POLICING DEMOCRACY: COMMUNICATION FREEDOM IN THE AGE OF INTERNET kees brants Introduction At the end of 1995, a Munich prosecutor visited the German offices of the American on-line service CompuServe to complain about certain sexually explicit material on Internet, as this was illegal under German criminal law.1 Although CompuServe as a self-contained electronic community is only an access provider and thus not, they claim, responsible for the origination or nature of content on the Internet, the prosecutor wanted to block access to more than 200 of the about 10,000 newsgroups. Most of the two hundred were identifiable by the prefix "@sex" and ranged from pedophilia with boys and bestiality to material not illegal for (German) adults, but of such an explicit nature that it is for (German) minors. CompuServe complied, thereby not only restricting access to the German users, but to all the 500,000 members in Europe and the four million in the US and the rest of the world. Although some on-line services have rules governing the language and materials used on their own services, the companies have no control over Internet to which they all offer access. CompuServe is working on a technology enabling restriction of access according to geography. Others, such as Microsoft and America Online, offer parents screening software to limit children's access to sexually explicit material. In an attempt to crack down on pornography on its own service, America Online (which started a European partnership with Bertelsmann and Deutsche Telekom) not so long ago restricted access to users who included the word "breast" in their user profile. The action, however, inadvertently affected women who had identified themselves as hav- Kees Brants is lecturer at the Department of Communication, University of Amsterdam, Oude Hoogstraat 24 1012 CE Amsterdam. t£> cri cTt ro "5 > r-in ing an interest in breast cancer, and the service later dropped its ban (San Francisco Chronicle, Dec. 12,1995). These measures of self regulation have not been enough for the US House and Senate, which on February 1 passed a Communications Decency Act. This amendment to the new Telecommunications Bill requires all content on the Internet to be acceptable for distribution to young children. Distribution of pornographic material ("in terms patently offensive as measured by contemporary community standards ... regardless of whether the user of such service placed the call or initiated the communication;" sec. 502) is prohibited with a maximum penalty of two years. Although the enforcement of the Act is expected to be problematic, civil liberties groups in the US see it as censorship and a violation of the First Amendment. The recent examples from Munich and Washington show — be it only from the moral panicking side of powerholders — two important and related issues of the new information and communication technology (ICT): the pros and cons of its anarchistic nature and emancipating potential on the one hand and the far reaching ways and means of (self-)regulatory control and other restrictions on the other. New technologies of communication have always been greeted with enthusiasm and looked at suspiciously at the same time, but there now seem to be at least three stands regarding the workings of Internet. The normatives see a world wide web of filth, a global highway lined with hookers, pedophiles and (other) dirty old men. Euphorics or Utopians, on the other hand, will say that the ultimate result of the ICT will be an abundance of information, intellectual pluralism, direct democracy and personalised control over communication. Where these optimists see a new Athens with more and more people participating on the electronic agora, critics of such "video-utopians" like to refer to an Orwellian state. A new reality where Big Brother (hiding himself in an ever growing information gap) not only registers what people think, but provides legitimating ideology for global capitalist enterprise. The new electronic media will extend the power and reach of large, multinational corporations and generally accelerate forces of tansnationalisation. Interactivity in general and the newsgroups in particular might potentially revive a democracy troubled by political cynicism and voter drop out, but in practice, the cost of it will strengthen inequality and reaffirm political elitism. All rhetoric aside, the Internet and new interactive media in principle can, and do provide an unregulated platform for the exchange of ideas, but at the same time, a part of its contents seem to trigger kinds of regulation which we thought had been passe in western democracies. Will ICT and its far reaching possibilities for freedom of expression be the panacea for, or the nightmare of, democracy? What are the needs, possibilities and limits of regulating it? Communication Freedom and Regulation Regulation of communication in liberal democracies is mainly based on traditional principles of freedom: freedom from state interference, that is to say, no government action to prohibit a publication before it has taken place (no censorship or, in the American legal terminology, no "prior restraint") and, the other side of the coin, the right or freedom to express oneself. Over the years, in most countries in Western Europe the "freedom to" principle has come to be safeguarded by the state as well: governments do have an obligation to enable freedom of expression and a diversity of ideas. The idea of diversity or plurality—as both the result of freedom of expression and a prerequisite for democratic discourse — refers, firstly, to the number and variety of providers of information. Thus state support for ailing (privately owned) newspapers in order to guarantee a certain level of pluralism in the production of information has become a common feature of many a Keynesian welfare state. A concern for pluriformity might also lead governments to stimulate new entrants and thus competition in the telecommunications sector. Another aspect is the diversity of information contents. It is defined as an obligation of many European public broadcasting corporations which are expected to provide a "reasonable ratio of information, culture, entertainment and education," as stated in the 1967 Broadcasting Law in the Netherlands. A third and central aspect of the freedom of expression and the plurality of information is the right of access to channels of communication, enabling reflection or representation of the prevailing differences of opinion and culture in a given society. The fundamental conditions for effective access are freedom and opportunity to speak out, and thus a sufficient number of independent and different channels, plus autonomy over media access opportunities (McQuail 1992,145). Access to the network for all on equal terms, and a guarantee of continuity of the "universal service" performed by public telecommunications operators are typical examples. As all these aspects of guaranteeing diversity imply some form of government action — be it prohibiting, prescribing or supporting — based on concrete decisions and an abstract notion of the "general interest," there is by definition an inbuilt tension in the traditional idea of communication freedom. In order to improve the "freedom to," one might well have to be a bit lenient with the "freedom from." And there are other, more legal, limitations to the principle of freedom. To put it bluntly, freedom of the one person or group usually meets its boundary where it interferes with the freedom of the other. Communication freedom is thus legally limited in most countries by privacy laws, property rights (copyright), libel laws, a perceived threat to state security and public order, and some norms of taste and decency. Although we are facing the beginning of a supranational regulation (e.g., the Right to Reply in the EC-Directive Television Without Frontiers), limitations on communication freedom usually differ from one country to another. Regarding taste and decency, the cultural differences between countries are most clear. What is tolerated or even appreciated in one country might be a sin in another. Especially in the area of "explicit material," countries like the Netherlands and Denmark have a more liberal stand than, for example, the UK or the US.2 Information Services To understand the complexity of different areas and kinds of the existing communication regulations, it is useful to distinguish between different types of information services. According to Bordewijk and Van Kaam (1982), communication patterns may be characterised in terms of two dimensions, i.e., central versus individual control of information storage and central versus individual control of timetable of distribution and choice of subject. The combination of these two dimensions results in four basic communication patterns: allocution, consultation, registration and conversation (see Table 1). Table 1: Model of Communication Patterns Central information storage Individual information storage Centre chooses subject and timetable allocution registration Individual chooses subject and timetable consultation conversation Source: Bordewijk and Van Kaam 1982. In allocution3 — one-to-many communication with usually little personal feedback opportunity — information is distributed from a centre that decides the timetable of communication simultaneously to many peripheral receivers. The most common example is broadcasting, where programmes are received by large numbers of scattered individuals at the same time. Consultation refers to a many-to-one-to-many situation in which individuals at the periphery look for information at a central storage of information, like a library, database or teletext. Registration is in effect, the consultation pattern in reverse, in that a centre searches for information from a participant at the periphery; with the centre usually determining the content and occurrence of the communication traffic and often without the awareness of the individual. This many-to-one communication applies to all systems of surveillance and relates to a variety of services: from automatic recording at a central exchange of telephone calls to audience research (the "people-meter") and for purposes of charging consumers. In cases of conversation, individuals interact directly with each other, bypassing a centre or intermediary and choosing their partners, time, and topic of communication by themselves. It is a one-to-one pattern of communication, usually with symmetry and balance between the parties, like in an exchange of personal letters or the use of electronic mail. Another relevant dimension in comparing these different information services or communication patterns is that between the public and private domains, areas where traditionally different fields of regulation apply: privacy and property rights more in the private domain, public order and taste and decency more in the public domain. Mass media contents, for instance, which are widely available to all without restriction are most public by their nature, while the registration and storage of data concerning people or organisations are least public. Conversation services (like the telephone) are more likely than consultation services to belong to the private sphere, although the ostentatious use of portable phones in public places seems to point to a cultural shift. — The allocution pattern, more than the other three, is associated with the "old me- «£> dia" of mass communication. Not that old necessarily means dying in due course — as some say is the case with newspapers. The allocutive pattern remains important and may still grow in absolute amount of traffic. Especially consultation and conversation however, have been able to grow because of new telematics — combining telecommunications, informatics and digitalisation — and the diffusion of video and sound recording equipment. The steady increase of subscription and pay-per-view channels, as part of the explosion of cable and satellite television, will further contribute to a relative decline in allocution. At the same time, computerisation and extended telecommunication connections have favoured the growth of registration potential. Three Regulatory Domains New information and communication technologies are said to demonstrate three basic trends: the redistribution of information traffic from allocutary to conversational and consultative patterns; the shift of emphasis from the public to the private domain; and the convergence, overlap and interconnection of communication functions and technologies (Van Cuilenburg and Slaa 1993). In spite of these trends, many national communication systems are still regulated according to their basic technology and often for historic reasons (Pool 1983). Thus, print media are almost entirely free from regulation or control and often protected and privileged, as they are (were) judged to be the prime vehicle for the expression of opinions and the control of decision makers in a democracy. Even with the increase of mergers, take-overs and cross-media ownership, and the threat to diversity it might have, many governments still hesitate to set rules for, and limits to, press concentration. Prohibitive interference in an imperfect market is still seen as a violation of the freedom to publish, a freedom generally regarded as co-extensive with freedom of speech. The principle of freedom of communication got a somewhat different interpretation when, during the nineteenth century, telegraph and telephone, and later telecommunications became popular. To solve market-imperfections, these so-called "common carrier" media were often regulated as to their infrastructure, ownership and pricing, while there was no regulation of contents. They were considered a natural monopoly which should provide universal service. All over Europe, the regulatory regime of cable networks and telecommunication services has recently been shifting towards a press model, with liberalisation of the market and privatisation of (some of) the services. Broadcasting, historically due to the physical scarcity of frequencies, has always been much more subject to regulation. In public broadcasting systems, the scarcity usually meant a semi-monopoly situation, where governments or government controlled bodies decided on who was allowed the entrance, and on what grounds. At the same time, a central collection of license fees had to be organised in order to finance programming. Together with the supposedly intrusive and at the same time distractive nature of the medium television, this prompted governments to set all kinds of norms and standards for entry and performance. Liberalisation (breaking the monopoly of the public broadcasters and allowing commercial channels) and privatisation (e.g., TF1 in France) represent a de-regulatory trend in terms of the structure of the broadcasting market; the content, however, remains an area where rules are eminent. Although there is a hesitant trend towards convergence of these three policy domains, new information and communication technologies go beyond the separate boundaries and, even more so, beyond those of the separate countries. This poses special problems. Does the common carrier element of much ICT ask for regulation of infrastructure and access? Should the press or broadcasting model be applied to the content? Is ICT the new battleground for deciding the "openness" of a society? All questions relate, in one way or the other, to the problem of communication freedom, especially where ICT is applied in the field of political communication, participation and democracy. ICT in Politics: Medicine for a Mid-life Crisis One should not forget that many ICT, and the Internet in particular, were first implemented in the area of defence and surveillance. At the moment, they are primarily treasured for their economic potential, and regulation of access and content is hardly a part of the industrial logic. The relative shift in information traffic from allocution to consultation and conversation is also largely a consequence of its economic use. Most large enterprises in the US and more than half in the Netherlands have a site on the Internet. In taking over the "backbone" of the Internet from the National Science Foundation in 1995, the US telecommunication giants Sprint, American and Pacific Bell were more interested in the commercial than possible democratic opportunities. American and European publishers and newspapers get "on line" not because this is an extra medium to inform, but because it is a potential market and a means to avoid distribution and paper costs. They are also afraid that others might take off with their classified ads and thus with an important part of their income.4 The technology-push policy of the Dutch government, spending ECU 30 million on infrastructure and the development of consumer services, in order to get one million Dutch on the Internet by 1997, is particularly inspired by the social (employment) and economic potential of telematics development in Europe, and the aim to be in its forefront. For most political decision makers, the applications in political communication have been a spin off from the industrial policy. Not so for the grass roots movements which have hailed the new medium as a means of enriching democracy. Its flat, open and unstructured organisation, its non-hierarchical and potentially bottom-up structure and the possibility to combine allocution, consultation and conversation through interaction is seen as bringing back direct democracy. A reference is often made to the Greek city states where all citizens actively participated in political debate and opinion forming, and often were even obliged to fulfil political tasks in rotation. Many centuries later Rousseau, reminding us in passing that participation in Athens was only by male citizens and made possible because they had slaves and thus a lot of spare time, rephrased the idea of direct democracy in that volonte général could only be achieved by all sharing in its formation. What is often forgotten in references to the Contrat Social is that, according to the French philosopher, direct democracy would prosper best in small communities where citizens are relatively independent and the difference between rich and poor is small. Rousseau's ideas about democracy paved the way for universal suffrage in the 19th and 20th century. But where he saw an obligation for the people to actively participate in politics, the earliest election studies in the US and the UK during the Second World War and shortly after it demonstrated that even voting, the minimal participation in the political process, was not common among all and interest in the po- liticai affairs was an exception. While some saw non-voting as a serious problem in a democracy, modern democratic theorists such as Joseph Schumpeter labelled notions like the "will of the people" and the "general will" a myth. He criticised the classical idea that the final goal of democracy is the emancipation of the dignity and happiness of the individual through significant participation in decision making. Schumpeter proposed instead a democratic "method" (which in fact legitimised non-participation), an "institutional arrangement for arriving at political decisions in which individuals acquire the power to decide by means of a competitive struggle for the people's vote" (1943,250). Although representative democracy is often seen as the second best solution, democracy as a method is the political form liberal democracies have taken. At the same time, the sometimes nostalgic and romantic (and selective, for that matter) references to Greece and Rousseau come from a realisation that at the fin de siècle, parliamentary democracy is suffering from a sort of mid-life crisis, may be even a terminal decease. Its symptoms being: a diminishing importance of political ideologies as the "cement" of society and growing fragmentation and individualisation of its population; a decline of voter turn out and extensive political cynicism towards policies, politics and politicians; citizens turning their backs on the present day bearers of the volonté général: political parties, in favour of single-issue social movements. All this has led to a call for a return to direct democracy as a favoured cure and ICT, readily available with its communicative possibilities, as its vehicle. It is here that political parties and grass roots movements have found each other in ideas about the role of interactive communication in the political process; bien etonnés de se trouver ensemble, because the former see chances for reviving contact with voters while the latter see it as a means to circumvent traditional politics. Improving Democracy While Internet is renowned for its global reach and lack of frontiers, it is worth noting that the more interesting, democracy related applications and experiments with ICT are taking place at the local level. Several projects and experiments show that the Netherlands, with a cable density of around 85 per cent, a PC in every four out of ten households and in 1995 around 300 thousand people (two per cent) using the Internet, is a European forerunner in interactive political uses of ICT.5 The so-called City Talks (Stadsgesprekken) in Amsterdam were a prudent attempt to combine allocution with conversation and consultation. Based on a limited form of two-way communication, the local municipality initiated from 1989 a series of live discussion programmes on "hot" issues on the local television channel "Salto." Although television is by definition an allocutory, one-way medium, this public access channel has for years facilitated minority groups to voice ideas and broadcast cultural experiments. In the "Talks" politicians and representatives of non-governmental organisations discussed a chosen topic (drugs, crime, housing), while the home audience could get extra background information via teletext and react to the opinions voiced (through telephone and later computers placed in public places like libraries and the town hall). These reactions were then included in the debate, while the viewers could also vote for certain statements or policy options. The philosophy was "information in reverse:" citizens telling politicians what they think of certain issues instead of politicians gaining support for decisions by explain- ing and persuading. It was a limited form of input-democracy, with wants and desires articulated and formulated bottom-up. In spite of its open structure, or may be because of it, none of the actors were really happy with the space they got to make their say. The non-politicians were also dissatisfied with the lack of follow-up in terms of political consequences. In 1995 the programme was moved to the local semi-commer-cial channel AT5 and the interactive element taken out to give it more the format of a talk show. Consultation services — especially in the area of consumer use — are probably the most publicised where it comes to ICT applications. Electronic banking, teleshopping, electronic bulletin boards, etc. have not yet become common place but are beginning to be part of everyday discourse. D-bases of newspaper archives, of laws, (local) government reports, parliamentary questions and minutes, of libraries (from the small and local to the extensive US Library of Congress), of demographic statistical data, of all kinds of information relevant for opinion forming have become accessible for everyone, and in some cases only for a minimal price. Political parties have begun bulletin boards, where voters can retrieve information (not that there is much need for that, it seems). All parties in the Netherlands have e-mail addresses, some have set up a mailing list to which one can subscribe while others have opened up home pages on the World Wide Web. The way and speed in which data can be retrieved, stored, analysed, manipulated and sent has certainly affected the registration and surveillance potential of civil servants and politicians. Tele-polling and other computer aided survey techniques have made it possible to use knowledge about public opinion in the policy making process. In an experiment in Rotterdam with two-way cable citizens were electronically questioned about wants and grievances and plans are now drawn up to organise instant referenda and electronic discussions (Depla 1995,50). It is this development which enables advanced marketing techniques in election campaigns which are often seen as another (negative) example of the "Americanisation" of electoral politics in Europe. According to Newman (1994), Clinton based his programme, choice of and stand on issues, and important political statements on the results of computer aided opinion polls. In Amsterdam the Green Left party "manipulated" a group of 5,600 Amsterdammers who had filed a protest against the building of a tennis court in a city park. The protesters had been registered and stored in a database of the municipality and Green Left — which supported the tennis park — explained its stand to the protesters through direct mail. Socialising Democracy The majority of examples to be found about the use of ICT in relation to democracy is usually of an instrumental kind: applications aim at improving the existing representative democracy. In general they are initiated from the (political) top and mostly based on giving more and better (user-friendly) access to more information. Only very few of the ICT applications go for a more communicative use, socialising democracy, improving the "responsiveness of political representatives, active citizenship, communication and public debate instead of information, consultation and registration, bottom-up instead of top-down, responsibility shared by decision makers and the public" (Brants at al 1996,238; cf. Depla 1995). Table 2: ICT Applications Aim Communication Patterns Allocution Consultation Conversation Registration Instrumental Communicative Public access tv Bulletin boards Interactive tv-debates Eledrcmic polling Digital City With socialisation of democracy, allocution, conversation and consultation are interactively combined (see Table 2). Such is the case with most of the now seven existing so-called Digital Cities in the Netherlands. The first opened in Amsterdam, in January 1994, organised by a group of citizens, actively involved in cultural, (nonparty) political and the hacker's movement. Essentially, this Digital City is a site on the Internet, representing Amsterdam as a virtual city and a virtual community with, for example, a library, coffee bar (with virtual soft drugs?), post office, museum. The idea is that citizens build there own city, there are streets and squares sharing a certain (cultural, political, etc.) "sphere," people can "rent" flats and "construct" houses. The City's philosophy is bottom-up, non-hierarchical and voluntary, and the interface is user-friendly. Discussion groups have started in virtual public spaces on a variety of topics, of which the traditional political issues form however a minority. The emphasis is on horizontal debate between citizens, without intermediaries, and between citizens and politicians and it thus functions as a community network, articulating wants and grievances. Digital City's town hall provides access to the administrative information system, while political parties publish their programmes, policy documents and political stands. For access one needs a PC and a modem and for full use one has to log in via an Internet provider. Public terminals have been placed in libraries, museums, the town hall and homes for the elderly. Although access is free (except for telephone cost), the users form a relatively elite group. Digital City claims a population of some 35 000 and around 5,000 consultations a day; the majority of the citizens is young (58 per cent under 30), well educated, employed or studying and male (85 per cent). Politics in ICT: The Other Site of Teledemocracy The issues coming up in discussions about the instrumental or socialising applications of ICT partly overlap with the recent hype around and prohibitive regulations of Internet. In the former the undertone is concern for participation and political consequences, while in the latter the worry is about content and sometimes clouded by a moral panic. But the discussion is also clouded by the fact that the "euphorics" are allergic to the word "regulation," while those in favour refuse to acknowledge that ICT might be a "technology of freedom." The adherents of the two positions usually talk at different levels and thus fail to see that there is more between heaven and earth than dirty pictures or the interfering state. At the heart is a question about the conditions for and limitations of communication freedom. Broken down, the issues at stake in the role of ICT in politics seem to be about (the relation between) access, privacy, content, political representation and control of politics: is one to worry about the discrepancy between optimal access and limited participation; does the unregulated structure revive the "old" contradiction between communication freedom and privacy; should norms be enforced for content and who is responsible; and what will be the effect on democratic control with the changing role or even disappearance of the traditional actors in political communication: political parties and journalists? Access Access refers, in the first place, to the openness of a network or technology to both information providers and information users. Where in the economic applications there is a growing number of providers (and a possible threat of future concentration in the hands of a few large ones), with the political applications ideally the difference between providers and users disappears. With the non-hierarchical, horizontal organisation structure and the interactive possibilities senders become receivers and vice versa. Secondly, access is operationalised as the physical, affordable, user-friendly and reliable opportunity to make use of a channel or network. At the moment access to ICT is usually restricted to either terminals placed in public places or offices which are "on line." It is not surprising that most users enter the Digital City during office hours. If ICT is considered crucial in the realisation of socio-political and economic goals in society — and e.g. the Dutch government's subsidy for Internet-use points to that underlying discourse — then it should be considered as a common carrier type "public service," guaranteed by the state and apparent in universal geographical and affordable access, and universal quality and tariffs (Garnham 1989). That does not necessarily mean universal use, let alone maximum participation. At the moment ICT is very much for an elite and applications in political communication point to a use by those already able to find their way into the decision making arena. The question is whether a minimum percentage and representativeness is necessary in order to talk about socialisation of democracy through ICT, or whether different levels of participation only point to a more flexible citizenship. Certainly, the situation at the moment is far from direct democracy, but discussions about the lack of representativeness and low participation tend to turn opportunity into obligation: instead of imposing conditions for participation one imposes participation itself. There is also a problematic assumption in this Athenian style obligation to participate, namely that people are political animals. Apart from consulting government services, most people are seldom involved in politics and rarely searching purposefully and actively for political information. The interactive possibilities of ICT, the sensitivity to new social issues that can derive from the horizontal structure and the issue-orientation of newsgroups might contribute to more information and involvement in debate, but it runs the risk of a self-defeating prophecy and thus turning the high hopes around ICT into a self-fulfilling disappointment. The danger is that the potential for participation in debate is enlarged, but that it takes place in a public 15 vacuum: active participation grows but passive participation (which is the character- istic of allocutory political communication) diminishes (see Van Praag 1996). Many newsgroups on the Internet or in the Digital City drown in the cacophony of their own communications without many people listening or hearing. Privacy The freedom of the one might well be the insult of the other, or touching on his freedom. A major boundary of communication freedom is set by privacy (laws), a boundary which is not static but constantly under pressure. Journalists test the privacy of public figures in the tabloid press or through muckraking in investigative reporting. Police and other repressive state apparatuses rock the balance of proportionality between their tracing mandate and their privacy protection responsibility. The surveillance potential of the new technologies in Europe has probably found its apotheosis in the Schengen Information System (SIS). European integration and the opening up of frontiers has led to synchronisation of the respective computer systems of immigration, security and police services in the different EU-countries to control the movements of non-EU-members. Knowledge from computer aided polls, discussion groups or interactive programmes like the Amsterdam City Talks does give invaluable information to decision makers about people and their wishes. It might well influence the political agenda from below and politicians have to take stock of these opinions. The data from mass surveys and the deduction of profiles from the many d-bases available on demographic data and people's consumer behaviour can, on the other hand, be combined to target specific groups in society with specific information and thus turning "co-production" into allocution, debate into persuasion again. The tension between communication freedom and privacy is not always recognised or admitted here. Complaints about the misuse of privacy in the case of the Amsterdam Green Left party and their direct mail campaign to the tennis park protesters were turned down by the Amsterdam municipality. Targeted communication is becoming more and more popular. According to Selnow (1994) all presidential candidates in the 1992 US-election applied it and with the advancement of technology in the past four years, the 1996 election campaign will also see a combination of databases, e-mail, direct mail and other forms of targeted campaigns with restricted information. Content Is it not a paradox that in telecommunications, the recently most liberalised and privatised domain, prior restraint is being reintroduced by netprovider CompuServe with the blocking of access to the Internet of some 200 newsgroups? The pornographic and in some cases racist content of certain newsgroups seem to be the most pressing issue in relation to communication freedom and regulation. The problem with Internet is its cross-frontier character: information "unwelcome" in one country "migrates" to another to find a free area. Originally national prohibitions did not count because Internet does not pay attention to local boundaries (Rodriquez 1995,127). But as long as there is no geographical way of excluding particular newsgroups, chances are that national regulation catches up with this migration. The most restrictive policy then becomes the dominant one for all, be it sex in Germany or the US, racism in Germany or the Netherlands, and may be in the near future religion, politics, etc. in other countries. The moral panic over some of the content of the Internet and the call for (self)regulation brings back memories of the introduction of the printing press, newspapers, film, radio and television. Restrictions and often some form of regulation or even censorship of their content have been common place in the first phase of new media. And it was mostly the provider that was held responsible: the printer, publisher, bookshop or bookstand, the telephone or cable company and now the access providers on Internet. The choice seems to be between a tightly regulated information network where the most restrictive national values of one country are applicable to all or a network the content of which is constantly challenging the limits of free speech in a democracy or of the sexual norms in certain countries. Experimental forms of teledemocracy take place at the local level, but it is the global nature of the Internet that is threatening communication freedom. Although the US Communications Decency Act and the actions of CompuServe point to a serious tension between theory and practice of free speech in the information society, civil liberties movements tend to idealise the Internet newsgroups as a sort of vehicles for world direct democracy. Internet is important for consultation to strengthen democracy, newsgroups less so for conversation, let alone allocution. Rousseau was right when he said that direct democracy prospers best in small communities where citizens are relatively independent and with little difference between rich and poor. Political Representation There is another paradox in the arguments about ICT in politics. Where many applications were enthusiastically greeted by party politicians with a silent hope that interaction, debate and more information would lead to a closing of the gap between citizens and politics and thus to a revitalisation of parliamentary democracy, these same applications are said to result in the end of political parties themselves. It is a notion heard among both euphorics who see this as another proof of the direct democratic potential of ICT, and party politicians who suddenly fear they fetched in the Trojan horse. Electronic referenda and other forms of direct contact between decision makers and the electorate would render the intermediary function of political parties obsolete. The arguments of both are flawed, or at least one-sided. Where political parties are traditionally the brokers and articulators of what the wants and grievances in a society are and the translators of these into political demands, newsgroups on the Internet and discussion groups in the Digital City tend to strengthen single-issue politics. Moreover, their voluntary nature are not only their strength but also their weakness: no one is committed to the outcome. Debate in newsgroups might lead to consensus over (the solution of) certain issues, but as long as passive participation diminishes they are not much more than the traditional soapbox allowing for screaming in the dark. Improvement of consultation potential will make government and administration more transparent and probably more responsive to questions asked. It will remain the function of political parties and elected representatives to make choices between the different particular interests on the basis of a notion of the general interest. That does not mean that these choices should take place in an ivory tower. More than before, the existence of interactive media will mean that listening to and participating in debates — and thus making oneself accountable — becomes a part of the political trade. And 05 politicians could feel a bit uncomfortable with that, as they are more used to persua- siveness than responsiveness. The pallbearers of the political parties forget three more things. Firstly, that political parties are a recruitment base for future decision makers and power holders, and thus will remain needed. Secondly, for newsgroups and discussion platforms to take over the role of intermediary organisations people will have to become political animals. As we have seen, most of them are not even consumers of political information, let alone political actors. The anarchistic structure of Internet and the open, non-hierarchical structure of initiatives like the Digital City can, however, be an incentive for participation. Thirdly, political parties are needed as the organised control over power holders as, with electronic polling etc. they might base their decisions more on political marketing than ideologies and weighing different interests. Political Control The other controllers of power holders are the media, the watchdogs of democratic politics. Here again the argument goes that their role becomes less important now that people can consult all kinds of sources themselves. Citizens are not dependent anymore on the journalistic selection criteria based on sensational news values, databases are open for them and not only for journalists. And again this assumption is based on a notion of the public actively and consciously searching for (politically relevant) information. In reality political communication as well as communication use have traditionally been allocutive; over the years we have been allocutively socialised. The noted trend from allocution to more consultation and conversation is first of all a relative one and secondly more the result of industrial use. With the commercialisation of television, the popularity of video as a time shifting apparatus and remote control enabling to "zap" around informative television programmes, political communication becomes more than ever: politicians in search of an audience. And moreover, with the abundance of information, political marketing and targeted political communication, the critical and selective role of journalism is probably more needed than ever. Discussion and newsgroups can contribute to debate and opinion forming, but with their non-compelling nature they run the risk of remaining in the discursive public sphere, while journalism can lift issues more easily into the decision making sphere and on the political agenda. Conclusion With the same ease that notions like tele-democracy have been introduced, when new information and communication technology was applied in the domain of politics, one now talks of censorship and a return to the dark ages. A German prosecutor, CompuServe and the US-government have certainly triggered a debate over communication freedom in the age of Internet. And not only a debate. Internet, with the characteristics of telecommunications and the free speech claim of the press, has triggered a kind of regulation which may turn out to be more strict than most broadcasting laws in Western Europe. Access and content rules with regard to broadcasting were legitimised with the argument of scarcity of channels and the supposedly intrusive nature of the medium. Neither can really be applicable to Internet and other interactive ICT applications. There is an abundance of channels and potentially wide access while the elite character of the users can withstand intrusiveness, if that is the case. Only the position of children is comparable, but it remains a question whether this is a responsibility of the state or of their guardians. The panic, both over the content and regulation of Internet, overshadows the fact that a notion of tele-democracy can really only be used at local applications of ICT, and even than with reluctance. More than content, questions of access and participation, of privacy and registration, of control and representation come here to the fore. A discussion about fact or fiction of tele-democracy should take the complexity and in-terrelatedness of these questions into account. Notes: 1. Early February 1996 the German prosecutor started another investigation, this time into CompuServe enabling subscribers access to a neo-Nazi web site. 2. Early 1996 the Dutch government decided not to introduce new legislation with regard to the content of the Internet, after questions from two Christian Democratic MP's about sex and soft drugs on the Net. They were though more worried about the image of the Netherlands abroad than the effect the content might have on children. 3. The word allocution is derived from the Latin for the address by a Roman general to the assembled troops. 4. The majority of newspapers, both in the US and Europe, go "on line" without a really knowing why, but just afraid to "miss the boat" whichever way it is sailing. The same goes for a lot of other big businesses: they are in there for the money, although it might still take a while before it begins to pay. 5. Some of the examples discussed here are reviewed in more detail in Brants et al. (1996). There are of course other examples in Europe, notably the CityCard project in Bologna, Italy. This is an interactive system, accessible via Internet and kiosks all over the city, which offers residents the opportunity to send messages to and request information and services from the local authorities, and to set up discussion groups with or without the participation of the local authority. References: Bordewijk, J. L„ and B. van Kaam. 1982. AHocutie: enkele gedachten over communicatievrijheid in een bekabeldland. Baarn: Bosch & Keuning. Brants, Kees, Martine Huizenga, and Reineke van Meerten. 1996. The New Canals of Amsterdam: An Exercise in Local Electronic Democracy. Media, Culture and Society 18, 233-247. Depla, Paul. 1995. Technologie en de vernieuwing van de lokale democratie. Den Haag: Vuga. Garnham, Nicholas. 1989. Universal Service in European Telecommunications. In N. Gamham (ed), European Telecommunications Policy Research. Amsterdam: IOS Press. McQuail, Denis. 1992. Media Performance. Mass Communication and the Public Interest. London: Sage. Newman, Bruce. 1994. The Making of the President. Political Marketing as Campaign Strategy. Thousand Oaks: Sage. Pool, Ithiel de Sola. 1983. Technologies of Freedom. Cambridge, MA: Harvard University Press. Rodriquez, Felipe. 1995. Internetperikelen. Mediaforum 11/12, 127. Schumpeter, Joseph. 1943. Capitalism, Socialism and Democracy. London: Allen and Unwin. Selnow, Gary. 1994. High-tech Campaigns: Computer Technology in Political Communication. Westport: Praeger. Van Cuilenburg, Jan, and Paul Slaa. 1993. From Media Policy Towards a National Communications Policy: Broadening the Scope. European Journal of Communication 8, 149-176. Q Van Praag jr., Philip. 1996. Teledemocratie: de nieuwe illusie. Christen Democratische Verkenningen r^ 1,22-32.