1. Introduction

In this paper I will try to develop a framework for analysis of democracy in the information age. In developing this framework I will put forward five arguments. Firstly, I will argue that in trying to understand the role which ICT´s plays in the ’political world’ or for our democracies we need to move social practices to the center stage in our analysis. However, in order to avoid any kind of technological determinism, and also to be more precise about the character of these (political/democratic/technological) practices, I prefer to talk about these as ’technologically mediated innovations in political practices’ or TMIPP’s. Secondly, the use of this concept is a part of trying to argue that we need to see such practices in a constructivist perspective. Thus, even though practices are in a sense ’given’ by themselves, they are always interpreted and understod in certain and sometimes differing ways by their agents and other actors involved in the area in question. Thirdly, I will argue that TMIPP’s as ’discursive practices’ always (already) embody certain ideas on how ICT can and should be used in politics, and the meaning it therefore has for democracy. Another way to express this idea is to say that TMIPP’s is a simultaneous articulation of the perceived characteristics of certain technologies, ideas on citizenship and democratic procedures. Fourthly, I will argue that the central elements in understanding citizenship and democratic procedures are the concepts of ’individual’ (or ’individuality’), ’community’ (or communality), ’political leadership’ and ’laymanship’ (or ’political amateurism’). These are the central concepts around which traditional democratic theory has evolved, and even though one can argue that the content of these concepts has changed, and (probably) will change considerable in the information age, I still find it wise to take these concepts as a point of departure for the analysis of democracy in the information age. However, we have to be aware, as noted above, that even though these concepts have a reasonable precise meaning in traditional theories of democracy or citizenship, they are also subject to different interpretations and different articulations with ideas on the use of ICT in ’the political world’. Fiftly, I will argue, on the basis of current litterature on changes in our political/administrative systems, and the use of ICT in this connection, as well as a number of case studies undertaken, that at present we see four competing discourses on ICT and democracy emerging. These we have elsewhere (Hoff, Horrocks & Tops op.cit.) called "emerging models of democracy for the information age", and suggested that they should be called ’the consumer democracy model’, ’the demo-elitist democracy model’, ’the neo-republican democracy model’ and ’the cyberdemocracy model’. In fleshing out these models I shall try to describe the certain understandings and uses of ICT involved in the TMIPP’s connected with each of these models.

As I have been asked to clarify both my argument that the traditional concepts of democracy (individual, community, political leadership and political ’laymanship’) might still provide us with some analytical milage when trying to understand democracy in the information age, as well as my ideas on the four ’models of democracy’ in the fruitful comments I have had to the ’democracy platform’ by especially Mr. Anttiroiko, I will start by presenting the traditional views on democracy, in liberalism, republicanism, communitarianism, and in the less traditional view of radical-democractic theory (par.2). I will then move on to see how the relationship between ICT and democracy is understood in a utopian or dystopian perspective (technological determinism), and in a realist perspective, and I will then move on to see how this relationship is perceived in a constructivist perspective (par.3). Finally, I will then try to make the case for the four emerging ’models of democracy for the information age’, and try to describe the content of each of these models (par.4).

2. Discussing ICT and democracy. Can we take our point of departure in the traditional concepts?

In this paragraph I will demonstrate that the concepts of the ’individual’, ’community’, ’political leadership’ and ’laymanship’ are the central concepts of the traditional theories of democracy. Thereafter I will argue that these concepts (or dichotomies) still constitute the most important bearings in contemporary discussions on ICT and democracy, and that it therefore seems prudent not to give them up overnight, but rather to try to understand the new meanings these concepts are given by different actors in contemporary discourses on for example ’electronic democracy’, ’cyberdemocracy’, etc.

Liberalism

The liberal tradition has many faces, from Locke (1963), through social liberals such as J.S.Mills (1982), to modern neo-liberals such as Nozick (1974). However, the paradigm statement of contemporary liberal thought is undoubtly formulated by John Rawls (1971, 1985). For Rawls, as for all liberals, personal autonomy or freedom is a dominant concern. The self is, as Mulhall and Swift put it (1992:41,158) ’antecedently individuated’ and people’s goals or preferences are formed independently of others. Individuals should be free to define and pursue their own ’life projects’ although for Rawls it is also essential that all have equal opportunities to do that.

This abstraction from citizen’s social positions and particular conceptions of ’the good’ is seen as necessary to allow room for cultural diversity. That is, not to prioritize any ’life projects’ or ’life styles over others. This also means that the state should be neutral; basically concerned with citizens (equal) rights and justice. In radical, neo-liberal versions politics are even regarded as evil, as any sort of binding collective decisions reduces the autonomy of the individual. Thus, neo-liberalism in particular stresses freedom of choice (exit) rather than influence on decision making (voice). Liberal theories therefore tend to see society as a venture necessary for mutual advantage; something in which individuals basically participate in order to derive personal benefits. For this reason liberal theories of democracy and citizenship put little, if any, stress on civic identities and civic virtues. Thus, democracy in the liberal tradition, can be said to stress the importance of the individual and individual freedom guaranteed in particular by the forms of civil and political rights. These civil and political rights are elaborated more in detail by for example J.S.Mills (representative government, minority rights, etc.), and liberalism as a democratic doctrine can therefore be said to emphasise especially the relation between the individual and democratic procedures (e.i the character of political leadership). I have tried to illustrate this by relation I in fig.1 below.

Republicanism

The republican conception of democracy and citizenship is connected with political philosophers such as Aristotle, Machiavelli, Rosseau, Hegel, Tocqueville, and, in more recent times, Arendt and Habermas (Oldfield 1990). According to the republican ideal of citizenship, civic identities are broad; that is citizens are strongly oriented towards the political community. Participation in public affairs is a way of life – a part of being a ’good citizen’. And citizen virtues give emphasis to moral duties: a good citizen is a loyal, active participant in politics, oriented towards the common good of the political community. The republican tradition shares with the tradition of participatory democracy the idea of the educative function of political participation, and, in particular the belief that participation in associations and community affairs, enhances the resources for citizenship at the national level (Held 1987). The republican tradition is rooted in a perception of politics as ’problem solving’. That is, finding good and prudent decisions for the political community as a whole. Thus, the dialogue over public issues in the public sphere is a central aspect of democratic practice. As far as democracy and citizenship is concerned the republican tradition can therefore be said to stress the values of political participation, public deliberation, as well as political rights and duties. In my understanding the republican tradition is thus putting a special emphasis on the relation between community, or rather political community and political leadership; a leadership which will by nature be (more or less) ’prudent’ meant to guarantee citizens their civil and political rights, and to ’educate’ them to fulfill their political obligations. I have tried to illustrate this by relation II in fig.1 below.

Communitarianism

Modern communitarian thinking as it has been depicted by, for example, Mulhall and Swift (1992) is quite heterogeneous, and has formulated its point of view both in a direct criticism of modern liberalism, particularly as formulated by Rawls (see above), but also as a more encompassing criticism of Western morality and political culture. Particular attention has been paid to Rawls’ work by the likes of Sandel (1982) and Walzer (1983), while the more general criticism of Western morality and political culture can be found in the work of Taylor (1990) and MacIntyre (1981, 1988). Despite the heterogeneity these writers have enough in common to warrant the label communitarians. Firstly, like the republican tradition, the communitarians stress the importance of community for personal identity, interests and self realisation. However, the important community here is not the ’artificial’ political community of the republican tradition, but the ’real’ historically specific and culturally diverse communities of language and everyday practices. Also, the citizens of the communtarian tradition are not the ’good’ and politically educated citizens of the republican tradition. Rather, they are seen as ’political amateurs’ located in their culturally diverse communities, equipped with world views and needs heavily coloured by their location in such communties.

In contrast to the universalism of the liberal tradition, in terms of civil and political rights and justice, communitarians stress that such rights (as well as other ’social goods’) are the inevitable product of a particular political, economic and cultural development. This particularism probably has its clearest expression in Walzer (1983). Secondly, and also in contrast to the liberal tradition, the communitarians think it possible to formulate common conceptions of the ’good’. Politics, and the state, should be the vehicle for this common good, and in the case of conflicts between the common good and individual life-projects the common good must have priority. Thus, insofar as democracy and citizenship is concerned, the communitarian tradition can be said to stress the importance of (local, virtual?) community as the ’natural’ reference point for developing an understanding of oneself, and society, and thus also as the ’natural’ locus for the formulation of political demands and the implementation of policies.

As I read the communitarian tradition it is emphasising especially the relationship between the self and community. In political matters this self is however a ’layman’ or ’political amateur’ trying to solve matters of common concern primarily in his/hers own ’natural’ community. Being a ’layman’ in politics is seen as a virtue rather than a problem, as it means trying to define politics and policies in your own terms, and trying to distance yourself from the ’expert politics’ of for example a national, political leadership. This special relationship is illustrated by (roman) III in fig. 1 below.

Radical democratic theory

Besides the three traditional theories of democracy there are also a couple of modern ones’. However, the only one I see as containing something genuinely new is the theory on radical democracy. The radical democratic conception of democracy and citizenship is first and foremost associated with the works of Laclau and Mouffe (Laclau & Moufe 1985, Mouffe 1991, Mouffe (ed.) 1992). Their point of departure is the crises of class politics; the collapse of Soviet-style socialism, and with it the ideas of the emancipatory potential of the working class and the revolutionary party. For Laclau and Mouffe this has led to a re-evaluation of liberal democracy, and thus to the formulation of a radicalised version of the modern liberal democratic tradition. Central to this formulation is the nature of the political community and the shaping of political identity (citizenship). While defending liberal pluralism (i.e. the idea of individual rights and the principles of justice) both the liberal and communitarian conception of political community are criticised: the liberal conception for being instrumental and the communitarian for being pre-modern and potentially totalitarian. Instead, an idea of political community is constructed where individuals are seen as linked to each other through the recognition of the authority of the condition specifying their common concern; a moral bond expressed through ’a practise of civility’ (Mouffe 1991:76).

However, for Laclau and Mouffe it is clear that a political community will always be shaped by conflict and antagonism, and it will therefore always be the product of, or represent, a given hegemony. In their discourse-theoretical understanding of the construction of political identity and hegemony this means that they come to see citizenship as ’an articulating principle that affects the different subject positions of the social agent, while allowing for a plurality of specific allegiances, and for the respect of individual liberty’ (Mouffe 1991:79). Thus, citizenship is not a concept with a completely fixed content, but a concept open to different and competing interpretations. The interpretation that Laclau and Mouffe suggest is one that radicalises the principles of liberal democracy – liberty and equality for all – and uses these as a point of departure for creating a radical political ideal meant to appeal to all repressed groups in society (Laclau & Mouffe 1985). Thus, democracy and citizenship in the radical democratic tradition can be said to stress the importance of individual liberty. However, what makes it different from the liberal tradition is that it puts a stronger emphasis on equality: first that everyone should be guaranteed the same civil, political and social rights. And, second, that one should be ready to accept that others might use their rights very differently from oneself. Thus, accept of diversity is also an important value in this theory.

Now, it is not very difficult to demonstrate that by far the largest part of the litterature on ’teledemocracy’, ’electronic democracy’ or ’cyberdemocracy’ also use the concepts of ’individual’, ’community’, ’political leadership’ (political institutions) and ’political amateurism’ as their main points of reference in terms of the democratic elements of their theories:

Thus, the first generation of ’teledemocracts’ (Etzioni 1975, Toffler 1982, Naisbitt 1982), writing in the 1970’s and –80’s, under the impression of the explosive growth in cable-TV and the shift to digital phone connection, drew a picture of a future democracy where democratic dialogue and decisionmaking would take place using modern communication technology. This, they believed would renew democratic institutions (political leadership), and pave the road for more ’democratic’ forms of participation (the relationship between the individual and the political leadership).

Some of these writers turned into ’cyberdemocracts’ in the 1990’s. Among the ’cyberdemocrats’ there are roughly two main groups (Hagen 1997): a rightwing, libertarian group, and a ’softer’ leftwing group. The ’cyber-libertarians’ are oriented very much towards the individual and individual freedom. They are represented by organisations like the Electronic Frontier Foundation, which is fighting for (among other things) the right to free encryption. The ’softer’ leftwing group argue that for example the Internet can improve citizens’ social capital by offering them a type of community which would never have been possible without the Net. Thus, for the ’softer’ cyberdemocrats ’community’ (political, social) is important in much the same way as in the communitarian tradition described above. Therefore increased decentralisation and the construction of local community networks is a part of their strategy (see for example Rheingold 1993).

Also, and maybe less surprising the proponents of ’electronic democratisation’ relate themselves to our four democratic ’pillars’. These proponents, among which we find many bureaucrats and politicians, take their point of departure in the increased access to public information via the Net, the possibility of public electronic ’self service’, and of e-mail exchanges with politicians and civil servants, through which they see a possibility of revitalising and renewing the existing representative political system (Danish Ministry of Interior 1993). The focus is here on the wellknown relations between the individual as citizen/consumer, political institutions (leadership), and community.

Demonstrating that our four democratic ’pillars’ have been the main democratic points of reference in most of the discussions on ’cyberdemocracy’ is of course not in itself an argument for the continued relevance of these concepts. A few authors, like for example Poster (1997) has directly argued against this. Poster claims that to use the term ’democracy’ is to imprison our conception of cyberspace in outmoded modernity. Thus, for him the essential point about the ’second media age’ is that it promises to displace human connectivity from its institutional location, returning the very formation of culture into the people’s power. Therefore, the idea that democracy in the ’second media age’, i.e cyberdemocracy could be thought of or modelled a priori, through the use of for example our concepts, is selfdefeating.

Also, I have myself recently worked in another direction than the one sketched in the beginning of this paper (Hoff, Löfgren & Johansson 1999). Doing a democratic evaluation of an Internet concept launched by the Municipal Software House [Kommunedata] and the National Association of Municipalities [Kommunernes Landsforening] in connection with the local elections in Denmark in 1997 we took our point of departure in the tendential shift from ’government’ to ’governance’, or from the socalled ’Whitehall model’ of political steering towards more network –oriented, heterarchical, forms (Rhodes 1997, Kooiman 1993, Stoker 19xx). This shift, we argued, has also meant an undermining of traditional democratic norms of governing like for example equality in access to influence, public debate concerning political decisions, government’s capability to act, bureaucracy’s responsibility and accountability, etc., and the tendential appearence of a new set of norms. In describing these new norms we have been much inspired by the work of March & Olsen (1995) on ’democratic governance’, and we present these norms as: a) the development of democratic identities of citizens and other political actors, b) the development of capabilities for appropriate political action among citizens and other political actors, c) the development of accounts and accountability for all political institutions, and d) the development of an adaptive political system able to cope with changing demands and changing environments.

In our work we then proceed to use these norms as our criteria for evaluation of the mentioned Internet concept. For this purpose the norms seem to work quite well, and produce some interesting results. This could indicate that it would be possible to use the essence of these norms as a point of departure for constructing a theory of democracy for the information age. The essence would then be the concepts: a) (democratic) identity, b) capability (empowerment, e.i resources and competences), c) accountability, and d) (democratic) adaptability. These concepts could seem more apt than the ’traditional’ ones proposed above in catching essential aspects of ’democratic governance’ in the information age, as they seem geared to throw light on essential institutional dynamics (accountability, adaptability), and to be able to deal with the complex issue of individuality (identity, capability) in the information age, and maybe especially to handle the relations between the two arenas.

Here I will argue that even though March & Olsens framework seem to fit well with many current democratic problems, and even though Posters thinking seem congruent with ideas about the processual and developmental character of democracy in contemporary deliberative and participatory democratic theory, both miss a very important element in their theoretical frameworks. Thus, none of them are very precise about what exactly the ’democratic’ element in their framework is. I see this as a serious flaw, and will insist that we need to have a notion of what the essential ’democratic’ in democratic theory is. Otherwise the whole point in establishing a democratic theory is lost, and we might just as well call it a theory of political decisionmaking, of political identity formation or whatever. What I am saying is in other words that democratic theory must have a normative foundation; or that I don’t think it possible to construct a theory of democracy without such normative foundation.

Having made this claim what then should be the normative foundation for democracy in the information age? My personal bid is that it should be ’accept and acknowledgement of diversity’, and that this is the only rock-bottom norm that we really need (see also Bang & Dyrberg 1999). The centrality of this norm is dealt with above in the exposition of radical democratic theory, and also found in parts of Rawls thinking. Thus, it seems to be the central and necessary norm in connecting the individual to both community and political leadership. By saying this, I am also stressing the point that our democratic model (fig.1) is a relational model. This opens up the field for a host of questions which seem to be the relevant questions to ask democratic theory in the information age. For example: how does the individual relate to the forms of virtual, local or global, political and social communities we see forming in the information society? And what does this mean for the individual’s politcal identity? What are the possibilities for the individual, using ICT’s, to elect, influence, lobby, etc. the political leadership in whatever multi-centered, decentralised, local or supranational form this has? What kind of civic (democratic) identites are this political leadership/political institutions shaping and reproducing, and do ICT play a role in this? What is the role of ICT in shaping political communities, and what does this mean for political leadership? etc., etc.? These questions are probably not very different from the questions one would ask concerning the use of ICT in the ’political world’ departing from either March & Olsens or Posters framework. However the advantage of our framework is that it has, or can be given, a clear normative foundation, and that it operates with a clear set of relations which seems to be able to generate a lot of relevant questions when it comes to analyse the role of ICT in the ’political world’.

3. A constructivist perspective on information age democracy.

If we take a look at figure 1 again we see the term ’ICT’ in a circle in the middle of the figure. This is an attempt at illustrating the way in which the relation between ICT and democracy is perceived by the vast majority of persons dealing with these questions. For the utopists/dystopists the relationship is simple and straightforward. The new technologies are affecting individuals, communities and political leadership in a direct and causal way. The output is given – technology has already determined the configuration of future (athenian/orwellian) democracy. For the majority; the ’realists’, this is too crude a way of looking at things. Also other factors than ICT are seen as playing a role for the development of democracy, and the complex interplay between ICT and these other factors (political decisions, technocratic power, etc.; see for example Danziger et.al. 1981) must be analysed and understood. However, the ’realists’ still tend to see ICT as an exogeneous ’force’ in relation to democratic processes and institutions. For them it is therefore still meaningful to look for the ’impacts’ or ’effects’ of ICT on democracy. However, as ICT is only one piece in a complicated puzzle these ’impacts’ must be studied empirically, and the ’good’ and ’the ’bad’ impacts carefully mapped.

Such perceptions of the relationships between ICT’s and democracy are naive at best, damaging at worst, as they are blocking a deeper understanding of the processes and forces really at work when ICT’s becomes wowen into the fabric of the democratic wardrobe. As mentioned on p.2 a good starting point of a very different way of perceiving these relationships are new or changing political practices in which some measure of ICT play a role; the ’technologically-mediated-innovations-in-political-practices’ or TMIPP’s. However, such practices never take place in an ahistoric or acontextual vacuum, but are always practices within a 'duality of structure’ (cf. the Vehviläinen & Heiskanen paper). As such they express the constraining/enabling characteristics of the institutions within which they take place; their historical legacies, their rules, resources, etc. Practices are by definition set in motion by actors/agents, who could always ’have acted otherwise’. They therefore represent the unexpected, the new, even the revolutionary, but also of course, and indeed most of the time, the dull repetiton of routines within existing institutions (e.i. reproduction). Institutions on their side are, as Vehviläinen & Heiskanen point out (citing Giddens), just patterns of practices; without their recursive character institutions would vanish.

March & Olsen (1989) claim that practices also, in their reproductive mode, represent a certain ’logic of appropriateness’ (March & Olsen 1989), e.i. they express the dominant interpretation within a field/an institution/organisation on how a certain job or function should be carried out. This means two things of importance for us: a) that practices cannot just be seen as ’acts’ but also represent a certain understanding of oneself and others in a given organisation; an understanding which typically also reflect (or reflect upon) the expectations of actors outside the organisation, the ’demands’ of society, of the economy, of globalisation, etc., b) if a ’logic of appropriateness’ is a dominant interpretation there must also be other, non-dominant interpretations, or rather, that there is a certain ’interpretative flexibility’ in relation to the practices.

Starting to see practices as interpretations or meaning systems, and beginning to discuss how certain interpretations have become dominant, opens up a research agenda, which is very different from the one pursued by the ’realists’ mentioned above. The discursive character of practices is brought forward, and an adequate analysis of them becomes a question of understanding how they are (socially) constructed.

In understanding how practices are socially constructed, and in undertaking an actual deconstruction/construction analysis, the whole repertoire of ’deconstruction’ or ’social construction’ analysis is at our disposal. Elsewhere I have found it very helpful to work with the analytical framework of the ’social construction of technology’ (SCOT) approach (Bijker 1995, also Woolgar, etc. etc.) in relation to the question of ICT and democracy (Hoff, Horrocks & Tops (eds.) op.cit.). I am not going to repeat this analysis here, but just point to some concepts in the SCOT approach which seems useful for our purposes, and also to where I see some shortcomings as far as a democratic theory for the information age is concerned.

The concepts I find useful are the concepts of ’technological frames’, ’flexible interpretation’, ’stabilisation’ and ’closure’:

A ’technological frame’ represent a certain interpretation of problems and solution in relation to a certain technology. A ’technological frame’ is normally connected with a certain social group or certain actors, who actively share the same interpretation. In relation to our TMIPP’s we might also talk about ’democratic frames’, as specific and different interpretations of the democratic dimensions of TMIPP’s. The best would probably be to talk about ’demo-technological frames’, in order to stress how closely interwowen and in fact inseperable technological and democratic considerations are in the TMIPP’s.

The concept ’flexible interpretation’ represent the opposite side of the coin: the idea that a given technology can always be interpreted in different ways. Indeed, it might be possible to talk about as many technological artifacts as there are interpretations.

Having deconstructed a certain technological artifact the SCOT analysis then moves on to an analysis of the social construction of the artifact; or towards an understanding of how the degrees of freedom in interpretations are gradually reduced. This reduction takes place through the two processes that the SCOT theory calls ’stabilisation’ and ’closure’. ’Stabilisation’ can be determined by a rhetorical analysis, first employed in science studies by Latour & Woolgar (1979), showing how, in the construction of scientific facts, ’modalities’ are attached or withdrawn from statements about facts, thus connoting the degree of stabilisation of the fact. ’Closure’ is the process which marks a decrease in the flexibility which used to characterise a technological artifact. Thus, ’closure’ is the end of controversy about the interpretation of the meaning of the artifact, and closure involves power relations and power strugles.

Above I have already demonstrated how the concept of ’technological frames’ seem useful in relation to TMIPP’s. It thus clarifies, that there are indeed different interpretations of how certain technological artifacts should be used in relation to politics, and that these interpretations are typically tied in with certain groups or actors (certain ministries, task forces, grassroot movements, etc.). But as noted, TMIPP’s also contain interpretations of democracy, thus the idea about ’democratic’ or ’demo-technological frames’.

’Stabilisation’and ’closure’ also seem very useful in order to understand how the interpretative flexibility characterising certain TMIPP’s are gradually reduced, as a result of the strategies of different actors, power games, etc.. Indeed, in my interpretation one of the challenges of working in the area of democracy and ICT has been that ’interpretative flexibility’ in relation to the TMIPP’s has until now been high, resulting in many different and often contradicting visions, policy proposals and prophecies about the future. However, my hypothesis is that we are now in a period of ’closure’ in relation to ICT and democracy both in the sense that certain clusters of TMIPP's are becoming tied to certain models of democracy (see below), but also in the sense that certain models of democracy (notably the consumer model and the demo-elitist model) are moving even more center stage in the political process, while others are becoming (further) marginalised (notably the neo-republican and the cyberdemocracy model).

There are also some shortcomings in the SCOT theory of which I will only deal with the ones that are directly relevant for the topic discussed here. The first one is the lack of a concept of institution in the theory. Thus, the theory is not really able to account for why some outcomes in the stabilisation and closure proces are much more likely than others. What we miss is in others words an understanding of how institutions, through their ’historic’ interpretations, use of resources, etc., create ’path dependencies’ which makes some developments more likely than others. This shortcoming can be remedied, I argue elsewhere, by ’importing’ a concept of institution from the socalled ’new institutionalism’ of for example March & Olsen (1989). Related to our discussion this point is important, as it makes us aware that TMIPP’s, however ’novel’ innovations they are, are practices with a history; a history because TMIPP’s integrate (ideas about) democracy, (interpretations of ) the capabilities of certain ICT’s, and (interpretations of) experiences with political practices (or at least that is my postulate).

The second shortcoming in the SCOT theory is its relation to social change. What happens after ’closure’ is not clear; if the answer isn’t eternal reproduction. In my view a better explanation of this process, and one which is perfectly compatible with the concepts dealt with above, is found in the discourse theory of especially Laclau and Mouffe. Thus, they offer interesting ideas on why no identity or meaning can ever be fully constituted, and why ’closure’ is therefore always open to rearticulation (fodnote om ontologiske problemer).

As I will use Laclau and Mouffes discourse theory as the point of departure for my description of different and competing models of democracy for the information age, and for my understanding of how certain clusters of TMIPP’s become ’tied to’ certain democracy models, it is to their theory which I will now turn.

Laclau & Mouffe define discourse as: ’… a given collection of phenomena through which the social production of meaning takes place’, or as ’…. [a] structured totality resulting from an articulatory practice (Laclau & Mouffe 1985:105). An articulatory practice is defined as ’… any practice establishing a relation among elements such that their identity is modified as a result of that practice (ibid). Discourses are therefore created through a number of articulations which establish a relation of difference among elements in such a way as to mutually modify their identity. Thus, a discourse can be seen as a relational system of differences in which the single element only has an identity in terms of its relation to the other elements. An example could be a ’family discourse’ in which a ’father’ is only a father because their are also mothers, sons and daughters. Is, in other words, through the articulatory practice that the meaning and identity of the elements and the discourse itself is established.

As mentioned both in the ’democracy platform’ as well as in the introduction above, I think that a fruitful way to understand the relationship between ICT and democracy is to see it as different articulatory practices coming from a host of different ’voices’ (actors) who are trying to connect (clusters of ) TMIPP’s with ideas on, and strategies for democracy. Hereby different, and competing, structured totalities or discourses are produced, which I (we) have chosen to call ’emerging models of democracy for the information age’ (see fig.2) .

For Laclau & Mouffe, however, the concrete discourses existing at a certain time are only able to partially fix or freeze existing social elements or identies. The elements which are at a certain time not fixed in a discourse constitute ’… the field of discursivity’ (Laclau & Mouffe op.cit:111). These elements are also described as ’floating signifiers’ which, because of their surplus of meaning are seen as the raw material for the creation of new discourses or for the re-articulation of existing ones. Thus, discurses are never closed entities, and no identity can ever be fully constituted nor can any meaning be ultimately fixed. What the discourses do are to partially fix meaning. Discourses are, in other words, creating ’islands of order’ in a sea of chaos by stopping the flow of differences and constructing a centre. Laclau & Mouffe refer to such centres as ’nodal points’ or ’master signifiers’. The function of such point is to a tie number of elements together in a ’knot of meaning’. Thus, a nodal point becomes a kind of mirror in which the discourse can be seen in its totality. In the description of the models of democracy below I shall return to this question about the ’nodal point’.

Apart from offering this understanding of ’closure’, which I think is superior to the one offered by the SCOT approach, I also think discourse theory is superior when it comes to conceptualizing the question of power, and opening an research agenda for the study of power in relation to for example ICT and democracy. As this question is not the main theme for our endavour here, I will touch on this question only briefly.

For Laclau & Mouffe, as for Foucault, discourses are also expressions of power. As discourses are never closed entities there will always be different possibilities for action when different actors are trying to constitute or ’close’ a certain discourse. These types of decisions are expressions of power, which always include the repression of other possible decisions (Laclau 1990:34). Thus, power can be seen as the setting or construction of (a set of) differences which constitute a ’field’ in which certain actors can pursue their goals. Power is the creation of a demarcation line between a discourse and its surroundings, and the way in which this demarcation is made self-evident in the discourse. This is the reason why Foucault comes to see discourses as the simultaneous production of knowledge, truth and power. In my view this means that, seeing ’the models of democracy’ below as discourses, we also necessarily have to confront the question of power. Thus, in the competition and conflict between the different elements and models we cannot avoid asking ourselves the questions of: by whom, for whom, and with what (anticipated and unanticipated) concequences?

4. Emerging models of democracy for the information age

In this paragraph the four emerging models of democracy will be spelled out in more detail. I will stress that the models must be seen as hypothetical and provisional, as a means of establishing an iterative process through which the process of constructing an analytic framework can both inform and be informed by the experience of further case studies. As mentioned on p.2 the starting point for the models lie in both a set of texts and a number fo case studies. The texts are: the contemporary litterature on citizenship and democracy, which is mainly grounded in political thought, and the contemporary litterature on technology-mediated communications, must of which is grounded in the fast-growing dicipline of media studies. The case studies are a number of case studies carried out in the UK, the Netherlands and Denmark as a part of the socalled GADIA-project. They represent a number of different TMIPP’s, of which some are comparable across countries.

As will be clear when I describe the models, and if one looks at table 1 (below), each of the first three models (from the left) of information age democracy may be regarded as an attempt to restore a particularely valued dimension of the constitutionalist tradition. Only on model; the cyberdemocracy model can be said to have some features which point towards genuinely new ways of organising relations between the individual, community and political leadership. The models are arranged along a continuum depicting the varying degrees to which they tend towards a narrow definition of democracy, one in which ’the parliamentary chain of steering (the ’constitutional’ or Whitehall model) still plays a critical part, or point to new kinds of democratic practice.

The consumer democracy model

Consumer democracy shares with demo-elitism an uncritical acceptance of many well-established features of constitutional democracy, including the role of parliamentary institutions, elections and parties. At the same time, however, there is a clear understanding that for most electors the public affairs of the modern state occupy a far lower place than the private concerns of family, home and work. These models reflect the work of Schumpeter, public choice theorists (Olson 1965, Downs 1967) and other ’realistic’ accounts of modern democracy in supposing that the only important political action of most people is to cast a vote. These two models also place considerable emphasis on the growth of the bureaucratic state, recognising the increasing penetration of the administrative and regulatory functions of governance into the everyday lifes of ordinary people. In the consumer democracy model, therefore, the main priority is to give individuals more information, more choice, and thus more power in dealing with public bureaucracy.

The consumer democracy model takes for granted, but therefore largly discounts the input side of the parliamentary chain of steering. Instead, it focuses on the reconstruction of the output side, on the interface between services and their consumers. In other words consumer democracy aims not so much to challenge as to bypass the tired institutions of representative democracy. It seeks, in effect, to recentre democracy from the political nexus, a nexus formed around parliamentary and electoral processes, onto the consumer nexus, a nexus formed largely around the consumption of public services. Consumer democracy can therefore be regarded as the culmination of a strategy for legitimating and controlling the growth of the administrative state, by establishing greater popular control over the consumption nexus.

The consumer democracy model epitomises the old saw that ’information is power’. It seeks to redress the balance of power between producer and consumer by promoting such principles as choice, access, redress and voice. The effectiveness of these principles depend in turn, however, on a meta-principle: access to more and better information. Consumerism therefore establishes as a fundamental procedural norm the principle that there must be effective flows of information to consumers, enabling service users to become more competent, more discriminating, more powerful clients of government. This model draws attention, therefore, to important questions of public policy: for example, what information can governments make freely available? How does this impact on policies relating to the commercial exploitation of information? How can ICT enable consumers to become more active, more selective, more purposeful users of public service information?

The demo-elitist model

Like the consumer model the demo-elitist model assumes that public opinion is less active than the constitutional model suggests, and that its prime function is to constrain and legitimate government rather than actively to steer policy. It suggests that popular legitimation depends less on widespread agreement about procedural norms and more on policy outcomes than the liberal project acknowledges. It proposes, therefore, that succesful welfare states are those that manage both to satisfy a sufficient range of wants, and to do so in ways that mediate most acceptably between competing claims.

Demo-elitism gives a prominent place to the role of experts and external interests in meeting the demands imposed by the search for popular legitimation. The policy communities and issue networks that form around governments are not, therefore, to be easily dismissed as an excrescence on the body politic, but accepted as a normal, routine part of governing. It follows that they should also be fully integrated into accounts of contemporary democratic practice.

The extended networks of contemporary governance systems (Kickert et al. 1997) are critical to the processes of building consensus and co-operation between the interdependent interests that have become entagled in the modern state. In this model, then, the processes of representative democracy, on the one hand, and the institutions of governace, on the other, share responsibility for the quality of late modern democracy. The former serve to legitimate the state, and act as a means of renewing political elites, while the extensive networks of functional constituencies provide the means of renewing the co-operation and accomodations which enable this legitimation to be sustained.

Despite this democratic rationale for demo-elitism, it cannot be denied, too, that this model also depicts a state of affairs which is prone to degeneration from a democratic point of view. Demo-elitism points equally well to an oligarchical form of government, in which a narrow range of sel-perpetuting elites become steadily more detached from controlby the constituencies they claim to represent. In this more pessimistic scenario, power over public policy comes to be concentrated in closed networks, just as techniques for managing the representative process is monopolised by party and media bosses. Wheras consumer democracy may be regarded as an attempt to by-pass the problems of contemporary democratic practise, demo-elitism aknowledges the danger that the problems of mass politics will be intensified by ICT.

Demo-elitism also captures, however, many reforms on the active agendas of contemporary governments. In the first place the demo-elitist model is designed to acknowledge a range of proposals for enhancing the democratic quality of electoral politics, thus strengthening the representative side of the demo-elitist balance. These proposals range from the British Government’s proposals for allowing voting in unconventional locations – such as supermarkets and bus stations – to proposals for putting party manifestos, candidates’ voting records, and position statements on the World Wide Web.

What is distinctive about the demo-elitist model is that it also emphasises the importance of reform on the elitist side of the demo-elitist balance.In part, our thinking here echoes that of Etzioni-Halevy (1993) in proposing that contemporary democratic theory s hould focus on such questions as the responsiveness, accountability and openness of elites, and the extent to which elite structures are compatible with egalitarian values. In this respect, demo-elitism point up such issues as the relative autonomy of elites (both from each other and the state); the circulation and interchange of elites , sub-elites and the public; the inclusivity of corporatist structures; and the transparency of corporatis processes. In so doing it raise several issues relevant to our concerns. First, the emphasis on openness and transparency brings into play such policy issues as freedom of information and the regulation of the press. Second, it questions the extent to which the economic interests and social agencies which become enmeshed in neo-corporatist structures are themselves subject to democratic control and renewal. In the work of Etzioni-Halevy (1993) and Putnam (1993, 1995), for example, the efficacy of modern states in promting a consensual, inclusive, effective style of governing is shown to depend on such factors as the inclusivity of neo-corporatist structures, the relative autonomy of economic and social elites, and the extent to which those elites are e,mbedded in dense networks in civil society. As a model of information-age democracy, demo-elitism thus draws attention to the contribution whcih new information and communication technologies might make in building and invigorating such networks. Indeed, the complex, diffuse networks of modern governance could be controlled not simply by strengthening accountability through the electoral chain of command or by securing greater democracy within specific neo-corporatist structures. We would wish to explore, too, the possibility of countering the pervasive power of governance networks through the equally pervasive and flexible alliances between the clients and stakeholders in these networks.

As a model of information-age democracy, demo-elitism therefore focuses mainly on opening up and strengthening vertical rather than horizontal flows of information and communication: between voters and their representatives, between representative institutions and government; between government and external elites; and between elites and their stakeholders. It is preoccupied mainly with macro- and meso-level processes.

The neo-republican model

The cyber-democracy model

5. Final remarks

In this paper I have tried to develop a framework for analysis of democracy in the information age. This framework is explicitely synthezising and constructivist. However, its methodology sofar has basically been deductive, taking its point of departure in fundamental categories and relations in democratic theory (individual, community, leadership and laymanship), and in socalled technologically-mediated-innovations-in-political-practices (TMIPP’s). It is then postulated that there are different articulatory practices going on which tries to connect TMIPP’s with ideas on, and strategies for democracy. Hereby different and competing discourses are produced, which I have choosed to call ’emerging models of democracy for the information age’.

As mentioned already, these models should be seen as hypothetical and provisional, basically meant to establish an iterative process through which this analytical framework can both inform and be informed by the experience of further case studies. The models are therefore neither to be seen as ideal types in a weberian sense nor are they testable hypotheses in a more rigerous understanding of the term. Rather, they should be seen as heuristic devices, offering inroads to an analysis of democracy in the information age, and encouraging further case studies along these lines. Also, they serve to clarify the position of the researcher, who, with the analytical arsenal offered here, should be in a position to ask the critical questions: by whom, for whom and with what (anticipated and unanticipated) consequences?

Table 1 Emerging models of democracy for the information age

Institute of Political Science, University of Copenhagen,

Rosenborggade 15, DK-1130 Copenhagen K., Denmark

Phone: + 45 35 32 33 97

Fax: +45 35 32 33 99