Resumé

A host of literature describes Sweden as the epitome of a consensual policy style country founded on rational and anticipatory behavior. However, recent research holds that consensus has yielded to a more conflict-ridden climate. Earlier research saw a consensual atmosphere as logically connected to anticipatory behavior, whereas conflict was connected to reaction. This article questions these linkages and claims that the present usage of the concept of anticipation does not fully acknowledge the strategic implications of policy style. The point is that policy style may be studied by examining how politicians set the political agenda. Designing a typology for measuring agenda setting in the Swedish Parliament's standing committee system gives us a research tool for studying the development of Swedish policy style from 1973-1991. The results show that the trend towards less consensus and more reactive political behavior in the Swedish society does not automatically amount to a less anticipatory policy style. On the contrary, real and open political antagonisms about the agenda give the parties strong incentives to use strategic anticipation to set the future agenda. Seen in this light, anticipation is not necessarily opposed to reaction. Growing conflictual reaction has not eroded anticipation in Sweden. Instead, both trends exist alongside each other. The parties do indeed make vivid use of their anticipatory means which may even strengthen democratic legitimacy in Sweden.

The Thesis: From Anticipatory Consensus to Anticipatory Conflict

In a host of literature, Swedish politics are described as the epitome of a consensual policy style founded on rational and anticipatory political behavior.¹ However, recent research holds that instead, consensus has yielded to a more conflict-ridden climate. While earlier research indicated a necessary link between consensus and deliberative and thoughtful rational anticipation, later research points to a similar link between conflict and reactive, shortsighted behavior. This article questions these linkages. It claims that policy style may be characterized by studying a country's agenda setting process, and the results show that growing conflictual reaction does

---

* Jonas Hinnfors. Department of Political Science, Göteborg University. Sprängkullsgatan 19. S-411 23 Göteborg, Sweden.

not necessarily mean the erosion of anticipation in Sweden. Instead, both
trends exist alongside each other.

Since one of the foundations of Swedish democracy (as defined in the Swedish Instrument of Government Chapter 1, §6) rests on the ability of the government to govern rather than just react, the alleged changes in style could have far-reaching repercussions for the quality of democracy. Surprisingly, opinions about Swedish policy style tend to be supported by rather scant empirical data (Anton 1969; Richardson 1979; Anton 1980; Ruin 1982; Elder et al. 1982; Särlvik 1983, 146f.; Jerneck et al. 1988; Maktutredningen 1990; Sjölin 1993).

Apparently, consensus has been seen as logically connected to anticipatory behavior, whereas conflict has been connected to reaction. But is there really a case to be made? Can we be sure that the element of reaction has been constantly growing as an automatic consequence of increasing levels of conflict, to the detriment of the element of anticipation, one of the pillars of Swedish state and society? This article says no, based on the following considerations. When assessing policy style, research has typically concentrated on the content as well as on the organizational setting of particular policy areas. However, the political centrality of the political debate has been underestimated. The fact that all relevant Swedish political actors used to be organizationally involved in the decision making process does not automatically imply that the process has also been characterized by consensual consideration. Shifting the analytical focus to the actual debates gives us a better research tool for estimating the degree of anticipation in Swedish politics and its possible connections with consensus or conflict.

In debates between politicians, the strategic activity aimed at setting the future political agenda is central.² When the policy climate becomes more conflictual, the politician will be less and less certain that future issues are settled to his advantage. Therefore, in a conflictual climate, politicians increasingly need to rationally calculate and anticipate the setting for future political debates. Thus, we have reason to believe that, in order to be strategically prepared, politicians try to set the future political agenda.

The concept of policy style implies a durable and systematic approach to public problems (Freeman 1985, 474). As defined by Richardson et al., policy style stands for "standard operating procedures for handling issues which arrive on the political agenda" (Richardson et al. 1982, 2f.; cf. Mayer 1972, 119). According to Richardson et al., a nation's policy style consists of two aspects. (1) "A Government's approach to problem solving," and (2) "A Government's relationship to other actors. ... [Is it] concerned to reach a consensus with organized groups, or is it more inclined towards imposing decisions notwithstanding opposition from groups?" (Richardson et al. 1982, 12f.). Thus defined, the thoughtful, rational, open, and consensual Swedish

policy style has apparently changed towards more conflict and less
anticipation.

In this article, I first set out to qualify the concept of anticipation. Its present usage does not fully acknowledge the strategic implications of policy style. I contend that the contradiction between reaction and anticipation is based on the wrong premises. Surely, a political actor commands a range of possibilities to anticipate future outcomes while reacting! Thus, one of the major elements of anticipation must be that politicians strategically calculate future outcomes of present acts.

Secondly, I apply an elaborated concept of anticipation to a case study. This case study is designed to focus on the fact that the anticipation of future political agendas is essential to political strategy. The point is that policy style can be studied by examining how politicians set the political agenda. The results in this article rest on the assumption that a formalized forum exists where agenda setting can be studied. Recent research findings emphasize the agenda setting capabilities of national parliaments generally, and the committee systems in particular.

Third, I set out to establish a solid research tool for measuring policy style by designing a typology of the various standing committee majority ways, i.e. "standard operating procedures" of denying the minority's agenda proposals. To make conclusions about the development of Swedish policy style possible, the typology is applied to Parliament election years spanning the time period since the reorganization of the standing committee system (1973-1991).

Before embarking on the study, a word on Swedish parliamentary procedure is required. Compared to Anglo-Saxon systems, activity in the Parliament is focused on political decision making rather than on criticizing and controlling government (Sannerstedt & Sjölin 1992, 99; cf. Esaiasson & Holmberg 1996, 216). A central parliamentary institution is the committee system which is composed of 16 standing policy-specialized committees (17 members each) selected by the Parliament among its members according to overall party strength in the Parliament. "Any matter raised by the government or by an MP has to be considered by a committee before the final decision is taken [by the entire Parliament]. This consideration results in a written report, and the decision of the Riksdag concerns this report." (Sannerstedt & Sjölin 1992, 110). Proposals and motions cannot be killed in committee. All bills must be dealt with and commented upon within a relatively short period of time, which means that the committee reports contain more or less extensive majority comments concerning the opposition's motions.

Politicized issues in the committees usually arise as a consequence of a
government proposal being challenged by "party motions." Contrary to the
great mass of independent, private members' motions, these "party motions

are introduced by the opposition parties on matters of principle and party politics. . . [Consequently, in dealing with party motions, the committee members] are obliged to follow the official party line ... In practice, proposals put forward in private members' motions are seldom accepted by the committee" (ibid., 116). Between 1973-1991, the number of motions introduced rose substantially from about 2000 per annum to around 5000 per annum (the number of "party motions" has gone up from roughly 200 to about 500). However, the amount of opposition "party motions" being rejected by the majority has stayed at about 80 percent (Sjölin 1991, 142).

Although formally independent, committee reports on politicized issues "are de facto decided upon by the party groups" (Sannerstedt & Sjölin 1992, 111) and the argumentation from the committee reports is usually repeated in the ensuing parliamentary debates. Thus, committee reports precede floor debates as important arenas where agenda setting considerations have to be elaborated among the parties. In most cases, the real political job - including the parties' fundamental strategic considerations - has to be done in the committees.

In politics, agenda setting may generally be achieved either by initiating one's own issues or preventing other actors from initiating their issues. According to Weaver's view that "when push comes to shove, most officeholders seek above all not to maximize the credit they receive but to minimize blame" (Weaver 1986, 372), it would seem that the act of denying and preventing challenges to agenda setting attempts constitutes an important part of political life. Bearing in mind that the bulk of the Swedish Parliament's work (including committee work) actually concerns the majority's refusals of the minority's agenda challenging bills, it appears fruitful to use these refusals in a study of agenda setting behavior on the elite level.

In summary, Swedish parliamentary committee reports are founded on strategic party considerations. They constitute empirical material that is central to politicized debate and relatively easy to handle. The following results are exclusively based on politicized issues where policy setting "party motions" stand against government proposals in four important standing committees: Agriculture (environmentally related issues only), Housing, Social Security, and Trade and Industry.

The committee reports that deny the minority motions will be classified according to the typology developed in the following sections and are graphically displayed in Table 1. The selected election years (1973, 1978/79, 1984/85, and 1990/91) span the entire period since the major 1970 transformation of the Swedish Parliament from bicameralism to unicameralism.

The Concept of Anticipation

Characterizations of Swedish policy style as "deliberative," "anticipatory," and "rationalistic" are all based on the assumption (explicit or implicit) that political actors anticipate "the objective reality" (Anton 1969, 93f.; 1980, 182; Elder et al. 1982, 180ff.; Ruin 1982, 142, 151). Scholars seem to refer to a set of ideas which are here specified as fact anticipation. Swedish political actors are depicted as masters of the art of rationally judging and assessing future societal directions. Based on the knowledge of seemingly "objective" social traits, parties formulate policies and platforms by heeding the presumed changes. Anton pictures the anticipatory ways of Swedish parties as so thoroughly rational that they, in fact, appear to react to societal changes - though in advance (Anton 1969, 93f.; cf. Gustafsson 1989, 165).

Without exception, the research that supplies the image of an anticipatory style is based on the assumption that anticipation is confirmed by "rational consensus," "compromises" and "harmony." Richardson et al. juxtapose reaction with anticipation: "Most post-industrial societies may be moving [to] where governments are increasingly reactive rather than anticipatory in their approach to problem-solving" (Richardson et al. 1982, 14). However, their argument also states that each of the two categories may be characterized by either consensual or imposing politics (ibid., 13).

Important as such distinctions may be, there is room for further elaboration. Acts of anticipation need not exclude reaction. Nothing says that an actor cannot or does not anticipate the effects of his reactions, thereby improving rather than impairing his prospects of governing.

My point is that the earlier definitions of policy style are flawed to the extent that they ignore strategy as an important aspect of political life (Sjöblom 1968, 30). European unification is a good example of the difference between "strategic" and "fact" anticipation. After the Second World War, the French statesman, Jean Monnet, formulated his vision of a united Europe. However, Monnet did not enter this gigantic enterprise by flatly proposing a union. Instead, he initiated much smaller, though still important, measures. These were based on extensive and advanced strategic thinking. Monnet anticipated a strong relation between the means of economic integration, and the goal, a European political union. His plan was that a sufficiently all-embracing integration in a vital economic field - steel and coal - would spill over into other economic areas, and later, cooperation would also reach political organizations and parties. Gradually, European central banks and bureaucracies would be cooperating more closely (Haas 1958; 1964).

Monnet's basic idea was that preliminary reform would prove further
reforms necessary. As economic integration acquired an inner dynamic, it
would make it natural for politicians to widen cooperation in other areas.

Step by step, new reforms would be enforced. Instead of immediately heading for the final target and thus risking political opposition, Monnet utilized and mastered the art of anticipating the strategic consequences of small reforms.

The very essence of politics is found in the actors' preference for holding the initiative and not being held at bay by others. They do not wait for things to happen. In order to avoid unforeseen occurrences, they try to prepare themselves as best they can. In fact, Jean Monnet used what is here specified as strategy anticipation. His endeavor was to foresee the consequences of his own activities not only for state and society, but also for his political opponents.³

Fact anticipation, on the other hand, means that actors calculate possible future developments in society and its various sections, regardless of what other political actors might do. With such an approach to the concept of anticipation, earlier research comes close to neglecting the fact that political actors are part of the political game which is supposed to change the surrounding society. But, in order to make things happen the way they want them to, political actors need to calculate the consequences of their own activities, and in particular relate them to the expected strategic calculi and actions of other actors. Fact anticipation disregards all this and thus only covers part of the truth about Swedish policy style.

As long as political life remains stable and without deep, divisive conflicts, politics can be seen as a question of rationally reflecting on the objectively best alternative. Fact anticipation becomes possible. On the other hand, when clear-cut conflicts force one actor to emphasize his specific merits, the need for strategy anticipation grows imminent. Obviously, one important means for success in these matters is to have a strategy for how to handle the expected activities of your opponents.

Agenda Power

The political game that precedes the concrete allocation of values often concerns more abstract matters than simply deciding on specific issues. Earlier, it was common wisdom that agenda setting is a result of factors beyond the influence of politicians. Later studies of the American Congress indicate the opposite. A great deal of the struggle between members of Congress concerns the political agenda and how to influence it. Moreover, politicians are quite successful "fram[ing] ... the debate." (Sinclair 1989, 213f.; 1985).

The basic and most interesting element of governing in modern political
life is thus the activity that surrounds agenda setting. Controlling the future
agenda might be as efficient a political tool as trying to control the outcome

of the issues already on the agenda (Bahrach & Baratz 1962; Hammond
1986, 381; Howlett 1991, 1).

Actors trying to avoid unwanted issue outcomes have a number of strategies at their disposal. Such actors necessarily need some kind of means to control the political agenda. The actor might try to deny the legitimacy or the competence of the assembly. The agenda can be changed and decisions adjourned. All else failing, the politician must finally choose the least negative outcome (Lewin 1988, 12ff.). These various means of control constitute different approaches to problem solving. Therefore, using the definition suggested by Richardson et al. - "standard operating procedures for handling issues which arrive on the political agenda" - the various strategic means by which an actor may control the agenda comprise various policy styles.

In light of the earlier discussion, I deem it fruitful to concentrate agenda
setting studies on the means of strategic control available to those who wish
to set the agenda (Burstein 1991, 335; Pfeffer 1992, 187f.).

Agenda Setting Strategies

Agendas are far too important for the actors to indulge in the kind of harmonious consensus attitude claimed to epitomize Swedish policy style. First, the actors need to prevent challenging actors from setting the agenda. We can say that "[t]he powers over timing and agenda appear to be more negative than positive. It is easier to block a piece of legislation in committee than it is to steer it through the legislative process" (Kingdon 1973, 118; cf. Stringer & Richardson 1980). Second, they have to favor their own agenda content as best they can. This article claims that acting strategically gives you a wide range of possibilities to reach the future consent from your opponents by denying their agenda propositions today. This report differs from other studies of negotiation strategies such as those made by the American scholars Fisher and Ury, because it does not deal explicitly with what happens at the negotiation table. Fisher and Ury formulate conclusions about relations between political opponents based on the opposite assumption: In order to strike the deals you want, you should refrain from openly turning your antagonist's wishes down. By saying yes to, at least, some requests, chances increase that the negotiations can be characterized by mutual compromises (Fisher et al. 1992; Ury 1992).

However, the act of denying may work as an ideological device much in the same way as when managers impose their "definition of reality upon discourse and conduct within and around [an] organization" (Izraeli & Jick 1986, 173). The leader's "definition of the situation serves as a basis for action by others - actions oriented to the achievement of ends desirable from

the leader's viewpoint" (ibid.)- The proof of the pudding lies in turning the
opponent's challenging agenda suggestions down in such a manner that you
are also able to set the agenda in the future.

How, then, could studies of the art of saying no to challenging political agendas be done? In a general way, Hirschman elaborates the concept of denial in The Rhetoric of Reaction. Over the years, various "reactionary" patterns have had the purpose of rejecting reforms.⁴ Hirschman discovers three waves of reaction against modern Western democracy. The first one opposed equal legal rights generally. The second wave criticized general suffrage. Today's criticism concerns the welfare state (Hirschman 1991).

The "reactionaries" have used three kinds of arguments. The "perversity" argument holds that the reforms will come to nothing, their content is questionable and will have unintended side effects. The "futility" argument claims that the reforms may be well meant, but they are inevitably insufficient. The world is run by eternal rules which cannot be changed by man. The "jeopardy" argument describes the reforms as important, but they happen to cause unacceptable costs that ruin earlier achievements (ibid., llff., 43ff., 81ff.; Hirschman 1993, 293-307).

Hirschman's categories may be used as a starting point for a discussion on how to execute power over the political agenda. In the first place, these categories contain arguments against the content of reforms. Secondly, his typology includes procedural arguments which, although they accept a particular reform's content, emphasize the related risks if decisions are forced through without checks and/or delays. Drawing on Hirschman's categories of denial, the following discussion concentrates on what these concrete measures might be for the agenda setting actor.

Governing Signals

Let us consider the agenda setting activity as being equal to communicating a certain message to relevant parts of an organization. The actor in charge of the design, content, and rhythm of the "governing signals" is in charge of the "organization," i.e. the agenda (Deutsch 1966, xxvii, 76ff.; Weaver 1989, 333; Katzmann 1989, 288ff.).

However, the governing actors do not stand uncontested: "Given the huge number of groups and individuals actively attempting to focus attention on an issue and get it onto institutional agendas, the selection process has become critical" (Sinclair 1989, 213; cf. Kingdon 1973, 117; Sinclair 1986, 76; Pfeffer 1992, 227ff.). The point made here is that it is exactly this process of selection that political actors try to control in their own favor. With power over the selection process, they have good prospects for future influence

over the ensuing concrete issue decisions. Parliamentary committee
majorities have many opportunities to influence this selection process.

In order to establish a categorization of different governing signals, I use the original garbage can model. This model emphasizes three essential, procedural and actor related factors that can change decisions in organizations. First, the "access/decision structure'" of the organization can be controlled. Second, the "entry times" for choices can be regulated. Third, the "net energy load" of the organization may be adjusted (Cohen et al. 1972, 2ff.). Actors who are able to determine who is a legitimate participant in the decision making, where the decisions should be taken, and influence when the legitimate decisions should be taken, increase their opportunities to set the agenda.

Behind the garbage can model, it is in fact possible to discern purposeful, consciously acting actors, struggling to succeed in "intelligent decision making under ambiguous circumstances" (ibid., 2). With constantly changing "problems," "policy alternatives" and "politics," politicians live in an apparently unsafe world. They risk getting into dilemmas at any time. However, my claim is based on the belief that politicians are intentional creatures, capable of choosing for the better rather than the worse. They are prepared and ready to seize various opportunities as these develop.⁵ An essential tool of government is therefore the formal rules and procedures by which actors regulate when, where, and by whom the agenda may be set.

However, it is not very likely that these signals actually affect or change the preferences of competing actors at the elite level, at least not in the short run. Nor is it highly probable that political skills will change in the short run. On the other hand, strategically rewarding activities would be those measures that decide the timing, as well as the place, of future agenda decisions (Underdal 1992, 222; Weaver 1989; Torgersen 1991). The logic behind this line of strategic reasoning lies in the need of blame avoidance. Opposition proposals should not be flatly turned down. Instead, they should receive a decent amount of attention. Decisions involving high political costs may be avoided by refraining from taking any decision at all. Further investigation into the matter to postpone the final verdict may be a superior alternative to an immediate all-out political confrontation. Should any of these strategies be unsuitable to the actor, the right to decide can be transferred to an alternative forum or authority (Nelson & Lindenfeld 1978. 21; Weaver 1986, 372ff.; Hood 1986, 124f.; Lundquist 1987. 37ff.; Linder & Peters 1989, 42ff.).

So far, research concerning conflict levels in Swedish political life has been based on the increasing tendency of opposition parties to make reservations to government proposals (Sannerstedt & Sjölin 1992. 112). When it comes to majority parties denying minority party bills, no

corresponding development has been found. The rejection rate remains around 80 percent (Sjölin 1991, 142). However, my point is that major changes are hidden behind this apparent stability. By focusing on the character rather than on the level of the majority denials, this article shows changing levels of governing-signal conflict behind the veil of stability.

A Typology for Measuring Policy Style

The following case study is based on the assumption that agenda setting is a suitable focus in an analysis of changes in Swedish policy style. To test the importance of formal determinants of agenda setting strategies, the Swedish Parliament's standing committees will be the object of study. A typology concerning the "art of saying no" is designed in order to be able to say something about the conflict/consensus and reaction/anticipation aspects of the Swedish policy style.

The typology is built around the general elements suggested by Hirschman. A committee majority may regulate the future agenda by letting its own definition of appropriate content dominate the committee's drafts and reports, be it ideological or factual. Besides, the majority can use procedural arguments as its rejection tool. The majority may claim that the issue belongs to other assemblies, e.g. regional or local councils, thereby defining where legitimate decisions should be taken. Lacking any of these arguments, the majority may instead claim that the issue is brought to the committee prematurely. This last type of argument can be turned upside down, meaning that the committee claims the challenging proposal has already been met. These two final types of argument enable the committee's majority to define when an issue is legitimately brought to the agenda.

In relation to the earlier statements in this article, please remember that all categories used here are applied to those cases where the majority denies the minority bills. Thus, the typology measures instances that could be formally characterized as conflictual, i.e. the bills are denied. A high level of majority denial belongs to normal parliamentary procedure, and this applies equally well to the consensual 1960s and to the 19905. Therefore, knowledge of the levels alone does not reveal much about the extent of conflictual behavior. Instead, by studying the character of denial, we obtain a more subtle tool for measuring conflictual behavior. Consequently, the typology measures two things, i.e. the degree of anticipation (cf. Table 1; higher: categories number 2, 4; medium: categories number 1, 3, 5, 6; lowest: category number 7) and the character of conflict.

Findings: A Changing Policy Style in Sweden?

The findings shown in Table 1 corroborate the need to qualify the descriptions of recent developments in the Swedish policy style. Of the three main categories in Table 1, the "No, the timing is wrong" category is the best measure of anticipatory strategic agenda behavior, and it shows remarkable stability. The percentage stays close to the 50 percent level in all years. The "Content" category, on the other hand, increases slightly over time.

Perhaps surprisingly, the category that measures the Parliament's willingness to phrase its denial argumentation by referring to a lack of formal decisive competence ("No, we are not the proper assembly") has fallen from around 12 percent to roughly 6 percent. In spite of dramatically growing work loads (e.g. the sharp rise in the number of motions) in an age of alleged rubber-stamp parliaments, the Parliament obviously refrains from giving away its decisive privilege - at least orally.

When we split the main categories into sub-categories, our above conclusions need to be qualified. Some unequivocal change has indeed taken place. In spite of the remarkable overall "No, the timing is wrong" stability, we find that the decidedly "future-oriented" sub-categories (2 and 4 in Table 1; the two most clear-cut anticipatory measuring indicators of the seven subcategories) suffer a rather substantial decrease, dropping from a sum total of around 41 percent to about 30 percent. Still, the overall conclusion seems to be that even though it may be decreasing, anticipatory strategic behavior still constitutes a main element of the majorities' acts towards their opponents.⁶

There is a definite trend towards a much more curt atmosphere in the committees' policy style. Above all, this trend is indicated by the fact that the sub-category "flat refusals" has risen dramatically to comprise 1/4 of all denials by the early 19905. Today, majorities not only deny the minorities' bills. They do it without bothering to give their reasons. In an institution where the established norm has been deliberate and elaborate "matter-offactness" (Anton 1969; Sannerstedt & Sjölin 1992, 11 If.), the results from the later part of the research period indicate a clear breach with old habits.

The increase may be compared with the slightly decreasing ideological levels. At the same time, the majority seems increasingly content with its own record. "No, it is too late" statements increase while the "No, it is too early" denials decrease.

Apparently, compared to the politicians of twenty years ago, today's conflictual party representatives seem to be struck by a sense of "fin de siécle," a stark contrast to the "happy" era of belief in development and harmony that was so apparent in the 1950s and 19605. They might have reason to feel insecure. The above results can be used to evaluate the longterm effects of the 1970 constitutional reforms which transformed the

Parliament into a unicameral assembly. At the same time, strict electoral proportionality was introduced. The ensuing lack of clear parliamentary majorities has enabled a power shift from Cabinet to Parliament. Likewise, today's decidedly more volatile electorate has contributed to a situation where the parties are obviously increasingly insecure as to their future competitive strength (Kirchheimer 1966; Granberg & Holmberg 1991: Sjölin 1993). The need to maintain freedom of action has put pressure on the parties to wield as much future agenda power as they can. To some extent, my results support this observation. The parliamentary standing committees certainly pay more than lip-service to their constitutional obligation to comment on all parliamentary proposals.⁷

The results seem to corroborate Harold Lasswell's thesis about the similarities between the conditions for policy style and those of a theater of war. The harder the crisis on the battle field as well as in politics, the harsher the fighting, and the more important the need to maintain freedom of action. Only when facing more favorable prospects, soldiers as well as politicians who deny bills might resort to more lenient measures (Lasswell 1949b, 22-28; 1949a, 19). No doubt, the impressive increase in the rate of flat refusals is a good indicator of today's more harsh and conflictual political fighting.

Implications for Further Research

The results presented here rest on three assumptions. First, policy style may be simultaneously anticipatory and reactive. Second, relevant "standard operating procedures" can be found within agenda setting in a nation's parliament. Third, conclusions on changes in policy style are more easily reached by studying the character of the debate preceding the acts than by studying the content and organizational setting surrounding the actual policy choice in Parliament.

These assumptions are not entirely uncontested. Using "reaction" in conjunction with "anticipation" may appear to form an overly waterproof combination of criteria that would meet any imaginable situation. On the other hand, it has been shown that the typology used here in no way prevents the empirical data from varying between degrees of anticipation. The methodological logic behind the research is the following. If political behavior were to be either reactive or anticipatory, a standpoint contested in this paper, a truly reactive decisive forum would show no signs of anticipatory behavior. The parliamentary standing committee system is a forum that is formally obliged to "react." In such a system, we would not expect to find anticipatory behavior as a matter of course. But if we find traces of anticipation even in those circumstances, we have a better

foundation when we claim that the character of what is usually called reactive behavior may very well be anticipatory as well. By applying the typology used here on instances of reaction, we can say something about the extent to which the reaction is also anticipatory.

The results clearly show that the typology actually worked as expected, i.e. as an operationalization of the Hirschman categories of denial. In order to test the generalizability of this operationalization, the typology could be applied to other decision contexts, e.g. decisions within party organizations, parliamentary debates, etc.

Furthermore, a large part of a nation's political debate concerning agenda setting falls outside the agenda as operationalized here. Surely, such contextual factors as debates in the media may have far-reaching implications for the policy agenda. It has been suggested that the character of the issues, the work load of the politicians, and the issue life cycles heavily influence the agenda (Downs 1972; Walker 1977; Kingdon 1984; Polsby 1984; Kelman 1987; Schneider & Ingram 1988; Stone 1989). On the other hand, issues simply have to make their way into the authoritatively deciding bodies if they are to become authoritatively binding decisions. The Parliament is a body that can make such decisions.

Limiting my case study to those instances where a committee majority denies the minority bills might seem too narrow. However, contrary to earlier agenda studies, I measure the strategic deliberation and anticipation before the agenda has been set. Earlier agenda studies have been limited to offering "convincing retrospective explanations of why an issue arose" (Burstein 1991, 335). By introducing the actors' prospective reasoning before the agenda setting and comparing these acts with what actually became of the agenda, research would stand a chance of enhancing its predictive potential.

As operationalized here, statements from the parliamentary minutes are used as motive indicators of the politicians' strategies. The results are thus based on the somewhat precarious premise that politicians actually mean what they say. Such a research approach is, however, no more uncertain than one where politicians' actual choices of policy instruments or the organizational setting are taken as motive indicators. In the latter instance, the risk of circular explanations looms large (Hadenius 1983).

Finally, the parliamentary power balance has already been pointed to as something that might affect the policy style. Though it is beyond the scope of this article, another circumstance that affects the level of conflict is the institutional setting. During the time period studied here, the centerpiece of "the Swedish model," the Royal Research Commissions, have peaked and begun to decrease in importance (Maktutredningen 1990; Johansson 1992). Because of their status and parliamentary composition, the Commissions have constituted an important forum for political minority views. As a result

of the dwindling commission system, minority views now have to be
presented elsewhere, e.g. in Parliament.

Policy Style and Democracy

As indicated in the beginning of this article, a changing policy style may have repercussions on the quality of democracy. Previous research concerning Swedish policy style includes an underlying normative message. It is asserted that rational, consensual and deliberative decisions are democratically good. The authors are impressed by the Swedish style and pity less fortunate, more conflictual countries. According to such norms of harmony, increasing levels of conflict would be precarious from a democratic point of view.

Such views seem to go well with the opinion that the foundations of democracy rest on a society where the important institutions are based on an open and reflecting policy style: "The aim of the deliberative process is to broaden the participants' information and enable them to discover their own preferences" (Manin 1987, 352; Hirschman 1991, 170). A decreasing amount of reflection and increasing levels of harsh words would indicate a less desirable development.

On the other hand, Manin brings our attention to the other equally important aspect of democracy: "Deliberation requires not only multiple, but conflicting points of view because conflict is the essence of politics" (Manin 1987, 352). Is it possible that the image of harmony and consensus, so long hailed as the hallmark of Sweden, has had the unfortunate side effect of hiding real antagonisms between the political parties?

In the light of Lijphart's works on democracy, Swedish (and Scandinavian) politics of the 1960s and 1970s appeared "to have moved the farthest in the direction of the depoliticized type" of democracy characterized by a homogeneous society in conjunction with coalescent elite behavior (Lijphart 1977, 111). The results in this article indicate that it would be more adequate to describe Swedish policy style in the 1990s as belonging to Lijphart's "centripetal democracy" where elite behavior has turned adversarial. However, as long as opposition and conflict are maintained within the parliamentary framework, Lijphart deems such a development favorable from a democratic point of view (ibid., 108).

The legitimacy of democracy is equally founded on a clear-cut demonstration of conflict and open, rational reflection. The style of give and take, which has emerged and grown in recent years, might then be a new source of democratic legitimacy in Sweden (cf. Tingsten 1966, 46ff.).

Attempts by political actors to set the future policy agenda may thus be
related to democratic legitimacy. Legitimacy exists if (a) a democratic

regime gains legitimacy as a result of open and reflective debate in the deciding assemblies, and (b) the "deliberation" ensures that open conflicts are brought to the table but do not place the participants in a deadlock. In addition, (c) the population should hold the view that their elected representatives are allowed to base their opinions on the outcome of the open debate (cf. Hirschman 1991, 169). The efforts to set the future agenda thereby form an important part of democracy by allowing the actors the freedom to refrain from certain present standpoints. Instead, the actor has the opportunity to keep an eye on his future "privilege of interpretation."

The thesis presented here is that the trend towards less consensus and more reactive political behavior in Swedish society does not automatically amount to a less anticipatory policy style. On the contrary, real and open political antagonisms about the agenda give the parties strong incentives to use their strategic anticipation to set the future agenda. Seen in this light, anticipation is not necessarily opposed to reaction. The parties do indeed make vivid use of their anticipatory means which may even strengthen democratic legitimacy in Sweden.

---

NOTES

1. This article is a brief account of a Bank of Sweden Tercentenary Foundation sponsored research project, Mortgaging the Future Agenda (Att inteckna den framtida dagordningen). A full account is given in Hinnfors (1995).

2. The concept of agenda is subject to a confusingly abundant number of definitions. As used here, the concept stands for "the list of subjects within the governmental agenda that are up for an active decision." (Kingdon 1984, 4; Walker 1977, 424f.; Asp 1983, 333ff.; Korsmo 1990; McClain 1990; Burstein 1991, 334ff.).

3. Here, strategy is defined as an "actor's extensive and comprehensive planning of the use of available means with the object of attaining certain goals attempted in competition with others" (Sjöblom 1968, 30).

4. The no arguments are not by axiom basedon "reactionary" deliberations. Any group of actors may use them as a means to criticize competing standpoints (Hirschman 1991, 7).

5. Obviously, it is important that the actors are prepared by making sure that ready-made solutions are available. These solutions should be thoroughly prepared, but also confer legitimacy to the actor himself rather than to the opponent parties (Durant & Diehl 1989, 180).

6. A more elaborate version of Table 1, not shown here, indicates that while "strong" future-oriented behavior comprises 24 percent of the 1991 sum total of future-oriented categories, the corresponding figure for 1973 was only 14 percent. Figures on "weak" future-oriented behavior show a similar but opposite trend in that the 1973 amount of "weak" future-oriented behavior was as high as 14 percent, whereas the 1991 share comprised 9 percent. These more elaborate findings corroborate the overall conclusion that anticipatory behavior still constitutes an important aspect of Swedish majority behavior.

7. As expected, the results vary somewhat between the four different sectors (cf. Padgett 1990). For example in the Housing Committee, categories number 3 + 4 (timing) fell from about 67 percent in 1973 to about 30 percent in 1991, while categories number 5-7 (content) rose from about 23 percent in 1973 to nearly 60 percent in 1991.

REFERENCES

Asp, K. 1983. "The Struggle for the Agenda, Party Agenda, Media Agenda, and Voter Agenda
in the 1979 Swedish Election Campaign," Communication Research 10, 333-355.

Anton, T. J. 1969. "Policy-Making and Political Culture in Sweden," Scandinavian Political
Studies 4, 88-102.

Anton, T. J. 1980. Administered Politics. Elite Political Culture in Sweden. Boston: Martinus
Nijhoff Publishing.

Bachrach, P. & Baratz, M. S. 1962. "Two Faces of Power,"" American Political Science
Review 56, 947-952.

Burstein, P. 1991. "Policy Domains: Organization, Culture, and Policy Outcomes." Annual
Review of Sociology 17, 327-350.

Cohen, D. et al. 1972. "A Garbage Can Model of Organizational Choice," Administrative
Science Quarterly 17, 1-25.

Deutsch, K. 1966. The Ner\>es of Government. Models of Political Communication and
Control. New York: The Free Press.

Downs. A. 1972. "Up and Down With Ecology - The 'Issue-Attention Cycle."" The Public
Interest 28. 38-50.

Durant. R. & Diehl, P. 1989. "Agendas. Alternatives, and Public Policy: Lessons from the U.S.
Foreign Policy Arena," Journal of Public Policy 9, 179-205.

Elder, N. et al. 1982. The Consensual Democracies? The Government and Politics of the
Scandinavian States. Oxford: Martin Robertson.

Esaiasson, P. & Holmberg, S. 1996. Representation From Above: Members of Parliament and
Representative Democracy in Sweden. Aldershot: Dartmouth.

Fisher, R. et al. 1992. Vagen till Ja. En nyckel till framgångsrika förhandlingar [The Road to
Yes. A Key to Successful Negotiations]. Malnio: Liber Ekonomi.

Freeman, G. P. 1985. "National Styles and Policy Sectors: Explaining Structured Variation."'
Journal of Public Policy 5. 467^196.

Granberg. D. & Holmberg, S. 1991. "Electoral Campaign Volatility in Sweden and the United
States," Electoral Studies 10, 208-230.

Gustafsson, G. 1989. "Challenges confronting Swedish decision-makers: political and policy
responses 1974-1987," in Damgaard, E. & Gerlich. P. & Richardson. J J.. eds.. The Politics
of Economic Crisis. Lessons from Western Europe. Aldershot: Avebury.

Haas. E. B. 1958. The Uniting of Europe. Oxford: Oxford University Press.

Hadenius. A. 1983. "The Verification of Motives." Scandinavian Political Studies Vol 6 New
125-150.

Hammond, T. 1986. "Agenda Control, Organizational Structure, and Bureaucratic Politics."
American Journal of Political Science 30, 379-420.

Hinnfors. J. 1995. På dagordningen? Svensk politisk stil i forandring [On the Agenda? A
Changing Swedish Policy Style], Stockholm: Nerenius & Santérus.

Hirschman. A. O. 1991. The Rhetoric of Reaction. Perversity, Futility. Jeopardy. Cambridge.
Massachusetts: Belknap Press.

Hirschman, A. O. 1993. "The Rhetoric of Reaction - Two Years Later." Government and
Opposition 28, 292-314.

Hood, C. 1986. The Tools of Government. Chatham: Chatham House.

Howlett, M. 1991. "Policy Instruments, Policy Styles, and Policy Implementation: National
Approaches to Theories of Instrument Choice." Policy Studies Journal 19. 1-21.

Izraeli, D. M. & Jick, T. D. 1986. "The Art of Saying No: Linking Power to Culture."
Organization Studies 7. 171-192.

Jerneck, M. et al. 1988. "Internationalization and Parliamentary Decision-Making: The Case
of Sweden 1970-1985," Scandinavian Political Studies 11, 169-194.

Johansson, J. Del statliga kommittéva'sendet. Kunskap. kontroll, konsensus (The System of
Royal Commissions. Knowledge. Control. Consensus]. Edsbruk: Akademitryck AB.

Katzmann, R A. 1989. "The American Legislative Process as a Signal." Journal of Public
Policy 9. 287-306.

Kelman. S. 1987. Making Public Policx. NewYork: Basic.

Kingdon, J. 1973. Congressmen's Voting Decisions. New York: Haiper & Row. Publishers.

Kingdon, J. 1984. Agendas, Alternatives, and Public Policies. Boston: Little, Brown and
Company.

Kirchheimer, O. 1966. "The Transformation of Western European Party Systems," in
LaPalombara, J. & Weiner, M., eds., Political Parties and Political Development. Princeton:
Princeton University Press.

Korsmo, F. 1990. "Problem Definition and the Alaska Natives: Ethnic Identity and Policy
Formation," Policy Studies Review 9, 294-306.

Lasswell, H. 1949a. "The Language of Power," in Lasswell, H. & Leites, N., eds., Language
of Politics. New York: George W. Stewart, Publisher, Inc.

Lasswell, H. 1949b. "Style in the Language of Politics," in Lasswell, H. & Leites, N., eds.,
Language of Politics. New York: George W. Stewart, Publisher, Inc.

Lasswell, H. & Kaplan, A. 1950. Power and Society. New Haven: Yale University Press.

Lijphart, A. 1977. Democracy in Plural Societies. New Haven and London: Yale University
Press.

Linder, S. & Peters, G. 1989. "The Instruments of Government," Journal of Public Policy 9,
35-58.

Lundquist, L. 1987. Implementation Steering. An Actor-Structure Approach. Lund, Bromley:
Studentlitteratur, Chartwell-Bratt Ltd.

Maktutredningen. 1990. Demokrati och makt i Sverige [Commission on Democracy and Power
in Sweden], SOU 1990:44.

Manin, B. 1987. "On Legitimacy and Political Deliberation," Political Theory 15, 338-368.

McClain, P. 1990. "Agenda Setting, Public Policy, and Minority Group Influence: An
Introduction," Policy Studies Review 9, 263-272.

Nelson, B. J. & Lindenfeld, T. 1978. "Setting the Public Agenda: The Case of Child Abuse,"
in May, J. & Wildavsky, A., eds., The Policy Cycle. Beverly Hills, London: Sage.

Padgett, S. 1990. "Policy Style and Issue Environment: The Electricity Supply Sector in West
Germany," Journal of Public Policy 10, 165-193.

Pfeffer, J. 1992. Managing with Power. Boston: Harvard Business School Press.

Richardson, J. J. et al. 1982. "The Concept of Policy Style," in Richardson, J., ed., Policy
Styles in Western Europe. London: George Allen & Unwin.

Riksdag minutes.

Ruin, O. 1982. "Sweden in the 19705: Policy-Making Becomes More Difficult," in
Richardson, J., ed., Policy Styles in Western Europe. London: George Allen & Unwin.

Sannerstedt, A. & Sjölin, M. 1992. "Sweden: Changing Party Relations in a More Active
Parliament," in Damgaard, E., ed., Parliamentary Change in the Nordic Countries. Oslo:
Scandinavian University Press.

Särlvik, B. 1983. "Coalition Politics and Policy Output in Scandinavia: Sweden, Denmark and
Norway," in Bogdanor, V., ed., Coalition Government in Western Europe. London:
Heinemann Educational Books Ltd.

Schneider, A. & Ingram, H. 1988. "Systematically Pinching Ideas: A Comparative Approach
to Policy Design," Journal of Public Policy 8, 61-80.

Sinclair, B. 1985. "Agenda Control and Policy Success: Ronald Reagan and the 97th House,"
Legislative Studies Quarterly X, 291-314.

Sinclair, B. 1986. "The Role of Committees in Agenda Setting in the U.S. Congress,"
Legislative Studies Quarterly XI, 35^5.

Sinclair, B. 1989. The Transformation of the U.S. Senate. Baltimore: The Johns Hopkins
University Press.

Sjöblom, G. 1968. Party Strategies in a Multiparty System. Lund: Studentlitteratur.

Sjölin, M. 1993. Coalition Politics and Parliamentary Power. Lund, Bromley: Lund
University Press, Chartwell-Bratt Ltd.

Stone, D. 1989. "Causal Stories and the Formation of Policy Agendas," Political Science
Quarterly 104, 281-300.

Stringer, K. & Richardson, J. J. 1980. "Managing the Political Agenda: Problem Definition
and Policy Making in Britain," Parliamentary Affairs 33, 23-39.

Tingsten, H. 1966. Från idéer till idyll. Den lyckliga demokratien [From Ideas to Idyllics. The
Happy Democracy]. Stockholm: Norstedts.

Torgersen, U. 1991. "Housing Policy - Goals, Rhetoric, Information and Regularity,"
Scandinavian Housing & Planning Research 8, 105-111.

Underdal, A. 1992. "Designing Politically Feasible Solutions," in Maines, R. & Underdal, A.,
eds., Rationality and Institutions. Oslo: Universitetsforlaget.

Ury, W. 1992. Vagen forbi nej. Attförhandla med besvärliga människor [The Road Past No.
Negotiating Troublesome People]. Stockholm: Svenska Dagbladets Forlag.

Walker, J. 1977. "Setting the Agenda in the U.S. Senate: A Theory of Problem Selection,"
British Journal of Political Science 7, 423-445.

Weaver, R. K. 1986. "The Politics of Blame Avoidance," Journal of Public Policy 6,
371-398.

Weaver, R. K. 1989. "Setting and Firing Policy Triggers," Journal of Public Policy 9,
307-336.