The positivist and behaviorist emphasis in mid-twentieth-century political science has often seemed in conflict with an older historical and legal-institutional approach. Yet the work of Dahl, Deutsch, Huntington, Rokkan, and others shows that precision of language, explicit testing of hypotheses, and even quantification can be combined with historical depth and attention to institutions.¹

In coming decades the most fruitful work in political science may well come to be done in this combined sphere. As Dahl has noted, the "behavioral revolution" came to an end just because its ideals of precision in theory and empiricism have been acknowledged on all sides.² Meanwhile, a number of developments have combined to remind political scientists of the limitation of a purely contemporary and behavioral analysis. The end of Western colonial rule in Asia and Africa and the global involvement of their government drew the attention of American political scientists to the Third World and hence to processes of political transition. As European scholars began to take up the systematic study of politics in larger numbers, they brought to the field a richer historical perspective and a sharper awareness of the ideological dimensions of politics. Even in the United States, where premature obituaries had celebrated The End of Ideology,3 the illusion of technocratic consensus gave way to the Vietnam War, the race question, and the student risings of the sixties. As a result American social scientists once again became alive to questions of value that a misunderstood Weberianism had long banished.

The result has been a new attention to change as a central theme of politics. The pure behaviorists (one is tempted to exclaim) used to explain the world, with this bit of jargon or that bit of computerized correlation; the point, however, is to change it. To help produce, or for that matter avert, change on any large scale we must seek explanations that account for causal links rather than mere correlations, that envisage developments over decades rather than phenomena of the day, that combine imaginative hypothesizing with careful testing, and that acknowledge once again the purposive nature of human action.⁴

The recent literature on democracy reflects this methodological situation. Phenomena

such as the collapse of the Weimar Republic in Germany and the Third and Fourth Republics in France had prompted a search for the conditions of stable democracy, but this search, among American scholars at least, commonly led to functional correlations between democracy and a number of simultaneous social, economic, or psychological factors. Lipset nominated affluence, urban residence, and high levels of education as "requisites of democracy," whereas Almond and Verba discovered the basis of stable democracy in a mixture of "participant," "subject," and "parochial" attitudes that they labeled the "civic culture." The former hypothesis was derived from a tabulation of aggregate national statistics for the early or mid-1950'_5, the latter from a single, ambitious sample survey of political attitudes in the United States, Great Britain, Germany, Italy, and Mexico.s Eckstein, in a set of hypotheses illustrated on the example of Norway, suggested that democracy will thrive in societies where structures of authority prove the more nearly democratic the more directly they impinge on processes of government.⁶

The most widely held assumption probably is that democratic stability depends on the existence of a consensus among the citizenry. This notion among American political scientists was first formulated by Abbott Lawrence Lowell at the turn of the century and has since that time become part of the commonly accepted lore of the profession.7 There has been some debate, however, on the exact nature of the required consensus, with one school seeing a need for substantive agreement on some set or other of fundamentals, and another restricting the necessary agreement to the rules of democracy themselves.⁸ There is also disagreement whether the necessary consensus must be held by the entire citizenry, by the majority, or only by a political elite tied to the common citizenry through parties and other organizations.⁹ Finally there has been a small but growing group of authors who have rejected consensus as a requirement of democracy and have instead stressed the potentially creative and integrative nature of political conflict.lo

Several things are noteworthy about the prevailing behaviorist explanation of democracy. It is, first and foremost, a functional rather than a genetic explanation. It focuses on democracies assumed to be already in existence and has little or nothing to say about the process by which an oligarchy or dictatorship might be transformed into a more democratic regime. The favorite evidence adduced is synchronic and contemporary, evidence which by its very nature cannot solve any problems of causal explanation. For example, even if Lipset's data are accepted at face value,¹¹ they leave it entirely open whether affluent and well-educated citizens make the better democrats or whether, on the contrary, democratic government encourages economic development and more intensive efforts at public education. The assumption throughout most of the literature is that all the causal relationships flow from economic and social antecedents to political effects, or else from opinions and attitudes to political actions and institutions.

Insofar as the prevailing theory touches on questions of historical development at all, the model envisaged is that of unilinear, cumulative progress in a single steady direction. This is a natural consequence of the preoccupation with linear statistical correlations.

The Marshall Plan of the Truman years and the Alliance for Progress of the Kennedy period reflected this notion of democracy as a linear function dependent on economic or social factors. Any contributions to industrial recovery in Europe or public education in Latin America would bring a gain pro tanto for democracy. Or as Lipset phrased it, "the higher the education,. .. the better the chances for democracy"; and "the more well-to-do a nation, the greater the chances that it will sustain democracy."12 Such generalizations were rarely combined with any inquiry into the circumstances that made a country deviate from the statistical mean. This aversion to particulars reinforced the prevailing unhistorical attitude.

It is true, of course, that Almond and Verba in writing about Great Britain, Eckstein about Norway, and Dahl about New Haven did provide some discussion of the evolution of the social or psychological patterns that they analyze. Their major propositions, nonetheless, remained couched in the present tense.¹³

In short, the American literature has typically dwelled on the economic, social, psychological, and historical backgrounds but rarely taken up the dynamic processes that connect these to the political foreground. The danger with such a view is that it will leave democracy free-floating and disembodied, remote from any possibility of causal knowledge or purposeful action. This in turn can easily lead to an attitude of self-congratulation on behalf of those countries that already have achieved stable democracy, and of resignation and quietism in contemplating the less happy condition of those that have not. For example Eckstein has spoken of "those calamitously improbable combinations of circumstances that make democracy work," and even Dahl stresses that "legal party opposition ... is a recent unplanned invention" - a finding that might have come as a surprise to generations of democrats and liberals who fought precisely for the legalization of opposition and political criticism _l4

Fortunately there has been a reexamination of some of the functionalist and behaviorist assumptions within the field of democratic theory, just as there has been in political science more generally. Deane Neubauer has substituted the notion of a threshold for the earlier assumption of statistical unilinearism. Quantitative researchers like Flanigan and Fogelman and Arthur Banks have turned to historical statistics.¹⁵ Lijphart has presented a sophisticated analysis of the Dutch political system as one of conflict and accommodation.l6 Binder and others have described the origins and workings of political pluralism in Lebanon.¹⁷ And Dahl himself, in his recent works, has adopted a more historical perspective, tracing the evolution of party opposition on both sides of the Atlantic and making "Conflict and Consent" the dual theme of his interpretation of American constitutional government from the days of the Philadelphia Convention to the present.¹⁸

2

The process of transition to democracy is today of considerable practical interest.
Many countries of Latin America and Asia have recently moved from oligarchy

(military rule, one-party government, colonialism) to a government based on popular choice between freely competing parties; in others, oligarchy and democracy are engaged in a recurrent seesaw struggle. In Eastern Europe, tendencies toward liberalization, and perhaps future democratization, were apparent in the 1950's and 60's - until they were squelched by the Russian occupation of Czechoslovakia in 1968. In settings such as these, it is clearly not enough to ask the functional question (What forces can sustain a democracy already in existence?); rather, we must ask the genetic question: What brings a democracy into existence? But a tenable answer to the second question may well shed further light on the first. Democracies do not lose their dynamic quality once established. Moreover, if democracy cannot be sustained it may have to be reestablished all over again. From such a perspective, the question as to the forces that make for a transition to democracy also assumes importance for countries like the United States and those of Western Europe, where the transition to democracy was made generations ago, but where important segments of youth and of the intellectuals now challenge its current workings.

Practical considerations aside, the question of transition to democracy holds genuine theoretical interest. Social scientists have learned a great deal about the process of modernization conceived as a broad and interrelated set of social changes.¹⁹ Karl Deutsch has offered a suggestive set of hypotheses on the origins of nationality in new patterns of social communication.²⁰ Political scientists, psychoanalysts, and others have contributed to a theory of leadership as a dynamic process.²¹ Sociologists have investigated the integrative as well as the destructive aspects of conflict, and economists such as Albert Hirschman have presented the formulation of major economic policies as a historic process of trial, error, and cumulative learning.22 Leon Festinger's theory of cognitive dissonance has its obvious applications to any process of transition from one political system to another.²³ All these strands may well come to be interwoven in a theory of democratic origins that transcends the simple-minded unilinear hypothesis.

In any such inquiry the example of Sweden holds an important place. As the accompanying table indicates, as recently as the beginning of this century, Sweden had one of the most restrictive suffrage systems of any country with a representative constitution: it was a stable oligarchy. Yet by 1921 the right to vote had been extended to adult men and women, and the accountability of the cabinet to the democratically elected legislature was beginning to be accepted. The transition to democracy, then, was late and rapid. This means that the process can be observed in Sweden in condensed form and in a setting that may have some relevance to other countries attempting such a transition in an industrial age.

The Swedish transformation, moreover, occurred mainly in response to domestic political pressures. There was no large-scale immigration, such as would complicate the analysis of democratic origins in the United States, Canada, Australia, New Zealand, or Israel. Nor was there any foreign military occupation aimed in part at imposing democracy, as there was in post-war West Germany and Japan.

The following pages will not attempt to recite the facts of the Swedish transition
to democracy; on this a sizable literature is available.24 Instead, selected aspects of

this factual development - from the formation of the suffrage movement in the
1890's to the franchise reforms of 1909 and 1918/21 - will be referred to in order to
illustrate, positively and negatively, some old and new hypotheses about democracy.

o

3

The Swedish experience lends no support to the notion that democracy presupposes consensus about democratic fundamentals or democratic procedures. Three periods may be distinguished. Before the 1890's there was a broad consensus in support of the existing oligarchic regime. This was an active consensus among the political classes such as the landowners, high officials, and industrialists who dominated the Upper Chamber and the farmers who dominated the Lower Chamber. It took the form of passive support or acquiescence among the rest of the population. Democracy remained an extreme minority position, actively espoused by only a handful of

political leaders such as Adolf Hedin, Anders Wilhelm Uhr, and other New Liberals
in the late 1860's.

This oligarchic consensus was shattered in the decades from 1890 to 1920. Democracy vs. oligarchy became the major theme of political conflict. "Shall kingly power with lordly power or kingly power with people's power rule in the Swedish realm?" - as a leading protagonist put it on one famous occasion.₂₅

The adoption of the major suffrage acts in 1907 and 1918 might be cited as evidence of democratic consensus. But this would be little more than a roundabout way of saying that the transition in Sweden was peaceful: there was no attempt to impose democracy by domestic force (as in the English or French Revolutions) or by outside intervention (as recently in Germany and Japan). Being the product of peaceful internal evolution, democracy in Sweden was indeed approved by the existing power-holders, including majorities in each of the chambers and among their constituents.

There was democratic consensus, that is to say, only in the trivial or tautological sense that, when democracy was adopted, those whose concurrence was required for its adoption did at last concur. We shall later examine some of the considerations that persuaded the defenders of oligarchy to accept the new political dispensation. The conviction that democracy was the best or most just government was not prominent among these.

The third period, since the 1920'5, has indeed been one of widening consensus. There was, first of all, a growing acceptance of the democratic form of government. Conservatives in the days of Erik Gustaf Boström and Christian Lundeberg had considered democracy a danger; those of Arvid Lindman's and Ernst Trygger's generation had accepted it - with grace or reluctance, as the case may be; those of the 1930's and 40's came to take it for granted. There were recurrent sharp disagreements about substantive issues - about defense, labor relations, and tariffs in the 20's, about anti-depression policies in the 30's, about aspects of the welfare state in the 40's. There were also continuing latent conflicts, such as the one about neutrality and royal prerogative which came to the fore in the abdication crisis during World War Two. Whenever the disagreements were openly fought out, however, these conflicts were eventually resolved. By the 1930'5, for example, the Social Democrats had given up their anti-militarism, their republicanism, and their anti-clericalism; the Conservatives had set aside their objections to the League of Nations. By the 1940's the clash between Social Democratic advocates of nationalization and bourgeois champions of free enterprise had been resolved within the framework of a welfare state.

A growing measure of substantive consensus, that is to say, was the product rather than the precondition of democracy. As August Sävström, the long-time speaker of the Lower Chamber, remarked in looking back on four decades of parliamentary service: "The struggles which formerly were bitter - far too bitter - have given way to a willingness to cooperate. Everything now goes more smoothly than it used to."²⁶ The Swedish experience in this respect corresponds closely to Bernard Crick's notion of consensus not as a prerequisite of democracy or as "some external and intangible

spiritual adhesive" but rather as "the activity (the civilizing activity) of democratic]
politics itself."27

Behind this movement from oligarchic consensus first to conflict and then to
democratic consensus, was there any other agreement on matters more fundamental
perhaps than the form of government?

There was, first of all, territorial unity unchallenged since the Middle Ages and within unvarying frontiers since 1658. (Finland, Norway, and the Baltic possessions of the seventeenth and eighteenth centuries are not relevant in this context, since they were separately administered.) There was unity of language and of religion - although the latter was disrupted in the nineteenth century by the challenge of the "free churches" to the official state church and later by the rapid spread of agnosticism. There was, finally, a common bureaucratic structure, developed by the Vasa kings and reaching down to the entire population by means of the land tax and the peculiar conscription system known as indelningsverket.

All this added up to a strong sense of national unity. As Deutsch has pointed out, such a sense of unity rests less on uniformity of opinion rather than on complementarity of habit.2B In the oligarchic, militarist Sweden of the sixteenth to nineteenth centuries, for example, it was supported by the nobility's habit of command, the peasantry's habit of obedience, and the religious doctrine preached by the Lutheran state church. Some of the tacit assumptions underlying the habit of national unity found articulation only when they seemed challenged. A perceived challenge of this sort was the "union crisis" leading to Norway's secession in 1905. It was this patriotic issue, coming in the midst of the protracted suffrage crisis, that prompted the formation of the Lundeberg government (August-November 1905), the first to accept the principle of the cabinet's accountability to parliament and the only one (except for the cabinet of 1939—45) to include representatives of all groups in the legislature.

Other elements in the Swedish political tradition that proved of crucial importance were the deepseated devotion to legal rules, the liberal content of the rules long in effect, and the practice of division of powers within the government. Next to England, Sweden has the longest tradition of constitutional and representative government. A Freedom of the Press Act has been among the fundamental laws since the eighteenth century. This meant that the supporters of democracy, in pressing their case, could make fullest use of freedom of expression and assembly and also of other traditional rights such as that of petitioning the king. The long established division of powers between king, cabinet, Upper Chamber, Lower Chamber, and a bureaucracy consisting of numerous independent agencies meant that the defenders of the oligarchic system rarely met the democratic challengers with a fully unified front.

National unity, legality, liberalism, and complex constitutional government thus facilitated Sweden's political transformation. They are not, however, enough to explain it - precisely because they were traditions that had been in effect long before the transition to democracy began. They are best put in the category of background conditions that came into play once the process of change was set off bv other stimuli.

4

Among the precipitating factors, one may well look for certain social and economic changes. In the last third of the nineteenth century the proportion of Swedes employed in manufacturing doubled (14.6% in 1870, 27.8% in 1900) and that of urban residents tripled (3.2 % living in towns over 25,000 in 1865, and 10.5 % in 1906).²⁹ In the half century before World War One, Sweden experienced one of the most rapid phases of economic growth of any industrial country - the increase in per capita national product per decade (26.2 %} being more than twice that of

the United Kingdom (12.5 %} and only slightly less than that of the United States (27.5 %}. Lest this be taken as prima jade confirmation for a correlation between economic development and democratization, it should be added that Japan's pre War One growth rate (33.7 %} exceeded all others.³⁰

In absolute terms, Sweden's economic and social development at the turn of the century, as measured by such indicators as urbanization, mail flow, number of telephones, and school enrolment was still far behind that of many other democratic and non-democratic countries. If such earlier data are compared with recent figures for countries included in Lipset's tabulation, Swedish primary school enrolments as early as 1882 exceeded the average level for recent "European stable democracies."

But in other respects Sweden in 1900 ranked below recent "Latin American stable dictatorships" - specifically with regard to urbanization, telephones per capita, and levels of enrolment in secondary and higher education. Needless to say, Sweden found itself in good company (see Tables II and III). Indeed nothing perhaps indicates the limited value of Lipset's hypotheses so strikingly as a simple exercise in comparison over time.

The connection between economic growth and democratization, then, needs to be formulated with some subtlety. One composite hypothesis would be that rapid social and economic development within a framework of national unity and of a libertarian legal order sets off pressures toward democracy, whereas in settings such as Bismarck Germany, Tsarist Russia, and Meiji Japan it would strengthen a paternalistic or authoritarian structure. Such a statement is reminiscent of Barrington Moore's assumption that industrialization will have very different political effects depending on the initial social constellation,³¹ though in making it one need not subscribe to Moore's more specific neo-Marxist formulations or to his somewhat facile characterization of countries as democratic, communist, or fascist.

There is one economic influence on democratization which can be traced causally and not through some correlation of unspecified character. The Parliament Act of 1866 limited the suffrage in monetary terms - for the Lower Chamber to taxpayers with an income of at least 800 kronor per year, owners of real estate assessed at 1,000 kronor or more, and leaseholders on agricultural property assessed at 6,000 kronor or more. This made the size and composition of the electorate dependent on economic growth and the value of the currency. "A sudden change in the value of money or in the wage level," as Professor Andrén has noted, "would have produced something that would have pretty much resembled a political revolution."32

Nothing quite as sudden occurred; still, the variations were pronounced, especially toward the end of the oligarchic period. The agricultural depression of the 1880's somewhat reduced the rural electorate, whereas the steady rise of the wage level in one branch of industry after the other increased the urban electorate in the 1890's and early 1900's. These gradual changes within the existing suffrage order were of a similar dimension to those brought about by the reform acts of 1907/9 and 1918/21. For example, the proportion of urban residents entitled to vote rose from 6.7 % in 1896 to 12.0 % in 1908, that of citizens generally from 6.3 % in 1896 to 9.4% in 1908 to 19.3% in 1911 - whereas in 1921 as many as 28.4 % of all men (as well as a somewhat smaller proportion of women) were included in the electorate.33

The progressive inclusion in the pre-reform electorate of social groups that had long felt unjustly excluded from political participation added substantially to the reform impetus. The fact that for some of them this inclusion seemed precarious - reversible if bad times reduced their wages below the suffrage limit of 800 kronor - is likely to have sharpened their sense of urgency.

O-j

5

It was noted earlier that the ancien regime was not monolithic and just now that the electorate, by the very terms of the old order, was not rigidly fixed but slowly expanding. All these are still in the nature of background conditions, for as James Bryce noted long ago, "One road only has in the past led into democracy, viz. the wish to be rid of tangible evils."34

The tangible evils of which Swedish citizens in the half century from 1870 to
1920 wished to rid themselves varied. Successive issues gave rise to different alignments,
thus creating a dynamic factor not to be underrated.

The parliamentary battles of the 1870's and early 80's pitted the farmers who controlled the Lower Chamber and wished to reduce the land taxes and devolve the indelningsverket against the high officials and industrialists of the Upper Chamber who wished to maintain a strong military system. The fight continued for over two decades, even though two partial reductions were agreed in 1873 and 1885.

By the late 1880's, the agricultural depression overshadowed this earlier alignment and split the solid front of the farmers earlier organized in the Ruralist Party (Lantmannapartiet). Grain producing farmers joined with Upper Chamber Conservatives in voting for a protective tariff in 1888. Livestock farmers and representatives of the urban consumers opposed it.

Out of the remnants of these two successive alignments, that superb tactician Boström fashioned a temporary alliance (including Conservative senators and urban free traders interested in a strong defense and Ruralists of both the protectionist and free trade factions) in 1892 that replaced the land tax and subdivision system with a 90-day period of military training.

When the suffrage question at last became paramount in parliament, toward 1900, the same sort of realignment occurred. The new Liberal party (Liberala samlingspartiet) included elements of the Old and New Ruralist parties along with their erstwhile urban opponents. Crucial support from farmers' representatives and free church and temperance groups thus boosted the once chiefly urban ranks of the suffrage advocates. Yet the alliance of radical Liberals and Social Democrats, fashioned in the 1890's when both groups were still largely excluded from the parliamentary system, held up despite recurrent strains.

A new split occurred when various groups of farmers, inside and outside the Liberal party, turned against Karl Staaff's demand for manhood suffrage in majority elections and instead endorsed the Conservative demand for proportional representation. By this time the issue, for the leading Conservatives, was no longer whether universal suffrage should be accepted but rather on what terms.

The defense question in the 1890's not only provided a precedent for far-flung compromise such as was repeated in the reform of 1907; it also provided a direct contribution to the suffrage issue. To many participants in the public debate from the 1860's to the turn of the century, whether Liberals or Conservatives, there was a very direct connection between a citizen's duty to bear arms and his right to vote; conversely only a popular state could muster an effective defense.³⁵ Nationalist

arguments thus bridged the gap between the militarism of the extreme right and
the democratic ideals of the extreme left.

The fluid alignments on various issues corresponded to a pattern of social recruitment of political leaders of unprecedented fluidity. The unreformed parliament before 1866 had been divided into hereditary estates, and in the cabinets of the 1870's the nobility still prevailed. This made for a narrow base of recruitment that gave a sharp point to the anti-nepotism provisions of the 1809 constitution.36 It was not until 1883 that Sweden had its first commoner in the prime minister's post, and the appointment of two ministers with a lowly "farmer's name" such as Petersson caused something of a sensation in 1905.

But the transition from estate parliament to two plutocratically elected chambers brought about some social shifts, with a high aristocrat such as Count Arvid Posse and a bourgeois landowner such as Erriil Key becoming the chief leaders of the oppositional farmers in the Lower Chamber.

The same championship of lower class causes by dissidents from the upper class recurred in the evolution of the Swedish Labor movement. "The Socialist press might boast that the party congress of 1889 consisted almost entirely of workers and artisans; yet of the two delegates of middle class origin one was Fredrik Sterky, later the first chairman of the Swedish Federation of Labor, and the other was Hjalmar Branting."37

6

Prominent individuals might thus smooth the political transition by making the skills of a declining elite available to a rising movement of lower class protest. The major question in explaining a peaceful transition to democracy still remains: What can persuade the elite of an ancien regime to abdicate its position of power peacefully?

The Swedish experience suggests two major factors. The first is a systematic attack on the old system by well-organized forces - the attack including at least implied threats of violence. The other are a number of considerations of prudence that weaken the old elite's will to resist.

The Swedish suffrage agitation went through a number of stages. Suffrage associations were formed in several provinces in the late 1880's and on a national basis in 1890. In 1893 and 1896 they called two "People's Parliaments" and in 1899 circulated a mammoth petition with 364,000 signatures. Vallinder's researches have demonstrated how much systematic effort, by David Bergström and others, went into these undertakings.38 In the People's Parliaments, moreover, the newly formed Social Democratic party (1889) with its trade union affiliates, took a leading part.

The early suffrage advocates had formed rifle societies, one of the "means of pressure" (along with others of more doubtful legality, such as a refusal to pay taxes or perform military service) the discussion of which was to preoccupy the People's Parliaments. By the beginning of the century, the Liberals were firmly

committed to electoral and parliamentary action within the existing order whereas the Socialists sponsored the political strike of 1902, which involved as many as 116,000 workers. The pressures from within and outside the system, presumably, both contributed to the ultimate victory.

In the changing attitudes of the defenders of the old regime, several of the factors listed earlier played their part. The liberal principles of the legal order gave full scope to the arguments that invoked basic principles of justice, Christianity, and patriotism in favor of political equality. The plutocratic franchise of 1866, moreover, had only recently replaced the estate parliament so that it commanded no time-hallowed loyalties. The late advent of capitalist plutocracy facilitated its early demise.

The crucial change in attitude was reached when leading Conservatives, by the turn of the century, began to feel that universal suffrage, at least for the Lower Chamber, would prove inevitable. The entry of the working class into the plutocratically defined electorate must have contributed to this impression. So, surely, did the mounting organized extra-parliamentary pressure. Foreign events drove home a similar lesson: Norway widened its franchise in 1884 and 1897, Great Britain enacted the third reform bill in 1884, Belgium adopted manhood suffrage in 1893, and Finland in 1906 became the first European country to give the vote to women as well.

It was in this frame of mind that leading Conservatives were beginning to look for what they called "guarantees" - not to prevent the adoption of universal suffrage but to alleviate its effects. As early as 1892 Bishop Gottfrid Billing expressed this attitude: "Rather stronger guarantees and a further extension of this suffrage than weaker guarantees and a lesser extension."³⁹ And a decade later Gustaf Axel Berg added the characteristic note of urgency. "If we wish to get universal suffrage with guarantees, we must get it now - if we tarry, make no mistake about it, we shall assuredly get it without guarantees."40

The great compromise of 1907 involved precisely manhood suffrage for the Lower Chamber together with Billing's favorite guarantee of proportional representation - except that the latter was applied to a less plutocratically constituted Upper Chamber as well.

If the suffrage extensions elsewhere in Europe at the turn of the century contributed to a sense of inevitability, the collapse of the German empire in 1918 in defeat and revolution created among anti-democratic Swedish Conservatives a sense of doom. This at least was the estimate of Prime Minister Nils Eden who used what he considered the crucial psychological moment in preparing the final set of democratic reforms for a special session of the Riksdag.41

Proportional representation lived up to the expectations of its sponsors in delaying the advent of a solid parliamentary majority of lower middle class and working class representatives. Universal suffrage in single member constituencies would very likely have resulted in a Liberal-Socialist coalition at first and a Social Democratic majority some time in the 1920'5. (No mere recalculation of election returns would indicate the exact effect of majority or plurality elections. For party divisions

among leaders and voter preferences are themselves in part a response to certain features of the electoral system, such as the överrepresentation of large parties under the plurality system.) Instead, proportionalism, applied in smaller constituencies after 1921, encouraged a four (or four-and-one-half) party system.⁴² In the absence of solid parliamentary majorities, the most successful tactic was that of the Liberal (Frisinnade) leader Carl Gustaf Ekman, who sided with the Social Democrats in reducing Conservative motions for higher military expenditures and with the Conservatives in defeating Social Democratic labor and welfare policies. By the time the Social Democrats did come to power, with Agrarian support in the 1930's and mostly by themselves since 1945, their more extreme positions (pacifism, republicanism, anti-clericalism) had all been abandoned.

Nonetheless, it seems likely that in the long run the competition for the crucial swing vote in a three party system (Social Democrats, Liberals, Conservatives) and later a two party system (Socialist vs. bourgeois) would have produced much the same moderating effect.

7

Consensus on democratic values or democratic procedures is an implausible precondition of democracy. The theoretical task is not to explain how democratic institutions can benefit from the acceptance of democratic ideals or the prevalence of democratic attitudes. Rather, a genetic theory of democracy must explain how that form of government can emerge out of a setting in which neither the institutions, nor the popular attitudes, nor the accepted values are as yet democratic.

The Swedish experience suggests that a long tradition of national unity, and the prior existence of liberal and constitutional procedures is a helpful, maybe an essential, precondition. The actual impetus to the transition from oligarchy to democracy, however, was a prolonged and indecisive conflict - in this case of the disfranchised lower middle and working classes against the oligarchic representatives of high bureaucrats, industrialists, and farmers.

Frequently shifting social and political alignments in a period of rapid economic development kept battlefronts fluid and suggested the possibility of compromise and gradualism even on as fundamental and divisive an issue as universal suffrage. Examples of democratization in other countries and the widening of the electorate under the existing census franchise persuaded many Conservatives that democracy in one form or another would be inevitable. They thus began to concentrate less on opposing democracy than on finding its most innocuous form, and thereby crucially contributed to a peaceful compromise solution.

Once the commitment to democracy was made, democratic convictions spread both among the transitional politicians (of Lindman's and Trygger's generation) and especially among their successors who came to spend most of their active lives under the new system. This adaptation of opinion to action (or rationalization) is a nexus little recognized in current opinion theory but quite adequately accounted for within Festinger's scheme of cognitive dissonance.

Consensus, then, was chiefly a product of democracy which had its genesis chiefly in conflict. These conclusions, suggested by the empirical evidence of Sweden and other countries, also seem to be in accord with an important strand of democratic political philosophy - indeed the prevalent strand until the 1930'5, when the tribulations of depression and war made many democratic thinkers take refuge in the somewhat vague and often misleading notion of consensus.

"The basis of democracy," as I have written elsewhere, "is not maximum consensus. It is the tenuous middle ground between imposed uniformity (such as would lead to some sort of tyranny) and implacable hostility (of a kind that would disrupt the community in civil war or secession). In the process of genesis of democracy, an element of what might be termed consensus enters at three points at least. There must be a prior sense of community, preferably a sense of community quietly taken for granted that is above mere opinion and mere agreement. There must be a conscious adoption of democratic rules, but they must not be so much believed in as applied, first perhaps from necessity and gradually from habit. The very operation of these rules will enlarge the area of consensus step-by-step as democracy moves down its crowded agenda.

"But new issues will always emerge and new conflicts threaten the newly won agreements. The characteristic procedures of democracy include campaign oratory, the election of candidates, parliamentary divisions, votes of confidence and of censure — a host of devices, in short, for expressing conflict and thereby resolving it. The essence of democracy is the habit of dissension and conciliation over everchanging issues and amidst ever-changing alignments. Totalitarian rulers must enforce unanimity on fundamentals and on procedures before they can get down to other business. By contrast, democracy is that form of government that derives its just powers from the dissent of up to one half of the governed."43

---

NOTES

1 Robert A. Dahl, ed., Political Oppositions in Western Democracies (New Haven: Yale University Press, 1966); Robert A. Dahl, Pluralist Democracy in the United States: Conflict and Consensus (New York: Rand McNally, 1967); Karl W. Deutsch et al., Political Community and the North Atlantic Area (Princeton: Princeton University Press, 1957); Samuel P. Huntington, Political Order in Changing Societies (New Haven: Yale University Press, 1968); Stein Rokkan, Citizens, Elections, Parties: Approaches to the Comparative Study of the Processes of Development (New York: McKay, 1970).

2 Robert A. Dahl, "The Behavioral Approach in Political Science: Epitaph for a Monument to a Successful Protest," The American Political Science Review (Dec. 1961), pp. 763—772.

3 Daniel Bell, The End of Ideology (New York: Free Press, 1960).

4 For pleas to center the study of politics around the theme of change, see e.g. Samuel P. Huntington, "The Change to Change: Modernization, Development, and Politics," Comparative Politics (April 1971); and three articles by Dankwart A. Rustow, "Communism and Change" in Chalmers Johnson, ed., Change in Communist Systems (Stanford: Stanford University Press, 1970), pp. 343-358; "Modernization and Comparative Politics: Prospects in Research and Theory," Comparative Politics, vol. 1, no. 1 (October 1968), pp. 37-51; "Change as the Theme of Political Science," International Political Science Association, Round Table, Torino, Sept. 1969 (mimeographed).

NOTES

1 Robert A. Dahl, ed., Political Oppositions in Western Democracies (New Haven: Yale University Press, 1966); Robert A. Dahl, Pluralist Democracy in the United States: Conflict and Consensus (New York: Rand McNally, 1967); Karl W. Deutsch et al., Political Community and the North Atlantic Area (Princeton: Princeton University Press, 1957); Samuel P. Huntington, Political Order in Changing Societies (New Haven: Yale University Press, 1968); Stein Rokkan, Citizens, Elections, Parties: Approaches to the Comparative Study of the Processes of Development (New York: McKay, 1970).

2 Robert A. Dahl, "The Behavioral Approach in Political Science: Epitaph for a Monument to a Successful Protest," The American Political Science Review (Dec. 1961), pp. 763—772.

3 Daniel Bell, The End of Ideology (New York: Free Press, 1960).

4 For pleas to center the study of politics around the theme of change, see e.g. Samuel P. Huntington, "The Change to Change: Modernization, Development, and Politics," Comparative Politics (April 1971); and three articles by Dankwart A. Rustow, "Communism and Change" in Chalmers Johnson, ed., Change in Communist Systems (Stanford: Stanford University Press, 1970), pp. 343-358; "Modernization and Comparative Politics: Prospects in Research and Theory," Comparative Politics, vol. 1, no. 1 (October 1968), pp. 37-51; "Change as the Theme of Political Science," International Political Science Association, Round Table, Torino, Sept. 1969 (mimeographed).

---

5 Seymour Martin Lipset, Political Man (Garden City, N. Y.: Doubleday, 1960); Gabriel Almond and Sidney Verba, The Civic Culture (Princeton: Princeton University Press, 1963).

6 Harry Eckstein, The Theory of Stable Democracy (Princeton: Center of International Studies, 1961) and Division and Cohesion in a Democracy (Princeton: Princeton University Press, 1965).

7 For the earliest use of "consensus" in American political science see A. Lawrence Lowell, Public Opinion and Popular Government (New York: Longmans, Green, new edn. 1914) p. 9, and Government and Parties in Continental Europe (Boston: Houghton Mifflin, 1896), vol. 1, pp. 101 ff. For a representative statement of contemporary doctrine see Edward Shils, "Consensus," International Encyclopedia of the Social Sciences (1968), 111, pp. 260—266. Cf. Rustow, "Agreement, Dissent, and Democratic Fundamentals," in Klaus von Beyme, ed., Theory and Politics - Theorie und Politik: Festschrift zum 70. Geburtstag von Carl Joachim Friedrich (The Hague: Nijhoff, 1971), pp. 327-342.

8 Ernest S. Griffith, John Plamenatz and J. Roland Pennock, "Cultural Prerequisites to a Successfully Functioning Democracy: A Symposium," The American Political Science Review (March 1956), pp. 101-137.

9 See e.g. Robert A. Dahl, Who Governs? (New Haven: Yale University Press, 1961).

10 Carl Joachim Friedrich, The New Belief in the Common Man (Brattleboro: Vermont Publishing Co., 1942); Bernard Crick, In Defense of Politics, rev. ed. (Penguin Books, 1964); Lewis A. The FunctionSocialFree

Coser, of Conflict (Glencoe: Press, 1956). 11 It should be noted, for example, that Castro's Cuba, Hitler's Germany, and Lumumba's Congo would have ranked high among countries in their continents on most of Lipset's scales.

12 cit.

Lipset, op. 13 Dahl, Who Governs?, pp. 11-31, and Almond and Verba, The Civic Culture, pp. 5-6, Eckstein, Division and Cohesion, 6.

chapter 14 AStable47.

Eckstein, Theory of Democracy, p. 15 Deane E. Neubauer, "Some Conditions of Democracy, "The American Political Science Review, vol 61, no. 4 (December 1967), pp. 1002-1009. Arthur S. Banks, Cross-Polity Time Data (Cambridge: The MIT Press, 1971) and work in preparation by William Flanigan and Edwin

Fogelman. 16 Arend Lijphart, The Politics of Accommodation (Berkeley and Los Angeles: University of California

Press, 1968). 17

Leonard Binder, ed., Politics in Lebanon (New York: Wiley, 1966). 18 Dahl, Political Opposition in Western Democracies, and Pluralist Democracy in the United States.

19 Cyril E. Black, The Dynamics of Modernization (New York: Harper, 1966); Huntington, Political Order in Changing Societies, and Rustow, A World of Nations: Problems of

Political Modernization (Washington: Brookings Institution, 1967). 20 Karl W. Deutsch, Nationalism and Social Communication (New York: Wiley, 1953) and Deutsch et theAtlantic

al., Political Community and North Area. 21 Lewis J. Edinger, ed., Political Leadership in Industrialized Societies (New York: Wiley, 1967); Erik H. Erikson, Gandhi's Truth (New York: Norton, 1969); Rustow, ed.,

Philosophers and Kings: Studies in Leadership (New York: Braziller, 1970). 22 Coser, The Function of Social Conflict; Albert O. Hirschman, Journeys Toward Progress 'New York: Twentieth

Century Fund, 1963). 23 Leon Festinger, A Theory of Cognitive Dissonance (Stanford: Stanford University Press,

1957). 24 Dankwart A. Rustow, The Politics of Compromise: A Study of Parties and Cabinet Government in Sweden (Princeton: Princeton University Press, 1955); Douglas A. Verney, Parliamentary Reform in Sweden, 1866-1921 (London: Oxford University Press, 1957); Gunnar Wallin, Valrörelser och falresultat: Andrakammarvaler i Sverige 1866-1884 (Stockholm: Ronzo Boktryckeri, 1961); Torbjörn Vallinder, / kamp för demokratin; Rösträttsrörelsen i 1886-1900Natur 0ch1962).

Sverige (Stockholm: Kultur, 25 "Skall konungamakt med herremakt eller konungamakt med folkemakt styra i Sveriges rike?" Prime Minister Karl Staaff in the Lower Chamber, May 15, 1906, after rejection of

his government's suffrage bill by the Upper Chamber. 26

August Sävström, En talmans levnadsminnen (Stockholm: Lindfors, 1949), pp. 228 f. 27 Crick, In Defense of Politics, p. 24.

5 Seymour Martin Lipset, Political Man (Garden City, N. Y.: Doubleday, 1960); Gabriel Almond and Sidney Verba, The Civic Culture (Princeton: Princeton University Press, 1963).

6 Harry Eckstein, The Theory of Stable Democracy (Princeton: Center of International Studies, 1961) and Division and Cohesion in a Democracy (Princeton: Princeton University Press, 1965).

7 For the earliest use of "consensus" in American political science see A. Lawrence Lowell, Public Opinion and Popular Government (New York: Longmans, Green, new edn. 1914) p. 9, and Government and Parties in Continental Europe (Boston: Houghton Mifflin, 1896), vol. 1, pp. 101 ff. For a representative statement of contemporary doctrine see Edward Shils, "Consensus," International Encyclopedia of the Social Sciences (1968), 111, pp. 260—266. Cf. Rustow, "Agreement, Dissent, and Democratic Fundamentals," in Klaus von Beyme, ed., Theory and Politics - Theorie und Politik: Festschrift zum 70. Geburtstag von Carl Joachim Friedrich (The Hague: Nijhoff, 1971), pp. 327-342.

8 Ernest S. Griffith, John Plamenatz and J. Roland Pennock, "Cultural Prerequisites to a Successfully Functioning Democracy: A Symposium," The American Political Science Review (March 1956), pp. 101-137.

9 See e.g. Robert A. Dahl, Who Governs? (New Haven: Yale University Press, 1961).

10 Carl Joachim Friedrich, The New Belief in the Common Man (Brattleboro: Vermont Publishing Co., 1942); Bernard Crick, In Defense of Politics, rev. ed. (Penguin Books, 1964); Lewis A. The FunctionSocialFree

Coser, of Conflict (Glencoe: Press, 1956). 11 It should be noted, for example, that Castro's Cuba, Hitler's Germany, and Lumumba's Congo would have ranked high among countries in their continents on most of Lipset's scales.

12 cit.

Lipset, op. 13 Dahl, Who Governs?, pp. 11-31, and Almond and Verba, The Civic Culture, pp. 5-6, Eckstein, Division and Cohesion, 6.

chapter 14 AStable47.

Eckstein, Theory of Democracy, p. 15 Deane E. Neubauer, "Some Conditions of Democracy, "The American Political Science Review, vol 61, no. 4 (December 1967), pp. 1002-1009. Arthur S. Banks, Cross-Polity Time Data (Cambridge: The MIT Press, 1971) and work in preparation by William Flanigan and Edwin

Fogelman. 16 Arend Lijphart, The Politics of Accommodation (Berkeley and Los Angeles: University of California

Press, 1968). 17

Leonard Binder, ed., Politics in Lebanon (New York: Wiley, 1966). 18 Dahl, Political Opposition in Western Democracies, and Pluralist Democracy in the United States.

19 Cyril E. Black, The Dynamics of Modernization (New York: Harper, 1966); Huntington, Political Order in Changing Societies, and Rustow, A World of Nations: Problems of

Political Modernization (Washington: Brookings Institution, 1967). 20 Karl W. Deutsch, Nationalism and Social Communication (New York: Wiley, 1953) and Deutsch et theAtlantic

al., Political Community and North Area. 21 Lewis J. Edinger, ed., Political Leadership in Industrialized Societies (New York: Wiley, 1967); Erik H. Erikson, Gandhi's Truth (New York: Norton, 1969); Rustow, ed.,

Philosophers and Kings: Studies in Leadership (New York: Braziller, 1970). 22 Coser, The Function of Social Conflict; Albert O. Hirschman, Journeys Toward Progress 'New York: Twentieth

Century Fund, 1963). 23 Leon Festinger, A Theory of Cognitive Dissonance (Stanford: Stanford University Press,

1957). 24 Dankwart A. Rustow, The Politics of Compromise: A Study of Parties and Cabinet Government in Sweden (Princeton: Princeton University Press, 1955); Douglas A. Verney, Parliamentary Reform in Sweden, 1866-1921 (London: Oxford University Press, 1957); Gunnar Wallin, Valrörelser och falresultat: Andrakammarvaler i Sverige 1866-1884 (Stockholm: Ronzo Boktryckeri, 1961); Torbjörn Vallinder, / kamp för demokratin; Rösträttsrörelsen i 1886-1900Natur 0ch1962).

Sverige (Stockholm: Kultur, 25 "Skall konungamakt med herremakt eller konungamakt med folkemakt styra i Sveriges rike?" Prime Minister Karl Staaff in the Lower Chamber, May 15, 1906, after rejection of

his government's suffrage bill by the Upper Chamber. 26

August Sävström, En talmans levnadsminnen (Stockholm: Lindfors, 1949), pp. 228 f. 27 Crick, In Defense of Politics, p. 24.

---

28 Deutsch et al., Political Community and the North Atlantic Area, p. 90.

29 G. A. Montgomery, The Rise of Modern Industry in Sweden (London: P. S. King, 1939) p. 141.

30 Simon Kuznets, "Quantitative Aspects of the Economic Growth of Nations, I," Economic Development and Cultural Change, vol. 5 (1956/7), p. 13.

31 Harrington Moore, Jr., Social Origins of Dictatorship and Democracy (Boston: Beacon, 1966).

32 Georg Andrén, Tvåkammat_systemets tillkomst och utveckling, Sveriges Riksdag IX (Stockholm: Victor Petterson, 1937), p. 207.

33 Ibid., pp. 205, 595.

34 Tames Bryce. Modern Democracies (New York: Macmillan. 1921). 11. p. 602.

35 Such sentiments were voiced by Adolf Hedin as early as 1868 and echoed by Harald Hjärne and Verner von Heidenstam at the turn of the century. Cf. Rustow, Politics of Compromise, pp. 59-61.

36 According to Article 4 of the Regeringsformen, two brothers could not at once serve in the king's council. But it was not uncommon to find cousins, brothers-in-law, or uncles and nephews in the same cabinet.

37 Rustow, Politics of Compromise, p. 162.

38 See Vallinder, I kamp för demokratin.

39

Cited Andrén, Tvåkammar systemets tillkomst och utveckling, p. 305. 40 Cited ibid., p. 346.

41 Cf. Ibid., 535; Rustow, Politics ofBs, and literature cited.

42 On the effects of proportional representation, see Rustow, Politics of Compromise, pp. 123 ff.

43 Dankwart A. Rustow, "Transitions to Democracy: Toward a Dynamic Theory," Comparative Politics, vol. 2 no. 3 (April 1970), p. 363. The article spells out in a more general context some of the theoretical and methodological reflections which have here been applied specifically to Sweden.

28 Deutsch et al., Political Community and the North Atlantic Area, p. 90.

29 G. A. Montgomery, The Rise of Modern Industry in Sweden (London: P. S. King, 1939) p. 141.

30 Simon Kuznets, "Quantitative Aspects of the Economic Growth of Nations, I," Economic Development and Cultural Change, vol. 5 (1956/7), p. 13.

31 Harrington Moore, Jr., Social Origins of Dictatorship and Democracy (Boston: Beacon, 1966).

32 Georg Andrén, Tvåkammat_systemets tillkomst och utveckling, Sveriges Riksdag IX (Stockholm: Victor Petterson, 1937), p. 207.

33 Ibid., pp. 205, 595.

34 Tames Bryce. Modern Democracies (New York: Macmillan. 1921). 11. p. 602.

35 Such sentiments were voiced by Adolf Hedin as early as 1868 and echoed by Harald Hjärne and Verner von Heidenstam at the turn of the century. Cf. Rustow, Politics of Compromise, pp. 59-61.

36 According to Article 4 of the Regeringsformen, two brothers could not at once serve in the king's council. But it was not uncommon to find cousins, brothers-in-law, or uncles and nephews in the same cabinet.

37 Rustow, Politics of Compromise, p. 162.

38 See Vallinder, I kamp för demokratin.

39

Cited Andrén, Tvåkammar systemets tillkomst och utveckling, p. 305. 40 Cited ibid., p. 346.

41 Cf. Ibid., 535; Rustow, Politics ofBs, and literature cited.

42 On the effects of proportional representation, see Rustow, Politics of Compromise, pp. 123 ff.

43 Dankwart A. Rustow, "Transitions to Democracy: Toward a Dynamic Theory," Comparative Politics, vol. 2 no. 3 (April 1970), p. 363. The article spells out in a more general context some of the theoretical and methodological reflections which have here been applied specifically to Sweden.