Hypertext reading as practical action

-notes on technology, objectivation and knowledge

By Terje Rasmussen

A) Electronic mass media and digitised telecommunications: Towards a unified media matrix?
Tools and objects - perceptive and hermeneutic mediation
B) Agency
The principle of habitus
Agency and the world of artefacts
C) Objectivation of knowledge
Practical action and the 'art of using'


The concept of 'hypermedia' has perhaps superseded the actual technical steps toward such a technology. More is said about its revolutionary potential, than demonstrated. Both pioneers within hypermedia like Ted Nelson (who coined the term 'hypertext' in the 1960s) and literary theorists argue that we must abandon conceptual systems founded upon ideas of centre, margin, hierarchy and linearity, and replace them with the ones of multi-linearity, nodes, links and networks (Landow 1992, 2). Electronic texts are seen to include features which were never to be located in the printed book. The text opens itself to changes which automatically reduce the centrality of lines and chapters. It signifies the opportunity of non-sequential writing, to use one of Nelson's terms.

If this is correct, we need to investigate further what these new systems do to reader practices and what the reader can do with the systems. In the following, I would like to consider the transformations of the role of the reader hypertext seems to encourage due to the new hyper form of the text. My suggestion is that this new role enforces a new conception of reading and that hypertext reading can be better understood if we consider it as action, not very different from other actions and practices in everyday life. The new features of hypertext, the integration and the hyper aspect, change the role and the experience of the reader in that it enables stronger ties between the reader and the medium, which again may have some interesting consequences. In this light, we can also better grasp the historical process of objectivation (a term I shall explain later), which hypermedia seem to bring to a new level. Before that, however, I will consider some similarities and differences between telecommunications and the mass media, from which we can map the strange hybrid character of hypermedia; the merge of mass media like the book or television, and two-way and interactive media like the telephone or the database.

This article then, consists of three main parts: A) a discussion of the relationship between telecommunications, which results in hypermedia and other new media forms, B) a discussion of the nature of human action and interactive media like hypertext, and C) a discussion of objectivation of knowledge, that is, the inclusion and containment of knowledge in technological systems like hypertext. I will begin by considering some features of hypertext which make this medium intriguing in a socio-psychological and sociological light.

Hypertext is made of blocks of text - in the form of written text, pictures, video and sound, chained together by electronic links. The ambition of these systems is that the reader may browse through linked, cross-referenced, annotated texts in a non- or multi-sequential manner without becoming lost in the universe of texts, in hyperspace. The ideal, one may say, is order without linearity, or multi-linearity, multi-textuality (texts, animation, sound, etc.) and multi-sequentiality without chaos.

A full hypertext system enables the reader to select, rearrange and comment on the meta-text. One may choose one's own, individual reading paths, which make the 'book' a unique item. The active reader explores a text which in fact does not exist readily as a text until the reader composes it according to his or her requirements. In quite a different way than the typed text, the reader becomes more of a co-author, a new role which accompanies his role as a reader. Rather than simply reading a book, he draws from a book numerous of citations and put them together in new ways. The transition from typed text to hypertext is also a change from physical to virtual text, to potential textuality. In this sense, the author looses some of the control of his work, because the reader recomposes and makes it a new montage-like text each time. There is no given centre, and hence, no given hierarchy, of the narrative. The reader must make a centre from which he orients himself and investigates the text, or alternatively, he may 'travel' in the system, making use of various items of orientation, such as maps, menus, icons, and micons.

In the hypertext literature, the terms 'hypertext' and 'hypermedia' are often used interchangeably (see for instance Landow 1992). It seems to me, however, that the terms put the emphasis on somewhat different aspects of the new medium. Hypertext emphasises the integrational, and the hyper aspect: The first refers to the blending of verbal and non-verbal texts and the second refers to the possibility of multi-linearity and multi-sequentiality in the composing and reading of texts. The concept 'hypermedia', however, seems to refer more to the hypertext technology which the texts derive from, primarily the computer and the necessary additional equipment. From this, one may say that hypertexts stress the literary or textual element, whereas hypermedia stress the man-machine relationship, the relationship between the author, the medium and the reader. This, however, is far from a clear-cut distinction and the two aspects are of course mutually dependent on each other.

If we for a moment associate Marshall McLuhan's credo that 'the media is the message' to hypertext, it is clear that the hyper-message goes through significant changes. The hyper-message seems to be more of a product of its technology, compared to the message in a typed book or on the television screen, in addition to being a product of the author. The message becomes a message only through the reader's wandering through sequences, links, and networks. The multi-linearity and hyper forms which the new medium enables, visualise clearly McLuhan's point: the text is no longer an isolated idea or argument delivered to the world by a detached author. Perhaps it never was like that. The hypertext is a fundamental socio-technological product - a merge of the mental and the material, of ideas and technology. In fact, hypermedia reveals that many commonplace ideas of messages, content, of texts and literature, are really historical effects of the specific forms of information technologies, from the parchment manuscript and the feather pen to the telephone and the hypertext computer. As also McLuhan notes, this fact becomes most evident when the medium is new and not yet interwoven with the assumptions in everyday habits and conventions. This is probably why hypermedia show its technological distinctiveness so clearly to us these days.

Another change is interlinked with the technological authority over the message. The electronic text is not an autonomous text like the printed text. Rather, it is one of innumerable possible versions of a typed text. As the message becomes a message only through the practice of the reader, the 'message is the reader', to rephrase McLuhan. The authoring function of the reader erodes the distance between the message and its reader as is present in the case of ordinary reading.

This means that the text is no longer simply an object as is the case with the typed book of text. The word 'book' refers to the text, the particular paper technology and to a physical entity. The book does not refer to a particular function, although we use books in functional practices like education, entertainment, etc. The book is an object to which the subject (the reader) directs his attention. However, this is not quite the case with hypermedia. Hypermedia is something more. It is also a tool, like the pen or the text editor. It is a tool of production, a tool of a special kind. It is a tool which enables communicative interactivity with the original author. We may compare it with the telephone, which is precisely a tool for communication.

Thus, the textuality is neither instantiated by the physical object of the book, nor limited to it (Landow 1992, 41). It cannot simply be called an electronic book since the text is not inherently unified with the computer (as the typed text is with the book) and since the text does not present itself in a ready-made fashion. Consequently, the user is not merely a reader in the commonplace meaning of the word. The user does not receive and interpret a complete unchangeable text, but applies an additional authorial role in modifying and recomposing the text into his text.

The hypertext computer is also a tool that must be instrumentally operated more actively and in a much more complicated way than in the reading of the book. And more important, it must be operated in a way that is much more intimately linked with the interpretation of the meaning itself. One may argue that if hypertext is composed in ways that are much closer to the ways the human intellect actually makes sense of texts, the act of reading, the reading as a practice, tends to approach the electronic structure of the text. This means that the act of reading and the text approach each other, even if they do not entirely merge.

As I will return to later, tools can be defined as 'material ways of doing things'. This definition stresses the action of the user more than the 'thing' itself, because it is the use that defines it as a function, as a tool. An axe would not be a tool for the cutting of trees if no-one knew what it was for. Clearly, the hypertext computer is a tool, because it requires certain operations of the user in order to reach some ends. These operations with the computer as a tool proceed as an interactive process, where the user continually sees the result of each command or operation on the screen, from which he decides on new steps. The process is not very different from the cutting of trees or the paddling of a canoe with an oar. One central distinction, however, is that wood cutting and paddling proceed manually, whereas the wandering down the paths and passages of hypertext is manual/communicational.

I would like to argue that hypermedia merges these two forms of orientation, the instrumental orientation enhanced by tools, and the interpretative enhanced by objects of meaning. Through this blending of functions, it radically changes our conceptions of reading, writing and of what a 'text' is. And equally important, it involves new rules for meaning-constitutive practices of the reader. The reader becomes much more of an actor or agent, directing his competence and cognitive skills in and through this instrument/object. Our current vocabulary does not offer any appropriate terms for this new object/tool and its reader/user. I suggest that we simply settle for the term media, even if it normally refers to one-way mass media which does not require such interactivity. We should therefore be keeping in mind the two aspects of object and tool inherent in the medium of hypertext.

Ideally, hypertext enhances an active, selective and independent reader. However, in so doing, it also creates closer ties between the reader and the hypertext. The reader may select chunks of text from the meta-text and create links wherever he desires:

Some of this autonomy, according to Landow, is transferred to the reader. The reader is assumed as more independent, possessing more authority over the text. The active construction of meaning from texts that all readers are involved in, becomes extended even to select and sequence the very text itself. One may say that the reader collaborates with the author in producing a text by autonomous choices. This is also the case with a database search (Landow 1992, 75). It permits the reader to enter the author's text at any point and not at the point of the text, and so decide what is the 'beginning' and the 'end'.

Contrary to printed text, hypertext suggests integration and openness rather than self-containment. But precisely through this openness, it looses the quality of containing a truth, a definite message. The text is not a definite thing or object, but more like oral, evanescent utterances that constantly appear and vanish in flows of voices.

A) Electronic mass media and digitised telecommunications: Towards a unified media matrix?

To understand the background of hypertext, we should consider the two lines of technology and research that gave birth to it. Here, I shall particularly emphasise the social research on the two forms of media. Until the mid-eighties, telecommunications were largely ignored as an object in the field of media research, and so social and cultural questions related to telecommunications were largely left unaddressed. Apparently, the problems of telecommunications seemed to be merely technical, economic, and policy-oriented. Several theoretical reasons may account for this: First, media research was caught up with the explosive change within mass communication, particularly the electronic media. The growth of media research was legitimated on the cultural effects of media containing specific social and cultural messages.

Second, the field of media research was structured around theories and models dedicated to the particular structural features of one-way mass media and mass communication. There was an implicit tendency to view telecommunications as devoid of any cultural or political meaning 'content-less' as they were. The structure of the media made it irrelevant to distinguish between powerful senders and dominated receivers of messages. Thus, the sociological topic of power was rarely pinpointed. Neither text- and gratification-oriented media research, nor cultural studies could 'find' any cultural content to analyse. The social and cultural impacts of telecommunications were impossible to grasp with models that assumed mass audiences and mass distributed messages. In the distinction between interpersonal communication and mass communication, the telephone and other communication technologies simply disappear, because they enhance communication in a fashion beyond both communication types.

Third, media research was not sufficiently connected to more general questions of modern change. For instance, the early days of media research did not provide perspectives on the successive transformations from face-to-face communication towards technical mediation of meaning. For many reasons, the insights of Harold Innis and Marshall McLuhan were (partly due to their rhetoric?) rarely seen as an important point of departure for further theorising.

Fourth, the most widely distributed and used telecommunications medium, the telephone, appeared 'invisible' in everyday life. The discrete and mundane success of the telephone may also explain some of the lack of concern in media research for telecommunications.

A fifth reason for this lack of interest may be that the telephone has had a feminine image, deriving from its connection to women and the explicitly domestic. The telephone enhanced the widening gaps between private and public spheres of industrial society, and simultaneously provided means for female interaction. As a domestic medium, it became a 'feminine' medium, available for home-working women with the expressive and integrating duties of the family in relation to relatives, other family members, friends, etc. Therefore, it is no coincidence that the emerging social research of the use of the telephone is closely connected to gender issues. (See Fischer 1992, Martin 1992, Rakow 1992.)

Lastly, the development and implementation of telecommunication services is predominantly driven by a technical reason of engineers. The critical problem since the days of A. G. Bell has been the technical feasibility and improvements of the systems. In contrast, while also television is normally characterised as a 'high technology', it is fundamentally understood as a mediator of meaning and culture. Its technological features are seen as irrelevant compared to its cultural significance. Or rather, the cultural significance of mass media was seen as something detached from technology. Clearly, the diverging interpretation is related to the fact that the 'content' is created by the users themselves. Whereas mass media is dependent on a large, professional and highly innovative programming industry, there was no such industry in the telecommunications sector. No culture industry existed to balance the technological dimension. The considerable technical constraints in the telecommunications industry regarding diffusion of services, then digitalisation of the networks only enforced this. As one consequence, telecommunications, in spite of their role as mediators of meaning, were rarely discussed in relation to cultural issues.

Over the last decade, however, many of the distinctions between mass media and telecommunications have gradually evaded. This is due to technical changes in both kinds of media:

First, telecommunications adopt prestructured 'content' with narrative meaning. Emerging trends indicate that telecommunication services extend its area from the strictly personal and interpersonal to the 'mass' level. This results in three kinds of changes; in use patterns, technology and policy. The telephone and the telefax are used for distribution of standardised messages, as in marketing and fund-raising. 'Junk-calls' and 'junk-fax' and 'junk-E-mail' clearly approach the principle of mass communication. In professional life, there is clearly a trend to substitute answering machines for 'live' human voices. Voice mail and voice processing devices and messaging services may make it harder to reach a human being on the telephone. Instead, a standardised, preproduced message is presented. The similar trend can be seen in the increasing use of electronic bulletin boards, Videotex, and other databases and services as public information devices, and the coming service 'video-on-demand'. To the extent that hypermedia is linked to networks, they are certainly a prime example of the same. In the mid-eighties, one could readily refer to 'mass telematics' and telematic services as mass media without modifying accepted definitions of mass media. Telecommunications acquired preproduced 'contents', available for an unidentifiable 'mass' audience to interpret. While the traditional culture industry is not involved to the same extent in telecommunications, a software (and marketing) industry is expanding more than most industrial sectors. It seems like the information technology industry is developing a 'culture industry' of its own. The increasing affiliation to software means that communication technology acquires a cultural 'content'.

Second, and related, is the trend of telephone conferences (Tele-markets, chat-lines, etc.), video conferences, and computer conferencing to depersonalise telecommunication. More indirectly, electronic surveillance in public spaces and databases containing personal data also suggest that telecommunications to a greater extent have collective significance in that they involve the 'mass'. Telecommunications prove to enhance group or mass meaning because they increasingly operate in junction with mass media. In distance education for instance, electronic media are normally combined with printed media.

Third, as telecommunications become deregulated, they find themselves in the political and the commercial heat: As they become subject for private initiative (due to deregulation), they also become noticeable as subject for political conflict, leading to various problematic dilemmas. For instance, freedom of expression hitherto related to the mass media, relates increasingly to Caller-ID, censorship of chat lines ('dial-a-porn') the French 'messagerie', etc. To an increasing extent, they take part in the transformation of temporal and spatial conditions for social practices leading to ethical and political dilemmas. Just as the concept of 'public service' is at stake in broadcasting, the principle of 'universal service' becomes more visible for public attention when it is increasingly subject for deregulation.

Fourth, services based upon telecommunications increasingly join the matrix of media in domestic life and in the office. As the telephone receives new functions, computers, minitels, answering machines, and telefaxes find their way into the domestic market, they share a social context with the mass media. When telecommunications store messages of meaning, they escape evanescence. The growing use of technical equipment to enable storing of real-time media like television and telephone (answering machines, phone-mail, video recorders), erode the differences between real-time and asynchronous media. Certainly, the distinction between real-time and asynchronous media does not correspond to the difference between mass media and telecommunications.

Tools and objects - perceptive and hermeneutic mediation

All these socio-technical changes promote the perception of telecommunications, from something 'empty', individual and technical, to something with a collective and cultural significance. The trends increasingly locate telecommunications (factually and mentally) in a new, shared media matrix.

And yet there are substantial differences. It remains a fact that telecommunications are normally interactive (as in the case of data-bases) and two-way directed (as in the case of telephone). Clearly, they require a different response from the part of the user than do mass media like radio or television. Apparently, communication technologies require a much higher degree of involvement from the user, as they open up for instant, interactive and immediate dialogical processes. The difference is often conceived of as one of different levels of commitment from the part of the user. It requires more initiative and conscious action to make a telephone call than to watch television. As the receiver also is a sender, (or rather, as the sender-receiver scheme renders obsolete) communication technology requires determination about the intentions of the process than the more habitual media reception. This is the case both if one is involved in an interactive process with a hypertext terminal, or if one participates in direct, communicative processes with others through the telephone.

However, the intuitive 'observation' that communication technology requires a higher degree of determination than do mass media, only masks deeper and more significant user differences between the two kinds of media. However, the social and cultural significance of the various media is not explained by pointing out merely technological differences and similarities. Marshall McLuhan suggested a way of differentiating between media on a continuum between 'hot' and 'cold', according to their capacity to involve and engage the user emotionally (McLuhan 1964, 36). His idea was that while some media technologies furthered isolation and detachment from the community, others encouraged involvement into common affairs. The last kinds were those which resembled and simulated the direct, multi-sensed, non-linear and spoken communication patterns of the pre-modern village. From this, he derived that one-sensed, linear and visual media as printed media were 'hot' media, in that they did not need the 'warming' of the public. They favoured mental distance rather than involvement. At the other extreme of the hot-cold continuum was the 'cold' television, (and to a lesser extent, other 'electric' media) which promoted visual/auditive non-linear verbal communication. Hence, McLuhan's optimism derived from the appropriation of electric media at the expense of print media, leading toward a new tribal, 'global village'.

This distinction was not considered very successful. Suffice it to say here that the model relied primarily entirely on mental significance of the human sensorium as such, and less on the social and communicative relationship between the subject and human or symbolic objects. It was model of shared spirit more than of social solidarity. In the aftermath, it seems as the model was biased by the a priori aspiration to categorise the most powerful medium, television, as the promising new sensory and environment of human understanding.

Mass media enable relative polysemy, a cultural openness in relationship between producers and consumers: Along with (other) consumer goods and services, they involve cultural interpretation only relatively dependent on intentions on the producer side. Compared to telecommunications, cultural interpretation of mass media in everyday life proceeds more explicitly according to cultural values. The meaning inherent in telecommunication leans more to specific, direct and verbal constraints and response. Most telephone calls involve immediate co-ordination of meaning which can only take place in an ongoing conversation. The same is the case in computer-mediated communication. Although it would not be correct to label the telecommunication process a social system, it depends less on cultural values in society, and more directly on the immediate communicative interaction between specific individuals. It seems as in mass communication, on the other hand, the receiver possesses more interpretative autonomy in relation to the verbal message and less autonomy in relation to general norms, than do participators in telecommunication processes. In such processes the users are to a greater extent subjects, that is, producers of meaning that interweave, interpret, correct each other directly.

Consequently, it is often more appropriate to understand interaction through telecommunications as one interweaving interpretation process. For instance, rather than analysing a telephone call or computer-mediated communication as individuals delivering separate speech acts or 'notices', one should see them as interwoven and irreducible meaning-constitutive processes where contributions merge into mediation processes. This is one idea behind Everett Rogers' convergence model of communication (Rogers 1986, 200) and can also be traced to Bakhtins as well as Habermas' ideas of dialogue and communication. While both mass media and communication technology entail polysemy, the mass media present messages as something relatively independent of the producer and requires interpretative effort from the user. The interpretative space is more 'cultural' than in the case of communication technology; they draw upon larger circles and more complex forms of cultural values. Hypertext seems to develop towards a peculiar hybrid of the two mediation mechanisms. It is more connected to an established 'work' of an author than is telecommunication, but leaves far more autonomy to the reader than do most mass media. It merges perceptive and hermeneutic interpretation forms.

Technology is always practically used, perceived or developed in relationship with a subject. As previously stated, the role of technology in agency can be categorised

Here, I wish to focus on a) and b) although the contextual element is increasingly important. In technology-mediated action involving tools, an inherent contextual and purposive junction between the agent and the medium is at play as opposed to the world on the other. This, however, varies inherently between technologies. I would like to argue that this addresses one of the most important differences between mass media and communication technologies. Whereas technology-mediated action/perception inherently unites the subject and the medium toward the world, mass media, as projections and images of the world, distance themselves from the subject. Thus, the subject - medium - world relationship is quite different.

Both cases are forms of mediated everyday practices that principally entail the dimensions of agency. However, whereas perceptional technologies like communication technologies involve action predominantly towards the world through technology, mass media involve predominantly action towards the technology through the world. It is from this we can distinguish between two different forms of social integration processes; perceptive and hermeneutic integration.

I should here propose a preliminary definition of tools. With reference to Olav Asheim, Tom Johansen defines technology as 'materially embedded (fastlagte) ways of doing things' (Johansen 1992, 123). The definition, which stresses technology as socially constructed, is wide enough to encapsulate both the particular historical technologies in different cultures and the changing of ways of doing things, that is, the rationality involved. The definition is closely related to the subject in the complex of agency, and thus suitable for our perspective related to technology-mediated action. However, while in all its generality it is hardly incorrect, it is insufficient. In my own terminology this definition suits more as a definition of tools, than of technology as such. This is because it does not address the significance of technology beyond the level of action. We may define tools as one aspect of technology and as material ways of doing things. Communication technologies as tools, implies material/communicational ways of doing things in an extended time/space.

B) Agency

A study of hypertext and technology-mediated action implies an analysis of the alternative possibilities of action and the potential consequences for the ones involved. It should account for implications and consequences of alternative compositions of text. Clearly, this is not the way media researchers normally think, which shows that the problems posed by the new media differs substantially from the ones normally posed in mass media research. While it can account for some of the non-subject and non-technological aspects of the new media, i.e. through textual analysis of databases and computer conversations, it cannot explain the position of the subject in relation to the medium and co-subjects.

Thus, it seems like the use of the communication technologies most of all constitutes a theoretical challenge to theories of action. What we may call technology-mediated action constitutes an expanded and more dense dimension of social action in advanced societies. Only by relating the use of communication technology to sociology of action, of using in day to day life, one may profit from both media theory and theory of technology. Exploring the nature of technology-mediated action involves a sociological inquiry of contemporary forms of action orientations. The next step is therefore to relate such tools to the acting individual.

The tool, the context and the knowledgeable agent meet in agency. In the following I will discuss agency from different perspectives. Central here is the understanding of agency as composed of subject, tool and context in practical everyday conduct.

Let me start by considering some conceptions of action in sociology. The basic object of sociological enquiry, according to the prominent British sociologist Anthony Giddens, is neither the subjective experience of the individual actor, nor 'society' as a social totality (Giddens 1984). Rather than focusing on subjectivity or objective structures, one should connect all aspects of agency and relate them to social change. This is because practices are not just created by the individual actor, but reproduced objectively in ways that confirm the subject as subject. Social life is reproduced

In and through their own re-creation, practices recursively reproduce conditions for future events. To Giddens, the notion of intentionality may overstate the rational aims of the actor. The role of the actor should be understood in terms of knowledgeability and capability; her power to produce effects, and her awareness of her capacity to achieve outcomes (Giddens 1984, 257-258):

Giddens' theory represents an attempt to balance individual preferences for action with collective norms. As Cohen points out, Giddens' theory involves a 'decentering' of the subject in favour of a concern for the nature and consequences of the activities in which social actors engage during their participation in day-to-day life (Cohen 1989, 11). Thus the processes of practices as such are more central to Giddens than the actor herself. Though the subject is the interpretative agent who produces social life and its events, daily practices may be carried out without being directly or discursively motivated. They are neither carried out with the automatics of natural processes, nor with perfect consciousness. Social practices are recreated with what Giddens calls human knowledgeability, normally in the form of tacit knowledge. They are a part of the 'flow of life'.

However, this does not imply behaviourism or any other theoretical reduction of the subject. The acting and knowledgeable subject is vital to an encompassing notion of agency. Agency involves not only practices and interaction as such, but also the conditions, forces and means of action. The theoretical convenience of the notion of agency is that it involves much more than the actor him-/herself. Agency equates the constitution of social life, that is, all elements involved in, and conditions for action. This broad notion of agency is motivated by a concern in the theory of structuration to express the entanglement of individual and collective action, as well as the inherent connection between the conduct and its conditions.

Agency may be both individual and collective. As Sewell notes, it entails the ability to co-ordinate one's action with others and against others. Social movements for instance, are agents of themselves, and they are important foundations of individual agents.

Power means to possess transformative capacity. Thus, the power of agency means the capacity to intervene in a course of events or in a state of affairs, which requires 'causal powers' including tools. Through the capability of the actor to 'make a difference' as an inherent feature of agency, social practices always mediate power, and are always oriented towards potential change. An actor may in principle always change a course of events or a pattern of interaction. To speak of agency means that the individual could have acted differently:

This is implied in the recognition of social agents as knowledgeable and with practical (or even discursive) consciousness. The production of social life is a skilled performance even if the skills are only tacitly recognised. The two forms of integration (perceptive and hermeneutic) involve two kinds of user power. Clearly, they are both devices of power and control. But the power of the user of communication technologies reach beyond the on/off ritual, the opinion leader function or to the hermeneutic practices of media reception. The power to intrude, awake, register, harass, collect, survey, process, and distribute information, puts the user of communication technology in a critical position within agency because they precisely enable the power to change the course of distanciated events, to intervene into social processes in extended time/space.

As Giddens, the French sociologist Pierre Bourdieu attempts to transcend the usual emphasis on individual choice while not relapsing into the contrary position of structuralism (Bourdieu 1977, 1990). He wants to overcome both determinist and voluntarist images of action. Both modes of knowledge are equally indispensable to the social sciences, but to apply the insights of both, they must be transcended.

For example, what is economically sound or most efficient may not be what guides individual action in practical life. Tradition and culture developed under different conditions continuously guide action where they are not economically rational. And reversibly, the symbolic character of practices may be as important to the reproduction of a social formation as their material character. Thus Bourdieu wants to develop a mode of knowledge and practice that is genuinely practical. Rather than defining action internally (choice) or externally (structure), Bourdieu wants to explain the dialectic of incorporation and objectivation: The internalisation of objective structure (culture as well as materiality) and the externalisation of internal competencies. Only in this way, one can understand social order and social change.

The principle of habitus

The concept of habitus represents the crux of this argument. Habitus is the social logic of which daily practices are guided. It coins the 'rationality' of everyday practices. One may conceive it as an ethos or disposition that orient practice according to an unrecognised symbolic economy, but it is also a principle of social and cultural competencies of the individual, adapted creatively to specific contexts in everyday life. It reflects the social condition under which it is unfolded, but it also leaves space for individual variability in thoughts, perceptions, and actions. It represents the 'collective unconscious' and defines the repertoire of action in specific contexts. It neither guides automatic ritual, nor autonomous choice. Rather, it composes practical conduct by mediating between the internal and the external, for instance between personal style and collective norms. As Thompson holds,

In socialisation processes, the individual adapts new experiences to already internalised sets of preferences that then also become affected. Habitus is a product of socialisation in an established cultural order. Miller emphasises this when he defines Habitus as

Patterns of morality, exchange relations, architecture, art and other approximately established cultural systems become part of the individual's personal competence through social interaction. Habitus thus mediates between material conditions and practices of social segments. It represents a constructionist competence that is systematically differentiated in society, and unites and separates social groups. The dialectic of Habitus is amply expressed in a definition of habitus as:

And similarly:

The cultural principles that constitute a habitus are socio-economically differentiated, for instance in terms of gender, class or urban/rural background. The habitus therefore tends to reproduce social differences through defining and homogenising cultural boundaries. This sociological fact represents the structural or given side of habitus. In the area of cultural consumption, cultural capital (based on time invested in obtaining cultural knowledge) structures levels of habitus and the following differentiation of taste (Bourdieu 1990, 80).

And yet reproduction is entirely dependent on individual interpretation of established classificatory systems, resulting in various practices that can only be understood subjectively, through a constructive approach. Bourdieu's perspective enables a view on structured sets of dispositions, and their consequences for the reproduction of power and social control without relapsing into objectivism. Social change are accounted for through the emphasis on indirect, embodied rules that regulate action. There are no other structuring than the reproduction of conditions that reproduce cultural homogeneity (Bourdieu 1990, 80). Objective structures are themselves products of historical practices. Although socio-economic structures can be seen reproduced in cultural practice, (as can easily be demonstrated through statistics), modified practices and 'readings' of the world always break away from hegemonic patterns.

The principle of habitus has been applied to a number of areas of culture by Bourdieu and others. It helps to explain how and why specific cultural patterns reproduce themselves. It connects culture with social control because it exposes the relationship between cultural practices and economic and cultural capital. In media research, for instance, it clarifies that there are not only different things to select from on television, there are also different ways of seeing and understanding television content, which co-varies relatively with the capital-differentiated habitus. The hermeneutic point of meaning is refined and connected to socio-economic analysis of audiences, and the incorporation of stratified life conditions can then be traced in the social use of the media. Though the use and understanding of media content are individual and unique, it also demonstrates inscriptions of structural distinctions. The embodying of cultural-economic structures in individual dispositions may provide systematically differentiated cultural media practices.

Bourdieu's position relates the unconscious reproduction of class divisions by individual practices in various situations. This structural aspect of practice is thus invisible for us in daily life, and can only be exposed through empirical research that connect accumulations of subjective interpretation to conditions of capital. As a principle of orientation, habitus blends ideas and behaviour in the practical because the roots of both aspects are incorporated in us through language and socialisation. It also blends the material and the symbolic. In the case of the social use of communication technologies, the practices are guided by the merge of tools and tacit knowledge.

There is a close connection here to Anthony Giddens' structuration theory. Structuration is precisely a process that embodies a mediation between action and system without distinguishing between the two with separate 'levels'. Structure is the medium and outcome of action (Giddens 1984, 25). Bourdieu's combining of constructionist analysis and structuralism provides understanding of how the subject relates to material-cultural products, and how materiality mediates social relationships. Bourdieu has been more successful than many others to explore how social action involving objects is related to culture, and to demystify this empirically. It resembles structuration theory in that it seeks to combine structural and action theory into one approach where it becomes apparent that materiality plays an important mediating role. It may then be possible to avoid seeing technology as either neutral tools or total systems of technique. This is perhaps the major theoretical challenge of a sociology of communication technology.

Agency and the world of artefacts

The individual learns to apply principles of practice toward cultural products, on new areas. It acquires the competence to reproduce certain cultural principles and to apply them pragmatically in new contexts. Habitus also inspires subjective probabilities that give aspirations and prospects toward social hierarchies. For instance, habitus provides the individual with classificatory schemes toward art and cultural artefacts. To a variable degree, artefacts have symbolic value. Habitus mediates differentiated appropriation of external facts. Accordingly, it provides systematically different interpretations of cultural artefacts.

Technology and materiality play a prominent part in the understanding of agency. Such a broad understanding of agency as a theoretical precondition presumes the inclusion of tools. Among other things, agency not only refers to choices of tools but also to the tools themselves. Material and social conditions of social life have constraining as well as enabling influences on the acting subject and on the recurrent day-to-day practices. Day-to-day practices increasingly imply routines involving technologies and materiality. Technology-mediated action therefore, implies an inherent (but not ontological) connection between the actor and the tool. The choice of a different tool implies what aims she intends to realise. The 'choice' of different tools implies that the agent acts differently, a completely different form of conduct. Subjectivity and tool negotiate each other in the constitution of agency.

As one dimension of the mediation of communication then, the tool is medium and outcome of technology-mediated action. This means that the subject and the tool should be distinguished ontologically, but seen as an analytical totality. They inherently condition and affect each other in the course of daily life conduct. Material means and capability/intentionality prescribe each other mutually. As emphasised, for example to send a telegram, fax or electronic mail to my colleague cannot be considered as the same practice, merely involving alternative means. The acting subject does not choose between means but between actions. The tools are an inherent part of the particular action, hence, of agency. The tool 'suggests' its contextual availability to the subject. This point is often forgotten, precisely because the technology itself 'withdraws' from our attention.

However, this does not mean that the notion of the subject in technology-mediated action can be left out. The subject makes pragmatic choices according to social rules, that cannot be reduced to the power of the tool or of technological systems. Therefore the ontological distinction between the human subject and the material tool cannot be wiped out. But, neither can the tool be reduced to its everyday sense as neutral assistance, leaving the subject in complete power of singular practices. To understand everyday action, one must address the merging powers of subject, tool and context in the flow of practical everyday conduct. Only in this way (human) subjectivism and (material) objectivism can be transcended.

The distinction between tools and action is important for several reasons. First, the actor has the choice of applying technical means or to ignore them. Giddens' ontological notion of the actor's ability to make a difference for example, implies that the actor can principally abstain from technology-mediated action, though this is often impossible in modern daily life. Second, the actor may choose between different means of interaction, such as the letter, the telephone, the train etc. It is not suggested here that the identical message can be transmitted through several media. In this respect, McLuhan's famous credo about the media and the message is quite accurate. However, in most cases, it is not vital to the sender (or the receiver) to transmit the message in an exact form, but to get the message, in a pragmatic sense, across. The medium is not the message from the point of view of the sender, because the intention of a message can be delivered through different media. To emphasise the analytical distinction between tools and actions means to highlight the degrees of freedom that potentially are at the disposal.

C) Objectivation of knowledge

In a discussion of the interrelationship of action and hypertext as object/tool, the problem of objectivation cannot be avoided. In technological theory, the problem of objectification (or externalisation) is usually addressed as a large-scale, cultural process, and rarely in contrast to agency. Here, I shall stick to the individual problem of objectivation of habitus, of the vital recourse of agency called tacit or implicit knowledge.

The Wittgenstein of Philosophic Investigations (PI) turns away from his attempts to account for the formal structures of language in order to investigate its use (Wittgenstein 1978). Subsequently, he concludes that the users/subjects are central determinants of language. Language as an analytic tool for experience depends on language-in-use, the practical everyday use of it (PI, p 30). A central concept in PI is language games, although it is never clearly defined (PI, pp 7, 71). However, it becomes clear that it is the language-in-use of social interaction.

Inspired by Wittgenstein, Michael Polanyi elaborates the concept of tacit knowing. This is individual knowledge that cannot readily be accounted for through language. To use one of Wittgenstein's examples; most of us know what the clarinet sounds like, but it is almost impossible to explain (Wittgenstein 1978, 78). Polanyi's argument is that not all knowledge is formal knowledge, that it can not be isolated from its context and that formal knowledge can account fully for all aspects of practices. Non-verbal competence and knowledge are aspects of all personal knowledge (Kirkebøen 1993,363). This insight appears in various understandings of the life world, in Bourdieu's concept of Habitus and in Giddens' theory of agency in structuration.

A central problem here is whether individual competencies (i.e. tacit knowledge, habitus) become narrowed, specified and included into technology and the world of objects, and thus becomes what we normally label information. Tradition seems increasingly relieved and detached from its educating and normative role. In earlier times, one learned how to swing the axe by one's father. Today, the computer itself instructs in detail how it should be used. Important aspects of the cultural task of socialisation related to technology in everyday life, are transferred to the technology and materiality itself. Human competence becomes marginalised by technology and materiality. This thesis is suggested by Tom Johansen (1992) and I will discuss the relationship of knowledge and tools considering his arguments.

Does the displacement of rules into technology mean that tacit knowledge loses its significance in learning and application processes? What is left of personal competence when technology emerges as its own cultural mediator? For instance, the ideal of starting to use a computer with modem and a communication software for the first time without hands-on instruction, may in fact suggest that technology, step by step, takes the power over central aspects of agency. Clearly, for the user this means fewer frustrations. 'Everyone' can take advanced technology into use, irrespective of prior competence. The seamy side of this is that the user apparently looses the possibility to act from personal informal knowledge.

If tacit knowledge must be acquired through inter-personal learning, through learning by watching and doing, and not through theoretical knowledge, one would expect tacit knowledge to become of marginal significance. Tom Johansen argues that tacit knowledge (implisitt kunnskap) dries out and with it, the relational understanding it provides. As materiality and technology increasingly take care of their own implementation and use, action of this sort becomes independent from context and tradition. Technification of everyday life implies increasing redundancy of tacit knowledge:

This implies that the significance of tradition for everyday mediated practices becomes subordinate. Instead, what increasingly dominates daily life is specialised expert knowledge. As a consequence, daily conduct becomes increasingly dependent on objective technology and materiality at the expense of subjective flexibility and direct social interaction. In short; technology colonises personal competence - systems-technical integration substitutes for social integration.

Quite correctly, Johansen writes that:

He then goes on:

Johansen differs between technically mediated and habitus-mediated action, implying that the first mediates action in a narrow, technical way, whereas habitus mediates action through tradition, context and the understanding of the agent (Johansen 1992, 121). The first separates meaning from use through objectivation and reduces the agent to just another means. In contrast, habitus, as shown in the discussion above, mediates culture and tradition internally through personal competence. The first mediates formally, the latter through meaning. Both types mediate culture; the first through materiality, the second through the agent.

According to Johansen's understanding of the term, habitus provides the possibility to organise reality through practices according to cultural norms about right and wrong, honour, integrity, etc. However, it should not be conceived as merely action according to cultural norms and values in an Durkheimian sense. It is essential to the logic of Johansen's objectivation theory that habitus provides the capacity to act differently in different situations and to carefully supersede what is normatively defined. Indeed, from my reading, it seems that the point of creativity and improvisation is more explicit in Johansen's than Bourdieu's discussion, the latter being frequently accused of relying too much on structuralism. Involving creativity as much as conformity, Habitus is a principle for handling uncertainty and for interpreting ambiguous contexts. It is the everyday art of applying flexibility from situation to situation. It involves mental understanding and common sense, if not the reflexivity to formulate the choices adequately.

Johansen's argument is that the reproduction of culture has step by step moved from internal to external mediation, from mediation through habitus to mediation through materiality. Johansen turns away from the Weberian categorisation of action orientations and rationalities, to focus on the transitory nature of mediation of action, and the subsequent historical change from human to material mediation. Habitus becomes increasingly irrelevant for a number of everyday life situations in a society textured by technology. For instance, to drive a car through the city involves that one follows a wide range of formal rules in a very detailed fashion. From the fact of technological society, Johansen concludes that tacit knowledge, inherent in the principle of habitus, renders increasingly obsolete.

Technical civilisation involves the use of things in most activities. 'Material structure', according to Johansen, is the sum of external forces in action when artefacts are taken into use. Johansen assumes that all forms of action and interaction (samkvem) that somehow apply materiality, must somehow relate to this material in the same practical/technical way, irrespective of whether the action is religious, ritual, work, etc. This, in contrast to all forms of action that are not mediated by materiality, irrespective of context, purpose and social sphere. Johansen thus develops a fundamental distinction between materially and textually mediated practices - those which are mediated externally through materiality, and those that are mediated internally through habitus and tradition. Action is guided by the material structure or by the cultural arsenal. The development of materially mediated action means increasingly closed rules for the user, and narrow material 'tracks' for later generations of users/agents.

Johansen suggests that this leads to an increasing technical understanding at the expense of a cultural understanding of the world. He seems all too easily to equate technology-mediated action and technical understanding. The vast artificial and technological landscape in modern society indicates to Johansen that technical understanding or technical rationality dominates everyday life:

Johansen also holds that everyday knowledge incorporates the instrumental-analytic attitude of science, for instance by reducing problems to its singular elements or by functional division of labour (Johansen 1992, 104). This is one of the ways science legitimates itself. This also implies that everyday lives adapt an asymmetric and controlling attitude towards nature and increasingly towards other humans, as evident in the expert-client relationship. Whether everyday knowledge structures actually become in some sense scientific, cannot without implications be reduced to an empirical question because it means application of science to evaluate 'itself'. However, Johansen cannot explain why we do not act like scientific operators or 'replicants' in everyday life. It seems as if Johansen confuses technification and materialisation in everyday life with scientific rationality.

Alternatively, I hold that in everyday life, precisely due to our tacit competence, science disappears in technology. I do not necessarily refute that technologies are application or objectification of science (although the reverse is as much the case), that technology is what Marx called 'frozen science', nor to address the (naive) claim that two different projects are pursued; that science is directed toward truth and technology toward efficiency. However, from the perspective of the user, only the technology (or rather; its tool dimension) is visible. Science becomes objectified in everyday technology. Science 'hides' itself in the guise of tools and through this, opens itself for tacit knowledge in day-to-day practices. This is why science in the form of scientific reasoning is peripheral in day-to-day life. That modern everyday life is increasingly influenced by scientific knowledge, (such as medicine and psychology in child care, economics in our personal economic management, knowledge about nutrition in our cooking, etc.), does not imply entirely scientification and colonisation of everyday knowledge mediated by scientific rationality. On the contrary, in social life, people receive and apply such knowledge in a tacit, non-scientific way.

What remains is the tacit resources of everyday life that produce non-technical understanding, and in fact do so by drawing upon science and modern artificiality. Johansen cannot account for the translation of scientific and technical knowledge into practical common sense of the life world. One reason for this confusion is that Johansen develops a theory of technology and materiality in everyday life without any ontological concept of interpretation in our 'lifeworld'. He believes that formal/objectivated knowledge is complementary to tacit knowledge, while the two forms in fact precondition each other.

Materially mediated action takes place in and through materiality and technology. This means that the use of tools or the living in materiality can only be understood through an encompassing concept of agency that includes various forms of knowledge and the materiality involved. To drive a car through heavy traffic or to operate a computer requires as much habitus and tacit knowledge as the use of the stone axe once did. Technology-mediated action prescribes as much tacit or implicit understanding as other forms of action in everyday life. Advanced technology cannot be used exclusively through external instructions. Throughout history, habitus and tacit knowledge has been in a process of continuing transformation according to new environments. However, with the transformation toward advanced modernity, habitus provides resources for how to go on in everyday life as much as in pre-modern times and cultures. In fact, what Johansen calls strategic rules resemble precisely habitus in a technological environment.

The process of objectivation of knowledge should be reformulated to a problem of objectivation of culture. What is taking place is that relatively specific 'knowledge' in the intersubjective cultural world is incorporated into materiality and culture. One may distinguish between two processes: a) the professionalisation of general cultural knowledge to expert knowledge, and b) the individualisation and materialisation of professional knowledge to 'self-service' knowledge (Gershuny & Miløs 1983). This means that the objectivation of knowledge - which is what we simply call information - is normally paralleled by personal, tacit knowledge of how to orient and handle the materiality. The most important case is alphabetisation: With the explosion of objectivation of knowledge into information as typed text since the late 15th century, a general increase of the competence to read and write has followed. Johansen fails to distinguish between objective cultural knowledge as information and personal, tacit knowledge. When the first kind becomes increasingly objectified the latter kind automatically follows in Johansen's scheme, causing unfortunate reductionism.

The dilemma between implicit knowledge and materially mediated action as formulated by Johansen is therefore misconstrued. In traditional society, there were good or bad tailors and locksmiths. Both their efficiency and their artistic performances varied. Today, there are good and bad programmers and car drivers. In a technologically advanced society, the problem of competence is just as relevant as then. There is still a question of using technological tools with reasonable flexibility and competence, just as it is still a question of acting with reasonable flexibility in material contexts.

Generally, to distinguish between materially mediated action and habitus irrespective of use or context does not uncover the varying challenges technology and materiality constitute for the modern individual. It does not account for how individuals go on in a highly technological society. Neither does it account for how habitus is reproduced through technology in modern society. Johansen's rejection of Weber's differentiation of rationality types leads him to distinguish between two forms of rationality; technological and non-technological rationality. This unfortunate reduction of the role of materiality and technology leads to unjustified technological determinism, since there appears to be only one way of using technology; following Johansen, society will increasingly act according to distinct technological rules. A technological society can only proceed according to the formal and reductionist rules and preferences inherent in its technology. Hypertext and similar media, according to this view, lead to an emptying of personal competence - irrespective of the quality or nature of the media. Alternatively, I hold that we must keep tacit knowledge separate from information, that tacit knowledge is a necessary requirement for handling technology in everyday life, and that tacit knowledge changes along with new and increasing technology in daily life.

Practical action and the 'art of using'

Maurice Merleau-Ponty notes that life-world perception competently gears the body through the world of artefacts. The perceptual awareness he describes here is precisely tacit knowledge inherent in practical conduct:

It is impossible to understand technology-mediated action in everyday life without seeing it in relation to tacit knowledge. Tacit knowledge is informed by both expert knowledge necessary to master new communication technologies, and by the implicit intention, such as to transmit a message, to communicate with others, to entertain oneself, etc. Tacit knowledge fuels and reproduces the meaningfulness of technology-mediated action and thus is a complete part of it.

Success for communication technologies in everyday life implies that they become mundane and 'trivial' instruments embedded in the flow of everyday practices. The 'easiness' of computer technology in everyday life is really a question of whether the technology can be adequately translated to the level of tacit knowledge. The private telephone, for instance, appears as an undisputed success, while computer-mediated communication is associated with more technical trouble. Tacit knowledge is a vital aspect relating the subject to her socio-material environment. 'Easiness' means wider 'space' for subjective intentions. Tacit knowledge is guided by common sense, 'between' automatic and calculated action, and by social rules integrated with technological tools. When we say that a computer is more or less 'easy to use', we really refer to the tacit correspondence of the rules and recourses of the technology, and tacit knowledge inherent in practical action.

Intersubjectively produced common sense cannot be substituted by user manuals and instruction courses. Technology-mediated action can only be produced through meaningful and practical application of tools and the non-discursive reproduction of mutual knowledge that come close to routines and procedures in everyday life. Only then, communication technology seizes its alien character as simply materiality and becomes translated into tools. In this light, the task for hermeneutic social science is, for instance in the case of hypertext, to investigate the technical and contextual conditions for this translation.

Practical action, then, does not emerge from pre-programmed responses, nor from subtle reflections. Tacit knowledge is inherent in the practicality of daily action itself. Communication technologies are part of the action they mediate through the rules and recourses of their symbolic and material character. Let me give one example. If I want to communicate to my mother while I am at home, I do not bracket this down to: 1) communicate to my mother and 2) different options available for doing so (write a letter, take the bus to her home, make a phone call, etc.). Rather, the decision is communication-through-telephone, communication-through-letter or communication-face-to-face, and so on, which are quite different experiences, for both parts. The intended communicative process and the technical means for accomplishing it therefore, are inseparable in a practical context. The choices and use of communication technologies define the particularities of mediation with its distinct communicative characteristics (spoken, written or typed words, monological or dialogical feasibility, etc.). Thus, communication-technological mediation is the medium and outcome, of all aspects of technology-mediated action. Therefore it is irrelevant to distinguish too sharply between the use of terminals (giving commands, dialling, etc.) and the distanciated process of mediation. Technology-mediated action reproduces mediation both through manual and linguistic operations. Both are inherent in technology-mediated action as agency.

Rather than comparing and contrasting tacit knowledge and objectified knowledge, the challenge is to understand what human, moral, knowledgeable (individual and collective) agents do with their tool-systems and with their objectified knowledge, giving various unrecognised consequences. In other words, tacit and objectified forms of knowledge are not complementary but preconditioning in various compositions and contexts. Practical actions are always necessary in the wrestling with the new objectified contexts and mediators of social relations. In a technology-textured society, the question is not how objectified knowledge excludes tacit knowledge, but how they modify each other into technology-mediated agency. Lundequist summarises this aptly:

Tacit knowledge is the 'invisible software' always at play in agency. Day-to-day life is composed by activities guided by relatively taken-for-granted assumptions, whether one carries traditions and routines further, or whether the activity involves creative impulses. Tacit knowledge is composed of daily competencies, confidence toward other humans and technology and experience. It is knowledge yet to be formalised or made explicit by for example science. To formalise knowledge in hypertext systems, for instance, does not mean that tacit knowledge becomes correspondingly reduced. Rather ironically, the 'critical' proponents of the objectivation of knowledge-thesis, which fear a reduction of personal tacit knowledge, tend to conceptionally reduce 'knowledge' to information. In so doing they join forces with 'Scientific Management', schools of behaviourism as well as a number of works on Artificial Intelligence which stand accused of reducing knowledge to formal knowledge, and agency to instrumental operations for the more or less explicit purpose of targeting and testing knowledge 'scientifically'.

To use communication technology implies both technical mastering and the relationship between the use process and the conventions of everyday life. Formal knowledge is not sufficient. One needs to know implicitly how the technology is interwoven in wider processes of meaning, the 'place' of the technology. To learn how to use technology is to learn about the world which the technology inhabits. Tacit knowledge includes a certain understanding of what we do when we use. The articulated part of the required knowledge is only the tip of the iceberg.

As a field of play, a framework of activity, communication technology is only partly in the hands of the systems that created its technical facilities (Feenberg 1991, 88). In spite of user manuals, commands and instructions, it can never define its own use completely. Communication technology escapes partly the totality of technocratic logic and opens itself to differentiated and undetermined practices. Andrew Feenberg writes:

Contrary to the argument that advanced technology implies a sort of necessary Taylorism contradicts their flexible and dynamic potential. Rather that being based upon inherent, rigid controls as in the case of mechanic technology, computer technology employs a wide range of electronic commands separated from the machine itself. This allows for an unprecedented flexibility in selecting goals and for the ways to reach them. In other words, it increases the role of the mind, skill and will of the agent. Technical design permits this in order to make both technology and labour more flexible in a rapidly changing production life. The flexible adaptations of technology are supported by its communicative dimension that may give work a learning dimension which blurs the distinction of mental and manual labour, of education and work (see Hirschorn in Feenberg 1991, 94, Feenberg 1991, 95). In fact, in production life as well as in everyday life, automation that extends the function of the machine and reduces choices of the agent may simply be irrational even from an instrumental and productivist perspective. Rather than subscribing to the 'objectivation' thesis presented above, I propose that communication technology is located in a sphere of competing perspectives. The following statement by Hirschorn is as much relevant for consumption, as for production technologies:

Communication technologies as hypertext are ambivalent technologies available for a plurality of alternative developments (Feenberg 1991, 96). Software, design and language embody assumptions that can either invite or extinguish human skills and involvement. Besides the argument that communication supports flexibility, there is an argument related to context: In everyday life, and in domestic contexts in particular, the space of autonomous manoeuvres are larger than in working life. In the household, most aspects of activity (context, leadership structure, goals, motivation), are less formal and explicit. The hidden cultural agenda of communication technology can only marginally be detected in the technical design.

Insofar as agency and habitus contain formal knowledge and that habitus is differentially reproduced, however, makes it plausible that technical integration creates new social problems of inequality. Habitus may prove insufficient for the mastery of communication technology for some social groups. However, this regards the formal stock of knowledge. Further, it is related to that computers and advanced communication technologies are associated with certain (highly educated) segments, reproducing images of a technology reserved for some groups.

That technology is socially constructed in the production as well as in the use process means that social and cultural (symbolic, mental) force makes communication technology into what it is to us. Further, technological development and growing complexity implies that technology is increasingly culturally constructed and subsequently less defined by the techniques 'itself'. This is not to say that tacit knowledge or habitus in general become gradually transferred from culture to techniques. It means that one needs cultural supported competence to recognise the purpose, reason, relevance and applicability of technology. It is probably easier for a complete stranger (due to the current state of globalisation, we probably would need an alien from outer space) to guess what an axe can be used for than a telephone or a computer. Little is given from the techniques itself, because of its growing 'closeness'. With the case of the computer, it is not at all clear where or how the technology meets its environment (natural or social) as it is with the blade of the axe. And naturally, this cultural competence, the habitus, is socially differentiated in a society of inequality.


According to Bolter (1991), we live in the late age of print - a diagnosis I do not subscribe to. However, it is clear that new forms of writing and reading emerge, hypertext being only one case. This paper started from the assumption that we must modify our conception of 'reading' in relation to hypertext and similar forms to a hermeneutical (interpretive) notion of agency or action.

Since I wanted to clarify the tool aspect and perceptual mediation, the paper attempted to clarify some central aspects of the term 'technology-mediated action'. First, technology-mediated action was culturally specified (and distinguished from the usual mass media) by relating it to tools and perceptual mediation. Second, technology-mediated action was connected to the acting subject through the notion of agency.

One central problem in the interrelationship between action and tool was discussed; the thesis of objectivation of habitus. As an alternative position, I argued that tacit knowledge and habitus live on in a highly technical integrated society as in previous phases of history. While formal knowledge is increasingly implemented into technology, tacit knowledge always transforms itself according to new technological environments. The problem is not the tacit knowledge required to handle advanced technology, but the symbolic character of such technology that may exclude it from groups with less symbolic, cultural or economic capital. Advanced communication technology, such as hypertext, is ambivalently located in a contested terrain of multiple definitions, contexts and interpretations, where it may follow roads leading to both autonomy and control.


Arendt, H. The human condition. Chicago, University of Chicago Press, 1958.

Blumer, H. Symbolic interactionism - perspective and method. Englewood Cliffs, N J, Prentice-Hall, 1969.

Bolter, J. Turings man, western culture in the computer age. Chapel Hill, N. C., The University of North Carolina Press, 1984.

Bolter, J. Writing space: the computer, hypertext, and the history of writing. Hillsdale, NJ, Lawrence Erlbaum, 1991.

Bourdieu, P. Outline of a theory of practice. Cambridge, Cambridge University Press, 1977.

Bourdieu, P. The logic of practice. Cambridge, Polity Press, 1990.

Bourdieu, P. Kultursociologiska texter. Stockholm, Brutus …stlings Bokfšrlag, 1991.

Bourdieu, P, Coleman, J S. (ed.) Social theory for a changing society. Boulder, Westview Press, 1991.

Bråten, S. Dialogens vilkår i datasamfunnet. Oslo, Universitetsforlaget, 1983.

de Certeau, M. The practice of everyday life. Berkeley, The University of California Press, 1984.

Cohen, I J. Structuration theory: Anthony Giddens and the constitution of social life. New York, St. Martins Press, 1989.

Dreifus H. What computers can't do. New York, Harper & Row, 1979.

Dreifus, H. Mind over machine: the power of human intuition and expertise in the era of the computer. New York, The Free Press, 1987.

Feenberg, A. Critical theory of technology. Oxford, Oxford University Press, 1991.

Fischer, C S. America calling: a social history of the telephone to 1940. Berkeley, University of California Press, 1992.

Gershuny, J, Miles, I. The new service economy. London, Pinter, 1983.

Giddens, A. New rules of sociological method. London, Hutchinson, 1976.

Giddens, A. Profiles and critiques in social theory. London, The Macmillan Press, 1982.

Giddens, A. The constitution of society. Cambridge, Polity Press, 1984.

Giddens, A. Social theory and modern sociology. Cambridge, Polity Press, 1987.

Giddens, A. The consequences of modernity. Cambridge, Polity Press, 1990.

Giddens, A. Modernity and self-identity. Cambridge, Polity Press, 1991.

Goffman, E. The presentation of self in everyday life. New York, Pelikan, 1959.

Gšranzon, B. (ed.) Datautvecklingens filosofi: tyst kunskap och ny teknik. Stockholm, Carlsson & Jšnsson Bokfšrlag AB, 1983.

Held, Thompson. (ed.) Social theory of modern societies: Anthony Giddens and his critics. Cambridge, Cambridge University Press, 1989.

Ihde, D. Technology and the lifeworld: from garden to earth. Bloomington, Indiana University Press, 1990.

Inquiry, Journal of philosophy, (Symposium on Technology and Human Values) Scandinavian University Press, Oslo - Stockholm. Vol. 35, Nos 3/4 September/December 1992 .

Johansen, T. Kulissernas regi. In: Andersson, Johansen, Nilson & Østerberg: Mellan mŠnniskor och ting. Gothenburg, Bokfšrlaget Korpen, 1985.

Johansen, T. Kulissenes regi: mellom ting og tekster. Oslo, Universitetsforlaget, 1992.

Kirkebøen, G. Psykologi, informasjonsteknologi og ekspertise. Avhandling for dr.philos.-graden. Oslo, Institutt for lingvistikk og filosofi, Universitetet i Oslo, 1993.

Landow, G P. Hypertext: the convergence of contemporary critical theory and technology. Baltimore, MD, Johns Hopkins University Press, 1992.

Martin, M. 'Hello, Central?' Gender, technology, and culture in the formation of telephone systems. Kingston, Montreal, McGill-Queen's University Press, 1992.

McLuhan, M. Understanding media: the extensions of man. New York, McGraw-Hill, 1964.

Miller, D. Material culture and mass consumption. Oxford, Basil Blackwell, 1987.

Nielsen, J. Hypertext and hypermedia. San Diego, Academic Press, 1990.

Rakow, L F. Gender on the line: women, the telephone and community life. Urbana, Ill., University of Illinois Press, 1992.

Rogers, E M. Communication technologies: the new media in society. London, The Free Press, 1986.

Sewell, W H. A theory of structure: duality, agency, and transformation, American Journal of Sociology, 98 (1), 1-29, 1992.

Thompson, J B. Ideology and modern culture. Cambridge, Polity Press, 1990.

Winograd, T, Flores, F. Understanding computers and cognition: a new foundation for design. New York, Ablex, 1986.

Wittgenstein, L. Filosofiska undersøkningar. Stockholm, Bonniers, 1978.

Østerberg, D. (ed.) Handling og samfunn. Oslo, Pax forlag, 1990.