Words Theoretical development of information science: a brief history Birger Hjørland
The diversification of viewpoints
Download 225.5 Kb.
|
- Navigate this page:
- 7. Too much theory
- 8. What influences the theoretical development of information science
- Table 1 . Information science’s theories and traditions. General philosophical level
- Meta-level/ paradigms: Information sciences’ paradigms and traditions
- General social science/humanities theory used in information science
- Theory level
- Application level (practical activities done by information specialists)
- 9. A social and cultural turn in information science
6. The diversification of viewpoints Information science seems today to be in an extremely diversified and fragmented theoretical position. There are many indication of this. Marcia Bates’ (2005, p. 10-14) overview of metatheories in the field briefly presents thirteen different approaches: (1) A historical approach (2) A constructivist approach (3) A constructionist or discourse-analytic approach (4) A philosophical-analytical approach (5) A critical theory approach (6) An ethnographic approach (7) A socio-cognitive approach (8) A cognitive approach (9) A bibliometric approach (10) A physical approach [which Bates exemplify with Shannon’s information theory] (11) An engineering approach (12) A user-centered design approach (13) An evolutionary approach. We cannot in this paper present each of these 13 perspectives. Bates’ own descriptions are very short and catalog-like8. The situation is, however, even more fragmented than these 13 different approaches suggest. Fisher, Erdelez and McKechnie (2005) have provided an introduction to ‘theories of information behavior’. It describes 72 ‘theories’. However, it is arguable whether all 72 are really theories of LIS? Some appear to be theories from other disciplines and others appear to be subfields rather than theories. Also, many theories are missing (e.g. information theory itself, quantum theory, (cf., Bawden, Robinson and Siddiqui in press) and evidence based library and information practice (cf., Booth and Brice, 2004)). This means that in spite of the broad coverage the book is not suitable for providing an overview of all the theories in information science. The number (72) indicates that LIS has become very fragmented: there are few overall views, but extremely many isolated and fragmented proposals. A third example is Leckie, Given and Buschman’s (2010) Critical Theory for Library and Information Science, which contains introductions to 26 ‘critical theorists’ presented in 23 chapters. A fourth example is Arnold (2013), an edited book published in the series Routledge Studies in Library and Information Science about Traditions of systems theory: Major figures and contemporary developments. There are no authors from LIS among the authors or editor, there is no chapter relating systems theory to information science, and the literature about systems theory from our field is not cited. We do have systems theory in information science and people believing in its fruitfulness. But these voices tend to disappear in the overwhelming number of other voices and of voices like Arnold (2013) in which one would expect them to be present. These four examples do not overlap much, and there is no sign that they together cover all the approaches, or even most of the approaches which have been put forward in the literature, as a little browsing easily produces many other theories. Thus, many theories may be considered to be ‘a flash in the pan’. Howard D White expressed a similar concern: I see the field of library and information science (L&IS) as highly centrifugal and greatly in need of high-quality syntheses. Library and information science has always been easy to enter by persons trained in other disciplines, particularly if they bring quantitative skills. The pattern has been many fresh starts by new entrants rather than strong cumulation. Nor is there full agreement as to which work is paradigmatic. Therefore, I would give warm encouragement to writers who show a talent for creative integration and criticism of ideas already embodied in the literature. Their efforts should indeed go into reading and organizing existing claims, rather than gathering new data (White, 1999, p. 1052). This fragmented state of the field may also be related to the high level of uncited papers in the field: Schwartz (1997) found that 72% of the articles published in library and information science are uncited within five years of publication. Under the heading ‘The diversification of viewpoints’ Wersig wrote: At the beginning of the 1970s, when information science started to establish itself, it was faced with the problem that while nearly everybody used the term information, nearly everybody meant something different by it. The problem was complicated by the fact that most of the users of the term thought that everybody else would understand and therefore they very often did not define which kind of meaning they had in mind (Wersig, 2003, p. 312). Wersig here echoes Machlup and Mansfield's suggestion that information science is ‘a rather shapeless assemblage of chunks picked from a variety of disciplines that happen to talk about information in one of its many meanings’ (Machlup and Mansfield, 1983, p. 22). Thus, one major cause of the fragmented cognition of the field is that the term ‘information’ is used in many fields and that anyone from any perspective feels free to consider himself or herself a part of information science without seriously relating to discourses that have taken place or are taking place in the institutionalized field of information science. Another major cause is the practicalist influence, which tends to reduce information specialists to people with ‘how-to’ knowledge about certain systems or information sources. As expressed by Buckland: One might have thought that, for so important a field, a general introduction would be easily written and redundant. This is not the case. Each different type of information system (online databases, libraries, etc.) has a massive and largely separate literature. Attention is almost always limited to one type of information system, is restricted by technology, usually to computer-based information systems, or is focused on one function, such as retrieval, disregarding the broader context. What is published is overwhelmingly specialized, technical, ‘how-to’ writing with localized terminology and definitions. Writings on theory are usually very narrowly focused on logic, probability, and physical signals. This diversity has been compounded by confusion arising from inadequate recognition that the word information is used by different people to denote different things. (Buckland, 1991a, p. xiii). I consider all these examples as indicating a serious condition in information science. Is it necessary to argue why? Can the opposite view: that such plurality is a good thing, be claimed with equal right? I do not think so. I would use the metaphor of a badly managed chatroom in which all participants are speaking in the mouth of each other with first listening to what has already been said. An argument for such pluralism has forgotten that the goal of a scholarly field is to provide theoretical clarity and that such a clarification is a collective responsibility. The accumulated knowledge of a field in its published literature represent much of its heritage and thus what we and our students are supposed to live by. Therefore it should be carefully nursed. At the present LIS seems not to be in a process where the relative fruitfulness and failures of different approaches are clarified over time. People who are interested in a given approach mostly fail to consider their own view in relation to other views, and they also mostly fail to present their view in relation to the kinds of problems information science is trying to solve. 7. Too much theory? “I consider myself a theorist. That is, my inclination is to theoretical argument, to achieving theoretical understanding, in information retrieval as in other realms” Robertson (2000). We have already considered Melvil Dewey and what we termed practicalism, an attempt to improve practice without deeper scholarly involvement. From the point of view of a research based field, it is disappointing that his DDC system today is the most used library classification in the world and that other systems, based on research, such as the Bliss Bibliographical Classification 2nd edition have not been able to gain foothold. Of course research and theories should make a difference, they are not developed just out of some interest in research and theory for its own sake. It is important to consider that much academic knowledge is not just explicit theory, but also, for example, procedural knowledge. Academics in languages are, for example, supposed to master the language, not just know theories of it. Information specialists are supposed to be able to use bibliographic databases professionally and master specific standards (e.g. metadata standards) etc. Theoretical studies and such practical issues should support each other. It should also be considered that theoretical knowledge may is often built into practical systems, and when you learn to use such systems, you learn implicit about the theoretical knowledge on which they are based. In the following quote Stephen Robertson makes a distinction between theory-driven versus pragmatic driven information science: “I have to admit that the field of information retrieval in which I have chosen to be a theorist is not a very theoretical one. This is true in two senses: in a negative sense, there are few strong theories in IR, and certainly no overall theory of IR, to which one might appeal to solve all difficulties. In a positive sense, the field is very strongly pragmatic: it is driven by practical problems and considerations and evaluated by practical criteria. Actually, the pragma of IR comes in two distinct forms. On the one hand, we have commercial pragmatism: IR, systems and services operate in the marketplace, and stand or fall by market forces - customer satisfaction, willingness to pay, competition etc. On the other hand, we have technological pragmatism: we design systems to perform certain tasks, and provided we have the ways and means to measure success or failure in the performance of these tasks, then we can try out mechanisms and techniques to our heart's content, selecting those that help in the pursuit of performance, and rejecting those that do not. Why they work or do not work is a secondary consideration” (Robertson, 2000, p. 1-2). That there are “few strong theories in IR, and certainly no overall theory of IR” is in my view something like an integration of social epistemological, semiotics, activity theory9 and domain analysis might contribute to develop (or in short: a point of departure in a social and cultural perspective). We should be aware, however, that such a strong theory may have to be modified by a large number of narrower theories, such as theories about specific languages (such as Swedish, cf. Hedlund, Pirkola and Järvelin, 2001), genres, domains etc. In other words, we should not in beforehand assume too universal theories. The great success of classical IR-approaches may be seen as easy pickings due to the establishing of huge databases (this said without disrespect for the highly qualified work done in the field). To improve things from here may involve more specific solutions based on a deep understanding of language, knowledge, genres etc. Also, I do not see an opposition between theories versus pragma. The purpose of theories is to be practical: “there is nothing as practical as a good theory” (Lewin 1943:35). A given theory may, for example, say how best to utilize a given technology. Robertson specifies the kind of theory he is skeptical about: Well, we have cognitive science, and linguistics, and epistemology or ontology, and probability and statistics, and probably other things. There is some tendency to regard these as alternative ways of looking at IR, but of course they are really complementary. And their complementarity resides precisely in the combination of low-level logic and experience. It seems unlikely that we can find a Grand Theory that will tell us exactly when we should be worrying about the linguistics and when, by contrast, we should take the linguistic entities we have identified at their face-value and treat them as statistical clues. I'm not claiming that such a theory is impossible - just that it's a tall order. This is not at all to say that the search for theory is futile - far from it. I believe that the models we have at present can indeed be extended, by theoretical argument as well as by both kinds of pragmatism, to cover more ground than they do at present and to be more useful as tools. But when I read a paper (as one does, occasionally) which seems to make a claim to represent a Grand Theory, then I shall continue to take it with a pinch of salt. And if I myself should ever seem to make such a claim, you have my full permission, nay encouragement, to do the same to me! (Robertson, 2000, p. 9-10). My expectation are more favorable in relation to “Grand theory” compared to the view here expressed by Robertson, but I admit the task is “a tall order”. I also share his view we often encounter such grand theories without proper examination of their fruitfulness to the tasks they are supposed to solve. As I concluded elsewhere: Buckland (2012:5) has stated that Brier’s (2008) cybersemiotics provides a coherent unifying theory for an existing field, namely LIS. However, this has to be shown by concrete studies, which have hitherto not been carried out by Buckland, Brier, or anybody else. Bawden and Robinson, in a similar way, have great expectations for the philosophy of information developed by Luciano Floridi, but this philosophy has not hitherto been applied in the specific areas of LIS described in the rest of their book. We need to go from the concrete to the abstract and back again. This is the most urgent task for research in LIS. (Hjørland, 2013c, pp. 230-231). The answer to the question in the section headline is therefore: We may have too many theoretical suggestions, which are proposed without proper examination. All too often it seems to be considered legitimate to take a theory from another field, introduce it and claim that it might be relevant. A presupposition seems to be that there are enough qualified people in the field to do the rest of the job: Prove the relevance of the theory (what I consider to be a wrong assumption: The burden of proof is on those who suggest a new theory). It is important that theoretical suggestions are compared with other theoretical suggestions and that there is a strong dialectics between theory and practice. 8. What influences the theoretical development of information science? At first sight, information science does not seem to be much driven by theory. It seems to be much more driven by external factors such as new technologies and overall trends in society. When a new information technology such as online databases, CD-ROMs, the Internet, social media and so on arises, it is followed by new research fields in information science. We need therefore to consider two main sets of factors influencing the development of science (and correspondently two tendencies in the history of science to explain the development of science): One set are the internal factors in science (one discovery leads to new discoveries, science is seen as independent and relatively cumulative10). Another set of factors are external to science itself (such as new technologies, the people attracted to the field or new developments in philosophy. Science is here understood as consisting of relatively more breaks and new approaches). In order to consider the development of approaches to information science we may need to consider a range of mutually interacting external and internal developments: A. Overall developments in society
These overall factors influence information science and the way researchers in this field look at their own field. B. Technological developments Most important here is the development from printed media to digital media, the challenges traditional libraries and other memory institutions face, and the challenges librarians face from computer scientists, the plurality of new information fields and professions and so on. C. Media and communication developments Media development is of course also the above mentioned development in the direction of digital media, but it is broader and involves the development of symbolic systems, including languages, and of communication systems considered from overall perspectives such as those offered by developments in science, commerce, education, the development of open access and so on (See the UNISIST model in Fjordback Søndergaard, Andersen and Hjørland, 2003). D. Developments in libraries, archives, museums and databases Developments in research libraries and other memory institutions include library automation, library shortcut and the changing economic and political conditions under which such institutions have to function; developments in the information professions are also important. E. Theoretical developments Interdisciplinary developments
In order to develop fruitful theories science must consider epistemology. As Albert Einstein wrote: The reciprocal relationship of epistemology and science is of a noteworthy kind. They are dependent upon each other. Epistemology without contact with science becomes an empty scheme. Science without epistemology is – insofar as it is thinkable at all – primitive and muddled (Einstein, 1949, p. 683). It is important to consider and evaluate the different theories and traditions in the field. In this connection it is also important consider how different traditions are related to overall philosophical perspectives, which practical problems information specialists are sufficiently educated to work with and the dialectical relations between theory and practice. Table 1 shows different layers of mutual interactions between philosophy, subject theory, interdisciplinary theories and information practices. The views expressed in the present article may be considered an emphasis on a kind of Grand theory (cf., Skinner, 1985); they are opposed to the view that good research can be done without considering the theoretical and philosophical issues. p
Problems at the application levels are connected to the theories that information professionals have (and which have influenced their tools, e.g. classification systems), which are again connected to metatheories and again in turn associated with the general philosophical level11. 9. A social and cultural turn in information science? RSLIS has in last five years had a goal of developing information science ‘in interplay with issues such as the contextual, social, philosophical or cultural perspectives’. This goal is expressed, for example, in announcements about the position of Professors at RSLIS12. This expressed goal of a leading international information science institution is one (albeit a small) indication of a social and cultural turn in information science. Other indications are views expressed by leading scholars in the field such as Blaise Cronin (2008) who wrote about the ‘sociological turn in information science’ while Michael Buckland (2012, p. 1) wrote that ‘…information science is […] a form of cultural engagement’. The above mentioned indications and views do not in themselves provide evidence of any social and cultural turn in the field. It is probably also the case that some (many?) colleagues and scholars in the fields do not share this view (at least for now). The idea of relating information science to cultural studies is one in a long range of ideas relating IS to other fields such as computer science, management studies, cognitive science, and many other fields. To claim that field A is related to field B is a hypothesis that has to be examined. What really counts, of course, are strong arguments that such a relation will enable the most fruitful development of the field, and in this regard also demonstrate the limitations of earlier approaches in the field such as information theory and the cognitive view. Arguments for a social and cultural turn may come from inside information science as well as from developments in other fields such as psychology, linguistics, philosophy and many others. One such argument is, for example, for methodological holism, about which cultural psychologist Carl Ratner wrote: Individualism says that the individual element is an independent entity that has self-contained properties, although of course it draws on resources around it. An example is the popular idea that the individual is responsible for his or her own fate. One's success and failure depend ultimately on how hard one works. Holism says that the individual element is inextricably tied to other individuals. Individuals are interdependent, and they are internally related in the sense that each is imbued with and constituted by the qualities of others. An example is a child in a family. The child's psychology depends utterly on the way he or she is treated. Any intrinsic tendencies are modulated and mediated by experience. From this perspective, the child is not entirely responsible for his or her behavior (Ratner, 2008), p. 514). Ratner further shows how holism is related to cultural studies. Ratner’s examples are all about human beings as elements, but this may of course be extended to ‘information’ and ‘documents’. I believe there is a great opportunity to develop information science from such a holistic perspective. Computer science has so far dominated the field of information retrieval mostly based on an atomistic/individualistic frame of reference (e.g. by applying ‘bag-of-words’ approaches). Perhaps this line of research has exhausted most of its potential? Recently, it has been claimed that there seems to be no overall progress in the field of IR (Armstrong, Moat, Webber and Zobel, 2009). If this is true, information science may have important perspectives to contribute, which may be related to methodological holism and to social and cultural perspectives that, as we have seen, have lain dormant in the traditions of LIS. From a social and cultural perspective, documents (or ‘information’) are always understood as being produced within a particular tradition, which contains important clues to its meaning and potential use. In Hjørland (2013a) I argued that bibliometrics may be able to provide such a historicist perspective. Finally, we shall return to social epistemology. The term is central because information science is about communication of knowledge (and thus social) and about criteria of what counts as knowledge (hence epistemology). There are today at least four versions of social epistemology: (1) an approach from analytic philosophy in which Alvin Goldman is a leading representative (2) a critical approach developed by Steve Fuller (see Collin, 2013, for an introduction and comparison of these two versions) (3) a version developed by information scientist Rafael Capurro on the basis of hermeneutics (see Capurro and Holgate, 2011) and (4) the approach developed by Thomas Kuhn (see Wray, 2011). We shall here shortly consider Kuhn’s version. According to Kuhn, we do not conceive the world directly, but – as Kant proposed – through lenses of the mind. However, contrary to Kant, these lenses are not hard-wired into our brain, but are changing with cultures and scientific paradigms: Kuhn maintained that his view was Kantian but “with categories of the mind which could change with time as the accommodation of language and experience proceeded” (Kuhn 1979, 418-419) (Andersen, 2012, p. 188). What are the implications of this view for information science? The first thing is that we cannot study “information” or “users” directly, we always have to consider the cultural or theoretical perspective from which observations are made and research is done (and technology is developed). There is no “view from nowhere” to use Thomas Nagel’s (1986) expression. More directly, the subjectivity of relevance judgments by experts and by users may partly be explained by different “paradigms”. Understood in this way, we have a third position between expert evaluations and user evaluations: We have expert evaluations and user evaluations made in some specific context influenced more or less by different theories and “paradigms”. We should not just ask: Is X relevant? We should ask: From which perspective is X relevant (and which interests does X support?). Download 225.5 Kb. Share with your friends:
| ||||||||||||||||