After Epistemology

Citation

Bade, D. (2011), "After Epistemology", Journal of Documentation, Vol. 67 No. 1, pp. 194-200. https://doi.org/10.1108/00220411111105515

Ten minutes ago I finished reading After Epistemology (in one sitting) and after the thrilling experience of all 187 pages (I even read the bibliography and index item by item) I have already asked myself if I am not entirely lacking in critical intelligence, for my only criticism is that it was only 187 pages long. I would have liked to read 187 or 1870 pages more. Perhaps an apology is in order for yet another rave review of a book by Professor Harris, but I have none to offer.

After Epistemology was even more of a surprise and even more enjoyable that I expected – and I always expect to be pleased and surprised by Harris – for some very personal reasons. First, some of the main arguments Harris makes and the conclusions he draws recalls those made – albeit in much less clear and concise a manner – between 50 and 100 years ago by two nearly unknown philosophers (Owen Barfield and Eugen Rosenstock‐Huessy) whom I have long admired but from whose ideas I was never able to discern “what next?” until now. Second, I have for a long time wished – and with my reviews secretly hoped to inspire – a book by Professor Harris on information science from the perspective of his philosophy of language. I assumed that to ask him directly would draw his response “Isn't that your task?” But here we have that book, and as I should have expected, he treats the topic with an entirely unexpected approach. And finally, there is page 172 on dancing. Why that matters so much to me will, however, remain personal.

Traditional epistemology, Harris writes in his Preface, “allows no scope for recognizing that the forms of knowledge recognized and valued in any society depend on the forms of communication practiced in that society” and that “Forms of communication make a crucial difference not only to what is known, but to what is knowable” (p. 2). McLuhan, Harris notes on p. 118, did see this, but, as Harris argued in Rationality and the Literate Mind, he drew the wrong conclusions. So the book begins where Harris left off in that book also published last year (2009).

The first eight chapters of the book (“Part I: epistemology in the Western tradition”) discuss the challenges to the ancient Greek philosophy of language underpinning epistemology as discussed in the West, beginning with Bacon's “idols of the market”. Those idols were, for Bacon, “names of things which do not exist … [and] names of things which exist, but yet confused and ill‐defined” (pp. 24‐5, quoting from the Novum Organum). Harris then proceeds to Wilkins, Locke and Hume, and then surprisingly, Darwin. In his altogether unexpected discussion of Darwin it is not the influence of early nineteenth century theories of linguistic evolution on his theory of evolution that Harris focuses, but Darwin's reluctance to define two key terms: species and instinct. Here he quotes Darwin's remark in The Origin of Species:

[…] in determining whether a form should be ranked as a species or a variety, the opinion of naturalists having sound judgment and wide experience seems the only guide to follow. We must, however, in many cases, decide by a majority of naturalists, for few well‐marked and well‐known varieties can be named which have not been ranked as species by at least some competent judges (Darwin, quoted by Harris on p. 50).

To this Harris responds “Leaving aside the question of how ‘competence’ in such a matter is to be determined, counting heads hardly seems to be a shining example of scientific method.” I single out this passage not only because it is a marvelous example of Harris' critical perspicacity, nor because his wit sometimes makes me double up in laughter, but more importantly for readers of this journal because of the implications of this witticism for the notion so prevalent these days of “the wisdom of the crowd.” Harris goes on to quote Darwin's concluding remarks on the term species as being a term “arbitrarily given, for the sake of convenience” (p. 53) and asks “why, if Darwin believes this to be so, he does not declare species to be worthless as a scientific term and reject it” (p. 53).

Part One concludes with three twnetieth century challenges to traditional epistemology, those of A.J. Ayer, Saussure and Harris' own approach, integrationism. He sums up his discussion of classical epistemology by remarking that it:

[…] treats knowledge as dealing with matters that are public, communicable and debatable. This move rests ultimately on the assumption that a public language is or can be put in place, on which discussion of these matters can be based. So Classical thinking in practice imposes an epistemological hierarchy, in which linguistic knowledge takes priority over non‐linguistic knowledge (p. 78).

While critical of this view of knowledge, he acknowledges that this is “a great advance on the more primitive view of knowledge as dealing with ineffable mysteries” (p. 79). Western ideas about that public language, whether imagined as being imposed on us from without (as prior consensus) or from within (some innate linguistic program in the brain) have always “ignored or marginalized the overriding role of the communicational context” (p. 80).

In Harris's view – the integrationist perspective – “there is no such thing as a decontextualized sign. What a dictionary lists as ‘words’ are decontextualized abstractions. Signs exist only in the particular circumstances in which they function as signs. Consequently the theoretical focus of linguistic inquiry shifts from using signs to the making of signs” (p. 80). The implications of this for epistemology are far‐reaching:

Just as every sign presupposes a context, every item of knowledge presupposes a context. There are no free‐floating, contextless items of knowledge. There are no processes of knowing that exist independently of what is known. Knowledge, thus understood, is a form of activity. Most importantly of all, this activity is seen as a constructive, creative process. It is no a passive accumulation or residue of any kind. The creative process requires an active engagement of the self. It is not ‘triggered’ automatically. It cannot be undertaken by the collectivity on behalf of the individual (pp. 80‐81).

And with this reconception of knowledge and what it means to know, Harris launches into “Part II: integrating knowledge”, which is in large part a critique of that understanding of knowledge which reduces, confuses or identifies it with information, this latter being, he writes, “the static, inert residue to which knowledge dwindles when subjected to persistent and systematic reification” (p. 116). That description of the relation between knowledge and information is “from the top down” so to speak. He also gives a description “from the bottom up”:

[M]aking use of information usually demands of the user (i) the interpretation of some static form in which the information appears, and (ii) that this interpretation be integrated into some further process of knowledge‐making in an appropriate context. Appreciating the potential utility of information is no more excuse for confusing information with knowledge than for confusing a symphony score with an orchestral performance. Information, of its own accord, does not somehow convert automatically into knowledge. Knowledge, like music, is made and in the making of knowledge, as in the making of music, the systematic integration of activities lies at the heart of the creative process (pp. 141‐2).

Harris' discussion of information as opposed to knowledge runs throughout Part Two, but is most directly treated in chapter 13 (“Knowledge and information”). In that chapter he clarifies that he is referring not to “the technical term information as used in mathematical ‘information theory’” (p. 116, referring to Colin Cherry's (1957) On Human Communication) nor to the same term used in linguistics (referring to Halliday's “systemic linguistics”), but to “information in the lay sense in which one consults a train timetable … or a telephone directory” (p. 116). He focuses first on the role of codes “in the reduction of knowledge to information” since these “‘work’ by imposing systematic simplifications on whatever is to be ‘encoded’” (p. 116). Codification attempts, he writes “to split knowledge into conveniently small encodable items, determined not by what is known but by the structural requirements of whatever code is being used. The code, in other words, imposes its own analysis and organization upon any materials encoded” (p. 117) and thus “what began as knowledge them becomes mechanized: it is progressively divorced from the creative processes that brought it into being and the human values that presided over its creation. The systematic accumulation of mountains of codified items can safely be left to machines (or librarians), for it is a purely mechanical operation” (p. 118).

These remarks on codes and codification are followed by a discussion of a recent book on knowledge management which he refers to as “This little sermon for our times”: Wenger et al.'s (2002) Cultivating Communities of Practice: a Guide to Managing Knowledge, published by the Harvard Business School. Harris quotes a couple passages, including the following in which the authors “rationalize the tricky connexion between knowledge and communication”:

You know that the earth is round and orbits the sun, but you did not create that knowledge yourself. It derives from centuries of understanding and practice developed by long‐standing communities. Though our experience of knowing is individual, knowing is not (Wenger et al. (2002), quoted by Harris on p. 120).

Harris then describes their epistemological presuppositions this way:

You, as an individual, know nothing. Knowledge is something outside you, to which you can only gain access (if you are lucky). You gain that access by means of communication with an appropriate source. But that to which you might (with luck) gain access has been put in place not by you, but by a historical consensus in the past. It is already at one remove from “the facts” (e.g. the shape and movement of the earth). Here we see in vivo what an integrationist would regard as the reduction of knowledge to information (p. 120).

The passage from Wenger et al. that Harris quotes leads him to ask “some obvious undergraduate questions”:

Presumably people in the past might have been wrong. So why rest content with that account of knowledge? And if that account of knowledge were right, it is hard to see why knowledge has just not remained unchanged for centuries. If knowledge equates with “the consensus of previous generations”, how could we ever come to know something our ancestors did not? (p. 121).

Of this “knowledge management” view of knowledge Harris concludes that it “takes knowledge all the way from the laboratory to the market place without ever allowing one crucial question to be raised: has the journey had no effect on the knowledge itself? Is what ends up in the commercial market place any longer knowledge at all?” (p. 122).

Harris then turns to “theorists who speak as if a great deal of information were somehow already present externally in what is observable” (p. 122) such as Eysenck and Keane (1995) who write of the information in a photograph. Of this view Harris remarks that photographs are not “repositories of information, still less of knowledge. It is the human observer who creates knowledge by interpreting the photograph … on the basis of certain assumptions about how” the photograph was produced. Perhaps, he suggests, this is a recent “misapplication of the ancient doctrine of natural signs” (p. 123). But “Smoke does not ‘mean’ fire. It is human beings who interpret smoke as a sign of fire” (p. 123).

It is in his discussions of metaphor, literal meaning and perception that Harris arrives at fundamentally the same claim that Owen Barfield made in various publications between 1926 and 1957: there is no literal meaning, and perception of a tree (for example) as a tree (rather than as some hitherto unknown object) involves an interpretation. In Harris' words “it is vain to look for an absolute distinction between the metaphorical and the non‐metaphorical” (p. 144) and there are only more or less successful “attempts to establish fixed codes” (p. 133) and to communicate with others. Regarding sense perception Harris writes “sense perception as such is not knowledge. Saying ‘I see a birch tree’ already goes beyond sense perception. Knowledge in such cases begins by identifying what is seen, … that already presupposes interpretation of sensory experience” (p. 94). The implications of this for any theory of signs is quite straightforward, a conclusion that Harris draws but Barfield did not: the identification of a sign as a sign is the act of contextualization and the making of the sign in one and the same act (p. 103).

At the end of the book Harris argues that we must “start thinking of words and meanings as spontaneous creations of the mind which function as tools for the contextualization of those activities in which you are engaged” and that this “requires abandoning the old idea that whereas knowledge may make a great deal of difference to the knower, it ‘makes no difference to what is known’” (p. 176, quoting Price, 1963, p. 23). “That conception of knowledge” Harris insists, “must go” (p. 176). Here again I hear echoes of Rosenstock‐Huessy and Barfield. (For me, the only intellectual pleasure greater than finding agreement among those who inspired my youth and those who inspire my old age would be to have them argue their disagreements face to face. That of course is no longer possible.)

It is not just Barfield and Rosenstock‐Huessy whose valuable ideas one can find here. In the notion of local knowledge Geertz is not far away, and his discussion of perception seems to present with an extraordinary concision the most valuable insights of Heinz von Foerster, Humberto Maturana and Francisco Varela. Integrationism itself could be understood as a more radical and more profound view of the world of living beings living together than that developed by von Bertelanffy, Bateson, Wiener and other figures in systems theory and cybernetics. Yet none of these authors appear anywhere in the book; Harris responds to Aristotle, Austin, Bacon and Black, Bruner and Darwin, Eysenck, Gombrich, Lakoff, Langer and Locke, and Vico, Vygotsky,Wenger and Wittgenstein – among many others – by seizing on some basic presupposition and making his judgements and remarks directly in response to that. Harris does not acquire his knowledge second‐hand; he gets where he is going entirely through criticism – and boy does he go places!

One of the key conclusions that Harris draws is that “all communicable knowledge is local knowledge” because “signs do not exist independently of their integration into the particular activities that gave rise to their creation” (p. 97). If we accept this – or at least consider it – then the questions that epistemologists should be asking – but have not – are questions such as “What is the difference between knowing what time it is and not knowing?” … What is the difference between knowing the date of the Battle of Hastings and not knowing?” (p. 98). The answer, Harris claims, is in knowing how to act. In the real world nothing “can be known without involving a knower, and knowledge makes a difference to people's lives” (p. 99). Here we see Harris' fundamental disagreement with Popper and his “third world” of objective knowledge (which he does mention), as well as with philosopher‐librarian Gilbert Varet's notion of knowledge as being only that which is documented so well that it cannot be lost. (Harris does not mention it, but Science et son information à l'heure d'Internet (Varet, 2001) is an interesting book and Varet an author to whom I am indebted. Unfortunately he developed a ludicrous epistemology that only a librarian could have arrived at.)

Harris gives four “basic propositions” for the demythologization of knowledge, the last of which is “signs, and hence knowledge, arise from creative attempts to integrate the various activities of which human beings are capable” (p. 162). Reading this and many other passages in this book I could not help but think about Web2.0, tagging, the frequently expressed desire for more interactivity and user participation in the creation and management of information in libraries. Harris' discussion of art and photography are full of provocative discussions for anyone interested in graphic interfaces or the problems of searching and creating metadata for non‐textual digital objects. The claims made for and hopes pinned upon techniques and technologies such as the LCSH, authority files, DOIs and URLs, the Semantic Web, ontologies, data mining, and automated classification and subject analysis often came to mind as I read of the flaw in treating “knowledge as an aggregate of infinitely many separately knowable items, between which the only structure of organization recognized is the a priori structure imposed by the way we assign names and verbal descriptions to them” (pp. 45‐6). This book may be after epistemology but it also makes a strong case for “After Information Science.”

We have all seen – some of us way too many times – the DIKW diagrams: data, information, knowledge, wisdom. We have read too many discussions of how data differs from information, and information from knowledge, etc. We have believed that the more information and the more metadata the more knowledge there is and the better the world will be. We have read that there is too much information and that there is not enough information and that everything is information and that everything is metadata and that everything is miscellaneous. We have even been told that information is a “plastic word” that means anything and therefore nothing at all (Uwe Pörksen – and it is this view that has often seemed nearest the truth to me). What Harris has done in this book – whether he intended to or not – is to provide the rudiments of a coherent theory of information as a product of social activities. Instead of some metaphysical theory of platonic forms or some ontologically distinct form of entities alongside matter and energy, we have a notion of information as knowledge reduced to a sign awaiting its interpreter. Instead of definitional chaos we have information theoretically defined as that publicly available sign created for communicational purposes. The creation of knowledge and of information can be described in terms of a theory of signs, a semiology that is not limited to linguistic signs but that covers all manner of signs: gestures, silence, documents, road signs, signatures, advertising, painting, music, theater, propaganda, film, photography, clothing, nipple piercing, dance and even the archaic experience of nature “speaking.” And this goes not only for the writer but the reader as well, not only for the web site designer but for the internet browser, not only for the cataloger but for the library catalog user. From perception to interpretation to knowledge to creating a sign for communication to information buried in the stacks or cyberspace to information discovered, interpreted and made knowledge anew in a different time and place. It is an understanding of information that takes account of time and circumstance, understanding and misunderstanding, normative practices and creative activity.

On pages 164‐6 Harris offers a different scenario of actions involving perception, knowledge, information and communication, and his is not of levels as in the DIKW models, but rather a timeline of human activity. Taking the ideas from About Epistemology and ordering them for an information science we might arrive at something like this:

• •

  Perception – by creatures of a time and a place with a memory of interaction and engagement with other creatures of its own kind and others.

• •

  Interpretation – perception being active rather than passive, the very things we notice being selected for their relevance to our situation and task as we already understand it prior to the perception.

• •

  Knowledge – “we do have first‐order knowledge of our own everyday experiences, because we actually create our own interpretations of those experiences” (p. 166).

• •

  Communication – if the knowledge we have is to be communicable knowledge we have first to initiate that communicative activity.

• •

  Sign making – communication requires the making of publicly available signs.

• •

  Information – those public signs which are recognized as signs (in the extreme case this would be “found art” where the art is in the eyes of the beholder, not the creator).

• •

  Libraries, documentation, archiving.

• •

  Sign making by the reader, performer, audience, etc. – a text, painting, performance is informative only to someone who first understands it as a sign.

• •

  Interpretation – having understood something as a sign – which is to say having made a sign of something – the sign must be interpreted, and this will be done in a manner appropriate to the reader/hearer/observer's current situation, interests and program of thought or action.

• •

  Information again – having interpreted the sign as meaning something, that meaning is now available for intellectual and practical action.

• •

  Knowledge – information that has been successfully incorporated into some individual's intellectual or practical project.

• •

  Communication – the knower in his or her excitement sets out to write the book or book review to share her or his knowledge with the rest of the world.

Knowledge comes before data and information because “reasoning about the world is based on knowledge of the world, not vice versa” (p. 47). Harris' understanding of communicable knowledge is not tied to any particular forms of knowledge – bicycle riding, the difference between Constable's knowledge of painting as revealed in his paintings as opposed to that found in his lectures on painting are two of his memorable examples – forms of communication or technical systems. Instead of an information science that often seems little more than an advertisement for some new technical development (or potential development), we have a broad view of human communicative activity through the use of signs of all sorts and any tools that we may use to make those signs.

One of the major differences between Harris' view of language and communication and other notable theories is his emphasis on time. In The Semantics of Science Harris (2005, p. 117) stated “such knowledge as we have is always of change”. If fixity of meaning is to be sought in the external world (reocentrism) we “can only do so on the supposition that this world of external things, from which words derive their meaning, remains constant” (p. 120). Obviously such a solution is not only undesirable, but also impossible in LIS since we are constantly reminded that the only constant in libraries is change. Yet the recognition of change calls into question some of the techniques upon which our hopes for the future of libraries and information systems are pinned. Perhaps we should be asking ourselves how today's metadata and today's ontologies can be anything other than straitjackets for tomorrow's citizens?

Read After Epistemology and see what you think. Then let me know what you think, because I would love to get this discussion going.