In his writings on democracy Johan P. Olsen takes as his starting point a fundamental principle of the European and Western cultural heritage (Olsen 1990, 1991). Human beings are different in many ways, intellectually, morally and in terms of resources. Although such differences between people may be of great significance in many contexts, differences of these kinds have no relevance when applied to the right to self-government. The democratic creed is based on the willingness to accept that the design of regulations for the common solution of problems and conflicts in society is a decision that affects everyone to the same considerable extent and that every one is qualified to participate in an equal way in the making of such decisions. Democratic self-government is thus seen as a fundamental right. In a democracy everyone has the same right to be taken into consideration and it is the duty of the community to ensure that this is what happens (Olsen 1990, p. 24).

Two Interpretations of Democracy
Olsen goes on to say that there are two different interpretations of the concept of democracy. One ideal is centred on society, the other on the individual. The first ideal is based on the notion of the sovereignty of the people, the other is derived from the idea of the sovereignty of the individual (cf. also March & Olsen 1995).
The socially-centred notion of democracy emphasises the role of the individual as a member of society. The people are presumed to have a collective identity in the form of common customs, practices and moral principles. The interdependence of members of society and the feelings of belonging they share are emphasised. A socially-centred perspective lays great store in the fact that the life-chances of individual human beings are affected to a great extent by the social structures and the culture into which they are born and in which they live. These structures both create opportunities and impose limitations. They create the conditions for regularity and predictability. Through the process of socialisation, the members learn society’s common norms, values and expectations. Decisions are made on the basis of culturally determined standards and conventions about what is good and true.
The individually-based interpretation of democracy accentuates the distinctive character and primacy of the human being. The individual may not be subordinated to the needs and considerations of the collective. Every person is of equal value in moral terms and it is morally right to oppose any authority not based on the consent of the individual. The norm is that the individual shall be free to form an independent judgement as to what is true and good and be free to make decisions on the basis of these judgements. Individuals are assumed to have the ability to choose and to take responsibility for their decisions. Individual members of society also develop both morally and intellectually through the exercise of freedom and responsibility of this kind. An individually-based society is permanently changing as a result of the continual utility calculations on the part of individuals and as a result of the adaptations made by individuals consequent on these calculations.
On the basis of a review of the Swedish social debate, Olsen concludes that the prevailing view of democracy should most properly be characterised as socially-oriented. The Social Democratic understanding of democracy, in particular, contains important collectivist elements. This view of society emphasises the equal value of human beings, the need for active participation and an equal distribution of resources. In consequence the Swedish ideal of democracy lays great store by the sovereignty of the people and by equality. Solidarity between citizens is an essential precondition if a modern society is not to be fractured by internal conflicts between various groups, Olsen quotes from a Social Democratic text (Olsen 1990, p. 19).

The Swedish Model
When the Study of Power and Democracy summarised its conclusions, these made clear that the collectivist notion of democracy was closely connected with a particular stage in Sweden’s history which is usually labelled the "Swedish Model". It is assumed as part of this model that democracy is realised by decisions of a political majority, a large public sector and centralisation. However, as authors of the study, we were of the view that social changes meant that the load-bearing elements in this social system were being weakened. Several of these institutions were connected with a type of industrial society whose time had passed. This model of democracy was more appropriate for describing the real workings of government of society as it looked several decades ago. As a result the conclusion drawn was that the processes of change mean that society is not moving closer to the collectivist idea of democracy of the Swedish model but moving away from it (SOU 1990:44, p. 403).
This is not to say that democracy has deteriorated in a more general sense. Proponents of an individually-oriented perspective on democracy would be able to find much of positive value in the changes that have taken place in society. Despite the considerable differences that remained between different social groups, the general direction of change was clear. The qualities which were once heralded as the ideal of democratic citizenship, such as being well-informed and enjoying economic and social independence, were increasingly in evidence. According to one survey, forming an opinion independently of others was the civic virtue that was most highly prized. The preconditions for the autonomous taking of responsibility on the part of the citizens had been improved. From the point of view of an interpretation of democracy more clearly identified with the individually-oriented ideal, the conclusion was thus a fairly positive one. The direction of change was increasingly towards autonomous and independent citizens.
These transformations created new problems. A gulf grew between latent and manifest citizenship. The expectations citizens have of exercising personal influence increased more rapidly than did any real room for action. As a result the subsequent development has also been characterised by the powerlessness of the citizen. Many individuals have started to look for other routes towards exerting influence over their particular situation. Established organisations have found it increasingly difficult to retain their existing members and to recruit new ones (SOU 1990:44)
Any assessment of the vitality of democracy must largely be based on an evaluation of the commitment, knowledge, understanding and tolerance of members of the public. One of the main aims of political institutions is to promote positive citizenship. As a result participation on the part of citizens is built into the very definition of democracy. However, widely differing interpretations exist as to how the individual should play an active role in political decision-making processes.

Contending Approaches to the Study of Democracy
During large parts of the post-war period, the confrontation has been between two approaches to the study of democracy. Of these, the participatory democratic view , has made claims to be the classical one, while the other - the competitive democratic view - has been considered revisionist.
The school of participatory democracy emphasises the significance of the actions of the individual. Here, the key element of democracy involves all the citizens taking part in political decisions. The power of the people lies in the right to co-determination and the freedom to participate. The participatory democrats also make use of a broad definition of politics; the requirement for participation applies to all the areas of social life. Democracy should preferably be direct; through popular referendums, for example. Political awareness is fostered in individuals through their direct participation.
On the other hand the school of competitive democracy considers the most important aspect of democracy to be the competition by several elites or parties in free elections. The power of the people lies in the freedom to change the government. Frequently this view takes a restrictive definition of politics for granted; the emphasis is placed on the activities of the representative bodies. In consequence a negative view is taken of popular referendums and the emphasis is placed squarely on the freedom to call those elected to office to account. The voters are often considered to lack the knowledge required to make decisions on vital political issues.
In recent decades a third view has established itself as an alternative to both these schools of thought. It has been characterised either as discourse democracy or deliberative democracy (Fishkin 1991, Gutmann & Thompson 1996, Elster 1998). The key idea is that democratic legitimacy derives from the dialogue of politics. Only those decisions arrived at after the alternatives have been subjected to challenge in a free and open dialogue should be accepted as democratically valid.
All three lines of thought may be criticised. The participatory democrats underestimate the need for political leadership and for institutionalised forms for calling those in power to account. The competitive democrats tend to overlook the requirement of popular government for committed and responsible citizens. Discursive democracy risks neglecting the requirements of democracy for effective government and effective methods for making political decisions and implementing them.
The internal polemic between the three schools of democracy has made clear that all three are associated with shortcomings. Each line of thought highlights a certain element of the democratic polity but tends to neglect other elements which are also important. None of the three schools can thus constitute a valid alternative by itself.

The Problem of Political Participation
Even though there exists a close connection between participation and democracy, participation cannot be considered as the ultimate test of democracy. This would mean that a proper democracy would require maximum participation, an entire population in a permanent meeting. It was once observed that the conclusive argument against socialism is that it would take up too many evenings. The requirement that distinguishes democracy should be formulated instead in terms of rights. Democracy is realised to the extent that the citizens have real opportunities to influence the decisions that affect them, to choose their representatives, to form opinions and to join together to form independent organisations.
It could be said in more incisive terms that democracy does not necessarily require equal power, but equal rights instead (Walzer 1983). Here, rights mean the guaranteed freedom to exercise a minimum of power (the right to vote) and to exercise a greater measure of power (freedom of expression, freedom of assembly, the right to make demands). It is not legitimate for the winners, for the majority, to make use of their unequal power to deprive the losers, the minority, of their rights and freedoms. While the majority holds sway over the minority by virtue of the principle of majority rule, it would be tyranny to conclude on that basis that it has the right to rule for ever. Democracy is not founded solely on the rule of the majority but also on tolerance. Political rights are permanent guarantees. They constitute the foundations of a process without end, an exchange of opinions which is never completed. In a democracy, every situation is a temporary one. No citizen can claim to have convinced his or her kinsmen once and for all. New citizens are constantly coming into being and every citizen has the right to restart the discussion. It is therefore inappropriate to characterise the ideal model of democracy as one in which political power is equally distributed at any given moment. However, the requirement for equal distribution does apply to the possibilities of achieving political power. Every citizen is a potential participant, a potential politician.
A study of how democracy functions in practice must focus attention on three different aspects: what the citizens are allowed to do, what they are able to do and what they actually do. What the citizens are allowed to do represents their formal democratic rights. These are an essential, but insufficient, criterion to determine whether a democracy is functioning in real terms. The formal rights must also exist in tandem with resources and skills, with an ability to exploit these rights. However, a functioning democracy also requires something more than general knowledge, and that is active participation.
The relationship between participation and democracy is to some extent a paradoxical one. On the one hand not everyone need participate, while on the other some people must participate. Politics can never be reduced to an automatic, a self-initiating, set of operations, even if many utopias have cherished just such a hope, it is rather the creative actions of human beings. The political individual, homo politicus, is an essential component of every society. Political activity is not an essential part of every individual person but it is a vital component of society as a whole.
Even though the scale of participation should not be seen as a definitive criterion of democracy, data concerning who the participants are is nevertheless of great interest if the aim is to assess the workings of the democratic process. How does the political individual differ from other citizens?
Nor is political participation something that is cast in stone. Over time participation may alter radically, both for society as a whole and for individuals. In periods of calm, politics may be an affair for a minority of individuals, while in times of crisis a greater proportion of the citizenry may be involved in political action. As far as the individual is concerned, periods of enthusiastic commitment may alternate with periods of retreat from public affairs. Entire theories have come into existence about such cyclical swings between the public and the private (Hirschman 1982).
While no doctrine of democracy can enforce a duty to be political active in perpetuity, the proposition that the citizen of a democracy should possess a certain preparedness to get involved in politics can legitimately be made. Democracy presupposes that the citizen is potentially in a position to represent his or her interests in an active fashion and that in any case a minority is also prepared to involve itself in politics in a more lasting and committed fashion (Scharpf 1975). A study of political participation at any given point becomes in consequence a study of which citizens have on this occasion found reasons to convert their readiness for action into action itself.
The level of participation that is actually observed may easily be misinterpreted, tending both towards underestimation and towards overestimation. Concentrating on active citizens neglects those citizens who are prepared for action but have for some reason decided not to become involved. This fact forms part of the problem of interpreting silence. Passivity and silence may be an expression of satisfaction and
self-confidence as much as of powerlessness and frustration. A measure of the number of active citizens may therefore underestimate the latent political capability of members of the public.
Conversely the observed level of participation may lead to an overestimation. Activity need not mean influence. Attempts to exert influence may fail. A high level of participation may be a manifestation of the desperate attempts of citizens to assert their interests. The interpretation of the meaning of political activity is thus dependent on the particular assessments of the citizens, their feeling of being able to exert influence and the view they take of the legitimacy of the political system.
The question, then, is whether the level of political participation in Sweden should be described as high or low. The answer will depend on the point of comparison. A first possibility is to compare the real state of affairs with a norm of some kind. The theory of democracy fails to provide any unambiguous answer. Irrespective of which standpoint is adopted in this debate, it may be observed that the political institutions and (as a consequence?) the pattern of participation of the citizens lies closest to the competitive-democracy model. The ideal participatory democracy requires considerable more active citizens than those of today’s Sweden.
A second possible point of comparison is to relate Sweden to other democratic states. The statistics on electoral participation show that we currently lie close to the norm for states with proportional electoral systems. While other indicators are more difficult to compare, Sweden’s position particularly in relation to membership of organisations is a relatively high one. It is difficult to find forms of participation which lie on a lower level than elsewhere. The inevitable conclusion is that political participation in Sweden lies at much the same level as in other countries.
The third response to the issue of the level of political participation is to make a comparison over time: has the activity increased or decreased? While the answer to this question is not constant, it has also varied over time. A decade or so ago the conclusion was relatively clear (Petersson, Westholm & Blomberg 1989). While party membership and political participation in the institutions of representative democracy have remained stagnant, the trend in other areas was upwards. A growing number of Swedes had become involved in various forms of individual opinion formation, contacts and political manifestations. The proportion of those making some form of written expression of opinion in the media had doubled during the preceding twenty year period. The proportion of persons who had spoken at a meeting on some occasion had all but doubled in size. The same trend could be registered for collective expressions of opinion in the form of public demonstrations.

Stagnation and Decline
These observations were of great significance for the general conclusions reached by the Study of Power and Democracy. At issue was a very noticeable increase in citizen involvement. There was no support for the hypothesis that the level of political participation varies in a cyclical pattern. In actual fact the political climate was a great deal warmer during the "peaceful" eighties than during the "stormy" sixties. The trends that could actually be observed represented a long-term process of change rather than short-term fashion trends. The level of education rose, women became more independent and access to information improved considerably. The changes that had taken place in society must be considered a positive force from the point of view of democracy.
It was therefore with some excitement that the results were awaited of the corresponding study carried out a decade later, in 1997. Would the long-term, demographic forces have continued to raise the level of participation [---]? Given that the level of education has continued to rise, that more and more women have reached decision-making positions and that the media landscape has been a flourishing one, the expected hypothesis would be for participation to have risen. But reality had in fact taken a different path (Petersson et al. 1998).
Over the years interview surveys have contained questions designed to measure the extent to which members of the public are administratively competent and thus able to put their complaints into effect. At issue is whether the individual respondent feels able to undertake the composition of a written appeal against the decision of an administrative agency. Two thirds replied in the affirmative in 1997 while a third of the population do not see themselves as being capable of composing a formal written appeal. These people are not entirely helpless however. It turns out that the majority of the group are personally acquainted with an individual whom they know could help them in such a case. Those who consider themselves totally excluded, who are neither able themselves nor personally acquainted with someone else who could help them, make up a tenth of the population. Studies show that the level of administrative competence rose continuously from the end of the sixties to reach its highest point at the beginning of the nineties. The 1997 findings are particularly noteworthy in as much as that proportion is no longer increasing. Since previous analyses show that administrative competence is closely connected with the length of education, a likely hypothesis would be that this proportion would continue to increase concurrently with the rise in the level of general education. Such is, however, not the case. As a result the stagnation in the administrative competence of the citizen is all the more remarkable..
Some of the other survey questions are designed to provide a picture of the citizen as a participant in the opinion-making process. At issue here is whether the respondent has addressed a meeting of an association or organisation and whether the respondent has written a letter or an article for a newspaper or magazine. Measurements of individual opinion formation showed a fairly clear-cut and substantial increase between 1968 and 1987. Subsequently, however, this curve was to flatten out. Although there is no sign of any decrease, the positive trend that could still be noted at the end of the 1980s has now been interrupted.
Patterns of participation in public demonstrations have developed in a similar fashion. After the first measurement taken at the end of the 1960s, there followed a twenty-year period of continually rising participation. This trend was interrupted after peaking in 1987. During the 1990s, it has been a question of decreasing levels and stagnation.
While the question whether the respondent had at any time been in contact with an official in order to influence a political decision did not form part of the 1987 survey, the pattern for other indices is nevertheless largely in agreement with previous findings. The rising level until the beginning of the 1990’s has now been replaced by a slight decline.
As for party activities, the changes have followed a rather different course. Here the questions apply to participation at a political meeting or rally, membership in a political party and serving in any official capacity in a political party. The maximum level was recorded several decades ago. It was clear by the end of the 1980s that party activities had stagnated although it was not yet possible to record any clear decline. The contrast was nevertheless clear with other forms of participation which all showed rising trends at this point. Now, however, a clear decline can be noted. The summary index shows that party activity has halved over the last fifteen years.
For many years participation in national elections was at a relatively high level in international terms. Any large-scale mobilisation of the electorate is, however, no longer an issue. The record figure of almost 92 per cent had already been achieved by 1976. Voter turn-out fell by several percentage points at the 1988 election and the figure for the 1998 election of 81.4 per cent means that Sweden is now back to the levels of the 1950s.
The picture of political participation in Sweden must now be revised. The analysis made by the Study of Power and Democracy, which was primarily based on the citizen survey of 1987, showed that people had become more active, although their involvement was mainly in channels other than via the traditional parties. The change during the most recent ten-year period has been quite different. No increase has been recorded in any of the areas studied. The political involvement of the citizens has stagnated or is declining. Participation within the political parties has reached such a low level that it must be considered alarming.
Even when the perspective is further extended to include participation in associational life as well, the picture fails to become any brighter. Changes to the level of membership in associations between 1987 and 1997 mean that the proportion of those who remain wholly outside associational life has stayed largely unchanged. On the other hand the size of the group of those who are members of many associations is declining. In this regard, too, there are fewer and fewer enthusiasts. The changes between the two survey dates are more marked in terms of activity and official service than in relation to membership. The level of activity has fallen in an almost alarmingly uniform manner across various types of organisation. If only formal membership is considered, the decline in the popular basis of associational life has been relatively moderate. Behind the surface, however, a considerably more dramatic change has taken place. Compared with the situation ten years ago, what is clear today is that fewer Swedes are taking on associational duties and the proportion of those active in associational life has also fallen markedly.
The trend in relation to commitment to associational activities is in accord with the stagnation and decrease in political participation, particularly for activities within the political parties. The small group of citizens which keeps associational life going has become even smaller. This decrease brings into focus the question of where to set the minimum level for the social movement ideal which is founded on the notion of a large number of citizens becoming involved in organised forms of cooperation on a voluntary basis. Associational Sweden is starting to creak at the joints. Many studies have shown that the vitality of civil society is of major importance for both the workings of democracy and for social development in general (Putnam 1992). The view of the future now opening up is therefore a bleak one. The erosion of social capital may lead to negative consequences for society as a whole.

Representative Democracy in Transition
By extrapolation a worrying trend can be discerned. If the internal vitality of the parties continues to diminish, this will impair their capacity to formulate relevant and attractive alternatives. Elections risk being seen in consequence as less and less interesting. The result may be lower levels of voter turn-out. The stagnation of the parties may become a self-perpetuating, negative spiral. The difficulties posed by the need to recruit new individuals willing to take on political office are increasing. Isolated party leaderships are becoming increasingly dependent on the media and opinion polls. The problems for associational life and the political parties could develop into a crisis for representative democracy in general.
Representative democracy may be considered as a system of communication. Voters and representatives are linked together by an exchange of messages and signals. However, the means of communicating in society varies strongly over time, particularly as a consequence of technological change. From a historical perspective mass media have become the most powerful institution for mass ommunication. But the media do not simply mirror social changes, they also contribute to changing society, both directly and indirectly.
The media logic of modern journalism leads to greater attention being paid to particular issues rather than to coherent ideologies. The idea that a political party should stand for an overarching view of society is harder to realise in a consumer democracy. Politics becomes more of a impulse purchase than a forum for beliefs and for joint projects. The temporal perspective is foreshortened, the pace of the political process is accelerated. The media become more important than organisation. The party leadership can reach out directly to the voters without having to go through the lumbering party apparatus. Support from the public purse for the parties makes members unnecessary. Problems of recruitment for local parties are continually increasing. Considerations of publicity and media consultants are becoming ever more important.
Democracy is a form of society based on cooperation and mutual trust. Interview studies indicate that there is considerable support for civic ideals such as rationalism, obedience to the law, solidarity and participation. Even though the majority of people consider themselves to meet the ideal, there is a widespread notion that other people do not show solidarity and try, for example, to avoid tax and to cheat on benefits. There is an obvious distortion in the current process of opinion formation. The general public underestimates the readiness of the general public to cooperate. The risk is that a misjudgement of this kind will lead to a change in behaviour on the part of individuals. Since many people believe that others are failing in their duties, there is a diminution in personal readiness to be a good citizen (Petersson et. al. 1998).
When the Study of Power and Democracy concluded its deliberations in 1990, it was obvious that the country found itself in a period of institutional reorientation. The form of social organisation that had characterised Sweden for a considerable number of years had been conditioned by relatively stable social structures. The effect was to reinforce the impression of inevitability, control and permanence. The changes that could be discerned a decade previously mean that these structures have been weakened without any new ones being established. The process of social change appeared more open, uncertain and unpredictable as a result. Although political institutions are slow to change, they need to be adapted
to changing conditions (SOU 1990:44).
The idea that democracy is realised to the extent that a single national power centre can oversee and steer the entirety of social change is becoming increasingly unrealistic. In a continually more differentiated, specialised and internationalised society it is hardly credible that a single central body, such as the national parliament, will be able to fill the role of the dominant centre of authority. The fact that a single person or institution cannot have total control over the whole of society does not mean that social change cannot be influenced. It is scarcely probable that the democratic direction of a modern specialised and differentiated society can be based on a single dominant governmental principle or decision-making process. Government has to be able to take account of many, partly intersecting, considerations and interests. A strengthening of democracy will therefore depend on the extent to which society changes and institutionalises a new mixture of principles and processes.
A comparison of different governmental principles leads to a constitutional debate. A vital democratic constitution provides rules for which roles the various principles, participants and resources should be allocated. The rules clarify the allocation of duties, powers and responsibilities. They establish the power of the political centre and what rights and duties the citizens, their representatives, experts, bureaucrats and organised interests should have. In this way they provide guidance as to how democratic processes can be protected from the superior resources available to the rich, the well-organised, the well-educated, the well-informed and the superactive.
The task of constructing political institutions which promote and help develop the democratic consciousness of the public can therefore not be reduced to a choice between either a purely collectivist model or an purely individualist one. The issue in practice is to create institutions which realise the ideal of self-government. The democratic form of government presupposes that there are autonomous individuals able on the basis of critical reflection to take responsibility for living under a system of common rules.
There is therefore reason to repeat the concluding reflection of the Study of Power and Democracy, to which Johan P. Olsen contributed. A key issue is what should be made the object of collective decisions and what should be left to the individual, to groups of citizens or to autonomous institutions not subject to permanent intervention on the part of the central state power. A question of primary importance when it comes to finding a new balance between a socially-centred and an individually-based view is how to bring together the desire for social security on the basis of common rules and regulations with the desire for individual freedom of choice. Ultimately this a question as to how to combine the freedom of personal choice with solidarity and community.

Translated by Frank Gabriel Perry

References
Elster, Jon ed. 1998. Deliberative Democracy. Cambridge: Cambridge University Press.
Fishkin, James. 1991. Democracy and Deliberation. New Directions for Democratic Reform. New Haven: Yale University Press.
Gutmann, Amy and Dennis Thompson. 1996. Democracy and Disagreement. Cambridge: Harvard University Press.
Hirschman, Albert O. 1982. Shifting Involvements. Private Interests and Public Action. Princeton: Princeton University Press.
March, James G. and Johan P. Olsen. 1995. Democratic Governance. New York: The Free Press.
Olsen, Johan P. 1990. Demokrati på svenska. Stockholm: Carlssons.
Olsen, Johan P. ed. 1991. Svensk demokrati i förändring. Stockholm: Carlssons.
Petersson, Olof, Anders Westholm and Göran Blomberg. 1989. Medborgarnas makt. Stockholm: Carlssons.
Petersson Olof , Jörgen Hermansson, Michele Micheletti, Jan Teorell and Anders Westholm. 1998. Demokrati och medborgarskap. Demokratirådets rapport 1998. Stockholm: SNS Förlag.
Putnam, Robert D. 1992. Making Democracy Work: Civic Traditions in Modern Italy. Princeton: Princeton University Press.
Scharpf, Fritz W. 1975. Demokratitheorie zwischen Utopie und Anpassung. Kronberg: Scriptor.
SOU 1990:44. Demokrati och makt. Maktutredningens huvudrapport.
Walzer, Michael. 1983. Spheres of Justice. A Defence of Pluralism and Equality. Oxford: Blackwell.