examples of universals include properties such as the property of being red, and kinds such as the kind denoted by the word "apple."Examples of particulars include the apple I ate yesterday and the redness of that apple.
Lowe's distinction between substances and properties is among particulars.It is the distinction between objects and modes (or tropes).An object is "an entity which bears properties but which is not itself borne by anything else" (p.8), like the apple I ate yesterday; a mode or trope is a particular that is borne as a property by no more than one object (p.9), like the redness of that apple.
We might say objects "instantiate" kinds, "exemplify" attributes, and are "characterized" by modes.Similarly, attributes "characterize" kinds, and are "instantiated" by modes (see Figure 4).Another way to think of attributes is as property-kinds, in parallel with the substance-kinds that are instantiated by objects.Objects are substance-instances, characterized by modes as property-instances.It has sometimes been suggested that the so-called "four-category ontology" (Lowe, 2006) depicted in the so-called "ontological square" was first proposed by Aristotle, and on this basis Lowe and others in his camp are known as proponents of neo-Aristotelian metaphysics.(The version of the ontological square presented in Figure 5, by the way, was lifted from a 1997 article by Barry Smith, about whom we shall hear more in Part III.)The third basic distinction that Lowe draws in his top-level hierarchy is among objects (that is, among substance-instances), and it is between abstract objects and concrete objects.Concrete objects are those that exist in space-time (that is, are "datable and locatable") or at least exist in time, whereas abstract objects are those that do not.Examples of concrete objects include individual apples; examples of abstract objects (according to Lowe) include numbers, sets, and propositions.A different criterion that may not coincide exactly in its picking out of abstract objects is the capability of an object to enter into causal relations: an abstract object is one that is incapable of such interaction.
Controversy abounds in relation to the category of abstract object.For example, Lowe identifies propositions as a canonical sub-category of abstract object.But in what sense are propositions conceivable as particulars (that is, noninstantiable) rather than as universals instantiated by sentences (expressions of propositions) and in turn by utterances of those sentences in physical (spoken or written) form?Similarly, in what sense are works-often conceived as aggregations of propositions-comprehensible if not as universals instantiated by copies of those works in physical form?
Another bone of possible contention has to do with the category of concrete object.Two sub-categories of concrete object highlighted by Lowe (2003, p. 5) are masses, or material bodies, and living organisms: "Entities belonging to these two categories have quite different existence-and identity-conditions, because a living organism, being the kind of thing that is by its very nature capable of undergoing growth and metabolic processes, can survive a change of its constituent matter in a way that a mere mass of matter cannot.A mere mass, being nothing but an aggregate of material particles, cannot survive the loss or exchange of any of those particles, any more than a set can undergo a change of its members.As a consequence, it is impossible to identify a living organism with the mass of matter which constitutes it at any given stage of its existence, for it is constituted by different masses at different stages." Lowe does not clearly establish what other sub-categories of concrete object there are.The category of artifacts is one obvious candidate, but Lowe has little to say about artifacts in general or sub-categories of artifact more specifically, which means we are forced to speculate a little about where in his scheme certain entities might fit.In a 2014 paper entitled "How real are artefacts and artefact kinds?"Lowe distinguishes between utensils and machines as sub-categories of artifacts, arguing that machines and machine kinds, like natural kinds, are fully and mindindependently real, whereas utensils and utensil kinds-things like "tables, chairs, tents, cooking pots, knives, and hammers" (Lowe, 2014, p. 24)-are not.In the conclusion to this paper, he says "I should stress that I am not urging that machines are the only real artefacts.I am content to allow, for instance, that works of art may well qualify as real artefacts too" (p.26, emphasis in original).At the beginning of the paper, however, he had already stipulated that he was "setting aside here putative examples of abstract artefacts, such as musical scores, conceived as types rather than tokens" (p. 18, emphasis in original), thus leaving tantalizingly open the question as to whether works of art should be counted as particulars at all.
The important question for the would-be ontologist of documents, then, is the question of where in such a system of categories documents fit.Are documents universals or particulars?substances or properties?and so on?It's a question that's not as easily answered as it might at first seem.We might have an inkling of the difficulties now that we've conducted our survey of definitions of "document," and especially now that we've distinguished between the ideas of document-asmedium, document-as-message, and document-as-meaning.
In particular, if we cleave to a compound definition, we might expect to have to do some extra work in situating documents among what are typically conceived as exclusive categories.And if it turns out that the concept of "document" is too "complex" for easy placement in a top-level ontology, so be it; the ontology must be revised to accommodate our concept of the thing, not vice versa.
One way to proceed is to start at the top of Lowe's hierarchy, and attempt to justify our choice of placement of "document" on successive branches.Beginning at Level 0, as it were, the first question is, Are documents things or not?The answer is, Yes.Since the intention is that all things fall in the top category of "things," documents should be treated as things.So far, so good.
At the next level down, Level 1, the question is, Are documents universals or particulars?Almost immediately, we run into a problem.On the one hand, if we consider the document-as-medium option, it seems to be fairly clear that documents are particulars (that is, they are non-instantiable).On the other hand, if we take any of the other views (simple or compound) on the nature of documents as suggested by the survey, we implicitly commit to a conception of documents as instantiable, just as (some, but not necessarily Lowe, would argue) works and propositions are.
Let's continue, for the time being, on the assumption that the document-asmedium option is the more attractive.In that case, documents are particulars.At the next level down, then, Level 2, the question becomes, Are documents objects (substance-instances) or modes/tropes (property-instances)? The simple answer is that, since documents are the bearers of properties, and are not borne by anything else, they are clearly objects, not modes.Next!At the next level down, Level 3, the question is, Are documents abstract or concrete?Even for the document-as-medium conception, there is uncertainty here created by the absence of artifacts among the sub-categories of Lowe's concrete object.But this is probably a limitation of the presentation of Lowe's hierarchy and a reason to revise that presentation rather than a reason to place documents-asmedia in any category other than concrete object.So far, so simple.
If we make a retreat from our choice of document-as-medium, and consider how to handle documents-as-messages and documents-as-meanings, as well as documents-as-combinations, we need to go back to Level 1, and ask again, Are documents universals or particulars?In these cases, the conception is of documents as instantiable-that is, documents as texts or works that exist as types rather than as tokens.In the absence of a motivation to apply any theory that differentiates between the type/token distinction and the universal/particular distinction, our decision should be to consider documents as universals.
Moving down to Level 2, the question becomes, Are documents substancekinds or property-kinds (attributes)?Since documents are characterizable, the simple answer is substance-kinds.
So we have a situation where on the one hand, documents "live" in a new sub-category of concrete objects, possibly called artifacts, where the challenge will be to identify the qualities that distinguish documentary artifacts from nondocumentary artifacts (if there are such things as the latter); and on the other hand, "document" is placed in the category of "Kinds."There is actually a third possibility not suggested by our navigation through Lowe's hierarchy, but which is nevertheless suggested by Lowe's characterization of the category of abstract objects as including not just the sub-categories of numbers and sets, but also propositions.If works are considered to be aggregations of propositions, then it might seem that works should also be placed here; from work, it's a short step to document-as-meaning, and thus to thinking of documents as abstract objects.
Ideally we would use different words to refer to these different concepts.But we do not.We use a single word, "document," interchangeably in different contexts to mean different things.Perhaps this is no bad thing.It keeps us on our toes.But I want to suggest that much of the time, many people use the word "document" to mean something that is a universal, not a particular-something that's not necessarily material.And I think that that is important for our understanding of documents, and for our ideas as to where document theory could or even should be going.
In essence, my argument is an empirical one about the use of language.What do we talk about when we talk about documents?Much of the time, the properties of documents that we are most interested in are properties of documentsas-meaning, or if you like, documents-as-works, documents-as-universals, documents-as-types that are multiply instantiated by physical tokens.Of course, it is important to recognize the existence of, and to understand the nature of, the type/token relationship, but that doesn't necessarily mean that the sole or even the primary focus of document theory should be on the materiality of document tokens.Instead, or at least in addition, understanding the universality of document types should be high on our agenda.
We have reached the end of Part II.In the third and final part of the talk, I should like briefly to discuss what I take to be the most substantive contribution to the ontology of documents of the last few years.

III
Smith's ontology of documents The philosopher Barry Smith published a short conference paper called "The ontology of documents" in 2011, following up on a presentation on a similar topic from 2005.In these and several related papers on his so-called "theory of document acts," Smith has developed an account of the status of documents in the context of the picture of social reality painted by fellow philosopher John Searle over a period of several decades (see, e.g., Searle, 2010).
This work is not really about categories per se, but is ontological in the sense that it explains how documents-or more precisely document acts like signing a document, filling it in, delivering it, or archiving it (see Figure 6)-have the effect of bringing new entities into existence, that is, how document acts have ontological consequences.Smith builds on John Searle's ideas about speech acts-the things we do with words.Searle explains how certain kinds of speech acts-the ones he calls "declarations"-can bring about changes in the ontology of social reality, and Smith similarly describes how certain kinds of document acts-generally speaking, the things we do with rather than to documents-bring into existence not just physical entities like document tokens, and document-related artifacts like filing systems, but also document-related social practices and quasi-abstract entities ("at one and the same time subject to historical changes yet not made of physical parts," Smith, 2011, p. 2, emphasis in original), especially in the realms of commerce, law, and government.These quasi-abstract entities are things like organizations, contracts, laws, money, rights, obligations, identities, claims, privileges, corporations, capital, permissions, debts, trusts ... that is, they are entities that form vitally important parts of social reality.
Smith argues that we need to pay more attention to document acts than Searle does.His primary reason for doing so is to address issues relating to the "anchorage" of digital documents to the people who created them, by implementing systems for the certification of authenticity that are on a par with signatures and fingerprinting for physical documents.
But in the course of setting this up, Smith also makes important contributions to our understanding of the basic ontological categories to which documents belong.
Firstly, Smith points out (as others have done before and since) how Searle's categorization of speech acts may be applied to documents.Thus we may distinguish among documents that are representative, directive, commissive, expressive, and declarative.Representatives "commit the [writer] ... to the truth of ... expressed proposition[s]"; directives "attempt ... to get the [reader] to do something"; commissives "commit the [writer] ... to some future course of action"; expressives "express the [writer's] psychological state ... about a state of affairs"; and declarations "bring about correspondence between ... propositional content and reality" (Searle, 1975, pp. 354-361).
Smith is most concerned with declarations, those which are most clearly generative of new entities.Smith uses the term "creative" rather than generative, with the somewhat counter-intuitive result that he classes books (fiction and nonfiction), journal articles, maps, artworks as "non-creative"; on the other hand, certificates, contracts, receipts, banknotes, licenses, agreements, filled-in forms, passports, diplomas, medical records, meeting minutes, etc.-all have "creative power" in social reality (see Figure 7).Secondly, Smith follows Goodman in distinguishing those documents that are autographic, and those that are allographic.For Goodman (in Languages of Art), a work of art is autographic "if and only if the distinction between original and forgery of it is significant; or better, if and only if even the most exact duplication of it does not thereby count as genuine" (Goodman, 1976, p. 113).Painting, sculpture, and architecture are autographic; music, photography, and literature are allographic.
The autographic/allographic distinction seems to correspond at least roughly to the distinction between documents that exist as both types and tokens, and those that do not tokenize a type (see Figure 8).Briet's photograph of a star?Allographic.The stone in a museum, and the antelope in a zoo?Autographic.Many of the "creative" document-types listed above?Autographic.Whether the type/token distinction itself corresponds to the universal/particular distinction is a complex matter that might be better left for another day, although clearly any conclusions will be significant for our decision-making when it comes to situating documents in a top-level hierarchy of categories like Lowe's.Thirdly, and perhaps also significantly for that decision-making, Smith seems to conceive of document acts, including document production acts, like Searle's speech acts, as events or occurrents, notwithstanding that the physical records that are among the products of such acts are definitively classed by Smith as continuants (eliding the type/token distinction).Might there be scope here for further sub-categorization of concrete objects?
A more general question raised by Smith's analysis is, What might be the more productive route for DOCAM-style document theory to follow?On the one hand, we might imagine a future document theory that commits wholeheartedly to the distinction drawn by Smith between creative and non-creative documents, carving out a subfield that focuses on the former and on kinds of issues identified by Smith as critical for a digital social reality whose effective and efficient organization depends so much on reliable authentication of autographic creative documents.In this way we may contribute to the kind of "scientific understanding" that Smith says is necessary for arriving at an "intelligent appreciation of the changes in social reality that are being effected through the trillions of documents being created daily in the digital realm" (Smith, 2014, p. 30).
On the other hand, might it be productive to extend the notion of declarative documents' creative or generative power so that all documents, including those that Smith identifies as non-creative but that are traditionally the main concern of document theory, are considered to be creative in some respect and/or to some degree.Smith enumerates some of the kinds of things one can do to a document, such as sign it, fill it in, register it, and archive it.It is surely a short step to take to consider that some other such document acts include finding it, identifying it, selecting it, and obtaining access to it (just to choose those suggested by the FRBR final report), as well as organizing it, classifying it, and indexing it, and reading it, interpreting it, citing it, and using it, in many and various ways.Similarly, the products of such acts include quasi-abstract entities of many and varied kinds, including metadata, bibliographies, catalogs, result-sets, recommendations, rankings, metrics, and networks, to name just a few.
And there I shall stop, with a reminder of my three conclusions, a list of sources of definitions, and a list of references.I should like to thank my colleagues Greg Leazer and Julie Park for their helpful comments on an earlier draft of this talk, and to thank you very much for listening.
"A work recorded in language or symbols, or by other means."physical entity of any substance on which is recorded all or part of a work or multiple works.Documents include books and booklike materials, printed sheets, graphics, manuscripts, sound recordings, video recordings, motion pictures, and machinereadable data files.""A combination of a medium and the information recorded on or in it, which may be used as evidence or for consultation."The term applies not only to objects written or printed on paper or on microforms (for example, books, periodicals, maps, diagrams, tables, and illustrations) but also to non-print media (for example, artistic works, audio and video recordings, films, machine-readable records, and multimedia) and, by extension, to naturally occurring or humanly made objects intended to convey information (for example, zoo animals, plants in botanical gardens, museum collections of hand tools, etc.)." "A combination of a medium and the information recorded on or in it, which can be used for consultation, study or evidence."1976 Buchanan "Generic term for the information-bearing media handled by librarians --books, serials, sound recordings, films, illustrations etc." 1976 Unesco / Wersig & Neveling "A unit consisting of a data medium, the data recorded on it and the meaning assigned to the data."1987; Harrod [6th-10th "A record which conveys information; originally an inscribed or written record, but now considered to include any form of information --graphic, acoustic, alphanumeric, etc. (e.g.maps, manuscripts, tape, videotapes, computer software)."2000 Wellisch "A medium on or in which a message is encoded; thus, the combination of medium and message.