Direct democracy, communication and power relations

Preliminary version/110801

Institute for Social Research/Department of Sociology and Human Geography, University of Oslo

Presentation at the conference "Power and Communication in a Direct Democracy"

University of Oslo, November 9, 2001


In the following I shall present and discuss one case of direct democracy. The case is a socialist publishing collective, which I followed over an eight year period in the 1970s, first as a member of the collective – a sort of observing participant – and in the last phase as a full time observer, doing research on the collective. The research is reported in the book Likhet og styring. Deltakerdemokratiet på prøve (Equality and governance. Challenges to participatory democracy, Engelstad 1990).

The term ‘direct democracy’ may be defined in many ways. Frequent use of referenda, popular ownership of productive means, and general participation in decision-making are three common elements in such definitions. I shall only treat the latter aspect. The core of the Work Collective was the institution of the General Meeting, meeting every second week, where all members were participating on an equal basis. The members of the collective were not owners, but were acting as directors of a joint-stock company with very dispersed stockholder structure.

When discussing direct democracy, it is almost always interesting to make comparisons with ancient Athens. In the following I shall do that somewhat sporadically. At the same time, four significant differences between Athenian democracy and the Publishing collective should be borne in mind.

Scale. When discussing the differences between direct democracy and modern representative democracy, the point is often made that the modern state is much larger than the ancient city-state. Thus, experiences from the city-state cannot be transferred to modern societies. Here the point is the opposite, the collective was a very small group, varying somewhat in size over the years, but never comprising more than twenty persons at the same time.

Integration. Athens may be described in modern terms as a hyper integrated society, where sacrifice of the citizens’ own life was regarded as the first duty. This was the core message in religious rituals and artistic performances (Ober and Strauss 1990). Thus, Athenian citizens had a common destiny as members of the city-state. In contrast, the members of the Work Collective had common goals and common political aspirations, but at the same time they had a life outside the collective itself.

Turnover. A characteristic trait of the Collective was a rather high degree of turnover. Members stayed in the collective for an average of three to four years, some even shorter. Joining the Collective was often the first step in a career ladder. In contrast, membership in the Athenian state was strictly regulated, and foreigners had no chance to acquire full citizen rights.

Scope of decisions. The Work Collective was a production unit, and not a political organ. Its most important relationships to the external world was mediated by markets: product markets for books and printing services, the labor market, and capital markets. In Athens, some of the core issues in democratic decision-making were related to courts, to war and peace, and taxation. Members of the collective were of course never confronted by similar issues.

Even if it does not make sense to make direct comparisons between ancient Athens and the Work Collective, I believe the two cases may be related to a set of similar questions, demonstrating that they in some respects differ in interesting ways. Yet, the main point here is to discuss problems of direct democracy, not to make comparative analyses. The core problems taken up in the following have to do with power and power distribution among members. But they also include problems of collective powerlessness and lacking capacity to formulate and/or implement decisions bearing on the sustainability of the collective.

The exposition falls into two main parts. First I discuss changes in the mode of operation in a conventional organizational perspective, where the main elements are efficiency and transaction costs. Then I go more systematically into a communicative perspective, trying to track down elements in the communication structure that may have separately affected the distribution of power in the collective.

I. Equality, democracy and power structure

The Work Collective and power

In general, an organization may be characterized as a structural distribution of power. Who says organization, says hierarchy. The Collective, however, was in the opposite situation. It was an organization with the explicit goal of avoiding power differences. It had not always been so. In its first years it was dominated by a strong leader, who was also the founding father of the publishing company. When he left, the collective was instituted, in 1972. The period with collective leadership lasted until 1980, when the last of a series of economic crises forced the members to reintroduce a more conventional leadership structure.

As already mentioned, the core institution in the Work Collective was the General Meeting, where all members ware taking part, and where all important decisions were made. In principle, any issue might be discussed in the General Meeting. Full openness and free discussion were the highest values. Equality in decision-making was not accompanied by equality of tasks, however. The conventional division of labor found in any publishing house was preserved. Members were acting as editors, production and sales personnel, business manager, etc. Thus, in any discussion participants would act both as a member of the Collective and as a representative of her or his division.

When collective governance was instituted, absolutely all forms of leadership were abolished. Responsibility for setting agendas and chairing meetings was rotated between members, from one meeting to the next. However, this completely anarchistic system gradually and very slowly gave way to a more and more hierarchic order. In order to increase continuity, a working committee of three members was to take responsibility for preparing and chairing of meetings. All Collective members would rotate through this committee. At the outset the period of functioning in the group was very short, but later increased from one month to three months, and then six months. As part of this enlargement, members of the Working Committee were not selected by rotation, but appointed, partly as representatives of their division. One of the editors and the business manager became permanent members of the Working Committee. Thus, over a period of some five to six years, almost unnoticeably, despite the professed egalitarianism among members of the collective, a stable hierarchy was instituted.

The institution of this hierarchy, however, never undermined the sovereign position of the General Meeting. The Working Committee was responsible to the General Meeting, and all important decisions were made there. However, an informal hierarchy obviously existed in this context, basically along the same line as those drawn up in the selection for the leader group. A questionnaire survey conducted at the end of the 1970s revealed a very strong informal hierarchy, almost as strong as those found in companies with conventional management structures (Engelstad 1990:chap. 7). Similar investigations were not made at an earlier stage, but it seems plausible that the strength of the informal hierarchy developed parallel to the changes in leadership structure sketched above.

Direct democracy and power differentials

One possible explanation of the development of the informal hierarchy is that it is the result of the will to power among those who came out on top. They had a more or less conscious intention to build a power base for themselves, or at least did not support fully the idea of complete equality. Another hypothesis might be that conceptions of equality, or the weight accorded to equality, gradually eroded. However, both these hypotheses are contradicted by all available evidence. There are no indications that the most powerful persons acquired their position as a result of long term strategies. And likewise, there is no sign of members losing confidence in their adherence to equality. Power differentials should rather be seen as an indirect, and unintended result of a series of other choices, and constraints. The result was frustrations, which from time to time might be considerable, but not that members of the Collective deserted their values. In the following I shall discuss rather briefly five sets of such factors.

Division of labor

As mentioned, the conventional division of labor in a publishing house was upheld, even if decision-making authority was dispersed. Even if all aspects of the production process was open to scrutiny, everyone would have a better grip on their own jobs than others, and would be accorded confidence to the same degree. An analysis of the production process in publishing makes it clear that the most complex decisions are those made in the first stages of the process, by editors. Their tasks consist in scanning the market for prospective authors and manuscripts, selecting titles on rather diffuse and discretion based criteria, such as literary quality, ideological acceptability, and market potential. Every title would have to be decided by the General Meeting, but the editors were the ones with the greatest expertise, even if they in no way could dictate the decisions of the meeting.

Compared to the editorial division, where each title is a new product, jobs in production and in sales and marketing are more routinized, even if they require a substantive amount of creativity and imagination. To use the phrase of Herbert Simon (1957), in the collective editors are the foremost suppliers of premises. With the exception of the business manager, who also has a very complex job, in questions concerning the selection of titles, editors are experts, the others are interested laymen.

Human capital and ideological expertise

In line with the division of labor, there was a difference in formal education between editors and business managers on one side and the others, editors having the highest amount of human capital. However, the direct effects of these differences may not be too great. All members of the collective were recruited from milieus of student activism, and were used to taking part in political debates. They were not necessarily very impressed by the formal education of others, something that was enhanced by the egalitarian ideology.

On the other hand, publishing is not the same as political activism. Some of the most prestigious books published by a socialist publishing house, are related to the field of what is called ‘theory’, a mixture of Hegelian political philosophy and Marxian political economy, set out in various schools of critical theory and ‘Western Marxism’. The self reflection of a socialist collective to a certain degree at least, should be expressed in this very complex language. There is no necessary correlation between ideological expertise and formal education, but the overlap is usually fairly great. This is the stuff that editors and their likes must be able to master, whereas others on the average have a higher preference for listening to the exegeses.

Problems of learning and time horizons

Given the egalitarianism of the Work Collective, the differences among members created by the division of labor and education, could have been compensated by internal training. At any rate the Collective was a place for intense learning processes for all persons involved. Hardly anyone had the professional experience in advance, that is necessary for running a work organization as complex as a publishing company. Thus, everyone was profiting from the possibilities for developing their personal capacities. In the long run, less educated and experienced participants would have a good chance of arriving at a very high professional level.

The problem was that the members of the collective did not have a long term horizon. As mentioned, turnover was high, with an average period of 3-4 years. This led to the paradoxical result that those with highest education also had the highest learning, and that in many instances they gained more than those with a lower education, even if their increase in competence in the long run would have been the highest. This is explained by the assumption of a steeper learning curved, but within a shorter time span, for those with the highest initial qualifications. Given the high turnover rates, those with highest education came out as winners.

Two learning curves

Effects of external support groups

In addition to the decision-making process through the General Meeting, the Work Collective had two types of external support groups, which also had some effects on the distribution of power in the collective. Organized as a joint-stock company, the Work Collective formally had a board of directors. The board, however, did not operate as a separate body, but held its regular meetings in common with the members of the collective, in a body called the Company Assembly. Board members were selected as representatives of, or resource persons from, the political movement, but only exceptionally they had insight in publishing. In the Company Assembly they formed a minority. As a rule, discussions in the Assembly normally took the form of a dialogue between the most eloquent members of the Collective and the external board members. Ordinary members of the Collective mostly felt they were reduced to silence.

A parallel effect was caused by specific external support groups connected to the various divisions. Given the complexity of tasks, the editorial division had the highest number of such groups, and the most active ones. For external members of these groups, the main motivation for participation was the chance to influence the editorial profile of the Collective. The unavoidable result was a reduction of the decision-making power of the General Meeting. If the Meeting was reluctant to follow the suggestions of the support group, the argument would automatically come up: "If we refuse, central members of the group will withdraw." From time to time, frustration among collective members was looming large.

Restrictions on informal hierarchy?

Virtually all organizations contain some form of informal hierarchies. This informal hierarchy competes with the set of formal positions, the official hierarchy. From time to time tensions between the formal and informal structure arise, something that is a challenge to the incumbents of formal positions. In the Collective, the situation was slightly different. There was no formal hierarchy, to which the informal structure could be compared and assessed. Members were confronted with the discrepancy between official equality and informal hierarchy, and without anyone singled out as the one to take charge to make the two more compatible. The paradoxical result of egalitarianism was that hierarchy was made more visible and more problematic.

Sources of powerlessness

The mechanisms described above are connected to the emergence of a hierarchy, or in other words, a concentration of decision-making power. At the same time, other mechanisms contributed to weakening of the capacity of action in the Collective. These mechanisms are common to many forms of direct democracy, with ancient Athens as an obvious example.

Instability in the decision-making system

All humans have mixed motivations. This is even more true for assemblies of some size. Without the establishment of an elaborate set of checks and balances, they easily become myopic, dominated by short term considerations. Yesterday’s majority decision may easily be overturned today, either because decision-makers change their minds very fast, or because they do not realize the contradictory implications of different decisions.

In addition comes a general resistance to long term planning. Unpopular decisions are easily rejected, and long term problems are not perceived and handled before they become acute, and then it may be too late. This is particularly problematic in connection to questions where members of the decision-making body have special interests, which may be the case for instance in reorganization, or even downsizing of the organization.

Problems of authority and delegation

The hierarchy that emerged in the Collective was in many ways a conditional hierarchy, without very clear rules of delegation. Implicitly, delegation did not imply authority in matters concerning persons, only impersonal business questions. If one of the members of the Collective was having problems in their jobs, they did not get the support and correction, in other companies usually handled by close supervisors. In the Collective this would be the business of the General Meeting. At the same time, members would be very reluctant to bring such matters before the meeting, either due to insufficient information or to personal solidarity.

This also resulted in minimal control with members taking initiatives on their own, initiating processes that might go out of hand, in some cases with very serious economic consequences. In other words, the Collective was strong on ideological leadership, but weak on leadership in matters of personal relevance and job performance.

II. Direct democracy and problems of communication

It is a standard assumption in organizational studies that efficiency is a driving force shaping organizational structure. This may be based on theories of resource mobilization, or transaction costs (Williamson 1975), or both. As shown above, such problems were endemic to the Collective. However, they might be compensated in various ways, without giving up the basic vision of the organization. Even if there were efficiency losses caused by the collective structure, there were also substantive gains. Members were willing to work more than full hours without extra pay, and the collective form of organization mobilized their imagination and enthusiasm in ways normally quite uncommon in the ordinary working life. Thus, theories of efficiency, while useful, do not satisfactorily explain the troubles of the Collective. Probably theories of communication may shed more light on the problem.

Theories of power and communication usually take one of two forms. Power is either exercised by means of aesthetic effects, or through some sort of definition of the situation or exclusion of certain themes. I believe both approaches to be relevant (Engelstad 2001). However, they become more useful when seen in contrast to the idea of the ideal speech situation, which also may be formulated as a theory of the absence of power in communication (Habermas 1984).

The ideal speech situation

The concept of the ideal speech situation is not a prescription for empirical processes, but a normative model with the status of a regulative idea. Nevertheless, in discussions of the basis of the normative model, it is useful to have a clear understanding of the causes of deviations from the model. Only then can it be handled with a minimum of realism.

The core of ideals of direct democracy is parallel to this conception. The point is exactly to avoid alienating buffers between voters and representatives, and to enhance the possibilities of voters to enter directly into discussions with one another.

In its empirical version, an ideal speech situation may be conceived in two ways, depending on the size of the democratic body. In bodies with more than a very small number of members, the systematic distinction between speakers and listeners is unavoidable. This is typically the situation in the Athenian direct democracy. A group of highly trained speakers appear before the assembly, which shows its approval or disapproval by applauding booing, and finally, voting. This form may be characterized as ‘representative’ in the meaning that given speakers represent relevant arguments, even if they do not speak on behalf of a given group or party.

Alternatively, the speech situation may take the form of a conversation between a limited set of actors, where everyone is expected to take part in the deliberations on an equal footing. This is the situation found in the Work Collective. I believe that exploring the two sheds some interesting light on the possibilities both for non-authoritative communication and direct democracy in a strong sense of the word. In both cases, the analysis should take up problems connected to the production and sharing of information, effects of rhetorical expression, and the exclusion or formulation of problems.

Production and distribution of information

The variation of complexity between divisions and jobs in the Collective, leads to a problem which may be termed the inverse information flow: thos in jobs with high degree of complexity produce much information as input to discussions, whereas people in low-complexity jobs produce little information. This means that those who produce the least, must use most time and effort to go through and interpret what they receive. There is both a time aspect and a learning aspect to this. Producing information enhances the capacity to digest information from other sources. Thereby, people in low-complexity jobs are doubly disadvantaged, having to read more with less training.

Inequalities in communication are reinforced by time constraints on discussions in meetings. Those producing most documents also are the ones who are allowed most speaking time in discussions, simply by virtue of being the producers. They have to present their views to the meeting, and defend them in the ensuing debates. Thus, sharing of information entails a high degree of communicative specialization.

Is there a demarcation line between argumentation and seduction?

The theory of the ideal speech situation implies a strong adherence to the virtues of pure argumentation. The conception of the ‘power of the best argument’ is one of the central elements in Habermasian theory of communication.

In Athenian democracy, this question was very pertinent. The fear of empty rhetoric, used to seduce the people against their own interests, was lively in those considering themselves as "true philosophers" in contrast to mere "sophists", as exemplified in Plato’s portraits of Socrates. However, rhetoric was codified as a discipline in the Antiquity, with exactly the opposite goal. The ambition was that of efficient communication. Presenting the subject in a compact and understandable way, with sufficient pathos, was seen as a precondition for well-functioning democracy. Without well-formulated speeches, citizens would not be willing to spend time to listen to the arguments and take part in the deliberations (Andersen 1995).

Empty rhetoric?

The main problem of any form of rhetoric is that it may conceal more than it discloses. In the Pnyx, and probably in the Agora as well, accusations of empty rhetoric would primarily refer to distortion of facts, by hiding the speaker’s own intentions, or by seducing the people to make unwise decisions, for instance to declare war on purely emotional grounds, without understanding the consequences following from them.

The Work Collective presents an interesting contrast to this picture. In the deliberations of the General Meeting it would be hard to find empty rhetoric in the sense above. I know of no example of members trying to bluff the others out of self interest, or consciously hiding relevant facts in an ongoing discussion. Any motivation for doing so was absent, as hidden rewards or bribery were simply unfeasible. At the same time, it would be strange to overlook some mechanisms related to the problem of seduction, albeit on a somewhat different level.

Formulation of common goals

In an obvious sense, the Work Collective has a set of common goals – it is justified as a political endeavor. But all members do not relate to its goals in the same way and with the same competence. Most members take a rather pragmatic view of these goals. A few are more active in formulating, interpreting and reformulating the common goals. This does not necessarily stem from a desire for dominance. Political situations change continuously, and call for reinterpretation. And some members are better at formulating the goals than others, due to personal skills as well centrality of position in the production process. Usually, these two factors are closely related. Competence yields power through communicative skills. In reality there are only a few people who exercise the power of formulating the common goals. Thus, a hierarchy of persuasion may obviously be discerned in the Collective, even if the most powerful in the General Meeting have nothing like full control over decisions.

But why consider it as a hierarchy of persuasion, and not simply of argumentation? To the degree that a consensus exists that there is one politically correct solution to a given problem, there is no difference between persuasion and argumentation. The question arises when there are alternative problem formulations that are not taken seriously, either because they are suppressed, manipulated away, or simply not perceived.

Definition of the situation, agenda setting, discourse

These points lead up to a closer discussion of the concept of discourse (Foucault 1972) and related phenomena, such as the definition of the situation (Goffman 1959) and setting agendas (Bachrach and Baratz 1962). What these different approaches have in common is a focus on themes excluded from communication. Such processes of exclusion may be the result of deliberate action, or they may be regarded as the consequence of contradictory conceptions, wishful thinking or the like.

Bourdieu’s ideas of symbolic power posit as axiomatic that the powerful are the ones who gain from contradictory conceptions. Power holders shape their norms in such a way that it serves their own power (Bourdieu 1991). In the following I take the opposite view. The ideals constitutive of the Collective, i.e. its ideals of equality, are real enough and are shared by all in an non-strategic way. However, the adherence to these ideals entails a distorted view of some of the basic problems confronting the Collective. The main problematic in this connection is not that of problem formulation, but of problem avoidance.

Lack of authority – lack of communication

Let me start by returning to one of the points above, problems of authority and delegation. From the perspective of those who should be in charge of the problem, this is a question of authority. Seen from the other side, the problem may be formulated as one of definition of the situation. Goffman (1959) describes what goes on in the definition of the situation as a process of distinguishing between themes that are accepted and tabooed. Participants in a conversation have a common interest in setting up defenses against invasion into their private sphere. What happens in the Work Collective, is that the norm of equality becomes the point of reference, assigning who has a right to define the private sphere. If everyone is equal, nobody has the right to transgress the boundary of another person.

In principle, it is possible to imagine the opposite solution: complete openness. Equality gives everyone the right to break into the private spheres of the others. This would be in line with the rationalist, anti-taboo thinking prevailing in parts if the Left, also in the 1970s. ‘Politization of the private’ was one of the slogans in discussions on sexual relationships and the death of the family. However, in the Collective, which was a work organization and not a commune, this would not have been viable for long.

Similar processes of exclusion lie behind the evasion of long term organizational problems, likewise described above. The situation was not that these problems were discussed, and then rejected by a malcontent majority who refused to take them seriously. Some challenges were not formulated before it became impossible to overlook them. Others never reached the agenda, for similar reasons. Proposing solutions that would make some members disadvantaged compared to the rest, would imply breaking with the norm of equality. The main problem was that of formulating problems.

Tradeoffs and compromises

In a very strong sense, the Work Collective was a small socialist island in an enormous capitalist ocean. Despite being an anti-capitalist organization it had to find a modus vivendi with capitalist reality. But how to do this without becoming schizophrenic? The obvious solution would be to work out some sort of compromise, defining a balance between economic necessities and ideological and political goals. But it is very difficult to give such a compromise sufficient stability. First, it is difficult, but in no way inconceivable, to find a trade off between economic gains and political success, simply because they are incongruous. But when the goal of the organization is to fight capitalist economy, a compromise becomes virtually impossible. For, any attempt to make the company economically viable will entail the possibility of sliding toward the profit motive, which in itself would be unacceptable. When this is realized, the theme is excluded from further discussion.


One of the main points in the foregoing discussion has been to point out possible sources to the discrepancies between the ideals of equality and the existence of an informal hierarchy. Now, even if these discrepancies did exist, in themselves they were no serious threat to the functioning of the organization. The problem was not so much the discrepancies as the inability to admit their existence and to handle them. There is no reason to assume that the members of the Collective were more prone to self-deception than most other people. The basic problem was that they had no language for interpreting the existence of inequality, and trading a moderate amount of inequality off against necessary efficiency.

The result was a widespread confusion and frustration with the decision-making system. This was demonstrated quite dramatically in questionnaire interviews made at the end of the 1970s, relatively shortly before collective governance was given up. There it became clear that a majority of members of the Collective, despite their strong adherence to the norm of equality, wanted to give the organization a more hierarchical structure, and not less. However, no consensus existed, neither concerning the causes of present troubles nor the alternatives to the existing order. Without the existence of a credible alternative, the collective governance could not be reformed, only abandoned.

Ideals and reality – in practice

In some instances, the problems of equality were exposed and concretized in a striking way. In the daily routines equality was taken for granted, even if nobody could avoid noticing the informal inequalities that were present at the same time. These discrepancies could not be discussed openly, because that would break with the basic value of the collective. But normally the two conceptions could coexist without great problems, thanks to the tact and politeness exercised in the common definition of the situation.

But suddenly the clash between the formal and the informal structure could be exposed, if only for a brief instance. Somebody might be making an inappropriate remark, or being unaware of the consequences of a given suggestion. The resulting silence could be massive for a moment – as the saying goes, an angel went through the room. The taboo of the Work Collective came to the fore, perceptibly, while at the same time it could not be expressed.


The point of departure of the present discussion is a sketch of some internal problems of direct democracy of a strictly egalitarian brand. By distinguishing between factors connected to efficiency questions and those related to communication, a more complex understanding of power differentials becomes possible.

Efficiency and communication stand in a paradoxical relationship in the case studied. In order to cope with efficiency problems, the Collective developed a moderate hierarchical structure, which in itself did not signal a clear break with the egalitarian norms. However, in combination with the development of an informal power structure based on positions, personal capacities and ideological expertise, a general ambiguity emerged in the organization. This ambiguity became a contributing factor to the later demise of collective governance, because communicative patterns in the collective, rooted in the egalitarian ideals, showed themselves to be inadequate for coping with some of the basic problems confronting the Work Collective.

Admittedly, what is presented here is a single case study, with limited transference value. Nevertheless, several mechanisms discussed are quite common, and in this respect the case study may be of a more general value.

Moreover, I believe some of the observations made here may have a bearing on theoretical reflection. One is related to power and intentionality. In other writings I have insisted on the salience of intentionality to the interpretation of power. When talking about power, analysts should first look for intentions, and if possible avoid the notion of ‘structural power’ (Engelstad 1999). But what I have described here is to a large extent the relationship of power to unintended social patterns. First, by insisting that the power differentials resulting from the emergence of hierarchy and the division of labor were not the result of somebody’s purposive action. Second, by pointing to powerlessness, rather than power, as a major outcome of many of the communicative patterns in the Collective. In neither case intentions of domination played an important role.

This opens up for a precision of earlier discussions. The first point has to do with the relationship between emergence of a power structure and the exercise of power within the positions of that structure. What has been amply demonstrated is that the emergence of power structures may be the unintended consequence of actions pursuing quite different goals. But the exercise of power within these positions, in other words the use of resources generated by those positions, was of course intentional. When editors persuade the General Meeting to accept to include a certain title in the publishing program, they are undoubtedly exercising power, as a consequence of their greater credibility and competence within the field. What should be avoided, however, is the implicit assumption made by e.g. Bourdieu (1977), that an automatic connection and a homological set of interests exist between the emergence of the power position and the actions performed within it.

Moreover, notions of discourse usually take it for granted that they serve those who adhere to them, and above all those who interpret and defend them on ideological grounds. The present analysis has demonstrated that this need not be so. The dominant discourses in the Collective turned out to be a source of powerlessness, an impediment to efficient action, which would have been in the interest both of the Collective as a whole and its informal leadership.

On a more general level, the presentation invites a reflection on the notion of ideal speech situation. Two types of speech situations have been described above, the assembly and the conversation in the small meeting. In both, powerful communicators dominate the scene. It is a basic tenet in discourse ethics that all affected interests should be represented. However, experience makes it implausible that all interests can represent themselves equally well. And when the parties are unequal with respect to ability to formulate own interests, the ideal speech situation can hardly escape paternalism. Formulating the interests of the weaker will virtually always be left to some of the stronger parties.

Excursion: Power distribution by lottery?

A salient difference between ancient Athenian democracy and modern variants, lies in the use of lottery. In Athens virtually all positions were filled by drawing lots. Later in history, lottery was used in ancient Rome, medieval and renaissance Italian city-states, British parliamentary elections in the 16th century and in Switzerland up to the 19th century (Engelstad 1989). It may seem a little far-fetched, but I believe it is perfectly legitimate to ask why the use of lottery was out of question when filling positions in the Work Collective. By drawing lots Collective members would have instituted a mechanism for assigning people to positions that would have guaranteed the norm of equality in a mechanical and very simple way. As the matter was never discussed, the answer is necessarily speculative, but not necessarily uninteresting. I see three setsof relevant reasons for the rejection of lotteries.

First, rationality. A lottery employs other mechanisms than those of rational deliberation, and the consensus justified by rational arguments. Deviations from rationality accepted by ancient Athenians are felt as deeply problematic by most modern citizens.

Second, the weight accorded to the selection of talent. Lotteries have no traditional place in modern political organizing, where the weight on talent is overwhelming.

In principle, none of these arguments would necessarily be defended by members of the Collective. At the outset at least, the selection by talent was something to be avoided, for with relevant training anyone was expected to take charge in the Collective. Moreover, considerations of rationality based in recent political history hardly would be persuasive to people seeking untraditional forms of organization in order to pursue norms of equality.

Consequently a third reason may be added. Instituting lotteries would lead to less flexibility, something that would have forced the contradictions of the Collective out in the open. The absence of flexibility also makes it difficult to combine lottery with the free consent of being candidate for office. Variations in the capacity to assume responsibility in office probably made it necessary to introduce some use of discretion into the mechanics of sortition. This was possible in a polity as large as Athens, but would have been disastrous in a small group such as the Collective.


Andersen, Øivind (1995), I retorikkens hage. Oslo: Universitetsforlaget.

Bachrach, P. and M Baratz (1962), "The two faces of power." American Political Science Review, 57:947-952.

Bourdieu, Pierre (1977), Outline of a theory of practice. Cambridge: Cambridge University Press.

Bourdieu, Pierre (1991), Language and symbolic power. Cambridge: Polity Press

Engelstad, Fredrik (1990), Likhet og styring. Deltakerdemokratiet på prøve. Oslo: Universitetsforlaget

Engelstad, Fredrik (1989), "The assignment of political office by lot". Social Science Information, 28:23-50.

Engelstad, Fredrik (1999), "Maktbegrepet etter Max Weber". In: F. Engelstad, ed., Om Makt. Teori og kritikk. Oslo: AdNotam Gyldendal.

Engelstad, Fredrik (2001, in press), "Theories of power in communication. A critical assessment." In J. Gripsrud and F. Engelstad, eds., Power, Aesthetics, Media. Report, Makt- og demokratiutredningen. Oslo: Unipub.

Foucault, Michel (1972), The Archeology of Knowledge and The Discourse on Language. New York: Pantheon.

Goffman, Erving (1959), The presentation of self in everyday life. New York: Doubleday

Habermas, Jürgen (1984), Theorie des kommunikativen Handelns. Frankfurt: Suhrkamp Verlag.

Ober, Joshua and Barry Strauss (1990), "Drama, Political Rhetoric, and the Discourse of Athenian Democracy". In J.J. Winkler and F.I. Zeitlin, eds., Nothing to Do with Dionysos? Athenian Drama in Its Social Context. Princeton: Princeton University Press.

Simon, Herbert (1957), Models of Man. New York: Wiley.

Williamson, Oliver (1975), Markets and Hierarchies. New York: The Free Press.

Av Fredrik Engelstad
Publisert 25. nov. 2010 13:52 - Sist endret 5. nov. 2013 13:20