Resumé

Ever since the electoral turmoil of the early seventies, Scandinavia has been a market for alternative models. And, indeed, the left-right conflict dimension has fallen short of accounting for everything as evidenced by, for instance, the recent breakdown of the three-party bourgeois coalition government in Sweden.

This paper focuses on the conflict structure in Swedish politics, mainly the
genesis of the center-periphery dimension.

Methodological ambiguity in diffusion studies of the center-periphery conflict is illustrated through some empirical data. It is argued that the Green Wave should not be seen as a manifestation of an allegedly latent center-periphery cleavage, nor be construed as a sign of major transitions in Swedish politics. Explanations are offered as to why the comparatively strong counter-cultural movements (the free churches and the temperance movement) never managed to secure a party of their own.

Of late, Scandinavia has become a market for alternative models. Even in Swedish politics the left-right dimension has fallen short of accounting for everything, as evidenced by, for instance, the recent breakdown of the three-party bourgeois coalition government.

Whatever ideological and/or tactical reasons lay behind this demise it may serve as a starting point for a historical analysis of the Swedish conflict structure, especially the center-periphery cleavage. Three approaches to the center-periphery conflict will be dealt with in the subsequent analysis: the diffusion, economic, and cultural models.

---

* Helgeandsholmen is a small island in the Stockholm Stream where the Parliamentary Building was raised in 1887-1906 to accommodate the bicameral Riksdag. Upon unicarneral reform, the building was abandoned in 1970. After the interlude in the glass and concrete structure at Sergei Square the Riksdag will reconvene at Helgeandsholmen in the early 1980s, when the reconstruction of the old Parliamentary Building is scheduled to be completed.

1. The Setting

The center-peri])hery and the left-right dimensions - conflicts which in intensity and chronology are sharply reflected in Danish and Norwegian party political history - do not readily unfold in Sweden. The agrarian Venstre movements never had a counterpart in Sweden. While Denmark and Norway experienced a determined conservatism with economic liberalism as the guiding light, Swedish conservatism was a patriarchalism of civil servants, manufacturing tycoons, and freeholders.

Notwithstand ng these differences, the final product was to become quite comparabh to the party systems in the rest of Scandinavia. So there is indeed some grounds for suspecting the center-periphery dimension of also having had an influence in Swedish politics.

Using the left-right model, Carlsson (1963) has explained 60 per cent of the regional var ance in the Socialist parties' strength between 1911 and 1940. These findings also hold for postwar elections (Gustafsson 1974, Lewin et al. 197;!). When attitude data from the 1976 election survey were subjected to factor analysis the left-right dimension yielded 56 and the center-periphery dimension 26 per cent of variance explained (Petersson 1977, 72). Considering the mood of the recent elections it is reasonable to play down the center-periphery cleavage as an element structuring the vote to somewhere below 26 per cent. Religion - one of the cultural aspects of the center-periphery conflict-has stood for merely 4 per cent of variance explained in party alignment, given social class (Särlvik 1974, 419). We leave t at that.

What about the center-periphery dimension's interplay with the Swedish conflict, left vs. right? That question puts a finger on the weak spot of multidimensional analysis: overlapping conflict concepts. Working-class voting for socialist parties must be regarded as stemming from the left-right division, irrespective of whether the behavior is discovered in heavily indus trialized cities or in the most unreachable of backwood areas (cf., Rose & Urwin 1975, 6f). Hinterland socialism has, however, been viewed alternatively as a product of left vs. right or center vs. periphery, or as a combination of them both. Empirically, it is difficult to assess the weight of two dimensions that are at an angle somewhere between 0 and ')0 degrees (see Figure 3 below).

Sweden and Norway come extremely close on a number of counts. Values and ideals in the electorates are strikingly similar (Petersson & Valen 1978). The number of members in the free churches is about the same across the border (Särlvik 1974,417, Valen & Rokkan 1974,331). Still, the countr es are way apart in terms of center vs. periphery. Table 2

bears witness to this. Membership in a free church congregation has a more profound impact on electoral behavior in Norway compared to Sweden. In Norway the Christian party wins 35 per cent of the religious voters, whereas the Christian party in Sweden obtains only 13 percent. So the Norwegian Labor Party is much worse off than the Swedish Social Democrats among Non-Conformists. Similarities in background variables on the macro-level should thus serve as a caution against the use of few and specific factors in explaining cross-national differences.

2. Diffusion and Center-Periphery

Political phenomena are sometimes seen as a function of the spread of impulses from the national center(s) to the surrounding countryside. Occasionally, the perspective has been broadened. For instance, the favorable position of the Liberal Party in Gothenburg and its vicinity is said to be the net result of the city's exposure to Great Britain. In a similar vein, Malmö is thought of as the Swedish intermedium of German socialism channeled via Copenhagen.

However, the exceptions from a pattern of concentric spread of ideas are obvious by a glance at the regional strength of the Socialist and the Liberal parties (see Figures 4 A and B). In Figure 4 C the class structure of the constituencies has been reproduced as an independent variable intended to reduce or remove the imbalances in the working class parties' support. As brought out by the residuals, the Social Democratic Party and the Radical Left were unduly strong in the constituencies of Norrbotten, Southern Norrland, Kopparberg, and in Skåne inclusive of the Four Cities. A thorough testing of the strictly spatial approach on center vs. periphery would require a very elaborate apparatus, not yet at our disposal.

// is apparent though, that the findings at the regional level are at
odds with the precepts of the initial model.

Carlsson has given the diffusion theory a fairer chance to prove its fruitfulness. His empirical data comprised the Agrarian Party's growth up to 1936 in a socioeconomically homogeneous part of a Mid-Swedish constituency. Even though Agrarian strongholds and weak spots formed clusters of precincts, they were too randomly scattered to comply with a diffusion-process pattern.¹

The task is to explain why a certain party has unequal strength in areas with identical class structure. Assessing the import of impulses for party regionalism then requires controlling for the economic variable. Reality may very well suggest an interplay between the independent variables (economy and impulses), but an overlapping of the spatial and states factors should be avoided. By way of example, it is likely that socialist 'oases' are a function of (1) an early industrialization of the region, making the objective necessities favorable, which in turn made for (2) a fertile ideological milieu for continuing industrial expansion; the trade unions (read: innovative agents) were lobbying for more jobs in the adjacent areas. On the other hand, the piety of Småland may have hindered the economic modernization process, thus hindering parties representing class interests. To pursue, then, an inquiry of why Småland acquired lay churches and temperance movements while, say, the constituency of Västmanland acquired industries is of course feasible. But that would somewhat absurdly imply that the determinant of party regionalism is to be found in the climate, topography, mineral structure, etc., of the regions. Geographical analyses of partisan strength in the French genre often end in such theses, e.g., that agrarian parties are not likely to gain much advantage in big cities. Undoubtedly, such a statement is to the point, but, alas, it is trivial.

Statements about the effects of diffusion processes on party regionalism, with firm requirements about the causal linkages, are therefore very vulnerable unless the analysis is circumscribed to areas with a homogeneous socioeconomic setting. Moreover, such analyses also have to include the activity of other, often competitive movements in the region in order to pin down the dynamics behind the imbalances. For instance, it is very unlikely that the Conservative Party adopted a passive posture vis-å-vis the Agrarian mobilization in the constituency referred to in the Carlsson study. So, in effect, the diffusion theory is up against very serious problems. Hardly any of the entries in the diffusion equation are conducive to being held constant: there is just too much feedback in the society.

Turning to the concrete, we will now consider the content of the party
map (Figure 4 C), a document which displays that everything in Swedish
politics cannot be reduced to a matter of left and right.

3. Economy and Center-Periphery

The economic center-periphery dimension can be considered as a conflict between the feeders and the bleeders. Political rallies in the preindustrial era drew on the farmers' (the feeding periphery's) resistance against the urban trade trusts, bureaucrats, and tax collectors. Facing the recession of the primary sector, the landed interest - the now bleeding periphery - had to take defensive measures to safeguard the remnants of its domains.

Only the Liberal Party explicitly set out to stand above the industrial class structure, summarily depicted as cells 1,3,4, and 6 in Figure 1. The preindustrial conflict has been added to that figure along the center-periphery dimension, cells 2 and 5 respectively.

Upon the abolition of the Four Estate Diet in 1885-86 the agrarian body
(cell 5) coalesced into the Yeomen's Party in order to counterbalance the
court and moneyed interest (Back & Berglund 1978). The Yeomen's Party

was to face resistance from the 'intelligentsia' (cell 2), later known as the Center Party, a label which served to distinguish it from the aristocrats of the Upper House. The bicameral representational system per se had the effect that the turnover of political groupings during the late 19th century defies enumeration. The more undemocratic of the chambers, the Upper House, lagged at least 25 years behind the Lower House in a socioeconomic and ideological reflection of the society at large. In general, however, there were '. . . gentlemen vs. farmers and peasants vs. urbanities' (Hadenius et al. 1975, 17).

At the turn of the century social changes in Sweden were unprecedented in Scandinavia (Lafferty 1971, Kuhnle 1975). In terms of political transition there was a polarizing development (see Figure 2). The Social Democrats organized the urban working class which grew increasingly numerous through the migration of the rural proletariat, tenants, and leaseholders (cell I+4). In addition, emigration overseas made cell 4 in Figure 1 less important as a basis for social and political conflicts. Facing a mounting Social Democratic and Liberal demand for universal suffrage, the urban entrepreneurs and farmers (cells 3+6) were united in 1912 in v.he National Party and the Yeomen's and Burghers' Party of the Upper and Lower Houses respectively. In Figure 2, cells 2 and 5 have been left outside the typology, partly for illustrative purposes. However, the digits could have been placed in the bourgeois box and be seen to form the platform of the Liberal Coalition (1900) (cf. section 4 below).

Three major party splits occurred between the three-party era and the 'freezing' of the five-party system in 1934. The founding of the Communist Party was a branch-off on the left-right continuum. The split down the middle of the Yeomen's and Burghers' Party in 1913 gave birth to the Agrarian Party (known as the Center Party since 1957 it will consistently be referred to as the Agrarian Party in this paper) and the Conservative Party. Add to that the Liberal split of 1923 on the issue of prohibition, and there are intimations that the center-periphery cleavage (2 vs. 5 in Figure 1) was still alive.

As late as 1920, forty-four per cent of the population lived off the primary sector, which exceeded the figures for Denmark and Norway (Berglund & Lindström 1978, 22). Within this large farming cohort the;re were, not unexpectedly, multi-faceted sentiments. As for an alignment with the conservative forces, there was certainly not a consensus among the farmers. Some 80 per cent of the acreage consisted of smallholdings. During the off-season many of these smallholders were on the payrolls of forest companies and lumber yards owned by people who were seen as

representatives of the increasingly stronger industrial right-wing of the Yeomen's and Burghers' Party. Furthermore, the farmers had ambivalent feelings about the high tariff policy preached by the Conservatives: those who produced for the dairy industries (most of the farmers in Northern Sweden) advocated a low tariff profile on fodder commodities.

The raison d'etre of a genuine agrarian party was to leave the farmers with a higher net income - their leading political motive. But the objective of raising profits could have been achieved just as easily through a horizontal split of the farmer movement. By force of their number the smallholders held the key to the majority balance in the Riksdag. And by thrusting the Social Democrats into government responsibility the smallholders would have a hold on that party for payoffs in one form or another. As for the big farmers of Mid and South Sweden, they were much more at ease with the Conservatives' protectionist policy than with an alliance with 'radical' farmers of the North. Events partly turned out this way. The Swedish farmers' party was to win only a halfhearted response from its class. Up to the 1956 election the Agrarians received only half of the farming vote: the remaining half was shared by the Conservatives (25%) and the Liberals and Social Democrats (25%) (Petersson & Särlvik

Undoubtedly the social and cultural Weltanschauung of the farmers was neatly incorporated into the traditionally Conservative trinity based on the undisputable role of the Family, the Church, and the Nation. But in practice, the Agrarian ideology was not all that sacrosanct. In fact, it even turned out to be compatible with a 'crisis coalition' with the Social Democrats in 1933. That marked the beginning of a twenty-five year period of intermittent Agrarian-Social Democratic parliamentary wheeling and dealing.

By supporting some Social Democratic welfare bills the Agrarians could feel safe that the generous appropriations to the primary sector would not be discontinued. The Agrarian Party's bourgeois allies were liable not only to impose free enterprise (de) regulations in agricultural business, but - still worse for the uncompetitive state of Swedish farming - to apply for membership in the EEC. Not until the 19605, when the Agrarian constituents had become a group of prosperous farmers, could the party return to the bourgeois fold (ef., Berglund & Lindström 1979 on the concentration of land ownership in the Nordic countries).

The flux of Agrarian policy underscores that the party is not a product of
a peripheral peasant culture. In a Scandinavian context populär traditions
among the Swedish farmers are almost non-existent. Sweden does not

have any peasant folklore that even comes close to the aura attached to names like Jaabæk and Grundtvig in Norway and Denmark respectively (Østerud 1978, 250 ff.). Instead, the Agrarians' ideological point of reference has been that the ends justify the means. The economic status of the center-periphery dimension is summarized in Figure 3.

In graph A the center-periphery dimension has a strong negative correlation with the left-right dimension. The solution represents the time when the farmers were organized in the Yeomen's and Burghers' Party, and when the Social Democratic Party was a clear-cut urban phenomenon. From the time of the founding of the Agrarian Party up to the 'crisis coalition' with the Social Democrats in 1933, the center-periphery dimension rotated clock-wise; the two dimensions became strongly positive correlated. During this process the Conservatives were to become the party of the urban bourgeoisie and high-ranked civil servants, whereas the Social Democrats and the Agrarians joined forces in an attempt to increase the demand for farming commodities through public works, unemployment insurances, etc. (Tingsten 1967 I, 319 ff, Hellström 1976, 484 ff).² Recent developments have once again made graph A appropriate as a reflection of the center-periphery dimension's relationship with the leftright cleavage. Figure 3 serves as a reminder that the center-periphery dimension has never been a stable cross-cutting cleavage (c-p) in Swedish politics.

The 1956 election marks a watershed in the history of the Agrarian
Party. Of its all-time low share of the vote (9%) farmers accounted for 77

per cent. Twenty years later this class stands for only 19 per cent of the total Agrarian vote (Petersson, 1977, 14). Instead of vanishing from the electoral arena the Agrarians have experienced their heyday in a society where farmsteads nowadays are also tourist attractions.

The Agrarians did not rise to prominence during the early 1970s by riding on a center-periphery cleavage, let alone on an economic centerperiphery conflict. Today's agriculture is a well-established business. Indicative of the enterprise code is the farmers' demand at the 1977 Agrarian Convention that the ban on the pesticide DDT should be lifted. Another issue on which the farmers take a pragmatic stance is the nuclear power debate. As far as the farmers were concerned the Agrarian Party could have accepted the energy policy of the Liberal and Conservative parties. Instead, the Agrarian leadership preferred to break the first bourgeois government in 44 years. This decision was, of course, taken as a measure to safeguard the non-agricultural part of the Agrarian Party. Without the nuclear power issue in the forthcoming electoral campaign (September, 1979) the party would stand without a distinctive profile with all that that entails in terms of retaining those Agrarian voters who, not surprisingly, have been comparatively uncommitted to their party (Petersson & Särlvik 1975, 72, Petersson 1977, 265 ff).

Though the nuclear power issue is a cross-cutting, center-periphery cleavage (Holmberg et al., 1977, Holmberg, 1978) it has been of secondary importance for the majority of the electorate. Only a handful of the Social Democrats with a negative attitude towards nuclear power actually switched party allegiance. For the overwhelming majority pocket-book matters, i.e., left vs. right, still ranked as top priorities. Furthermore, if the energy dimension had been superior to the economic cleavage the Communist Party - the only party to have an untarnished anti-nuclear reputation after October, 1976 when the Agrarians succumbed to taking a nuclear power station into production - would have stood to gain in the subsequent opinion polls. Things, however, did not turn out that way.

Eventually the energy topic will disappear from the political agenda. When it goes it will be accompanied by the lingering Green Wave which has been given a few extra years to survive thanks to the nuclear debate. As a dim naturalistic populism the Green Wave has been offered as a panacea against the drawbacks of postwar modernization in general and the changes of the sixties in particular.

Not counting the ideological framework of the few organic communes
where the Green Wave philosophy has been practised, the movement had
many attributes of a bourgeois heritage with few difficulties in being

analyzed in a left-right perspective. In short, those to be attracted to the Agrarian Party in the early seventies were white-collars and workers v/ith a non-working class family background, workers in less populous areas, workers in small businesses, recently enfranchized voters, and, above all, former Liberal laborers (Petersson & Särlvik 1975, Petersson 1977, Back & Berglund 1978, 129). The hard-core of the Social Democratic Party - middle-aged unionized industrial workers living in middle-sized cities - remained stable in its support. In formalized language, this development went towards reducing the explanatory power of the left-right model. But this does not necessarily imply that the void has been filled by a centerperiphery cleavage. As of now, the vacuum has only signified the obvious: the residual has had a momentum in the electorate. During the fifties the Liberal Party recruited floating voters en masse. That flux was not seen as the beginning of any new conflict dimensions. The case in favor of a reevaluation of the Swedish model remains unimpressive.₃

4. Culture and Center-Periphery

In this section we will seek for political indicators of a cultural centerperiphery cleavage. Here the center represents the secular, no-nonsense way of life whereas the periphery is supposed to house counter-cultural movements such as lay churches and temperance organizations.

Arithmetically, an 'ethical' party was as conceivable as an agrarian party at the turn of the century. In 1910 more than 10 per cent of the adult population - 500,000- were members of a free church congregation and/or a temperance lodge (Lundkvist 1977, 67). In relation to electoral turnout in 1911 - 600,000 - the political potential of the two popular movements was quite remarkable. One may conjecture that the idealists not only were entitled to vote, but that they also took advantage of their citizen right.

By and large, the Liberals became the party of the idealists. Paradoxically enough, what opposition there was against the ideas preached by the free churches and the temperance movement happened to choose the very same party as its platform. After the 1934 reunion of the liberal factions the Liberal Party was just as unholy an alliance as it had been prior to the 1923 split over prohibition. During the postwar years the party became more and more secularized: today its electoral profile is genuinely urban. Did the conflict dimension based upon the free churches and the temperance movement perish with the demise of the pro-prohibitionist party, de frisinnade (which incidentally coincided with the forming of the Christian People's Party in Norway)?

Although the proportional strength of the two popular movements has decreased somewhat since the 19205, they cannot be ignored. However, the prohibitionists never managed to cultivate the values of the ethical movements into a conflict sufficiently cross-cutting to constitute a salient dimension. In fact, a distinct secular pole has been missing from the politico-cultural axis. Sweden has had nothing reminiscent of the Norwegian Conservative Party's strong opposition against restrictions on alcohol. In the 1976 electoral survey the 'piety factor' accounted for three per cent of the variance explained. The dimension did not discriminate very much among the parties. Somewhat surprisingly, what differences there were showed the Social Democrats to be more orthodox than the Liberals (Petersson 1977, 109).

The reason why there has never been an anti-clerical, anti-teetotalist front is that all parties except the Communists have had ties with the free churches and the temperance organizations. True, Social Democracy voiced an anti-religious propaganda in its infancy, 'but it soon became less conspicuous and less representative' (Tingsten 1967 11, 290). In the early twenties Christian Social Democrats founded a within-party association, the Fraternity Movement. In 1972 this counted close to 9,000 members. On the party executive board one chair has been reserved for the Fraternity Movement. The relationship between the lay church and Social Democracy dates back to the time when the party was mobilizing the masses. At that time the political activity of the free churches and the temperance lodges was straightforward. They called upon their members to vote for candidates who, regardless of party affiliation, were in favor of the movements' demands. Especially prior to the enactment of proportional representation (PR) in 1909, the agitation of the two ethical movements turned out to be crucial for the electoral outcome; they could tip the balance by siding with one of the two candidates running for office. The relationship between the free church and Social Democracy should not be overemphasized; but the fact remains that the Social Democratic Riksdag caucus did include a few religious members prior to 1920. As for the early ties with the temperance lodges the Social Democrats were fully committed. No ideological ambivalence can be traced to this cooperation, which, of course, had been the case when the party was confronted with sacral concerns. The struggle against alcohol was part and parcel of working class liberation. During the 1910s more than 80 per cent of Social Democratic MP's were teetotalists (Lundkvist 1977, 175 ff).

The Liberal Party was to face competition in the electoral arena as well
in its attempt to become the sole contender for the moral cause. This is

hinted at by the ecological correlations in Figure 4. It is also apparent that the two movements were not as interwoven as to make for a united political bloc. The center of gravity of the free churches was located in the southern parts of the country, whereas the temperance movement was at its strongest in southern Norrland. In spite of the relative weakness of the lay church in Västerbotten (the second northernmost constituency) during the twenties, the Liberal Party was by far the most dominating political force. In Småland, on the other hand, the party did quite well thanks to the position of the temperance movement. In the constituency of Gävleborg the Social Democrats were astonishingly well-off in view of the counter-cultural movements' strength. These ecological interpreta

tions run counter to what has been unraveled in analyses based on sample
surveys of the last two decades (ef., Särlvik 1974, 415-419, and Table I
above).

The established party system per se has been another factor working against the manifestation of a moral conflict dimension. Only with utter reluctance have the parties brought religious questions onto the floor. The third public investigation into the separation of the Church from the State is currently looked upon as a fatally contaminated object, which precludes that the investigation will appear as a government bill in the forseeable future. Likewise, alcohol matters always produce temporary strains in party cohesion since the temperance movement is still overly represented in Parliament.

As a result of the suppression of moral issues among the established
parties the Christian Democratic Union was formed in 1964. At that time
the process of secularization was about to gain impetus; ten years later the

moral disarmament had gained further inroads, such as the legalization of abortions and pornography. However outraged the Christian Democrats were, they never managed to transform the protest into votes. As of late the Christian party has entered negotiations with the Agrarians about forming electoral alliances in the forthcoming (1979) election. This is yet another example of how moral conflict potentials are frozen. The inferior status of the cultural dimension in relation to the left-right cleavage is also visible in the electoral arena. A majority of the bourgeois voters prefer to see their children taught in a Christian spirit at school, many bourgeois voters cannot reconcile with free abortions, and so forth. However, ideals of the Christian party may turn out too costly to support. A vote for the CDU may be a vote wasted since the party may fall short of the four-percent clause for parliamentary representation: the action may turn out counterproductive to the intention.

5. A Scandinavian Conclusion

When returning to the intriguing differences between Norway and Sweden in terms of center vs. periphery one is well advised to look for historical explanations. By and large, the postwar period stands out as an era of continuous prolongation of the immediate past.

With the 1814 Constitution, the Norwegian peasantry was enfranchized. Thereafter those with a stake in politics had everything to win by organizing the farmer (= one vote). And, indeed, the roster of Norwegian peasant leaders was to become impressive. Their opposition againsX Kristiania - with all that the Capital represented in terms of tax burdens, bureaucracy, servility to the king in Stockholm, and contempt vis-å-vis the farmers - gained relatively widespread support among individual farmers. The Nynorsk movement, which defined the written standards of the backwoods dialects, was but one of the rural societies to be founded in these years.

Nothing of the kind existed in Sweden. Here the franchise was restricted up to the abolition of the Estates, of which one had been the farmers' forum. And this representational format had not encouraged mass participation. Besides, the extension of the suffrage was far from complete. Approximately 23 per cent of the rural population met the norms between 1872 and 1900 (Hadenius et al. 1975, 295). In fact, it took the Swedish farmers one hundred years to acquire equal footing with their Norwegian brethren. Turnout statistics are informative in pointing to participation discrepancies: around 1880 some 50 per cent of the adult

rural population in Norway took part in the national election, whereas the comparable figure for Sweden was 5 per cent (Østerud 1978, 217, 220). In effect, the Norwegian - as opposed to the Swedish peasantry - was already socialized when the Industrial Revolution knocked on the door.

At long last, when the Swedish rural class was awakened out of its political coma, it was recruited to defend the status quo, which marks yet another contrast to Norway where for most of the time the farmers represented the outsiders in politics. Therefore the Norwegian farmers heavily relied on their own leaders in improving agriculture as a business. With the founding of the Norwegian Farmers' Association, Norsk Bondelag, in 1896, co-ops, etc., were added to the already existing cultural and political networks. In Sweden agricultural organizations are of a much more recent vintage (1929) and they are predominantly geared to economic concerns. Along with the gradual adaptation and receding of the original cleavage, the organizational framework has gained in importance as a cue-giver, i.e. the economic center-periphery dimension more and more became a function of the established peripheral institutions (producer co-ops, etc.), rather than vice versa. So, quantitative longitudinal data on the strength of the agrarian apparatuses in the two countries are badly needed.

Between 1910 and 1915 Sweden was mobilized in the proper sense of that word; turnout soared up to 70 per cent, and the Social Democratic membership figures were closing in on the 100,000 mark (Berglund & Lindström 1978, 28). Since this boost in participation occurred in the wake of the most transitional period in the economy, and given the farmers' idle record, political alignments followed class interest above anything else. Norway, on the other hand, entered the modern era with a /ay polar political structure very neatly mirroring the economic and cultural divisions at work (ef. Valen & Rokkan 1974, 333). The Liberal Party first and foremost represented the countryside, whereas its Swedish sister party also had a viable urban component. Consequently, the Conservative Party was primarily urban; its Swedish sister party remained semi-rural for many years to come. In addition, Labor in Norway toyed with revolutionary rhetoric (and thus irreparably repelled many former and potential voters in the teetotalist and, particularly, in the free-thinking communities?), whereas the Swedish Social Democrats were concerned not to lose support within the moral movements.⁴

The successful build-up of the Christian People's Party between 1933
and 1945 constitutes the deviant case in the Scandinavian partisan context.
As mentioned in section 1, it is more than sheer macro-level arithmetics

that account for the rise of that party. The Christian party has its roots in the county of Hordaland in Western Norway. Until very recently the party was meagerly represented beyond the South-West. In this part of the country the counter-cultural movements have had some fifty per cent 'coverage' of the population (Valen & Rokkan 1974, 336 f), which should be compared to the even regional distribution of free church and temperance membership in Sweden (see Figures 4 D and E).

In the South-West milieu, the Christian party's rationale was obvious. Furthermore, there was no doubt as to which party was bound for defeat in the event of a Christian momentum in the electorate. And in due course the Christian People's Party was to replace the Liberal Party as the spokesman of the periphery's cultural values. In contrast, the more low-keyed Christian Democratic Union, and the prohibitionist party before it, have had to compete with four parties in their attempt to become the voice of the moral cause. That made the CDU programs somewhat ambiguous. In effect, the party refrained from siding with the bourgeois bloc, and was thus left in a kind of limbo.

By the very nature of the setting, two cases and a multitude of potentially relevant variables, the above comparison should be seen as tentative. It serves to draw attention to the historical as opposed to the contemporary in explaining why the center-periphery dimension has carried so little weight in Swedish politics: (1) Sweden has close to no tradition of peasant mass politics, (2) the spread of the counter-cultural movements has been cross-nationally homogeneous, which (3) constituted something of a barrier against the free church and the temperance movement being monopolized by one single party.

---

NOTES

1 A comparable analysis of the spread of the Finnish Rural Party between 1959 and 1970 in North Carelia yielded only a low .27 gamma coefficient. Y = party strength in the 283 precincts, X = year of founding of local party organization, 1959-1969 (Hautamäki & Sänkiaho 1971, 12,22).

2 The accurate description of the dynamics behind the 'crisis coalition' is that the two parties approached each other, which is to say that the left-right axis in graph B rotated counter clock-wise.

3 For a more elaborate discussion of multi-dimensional analysis, see Holmberg (1978, 68), and Berglund & Lindström (1978, 169-173, 1979). That the left-right cleavage defends its top position is underscored by recent opinion-poll analyses. Those who since the 1976 election have defected from the Agrarian Party to the Social Democrats are primarily workers (Dagens Nyheter, October 15, 1978).

4 True, there was not an ideological conflict between being a teetotalist and a radical Socialist - on the contrary (cf. Lafferty 1971, 257 ff). Sociologically, however, the dynamics behind Labor radicalism and teetotalism is more complex. By and large, those who were members of a temperance lodge - and a free church congregation, for that matter - were often well-to-do workers and middle-class citizens. As such they might have been repelled by Labor joining the Comintern.

4 True, there was not an ideological conflict between being a teetotalist and a radical Socialist - on the contrary (cf. Lafferty 1971, 257 ff). Sociologically, however, the dynamics behind Labor radicalism and teetotalism is more complex. By and large, those who were members of a temperance lodge - and a free church congregation, for that matter - were often well-to-do workers and middle-class citizens. As such they might have been repelled by Labor joining the Comintern.

REFERENCES

Back, P-E., and Berglund, S., 1978. Det svenska partiväsendet. Stockholm: AWE/Gebers.

Berglund, S., and Lindström, U., 1978. The Scandinavian .Party System(s). Lund:
Studentlitteratur.

Berglund, S., and Lindström, U., 1979. 'The Scandinavian Party System(s) in Transition(?),'
European Journal of Political Research. In press.

Carlsson, G., 1963. 'Partiförskjutningar som tillväxtprocesser,' Statsvetenskaplig Tidskrift,
2-3, 172-213.

Gustafsson, G., 1974. Partistyrka och partistyrkeförskjutningar. Lund: Studentlitteratur.

Hadenius, S., Molin, 8., and Wiesiander, H., 1975. Sverige efter 1900. Stockholm: Aldus/
Bonniers.

Hautamäki, L., and Sänkiaho, R., 1971. 'Some Features About the Support of the Finnish
Rural Party in Years 1962-70,' Fennia, 109.

Hellström, G., 1976. Jordbrukspolitik i industrisamhället. Stockholm: LTs Förlag.

Holmberg, S., 1978. 'Kärnkraften och vänster-högerdimensionen,' Statsvetenskaplig
Tidskrift, 2, 67-79.

Holmberg, S., Westerståhl, J., and Branzén, K., 1977. Väljarna och kärnkraften. Stockholm:
Liber.

Kuhnle, S., 1975. Patterns of Social and Political Mobilization: a Historical Analysis of the
Nordic Countries. London and Beverly Hills: Sage.

Lafferty, W. M., 1971. Economic Development and the Response of Labor in Scandinavia.
Oslo: Universitetsforlaget.

Lewin, L., Jansson, 8., and Sörbom, D., 1972. The Swedish Electorate 1887-1968. Uppsala:
Almqvist & Wiksell.

Lundkvist, S. 1977'. Folkrörelserna idet svenska samhället 1850-1920. Uppsala: Almqvist &
Wiksell.

Østerud, 0. 1978. Agrarian Structure and Peasant Politics in Scandinavia. Oslo: Universitetsforlaget.

Petersson, O. 1977. Väljarna och valet 1976. Stockholm: SCB/ Liber.

Petersson, 0., and Särlvik, B. 1975. 'The 1973 Election', General Elections 1973, 3, Stockholm:
National Bureau of Statistics.

Petersson, 0., and Valen, H. 1978. 'Politiska konfliktlinjer i Sverige och Norge.' Paper.
Nordic Political Science Congress, Bergen.

Rokkan, S. 1975. 'Sentrum of periferi, økonomi og kultur: modeller og data i kilometrisk
sosiologi,' Studier i historisk metode, 10. Oslo, Bergen, Tromsø: Universitetsforlaget.

Rose, R., and Urwin, D. W. 1975. Regional Differentiation and Political Unity in Western
Nations. London and Beverly Hills: Sage.

Särlvik, B. 1974. 'Sweden: The Social Basis of the Parties in a Developmental Perspective,'
in Rose, R., ed., Electoral Behavior. New York: Free Press.

Tingsten, H. 1967. Den svenska socialdemokratins idéutveckling, I-11. Stockholm: Aldus/
Bonniers (Paperback Ed.).

Valen, H., and Rokkan, S. 1974. 'Norway: Conflict Structure and Mass Politics in a European
Periphery,' in Rose, R., ed., Electoral Behavior. New York: Free Press.