Revisiting "what is a document?"

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0022-0418.htm Revisiting “what is a document?” Revisiting “what is a document?” Bernd Frohmann Faculty of Information & Media Studies, The University of Western Ontario, London, Canada 291 Abstract Purpose – The purpose of this paper is to provide a reconsideration of Michael Buckland’s important question, “What is a document?”, analysing the point and purpose of definitions of “document” and “documentation”. Design/methodology/approach – Two philosophical notions of the point of definitions are contrasted: John Stuart Mill’s concept of a “real” definition, purporting to specify the nature of the definiendum; and a concept of definition based upon a foundationalist philosophy of language. Both conceptions assume that a general, philosophical justification for using words as we do is always in order. This assumption is criticized by deploying Hilary Putnam’s arguments against the orthodox Wittgensteinian interpretation of criteria governing the use of language. The example of the cabinets of curiosities of the sixteenth-century English and European virtuosi is developed to show how one might productively think about what documents might be, but without a definition of a document. Findings – Other than for specific, instrumentalist purposes (often appropriate for specific case studies), there is no general philosophical reason for asking, what is a document? There are good reasons for pursuing studies of documentation without the impediments of definitions of “document” or “documentation”. Originality/value – The paper makes an original contribution to the new interest in documentation studies by providing conceptual resources for multiplying, rather than restricting, the areas of application of the concepts of documents and documentation. Received 11 February 2008 Revised 13 June 2008 Accepted 17 June 2008 Keywords Document management, Information science, Philosophical concepts Paper type Conceptual paper In two important papers (Buckland, 1991, 1997), Michael Buckland asked: “What is a document?” He showed us that the answer is complex, and ever since, the question remains open. It has reverberated through the four annual conferences of the Document Academy, in the form of demands for definitions of a document[1]. It is therefore worth reconsidering Buckland’s question. But rather than propose an answer, I aim to treat the question itself as my topic. What are we doing when we ask for definitions? Are such questions always in order? What is it to give a definition? Can we think productively about documents and documentation without definitions? There are three important philosophical motivations for wanting definitions. The first is instrumental, seeking to provide working definitions for specific purposes. These kinds of definitions do not occasion the philosophical anxieties of “real” definitions, because they do not purport to capture the real nature of the defined thing. Because I do not see particular difficulties with them for my purposes in this paper, I am satisfied with just this brief mention. The author gratefully acknowledges the generous support of Standard Research Grant 410-2005-0347 awarded in April 2005 by the Social Sciences and Humanities Research Council of Canada. Journal of Documentation Vol. 65 No. 2, 2009 pp. 291-303 q Emerald Group Publishing Limited 0022-0418 DOI 10.1108/00220410910937624 JDOC 65,2 292 The second is a philosophically realist motivation, seeking to define the real nature of the definiendum by specifying the significant characteristics of things by virtue of which scientific knowledge of them can be had. I’ll say something about this sort of definition with regard to John Stuart Mill’s views as expressed in his 1868 work, A System of Logic (Mill, 1868). Finally, and I’ll say much more about this third type, is a motivation arising from a foundationalist philosophy of language that assumes whenever we speak, there must be something attached to the word on the basis of which we succeed in making sense in using it. The things most commonly assumed to be so attached are meanings, concepts, rules for the use of words, or, for orthodox Wittgensteinians, criteria governing language use. Surely – so the thought goes – when I speak meaningfully about documents (for example), I speak on the basis of some knowledge about the “marks or features” that distinguish documents from chairs, cats or dogs. A definition that identifies these marks or features then shines sharp analytical light on what I see but dimly in the ordinary course of conversation. Moreover, I am responsible for it whenever I profess to be speaking seriously, being obliged to defend it when challenged. I will address this thought with the help of Hilary Putnam’s arguments against orthodox Wittgensteinians. He has a view of language – which he claims is genuinely Wittgensteinian – according to which we extend our uses of words without the support assumed by foundationalist views of language. Throughout this discussion of definitions, I’ll try to bring it back to documentation. But in the second half, I’ll appeal directly to an example of a special kind of document to illustrate, by taking seriously Paul Otlet’s and Suzanne Briet’s ideas of physical objects as documents, how we might think about documentation without definitions. The example also prompts some methodological remarks about how to extend the concepts of documents and documentation. The objects are those of the cabinets of curiosities of the sixteenth-century English and European virtuosi. This glance backwards in time will, I hope, move us forward to reconsider Buckland’s question, “What is a document?” Definitions A problem for Wittgensteinian scholarship concerns the role of criteria. A key idea for “orthodox” Wittgensteinians, who interpret criteria in terms of a framework of rules, is that criteria are what we provide in justifications of how we apply words to the world. For example, to justify referring to something as a chair, a tree, a cat or dog – or, of interest here, a document – involves appealing to rules according to which particular “marks or features” of these things constitute criteria for successful reference to them. But criteria and rules do not arise only in justifications of our talk. “Rules of language” are also pictured as the rules “we go on when we apply words to the world,” as one of its chief exponents puts it (see Hilary Putnam’s discussion of Stephen Mulhall in Putnam, 2001, pp. 10-13). The implication is that we never make judgments without criteria or rules. They are pictured as always already there, hovering in the background whenever we speak, which we in some sense – which then has to be explained – “follow” as we talk. Whenever we use a word, this view implies, just speaking coherently makes us responsible, somehow, for articulating the criteria, or the rules, according to which we manage to do it. Thus when one speaks meaningfully of chairs, cats and dogs – or documents – one is, simply by virtue of making sense, responsible for articulating those marks and features according to which one calls something a chair, a cat or dog – or a document. The idea that we never make judgments without criteria implies that it always makes sense to ask, in a very general, philosophical sense, for justifications for our uses of words. This picture of criteria is not far removed from a closely associated picture regarding definitions of what we talk about (I am not arguing that Wittgensteinian criteria are definitions). A folkloric philosophy sees the role of definitions as justifications for our uses of words by analogy to the more sophisticated, orthodox Wittgensteinian view of the role of criteria. It also pictures definitions hovering in the background, as what we go by when we talk, and as what we must mean by a word when we use it to refer to the world. And, it assumes a similar responsibility to deliver the definitions we go by, and that it always makes sense to ask for them. In the chapter on definitions in his A System of Logic, John Stuart Mill has a robust conception of such a responsibility. For him, although a definition is a “proposition declaratory of the meaning of a word” (Mill, 1868, p. 149), some definitions are better than others. Ordinary folk, Mill complains, typically use words vaguely and loosely, without strict regard for affirming “any determinate attributes” (Mill’s terms for “marks or features”) of the referents of their terms. At worst, a word commonly used “denotes a confused huddle of objects, having nothing whatever in common,” and therefore “has become quite unfit for the purposes either of thought or of the communication of thought” (Mill, 1868, p. 169). Proper definitions are not arbitrary, because to provide good definitions is to engage in analysis and classification according to attributes relevant to scientific understanding. Ordinary meanings are confused, but logicians shoulder the responsibility of “retouching”, or “logically remodelling”, our language (Mill, 1868, p. 170) for epistemological purposes. Their definitions are not statements of conventional meanings, but the products of inquiries into “the resemblances and differences among . . . things” (Mill, 1868, p. 171). The philosopher’s responsibility is to discover what the meaning of a term should be, in order to contribute to a scientific classification of things, and thereby to bring order to thought and communication. Thus, from historically and philosophically disparate points of view, both the orthodox Wittgensteinian philosophy of language and Mill’s philosophical realism rely on the assumption that a general, philosophical justification for using words as we do is always in order, whether framed in terms of rules of language or of Mill’s real definitions. Putnam has a very clear, short paper on criteria in The Legacy of Wittgenstein: Pragmatism or Deconstruction (Nagl and Mouffe, 2001). He and some other Wittgenstein scholars, prominently among them Stanley Cavell, are critical of the orthodox Wittgensteinian position, and it seems to me they are right. They argue that the question of justification does not always makes sense; rather, it arises only in connection with specific philosophical or empirical confusions. There are cases where it makes good sense to ask, for example, “what is your justification for calling that animal a dog?”, perhaps when walking in the country we see from afar what could be a dog – or is it perhaps a wolf? Some of us can spot the differences between dogs and wolves from a distance and some of us cannot. If I am unsure what that yonder beast is, my asking for your justification for calling it a dog makes sense. I want to know how you can tell the the difference, and perhaps learn to do it myself. Revisiting “what is a document?” 293 JDOC 65,2 294 But does the question make sense in general, beyond context and specific empirical or philosophical confusions? Are such justifications always in order? I might wonder, even after reading Briet, whether an antelope can be a document, and my skepticism might readily be forgiven before reading her, but still, it is hard to know what kind of answer could possibly satisfy a decontextualized, confusion-free demand for a justification – of how, for example, a document differs from a rose, or from anything else – based solely on a philosophical notion that there must be rules or definitions we go on in applying words to the world. As for the position Wittgenstein held, here is Putnam: The very idea of a general problem here, a general question as to how any of the things we say is justified (or as to how we are justified in saying them), whether we be chatting at the dinner table, arguing about the next elections, advising a client, performing an experiment, reproving someone for their behavior . . . seems to me one that Wittgenstein would certainly reject as senseless (Putnam, 2001, p. 20). Justifications of how we talk, when based upon rules of language, do not take us very far toward bearing a very general responsibility for coming up with definitions of the words we use. We can, of course, make up rules and definitions, and pledge to follow them, perhaps even enforce them, but then they are instruments, more or less appropriate to specific aims, not necessities arising from theories of language. When it comes to documents, it is certainly the case that very specific sorts of investigations require very specific sorts of definitions, but it would be a mistake to brandish a definition devised for a specific context and purpose it as settling what counts as a document in every situation. Justifications aside, however, strong temptations remain to worry that if there is not something like rules, criteria, definitions, concepts, ideas, or meanings we go by when we speak, our language becomes completely unruly, an anarchy where anything goes. If our concepts are not fixed, if an inherent indeterminacy bedevils rule-following, then does not meaning become ambiguous? Do not our words, and in the present case, the words “document” and “documentation, ” slip their moorings? Some orthodox Wittgensteinians express this kind of anxiety, worrying that going on without criteria “is making sounds to which no ‘normativity’ attaches, in effect, mere babble” (Putnam, 2001, p. 13) – an idea not far removed from Mill’s disquiet about how a “confused huddle” threatens thought and its communication. But how realistic is such a threat? In the same book of essays on Wittgenstein, Linda M.G. Zirelli (2001, p. 33) writes, “I find little evidence in Wittgenstein to suggest that the lack of fixity in our concepts amount to a permanent threat to our life with language and with each other” – as deconstructive readings tend to celebrate as Wittgenstein’s admirable Derridean moment. Our concepts can lack fixity (and to this degree Wittgenstein is Derridean), what counts as following a rule can be multiple, yet we still can communicate and speak meaningfully, about documents and anything else, because there is no general, inherent ambiguity of language, only specific, contingent ambiguities for particular speakers under specific conditions. Zirelli reminds us that Wittgenstein constantly confronts his readers with situations where “what I took to be the most common concept is not used by another person in the same way” (Zirelli, 2001, p. 35). What happens in such a situation? I could cling to a belief that there must be rules, fixed concepts, sharp meanings, and precise definitions, from a worry about a general threat to language as such. That leads to my defending my usage by insisting that the other person does not understand the rules or is attaching different meanings to my words. Or, I could listen to the other’s use of language, and try to attune myself to a new story, a new way of speaking, to see where it leads. It is in the latter spirit that Zirelli writes: The point here is that whether I do in fact find out that you and I do not use a word in the same way, not to mention what I do with that discovery, is a matter not of the generalized failure of communication or the intrinsic deviance of rule-following but of a certain attitude that I take with respect to you” (Zirelli, 2001, p. 36; emphasis added). This aspect of the matter is not epistemological, as Mill sees it, or located in theories of meaning, as orthodox Wittgensteinians and foundationalists see it, but ethical. If someone puts the concept of a document to use in understanding oral traditions (for example), whether I react with an openness to the idea to see where it goes or with an argument that it violates rules of language, fixed concepts, clear meanings, or precise definitions that settle, once and for all, what a document really is, are moral responses, based upon how seriously I take the threat posed by a generalized imprecision of language to be. The moral of my story so far is that criteria, rules and definitions can be specified (and enforced) for particular purposes, but are not always already there, guiding us as we speak, and we are not generally obliged to provide them, nor is there any generalized ambiguity of language posing a threat to thought and communication as such – even when imprecision and ambiguity might pose a threat to specific kinds of talk that rely upon well-formed definitions, fixed meanings, and clear rules. The topic of the definition of a document leads directly to questions about how to think about documents and documentation. What are we trying to do when we think about these concepts? What does thinking about documents mean? Specific domains employing different definitions can be accumulated. But are there not ways of thinking about documents that do not rely on definitions, and that succeed quite apart from any imperative to provide them? If the moral of my story is taken to heart, unless there is a demonstrable and overriding necessity to engage only in talk governed by well-formed definitions, fixed meanings, and clear rules when we think about documents, then given that we don’t generally rely on such precision instruments, there ought to be ways of thinking about documents and documentation that also get along without them. Before turning directly to what some of these ways might look like, I’d like to briefly consider Putnam again, this time what he has to say about “going on without rules” (Putnam, 2001, pp. 13-18), because it bears directly on extending our concepts, including the concepts of document and documentation, without rules, without definitions – or without worrying about the answer to the question, What is a “document”? Rules and stories Putnam’s most developed examples come from scientific revolutions, specifically the “natural extension” of physics to include the General Theory of Relativity. Before non-Euclidean geometries, it was mere babble to say that one could have a triangle with two base angles that are right angles. It also would have been mere babble to say that space was finite yet bounded. But after Riemann, Lobachevski and Einstein, saying such things is not only not mere babble, but any physicist who denied non-Euclidean geometries and their application by Einstein to physical space would, as Revisiting “what is a document?” 295 JDOC 65,2 296 Putnam observes, be “regarded as irrational, as at best an unreasonable reactionary” (Putnam, 2001, p. 17). But would they be violating new rules for the use of “triangle” or “space”? Certainly, the first physicist who accepted the General Theory is rightly regarded as rational, “indeed as displaying a high order of scientific insight”, but such acceptance could not plausibly be thought an instance of a regularity, much less following a rule. But if concepts such as “straight line,” “simultaneity,” and others are extended in new ways by the Einsteinian revolution in physics without rules, how could such extensions be not only non-arbitrary, but manifest the highest order of scientific rationality? The answer, as Putnam puts it, is that “Einstein told his story and showed how to apply it,” and that “Riemann and Lobachevski told their stories and showed us how to apply them” (Putnam, 2001, p. 16). There is no deeper and generalized basis, no basis on the level of foundation or justification, beyond what Einstein, Riemann, and Lobachevski provided. It is not as if, beyond their work, there must be a deeper, foundational change, such as changes in rules, criteria, meanings, or definitions, a point Putnam illustrates by analogy to student questions in courses on ethics: [. . .] students often ask ‘By what criteria can one tell when an ethical claim is justified?’, and are startled when I reply, ‘By just the considerations that we advance in a good ethical argument.’ It is as if, over and above the things we say when we argue for or against an ethical judgment, there had to be a more fundamental consideration, a philosophical consideration, which we ordinarily neglect to give, but which has to be given lest our ordinary arguments lack . . . what?” (Putnam, 2001, p. 21). Putnam also tells the story of a professor at a Catholic university who, after losing his faith, gave a paper at a Catholic university conference, opening his remarks with “I guess I am the lion being thrown to the Christians.” Even when it is completely new, it is a joke virtually everyone not only gets but even grasps the difference between appropriate and inappropriate ways to understand it. Putnam closes his discussion of this example with, “But rules? Come on!” (Putnam, 2001, p. 19) Later, he summarizes: “we do not need a ‘framework of rules’ to serve as a foundation for the ways we go on” (Putnam, 2001, p. 21). The same goes for definitions. The moral of Putnam’s stories is that if we project the word “document” onto new situations we need not do so on the basis of new criteria, rules, meanings, or definitions. Sometimes others will chime in, or will “go on” from our extensions as we do; at other times, in ways they see as flowing from them but that we did not anticipate. Some may refuse to go in our direction, or not find it relevant to their purposes to do so. But if we want to extend documentation to new situations – if we want to multiply the concept of the document – we can (like Einstein) tell a story and how to apply it, and connect our extensions to theories, arguments and relevant intellectual resources to try to bring about an enlarged understanding of what documents and documentation can be. We can also use specific tactics, such as beginning with a clear case of a thing we agree is a document, or an activity we agree is an instance of documentation, and then introduce new cases by analogy, similarity, and resemblance. When Briet suggests an antelope can be a document, the only basis we need to extend the concept in her way is her argument. That’s enough of a foundation. If a new extension points in promising directions, we certainly benefit from a rich array of resources to help us take that path, and we can be frustrated if we are offered just a few. But what matters are the resources, and the richer, the better. A definition by itself is a puny resource for a significant change of direction. There is no reason to suppose that the stories we tell will be the same, which is another way of saying that there is no reason to suppose there will be one, unified theory of documents and documentation. But that is not to say that theoretical resources can’t be used to elaborate the stories. In the remainder of this paper, I want to show what some of these stories might look like, with reference to a very specific example from just one category of documents: the objects of the cabinets of curiosities of the sixteenth-century English and European virtuosi. Things Thanks to Buckland (1991) and Boyd Rayward (2007), we are familiar with Otlet’s and Briet’s conceptions of documentation, according to which things – physical objects, such as antelopes – can count as documents. Briet wrote that an antelope in a zoo was a document because a zoo is a place where the antelope is available to support, or document, a claim (the claim I use with students is about what antelope horns look like; the claim that some tigers have no stripes is effective too). The importance of place for counting as a document, or for its having documentary properties, is not so strange when we think of museum objects. The idea that things are documents when located in places where they are readily available to provide evidentiary support for particular propositions casts a very wide net, as Buckland’s meditations on “information as thing” has shown by leading to anything at all counting as a document (Buckland, 1991). Yet providing evidence is not the only kind of documentary power wielded by things exhibited in specific places. Some places, rather than encouraging statements, claims, and propositions about the things in them, were designed to deliberately constrain, limit, and inhibit them. Such were the “cabinets of curiosities” of sixteenth century England and Europe. Their objects were intended to intersect with the universe of statements such a way as to draw boundaries around what could be said. They were not intended to inform in Otlet and Briet’s sense; they were not displayed to generate evidentiary connections to other objects, especially not, as the material equivalents of Baconian facts, to theories or generalizations. They lack, or at least constrict, what Ron Day, in his interpretation of Briet, calls “indexicality” (Day, 2001). Their power was directed elsewhere, towards policing quite specific discursive exclusions. Facing Briet’s antelope in a zoo, the visitor is invited to speak, to affirm connections between this document and others, whereas, facing the objects collected in cabinets of curiosities, “[w]ide-eyed with wonder and open-mouthed with surprise, the admiring visitor paid the collector the sincerest compliment of speechlessness” (Daston, 1988, p. 458). The speechless “wonder” that reflected the visitor’s proper, polite response to these curiosities operated along two discursively exclusionary lines, one scientific, aligning with the “theory-free facts” of a Baconian “neutral observation language” (Daston, 1988, p. 467), and the other social, aligning with the period’s “genteel male identity” (Swann, 2001, p. 78). Both lines draw limits and enforce silences rather than engender loquaciousness. Despite these general similarities, the details differ, and with them, the justifications for extending the concept of the document to these curious objects. In the first case, the curiosities operate very much as do the written reports of Baconian “matters of fact” found in the early issues of the The Philosophical Transactions of the Royal Society and the Society’s other publications. Much has been written on the writing techniques of what Lorraine Daston has called “the language of strange facts” (Daston, 1998), and Revisiting “what is a document?” 297 JDOC 65,2 298 what Steven Shapin and Simon Schaffer have called Robert Boyle’s “literary technology” (Shapin and Schaffer, 1985). The effect of such techniques is the production of matters of fact as singular, granular, theory-free, and unique, without any necessary connections between them. In the English anti-Artistotelian intellectual culture, connections between facts and theoretical statements were discouraged. “Part of what made strange facts strange”, write Daston and Katherine Park, “was that they baffled ready explanation” (Daston and Park, 1998, p. 240). The favoured and encouraged intellectual attitude was analogous to the polite speechlessness that complimented the virtuosi’s cabinets of curiosities. Although Daston (1988) notes that Bacon “was confident that these so-called miracles of nature would eventually ‘be reduced and comprehended under some Form or fixed Law’ and was contemptuous of those who ‘go no further than to pronounce such things the secrets and mighty works of nature, things as it were causeless, and exceptions to the general rules’,” she also observes: However skeptical Bacon might have been of the mentality that informed the cabinets of curiosity and wonder books that were their literary counterpart, he took his granular view of experience and of theory-free facts from their fascination with anomalies, their deliberate severing of connections by display, and their obsession with the brute ‘thing-ness’ of the objects. Thus, she goes on to say: [. . .] for the interim this natural history of ‘Singular Instances’ was to be a compilation of counter examples, of anomalies that challenged existing theories. Here the playful side of the cabinet of curiosities was put to work, its nominalism and diversity turned to scientific ends” (Daston, 1988, pp. 465, 466). The cabinets of curiosities, these strange, unique “wonders” that made polite speechlessness a virtue of gentlemanly company, were meant to resist the same temptations of generalization and theorizing as did written reports of the new, Baconian, scientific matters of fact. Continuities between curious things and a Baconian mode of documentation were strengthened by the recording and cataloguing practices of both the Royal Society’s “Repository” of its collection of physical objects, and the registers of the Fellows’ documentary scientific activities, that is, of their submitted books, manuscripts, and other kinds of papers (for the variety of such submissions, see Johns, 1998, pp. 483-487). Marjorie Swann (2001) recognizes these continuities in remarking that “the Repository served as an analogue to the Society’s register books”. The collected curiosities and the scientific papers were not distinguished in the Society’s recording and cataloguing efforts. On the contrary, they were part of the same documentary impulse, and they exercised their documentary power along the same, exclusionary line that inhibited the kind of evidentiary connections so important to Briet centuries later. I offer this case to gain some insight about how the stories we tell support extensions of the concepts of documents and documentation to new cases. We can begin a self-reflection on the path just taken by appealing to Wittgenstein’s response to the idea that all phenomena designated by a word (he talks about games, and numbers) must have something in common. This is the familiar text on family resemblances, where he says: And we extend our concept of number as in spinning a thread we twist fibre on fibre. And the strength of the thread does not reside in the fact that some one fibre runs through its whole length, but in the overlapping of many fibres (Wittgenstein, 1958, p. §67). In our example, we began with some remarks (one fibre) on Otlet’s and Briet’s concept of a document, in which the concept of evidence was yoked to the prevailing concept of the document, thereby extending it to include physical objects. Otlet’s and Briet’s connection between the concept of evidence and object (second fibre) led to the idea that documents invite us to speak, to make statements about the manifold of phenomena for which the document-as-thing provides evidence (third fibre). The idea of relationship between the document-as-thing and the production of statements led us to the concept of things that restrict, limit, and control speech rather than engender it (fourth fibre). The concept of evidence loses strength here because the strange, granular things comprising the curiosities in the cabinets of the virtuosi were properly appreciated by resisting evidentiary claims. If the curiosities stand as evidence of anything, it is a curious kind of evidence, because in an anti-Aristotelian, anti-deductive and anti-theoretical intellectual climate they stand simply for themselves, for their brute, granular, “thing-ness,” severed from any sort of claim that would connect them to others. We then noted another fibre: the curiosities and the manuscripts, books, and other papers of the Fellows of Royal Society of Restoration England are recorded and catalogued as intellectual resources for the development of a science of singular, granular, and Baconian matters of fact by the same documentary procedures. We thereby see that the objects to which we propose extending the concept of “document” perform some of the same significant functions, have some of the same significant effects, and exercise some of the same significant powers, as items we accept without question as documents. We see them aligned with documents in a Baconian science of “theory-free facts.” It is by twisting these fibres one upon the other that the concept of the document gets extended. Such an extension is not arbitrary, any more so than Putnam’s joke, which extends the concept of the rapport between Christians and lions. If the fibres remain invisible, we are left wondering what the relationship between the two deployments of “document” might be. But if the fibres are laid bare in the stories we tell, our extensions are supported. A second line along which curiosities exercised documentary power consists in their intersection with forces driving the construction of the English virtuoso identity. The urgency to fortify the identity of the English landed elite emerged from a crisis of the aristocracy that Swann notes was especially acute in the Stuart period, when “there were far more aspirants for court office than positions available, leaving most genteel would-be courtiers unemployed and thus stranded in the country on their estates” (Swann, 2001, p. 77). The social superiority of this self-styled elite was maintained by constructing a distinction between it and other groups aspiring to social, economic and political power. The distinction drew upon an already existing trope of the virtuoso as a man whose character exhibited a “highly developed sense of curiosity” (Swann, 2001, p. 76). “‘Curiosity’ and ‘curiosities’,” Swann notes, “were seventeenth-century class markers”. Such curiosity was of the leisured kind appropriate to landed wealth, rather than the instrumental and practical curiosity of the engineer, craftsman, or skilled tradesman. The cabinets of the virtuosi displayed the fruits of a leisured curiosity that distinguished gentlemen from social upstarts – who anyway could not afford these rare, unusual, and unique objects. The curiosities served this purpose because they Revisiting “what is a document?” 299 JDOC 65,2 300 eschewed any practical or instrumental purpose. As Swann puts it, “the conjunction of collecting and class consciousness central to the ethos of ‘curiosity’ among seventeenth-century English vituosos made a historically specific form of cultural capital out of a fundamental quality of the collected object: its nonutility” (Swann, 2001, p. 78). Insofar as the characteristics of the collected objects became those of the collector (curious objects are collected by curious men), the object’s nonutility inscribes a characteristic of the collector: a man of leisured, landed wealth, whose nonutility and refined outlook marks his social status. Analogous to the curiosities’ exercise of a limiting and exclusionary power in alignment with Baconian facts, when documenting social status they situate their collectors through nonutility, an isolating and enclosing power, which enforces limits, exclusions, and distinctions. In extending the concept of documentation to curiosities along this second line, we can appeal to Bruno Latour’s “sociology of a few mundane artifacts” (Latour, 1992), which argues for non-human agency – even moral agency – including automobile seat-belt systems, automatic door-openers, doors themselves, and their hinges. He offers this methodological principle: “As a . . . general descriptive rule, every time you want to know what a non-human does, simply imagine what other humans or non-humans would have to do were this character not present” (Latour, 1992, p. 229). Here is some of what Latour says about doors: Walls are a nice invention, but if there were no holes in them there would be no way to get in or out – they would be mausoleums or tombs. The problem is that if you make holes in the walls, anything and anyone can get in and out . . . So architects invented this hybrid: a wall hole, often called a door, which although common enough has always struck me as a miracle of technology. The cleverness of the invention hinges upon the hinge-pin: instead of driving a hole through walls with a sledge hammer or a pick, you simply gently push the door (I am supposing here that the lock has not been invented . . .); furthermore – and here is the real trick – once you have passed through the door, you do not have to find trowel and cement to rebuild the wall you have just destroyed: you simply push the door gently back (I ignore for now the added complication of the “pull” and “push” signs) (Latour, 1992, p. 228). We can adapt Latour’s principle to determine the documentary agency of the curiosities along this second of our two lines. What if the curiosities were not there? What would other humans or other non-humans have to do to achieve the same effects? We can easily imagine an official, paper document, perhaps bearing a royal seal, that authorizes and is recognized to authorize its holder as having been granted rights to any and all of the respect and privileges of the esteemed social distinction sought by the virtuosi. But that is not enough. We also have to imagine all the work involved in petitioning the royal courts for such a document, in the lengthy court proceedings to prosecute the insubordinate and seditious mavericks, who pay no attention to it, and a host of other institutional set-ups or alignments. Once those are all in place, our imagined document might have the intended effect. Yet historians of the period have argued that curiosities, these small useless objects carefully collected and displayed by these useless men achieved much of what our imagined document might do. Of course, their powers also depend upon set-ups, alignments, or assemblages. But in the end, they do what the imagined document does. When we compare the effects of the curiosities with those of what we would clearly take to be documents, there seems no reason to withhold the concept of documentation from them. Conclusion I have argued that at least some of the reasons often given for a generalized philosophical demand for definitions of the words we use are questionable, following Putnam’s arguments against orthodox Wittgensteinians on the issue of criteria and rules governing our uses of words. Extensions of existing concepts are not illegitimate by virtue of violating existing meanings, definitions, or rules, because our use of language doesn’t depend upon them. Of course, we invoke such entities when we argue for a particular way of talking, but in so doing we are engaged in policing behaviour, in trying to get others to as we do, for which we should feel obliged to offer much more justification than mere philosophical arguments from the philosophy of language can provide. As Zirelli (2001) notes, there are moral issues here. When extensions of concepts are supported by thought, reasons, and argument – sometimes very profound reasons, as Putnam’s example of Einstein shows – we need not fear a generalized linguistic anarchy. Once we are in the realm of argument, good arguments can win over bad, provided, of course, that the requisite social set-ups for a space in which rational arguments count are obligingly aligned. Such alignments, of course, raise issues of politics and power. Reflections such as these lead us to the question of how to think about what we are after in defining documents and documentation. Starting with definitions signals an approach to documentation where the aims are precision, accuracy, and a “scientific” representation of what these concepts denote. Such appeals to definitions – or rules of language, or meanings – can always be defended, but it helps when their intended purposes are honestly laid out, because that indicates their range of application. As the last pages of Andrew Pickering’s Mangle of Practice (Pickering, 1995) charmingly show, the temptations of a Theory of Everything are often irresistible. But there are other approaches to documentation, which have as their aims not so much the precision and accuracy of a scientific representation of what documents and documentation might be, but forging concepts in a Deleuzian spirit, seeking to enhance their power and force, with more concern for what they do than for what they mean or represent. The benefits of extending the concepts of document and documentation are located here, and with a closely associated aim, that of multiplying these concepts and seeking ways of also extending an encouraging hospitality to many different areas of their application. To succumb to an over-used Deleuzian concept, the reorientation suggested here to the question, “What is a document?”, is more like seeding rhizomes than growing a tree. Note 1. The Department of Documentation Science, University of Tromsø, and the Document Academy, in its annual conferences, for the first four years at the University of California, Berkeley and now at the University of Wisconsin-Madison, are the primary sites of the resurgence of scholarly interest in documentation, sometimes called the “new-“ or “neo-documentation”. Much of the credit in library science and information studies must go to Michael Buckland and Niels Windfeld Lund for their work for the Document Academy conferences, and for Lund’s pioneering work on documentation in Tromsø. Boyd Rayward’s work on documentation for over 40 years is indispensable; a recent example is his conference, European modernism and the information society: Informing the present, understanding the past (held at the Graduate School of Library and Information Science, University of Illinois, Urbana-Champaign, May 6-8, 2005), and his anthology of papers from Revisiting “what is a document?” 301 JDOC 65,2 302 it, of the same name (Rayward, 2007). For a recent example of anthropology’s interest in documentation, see Annelise Riles’s anthology, Documents: Artifacts of modern knowledge (Riles, 2006). For my own work on documentation, see Frohmann, 2001, 2004a, 2004b, 2007a, 2007b. References Buckland, M. (1991), “Information as thing”, Journal of the American Society of Information Science, Vol. 42 No. 5, pp. 351-60. Buckland, M.K. (1997), “What is a “document”?”, Journal of the American Society for Information Science, Vol. 48 No. 9, pp. 804-9. Daston, L. and Park, K. (1998), Wonders and the Order of Nature, 1150–1750, Zone Books, New York, NY. Daston, L. (1988), “The factual sensibility”, Isis, Vol. 79 No. 3, pp. 452-67. Daston, L. (1998), “The language of strange facts in early modern science”, in Lenoir, T. (Ed.), Inscribing Science: Scientific Texts and the Materiality of Communication, Stanford University Press, Stanford, CA, pp. 20-38. Day, R.E. (2001), The Modern Invention of Information: Discourse, History, and Power, Southern Illinois University Press, Carbondale, IL. Frohmann, B. (2001), “Discourse and documentation: some implications for pedagogy and research”, Journal of Education for Library and Information Science, Vol. 42 No. 1, pp. 13-26. Frohmann, B. (2004a), Deflating Information: From Science Studies to Documentation, University of Toronto Press, Toronto, ON. Frohmann, B. (2004b), “Documentation redux: prolegomenon to (another) philosophy of information”, Library Trends, Vol. 52 No. 3, pp. 387-407. Frohmann, B. (2007a), “Multiplicity, materiality, and autonomous agency of documentation”, in Sklare, R., Windfeld Lund, N. and Vårheim, A. (Eds), A Document (Re)Turn: Contributions from a Research Field in Transition, Peter Lang, Frankfurt, pp. 27-39. Frohmann, B. (2007b), “The Role of Facts in Paul Otlet’s Modernist Project of Documentation”, in Rayward, W.B. (Ed.), European Modernism and the Information Society: Informing the Present, Understanding the Past, Ashgate, Aldershot, pp. 75-88. Johns, A. (1998), The Nature of the Book: Print and Knowledge in the Making, University of Chicago Press, Chicago, IL. Latour, B. (1992), “Where are the missing masses? The sociology of a few mundane artifacts”, Shaping Technology/Building Society: Studies in Sociotechnical Change, MIT Press, Cambridge, MA, pp. 225-58. Mill, J.S. (1868), A System of Logic, Ratiocinative and Inductive: Being a Connected View of The Principles of Evidence, and the Methods of Scientific Investigation, 7th ed., Longmans, Green, Reader, and Dyer, London. Nagl, L. and Mouffe, C. (2001), The Legacy of Wittgenstein: Pragmatism or Deconstruction, Peter Lang, Frankfurt. Pickering, A. (1995), The Mangle of Practice: Time, Agency, and Science, University of Chicago Press, Chicago, IL. Putnam, H. (2001), “Rules, attunement, and “applying words to the world”,”, in Nagl, L. and Mouffe, C. (Eds), The legacy of Wittgenstein: Pragmatism or Deconstruction, Peter Lang, Frankfurt, pp. 9-23. Rayward, W.B. (Ed.) (2007), European Modernism and the Information Society: Informing the Present, Understanding the Past, Ashgate, Aldershot. Riles, A. (Ed.) (2006), Documents: Artifacts of Modern Knowledge, University of Michigan Press, Ann Arbor, MI. Shapin, S. and Schaffer, S. (1985), Leviathan and the Air-pump; Hobbes, Boyle, and the Experimental Life, Princeton University Press, Princeton, NJ. Swann, M. (2001), Curiosities and Texts: The Culture of Collecting in Early Modern England, University of Pennsylvania Press, Philadelphia, PA. Wittgenstein, L. (1958), in Anscombe, G.E.M. (Ed.), Philosophical Investigations, Macmillan, New York, NY. Zirelli, L.M.G. (2001), “Wittgenstein: between pragmatism and deconstruction”, in Nagl, L. and Mouffe, C. (Eds), The Legacy of Wittgenstein: Pragmatism or Deconstruction, Peter Lang, Frankfurt, pp. 25-41. Corresponding author Bernd Frohmann can be contacted at: frohmann@uwo.ca To purchase reprints of this article please e-mail: reprints@emeraldinsight.com Or visit our web site for further details: www.emeraldinsight.com/reprints Revisiting “what is a document?” 303 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.