Wittgenstein, Language and Information: “Back to the Rough Ground!”

Citation

Hjørland, B. (2007), "Wittgenstein, Language and Information: “Back to the Rough Ground!”", Journal of Documentation, Vol. 63 No. 2, pp. 281-286. https://doi.org/10.1108/00220410710737222

This book is the second by David Blair on Wittgenstein and information science; the first one came in 1990 (Blair, 1990). Blair has also contributed an ARIST chapter on this subject (Blair, 2003). This new book is first and foremost characterized by a close reading of the collected works of Ludwig Wittgenstein and a presentation of that work as well as a discussion of its importance for information science.

Structure of the book

The book is organized in three parts: I: Introduction (26 pp.); II: Wittgenstein's philosophy of language and mind (242 pp.); and III: Wittgenstein, language and information (90 pp.).

According to the table of contents, there should be an index on page 359. However, there is no index, not on page 359 nor on any other page. Also, there is no bibliography or list of references (only references as footnotes, scattered throughout the book). This lack of a list of references makes it difficult to overview and checks what sources Blair has relied on in writing the book. The lack of an index adds to those difficulties.

In the preface it is stated that a series of detailed footnotes make out a fourth part of the book. This is a little strange since the footnotes enlarge on the issues discussed at the place where they are put. They are integrated in the three parts rather than forming a fourth part. Also, the absence of an index makes their content rather unobtainable, especially when they are regarded a separate part of the book. It is argued (p. 2) that the footnotes allow the reader to identify the exact source of the quotations. It should, however, be regarded a banality in an academic text that precise bibliographic information is provided for each quote.

It is stated that the (first) three parts of the book can be read by itself with some profit, although they are related and do form a conceptual whole.

The book is not organized in chapters! Just the three part with different levels of sections and subsections. This structure makes it difficult to comprehend, although there is a detailed table of contents (with up to four levels of headings). It is often difficult to follow where one line of arguments finishes and a new one begins. In other words: the structure seems messy.

The introduction raises questions (and provide answers) such as “why is the study of language important for information science?”, “why is the study of philosophy?”, and “why is the study of Wittgenstein?” This part, together with the third part, is in particular what makes this book a contribution to information science. I shall return to this below.

Part II: Wittgenstein's philosophy of language and mind

Almost 68 per cent of the book is thus about Wittgenstein's philosophy per se. This field is rather competitive. British Books in Print lists 544 titles with the keyword WITTGENSTEIN of which 333 mention Wittgenstein in the title. Wittgenstein is probably the philosopher on whom most books are currently written.

Although this chapter is well documented and reasonably argued, one wonders what the intension is in relation to the already overwhelming literature by and on Wittgenstein. Is it intended as an introduction especially written for information scientists? Is the purpose to be a stand‐alone introduction to substitute the reading of Wittgenstein's own works and other authors' writings on Wittgenstein? Or is intended as a scholarly examination and discussion of Wittgenstein's philosophy of language and mind?

I do not believe that the book under review is the best choice of books on Wittgenstein. Recently I bought Hanna and Harrison (2003), which I believe provides more clear answers to some core questions such as “the meaning of ‘meaning’” and the relations between words and reality. I also believe that Hanna and Harrison provides a better job relating the thoughts of Wittgenstein to other important philosophers and linguists.

Concerning the possible role of the book as a scholarly discussion of Wittgenstein, the book misses much important criticism and debate. One of the most famous concepts of Wittgenstein is the concept of language‐games. Sutcliffe (1993, p. 42) writes from the point of view of Aristotelean theory:

2.2.1 Wittgenstein on “family” and “family resemblance”.

Wittgenstein (1953, pp. 31‐2), having had difficulty specifying defining conditions for the class language (language games), gave up the search for necessary and sufficient conditions, and then (without proof) asserted that:

These phenomena have no one thing in common which makes us use the same word for all… You will not see something which is common to all, but similarities, relationships, and a whole series of them at that … We see a complicated network of similarities overlapping and criss‐crossing: sometimes overall similarities, sometimes similarities of detail. I can think of no better expression to characterize these similarities than “family resemblances” for the various resemblances between members of a family.

After a thorough discussion of this problem, Sutcliffe (1993, p. 48) concludes: “The explanation of Wittgenstein's difficulty, then, is that in concentrating on ‘family resemblances’ he remained in the wrong context A₁, when, to find the needed genus‐definition for the monothetic concept language‐games, he should he should have shifted to context A₀ within which one can state the conditions which set off language‐games as a class from other things which are not language‐games”.

The criticism raised by Sutcliffe (1993) seems very important, why the omission of this point of view is regrettable. This is just one example of important discussions of the status of Wittgenstein's philosophy that have been neglected by Blair.

Concerning the possible role of the book as an introduction written for information scientists, the final part of part II about Wittgenstein and behaviourism may serve as an example that illuminates Blair's dilemma in writing this book. Although I find the section about behaviorism well argued, it is neither an introduction for information scientists nor a scholarly examination of Wittgenstein's relation to behaviourism. If it should be an introduction for information scientists, is should demonstrate how behaviourist norms are relevant from the perspective of information science, that behaviourist norms are still influential in the literature of information science. I think they are, but I do not think that information professionals can see how this is the case without philosophical guidance, which should be given in philosophical books written for information professionals. In other words: I find the critique of behaviourism discussed by Blair important, but in need of specific exemplification within information science. If, on the other hand, it should be a scholarly contribution is should relate to the literature about Wittgenstein and behaviourism (e.g. Bloor, 1999) and go into a closer dialog with the whole literature on that issue.

In the rest of this review I will concentrate on what seems most important, the implications for information science, i.e. parts I and II (which together is a lesser part of the book).

Part I: Introduction

One of the important questions raised in this part is: Why Wittgenstein?

Blair argues that language and meaning are important to the study of information systems. But this view is not followed by an analysis of how current research in information science is done, and the implications that a Wittgensteinian perspective has. What research is neglecting language and meaning? Is philosophy really needed or may, for example, statistical and bibliometrical approaches solve the problems without the need to study philosophy in general and Wittgenstein in particular? Although I agree with Blair that theories of language and meaning are neglected and are potentially very important; I still find the argument put forward rather vague.

I also find that Blair ignores some research, which try to make connections between language and philosophy on one side and information science at the other side. There is a growing interest in information science for, among other things, semiotics, social semiotics, pragmatism, activity theory, discourse theory, and genre theory. These are approaches that emphasis language and meaning and a social perspective, and thus share many of the qualities that Wittgenstein may provide, but they are not presented or compared to the potentials of Wittgenstein's philosophy. Blair's book is not the place to find a discussion of different philosophies of language and their relative strengths and weaknesses and is therefore not able to provide strong arguments for a Wittgenstenian perspective.

This first part also contains a presentation of 12 principles from Wittgenstein and a discussion of the implications for information systems. Graphically these twelve principles are not clearly separated in the book, although each principle is numbered and conclude with a headline in bold “relevance to Information Systems”.

The idea of presenting a selection of principles and discussing their implication to information systems is good. However, the result is disappointing. Only the first principle provided by Blair will be presented and discussed here:

1 …we don't start from certain words, but from certain occasions or activities (LC p. 3).

I agree that this is an important principle (which Wittgenstein shares with, for example, pragmatic philosophers such as John Dewey). What implication does Blair make of this statement from Wittgenstein? He writes (p. 10): “The underlying order of information systems, in so far they are linguistic systems, is not so much words and categories, but ‘occations and activities’. Yet often it is the relation to ‘occasions and activities’ that is lost when information is organized for retrieval, especially when it is placed in computerized retrieval systems”. Blair finds that “Consequently paper‐based information can remain close to the activities that produce or use it, and these activities can provide an interpretative context for than information, but when that information is computerized, the very act of computerization may have the effect of removing the information from the activity context that provides much of its meaning and interpretation”.

Such a conclusion is in my opinion frustrating. First, is Blair suggesting that we study philosophy of language in order to give up computerization? The underlying motive should rather be, I suggest, to show how computerization might be improved. Second, the conclusion is disappointing in making computerization the cause of loosing interpretative context. Hjørland (1998, p. 27) suggested that the repacking or merging of information in different levels, e.g. in journals and databases remove some contexts, which may however be reestablished. It is the identification and reestablishing of meaning lost during merging of information, that may be a way forward. In this way may computerization be a tool for progress rather than an obstacle.

Blair's other 11 examples are, in my opinion not much better. At the specific and concrete level, his suggestions seem not much helpful. The third part of the book provides other suggestions, to which I shall return below.

Part III: Wittgenstein, language and information

In this part the reader is eager to learn about the implications of Wittgensteins philosophy for information science. However, the start of this part consists of a summary of Wittgenstein's philosophy of language and a discussion of issues such as scaffolding and mental models. It takes yet about 16 pages to reach the beginning of a discussion of information systems.

Pages 294 and 311 is about the nature of representational indeterminancy in information systems, pp. 311‐46 is about the consequences of indeterminancy in information retrieval, while the rest of the book (p. 346‐58) is about practical consequences (this structure is produced by the reviewer and is not clear from the table of contents, in which different levels of headings are not given different insertions).

One of Blair's main ideas (which was also expressed at the CoLIS5 Conference, see Blair, 2005) is that of partitioning large databases, e.g. by printing year, author, publisher, “principal activities of the users”, document type, “normative taxonomies”, “information usage” or previous searches. This idea is strange in two ways. First, this is just what classical databases have done all the time by using field codes and experienced searchers have utilized those codes. Second, such formal partitioning is not very much in line with Wittgenstein. Wittgenstein is about organic connections rather than about artificial kinds of separation. This idea does not really bring information science forward because it lacks suggestions on how to uncover underlying organic connections.

The idea that I believe has the best future is the idea of “exemplary documents” (also put forward in Blair and Kimbrough, 2002). This idea should be further developed, e.g. by examination how to identify exemplary documents, when is a document a good candidate to serve as an exemplar? Is the view of what makes a document exemplary related to different views?

Conclusion

I have tried to demonstrate that the editorial process in the production of this book has not been careful enough. The lack of a bibliographical reference to the quote by Douglas van Kirk on p. 6 is another indication of this.

I believe the philosophy of language is important and that Wittgenstein belongs to a family of philosophers who can bring information science in better shape by strengthening the social perspective in language, knowledge, information, documents and so on. Blair has evidently wanted this time to concentrate on a detailed study of one philosopher. This might be relevant for a philosophical dissertation, but I am not sure this is the best thing to do for information science.

Blair has done a heavily work by a close reading of Wittgenstein and by presenting and discussion it. We need badly people like Blair to work with such issues. However, the book ignores too much to relate to research in information science, both research based on related approaches (such as, for example, semiotics, activity theory, genre theory and pragmatism) and research on conflicting approaches

Blair talks about computer systems in a very general way and claims, that they are based on a narrow logical design principle that is problematic according to a Wittgenstein‐perspective. I do not believe that Blair will convince hard core IR‐researchers that they might gain by considering questions related to language, meaning and mind. A first problem is that computer based retrieval are often based on natural language representations. We have Hypertext‐representations, full‐text‐databases, citation indexes and we have bibliographical records in which author‐generated information is mixed with indexer‐based information such as descriptors and classification codes. We have indexing made by generalists and by subject specialists etc. I have difficulties recognizing Blair's description of narrow‐logical based computers in the real world.

In a similar way talks Blair much about ordinary language (and Wittgenstein's points regarding language) in a very general sense. No distinctions are made, for example between the language of experts and novices, between the medical language in MEDLINE records on the one hand and questions posed by experts or novices on the other hand. In my opinion are both the problematic generalizations about computer logic and the problematic generalization about ordinary language a serious limitation in this book. I really doubt if the straw man that Blair uses Wittgenstein to attack has much to do with real life information systems.

It is important that we can argue that traditional approaches to information retrieval have limitations that can be overcome by the study of meaning and language. There is a need for a third book by Blair about Wittgenstein and Information, in which, for example, underlying behaviorist conceptions in information science are brought into the light and confronted with the criticism provided by Wittgenstein and related thinkers.