Go to Semiotic Review of Books Home Page
Go to SRB Highlights
Go to SRB Archives

SRB Highlights

This review appeared in Volume 9 (2) of The Semiotic Review of Books.

Managing Knowledge

by René Jorna

The Knowledge-Creating Company, by I. Nonaka and H Takeuchi. Oxford University Press, Oxford UK. 1995. ISBN: 0-19-509269-4

Knowledge Management: Organization, Competence and Methodology, edited by J. F. Schreinemakers. Ergon Verlag, Würzburg. 1996. ISBN: 3-93-2004-26-4

Although I have been working now for more than a decade in the field of management science, I am repeatedly surprised by the fact that within this field the fashionable trends keep succeeding one another and that there is no cumulative insights toward the development of a scientific framework. This critique especially applies to one of the latest trends in management science, namely the debate about knowledge, knowledge management and learning. One should expect that in discussions about these issues in relation to organizations much conceptual help could be found in neighbouring fields, such as biology, psychology or philosophy. This is hardly the case. Biology is neglected, psychology is only used as far as it concerns characterology or peak experiences and philosophy is mainly used by quoting secondary literature. The fact that epistemology and ontology, and -- since the beginning of this century -- cognitive psychology especially deal with knowledge structures, knowledge acquisition and knowledge processing does not seem to be common knowledge among management and organization scholars. There may be several reasons for this situation. Let us consider some possibilities.

To management scholars, the practice of management appears to be so unique that theories that do not originate from this practice are deemed inadequate. There may be some arrogance involved here. The enormous complexity of human artifacts, such as big companies and institutions, requires a completely new way of observing and reasoning. Concepts like dynamic, holistic, and processual connectivity are put forward and inspire various groups of managers and management scientists. The organizational practice is such that answers cannot be found in the existing sciences. One knows better because one is familiar with the real management practice. Although the complexity of large organizations is undeniable, I think this is a questionable line of thinking, if only because the universe is far more complex than the most complex of organizations. I surely gave a caricature here, but I think that this picture of the state of affairs within management science is not completely unrealistic.

There also may be a more serious reason why management science finds it difficult to establish links with other scientific fields. It may be the case that management scientists lack certain conceptual frameworks to describe, explain andpredict the behaviour of organizations. For example, it is already a topic of debate whether one can say that organizations behave. What is the unit of research, what kind of laws or regularities are relevant and which frames of reference can be used? Concerning these issues, management science is a difficult area of study. The study of "organization" is as broad as the study of "nature". However, the latter is subdivided into diverse disciplines as, for example, geography, biology, and quantum physics. A similar partitioning should also exist with regard to organizations, but ergonomics, motivation studies or group dynamics are not regular parts in many management study programs. It may be argued that the determination of rigid conceptual framewords with variables, measurable units and causal connections has not taken place, yet. Closely related is the problem that in management studies several levels of description are completely mixed up. Within management science it is possible to study and describe inter- and intra-organizational relations, groups and individuals and all these "entities" literally or metaphorically at the same time. Demarcating the domain of study is not easy. Does one need a mathematical, a sociological, a biological or a psychological perspective? Perhaps this is one of the very good reasons for following new fashions within management studies. There are no established frames of reference that may be used to reject new fashions and trends. One is not able to refuse so-called "new sense" and call it nonsense. Nonetheless, this situation does not relieve serious management scientists from the obligation to look around, to study adjacent fields, especially if the topics concern knowledge and learning. Both topics have a long outstanding tradition.

It is interesting to notice that "knowledge" is taken for granted in most present day discussions about knowledge management in management science. One characteristic aspect of knowledge, among others, is that whether it is inter-individual or supra-individual, knowledge is always connected with brains or cognitions of individuals. This implies that a user, whether he is the origin of knowledge or just a transfer desk, is always involved as an interpretation mechanism. Based on this, knowledge can be called "interpreted information."

It also is amazing to see that knowledge management has not become an issue in management science until very recently. Management is supposed to be about the control, the regulation, and the command of things. However, the topic only became fashionable some five years ago. The problem is that although knowledge is now at the centre of interest, it is not clear what is meant by knowledge. Therefore, it is hard to figure out the control and regulation of that which is not very well-defined.Consequently, several definitions of knowledge and knowledge management are used. The best one can say about this is that it resulted in creative anarchy.

Knowledge (management) can mean four things. First, it may be a fashionable, somewhat more mysterious word for information. Here knowledge management means the same as information management. Second, knowledge management is about the assessment of staff in organizations, in which case it is almost similar to (sophisticated) human resource management. Third, knowledge management can be about the control of the content of existing and new products and services, present in the minds of people. Fourth, knowledge management can be about the control of the form of existing and new products, present in the minds of people. Calling knowledge management innovative in the sense of information management or human resource management is difficult. However these meanings are often used, as I will show later. Knowledge management in the sense of (the control of) the content or form of interpreted information is new within management science. However, frames of reference suitable to deal with these interpretations of knowledge are lacking.

1. Knowledge-creating companies

Nonaka and Takeuchi's book The Knowledge-Creating Company (1995) now is very often mentioned in discussions about knowledge management. This is surprising, because the book hardly deals with knowledge management. Rather, it is about changing organizations by means of the presentation and representation of knowledge.

Nonaka & Takeuchi (N&T) initiate two important topics in their book. The first concerns the innovation of products and services, whereas the second deals with the question of how changes in organizational structures can stimulate this innovation.

N&T call the innovation of products and services knowledge creation and this enables them to develop a conceptual framework to compare Japanese, American, and European companies. Knowledge is defined in a classical philosophical way as "justified true belief". However, in contrast with the logical and epistemological tradition that emphasizes the "truth" part of beliefs, N&T want to exploit the "justification" or legitimization part. The acquisition of (new) beliefs, which is the basis of innovation, and the communication of these beliefs to oneself and one's collaborators is the same as learning. However, according to management scientists, there are two ways of learning; exploitation (single loop learning) and exploration (double loop learning). The former can also be called deepening your performance within existing frameworks, where-as the latter consists in looking for completely new frameworks. In the AI-literature the former is called chunking, whereas the latter is called creativity (Boden, 1994). N&T mainly focus on this last kind of learning.

New ideas do not come out of the blue. They are essentially there, in the minds of people as members of organizations. Learning new things or a new frame of mind does not mean that the organization learns, but that individuals learn (N&T, 11-13). However, the metaphor of the learning organization comes in where individuals express their new ideas and concepts to their colleagues, who in turn react to, rethink, and reinforce these ideas. So, knowledge creation starts with individuals, then becomes part of the group, changing its character and formulation, and finally is integrated into an individual's explicit knowledge repertoire. N&T call the conversion from tacit knowledge into declarative knowledge, knowledge conversion. N&T make a distinction into four phases that regulate the knowledge creation process in an organization. First, there is the phase of socialization, in which an individual tries to express to himself his tacit knowledge by using metaphors, diagrams, small drawings and even body gestures. Second, in the phase of externalization, this tacit knowledge isexpressed in ways that can be understood by other members of the organization, meaning that an individual is trying to work towards declarative knowledge. Third, the externalized knowledge is joined with ideas of group members about the same issue. This phase is called combination and it is characterized by declarative knowledge and group orientation. Finally, we have the phase of internalization in which the members of the organization each make the explicit knowledge internal to themselves. Knowledge becomes tacit again and because everybody does it, there is a group orientation. Then we are back at the phase of socialization, but as N&T say, at a higher level. So, the process continues. The advantage of this process of knowledge creation is that knowledge is (a) sympathized, meaning that is shared with other members of the organization, (b) conceptualized, that it is couched and expressed in (declarative) structures, (c) systematized, that is to say, it is to a certain extent made coherent and (d) operationalized, meaning that it has resulted in really innovative products or services.

The important point for every management approach is how organization conditions can be shaped so that knowledge creation will flourish. N&T propose what they call the "middle-up-down" management processes of which the examples can best be found in Japanese firms. This management approach is in sharp contrast with the top-down and bottom-up styles, which are mainly present in American and European organizations. In distinguishing top, middle and low levels in an organization, the middle management has a key position in the knowledge creation process. They are what N&T call the knowledge engineers. From them the process goes up to the so-called knowledge officers (top management) and down to the so-called knowledge practitioners (workers). From both groups the knowledge comes back in a reshaped form to the starters of the process. Thus, the process proceeds in time like a spiral. In this way knowledge creation can also be called the engineering of new knowledge or the remaking (or reshaping) of organizational reality concerning products and services. In organizational terms this means that classical bureaucratic business systems are combined with task force structures. Bureaucratic systems work the best in stable, large and secure situations, whereas task forces are best used in dynamic and uncertain circumstances. Combining the two leads to organizations that are well equipped to respond to whatever challenge the environment offers. According to N&T, this kind of organization is called: a hypertext organization.

Beside many appealing examples from Japanese, American, and European firms, N&T also offer a sort of manual to any organization that may want to transform itself into a knowledge-creating company or a learning organization. The manual consists of seven phases that essentially boil down to the following imperatives: (1) create a knowledge vision; (2) develop a knowledge crew; (3) build a high-density field of interaction at the front line; (4) piggyback on the new development process; (5) adopt middle-up-down management; (6) switch to a hypertext organization; (7) construct a knowledge network with the outside world. The whole time path to realize such a learning organization may take several years and many pitfalls have to be avoided. However, N&T argue that in a world that requires fast transformations, quick responses to new challenges and increasing dynamic environments, this is the unavoidable direction future organizations have to go.

N&T's book is thought provoking, and it may be a best-seller in management circles, but from a scientific point of view it is questionable on several grounds. First let me illustrate its weak points and then explain its provocative aspect. In the second chapter the authors give an extended description of the epistemological tradition in philosophy, but they mainly use secondary literature. For instance, Bertrand Russell's History of Western Philosophy is often quoted. Russell probably is one of the most important philosophers of this century, but except for his study on Leibniz, he is not a trustworthy historian of philosophy. Russell is biased as one can see from his treatment of Hegel or Nietzsche. Furthermore, the way N&T present the history of Western philosophy is amazing. The point N&T want to make is that in Japanese philosophy the unity of mind and body, man and nature and individual and group is realized. What they mean with this remains unclear, but they succeed in neglecting many well-known Western philosophers who have discussed these matters, such as Hegel, Pascal, Ryle, Wittgenstein or Kierkegaard. In a similar sense, strange passages can be found about psychologists and organizational theorists. What they say, for example, about Simon is appalling. "Simon, however, emphasized the logical aspect of human reasoning and of organizational decision making processes, and the limitations of human cognitive capacity." (N&T, 38). This is nonsense. Following Selz, Simon emphasized and conceptualized the psychological aspect of human information processing. And as can be found in today's economical and organizational articles about man's optimal and unbounded rationality this psychological aspect still is strongly underestimated. Similar mistakes occur in discussions of March and Schumpeter. Another strange thing is the absence of references to the works of Watson, Skinner, Piaget, Thorn-dike and Gagné, to name but a few learning theory experts. On the other hand one of the central issues in N&T's book, about tacit and declarative knowledge, has been extensively discussed in the early eighties in cognitive science and has faded away (Pylyshyn, Newell and Fodor). References are missing in N&T's book.

A second drawback in N&T is the lack of methodology. The book is based on case studies. These are interesting in as much as they highlight relevant aspects of management practice. However, case studies cannot be used for corroboration of theories and hypotheses. From a classical methodological point of view cases only lead to the generation of theories or hypotheses. They may illustrate an existing theory or they may lead to the development of a theory. N&T limit themselves to presenting cases. Therefore, the book mainly consists of success stories, whereas the perhaps much more frequent failures are missing. This is not a sound approach for constructing theories; it is rather the surest way to confirm biases. A speculation that results from such a procedure may be thought-provoking and stimulating, but it is not suitable for describing, explaining, or predicting.

The theory put forward by N&T still has two other shortcomings. The first concerns the level of description of the objects of analysis. If the discussion is about knowledge-creating organizations and learning and it is argued that this is impossible without individuals in the organization, then necessarily concepts have to be used that are located at the individual level. This is not the case with N&T. Second, and this is much more serious, N&T do not discuss the real topic of their investigation. The central debate is not about knowledge-creation, but about expressing knowledge, whether existing or new, or tacit or declarative. If one wants to use knowledge in organizations, then this knowledge has to be expressed in one of the available sign systems we normally use. The discussion should be about the kinds of expressions, formulations and locutions people use. They not only use literal expressions and formulations, they also use metaphors, diagrams and imaginations. The point is that, although the latter entities are different from the former ones, all belong to the same category of signs, which, however, require different kinds of sign understanding. N&T incorrectly identify declarative knowledge with language entities and tacit knowledge with non-language entities, whereas in both cases we are talking about signs as presentations and representations and about sign understanding. The various phases, discussed by N&T, such as socialization, externalization, communication, etc. really are about various ways of presentation and representation. This is missing in the book, as is a conceptual framework that might have been useful in dealing with signs, that is to say semiotics,the study of sign systems. I will come back to this point after the review of the book by Schreinemakers.

2. Knowledge Management: Organization, Competence and Methodology

Some groups working on knowledge management are trying to start a movement. One of these groups is the International Symposium on the Management of Industrial and Corporate Knowledge (ISMICK). Researchers from various disciplines participate in the ISMICK and this mix can easily be traced in the book that documents the progress of this group: Knowledge Management: Organization, Competence and Methodology, edited by Jos Schreinemakers from the University of Rotterdam. This collection comprises 24 contributions (including two addenda), divided into seven groups, or as it may be called, tracks. This metaphor adequately expresses the fact that the "train of knowledge management" uses several, sometimes parallel, sometimes crossing, and sometimes dead end tracks. The various contributions are from engineers, computer scientists, sociologists, and organizational and educational researchers. Cognitive psychologists and philosophers are conspicuously absent from this group.

Taking into account the different meanings of knowledge management mentioned earlier (information management, human resource management, the content of knowledge and the form of knowledge) yields the following results: thirteen articles deal with knowledge management as a synonym for information management, five with human resource management, three are about the foundations of knowledge and new approaches towards knowledge and only on is about the form of knowledge.

However, three articles stand out: those by Tolis; Stokke, Syvertsen & Tilset; and Jooste & Duffy.

Tolis describes in his "Business Modelling for Understanding and Change" a conceptual framework that deals with the many changes in organizations. Important is the fact that he explicitly states that models are used to describe the present and the future situation in an organization. A model, however, has to be chosen out of the many available models: logical, real, verbal, diagrammatic or symbolical. The question is when to use what kind of model. A basic assumption, according to Tolis -- and a good one I think -- is to treat models as equal to signs. A model (sign) is a coherent entity of which a designer (or user) supposes that it stands for something (object). Based on the work of C.S. Peirce, Tolis discusses one main dimension (underlying the models) subdivided into symbolic content, symbolic process, iconic content and iconic process. In line with Peirce, symbolic means model on the basis of convention, whereas iconic means model based on natural similarity. Tolis gives two other main dimensions, namely abstract and concrete activities that may be action-oriented or reflection-oriented, and consensus or conflict situations that may be subjective or objective. Discussions about change of work and tasks in organizations can be modelled on the basis of the various constellations derived from the dimensions. Thus, by modelling several perspectives the different views on the future situation of the organization become more visible. Tolis ends his insightful discussion of modelling in an abrupt way, but he clarifies that a model, whether it is used for information systems or to redesign business processes, is a human artifact and not a quantity given by nature. This artifact is the result of a combination of modelling tools and the experiences, preferences and habits of engineers and managers. A conceptual framework such as the one suggested by Tolis might help to understand the continuous confusion of tongues. However, this does not facilitate the modelling process itself.

Stokke, Syversten and Tilset entitled their contribution "Internet and a Virtual Value Chain for the New Industrial Revolution." According to them knowledge is a new industrial product. However, the usual metaphor of the marketplace, very dominant within management science and economics, is not suitable anymore to conceptualize this new product. In the near future we will have to talk about a representational market space, but we do not know what it exactly means, yet. Touching and grasping products will be replaced by presenting and representing information that only has an electronic basis. The authors shortly discuss some examples of new knowledge products in the form of digital services offered by Netscape, Microsoft Explorer, Anderson Consulting and SRI-International. No physical sales person is involved in marketing and offering these services. They are only available from the Internet. neither floppy's nor CD-ROM's are distributed anymore. Users, supported by help desks and powerful search engines, are supposed to judge the value of the products and services. The point Stokke, Syvertsen and Tilset want to raise is the following: What are the consequences of this value chain? This value chain may be relevant in discussing internal and external logistics, production processes, marketing, sales and service. However, we are talking now about information and knowledge elements and not about "physical" products. The comparisons the authors make are related to the composition of the new "product", the determination of its value and the distribution of this new "product".

Jooste & Duffy discuss the different opinions on knowledge management in South Africa in "Knowledge management competencies as the building blocks of a knowledge-rich organization." They collected questionnaires about knowledge and information competencies from 202 companies. These questionnaires were divided into several groups and these groups were compared with one another based on various criteria. They found interesting results, such as that most companies considered their information systems as essential for survival in the near future, but the formal level in the organization increased with the complexity and the integratedness of the systems. Surprising was the fact that most companies did not want to have knowledge systems, decision support systems, business process redesign, object-oriented programming or simulation. The organizations tried it, they did not understand it and they did not see opportunities for these so-called novelties. This attitude is not surprising, because similar things happen to many new things in their starting phase. After all, this also happened with Internet (Arpanet). One is inclined to say that happily there still are universities where immediate profit or utility is not the first priority. Jooste & Duffy's conclusion is that most companies require their employees to become familiar with computers. Furthermore, attuning the information systems and the strategic goals of organizations is important. Many companies only achieved this after many failures. Organizations try to gain advantage with their information systems. It was also found that the behaviour of employees to the way organizations stimulate employees are decisive in the success of information systems. Technically advanced systems were not trusted. Perhaps it has to be stated that the attitude of employees is a critical factor in the success of information systems. This means that we are returning to knowledge management in the sense of human resource management. People still constitute the strength or the weakness of a company (Boersma, 1996).

It is not possible to discuss all contributions in Schreinemakers' book. The articles are diverse but of unequal quality. Overall, the most interesting feature of this book is the confusion it perpetuates about what knowledge management really is. An encompassing conceptual framework is missing. If knowledge management is about control, the regulation and the stimulation of knowledge, and if knowledge is something that is located in the minds of people, then what should be discussed is how the cognitive architecture and the mental representations of humans relate to knowledge management. This seems to be something for psychology. However, psychology as a science, is not ready to fulfil this role, yet. This is not to say that psychology should be omitted when the discussion is about knowledge and knowledge management; on the contrary. Much of psychology is already missing in the field of management studies. What I mean is that regarding knowledge management, semiotics provides the most relevant conceptual framework for describing and analyzing the form and content of knowledge.

3. Knowledge management also is about presentations and representations.

A characteristic of today's semiotics is that many researchers from different scientific areas define themselves as semiotically oriented. Even computer scientists nowadays are using semiotics, because sign producing, processing and interpreting systems, whether they are natural or artificial, are a topic of study within semiotics (Jorna, 1990). Even in some parts of management science, for example marketing research, the interest in semiotics is growing. Good marketing depends upon the presentation of the product or service to the customers (Umiker-Sebeok, 1987). For similar reasons some designers of user interfaces started to look at semiotics. The expression of information is always in the form of signs -- it is the only vehicle we have for communication -- but the one sign system is different from the other. Everybody knows that a picture says more than a thousand words. However, pictures are ambiguous as the designers of instructions for medicine use in Africa and Asia discovered. Depicting a moon as an instruction for sick people to take one tablet at night has resulted in unforeseen nasty consequences. This means that in many contexts the way information is presented and represented has consequences for our understanding of, interpreting of and acting upon this information.

Semiotics not only is related to the categorization of kinds of signs and sign systems, it also is related to the internal systematicity of signs. To put it simply and generally, a sign stands for something it represents and is interpreted by a user who may decide to act. The reference of a sign to an object, however, does not necessarily have to be the case. This makes it possible to distinguish presentations that do not refer, and representations that do refer. For example, in a management meeting a certain diagram may only be used to distract or focus attention. It may also refer to the sales of a certain product or the predicted profit of a service. In the first case it is only a presentation, whereas in the second case it is a representation. However, in both cases it is perceived by many people who may interpret the sign according to their own frame of reference. It may probably be the case that the person who used the diagram could better have used a hand gesture or a text fragment for some members of theaudience. If on the basis of the diagram actions are also required, the sign works, as I would like to call it, as a "pragmation." The meaning of the sign implies an action. To take an example from a completely different domain, the "enter" button on a computer not only represents that an action should be carried out, it also invites (starts) the actual action.

In working with signs a choice is always made in favour of a certain sign (diagram, word, gesture) that can be used in different ways (presentation, representation, pragmation) and can be interpreted by different kinds of users in a different way. Semiotics has developed an enormous richness of categories to approach signs and sign systems (see Nšth, 1985). Nonaka & Takeuchi as well as Schreinemakers illustrate the point that the success and failure of organizations strongly depend upon the way the members of the organization succeed in giving expression to innovations and changes of products and services. Giving expression not only is important in communication with others, it also is extremely relevant in having discussions with oneself. Here we touch the essence of the book by Nonaka & Takeuchi and some contributions in the book by Schreine-makers. After all, what is the meaning of the phases of socialization, externalization or communication if it is not about the change of one sign form into another sign form? Not only verbal forms are relevant, here, but also rough drafts, sketchy pictures, vague gestures and complicated diagrams. Expressing, choosing forms of expression, interpreting and reinterpreting is the same as presenting and representing, meaning that this is the basis of any form of knowledge creation and knowledge management.

Although this conclusion may hold for the development of new products and services, it is even more relevant for something completely different, namely our future working situation. In the very near future the good old physical office will be replaced by more or less intelligent computer systems. Perhaps, it will take another ten years, but the office of the future will be a combination of what is between our ears and the power of the Internet. Knowledge will be determined by the way we are able to handle intelligent search engines that are able to get information from all over the world. That information, however, still has to be interpreted by us. Forms of giving expression to this information are important in three ways. First in relation to what we know ourselves, second in relation to the interpretation of this information in the sense of what it is, what it stands for and which action has to be taken and third in relation to the transfer of interpreted information to others. Adequate presentations, representations and pragmations are indispensable in realizing this.

It seems that the debate about knowledge management functions as thewell known box of Pandora. Management researchers (and managers) started a debate that they could hardly overlook. Knowledge and learning are complicated matters and it is not easy to grasp them. However, there also is a lack of conceptual frameworks to deal with these non-physical entities within management science. Of course, this depends upon the meaning of knowledge management one uses. If knowledge management means human resource management with a strong emphasis on the assessment of skills and competencies of staff it is just old wine in new bottles. The same can be said if knowledge management means (advanced) information management. But if knowledge management is about (the control of) the form and content of knowledge and if phenomena like "individual people with their minds and cognitive architectures" are central -- and I think this is the case in discussions about knowledge and learning -- it automatically leads to questions about the frames of reference that are necessary to grasp these entities. I have tried to illustrate that the conceptual framework of semiotics might be of help here. However, it is not a cure for all dysfunctions. Nonaka and Takeuchi wrote a best seller with their book about knowledge-creation, but completely missed the point, as I tried to explain. Because they started something that still needs to be finished, it is perhaps time for someone to write a new best seller... about semiotics and the management of knowledge.

Table of Contents: Knowledge management: organization, competence and methodology.
J.F. Schreinemakers, Ed.

Keynote Address
Jean-Paul Barthes: ISMICK and Knowledge Management I. Knowledge management in practice II. Ontological and Epistemological Aspects of Knowledge Management III. Information Technology Aspects of Knowledge Management IV. Knowledge, Learning and Management V. Inter and Intra Organizational Knowledge Management VI. Knowledge as a Strategic Resource VII. Knowledge Capitalization Addendum


Boden, M A, 1994. Dimensions of creativity. Cambridge, MA: The MIT Press.

Boersma, S K Th, 1996. Kennismanagement: een creatieve onderneming (Knowledge management, a creative enterprise). Rijksuniversiteit Groningen, Oratie.

Jorna, R J, 1990. Wissenrepräsentation in künstlichen Intelligenzen. Zeichentheorie und Kognitionsforschung. Zeitschrift für Semiotik, 12, 9-23.

Nonaka, I and H Takeuchi, 1995. The knowledge-creating company. Oxford: Oxford University Press.

Nšth, W, 1985. Handbuch der Semiotik. Stuttgart: Metzler.

Schreinemakers, J F (Ed), 1996. Knowledge management: organization, competence and methodology. Würzburg: Ergon Verlag.

Umiker-Sebeok, J (Ed), 1987. Marketing and Semiotics: New directions in the study of signs for sale. Berlin: de Gruyter.

René Jorna is associate professor in Knowledge Systems and Decision Support Systems at the Faculty of Management & Organization, University of Groningen, The Netherlands. He has published books and articles on Planning and Scheduling Support, Knowledge Systems, Semiotics and Cognition.

Go to Semiotic Review of Books Home Page
Go to SRB Highlights
Go to SRB Archives