"The Riksdag represents the Swedish people." This key passage in the Swedish constitution emphatically states that it is the responsibility of parliament to exercise the mandate of the people. When it comes to a precise description of how the Riksdag shall be organized in order to accomplish its task, however, the Swedish Code of Laws leaves us in the lurch. Like other constitutions, the Swedish Grundlag has little to say about which of parliament's various institutions and bodies shall be granted power and influence.

The silence of the constitution makes way for an internal struggle over power within parliament. Some parliamentary actors may benefit, for example, from a strong committee system, others from a strengthening of the position of the party groups. As the balance of influence between bodies change, so does the internal functioning of parliament. The outcome of the internal power struggle, in short, decides how the institution works.

Researchers interested in parliaments have so far been reluctant to analyze the internal workings of parliamentary bodies from the perspective of power and influence. There are, of course, a number of studies that assess the relative influence of two or three of the special parliamentary institutions (e.g. Mezey 1979; Arter 1984), but very few cover the whole spectrum of possible centers of power. Moreover, scholarly interest has been directed towards the distribution of factual influence and power. In

Resumé

This article is based on a mail questionnaire sent to members of the Swedish parliament (the Riksdag) in 1988. To increase our understanding of how the institution works, an analysis of members' perceptions of the distribution of power within the Riksdag is undertaken. Members were asked (1) how influential various groups and bodies are, and (2) how influential these groups and bodies should be. The results show that members want more power to be given to parliamentary party groups, committees and MPs as individuals, and less power to the party leaders and the chiefs of staff. Differences in perceptions along partisan lines are small. However, members of the Greens - an anti-establishment party - are more oriented toward strengthening the influence of individual MPs and toward weakening the power of party leaders than are members of the established parties.

order to understand how the institution works, however, it is just as
important to gain insight into the actors' perceptions of power.

In this article we try to break new ground by analyzing how Swedish MPs perceive the distribution of power and influence in the Riksdag. We investigate (1) which institutions and bodies are perceived by members as being the most influential; and (2) perceptions of the way in which power and influence should ideally be distributed. Drawing on this information, we then discuss whether there is a propensity for change in today's Riksdag.

On an a priori basis, there is good reason to believe that members' perceptions of power are likely to be related to macro-level factors, such as party grouping, as well as to more micro-level factors relating to characteristics the individual MP. With respect to the former, for example, members of opposition parties have limited influence over decision-making, which may color their thinking about power in the Riksdag. For similar reasons, members of anti-establishment parties, like the Left Party and the Greens, are more likely to be dissatisfied with existing conditions.

As for individual-level characteristics, backbenchers, newcomers to the parliament and perhaps female members may look differently on the workings of the Riksdag than do frontbenchers, veterans and male members. After all, a revolution from below was instigated against the procedures of the House of Commons in the 1970s (e.g. Norton 1980a; Judge 1981).

Perceptions of power, of course, are to some extent rooted in the factual distribution of power and influence. As a baseline for our analysis, we will therefore compare organizational structures of the Riksdag with parliaments other western democracies. Which features of the internal workings are unique to the Riksdag, and which are shared by most parliaments?

Data used in this article were collected within the framework of a larger study on the role of the Riksdag within Swedish representative democracy. Shortly after the 1988 parliamentary elections, mail questionnaires were distributed to all 349 elected members of the Riksdag. The response rate was 96 percent. In the spring of 1991, complementary personal interviews were conducted with a subsample of 20 MPs. The data analyzed here are basically from the 1988 study.¹

Institutional Centers of Power

Obviously, committees and party groups are possible centers of power and
influence in most parliamentary bodies, but there are other alternatives as
well. Party leaders, or specialist staff personnel, for instance, may also be

influential power-holders. Yet another possibility, usually associated with 19th century politics, is that parliament is dominated by individual members, who make decisions without ties to parties and committees. Finally, we need to take into consideration the relationship between parliamentand government. Even though we are dealing with the internal workings of parliament, power and influence exercised by the government are of crucial importance.

With respect to committees, scholars have identified a number of organizational which affect the relative power of a specific committee system, among these being size, the degree of specialization, permanence, stability of membership, and departmental parallelism. Thus, small, specialized committees with a stable membership and departmental parallelism are more influential than committees having the opposite characteristics (Shaw 1979; Olsen & Mezey 1991b).²

By comparative standards, the Swedish committee system can be placed in a middle position in terms of its power and influence over political decision-making. Because of their limited size, permanent status, relative specialization and departmental anchorage, the Riksdag's committees, together with the committees of the German Bundestag, are ranked as more influential than the corresponding institutions in the French National Assembly and the British House of Commons, although less powerful than the committees of the US Congress (Shaw 1979, 395-404; Olson 1980, 255-257; 1984, 204-208).

Parliamentary parties constitute a parallel and competitive institution to the committees. Decisions made in parliamentary party groups frequently serve to restrict committee autonomy. A rule of thumb is that power is centralized by strong parliamentary parties and decentralized by strong committees (e.g. Loewenberg & Patterson 1979).

Parliamentary parties can, in turn, be more or less hierarchal. In hierarchically parties, decision-making is concentrated in a small number of dominant leaders. Less hierarchal parties are characterized by a more even distribution of influence over decision-making among party members (e.g. Jewell 1973).

In Sweden, as in other parliamentary democracies, parliamentary parties hold a strong position in the legislature. As to the internal distribution of power, the Swedish parties have in international comparisons been placed in a middle position, with substantially less centralized processes of decisionmaking are found in British parliamentary parties for example.³ The domestic Swedish debate has followed other lines, however. "Party leader cult" is an expression that has been used to describe power conditions within the parliamentary parties - a phrase indicating strongly centralized decision-making and substantial influence on the part of the party leaders (Gahrton 1983).

Closely associated with party development is the principle of parliamentarism the government's strong position in relation to the legislature. Bryce (1921) introduced his noted thesis on the "decline of parliaments" in the 19205, there has been discussion as to whether there exists any power to be distributed within the legislatures. Critics allege that Western parliaments function solely as rubber stamps for governmental policy. Yet, in recent years the dominant standpoint has been that parliamentary do in faet have a certain independent power over decision-making.

Sweden has here again been placed in a middle position in international comparisons. As with the majority of Europe's parliaments, the Swedish Riksdag has been classified as a "reactive legislature" as opposed to an "active legislature" like the US Congress.⁴ More detailed Swedish studies have emphasized that the Riksdag has gained strength since the 19605, thanks in large part to developments occurring in reaction to weak minority governments (Isberg 1982, 1984; Arter 1984; von Sydow 1990; Sjolin 1991; Damgaard 1992).⁵

In contrast to committees, parties and governments, the staff is a body of decision-makers within parliaments that is relatively little researched. Parliamentary parties and committees all over the Western world employ a large group of specialized staff in order to manage the ever-growing number of complex tasks. With their knowledge and central positions in the bureaucracy, these experts have at least a potential influence over decision-making. The phenomenon of staff as a hidden power holder has received the greatest attention in the US, but is found in other Western parliaments as well, including the Swedish Riksdag.⁶

Assessing the Distribution of Power

One part of the story behind the cautiousness of scholars to analyze the overall distribution of power within parliament is the difficulty in finding criteria for comparison. It is perplexing enough to compare the power of two bodies having the same function. Arriving at a reasonable measurement for the amount of power of groups and bodies having different functions is even more worrisome a task. Researchers concerned with parliaments often wisely avoid specifying the criteria used in making their assessments.⁷

One feasible proceeding is of course to make use of the subjective reputational method. In his book Congressmen's Voting Decisions, Kingdon (1973) tries a variation of this subjective method by asking representatives what influenced their voting decisions and whether they had been in contact with various actors. Kornberg & Mishler used a more traditional reputational method by first asking Canadian members of parliament to

identify a number of influential persons and then analyzing their formal
positions (Kornberg 1967; Kornberg & Mishler 1976).8

Our approach differs from Kingdon's and Kornberg & Mishler's not only in that we focus on members' perceptions of power rather than on factual conditions, but also in that our measurement technique concentrates directly on the groups and bodies in question. In our survey, the members were asked to estimate the influence of the following eight groups and bodies on decision-making in the Riksdag, using an 11-point scale reaching from 0 (very low influence) to 10 (very influential): "the Standing Committees", Committee Chiefs of Staff", "Parliamentary Party Groups", "Parliamentary Party Leaders", "Parliamentary Party Group Leaders", "Parliamentary Party Chiefs of Staff", "the Government" and "Individual Representatives".⁹ The members were asked to mark how much influence the groups/bodies in question actually have within the parliament and also how much influence they should appropriately have.¹⁰

Our method has the advantage of offering comparisons on a unified scale, at least in theory. We left it up to the members themselves to define the concept of "influence over decision-making in the Riksdag"; no detailed instructions were given. Furthermore, the measurement technique makes it possible to cover a broad spectrum of conceivable power sources.¹¹ Although our simplified technique provoked certain objections on the part of the members, an absolute majority of MPs (between 218 and 275 depending on the group or body in question) agreed to give their views on the distribution of power within the Swedish Riksdag.¹²

Parliament According to Parliamentarians

Swedish MPs do not see themselves as kings of the parliamentary castle (see Table 1). In their view, it is the government and not any internal body or group that is most influential. Party leaders are second, followed by the committees and the parliamentary party groups. Farthest down on the list are the individual representatives - lower than both the parliamentary party staff and the committee staff. Thus, representatives see a Riksdag steered from the top. It is the government and the party leaders who control parliament, while individual representatives have even less to say than do the chiefs of staff.

Our interpretation, of course, is somewhat forced. The members primarilysee government-dominated Riksdag, but beyond this the differencesbetween influence of the party leaders, committees and parliamentary party groups are not particularly large. The Riksdag is controlled from the top, but such traditional collective bodies as the committees and the parliamentary party groups are also perceived to

have considerable impact. When we look at the bottom of the hierarchy,
however, it is evident that, above all, the party staff is perceived as having
significant power in relation to individual members.13

Members' perceptions of the distribution of power in the Riksdag both confirm and modify earlier studies of factual conditions. Pictures of a dominant government, strong parties and a relatively powerful committee system have been painted before. What is new is the impression of strict control from the leadership within the parties, the relatively influential position of the party staffs, and perhaps also the strong position of the committees vis-å-vis the party groups. Especially remarkable in relation to the international literature is that decision-making in the Riksdag is strongly associated with a small number of party leaders. It would seem that the Swedish party leaders have a perceived authority in parliament virtually equal to that of the almost omnipotent British Prime Ministers.

The scenario described is common to all parties (see Table 2). The government is perceived as most powerful by members of all parties, followed by either the party leaders (Social Democratic and Liberal members) or the committees (Left Party, Center, Conservative and Green members). Parliamentary party groups tend to fall into the fourth position in the various rankings. Only Social Democrats see the parliamentary party groups as being more influential than the committees. The parties also agreed on the lower end of their rankings. Individual members, for example, are perceived as least influential among all parties except among Social Democrats, who rank the committee chiefs of staff as having less influence.

Thus, on the whole, the results show great consistency in the perceptions of power. Complete consensus does not exist, but a calculation of the rank correlations demonstrates rather impressive similarities along party lines. The average rank correlation between the different views reaches +0.90, with +1.0 as the highest value (Center and Conservative members) and +0.76 as the lowest (Social Democrats and Greens).

One interpretation of the results is that MPs tend to differ in their perceptions of ideologically loaded aspects of the political world, but have similar perceptions of political procedures. Kingdon (1973) found only small differences between Democrats and Republicans in how they viewed the distribution of power in the House of Representatives - evidence that is consistent with our results.

Other studies, however, speak against this interpretation in terms of ideological and procedural issues. Depending on whether they belong to the party in or out of government, Canadian MPs show radically different perceptions of the influence of the committees over governmental policy (Rush 1979, 229). Scholars have also shown that members of different

parties have different perceptions of which norms are basic for work within
the parliament (Searing 1982; Kim & Patterson 1988; Hedlund 1985).
Obviously, there is need for more evidence before we can solve this matter.

The Ideal Parliament

According to the members, the greatest power in the Swedish parliament should be held by the party groups and the committees. The ideal Riksdag would thus be dominated by two traditional, collective institutions with, one may suppose, a capacity for mutual control (see Table 3). Today's primary holders of power, the government and the party leaders, should also play a role, but with significantly less influence than at present. The idea of a parliament dominated by independent individuals, however, receives only weak support; Members as individuals rank sixth in the order of suitable holders of power. Influence of individual members, it would seem, should be channeled primarily through party groups and committees.

The resistance to staff influence is evident in all parties; the members' views are that party and committee chiefs of staff should be considerably less influential than other groups in the Riksdag. There is a somewhat greater understanding for the influence of politically appointed party chiefs of staff, however, than for the non-political heads of the committees.¹⁴

Agreement between parties is not as great over who should have power as it was over perceptions of the actual distribution of power. Thus, as is evident from Table 4, views regarding ideal conditions are more divided along partisan lines than were perceptions of current conditions.¹⁵ It is

mainly the Greens, and the Leftists - i.e. the anti-establishment parties - that demonstrate divergent normative views of power. Other parties, in contrast, share similar notions of how the distribution of power in the Riksdag should be arranged.

As a political movement, the Greens emphasize the importance of breaking with traditional political organizations, and this ideological standpoint also reflected in members' view of the parliament. In accordance with the philosophy of minimizing the concentration of power, representatives the Greens would allocate the least amount of influence to party leaders in their ideal Riksdag. They prefer a Riksdag in which power is wielded by individual representatives and committees rather than one steered from the top by a handful of powerful leaders. It should be noted, however, that our survey was performed shortly after the Green Party's entry into the Riksdag in 1988. In our interview study in the spring of 1991, after three strenuous years of everyday parliamentary work, prominent representatives of the Greens showed greater sympathy for traditional forms of political organization.¹⁶

Members of the Left Party also have an individual-oriented view of power in the parliament. Individual representatives move to third place on the list of preferred holders of power, with average points almost as high as those among the members of the Greens (657 as compared with 670). Party leaders and the government are similarly ranked low in the desired hierarchy (in fifth and sixth place respectively). Indeed, the Left Party members' ideal Riksdag is formed more after the new left-wing soft principle after the democratic centralism of the old communist tradition.

Level of Dissatisfaction

The members of the Greens are most dissatisfied with the perceived distribution of power within the Riksdag. As shown in Table 5, the distance between norm and reality is clearly greater for representatives of the Greens than for representatives of the established parties.¹⁷ The average difference between perceptions and normative positions concerning power distribution among the Greens' representatives is more than twice that of any other party, with a relation between the ranking of actual and desired power among the investigated groups and bodies of nearly zero (rho = +0.10).

For the other parties, the greatest dissatisfaction is noted among members
of the Left Party, while Center Party members are the most content. Social
Democrats, Liberals and Conservatives show roughly the same results,

especially if one considers that the government was included among the bodies studied. Since the government was Social Democratic at the time of the study, members of the Social Democratic party are more likely to have been positive toward governmental influence. Members of the other parties, in contrast, graded the government lower on the scale.

Comparisons between perceived and desired power can also be made for each of the groups or bodies individually and for the individual MPs. By looking at the data from this angle, we can analyze whether the members feel that the different groups/bodies should have more or less influence, or whether the degree of influence is satisfactory. Such an analysis gives an idea of how the MPs would like to change the distribution of power in the Riksdag, and which representatives feel most urgently that a change should take place (see Table 6).¹⁸

The results suggest that representatives do not want a wholesale change. In only three instances do a majority of representatives want change to occur - i.e. individual representatives' influence should be increased (61 percent), while the influence of governments and parliamentary party chiefs of staffs should be decreased (60 percent and 51 percent respectively). Furthermore, there is relatively strong support for increasing the influence of the party groups and the committees (37 percent and 32 percent respectively). substantial proportion of the MPs would also like to decrease the influence of the standing committee chiefs of staff, the party leaders and the party group leaders (44, 40 and 26 percent respectively).

This analysis clarifies the earlier message: the MPs would like to have a

greater say, both as individual representatives and via the party groups and the committees. At the same time, they would like to decrease the power of government and party leaders. A relatively large number of MPs also feel that the influence of the chiefs of staff should be limited.

The results are essentially the same among the different parties, with the exception of the Greens (see Table 7). In their efforts to reform parliamentary work, the majority of Green members would like to decrease the influence of all groups and bodies other than individual representatives. Of particular note here as well is the weak Left Party support for increased committee power. Members of small parties with few representatives in committees tend in general to be less interested in strengthening the influence of these bodies. The greatest support for strengthening the influence the committees is found in the large Social Democratic Party.

As regards the question of institutional change, one cannot interpret our data as signs of a pending revolt. The results in Table 8 show that members whom one might expect to be inclined to upset traditional power relationships the Riksdag generally agree with the more established members. However, the differences that do exist run in the expected direction. The Riksdag's backbenchers would like to increase the influence of individual members, committees and party groups more than the frontbenchers.¹⁹ A similar pattern also emerges among female members and among newcomers to the Riksdag, who are particularly eager to decrease the influence of the party leaders and the government.²⁰ Yet, it should also be noted that newcomers feel less of a need than senior members to increase the influence of the committees, which may be explained by the fact that the newcomers have fewer permanent committee seats and less experience of committee work than their more established colleagues.

The same pattern can be found in the British House of Commons, where backbenchers showed increasing tendencies to protest against control from the party leadership during the 1970s (Norton 1980a, 1980b, 1985; Judge 1981; Wood & Jacoby 1984; Franklin et al. 1987). These protests were one of the factors which led to the decision in 1979 to create a number of new select committees (Drewry 1985; Marsh 1986). The relatively limited dissatisfaction among Swedish backbenchers may have to do with the fact that the Riksdag's 16 permanent standing committees offer nearly all its 349 members the opportunity to carry out meaningful work from the very start of their parliamentary careers.

Another important explanation for the limited discontent is the strong position of the parties. For the great majority of MPs, party loyalty is the most important norm. Career-oriented members also know that they generally need the support of their party in order to accomplish any particular goal. By comparison, British MPs have strong, if now somewhat dwindling, party loyalty as a basic political norm (Crowe 1983, 1986;

Rasmussen 1988; Patterson 1989; cf. Kornberg 1967; Cayrol et al 1976). It has been shown in earlier studies (e.g. Holmberg & Esaiasson 1988; Sannerstedt 1982; Sjolin 1993) that the party is also central for Swedish MPs, and analyses in this article clearly indicate that the party groups, together with the committees, are granted key positions in the ideal Riksdag. Thus, there is reason to look more closely at the Swedish parliamentarians'relationship their respective party groups.

Party Groups in the Center

As may be expected, MPs' perceptions of their own party group are more positive than their perceptions of party groups in general. The three analyses presented in Tables 9, 10 and 11 confirm that Swedish parliamentarians a good relationship with their party group and that they do not shrink from the idea of party power.

A very clear majority of members (79 percent) feel that the executive of their own party group has a satisfactory amount of influence over decisionmaking(Table An even greater majority (83 percent) are happy with the emphasis on cohesion and discipline within their respective parties. In fact, more members would prefer that party cohesion be increased rather

than weakened (Table 10). Nearly all members feel that they have a very good (50 percent) or fairly good (48 percent) chance of influencing their party's position within their own special field. The possibility of exerting influence is judged to be somewhat less outside these special fields, although a majority of members still feel that it is good even then (Table 11).

Differences along party lines are on the whole small. There is some tendency for the representatives of the larger parties to be less satisfied than other representatives. Social Democrats (with 151 representatives) and Conservatives (with 64 representatives) judge themselves to have the greatest difficulty in gaining attention for their standpoints within the party. The Conservatives, furthermore, have the greatest number of members who feel that their party executive has too great an influence in the party group, while ordinary members have too little. A relatively large number of Conservative representatives would also like to see a decrease in the demand for discipline within the party.

The size of the party is not the only possible explanation here, however. In the 19605, the Conservatives were the most individually oriented representatives, it is conceivable that this tradition has lived on to a certain degree.²¹ As regards Social Democrats, their position as the party in government is an alternate explanation for the members' perception that it is difficult to gain attention in their party.

Another difference along partisan lines is that 25 percent of the Leftists
would like to sharpen discipline within their party, thus revealing internal
conflict between traditionalists and reorganizers.22 Note also the Greens'

almost unanimously positive attitude toward each aspect of the work within
their own party group. In all probability, this unanimity reflects a new
parliamentary party's initial enthusiasm for its tasks.

These limited differences should not conceal the major result, however, i.e. that the majority of members of all the parties find themselves at ease with the distribution of power in their own party groups.²³ The parties have a very strong position among members. It may well be appropriate to speak of party decline among the voting population, but the trend within parliament is the opposite.²⁴

Our analyses, moreover, allow us to highlight the meaning of the members' wish to strengthen the influence of the party groups in the Riksdag. Here, the key concepts are specialization, division of work and shared responsibility. The party groups afford the members an opportunity to work within a specialized political field in which most see themselves as having the chance to gain attention for their ideas (Table 11). A further demonstration of the members' positive attitude toward specialization, division of work and shared responsibility is that the committee groups of the party, in which members of related committees hold regular discussions with one another, receive very strong support in all parties (Table 9).²⁵

A similar philosophy concerning the need for joint efforts and an equal distribution of responsibility has been demonstrated among British and Canadian MPs (Judge 1981; Jogerst 1991; Kornberg 1967). The concept of team work, where all members are given their individual tasks and the leader's right to decide is limited, seems to enjoy broad support among most Western parliamentarians. While corresponding data are lacking for the US Congress, it is highly unlikely that American legislators see a strengthening of party groups as an appealing reform proposal (Fenno 1973; Mayhew 1974; Frances 1985).

Summary and Discussion: Power in the Riksdag

In the eyes of the members, the Riksdag is an institution which is controlled from the top and where the greatest power is held by the government and the leaders of the parties. However, the standing committees and the party groups are collective bodies that offset this control. When compared with analyses of the distribution of factual power and influence, our study shows that Swedish parliamentarians ascribe stronger influence to the party leaders than is normal. The members also see the staff of the parties as being relatively powerful.

Overall, members agree along partisan lines on how to view power in
the Riksdag. It is mainly anti-establishment parties like the Greens and the
Left Party that differ. The party with the most divergent views, the Greens,

is also alone in seeing political organization as an integrated part of their ideology. The great consensus observed also applies to individual members who could be expected to have differing notions, such as backbenchers, female members and newcomers to the Riksdag. It would not be difficult to read into the results characteristics that are usually associated with "Consensual Democracies" - shared values and, most importantly, strong parties.

Swedish parliamentarians have clear wishes with respect to their ideal Riksdag. The standing committees arid the party groups are the two bodies which should in the first place be given greater strength, while the influence of primarily the government and the party leaders should be decreased. The Riksdag should be a working parliament for full-time professional politicians, where everyone plays his or her part. The traditional Westminster in which most MPs are content with supervising the government., and where there is an inherent opposition between parties and standing committees, receives weak support among Swedish parliamentarians.²⁶ The positive view of cooperation within collective organs also implies a dissociation from an individual-oriented American model. Swedish MPs do not wish to function as independent legislators.

In addition to the empirical results, our study has two major implications - one having to do with strategies for institutional analysis of parliaments, the other with the current debate over the need for curtailing the power of parliaments. On the first point, our strategy of asking for members" perceptions of the distribution of power within the Riksdag has proven to be useful. By allowing the protagonists to speak for themselves, we gain complementary insights about the workings of parliament. We have learned, for instance, that members tend to agree with the often heard description that the Riksdag is an elite-dominated institution. Of course, the analytical value of survey data of this kind will continue to grow as we gain further points of reference from repeated studies.

The second implication of our study is related to the calls heard in many western democracies for a strengthening of the position of the executives and a curtailment of the power of legislatures. Such views are heard both in the new democracies of Eastern Europe, and in the countries of Western Europe as well as in the US. The debate on this matter, moreover, is not only restricted to politicians; one can also find it among legislative researchers.

In Sweden, the Lindbeck Commission (1993) recently suggested a series of constitutional reforms with the purpose of limiting the possibilities for the Riksdag to interfere with propositions presented by the government. These proposals would, among other things, reduce the number of MPs by half, tighten up the budgetary process, and concentrate decision-making power in economic matters to the Finance Committee.

From our results it is clear that attempts to increase the influence of the government will run into trouble in the Riksdag. Members of all parties, Conservatives and Social Democrats alike, would prefer a radically different Self-assertive MPs want to lead, they do not want to follow; they do not want a parliament led by the government, they want a government led by the parliament. In essence, this is nothing new. This is a continuation of the world-wide and ongoing conflict between the legislative executive arms of government.

---

NOTES

1. For detailed information on the studies, see Holmberg & Esaiasson (1988).

2. Mezey (1991, 209) mentions certain reservations regarding the factor of departmental parallelism: "An additional caution is suggested by the case of the West German Bundestag where a strong committee system that parallels the bureaucratic structure does not seem to be associated with significant levels of parliamentary activity because of the decisive role of political parties." This objection, however, is not convincing. Under the assumption that we wish to isolate the significance of the organization of the committees, the relevant question is what significance the factor of departmental parallelism has in controlling for the influence of parties.

3. Olson & Mezey (1991b, 13); Arter (1984, 198-208); cf. Loewenberg and Patterson (1979, 125-140); von Beyme (1986); Norton (1981, 26-46); Rose (1986).

4. See Mezey (1979, chs 2 and 5). The Costa Rican Congress also belongs to the exclusive group of active legislatures in Mezey's classification. Other nations with influential parliaments are the Philippines, Uruguay, Chile, Italy and France during the Third and Fourth Republics. The list of reactive legislatures include, in addition to Sweden, United Kingdom, Canada, Australia, New Zealand, Norway, Denmark, Finland, Germany, Belgium, France and the Netherlands. Employing stricter frames of reference, German Bundestag is one that has been classified as more influential than the British House of Commons (Loewenberg & Patterson 1979).

5. A related comment in this context is that Bryce's study has increasingly taken on the character of a "straw man". It is considerably more controversial today to maintain that parliaments lack influence than to take the opposite position.

6. De Gregario (1988) and Aberbach (1990) are two of the most recent studies dealing with this topic. For an overview of the literature, see Hammond (1985).

7. Mezey (1979, 23) addresses the fundamental question of what type of power and influence is at issue here: "Although not always stated explicitly, such statements (about the strength or weakness of particular legislative institutions) usually refer to the importance of the legislative in the policy-making process relative to the importance of nonlegistlative institutions ..." For an unusually clear textbook approach to this subject, see Keefe (1980, 91-137). Cf. also Daalder & Rusk (1972, 165-8).

8. Eulau (1962) employs a similar method for evaluating the significance of individual factors such as "respect," "affection" and "expertise." Caldeira and Patterson (1988) and Hibbing and Thomas (1989) demonstrate new areas of use for the reputational method.

9. The category "Standing Committee Chiefs of Staff" refers to the collective of chief officials within the standing committees. Likewise. "Parliamentary Party Chiefs of Staff" refers to the chief executive for the different parliamentary parties' staffs. The "Parliamentary Party Leaders" are the same as the rive persons who stood at the head of their parties in 1988. It is customary in Sweden for the same person to be chairman for both party branches, as long as he is not also a part of the government, in which case the leadership of the parliamentary party is assigned to someone else (Esaiasson 1985, 1-12). Only the Greens refuse to designate one party leader, preferring instead one male and one female group representative who lack the traditional authoritative position of party leader. The "Parliamentary Party Group Leader" is the next highest official within the party, but is not nearly as well known outside the party as the party chairman.

---

10. The question read as follows: "Using the scale below, please state your view of the following two aspects of the influence of the groups and bodies given below on decision-making in the Riksdag: (a) their actual influence; (b) the influence you would consider appropriate."

11. A numberof studies have been made using similar points of departure. Jackson (1979), for example, asked Philippine MPs who had the greatest influence over commiltee decisions- the cornmittee chairman; committee members; the President of the Republic; speaker/President of the Senate/floor leaders; the staff of committees; or interest groups. See also Rush (1979); Frances & Riddlesperger (1982); Frances (1985).

12. The average internal refusal rate was higher than for other controversial questions in the survey, for example concerning the piacement of parties, voters and the members themselves on the Green dimension (on an average, 27 percent as compared with 9 percent). Also, the members were somewhat less willing to indicate their normative views about the Riksdag than to assess present conditions (average internal refusal rate of 33 percent as compared with 22 percent). After this predominantly positive evaluation, we would also like one critical voice to be heard. The academician and former MP John Mackintosh gives the following bitter view of the British representatives' ability for self-analysis: "This gap between what people say about themselves and what is the case is intensified by the faet that most performers in politics are poor observers; they can perform but not: describe the process" (quotation from Judge (1981, 187)).

13. The difference between the party leaders and party groups is significant at the 0.10 level. Other differences are either strongly significant (0.000-level) or clearly insignificant. The influence of the party chiefs of staff shall be seen in the light of the faet that the majority of actions on the part of the representatives, e.g. the introduction of private members' bilis, must be cleared in the party staff. For further detail, see Isberg (1982) and Arter (1984).

14. All group compansons named here are strongly significant (0.000-level).

15. We stumble again upon a non-trivial result. Analyses undertaken earlier regarding members' views of power in society at large (Holmberg & Esaiasson 1988) similarly show that agreement in cognitions is not always greater than agreement in norms.

16. The Swedish Greens have to a large exient taken their organizational principles from their sister party in Germany, die Grunen. This party has wrestled with how to handle the balance between ideological principles and parliamentary effectivity in the Bundestag as well (see Poguntke 1987; Frankland 1988).

17. The results shown in the table are based upon a totaled and averaged comparison of how much influence the members perceive the eight groups/bodies to have and how much they would prefer them to have. The interpretation of the measure is simple. The larger the distance between perceived and preferred reality, the larger the dissatisfaction and the desire to rectify the situation.

18. Members who gave higher points for preferred influence than for perceived influence have been classified as taking the position that the group in question should have greater power. Likewise, members who gave lower points for preferred power than for perceived power have been classified as taking the position that the group's or body's influence should be decreased. In the case of tied scores, the members' views have been logged in the middle category, "satisfactory as it is".

---

19. Frontbenchers are defined as committee chairmen, speakers of the Riksdag, and members of the party group executives.

20. All representatives of the Greens are included in the group of "newcomers". The conclusion does not change, however, if the analysis is carried out solely among newcomers from other parties.

21. Holmberg (1974, 243-249), on the basis of four different interview questions concerning toward one's own party, classified 35 percent of the Conservative MPs in 1969 as individual-oriented "Mavericks". The corresponding proportion in other parties varied between 11 percent (Social Democrats) and 21 percent (Liberals).

22. The party in which the greatest number of representatives would like to ease party discipline is the Liberal party. In our interview study in 1991, several representatives also called attention to the fact that the party leadership for the Liberals has unusually strict rules concerning e.g. the introduction of bills. The comments can be seen as an indication that our survey questions capture something politically essential.

23. The MPs' generally positive attitude toward their party groups was confirmed in the interview study in 1991. Three of four interviewees expressed their satisfaction with work within the party group. Only the Left Party Communist representatives, and to a certain degree the Green representatives, were dissatisfied, and this dissatisfaction then applied only to a lack of cohesion and leadership within the party. Perhaps not surprisingly, however, more than half of those interviewed had negative perceptions of the power distribution and internal workings of other party groups.

24. On the Swedish Parties' loosening grip over voters, see Gilljam & Holmberg (1993); Granberg & Holmberg (1988).

25. On the parties' committee groups in the Riksdag, see Isberg (1982) and Arter (1984).

26. It is now unusual openly to support the Westminster model. However, for a fervent defense using New Zealand as an example, see Jackson (1987).

ACKNOWLEDGEMENTS

Research for this article was supported by a grant from The Bank of Sweden Tercentenary
Foundation. We thank John Geer, Mikael Gilljam and Rick Herrera for valuable comments
on an earlier version of this paper.

REFERENCES

Aberbach, J. D. 1990. Keeping A Watchful Eye: The Politics of Congressional Oversight.
Washington DC: Brookings Institution.

Arter, D. 1984. The Nordic Parliaments: A Comparative Analysis. London: C. Hurst.

von Beyme, K. 1986. "The Role of Deputies in West Germany", in Suleiman, E. N. ed.,
Parliaments and Parliamentarians in Democratic Politics. New York: Holmes & Meier.

Bryce, J. 1921. Modern Democracies. New York: Macmillan.

Caldeira, G. A. & Patterson, S. C. 1988. "Contours of Friendship and Respect in the
Legislature", American Politics Quarterly 16, 466-485.

Cayrol, R., Parodi, J. C. & Ysmal, C. 1976. "French Deputies and the Political System",
Legislative Studies Quarterly 1, 67-100.

Crowe, E. 1983. "Consensus and Structure in Legislative Norms: Party Discipline in the
House of Commons", Journal of Politics 45, 907-931.

Crowe, E. 1986. "The Web of Authoritarian: Party Loyality and Social Control in the British
House of Commons". Legislative Studies Quarterly 11, 161-185.

Daalder & Rusk 1972. "Perceptions of Parties in Dutch Parliament", in Patterson, S. C &
Wahlke, J., eds., Comparative Legislative Behavior. New York: John Wiley.

Damgaard, E., ed., 1992. Parliamentary Change in the Nordic Countries. Oslo: Universitetsforlaget.

De Gregario, C. 1988. "Professionals in the U.S. Congress: An Analysis of Working Styles",
Legislative Studies Quarterly 13, 459-476.

Drewry, G., ed., 1985. The New Select Committees: A Study of the 1979 Reforms. Oxford:
Clarendon Press.

Esaiasson, P. 1985. Partiledarna infor vdljarna. Partiledarnas popularitet och betydelse for
valresultatet. Goteborg: Goteborgs universitet, Statsvetenskapliga institutionen. Forskningsrapport

Eulau, H. 1962. "Bases of Authority in Legislative Bodies: A Comparative Analysis",
Administrative Science Quarterly 7, 309-321.

Fenno, R. F. 1973. Congressmen in Committees. Boston: Little. Brown.

Frances, W. L. 1985. "Leadership, Party Caucuses and Committees in U.S. State
Legislatures", Legislative Studies Quarterly 10, 241?—257.

Frances, W. L. & Riddlesperger, J. W. 1982. "'U.S. State Legislative Committees: Structure,
Procedural Efficiency and Party Control", Legislative Studies Quarterly 7, 453-471.

Frankland, G. E. 1988. "The Role of the Greens in West German Parliamentary Politics
1980-87", The Review of Politics, 99-122.

Franklin, M., Baxter, A. & Jordan, M. 1987. "Who Were the Rebels? Dissent in the House
of Commons, 1970-1974", Legislative Studies Quarterly 12, 143-156.

Gahrton, P. 1983. Riksdagen inifrån. En studie av parlamentarisk handfallenhet infor eti
samhålle i kris. Stockholm,: Prisma.

Gilljam, M. & Holmberg, S. 1993. Vdljarna infor 90-talet. Stockholm: Norstedts.

Granberg, D. & Holmberg, S. 1988. The Political System Matters: Social Psychology and
Voting Behavior in Sweden and the United States. Cambridge: Cambridge University Press.

Hammond, S. W. 1985. "Legislative Staffs", in Loewenberg, G., Patterson, S. C. & Jewell,
M., eds., Handbook of Legislative Research. Cambridge, MA: Harvard University Press.

Hedlund, R. D. 1985. "Organizational Attributes of Legislative Institutions: Structure, Rules,
Norms, Resources", in Loewenberg, G., Patterson, S. C. & Jewell, M., eds.. Handbook
of Legislative Research. Cambridge, MA: Harvard University Press.

Hibbing, J. R. & Thomas, S. 1989. "The Modern United States Senate: What is Accorded
Respect", Journal of Politics 52, 126-145.

Holmberg, S. 1974. Riksdagen representerar svenska folket. Empiriska studier i representativ
demokrati. Lund: Studentlitteratur.

Holmberg, S. & Esaiasson, P. 1988. De folkvalda. En bok om riksdagsledamoterna och den
representativa demokratin i Sverige. Stockholm: Bonniers.

Isberg, M. 1982. The Role of the Unicameral Riksdag in Sweden During the 19705. Stockholm:
University of Stockholm, Department of Political Science.

Isberg, M. 1984. Riksdagens roll under 1970-talet. Stockholm: Akademiforlaget.

Jackson, R. J. 1979. "Committees in the Philippine Congress", in Lees, J. D. & Shaw. M..
eds., Committees in Legislatures: A Comparative Analysis. Durham, NC: Duke University
Press.

Jewell, M. E. 1973. "Linkages Between Legislative Parties and External Parties", in Kornberg,
A., ed., Legislatures in Comparative Perspectives. New York: David McKay.

Jogerst, M. 1991. "Backbenchers and Select Committees in the British House of Commons:
Can Parliament Offer Useful Roles for the Frustrated?", European Journal of Political
Research 20, 21-38.

Judge, D. 1981. Backbench Specialisation in the House of Commons. London: Heinemann.
Keefe, W. J. 1980. Congress and the American People. Englewood Cliffs, NJ: Prentice-Hall.

Kim, C. L. & Patterson, S. C. 1988. "Parliamentary Elite Integration in Six Nations",
Comparative Politics 20, 379-399.

Kingdon, R. 1973. Congressmen's Voting Decisions. New York: Harper & Row (first edition).

Kornberg, A. 1967. Canadian Legislative Behavior. A Study of the 25th Parliament. New
York: Holt, Rinehart and Winston.

Kornberg, A. & Mishler, W. 1976. Influence in Parliament: Canada. Durham, NC: Duke
University Press.

Lindbeck, A. 1993. Nya villkor for ekonomi och politik. Stockholm: Fritzes. SOU 1993:16.
Loewenberg, G. & Patterson, S. C. 1979. Comparing Legislatures. Boston: Little, Brown.

Marsh, I. 1986. Policy Making in a Three Party System: Committees, Coalitions and Parliament.
London: Methuen.

Mayhew, D. R. 1974. Congress: The Electoral Connection. New Haven: Yale University
Press.

Mezey, M. L. 1979. Comparative Legislatures. Durham, NC: Duke University Press.

Mezey, M. 1991. "Parliaments and Public Policy: An Assessment", in Olson, D. M. &
Mezey, L. M., eds., Legislatures in the Policy Process. The Dilemmas of Economic Policy.
Cambridge: Cambridge University Press.

Norton P. 1980a. Dissension in the House of Commons 1974—79. Oxford: Oxford University
Press.

Norton, P. 1980b. "The Changing Face of the British House of Commons in the 19705",
Legislative Studies Quarterly 5, 333-357.

Norton, P. 1981. The Commons in Perspective. New York: Longman.

Norton, P. 1985. "Behavioural Changes: Backbench Independence in the 1980s", in Norton,
P., ed., Parliaments in the 1980s. Oxford: Basil Blackwell.

Olson, D. M. 1980. The Legislative Process. A Comparative Approach. New York: Harper
& Row.

Olson, D. ML & Mezey, L. M., eds., 1991a. Legislatures in the Policy Process. The Dilemmas
of Economic Policy. Cambridge: Cambridge University Press.

Olson, D. M. & Mezey, M. L. 1991b. "Parliaments and Public Policy", in Olson, D. M. &
Mezey, M. L., eds.. Legislatures in the Policy Process. The Dilemmas of Economic Policy.
Cambridge: Cambridge University Press.

Patterson, S. C 1989. "Understanding the British Parliament", Political Studies 37, 449-462.

Poguntke, T. 1987. "The Organization of a Party: The German Greens", European Journal
of Political Research 15, 609-633.

Radice, L., Vallance, E. & Willis, W. 1987. Member of Parliament. The Job of a Backbencher.
London: Macmillan.

Rasmussen, J. 1988. "Will I Like It the First Time? The Maiden Speech In the British House
of Commons", European Journal of Political Research 16, 527-544.

Rose, R. 1986. "British MPs: More Bark Than Bite?", in Suleiman, E. N., ed., Parliaments
and Parliamentarians in Democratic Politics. New York: Holmes and Meier.

Rush, M. 1979. "Committees in the Canadian House of Commons", in Lees. J. D. &
Shaw, M., eds.. Committees in Legislatures: A Comparative Analysis. Durham, NC: Duke
University Press.

Sannerstedt, A. 1982. Forhandlingar i Riksdagen. Lund: Lund University Press.

Searing, D. E. 1982. "Rules of the Game in Britain: Can the Politicians Be Trusted?",
American Political Science Review 76, 239-258.

Shaw, M. 1979. "Conclusion", in Lees, J. D. & Shaw, M., eds., Committees in Legislatures:
A Comparative Analysis. Durham, NC: Duke University Press.

Shaw, M. 1990. "Members of Parliament", in Rush, M., ed., Parliament and Pressure Politics.
Oxford: Clarendon Press.

Sjolin, M. 1991. "The Decline of Parliament-tesen och den svenska riksdagens makt under
1970- och 1980-talet". Statsvetenskaplig Tidskrift 94, 125-148.

Sjolin, M. 1993. Coalition Politics and Parliamentary Power. Lund: Lund University Press.

von Sydow, B. 1990. "Parlamentarismens utveckling", in Att styra riket - regeringskansliet
1840-1990. Stockholm: Allmånna Forlaget.

Wood, D. M. & Jacoby, W. G. 1984. "Intraparty Cleavage in the British House of Commons:
Evidence from the 1974-1979", American Journal of Political Science 28, 203-223.