The early classics of elite theory - The Ruling Class by Gaetano Mosca (1939; originally 1883), Political Parties by Robert Michels (1915; 1911), and The Mind and Society by Vilfredo Pareto (1935; 1916) - all date back to the turn of the century. The best-known successive works within this tradition were published during the 1940s and 19505: The Managerial Revolution by James Burnham (1941); Community Power Structure by Floyd Hunter (1953); and The Power Elite by C. Wright Mills (1956). However, elite theory continues to provide a useful tool for the analysis of social power structures (see e.g. Etzioni-Halevy 1993). The argument here is not that classical elite theory should be taken to represent a scientific truth, but that it can help to generate relevant new hypotheses. Since the mid-1970s elite studies have indeed become increasingly popular in the field of research concerned with power structures (Moyser & Wagstaffe 1987, 5).

In recent years, blueprints for a European Union have given rise to widespread concern about the future of democracy. In particular, some people seem to fear that the transfer of national decisions to a single supranational bureaucracy will lead to a concentration of power and make it even harder for local people to be heard. In Finland many critics argue that if the country joins the Union, it will probably lead to the most

Resumé

The article discusses the main results of a study of the power structure of Finnish society. The purpose of the study was to investigate whether Finland is controlled by one closed, cohesive and unanimous power elite, by several rival elites, or whether, instead, the power structure is so decentralized that there is no justification for referring to a power elite in the first place. The working hypothesis was that Finnish society is controlled by one power elite. The study was based on a positional approach, using a survey questionnaire to collect data both from members of the power elite and from the population at large. The main conclusion of the study is that, with certain reservations, it is legitimate to speak of a fairly cohesive and unanimous, single power elite in Finland. Interaction among elite groups is close, and their attitudes are similar. In terms of recruitment patterns the Finnish power elite is fairly closed: the top social stratum is clearly overrepresented and the lowest underrepresented. On the other hand, no single stratum has a dominant position in the power elite.

profound changes in its power structures since independence was achieved
in 1917.

At the same time, popular attitudes towards politics and politicians, the traditional vehicles of democracy, have become more and more critical in Finland. According to a recent survey by Suomen Gallup, public confidence in parliament has declined sharply since the mid-1980s, whereas attitudes towards other institutions have remained more or less unchanged (Helsingin Sanomat, 5 April 1992). A deep rift has emerged to separate the people from top decision-makers, and it seems that the rift is getting deeper (Elinkeinoelåmån Valtuuskunta 1991). There are also indications of an ongoing concentration of ownership in various sectors (e.g. private enterprise, and insurance, mass communication), and this is known to strengthen similar tendencies as far as power structures are concerned.

The question that presents itself in this situation is, whether the people who occupy top positions in different sectors of Finnish society form a closed, cohesive and unanimous power elite which controls the life of the people independent of democratic rules and procedures? This article presents evidence relating to this issue. The article is based on results of a study that was carried out as part of an ongoing research project concerned with the power structures of Finnish society (Ruostetsaari 1992).

The concept of a power elite employed in this study follows in the tradition of C. Wright Mills (1956) but departs somewhat from that used by Mills. Mills's conception of a power elite is essentially a structural one: for Mills the power elite consists of people who occupy leading positions. Whether or not these people make decisions that affect the lives of others is secondary to whether they occupy positions with decision-making authority. The focus, in other words, is on formal rather than actual power. In this study we do not subscribe to the theory of structural determination: although social actors are constrained in their activity by certain structural limitations, they nevertheless are perceived to have relative autonomy to choose between different options (Lukes 1977, 6-7; see also Sweezy 1971, 131-132).

The Research Problem

As noted above, the purpose of the study was to investigate whether Finland is controlled by one closed, cohesive and unanimous power elite, by several rival elites, or whether the power structure in the country is so decentralized that there is no justification for referring to a power elite in the first place. The working hypothesis has been that Finnish society is controlled by one single power elite, i.e. that the power structure is centralized.

What are the implications if we find that Finland really is controlled by a single, cohesive power elite? Most importantly, a society that has such an elite can hardly be described as democratic in the classical meaning of the word. If, on the other hånd, there are several rival elites, then the conclusion is less obvious. A study that explored the power structures in Sweden (SOU 1990:44, 303) suggested that pluralist democracy presupposes society where there are several open elites. In this situation the majority of the people can exercise power by changing the ruling group for another one. In such a view, the essential thing about democratic government not the absence of elites but rather the existence of several rival. open elites. Through its adherence to this line of argumentation, the Swedish study endorses the key tenets of what has been termed democratic elitism (see Schumpeter 1959; Sartori 1962).

closed when its members are recruited from one single social class or stratum. An open elite has a broader recruitment basis, i.e. its members include women and men representing different social classes and different geographical regions. As to cohesiveness, the primary concern has typically been the pattern of continuous mutual communication. shared formal positions, and informal contacts. If there are several elites in society. then the power structure is determined by contacts both within elite groups and between them. by the network of interaction. In extreme cases elites are formally separate and independent of each other, but they have such close contacts that in actual faet they represent one single elite. Finally, with respect to the third characteristic. unanimity, the focus has typically been on the congruence of opinions, values and attitudes within the elite. The culture and values that tie the elite together may be more or less explicitly formulated. Applying these notions, we may conclude that there exists in Finnish society a power elite if the different sectors in society constitute a solid, cohesive bloc in the sense that recruitment into elite positions is selective and favours people in top social groups; if there is close interaction between the different elites; and if the opinions, attitudes and values of the people occupying top positions in these sectors are more or less consistent and differ from the opinions, attitudes and values of ordinary people. The hypothesis in the Swedish power study was that the country is controlled not by one or several separate power elites, but by two elites or, more correctly, by two major power blocs, i.e. economic power and political power. The growth of these two blocs reflects specific historical characteristics of Swedish society. Around the turn of the century there occurred two important breaks in the continuity of Swedish society. The first break was the separation of economic and political power: traditionally, the right to vote and eligibility for office had in Sweden been connected with economic standing. The second break was contemporaneous with the emergence of two new elite groupings, each of which specialized in its respective power sphere. The new economic elite was not deeply disappointed the loss of political power because it had never really possessed such power, no more than its precursors had. Results of the Swedish study supported this hypothesis (SOU 1990:44, 305-306).

The Structure of Finnish Society

Given the results for Sweden, one may ask if it is justifiable to start with the assumption that there is just one power elite in the Finnish case? Finland and Sweden are, after all, very similar modern societies, and they even share a common history. In addition, all Scandinavian countries, and

Sweden and Finland in particular, have traditionally had a large peasantry. which effectively prevented the growth of feudalism (Konttinen 1991. 76). Finnish society has always been characteristically bureaucratic and statecentred. Swedish rule, the Finnish nobility had little landed property and was therefore heavily dependent upon holding officers' posts in the army. Later, during Russian rule, when the army was abolished, the nobility was dependent upon holding civilian posts in public administration (ibid.. 92, 120). Thus, it was under Russian rule that the Finnish power structure and political culture began to develop in a different direction than in Sweden (see Ylikangas 1986, 111-112).

elites formed a relatively cohesive power elite in Finland (Allardt & Wiatr 1978, 66-67). Several factors were instrumental in creating and strengthening a sense of national consensus and in this way promoting the growth of a cohesive power elite in Finnish society (see, e.g., Salminen 1983, 3-4; Salminen & Haarala 1986; Ruostetsaari 1989, 210-212; Tiihonen 1990, 211-212; Hyvårinen 1990, 263). Different elites have been united, on the one hand, by social integration and, on the other hand, by the tradition in Finnish society of relying heavily on the state. Through its various regulation and support mechanisms, the state is integrated into civil society in many different ways and in many different spheres (Sånkiaho 1992, 20). From the very outset in 1968, for instance, an incomes policy adopted in Finland assumed more institutionalized and state-centred forms than in the other Nordic countries (Kosonen 1989, 173-190). Even the financing of art and culture is now dictated by central government through art administration (even though this operates on a corporatist basis). Finally, one cannot forget that the tendency towards concentration has affected not only public administration but private bureaucracy as well. Virtually all organizations and major business concerns have their head office in the capital city of Helsinki. It is obviously much easier to establish and maintain contacts at the personal level in this situation, which has the added benefit of providing several fora for such contacts. Many top decisionmakers, example, are members of several different organizations. It has been argued that communication and coordination among elites is easier if these are relatively small in size (Porter 1968, 211). The old adage that Finland is a small country where everyone knows each other, certainly holds true in this respect.

Composition of the Finnish Power Elite

Mills states that the power elite consists of people who occupy top positions in three institutions, i.e. the political directorate, major business corporations the Armed Forces. Although Mills speaks of institutions, it would in fact be more appropriate to speak of sectors, which cover broad sections of society (Bell 1971, 194). For purposes of the present study the assumption is that Finnish society can be divided into separate spheres or sectors, each of which have their own internal institutional structure. We do not argue that these sectors are independent of each other; on the contrary it is assumed that they are linked together through interaction and various interdependencies. Further, it is assumed that the models of communication and interaction may vary between different sectors. Finnish society is divided not into three but seven main sectors, which are further

divided into a number of sub-sectors. The main division here is identical with the one applied in the Swedish project (SOU 1990:44). The intent has been to investigate, first, whether there is close interaction between these sectors or whether they are separate spheres and, second, whether the people who occupy top positions in those sectors form a cohesive and unanimous power elite.

The method applied in this study is based on the positional approach (see Hoffman-Lange 1987, 30): a hypothetical power elite is constructed on the basis of the top positions identified in the main sectors. In the absence of any objective criteria for drawing a line between the top elite and the (ordinary) people, the definition of a formal elite necessarily involves some measure of subjective judgement. The definition of a formal elite adopted here is a fairly narrow one: the focus is on the very nucleus of power as we are looking at 1115 positions that are occupied by 997 persons. The difference in these two figures is explained by the fact that some people occupy more than one position in the power elite. The concern of our study is with top institutional positions in the respective sectors rather than with the individuals occupying those positions. This means that those people who occupy several elite positions in different sectors carry more weight than those who occupy only one position. The method of analysis precludes any consideration of the cumulation of internal positions within the sector. The composition of the power elite is outlined in Table 1.

A questionnaire was mailed to all people occupying institutional positions within the power elite as defined in Table 1 in November 1991. A total of 667 persons responded, yielding a response rate of 67 percent (calculated by individuals) and 63 percent (by positions). The corresponding figures in the study carried out in Sweden were 58 percent and 59 percent respectively ¹ By international comparison the response rate of the present study is probably higher than average (see SOU 1990:44. 315).

Recruitment to the Power Elite

Using the distinction and terms suggested by J. P. Roos (1987), presentday is ruled and governed by two more or less equally powerful generations: the generation of post-war reconstruction (born between 1927 and 1939), and the generation of the great transformation (born between 1940 and 1949). The Finnish power elite is, on the average, much older than the rest of the population; by far the largest generation in the total Finnish population is the suburban generation, i.e. people born in 1950 or later. Among the elites, this is the biggest generation in only one group - the political elite. The political elite is indeed clearly younger than the

other elites: the average age in this group is 43 years, while the figure for
the whole power elite is 51 years. On average, the Finnish power elite is
aiso two years younger than the power elite in Sweden.

sectors. Access into the political elite, where women currently constitute 33 percent, appears to be easiest. Women are also well represented in the cultural elite, occupying roughly one-quarter of all positions. By contrast. women are virtually excluded from the business elite, where they occupy no more than 4 percent of all positions. Furthermore, there is only one woman in the corporate elite proper. Women are very much underrepresented all other elite groups as well: organizations 12 percent, mass media 11 percent, science 8 percent and administration 5 percent. The biggest difference between Finland and Sweden in this respect is that the proportion of women in the administrative elite is more than twice as high in Sweden than in Finland (11 percent). It is important to stress here, however, that while all these figures for women's representation in the power elite are low, the picture is very different when we compare them with statistics for other countries. Women's representation in Nordic parliaments, example, has traditionally been one of the highest in the whole world.²

Although the statistics above lend support to the hypothesis that there exists a distinct power elite in Finland, recent trends indicate a slow change in the patterns of regional recruitment. The proportion originating from Uusimaa is significantly lower in the younger generations, dropping from over two-fifths among the oldest members of the power elite to less than one-third. The only county where there has been a linear increase from the youngest to the oldest generation in terms of the recruitment of power elite members, is the province of Oulu in northern Finland. In general, the vast majority of Finland's power elite comes from urban areas. For people who spent their youth in rural areas, the political elite has provided much easier access to upward social mobility than other elite groups. It may also be noted that there is a slight overrepresentation in the Finnish power elite of people coming from a Swedish-speaking family. Also, the basic education of the parents of power elite members is higher than average. A clear inheritance effect is in evidence in the case of cultural capital: the higher the parents' level of education, the higher, too, is the education of the power elite member. Again, upward social mobility through the political elite is easiest; the parents of those recruited into this group have a lower level of education than members of other elite groups. In terms of political views, the Finnish power elite differs quite clearly from the rest of the population. Most of the power elite's votes go to the right-wing Coalition Party (28 percent), followed by the Social Democratic Party (18 percent) and Centre Party (11 percent). Comparing the questionnaire with the 1991 parliamentary elections, the two parties with the highest "overrepresentation" in the power elite are the Coalition Party (+9 percent) and the Swedish People's Party RKP (+3 percent) whereas the Centre Party is most clearly underrepresented (-14 percent). The total support for socialist or left-wing parties in the Finnish power elite in 1991 was 10 percentage points lower than in the parliamentary elections held the same year. This pattern of support for political parties can at least partly be explained by the overrepresentation of southern Finland in the power elite. Party preferences in the power elite are largely inherited. The occupational status of parents also helps to explain the party preferences of people who have moved up into the power elite. In terms of social background, the Finnish power elite is fairly upper class, although probably less so than in Sweden (see SOU 1990:44, 320).³ Compared with the whole population, descendants of blue-collar workers and farmers are clearly underrepresented. This supports the hypothesis that there exists an exclusive power elite that is closed to lower social groups in Finland. A common feature shared by both Sweden and Finland, however, is that the political elite has provided a more open avenue for upward mobility for the children of blue-collar workers and farmers than other elite groups. The organizational elite comes rather close to the

political elite in terms of its social background and openness. The third most open elite group has been the administrative elite, although it follows at quite some distance. The combined proportion of blue-collar workers' and farmers' children is fourth largest in the science elite, but here the number of farmers' children is twice as high as that of blue-collar workers' children.

In the business elite and in the media elite the proportion of people coming from farming or blue-collar background groups is roughly the same. However, while the media elite has been equally open to both strata, the business elite has twice as many children of farmers than children of bluecollar In fact the business elite has been the most closed group to children of blue-collar workers, and, accordingly, the representation of the top stratum in this elite has been higher than elsewhere. Yet there are interesting differences in the openness of different sub-sectors within the business elite. Children of blue-collar workers and farmers have moved up into the power elite more often through state-owned firms and cooperatives than through private business corporations.

The social structure of the media elite comes rather close to that of the business elite. The culture elite seems to be more closed than other elite groups: children of blue-collar workers and farmers account for only onefifth the culture elite. While the business and science elites have been closed to children of blue-collar workers, children of farmers have been unable to move up into the culture elite (Tables 2 and 3).

However, if we look at how the situation has developed from generation to generation, then it is clear that the power elite has in fact been opening up to the lower social strata as well. Within elite members in the youngest generation, there is a much more balanced representation of different

social strata as defined according to fathers¹ occupation than in the oldest generation. The increased social openness of the power elite is evident in the fact that the combined proportion of the two lowest social strata grows consistently from the oldest to the youngest generation (although this increase is more linear in the case of children of blue-collar workers than children of farmers). By contrast, it would seem that the middle classes (i.e., upper and lower white-collar employees) are losing their position in the power elite, where their numbers decrease from the oldest towards the youngest generation - which is exactly the opposite of the situation in the total population. The upper class, on the other hand, appears to be experiencing something of a comeback: in the youngest generation the proportion of this stratum is the same as that in the oldest generation, after dropping off in the two middle generations. This is explained by the fact that the proportion of children of self-employed entrepreneurs in the power elite decreases from generation to generation, while at the same time the number of hired managers in top positions increases (Table 4).

Finally, as may be expected, income levels in the power elite are substantially than those in the total population. There are marked differences even between different elite groups: the business elite has by far the highest and the political elite the lowest salaries. Those in the top income bracket are typically 45-year-old male top business managers who vote for the Coalition Party or the Swedish People's Party and who live in southern Finland. However, Finnish business managers do not earn as much as their foreign colleagues.

Networks of Power

Apart from exclusive recruitment, another important condition for the existence of a distinct power elite is that there is sufficient cohesiveness within and between elite groups. Domhoff (1990), for example, underlines the key importance of cohesion, which is created by a shared value system and by social and institutional relations that serve to maintain that system. The discussion that follows will focus on the interaction that helps to sustain the cohesiveness of the Finnish power elite. Communication and interaction networks can be regarded as mechanisms that promote and strengthen attitudinal cohesion through processes of social comparison (Knoke 1990,

is to analyse contacts of members of different elite groups with different
institutions rather than find out direct contacts between members of different

actors. Interaction with central actors (i.e., actors with links to several organizations) presumably contributes more to the power of the actor than contacts with isolated organizations. Contacts with central organizations provide the actor with access to significant communication fora, which may serve to strengthen his or her influence in the network (Bonacich 1972, 177; see also Barnes 1979). In such a context, prestige concepts maintain the asymmetry of interaction: the actor may be defined as prominent if he or she receives several contacts but sends out few contacts. In this light, in other words, the quality rather than just the quantity of interaction is recognized as important: the weight and significance of the actor's contacts at least in part determine the weight and significance of the interaction itself. Knoke (1990, 13) goes so far as to argue that the centrality of the actor in the communication network is synonymous with the actor's power and influence. This prestige aspect is particularly relevant in the analysis of the power possessed by different positions in a network where orders or material resources are not exchanged on a reciprocal basis. The members of a cohesive group or a clique are in direct contact with each other through many intensive mutual communication connections. The more intensive and the more frequent the communication, the more likely it is that the members of the group shaie and foster the same social and political values and attitudes. Frequent interaction serves to socialize the group members to a common normative understanding with regard to collective interests (ibid., 11-12).

Structure of the Interaction Network

With this in mind we may now look at the positions occupied by different actors in the power elite as well as in the interaction networks of different elite groups. This is done by studying an index created on the basis of responses to the question concerning frequency of interaction. The three response alternatives were scored on a 0-100 range, with the mean being treated as "frequency" and the mean based on those respondents who score above 0 being treated as "intensity". This index is in fact a sum variable that takes into account both the frequency of interaction and its intensity.⁴

government; and 31 percent that their interaction with the government was of a more general nature. If the arrow were displayed in the opposite direction, it would mean that the political elite thought the government had chiefly operated in the role of disseminating information. A majority of contacts determines the principal direction of interaction concerning the institution. An asterisk after the index score symbolizes more general exchange of information or social interaction. This option is included for the reason that in many cases it is impossible to identify the main direction of interaction. A typical example of this type of interaction is the situation where a corporate manager and a politician meet briefly during a break at a conference and exchange views on some topical issue. As a rule neither party will consciously to influence the other in this sort of situation, even though the exchange of information may in fact influence decisions rather than merely serve the purpose of social interaction. If there is no symbol after the index score, there is no definite and explicit direction of interaction, but two or three types appear to be equally strong. The institution that occupies the most central place in the interaction network (with an index score of 80) is the mass media. The centrality of this institution is further underscored by the fact that no other institution reaches even the next index level of 70. The second most central institutions in the interaction network are business firms and banks, both of which obtain an index level of 60. At the next level we find the cabinet and the university system. This group of five institutions can be said to form the inner core of the power elite's interaction network, which is in a position to control information and by the same token to exert at least influence if not power in important political and social issues.

It was observed earlier that an actor occupies a position of prominence in a network if it receives several contacts but sends out few contacts. Out of the five institutions listed, the mass media and the government clearly qualify as prominent actors that are kept informed by the power elite. In the case of the three other core institutions, it is impossible to identify any predominant direction of interaction. About one-third of the members of the power elite usually inform a business concern (in which case the firm is the receiver of information) while another third receive information from business concerns; the remaining third have interaction of a more general nature with private business firms. However, private business firms are somewhat more often receivers of information rather than senders, with banking institutions providing an exception to the rule: banks disseminate information slightly more often than they receive information. The interaction goes on between the power elite and the system of higher education is typically of a more general nature.

All elite groups with the exception of administration, organizations and

culture, are represented in the inner core of the power elite. This means that even the core of the interaction network, the five most central institutions, integrate the different elite groups to a certain extent. This integrative effect expands to cover all elites if we also include in the nucleus the next level of the network. At an index level of 40 we find a fairly large number of actors: offices and public services, local communes or federations of communes, wage-earner organizations, other research institutes, trade and commercial organizations, state-owned firms, national party organizations. employer organizations, the Ministry of Education, the Ministry of Finance. and cultural organizations. "Offices and public services" refers here to offices other than ministries, and "other research institutes" to institutions other than universities. Apart from the fact that this "expanded inner core" includes institutions from all elite groups, it is also noteworthy that the corporative system has a very strong representation. The inner core also has a regional dimension: the interaction network includes communes and federations of communes. Further, the position of the Ministry of Finance near the centre of power is hardly surprising, but the Ministry of Education's position is.

Most of the political elite's contacts are with institutions within the political sector. The inner core of the interaction network is formed by the party's parliamentary group, national, district and local party organizations. mass media, the cabinet, parliamentary committee and commune or federation communes. With the exception of the last-mentioned institution. the political elite is itself the source of information; that is. the direction of information is chiefly from the political elite towards the core institutions. By contrast, communes or federations of communes usually inform the political elite. Actors from other than the political sector are only found on an index level of 50 — i.e. the Ministry of Education, the Ministry of Social Affairs and Health, the Ministry of the Environment, offices and public services, wage-earner organizations, private business companies, banks, other associations or organizations. In other words, the inner core of the political elite integrates all other elites with the exception of the science and culture elites. In the interaction between the political elite and institutions at this level, however, the political elite is primarily at the receiving end of information. An interesting curiosity here is that the President appears to be a fairly distant actor both in the whole power elite's and in the different elite groups' interaction networks. The interaction and exchange of information between the power elite and the President is chiefly of a more general nature.

Network Cohesion

Of further interest is the cohesiveness of different interaction networks.

We can begin by looking at any possible overlap that may exist between
different interaction networks and their inner cores, and then move on to
examine the cohesion of networks at a more general level.

and the science elites have fairly limited interaction with the political, administrative, business and organizational sectors. On the other hånd, in the light of this analysis - which eliminates potential effects of the size of main sectors - the mass media does not appear as a central actor in the power elite's interaction network as in the foregoing analyses. The description of the Finnish power elite's interaction networks up to this point has been based on data about contacts and the direction of those contacts provided by the persons involved. However, the cohesion of interaction networks can also be examined on the basis of a different, more "objective" set of data. As we saw earlier, the Finnish power elite comprises a total of 1115 positions occupied by 997 persons. This means that some people occupy more than one position in the power elite. The number of individuals occupying at least two elite positions either in their own or some other sector is 91, which is 9 percent of all members of the power elite. There are 40 people who occupy a position in at least two different elite groups. This is 4 percent of the power elite. The "strong links" carried by these top managers are of course just the tip of the iceberg of intersectoral overlap. Given the narrow focus and definition applied to the power elite here, it is clear that these 40 individuals have quite considerable power and influence in at least two sectors of social life. Proceeding further, there are just 17 men in Finland who belong to what may be termed the absolute core of power in Finland, that is, men who occupy at least three elite positions in one or more sectors. Out of this group the majority (10) work full time in the business sector, three in the science sector, and two in politics and organizations. In this centre of power there are no elite members who represent the administrative, media, or culture sector. The key individual figure in late 1991 was the director general of Kansallis-Osake-Pankki (a major merchant bank), Jaakko Lassila. who occupied seven elite positions. Even this nucleus of the power elite, it should be noted, is not a hermetically sealed stratum that is completely insensitive to external pressures. In just over six months from November 1991 to the summer of 1992, one of these 17 top decision-makers had been appointed director general of a major corporation, one had changed jobs and at the same time moved from one elite group to another, two had retired (one voluntarily) and one had been dismissed.

As well as frequency, intensity and content, interaction processes also have a shape or form. The broad consensus of opinion among our respondents that the most important form of influence in the endeavour to further one's interests is represented by personal, informal contacts, followed telephone calls and by the mass media. Contacts and official procedures through the politico-administrative system, and particularly through parliamentary committees, are thought to be of no real significance.

That mass communication is regarded as a fairly central form of influence

is consistent with the fact that this institution is a central actor in the power elite's interaction network. In particular, the political elite underlines the role of the media as a form of influence. This finding serves to emphasize what has been termed the "medialization" of politics (Asp 1986); that is, the ways of doing politics are adjusted and adapted according to the requirements of publicity. And it seems that politicians have indeed adjusted and adapted: in the political elite there are fewer people than in most other elite groups who feel that journalists have too much influence on public opinion formation.

Three-fifths of the members of the Finnish power elite have been part of some informal contact group. This sort of communication is most common in the political and business elite and least so in the media elite. The most common type of informal grouping is one that is organized around a certain profession or occupation, although there are also many different types of conversation groups and political contact groups. In fact it would seem quite legitimate to speak of a politicization of informal interaction in the sense that membership of a political group increases linearly from the oldest towards the youngest generation. It is also believed that informal contacts really make a difference. More than nine out of ten members of the power elite subscribe to the view that informal personal contacts have a major impact in the exercise of social power. This belief is strongest within the political elite. It also varies with age: the younger the generation, the more common is the view that informal personal contacts matter.

Conceptions of Influence

To explore how far the power elite in Finland is unanimous in its opinions, we may compare the views of both the power elite and the population on how influence in society is divided. It is important to stress here that the analysis is concerned with images of social influence rather than with the actual distribution of influence. The more consistent the power elite is in its views, attitudes and values, and the more it differs in these respects from the population at large, the stronger the support for the power elite hypothesis.

On the power elite's list the President of the Republic ranks as the third most influential institution in the country, while the people rank the President much lower at ninth. The power elite also has slightly more faith in the influence of elected officials in communes and in political parties than citizens. The views of the power elite and the people on the role of parliament, on the other hånd, are more consistent: the former rank it twelfth and the latter thirteenth. The people and the power elite are also in agreement in ranking the mass media as the secorid most influential bloc in society. The power elite believes that television and radio are the second most influential institutions and the print press the fourth most influential.

hypothesis is also supported by the fact that the power elite is fairly cohesive in terms of its views on the distribution of influence in society. By contrast, there are no marked differences between the views of the power elite and the people on how the influence of different institutions will develop in society during the 19905. A majority of both the power elite and the population at large expect that the influence of civic organizations and movements, television and radio, the print press, major corporations, universities and elected officials in communes will increase rather than decrease. In addition, one-quarter of the members of the power elite believe that the influence of private citizens will increase. One-fifth of the people expect to see banks and employer organizations increase their influence. It also seems that the power elite believes that executive authority and bureaucracy will gain a firmer grip on society during the next decade. It is noteworthy that both the power elite and the people believe that direct rather than representative democracy will be strengthened. Both groups anticipate reduced influence for parliament, the president, elected officials in communes, political parties, the central organization for agricultural producers, wage-earner organizations, banks, the church, the armed forces, and civil servants. An analysis of the visions of power elites and the people on the future development of influence in society, in short, lends less straightforward support to the power elite hypothesis than the analysis of current views on influence. There are even some (minor) differences between different elite groups. Looking at the broader picture, it would seem that the entire politico-administrative system as well as the corporatist system will progressively influence during the 1990s at the expense of the mass media and the economy.

Conclusions

The purpose of this study has been to investigate whether Finland is governed by one exclusive, cohesive and unanimous power elite. To draw a rough picture of this (possible) elite, we can modify the method of Scott (1991, 119) and cross-tabulate the three variables we have used here - exclusiveness, cohesion and unanimity - as shown in Figure 1. One of the typology's dimensions is the power elite's degree of openness, which may vary from low (with the power elite recruited from one single social stratum) to high (in which case the power elite is not dominated by any single stratum). The other dimension combines the variables of cohesion and unanimity.

attitudes and values. The power elite may be described as exclusive if it is recruited from one social stratum and if it is very cohesive, i.e. if its members have close contact with each other and they share the same social views. The power elite is segmented if it is recruited chiefly from one social stratum but its members have little interaction or its members do not share the same opinions, attitudes and values. The power elite is inclusive where it is not dominated by members originating from the same social stratum but if it is nevertheless cohesive, i.e. there is close interaction among members and their social views are more or less similar. The cell that remains empty in the figure is labelled by Scott as the fragmentary power elite. This description applies when the power elite is recruited from many different strata and when its shows no or little cohesiveness. However, it is not justified to speak of a fragmented power elite in the first place, since to do so is a contradiction in terms. Scott himself admits (1991, 120) that in this case the concept of power elite is almost an abstract category to describe rival, mutually balancing groups that take part in decision-making.

groups. In this regard there are no major differences between Finland and
Sweden.

On the basis of these results we may conclude that the situation in present-day Finland is different from that in Sweden, where there are - according to the Swedish power study - two dominant blocs or rival elites, i.e. economic power and political power, which are organized around private business and the labour movement. It is also noteworthy that the power study in Sweden found no major differences between the views of elites and the people on political issues. On the other hånd, the Swedish group of elites appeared to be quite heterogeneous in several respects. In certain piaces this distinction between "private business" and the "labour movement" was also visible in our analyses, but on the whole they seemed rather to form part of a broader and fairly cohesive network of social influence. During the past few decades the top stratum of decision-makers in Finland has been characterized by consensus rather than polarization, and a tendency to seek reconciliation among the interests of different parties has been evident. The most concrete manifestation of this has been the comprehensive incomes policy agreement, which coordinates the goals and interests of the government. employers, employees and agricultural producers.

institutions exercise the highest degree of social and political power in Finland? This question cannot be answered on the basis of the present results. It may be noted, however, that Mills (1956, 277) does not even accept the position that power is vested with either economic. political or military institutions. To avoid giving the impression that this could be the case, he uses the term "power elite" rather than, say, "ruling class". Mills starts out from the assumption that elite groups have autonomy. which means they have to take the most important social and political decisions jointly. This would seem to apply fairly well to the policy of consensus pursued in Finland, where the relationship between elite groups is characterized cooperation rather than by competition.

---

NOTES

1. The response rates (by positions) for different sectors were as follows: organizations 70. mass media 67, science 64, culture 63. administration 62. politics 60 and business 55. Because of the design of a questionnaire. the response rates of different sub-sectors are not known.

2. In 1991 Finnish women set a world record in winning a total of 77 seats (38.5 percent) in parliament.

3. In faet it is difficult to compare these countries because of different systems of professional In Sweden 42 percent of fathers of elite members are upper white-collar empioyees, the highest class (SOU 1990: 44, 320). In Finland the summed share of "leading positions" (highest) and "upper white-collar empioyees" is 38 percent.

4. To calculate the intensity of interaction, different coefiicients (weights) are assigned to the frequency of contacts. The coefficient for the option "at least once or twice a month" is 100; for the option "a few times a year" 50: and "less often or not at all" 0. Basedon this approach the same level of interaction (frequency) can be achieved in several ways, e.g. a few who have many contacts as well as by many who have few contacts. The concept of intensity separates these cases. This means that the weight of as high as in relatively infrequent interaction (a few times a year). The frequency of interaction in the last option is so low that it carries no weight at all in the intensity measure. The intensity of interaction between the power elite and institution A is obtained by multiplying by H)0 the proportion of those who have had contact with A at least once or twice a month and by adding to this figure the percentage. multiplied by 50, of those who have had contact with A a few times a year and by dividing this sum by the total number of contacts. i.e. the percentages of both those with a few monthly contacts and those with a few contacts. The maximum intensity value is 100. which is obtained if all members of the power elite have had interaction with A at least once or twice a month. The index describing the frequency of interaction is obtained division by 100 is done simply in order to obtain a comparable ligure. The maximum value for this index is thus also 100: the higher the score, the more frequent the interaction of the power elite with institution A (see Petersson 1989. 36-37.) For example, the intensity of interaction between the power elite and the president is obtained from the formula (1 x 100 + 20 x 50):21 = 52. The frequency will be 21 x 52: 100 = 11.

5. A question concerning the distribution of inrluence was presented to the power elite in a questionnaire. Conceptions among the population were collected simultaneously with the assistance of Suomen Gallup's so-called Finland Channel. The latter system was introduced in 1990. and permits collection of responses from a permanent group of people 15 to 65 years of age. Suomen Gallup has placed personal computers in 1000 homes in Finland and trained members of the households to use them. Responding to questions does not presuppose knowledge of data processing. A response rate is always more than 80 in the system. The sample represents all households in Finland, regionally and demographically.

REFERENCES

Alapuro, R. 1990. "Valta ja valtio - miksi vallasta tuli ongelma 1900-luvun vaihteessa", in
Haapala, P.. ed.. Talous, valta ja valtio. Tampere: Vastapaino.

Alestalo, M. 1986. Structural Change, Classes and the State. Finland in an Historical and
Comparative Perspective. Helsinki: University of Helsinki. Research group for comparative
sociology. Research reports No. 33.

Asp, K. 1986. Måktiga massmedier. Studier i politisk opinionsbildning. Stockholm: Forlaget
Akademi Litteratur.

Allardt, E. 1970. "Yhteiskunnan ryhmåt", in Waris, K. Allardt, E. & Wiio. O. A., eds.,
Valinnan vhteiskunta. Helsinki: Weilin + G66s.

Allardt, E. & Wiatr, J. W. 1978. '-Political Systems", in Allardt. E. & Wesolowski, W.. eds..
Social Structure and Change. Finland and Poland. Comparative Perspective. Warszava:
Polish Scientific Publishers.

Bacharach, P. 1967. The Theory of Democratic Elitism. A Critique. Boston: Little, Brown. Barnes, J. A. 1979. "Network Analysis: Orienting Notion, Rigorous Technique or Substantive Field of Study?", in Holland, P. W. & Leinhardt, S., eds.. Perspectives on Social Network Research. New York: Academic Press.

Bell. D. 1971. "The Power Elite Reconsidered", in Domhoff, G. W. & Ballard. H. 8., eds..
C. Wright Mills and the Power Elite. Boston: Beacon Press.

Bonacich, P. 1972. "Technique for Analysing Overlapping Membership", in Costner, H., ed.,
Sociological Methodology 1972. San Francisco: Jossey-Bass.

Burnham, J. 1941. The Managerial Revolution. London: Putnam.

Domhoff, G. W. 1990. The Power Elite and the State. How Policy is Made in America? New
York: Aldine De Gruyter.

Elinkeinoelaman valtuuskunta. 1991. Suomi etsii ilseåån. Raportti suomalaisten asenteisla
1991. Helsinki.

Etzioni-Halevy, E. 1993. The Elite Connection. Problems and Potential of Western Democracy.
Cambridge: Polity Press.

Helsingin Sanomat. 5.4.1992.

Hoffman-Lange, U. 1987. "Surveying National Elites in the Federal Republic of Germany",
in Moyser, G. & Wagstaffe, M., eds.. Research Methods for Elite Studies. Contemporary
Social Research 14. London: Allen & Unwin.

Hunter, F. 1953. Community Power Structure. Chapel Hill: University of North Carolina
Press.

Hyvårinen, M 1990. "Opiskelijaliike kertomuksenc", Politiikka 32, 261-270.
Knoke, D. 1990. Political Networks. The Structural Perspectives. New York: Cambridge
University Press.

Konttinen, E. 1991. Perinteisesti moderniin. Professioiden yhteiskunnallinen synty Suomessa.
Tampere: Vastapaino.

Kosonen. P. 1989. Hyvinvointivaltion haasteet ja pohjoismaiset mallit. Tampere: Vastapaino.
Lukes, S. 1974. Power: A Radical View. London: Macmillan.

Meisel, J. 1958. The Myth of the Ruling Class. Caetano Mosca and the "Elite". Ann Arbor:
University of Michigan Press.

Moyser, G. & Wagstaffe, M. 1987. "Studying Elites: Theoretical and Methodological Issues",
in Moyser, G. & Wagstaffe, M., eds.. Research Methods for Elite Studies. London: Allen
& Unwin.

Michels, R. 1915. Political Parties. New York: Hearst's International Library.

Pesonen, P. & Sankiaho, R. 1979. Kansalaiset ja kansanvalta. Juva: WSOY.

Petersson, O. 1989. Maktens ndtverk. En undersokning av regeringskansliets kontakter.
Helsingborg: Carlssons.

Porter, J. 1968. The Vertical Mosaic. An Analysis of Social Class and Power in Canada.
Toronto: University of Toronto Press.

Roos, J. P. 1987. Suomalainen eldmd. Håmeenlinna: Suomalaisen kirjallisuuden seura.

Ruostetsaari, I. 1989. Energiapolitiikanmdåråvty minen. Tampere: Tampereen yliopisto. Acta
Universitatis Tamperensis, Ser. A. vol. 278.

Ruostetsaari, I. 1992. Vallan ytimessd. Tutkimus suomalaisesta valtaeliitistd. Helsinki:
Gaudeamus.

Salminen, A. 1983. Tapaustutkimus lalouden, politiikan ja hallmnon keskinåisistå kxtkennbista:
Suomen iddnkaupan suuryritykset. Helsinki: Helsingin yliopisto. Yleisen valtio-opin laitos.
Tutkimuksia. Sarja A, N 63.

Salminen, A. & Haarala, S. 1986. "Ohjaavatko komissiot Suomen idånkauppaa?" Poluiikka
28, 69-83.

Sartori, G. 1962. Democratic Theory. Detroit: Wayne State University Press.

Schwarzmantel, J. J. 1987. Structures of Power. An Introduction to Politics. Brighton: Wheatsheaf

Scott, J. 1991. Who Rules Britain? Cambridge: Polity Press.

SOU 1990:44. Demokrati och makt i Sverige. Maktutredmngens huvudrapport. Stockholm:
Allmanna forlaget.

Sweezy, P. M. 1971. "Power Elite or Ruling Class?", in Domhoff. G. W. & Ballard. H. 8..
eds., C. Wright Mills and the Power Elite. Boston: Beacon Press.

Sånkiaho, R. 1992, "'Demokratian haasteet'. Kuuluuko kansanvalta kansalle?", in Anttila,
P. & Pukkila, T., eds., Suomisuosta. Kansallisen strategical aineksia. Tampere: Tampereen
yliopisto. Acta Universitatis Tamperensis, Ser. B, vol. 39.

Tiihonen, S. 1990. Hallitusvalta. Helsinki: Vapk-kustannus.