Man's capacity for justice makes democracy possible, but his inclination to
injustice makes democracy necessary.

Reinhold Niebuhr

The modern Nordic democracies - Denmark, Finland, Iceland, Norway, and Sweden - grew out of a tradition which combined royal absolutism with popular intolerance. In form, their political institutions are unitary, parliamentary, and highly centralized. In practice, three of them - Denmark, Norway, and Sweden - have been dominated, to a degree almost unique among Western democracies, by a single, reform-minded party for the last half century. Their political systems have grown in scope and resources to be among the most encompassing in the West.

Resumé

The combination of a tradition of arbitrary royal government and popular intolerance, with the existence of extensive, unitary and highly centralized institutions of government, and with the dominance of government by a single, reformist party for more than a half century would seem a potent recipe for governmental abuse of individual and minority group rights. That is especially true in the absence of the formal constitutional safeguards used to preserve rights elsewhere. Yet the Nordic democracies have in fact proved exemplary in protecting civil liberties. Why? This article challenges the view that Nordic political societies are consensual historically or by nature, and it describes the limited traditional institutional checks on abuse of individual and minority rights by government. Instead it finds the basis for the Nordic democracies' excellent record in civil liberties in the combination of (1) the development of new, non-traditional institutional checks on government, (2) the use of direct democracy as a check on parliamentary majorities, (3) the pervasive use of corporatist channels encompassing the major oppositional interest groups to draft and implement legislation, and (4) the effective constraints placed on decisions by parliamentary majorities by the pattern of fierce partisanship in the finely balanced parliamentary party constellation. These protections for individual and minority liberties arise primarily from the assumption that basic conflicts of interest are a fact of political life and that they should be institutionalized.

This would seem a recipe for governmental abuse of individual and minority group rights, especially in the absence of the institutional checks and balances used elsewhere. Yet the Nordic states have come to be among the most careful in protecting both individual civil liberties and minority rights.

Why?

Elements of the Paradox

Modern Nordic democracy is superimposed on a historical tradition that accorded vast power to the ruler, and by extension, to his subordinates. Royal absolutism, established in Denmark until 1849, had some fairly bizarre results. One of the bloodier episodes was the fashion in which the Danish King Christian II celebrated his ascension to the throne of Sweden in 1520: he executed 82 of his noble coronation guests on the charge of heresy. The ensuing Swedish revolt and the dissolution of the Union of Kalmar which linked Sweden with Denmark-Norway could be interpreted as a post hoc check on the abuse of royal power, but it was a trifle late for the unwitting nobles in Stockholm castle. Although the origins of modern Swedish parliamentary government can be traced back to 1435, royal supremacy often prevailed there as well. Thus Gustav HI in 1788 dealt with the parliamentary ban on waging offensive war on Russia by the simple stratagem of having Swedish troops dressed in Cossack uniforms borrowed from the Royal Opera attack a Swedish border post in Finland. Even after the creation of modern parliaments (1814 in Norway, 1849 in Denmark, 1866 in Sweden, and 1906 in Finland), royal governments which jailed opponents for blasphemy and lése-majesté continued to rule against majorities in the popularly elected lower chamber until the establishment of parliamentary supremacy in Norway in 1884, Denmark in 1901, Sweden in 1917, Iceland in 1918, and Finnish independence in 1917.!

The concept of a limited government with internal checks and balances built into its machinery which influenced those who drafted the American constitution had little chance to take root in Scandinavia. The early constitutions(1809 Sweden, 1814 in Norway, and 1849 in Denmark) did provide for a separation of powers, but it was a division between the monarch and the parliament that was designed to limit the powers of the king. It also left him with the power to appoint a government of his choosing. The struggle that began in the latter part of the 19th century between king and the democratic parties over parliamentary supremacy ultimately reduced the monarch to a figurehead, but it left no place for a democratic executive independent of parliament as long as the monarchy existed. Only in Finland, where a mixed presidential-parliamentary system

was created in lieu of the monarchy originally intended, does the executive have any independence of the legislature. The organization of parties antedated universal suffrage,² permitting the practice of responsible party government through a parliamentary system. There was never an opportunityto an ideal system from scratch except to a limited extent in Norway in 1814;³ it was a question of redesigning existing systems and reusing old building materials. Nordic constitutional compromises had to do with balance between king and parliament, between houses elected by popular or privileged suffrage, and between existing parties through the Social Democrats and Liberals' acceptance of proportional representation (which guaranteed the Conservatives a substantial block of seats) in return for the Conservatives' acceptance of universal suffrage. Besides, both the Conservatives, who were dominant in the end of the 19th century, and the Social Democrats, who rapidly rose to be dominant in this century, believed in a strong state; they did not equate the inability to govern with good government.

Thus power in the Scandinavian democracies came to be unusually concentrated. Not only were all but the Finnish relatively pure parliamentary with the concentration of power in the legislative branch that this implies, they were unitary in their organization, so that no tier of secondary sovereign entities - states or provinces - could dispute the will of the national legislative majority and its executive. Over time the structure of subnational government has been streamlined for the sake of efficiency in implementing national policies. Most recently, Denmark, Norway, and Sweden all carried out radical programs of consolidating municipalities in the late 1960s and early 1970s to create the institutional capacity in local government to implement major national welfare programs, some of which - like health insurance and employment services in Denmark - were acquired by the state from decentralized, semi-public or private groups. Professionalization and bureaucratization of these spheres followed. The rapid growth of the public sector - today roughly one employee in three in Denmark, Norway, and Sweden works for the public sector and sixty to seventy percent of the GNP passes through the public sector⁴ - extended the sphere within which the governors could govern. Outside of Scandinavia, only the state in the Netherlands has comparable relative size.

Unlike the Dutch system, Danish, Norwegian, and Swedish politics has also been dominated for a half century by a single party - the Social Democrats (in Norway: Labor)⁵ - which has given policy a singular degree of cohesion and consistency. Single-party dominant government also increases the likelihood of abuse of power. The only other examples of elected regimes with similar grasp on office are the Unionists in Northern Ireland and the Nationalists in South Africa and the Democrats in the

South in the U.S. between Reconstruction and the mid-19605. None of
those can be held up as examples of outstandingly fair government.

This potent combination of a tradition of centralized government, a strong state, and extended single-party dominance of government by a reformist party seems a recipe for majority tyranny or at least systematic governmental abuse of citizen rights. But that has not occurred in Scandinavia. Instead, we have the paradox that the Nordic democracies combine extraordinarily an active, interventionist state of the strongest and most centralized sort we know with what seems to be an expanding realm of individual freedom and minority autonomy.

How can this be explained?

The answer that Nordic political culture is by nature consensual and hence abuses of power will not occur is attractive but illusory. Like other isolated societies, nineteenth century Nordic popular culture was intolerant and xenophobic. Xenophobia is still alive and well in outlying areas of Finland, Norway, and Sweden and may be held to characterize the entire Icelandic society. Even in Denmark, the most cosmopolitan of the five, it is still easy to find rural villages where 'recent immigrants' - individuals who grew up 75 miles away, moved to the village thirty years ago, and speak a slightly different dialect - are regarded with lingering suspicion. Homogeneous, stable, isolated societies are anything but hospitable to those who are different.

Nordic governments were more tolerant of those who were different, as long as they served the purposes of the rulers. Political opponents, however, were not in that category. In the last century the democratic opponents of the royal regimes were occasionally jailed and the cavalry used to disperse demonstrations.

The early leaders of the Danish Social Democratic party, on their release
from prison, were offered a choice by the police between being jailed again
or receiving free one-way tickets to America.

Conditions of war and unrest in the 20th century strained the relatively civilized conventions of the Victorian age in political repression. This was made brutally clear in the brief civil war in 1918 after the Finns had declared themselves independent of Russia. The number killed in the fighting was dwarfed in the aftermath as the Whites, scarcely magnanimous in victory, executed some 8500 of the Red opponents and threw another 80,000 into prison camps where almost 12,000 died of malnutrition and disease in the summer of 1918.⁶ Subsequently, in the early 1930s the right wing Lapua movement undertook systematic terrorism that included murdering Communists kidnapping leading Social Democrats and even the bourgeois retired President and his wife; it was ultimately suppressed after an attempted coup d'etat.

While the level of political violence in the other Nordic countries bears

no comparison with that in Finland, the class struggle repeatedly threatened violence there as well. Sweden regularly topped world statistics in industrial strife, and, as late as 1931, the year before the Social Democrats took power in Sweden, the government ordered the army out against striking workers in Aadalen where it turned its machine guns on an unarmed demonstration, killing 5 demonstrators and spectators. During World War II the German occupation stripped away the veneer of domestic party balance in Denmark and Norway. Danish authorities aided the Germans in rounding up the Danish Communists after the German invasion of the Soviet Union in June 1941, providing both the lists and the police to carry out the operation. The Norwegian Fascists established their own government and were sufficiently energetic in their activities against their fellow countrymen to give the English language a new word: Quisling. The Resistance took on some of the characteristics of civil war, especially in Norway where the Liberation government ultimately investigated 90,000 individuals out of an adult, active population of perhaps two million for collaboration with the Germans and convicted 46,000.

Although the level of this kind of political violence has declined dramatically the postwar period, it has not been because of any sudden growth of consensus. Opinion polls today suggest that there is wide divergence views on practically every key political issue. Membership in the European Community in Denmark and Norway, nuclear power in Sweden, and defense policy in Denmark are cases in point. The 1970s and 1980s have seen pitched battles between housing activists and the police in Copenhagen, clashes between environmentalists and the government in Norway, and bitter wildcat strikes particularly in Denmark and Sweden. Even within democratic parties, traditions of tolerance are poorly developed, most parties prefer to expel active dissenters. The Nordic emphasis on compromise arises not from a consensual political culture but from the desire to avoid the consequences of an obviously nonconsensual one.

What, then, limits the abuse of individual or minority rights by the state
or by popular majorities in parliament?

Our thesis is that the avoidance of governmental abuse of citizen rights and majority tyranny is a consequence of nontraditional but effective restraints on the governors built into Nordic political practice and institutionswhich both the individual against the state and the minority against the majority. Our focus is on four sorts of restraints: (1) those which provide institutional protection for individuals and ethnic minorities, (2) those which check the power of majorities through direct appeal to the people, (3) those which develop broader consent in preparing legislation than necessary to pass it in parliament especially through integrating antagonisticinterest in making and implementing policy, and (4) those

which provide partisan checks on parliamentary majorities and on oligopolisticagreements all major parliamentary parties. These restraints have not developed out of a grand philosophical design for protecting individual and group liberties but rather out of an accretion of practice that is most frequently related to the party balance. They are grounded in the recognition that group conflict is the basis of political life and is best dampened by giving the antagonists the right of self-representation within an institutional setting. The consequence has been the provision of farreachingguarantees individual and minority rights.

Institutional Restraints on the Governors

Nordic political institutions offer few of the built-in restraints on the governors to which Americans are accustomed. As relatively pure parliamentary the executive branch is dependent on the legislature except in Finland, which has a mixed parliamentary-presidential system like that of France,⁷ but this division certainly did not check governmental abuse of the rights of political minorities in the 1920s and 19305. The concept of judicial review remains basically a foreign one in Denmark and Sweden and has atrophied since the 1920s in Norway.⁸

Generally speaking the Scandinavian constitutions of the 19th century paid only limited attention to civil and political liberties. The Norwegian Constitution, while the most innovative of the three in breaking with the previous practice, devoted little attention to individual liberties, and its provisions which barred Jews and Jesuits from admission to the country reflected the general intolerance of difference in Norwegian society. The prohibition against Jews was repealed in 1851 after considerable political agitation, but that against Jesuits stood until 1956.⁹ The Norwegian Constitution's guarantees of basic political freedoms were also strictly circumscribed.¹⁰ While the Danish Constitution of 1849 provided stronger guarantees of freedom of the press and assembly, those did not prevent the regular jailing of opposition journalists and agitators in the 19th century. The Swedish Constitution was more rigorous in its provisions for freedom of the press; indeed, the Freedom of the Press Act {Tryckfrihetsforordning), one of the three component basic laws that make up the Swedish Constitution, in its original formulation to 1766. None of them come close to providing the breadth of freedom of the American Bill of Rights, and given the nature of 19th century Scandinavian government, the rights that they provided were only enforceable to a limited extent.

The constitutional revisions of this century have extended and codified
civil and political liberties. This is most notable in the Danish and the
Swedish postwar constitutions. The Swedish Constitution, which came into

force in 1975 and was amended in 1976 and 1979, is the most sweeping in this regard. A number of the freedoms enshrined in it are protected against legislative action that might limit them; they, like similar American guarantees in the Bill of Rights, can only be changed by constitutional amendment. Procedures for constitutional amendment represent significant hurdles: constitutional changes require passage of identically worded resolutions by parliament with an intervening general election, and any constitutional change can be sent to a binding referendum by a minority (one-third) of the members of parliament. The Danish Constitution of 1953 provides a number of unusual freedoms in its Article 8, including economic rights and both the right to free public education and the right not to send one's children to school, provided they receive an equivalent primary education. Those are similarly protected by amending procedures that require referenda.

The expansion of constitutional protections of civil liberties, particularly in the postwar period, is paralleled by the provision in the same period for cultural and, in some cases, governmental autonomy for ethnic minority groups. The Danes have developed what is close to a surrogate for federalism granting local autonomy to groups that are distinctly different from the national majority. The obvious cases are the Faeroe Islands and Greenland, which are part of Denmark. Clearly both are distinct entities geographically and ethnically, but until the relatively recent past (1948 in the case of the Faeroes and 1979 in the case of Greenland) they were administered as part of the unified kingdom. In both cases, local autonomy extends even to foreign economic policy. The Faeroe Islands, for example, did not join the European Community with Denmark, and though Greenland did join as a Danish county in 1972, it withdrew unilaterally after attaining autonomy. However, unlike the extension of constitutional guarantees which reflected concern with the archaic nature of previous guarantees, the provision for ethnic autonomy is less a matter of philosophy than a political response to pressure for independence. The Finnish government, the pressure of sentiment from the Swedish-speaking Aaland Islands for union with Sweden,¹¹ had set a precedent by establishing limited local autonomy for Aaland in 1920. Aaland's special status was confirmed in 1951 with a new, broader autonomy law which has constitutional status and cannot be changed without the agreement of the Aaland local parliament. Likewise, the Faeroese autonomy agreement was drafted in response to a narrow popular vote to end the union with Denmark in 1946. Greenland's autonomy was provided to preempt secessionist sentiment.

A somewhat analogous situation exists as a result of international agreementon Danish-German border. The German minority in South Jutland, though lacking any form of regional autonomy, has substantial cultural autonomy financially supported by the German government; the

Danish minority in Schleswig-Holstein has a similar degree of cultural
autonomy.

The Faeroese, Greenlanders, Swedes in Aaland, and Germans in the Danish border areas have, of course, long historical traditions behind them. Potentially more problematic is the case of recent immigrants who flooded into Swedish industrial towns in the late 1960s and early 19705; today one Swedish resident in 20 is a foreign citizen. Norway and Denmark have had a smaller, but still substantial, immigration. While residents who are citizens of other Nordic countries are accorded rights equivalent to those of nationals of the country in question by intergovernmental Nordic agreement (and residents who are citizens of other EEC countries have roughly comparable rights in Denmark), Denmark, Norway, and Sweden have recently provided immigrants with substantial political rights, including the rights to vote and to hold office at the municipal and county levels after three years' residence without citizenship.¹² While the desire to encompass the immigrants in the political community without granting citizenship itself played a role in this expansion of the franchise, so did pragmatic party considerations. The Social Democrats reasoned that the largely working class immigrants were their voters.

Whatever the causes, voting rights should enable immigrants to defend their own rights, and a number of immigrant spokesmen have won council seats. What is perhaps most noteworthy is the fact that the extension of voting rights has led to the integration of the immigrants into the existing political parties, making them substantially less exposed to political attack. At the same time, there has been an attempt to extend to them considerable rights of cultural autonomy; the Swedes have gone farthest in protecting cultural diversity by guaranteeing the right to instruction in school in the language spoken at home.

An important supplement to formal constitutional rights is access to government documents and records. The Nordic countries have quite different secrecy and access traditions. Sweden has long had the greatest freedom of information in Europe, with public documents generally open to inspection unless they contain matters sensitive to national security or personal information about individuals. Given the scope of modern government, these are considerable exceptions, but numerous internal policy documents are available. Tax information about private citizens has long been a matter of public record causing some Swedes to have to worry about the balance between public access and personal privacy. Nevertheless, agencies at all levels of Swedish government routinely display current papers for inquiring journalists and others. Such access allows the press and other 'watchdogs' to supervise the process as well as the ends of public policy.

Denmark and Norway have a tradition of secrecy that has resisted even

recent legislation to open up the government. Public access laws in the

1970s maintained barriers to those wishing to see internal administrative papers, and many administrative agencies have interpreted access requirements conservatively. The closed nature of most legislative committee hearings is another obstacle to public scrutiny. The tradition of coalition government may explain the desire to maintain the confidentiality of political 'horse trading'.¹³

Clearly national security presents the greatest dilemma to freedom of information. Norway has had several controversial clashes between the government and external critics of security policy. Sweden and Denmark also have severe restrictions not only on technical documents, but also on foreign and security policy discussions within parliamentary and governmental With a widening definition of security to include foreign economic and energy policies, the public access debate will remain prominent.¹⁴

Last but certainly not least in the list of Nordic institutional protections for individuals is that Swedish innovation: the ombudsman. The origins of this institution lie in the Swedish experience with the abuse of royal power under the final and disastrous period of Swedish absolutism that began with the royal coup d'etat in 1772 and culminated in the war against Russia in 1808 that led to the loss of Finland, almost half the kingdom, in 1809. The constitution forced on the new king, Jean Bernadotte, created the ombudsman institution placed specifically under parliament to investigate and restrain royal government. Finland added an ombudsman at independence its constitution of 1919, and, since the war, the ombudsman system has been elaborated in Sweden and has taken root in Denmark and Norway as well.

The problem of protecting citizens against bureaucracy is as old as bureaucracy itself. Charles XII, who had considerable time to observe Turkish practices during his enforced stay there after his defeat at Poltava, established the post of Chancellor of Justice (justitiekansler, known popularlyas JK) in 1713 as a royal agent to check the abuses of civil servants; that position still exists and the overlap in duties between the Chancellor of Justice and the parliamentary ombudsman is generally resolved by consultation between the two. By 1809, however, parliament insisted on its need for independent control of misadministration, and the justitieombudsman,generally today as the JO, was the result.¹⁵ Elected by parliament for a four year term, the JO's duties include the supervision of public officials (including the courts) to see that they follow the law, the inspection of public agencies, and, if necessary, the prosecution of public officials. He cannot overturn decisions or provide redress, but he has wide ranging powers to gather the relevant information from the administration, to criticize officials, to prosecute them for flagrant abuses, and to recommendchanges

ommendchangesin law to parliament. While the Swedish JO responds to formal complaints - around 3,000 a year in recent years - he also has the power to raise cases on his own either on the basis of press or other reports or as a result of his personal inspections.¹⁶ The JO has initiated 200 to 300 cases annually in recent years of which roughly half have ended without admonitions or other action.

The JO's practice differs notably from that of American redress procedures the courts. First, it only infrequently involves any kind of court action: only 52 cases out of some 32,000 cases filed between 1968 and 1978 led to prosecution or disciplinary action.¹⁷ There is no cost to the complainant and no lawyers are involved on either side, except in the rare case that ends in court. The ombudsman's sanctions are such that he is encouraged to be flexible in seeking to resolve complaints which have merit (as 10 to 15 percent are found to have), and when he finds systematic problems it is his obligation to propose legislation to alleviate them.

With the growth of bureaucracy and the extension of the scope of the JO's authority, the number of cases taken to the ombudsman has exploded from fewer than 100 annually in the first part of the last century to the current rate of more than 3,000 per year. The JO's staff has expanded concomitantly, and the institution has been extended through the creation of specialized ombudsmen for consumer affairs, the press, competition, and equal opportunity with their own staffs. The system has proven singularly adaptable to the problems of bureaucracy in modern society because of its flexibility and non-litigious nature.

Finland, severed from Sweden by Russian conquest in 1809, maintained the Swedish legal system with the institution of the Chancellor of Justice (JK) as an autonomous constitutional Grand Duchy under Czarist rule. The procurator, as the office was termed in the Czarist period, was the Czar's direct representative, and the scope of his office theoretically included controlling the legality of the actions of the Czar's Governor General in Finland. The persistence with which procurators pursued this obligation during the attempted Russification of Finland at the end of the 19th century brought them into regular conflict with the Governor General and led to the replacement of more than one procurator, but it also invested the office with a great deal of public respect. With independence, the office of Chancellor of Justice was preserved in the Constitution of 1919 as a presidential appointment, and its powers were extended by requiring the holder of the office to attend all cabinet meetings; this obligation has given the JK a preventive role that extends to checking the power even of the president. The 1919 constitution added the office of the parliamentary ombudsman, who is elected by parliament like his Swedish equivalent and endowed with roughly comparable powers to investigate complaints and to inspect public institutions. Legislation in 1933 divided functional responsibilitiesbetween

bilitiesbetweenthe two positions, delegating most inspection and all complaints concerning the military and the prison system to the ombudsman. The JK has retained the function of being a chief legal advisor to the government and, in one noteworthy case in 1923, blocked legislation supported by the cabinet by obtaining a Supreme Court review; the Court concurred that the legislation was unconstitutional and the President vetoed it.¹⁸

The Danish Constitution of 1953 introduced a parliamentary ombudsman in that country as well. The Norwegians established a military ombudsman in 1952 and a general parliamentary ombudsman (Sivilombudsmann) in 1962. The powers and status of the Danish and Norwegian parliamentary ombudsmen are similar to those of the Swedish JO. Among the Nordic countries only Iceland, with its very small population base and high ratio of elected officials to voters, has not established a similar institution.¹⁹

Direct Democracy as a Check on Governmental Majorities

If the provisions discussed above protect individuals against the arbitrary or capricious acts of bureaucrats and provide basic cultural protections for some small ethnic minorities, how can the people at large be protected against an overzealous parliamentary party majority of some duration and stability?

The Scandinavians have not been strong believers in direct democracy for its own sake. Recall and initiative, for example, are unknown. They have, however, made increasing use of referenda to resolve constitutional issues and political questions so divisive that a parliamentary majority is insufficient basis for legitimate decision. These referenda have played a crucial part in the process of political decision-making in Denmark, Norway, and Sweden in the postwar period, but not in Finland and Iceland which held their last referenda in 1931 and 1944, respectively. Lacking the relatively permanent Social Democratic dominance of Denmark, Norway and Sweden, Finland and Iceland have had less need for extra-parliamentary checks on parliamentary majorities. Elections, and the frequent changes in government that resulted, played that role.

The importance of referenda as a check on majority rule is best establishedin With 15 national referenda since 1900, the Danes are second only to the Swiss in the use of the instrument. The Danish Constitutionrequires all constitutional changes need approval by two sessions of parliament with an election intervening and by a majority vote in a referendum in which the majority for the change exceeds 40 percent of the eligible voters. Furthermore, any legislation except finance and

tax measures can be sent to a referendum by one-third the members of parliament; rejection requires a majority against in excess of 30 percent of the eligible voters. This provision represented a quid pro quo to the more conservative parties of the abolition of the upper house of parliament in the 1953 Constitution. It was a sign of the times that popular referenda could be seen as a conservative counterweight to Social Democratic dominance in government. Moreover, any legislation involving a surrender of national sovereignty can be forced to a referendum by one-sixth of the members of parliament. While such Danish referenda are binding, Swedish and Norwegian referenda have been advisory, although the new Swedish Constitution includes the possibility of binding referenda on constitutional amendments (which also go through the process of being approved by two sessions of parliament with an intervening election) when one-third of the members of parliament demand one. The Icelandic Constitution calls for binding referenda on legislation which the president refuses to approve, to remove the president. and to change the established religion, but those provisions have not been invoked.

Before World War 11, the issues going to referenda involved one of two matters. Constitutional questions went before the voters with some regularity. Norwegians ratified independence - the vote was 368,208 for and 184 against - and the establishment of a monarchy in two separate referenda in 1905; Icelanders voted on their status in the union with Denmark in 1918; and Danes approved the sale of the Danish West Indies (Virgin Islands) to the U.S. in 1916, the reunification with North Schleswig in 1920, and 92 percent supported the effort at constitutional revision in 1939, though the turnout was not high enough to ratify the changes. And as the war drew near its close, Iceland ratified its independence and the establishment of a republic - attitudes having changed in the decades since Norwegian independence - with a referendum in 1944. The other matter to go before the voters was the thorny issue of prohibition. Prohibition was enacted by referenda in Iceland (1908) and Norway (1919) but rejected in Sweden (1922). Norwegian voters reversed themselves in 1926, however, under the pressure of threatened retaliation against Norwegian fish exports by the wine producing countries; the Finns repealed prohibition by referendum the country's only use of this procedure -in 1931; and Icelanders followed suit in 1933. The defeat of prohibition in Sweden and its repeal in Norway, Finland, and Iceland by popular votes marked the end of the campaign for legally enforced abstinence at the national level, although the restrictions on alcohol continued to be an important question.

In the postwar period, only the Danes have sent constitutional issues to the people. The Danes adopted a new constitution by referendum (1953) and have held five votes on lowering the age of suffrage, while special referenda have been held in the Faeroes and Greenland, respectively, on

independence (1946) and on autonomy (1979). On the other hånd, there has been an increasing tendency to put key political issues to referenda, for, in nations that accord sovereignty to the people, referenda can settle the most divisive issues in a way that parliament cannot. Indeed, voters have repeatedly reversed parliamentary majorities. A plurality of Swedish voters approved the Social Democrats' supplemental pension plan in 1957, although there was not at the time a majority in parliament in its favor. Danish voters in 1963 rejected a group of four laws regulating land use which parliament had approved. In 1969, at the height of the student rebellion, Danish voters used a vote on lowering the voting age from 21 to 18 to demonstrate their disapproval of the rebellious young; despite clear parliamentary approval (103 to 65), voters rejected it overwhelmingly (79 percent to 21 percent), sending it down to defeat in every electoral district in the country.²⁰ Most dramatically, Norwegian voters rejected membership in the European Community in September 1972 (54 percent to 46 percent) despite the overwhelming parliamentary majority for membership.²¹ Danishvoters membership the following week (63 percent to 37 percent) and, in 1986, more narrowly approved an expansion in the scope of European cooperation (56 percent to 44 percent). In the aftermath of Three Mile Island, the Swedish government sent in 1980 the question of the national nuclear program, almost as divisive an issue as the Common Market membership was in Norway and Denmark, to a referendum.

Despite the theoretical difference between the binding referenda in Denmark and the nonbinding referenda in Norway and Sweden, the latter two governments have always followed their voters' advice except in one peculiar case: which side of the road Swedes should drive on. In 1955, when the push for referenda from bourgeois parties shut out of power was at its height in Sweden, 83 percent of Swedes voting rejected changing from driving on the left to driving on the right, despite the views of the government common sense; after this advisory referendum, the Swedish government reaffirmed its intention to make the change anyway, but that was not done until 1967. The Swedish government also drew the lesson that offering voters two choices might lead to awkward results. In both the supplemental pension question in 1957 and the nuclear power question in 1980, voters were offered three alternatives, and in neither case did any alternative win a majority. Thus the outcome was sufficiently ambiguous that there was room for interpretation.

The Danish system of relatively common referenda which can be called on legislation by recalcitrant parliamentary minorities clearly enforces some restraint on majorities. Most conceivable referendum issues cut across party loyalties, and the possibility of losing a referendum threatens substantial voter defection in the next election.²² The effect is to persuade the parliamentarymajority a preemptive compromise with the obstreperous

minority to reduce it below the crucial one-third or one-sixth may be in order. Hence issues of substantial importance which affect deeply held minority interests tend to become issues requiring two-thirds majorities. Alternatively, the minority may hold the majority in check so long through the threat of a referendum that the minority becomes a majority, as happened in the case of nuclear power in Denmark. While that is not the case in the Norwegian or Swedish systems, referenda on extremely divisive issues permit a determination of the real (if perhaps fleeting) opinion of the people who are, after all, the source of sovereignty. The agreement to call a referendum also enables the government to escape mountingly disruptive demonstrations and, at least on the nuclear power issue in Sweden, a means to make as graceful as possible an about-face.

A characteristic of the three EEC and the nuclear power referenda campaigns is that, like the prohibition campaigns of the interwar period, they were the cause of mass mobilizations in what the Scandinavians call 'popular movements' (folkebevaegelser or folkrorelser). Popular movements in Nordic political mythology, a degree of legitimacy that rivals that of the established parties, in no small measure because the term denotes the agrarian and labor movements of the 19th century. While the new popular movements have much in common with the single issue movements of other countries, they have also served as a check upon parliamentary majorities. Through their mobilization, they can check the parliamentary majority by demonstrating that it is out of step with a significant portion of not only of voters but of the majority's own voters, threatening party loyalty on election day.²³

Mechanisms that Build Consensus

Less obvious but far more pervasive is the use of consultative processes in drafting legislation at the national level which offers a voice to all major interest groups; the corporatist patterns of representation on major governmentalcommissions restrict the power of political parties to legislate as they see fit.²⁴ The general pattern in drafting major reform legislation is to turn the investigation of the area over to a royal commission which includes not only representatives of the government but also of relevant interest groups. What is striking is the degree to which these commissions give representation to interest organizations likely to be critical of the government position. The Danish Low Income Commission, for instance, established by a Social Democratic government to offer suggestions on how to make redistribution more effective, included as many representatives from business and farm organizations, which could be expected to be critical, as from sympathetic labor groups. Major interest organizations,

including the union federation, the employers' organizations, and the
farmers' associations, are represented on the majority of the important
commissions involved in drafting legislation.

The consultative process in the drafting stage is complemented in Sweden by the practice of remiss, which entails sending legislative proposals out to interest organizations for comment and suggestions for amendment. The voluminous comments - both solicited and unsolicited - not only provide expert organizational feedback, but have also enabled Social Democratic governments to integrate the organized groups supporting the bourgeois parties into the decision-making process.²⁵ While their parties may be out of power, they retain means of direct access to the decision-making process in which their influence is directly related to the expertise which they can bring to bear. The process encourages an objective debate about the real merits of the government's proposals, charting areas of consensus as well as those of disagreement.

A second practice which integrates widely divergent interests is that of enacting what are known as 'framework laws' which are essentially legal skeletons to be fleshed out in implementation through local bargaining and agreement. A classic example is the Swedish Medbeståmmandelagen (Codetermination which is generally referred to as the MBL. Enacted in 1975, the MBL radically expanded employee rights of participation in the workplace, but the actual implementation of the act was left to collective bargaining. A series of MBL agreements was envisioned: a national agreement the trade union federation and the employers' federation, agreements covering industrial branches between branch employers' organizations industrial unions, and agreements between management and union locals in individual plants. The agreements were years in negotiation, and the process encouraged a great deal of flexibility and required substantial by both employers and unions. While neither abandoned their views, both sides have become quite skilled at what we may call 'antagonistic cooperation': continued advocacy of antagonistic interests within the context of the common pursuit of productivity and efficiency. The dependency of the Nordic countries on foreign trade and their sensitivity to competitive pressure from abroad help to constrain the expression of the antagonism.

The consequences of including representatives of the major - and antagonistic interest groups in drafting and, in at least some cases, implementing legislation are of interest. First, the process includes minorities as well as majority groups. If, during long periods of Social Democratic tenure, the influence of business interests on legislation in parliament through businessorientedparties been restricted, it has often been substantial through such corporatist arrangements. Second, it has permitted the Social Democratsto their opponents divided and to limit the vehemence of the

opposition, even when the opposition seemed permanently excluded from power. Third, it has encouraged the organization of the previously unorganizedby additional and appealing channels for influence. In short, the practice checks the majority while making the minorities more responsible.

Partisan Constraints and Protest Parties

Nordic government is party government. Elections do, in fact, constitute referenda on the policies of the government and on the proposals of the government and the opposition. But they are the only such direct check by citizens in normal times. Selection of party candidates is determined by internal party decision, not by the voters. Worse, from the point of view of control of governmental abuse, Nordic party organizations are very strong by international standards. The Swedish Social Democratic organization does reach practically all the party's voters during the campaign at work and at home. Watching it swing into action in the weeks before the election is as impressive a demonstration of the power of organization as American urban machines were able to muster half a century ago at their apogee.

Strong party organization, even when parties are internally democratic (which is not always the case in Scandinavia), is no guarantee that government not abuse citizen rights. If anything, it may well increase that likelihood, since it makes it probable that the governing party can win elections through its organization even when it has committed abuses. The relative stability of party loyalties, anchored by strong organizations, makes it even more likely that abuses will have few consequences for government, at least when abuses are limited to unpopular opposition groups.

Partisan checks in the Nordic countries are of quite a different sort. They rest on (1) the fineness of the balance between parties, (2) the regular possibility of shifts in parliamentary coalitions, and (3) the low threshold that has to be cleared for new entrants into party competition. Factor in a partisan press eager to pursue the other parties' scandals and a system that guarantees equal access to the media in election campaigns and you have a formula for partisan control of partisan abuse of government: a Nordic form of 'checks and balances'.²⁶

The Danish, Norwegian and Swedish party systems diverge from the Finnish and Icelandic ones in one crucial aspect: the dominant role played by the Social Democrats. While the labor movement has been outstandingly strong in all five countries, the key variable has been the relative strength of the Social Democrats and the Communists within the labor movement. In both Finland and Iceland, they have divided labor movement between them, and, as a consequence, each has only been one party among several

in forming cabinets. By contrast, in Denmark, Norway, and Sweden, Social Democratic hegemony in the labor movement has given the party a dominant position in parliament and in forming governments, and the problem this poses for minority rights in these countries is the subject of this section. Since coming to power in the late 1920s or early 19305, the Social Democrats have governed those three countries alone or in coalition for 80 to 90 percent of the period. These are cases of almost unprecedented one-party dominance in free, competitive elections. Yet this dominance has not produced governmental abuse of minorities' rights, as in other onepartydominant

Unlike the one-party dominant systems of Northern Ireland and South Africa which rest upon the perception of threat to the dominant religious or ethnic community or the one-party dominant system of the South in the United States which rested upon the experience of Reconstruction, the Social Democratic dominance in Denmark, Norway and Sweden has rested upon a combination of hegemony within the labor movement and the disunity among the bourgeois parties. In terms of our interest here, the latter is vital, for the Social Democrats retained their dominant position in government without actually winning electoral majorities. Indeed, the Danish Social Democrats and Norwegian Laborites together have managed more than three-quarters of a century of governmental power since the Depression without ever winning a majority of the vote in any election. The Swedish Social Democrats managed 44 consecutive years in power with actual popular majorities only in 1940 and 1968. While the election systems used in Sweden prior to 1970 and still used in Norway are sufficiently biased toward large parties that the Social Democrats could win a majority of the seats in parliament with 44 to 48 percent of the vote, the Danish Social Democrats managed more than four decades in power without ever having a parliamentary majority. That fineness of balance in elections between the Social Democrats and their opponents has made the elections more sensitive to relatively small movements of voters than is common in proportional systems.

This balance is made the more precarious by the Scandinavian system of relatively equal access to radio and television during the campaign without cost. While each party is permitted to present its views without cross examination in one program, the election programs that attract greatest viewer attention are those which pit party leaders against aggressive journalists against other party leaders. In recent years, the media campaign has concluded with a televised debate among all parties' leaders in which each party leader has the chance to interrogate his opponents. The opportunity to press embarrassment home is clear, even though most leaders' parliamentary with the give and take of interpellation debates give them good training in evasion.

That precarious balance is also reflected in the relatively frequent shifts in parliamentary coalitions. Given the distribution of voters and the election system, the last time any party won a majority alone in parliament was in 1968.²⁷ The immediate consequence of this situation is that the Social Democrats, while dominant, are sensitive to any issue that may forge a united front among their opponents. Since that typically requires that parties both to their left and to their right combine against them, Social Democratic governments are safe on all economic questions on which the left wing parties have to vote with them. Similarly on defense and foreign policy issues, an unholy left-right coalition against the Social Democrats is ruled out by the commitment of most parties of the right to defense compromises with the Social Democrats, while the Social Democrats can typically pick up votes to their left on any foreign policy issue on which they face a united right.

Questions of civil liberties and governmental abuse of citizen rights are, however, clearly a danger zone. The parliamentary parties to the left of the Social Democrats - the Socialist People's Party in Denmark, Socialist Left Party in Norway, and Left Party-Communists in Sweden - are far more libertarian than the pro-Moscow Communists whom they displaced in parliament in the 19605. Indeed, polls regularly demonstrate that, on most issues of protection of individual rights, their voters are at least as liberal as the most liberal of the bourgeois parties. Only on questions concerning property rights does this generalization not hold.

But what happens when all the major competing parties agree on a policy? Given the Scandinavian pattern of seeking broader than majority coalitions to pass major legislation (which means that the legislation will stay on the books after the opposition replaces the current government, since one or more opposition parties participated in shaping the legislation) even at the cost of significant compromise, this happens with some frequency. Under these circumstances of oligopolistic party collusion, the partisan checks just discussed cannot be expected to work.

Here the safeguard in the party system is the low threshold for new competitors. Not only is it easy to get on the ballot, it is also fairly easy for new parties to win seats. Any party that polls more than two percent of the vote in Denmark or more than four percent in Sweden gets the same proportion of seats in parliament as it does of the vote. The Norwegian election law uses medium-sized multimember districts without supplemental seats, so it is harder to generalize about the threshold, but parties typically can win seats with about three percent of the vote and achieve reasonable proportionality with about ten percent. Given the even balance between the labor and bourgeois blocs in Norway, even a small number of seats - the Socialist People's Party's two seats in 1961 or the Progress Party's two seats in 1985 - may be as decisive to give the party prominence as the six

or eight seats to which the party's vote might mathematically entitle it. The consequence is both that new parties win representation with some regularity and that old parties easily become sensitive to need to incorporate new demands into their own repertoire to avoid the formation of new parties.

Groups of the left, right, and center have all made use of the opportunity to organize new parties. Despite the low threshold for representation, most have failed. Pensioners' parties and environmentalist parties have been frequent contestants which have yet to succeed in winning parliamentary representation. The reason is the obvious one: both pensioners and environmentalists no lack of spokesmen in the existing parties.²⁸

On the other hand, parties reflecting interests not represented in the old parties because of their consensus have repeatedly broken into parliament. In the early 19605, it was the left socialists who were not tied to Moscow. The Social Democrats had moderated their positions over three decades in government, while the Communists were condemned to ineffectiveness by their foreign ties. The bourgeois parties, of course, made no pretense of representing radical working class interests, although at least in Denmark some working class radicals voted for the League for Justice, the Danish followers of Henry George, in protest in 1957. Independent radical socialist parties won representation in Denmark in 1960 and in Norway in 1961; the Swedish Moscow-loyalists compromised with the Eurocommunist wing in 1964 to avoid the latter's independent organization.

Similarly, Christian parties were organized in Sweden and Denmark in the 1960s - a comparable party had been organized in Norway in the 1930s - to represent the interests of more fundamentalist Christians against broadly supported social and cultural policies. The liberalization of legislation on abortion and pornography with broad parliamentary support left many cultural conservatives with the perception that the traditional non-socialist parties offered not a choice but an echo; they turned to the new Christian parties which succeeded in winning parliamentary representation in 1973 in Denmark and 1985 in Sweden. Access to free media time for even the smallest parties has played a role in their relative success.

Conclusions

The Nordic democracies are characterized by political institutions and patterns of power that make the abuse of power both by government officials and by parliamentary majorities more likely. The government is centralized in its structure, without the mechanical checks and balances that limit the abuse of power in part by limiting its possible exercise. Parliamentary majorities are inhibited neither by judicial review nor by the

independent powers of subordinate units of government. Not only are party loyalties strong and the parties well organized, but the dominance of the Social Democrats in government is such that they have set longevity records in winning free elections that parallel those of the best organized urban machines. Traditions of governmental restraint prior to the establishment of democratic government were not well established.

Yet the Nordic countries have paradoxically proven models in the area of the protection of civil liberties for individuals and for minority groups. We have found unpersuasive, for the reasons suggested above, the explanation the political culture was predisposed in this direction. The Scandinavians have developed some new institutional checks on the governors, with regard to individual and group rights, but the latter, in particular, are effective principally because of the balance of political forces in the society. With the exception of the Chancellor for Justice in Finland and Sweden and the ombudsman in four of the five Nordic countries, none of the checks can be termed politically neutral; these institutions principally offer protection for individuals against arbitrary acts or the caprice of bureaucrats.

The other checks rest fundamentally on the distinctly nonconsensual assumption that basic conflicts of interest are the norm in politics. These antagonisms are a fact of political life; the degree of bitterness with which they are expressed can be ameliorated, but the antagonisms themselves cannot be eliminated. If conflict is inevitable, its consequences are best limited by permitting every party to the conflict fair representation for its most articulate spokesmen. A number of the restraints which have been institutionalized have promoted a pattern of bargaining which has made traditional patterns of political confrontation less relevant by simply bringing expression of conflicts of interest inside organizations. That is more obvious in the workplace, but it exists also in the remiss and commission structures.

The existence of points of multiple and easy access for registering effective opposition has decreased the degree to which the even cohesive majorities can dominate the political system. Referenda blocked Danish land use laws, Norwegian entry into the European Community, and the further expansion of the Swedish nuclear power program. Glistrup's tax protests quickly dissolved the consensus on taxes that had gradually grown up among the major parties in Denmark. Given new parties' easy access to parliament, oligopolistic all-party agreements in Scandinavia cannot guarantee permanent majorities, even when supplemented by corporatist arrangements.

It is typical of Scandinavia that significant, perhaps transforming, reforms
have grown out of the accretion of practice, not out of some grand philosophicalor
design. What has emerged is a democracy that is

both more participatory (especially through corporatist channels) and more
limited than the traditional parliamentary model.

---

NOTES

1. The Norwegian Parliament attained supremacy in domestic affairs in 1884, but Norwegian foreign affairs and defense policy remained the province of the Swedish king untii Norwegian independence in 1905. The situation was similar in Iceland, where parliament acquired autonomy in domestic affairs in 1918 and control of foreign policy with independence from Denmark in 1944.

2. Universal manhood suffrage was attained in Norway in 1898, Finland in 1906, Denmark and Iceland in 1915, and Sweden in 1909 albeit weighted in relation to capital and estate ownership. Women were enfranchised in Finland in 1906, Norway in 1907, Denmark and Iceland in 1915, and Sweden in 1921.

3. During its brief independence in the Spring and Summer of 1814 after four centuries of Danish rule and before the Swedes established their sovereignty, the Norwegians drafted a constitution that drew on a variety of sources including the Swedish Constitution 1809, the French Constitution of 1791, the American Constitution, and the work of Montesquieu. The relative independence of the Norwegian judiciary is a result.

4. The public sector is somewhat smaller in Finland and significantly smaller in Iceland.

5. While the Social Democrats held power for intervals in the 19205. they solidified their control of government during the Depression. Since then, the Social Democrats have held office alone or in coalition in Denmark for 43 years since 1929; for 49 years in Sweden since 1932; and for 36 years in Norway since 1935. The German occupation accounts for two of the years that they were out of power in Denmark and five in Norway. In both Iceland and Finland, the labor movement split rather evenly between Social Democrats and Communists, and the Social Democrats were unable to exercise the same dominance.

6. Anthony Upton comments that "Scarcely a working class family in the land did not have some direct experience of repression or injustice at the hånds of the victorious Whites . . . and even today the resentments and hatreds have not wholly vanished." 'The White Terror was,' he notes, an example of majority tyranny; it was 'a popular, even a democratic phenomenon, in the sense that it represented what the minority of bourgeois activists wanted, and what the majority of rank and file White supporters were ready to approve" Upton (1980), p. 522.

7. In appreciation of German aid during the Civil War, Finland established itself as a monarchy in 1918 and offered the crown to Prince Friedrich Karl of Hesse, Kaiser Wilhelm's brother-in-law. He declined it in December 1918 after Germany's surrender.

8. The highest courts of Iceland, Denmark, Sweden, and Norway all claim some powers of judicial review. Only in Norway has the court actually struck down legislation, and that on procedural rather than substantive grounds. To avoid the constitutional problems that the passage of substantively unconstitutional measures would raise, the Norwegian parliament may request the view of the Supreme Court before enacting legislation, and the Swedish practice is to send many bilis to review by the Lagraadet (the Law Council) which includes members of the Supreme Court.

9. The latter Norwegian action was occasioned by embarrassing external pressure from the Council of Europe. Complete religious liberty in Norway followed eight years later. Cf. T. K. Derry (1973), p. 413.

10. Consider freedom of the press. Article 100 of the Norwegian Constitution provides that 'There shall be liberty of the Press. No person must be punished for any writing, whatever its contents may be, which he has caused to be printed or published, unless he wilfully and manifestly has either himself shown or incited others to disobedience to the laws, contempt of religion or morality or the constitutional powers, or resistance to their orders." (Italics added).

---

11. A 1919 referendum in the Aaland Islands resulted in 96 percent voting in favor of union with Sweden. The Finns jailed secessionist leaders and put the islands under military control. The islands were awarded to Finland by the League of Nations in 1921.

12. Finland and Iceland provide similar rights to citizens of other Nordic countries after two and three years' residence, respectively. Nordic citizens' rights in other Nordic countries are described in detail in Nordisk Ministerråd, Nordbors råttigheter i Norden (Oslo: Nordiska Rådet, 1986.)

13. Einhorn (1977).

14. Gleditsch & Høgetveit (1984).

15. Ombudsman means literally 'representative' and the term is used to refer to every conceivable sort of representative, including shop stewards and plant safety representatives. Hence the use of JO to refer to the rather pompous justitieombudsman or 'representative of justice'.

16. One of the most spectacular cases arose from the ombudsman's inspection of a debtors' prison in 1824. The ombudsman found a prisoner who had been held for 23 years following his bankruptcy in 1800. He had been sentenced to two hours of public humiliation and had appealed the sentence. The court never ruled on his appeal, and prison officials refused to release him because there had been no judgement. When the ombudsman went to the court, he found that the file had long been lost. (Personal communication, Judge Ulf Lundvik).

17. al-Wahab (1979), pp. 165-185.

18. For a discussion of the Finnish Chancellor of Justice, see Kastari (1965).

19. The ombudsman institution has been the subject of substantial interest outside Scandinavia, hence a good many accounts are available in English. For further information, see al-Wahab, op. cit., Rowat, op. cit., and the articles on the Danish, Finnish, Norwegian and Swedish ombudsmen in Caiden (1983). The Scandinavian model has influenced Australia, Britain, New Zealand, and a number of local governments the United States and Canada which have established variants of the ombudsman institution since 1960.

20. Cf. Nielsen (1970). Eventually the voting age was lowered in referenda to 20 in 1971 and to 18 in 1978.

21. For a blow-by-blow account of decision making in the EEC issue that culminated in the referendum, see Gleditsch & Hellevik (1977).

22. The clearest example of this threat are the back-to-back blows delivered to the Labor Party by the EEC referendum in 1972 and the general election in 1973. For a treatment of this, see Valen & Martinussen (1977). For a general account of Scandinavian referenda and their impact, see Nilsson (1978).

23. Direct citizen input at the local level is less common. The one notable example is in urban planning. The outstanding law is the Danish municipal planning act of 1975 which provides for neighborhood review of city and neighborhood plans for development and renewal. In recent years the city of Copenhagen has had its planners prepare alternative local redevelopment plans so that citizens have a basis for informed choice. Neighborhood are consultative rather than binding, but the procedure forces the planning bureaucracy to anticipate local protests and tailor plans to avoid them. For an account of this system, see 'Citizen Participation in Planning Decisions of Public Authorities', Notat 93 (Hørsholm: Statens Byggeforskningsinstitut, 1979). Similarly in Oslo, proposed changes in zoning go to consultative neighborhood committees.

24. Scandinavian societies are exceptionally well organized. The rates of organization among economic interest groups - farmers, workers, employers, shopkeepers, and evenconsumers-are the highest in the West. This is the source of need for corporatism in Scandinavia.

25. Cf. Vinde & Petri (1978), pp. 24-29.

26. Madison's expectation that 'ambition must check ambition' (Federalist 51) is thus operative in an unlikely manner.

27. In an election held immediately after the Soviet invasion of Czechoslovakia, the Swedish Social Democrats, appearing as the party of stability in an extremely tense international atmosphere, won a majority of both the parliamentary seats and the popular vote.

27. In an election held immediately after the Soviet invasion of Czechoslovakia, the Swedish Social Democrats, appearing as the party of stability in an extremely tense international atmosphere, won a majority of both the parliamentary seats and the popular vote.

REFERENCES

Caiden, G. E., ed. 1983. International Handbook of the Ombudsman: Country Surveys.
Westport, CT: Greenwood.

'Citizen Participation in Planning Decisions of Public Authorities'. Notat 93,1979. Hørsholm:
Statens Byggeforskningsinstitut.

Derry, T. K. 1973. Modem Norway. Oxford: Clarendon Press.

Einhorn, E. S. 1977. 'Denmark, Norway and Sweden', in Itzhak Galnoor, 1., ed., Government
Secrecy in Democracies. New York: Harper & Row.

Gleditsch, N. P. & Hellevik, O. 1977. Kampen om EF. Oslo: Pax.

Gleditsch, N. P. & Høgetveit, E. 1984. 'Freedom of Information in National Security Affairs',
Journal of Peace Research 21, 17-45.

Kastari, P. 1965. 'Finland's Guardians of the Law', in Rowat, D. C, ed., The Ombudsman.
London: Allen & Unwin.

Madison, J. The Federalist, no. 51.

Nielsen, H. J. 1970. 'Voting Age of 18 Years', Scandinavian Political Studies 5, 301-305.

Nilsson, S. S. 1978. 'Scandinavia', in Butler, D. & Ranney, A., eds., Referendums: A
Comparative Study of Practices and Theory. Washington, D.C.: American Enterprise
Institute.

Nordiska ministerrådet (1986), Nordbors råttigheter i Norden. Oslo: Nordiska rådet.

Upton, A. 1980. The Finnish Revolution 1917-1918. Minneapolis: University of Minnesota
Press.

Valen, H. & Martinussen, E. 1977. 'Electoral Trends and Foreign Politics in Norway: The
1973 Storting Election and the EEC Issue', in Cerny, K. H., ed., Scandinavia at the Polls.
Washington, D.C.: American Enterprise Institute.

Vinde, P. & Petri, G. 1978. Swedish Government Administration. Stockholm: Prisma, 2nd
Ed.

al-Wahab, I. 1979. The Swedish Ombudsman: An Instrument of Human Rights. Stockholm:
Liber.