For some decades the twin concepts of centre and periphery have been used rather commonly in political science. First, perhaps, by the late Stein Rokkan of Norway as a tool for the better understanding of differences in political culture within single countries. Later on by neo-Marxists, especially the Althusser school, as a means for pinning down the dependence of developing countries on the economy and politics of the developed ones, the struggle of the national, necessarily peripheral bourgeoisies of the Third World to free themselves from the dominance of the established centre bourgeoisies.

It seems to me that the concepts of centre and periphery are most convenient for grasping the crux of the matter concerning the state of the realm of Political Science. This is perhaps especially obvious to a late-comer into the field from a small country. I intend to look at, and to judge, the centres and their influences from the point of view of peripheral Danish political science.

A regular study of political science was not established in Denmark till 1959, at the University of Aarhus, other universities following the lead during the next decades. Initiatives in the eighteenth and nineteenth centuries had failed. Politics was till then a subject for historians, and, to some degree, for jurists and economists, and they studied it, of course, from their several and special points of view.

Thus, in 1959, the Aarhus founders (Poul Meyer, by education a jurist, and myself, a historian) were unhampered by any local tradition. We simply had to find out for ourselves what political science is about or, better, what we found it ought to be. We were, literally, a new-born and distant asteroid looking for some centre or centres to gravitate towards.

Perhaps I ought to remind the reader of the position of Denmark (and the
other Nordic countries) on the cultural map of Europe, bordering on three main
centres: Germany, France and Britain, heavily influenced by them but, having

---

* Paper presented at the Joint Symposium of the Finnish Political Science Association and the International Political Science Association in honour of the 50th Anniversary of the Finnish Political Science Association. Helsinki, 2-6 Oct. 1985. The theme was 'Development and Institutionalization of Political Science: Centre-Periphery and Other Crucial Concepts'.

some possibility of choice, never submerged by any one of them. So, to look around and to choose, to some degree consciously, is a customary Danish (and Nordic) way of life, the privilege of being peripheral with respect to several centres. Since around the First World War North America has added considerably to our possibilities of cultural choice.

As regards political science, in 1959 there was no question that America was the centre, presenting just at that time w,ith Messianic zeal the gift of behavioralism like some sort of Marshall aid to behind-lagging Europe, including quite undeveloped Denmark. We accepted the gift, greedily, for what it was worth, wondering perhaps a bit about its universal applicability.

Still, a Dane is a European, accustomed to looking to our Nordic neighbours (first and foremost Sweden with its venerable tradition in the field and the only specialized political science journal) and the three traditional centres. German Staatslehre was clearly outdated, any more up-to-date political science still in its birth pangs. France could offer the electoral geography of André Siegfried and Francois Goguel and the brilliant and typically French taxonomic logic of Maurice Duverger's Les partis politiques. In Britain Sir Ernest Barker was still a prominent figure, while D.E. Butler and his collaborators had started their reports on the British general elections and Samuel Finer had just recently treated the influence of interest organizations in his Anonymous Empire (as had Gunnar Heckscher in Sweden).

There was, of course, much more to be learnt from European colleagues (e.g. the Swede Herbert Tingsten). Still, a European map of political science around 1960 bore resemblance to the map of Africa a hundred years before, while the Americans seemed to be on their way to penetrate and eliminate all black areas, riding high on the tidal bore of behavioralism and welcomed by forerunners like Stein Rokkan of Norway and Erik Allardt of Finland.

This is not the place for writing an epitaph to behavioralism. That was done by Robert Dahl already in 1961, and a lot of his fellow-countrymen were eager to tell us what their new gospel meant. For us in the awakening Danish periphery it certainly contributed heavily to moulding our answers as to what political science is and ought to be. A few words will elucidate our situation.

First and foremost, political science had to be an empirical venture, taking advantage of any method, in being or to be invented, for precise information about political phenomena. This goal, including an ever on-going refinement of research techniques, has never been given up, even if, hopefully, it is always qualified by the Aristotelian warning (cited by Carl Friedrich (1963, 8), whose attitude to behavioralism may perhaps rightly be characterized as that of a sympathetic right-wing sceptic) that it is the mark of an educated man to look for precision in each class of things just so far as the nature of the subject matter admits.

Just as important for us was the behavioralist thrust for theory (even if often
enough only in the guise of conceptual frameworks) by which to master newly

harvested data and to be guided in search for new crops. The need for conscious theory was evident. There were micro-theories for the understanding of the decision-making process, like those of Herbert Simon and Richard Snyder, and broader macro-theories for the mapping of interrelations between the political system and society, offered by e.g. David Easton and Gabriel Almond. Ingenious minds were, perhaps, too eager to formulate the theory of political science. The result was that no single theory dominated the field. One had to choose.

Most promising seemed to me the endeavour of David Easton in The Political System (1953). I shall return to a discussion of some of its merits and demerits because such a discussion may, in my opinion, still contribute to the better understanding of core problems of our discipline.

But before doing so, I want to consider the role of theory, and especially general theory, at that time and today. What happened in this respect in political science was by no means an isolated phenomenon. It followed in the wake of what Parsons and Merton and others did in sociology and other social sciences. Of late, Easton has hinted (Easton 1985, 139-40) that the commitment to the introduction of science into the study of politics was partly due to menacing McCarthyism. It is not for me to deny that this may have been so. But I would like to add another Cold War perspective which seemed to me — and still seems — evident. I looked upon the ambitious theorizing of American social science, e.g. the Eastonian systems theory, as somehow, maybe often unconsciously, provoked by the blatant claim of Marxism to be the Social Science.

The allurement of that claim was to be felt around 1970. Marxian and neo theories presented themselves to many students and scholars as superior to bourgeois social science and, in the wake of the Vietnam War, as a means of fighting back against overwhelming American influence on modern social science, which was looked upon as favouring the Establishment. This onslaught was more impressive, perhaps, in Continental Europe than in the Anglo-Saxon countries, but it was of some consequence everywhere. That is why, when speaking of Danish political science in terms of centre-periphery, I should add to the three traditional centres of influence and the more recent and potent American one the neo-Marxist trends as a fifth, non-geographical centre (certainly not to be identified with Soviet Russia), which influenced many, obsessed some and could be ignored by none.

It would be childish to deny that Marxist theory instigated to some good — and quite a lot of bad — work. At the same time it spurred non-Marxists to think afresh, theoretically and otherwise. For some years the tide has been receding. We are left now with what is sometimes called post-behavioralism. An odd label! When I judge from what I see most of my younger colleagues doing I should say that this is not /røtf-behavioralism, this is behavioralism rejuvenated and revivified, rid of some naivety and, in many ways, broadened in scope and interests.

Recently, another bid has been made (March & Olsen 1984) for a new label:
the New Institutionalism. The article is certainly rich in valuable observations,

wise comments and suggestive indications of theoretical possibilities. I think, however, that its authors grossly underestimate behavioralist interest in institutions. No one, at any rate in the Scandinavian periphery, really doubted the basic importance of the study of formalized structures (or whatever institutions were called), and most of actual research focused on them. The authors want to deemphasize the dependence of the polity on society in favour of an interdependence between relatively autonomous social and political institutions'. A change of emphasis is no break with the past. The new institutionalism seems to me an enrichment of behavioralism, not a denial of its tenets.

One of the main differences from classic behavioralism may be less preoccupation nowadays with general theory. In the late sixties, already, it was easy to observe, and not quite impossible to understand, that grand theory was often spoken of as hypertheorizing (as opposed to the war-cry of the fifties against hyperfactualism); there was some tendency towards getting down to facts, once more. It is my impression that this tendency was strengthened by the Marxist wave, not only because it made people distrust any ambitious theory, but also because that wave carried with it a lot of those inclined to theorize, leaving to the more down-to-earth and empirically minded the carrying on of ordinary day-today research.

Though instinctively no believer in grand theory, it is my contention that the preoccupation of the fifties with general theory was not a futile obsession, but somehow a fundamentally healthy undertaking, even if the results were rather inconclusive, and that such an undertaking must be taken over as a commitment by today's sort of behavioralists. We may never know, or agree on, what political science is all about. Still, some striving for some kind of answer is a prerequisite to any such discussion as that of this symposium about centre-periphery and other crucial concepts. That is why I shall dwell obstinately on questions of that kind and their possible answers in what follows.

It is true, as David Easton has of late remarked (Easton 1985, 133-34), that political science has not yet arrived at a consensus on how to describe its subject matter at the most inclusive level. He reiterates his own definition (from The Political System, 1953) of political science as the study of the way in which decisions for a society are made and considered binding most of the time by most of the people, i.e. the authoritative allocation of values (valued things) for a society.

What I have always found most promising about the Eastonian definition is that it allows for what I call an aspect point of view, politics being not some separate kind of activity, but an aspect of a lot of activities which are simultaneously relevant in other respects, economically, culturally, and so on. In that way, the political system is looked upon as a sub-system of society, political science correspondingly as studying more or less the same phenomena as do economics, sociology etc. but in a different perspective. The aspect point of view stresses the unity and the variety of the social sciences (cf. Rasmussen 1970), a very major advantage. We might speak of which aspects of activities are central,

which ones peripheral to the various social sciences.

In Easton's 1953 book, decisions or allocations seemed to be authoritative for a society simply when considered binding most of the time by most of the people. Later on (Easton 1965, 349-50) he must be understood to mean by authoritative decisions those made by the authorities, seemingly identifying, as did Gabriel Almond and other structural-functionalists, authority with legitimate use of violent force. If so, the definition, even if couched in behavioral terms, is in fact equivalent to defining political science by the state so far as modern conditions are concerned. That would be a major set-back.

The state is of necessity a prominent subject of contemporary political science (it is paradoxical that behavioralism left the field of state theory to neo But it is just as evident that politics is a much wider subject whether defined as allocation of values for a society or in any other terms of activity. Some crucial allocation of values simply takes place in some countries within and in others outside the state apparatus and in still others at some time within and at some other time outside that apparatus, according to circumstances of each particular situation. This demonstrates that any definition of political science by means of institutional terms, however couched, will necessarily be inadequate as far as allocation of values for a society is concerned. Nor will, I presume, the New Institutionalists deny that.

The most striking example of this is provided by the various systems of bargaining about wages and other labour conditions. Recently, in his Dilemmas oj Pluralist Democracy (1982), Robert Dahl has stressed repeatedly as a major point the differences in this respect between the United States and, especially, the Scandinavian countries. Writes Dahl (p. 80):

... precisely because they make the problem so starkly visible, the systems of corporate pluralism that have evolved in the Scandinavian countries are of particular relevance ... a significant degree of control over crucial economic decisions, which in conventional democratic theory should remain with the citizens' representatives in parliament and cabinet, has plainly been transferred to a kind of non-elected parliament of industry consisting of the heads of the peak associations ... I find it hard to resist the conjecture that we are witnessing a transformation in democracy as fundamental and lasting as the change from institutions of popular government in the city-state to the institutions of polyarchy in the nation-state.

This 'transformation' is by no means a recent pheonomenon; in Denmark it even dates back to 1899. Nor is it so complete as Dahl seems to imagine; the fact is that bargaining is concluded sometimes by the labour market organizations themselves (in Denmark e.g. 1981, 1983), sometimes by state intervention (1977, 1979, 1985) and always (1985 being possibly an exception) in some sort of interaction between organizations and state authorities. But these qualifications just underline that if we concede that kind of allocation to be an important subject matter of political science in 1977, 1979 and 1985, it must be so also in 1981 and 1983. Put in another way: what matters is the conclusion, not the way in which the

bargaining is concluded. So, whatever institutions, public or non-public, partake
in the conclusion, i.e. in the allocation of values for the society, must be considered
components of its political (as well as its economic) system.

In the years before 1899 (in Denmark, somewhat later in the other Scandinavian countries), i.e. the decades of early industrialization, wages were settled not by the state, but, at any rate as late as in the 1880s, by the market, especially so after the abolition of the guilds (of medieval origin) in 1857. That period was the heyday of economic liberalism. Consequently, we are forced to operate with in principle at least three possibilities (which are intermingling): wages are settled 1) by the market, 2) by collective bargaining or 3) by state fiat. All three possibilities are relevant to political science (as well as to economics). For the sake of argument it may be convenient to bear in mind the fact that the mercantilist state did not shrink from settling or influencing wages and labour conditions, regulating for instance the guilds and introducing in 1733, during a severe agricultural crisis, some sort of mild serfdom (abolished in 1788). Such an example illustrates that allocation of values simply by the market is an historical incident, not a 'natural' state of affairs, but a way of allocating to be chosen or not. This is not a Marxist contention, it is shared by an outspoken anti-Socialist like Anthony de Jasay (1985) in his thought-provoking considerations about what he names the Adversary State, i.e. welfare democracy.

Absolute monarchs strived to uphold the unity of church, caring for the salvation of the souls of their subjects as well as the furthering of obedience to themselves. History as well as contemporary experience teaches us that human rights are not self-evident, their observance being an exception rather than a rule.

Religion, culture, economics, every aspect of human life is potentially political, and consequently pertinent to political science. The frontiers between a private or personal and a public sphere are ever-changing, and mighty structures like the church of the Middle Ages and the modern labour market organizations are prone to crop up to fill gaps left open by the state authorities. So, if we want to define political science as the study of the allocation of values for a society — and no much better definition is known to me — we have to take into consideration any allocation which is considered binding most of the time by most of the people, irrespective of the status of the decision-makers (or, possibly, the non-decision makers).

Allocation of values' runs the Eastonian phrase. He seems to have in mind, perhaps exclusively, valued things. Distribution in a wide sense of the word is certainly a most important task for a political system. But not the sole one. Ought society to be socialist or capitalist or something in between? Democratic or authoritarian? What about human rights? And foreign policy? Questions like these indicate that a major task is the setting of goals, the choice of values, which definitely affect the allocation of valued things but go far beyond that.

Consequently, when paraphrasing the Eastonian definition I prefer to talk of
politics as the choice between values and the allocation of valued things for a

society. That is a reading which was not contradictd by the original text of The
Political System.

Widespread among classic behavioralists (though not shared by Easton) was a belief in the desirability of value-free research, maybe seconded by the presumed withering away of ideologies, but not in consonance with the contemporary reiteration and detailing of the Weberian point of view by Gunnar Myrdal (1958) and Arnold Brecht (1959). Probably nobody will to-day doubt the inevitability of value-consciousness at every stage of the research process. This means that another naive notion of classic behavioralism has been given up, the fancy that modern political science was essentially different from and superior to the classical theorists. They are cited, nowadays, not for learned embellishment but as a means of clarification of fundamental issues. The realm of values regained its status as an important subject of political analysis.

Value-consciousness both weakened and strengthened traditional empirical research in face of the neo-Marxist onslaught. One had to concede hidden dependence on 'western' values, but even so was allowed and obliged to doubt the claim of Marxism to be objective knowledge. For some time it was rather fashionable to maintain any proposition by reference to some, per se irrefutable, value premise. However, even if it is not scientifically possible to refute any value premise, its empirical relevance is still a question to be discussed and perhaps sometimes settled by political scientists.

It is well worth noticing that Robert Dahl, without any explicit discussion of value theory per se, uses a great many pages in his Dilemmas of Pluralist Democracy to discard venerable value premises by way of trying to demonstrate their impracticability and, consequently, their irrelevance. If political scientists ought to engage themselves, as is now fashionable and, I think, laudable, more than did their predecessors some decades ago in the discovering and solution of the problems of their societies they will add to their chances of avoiding futile considerations by following the example of Dahl and a good many other levelheaded persons. Lack of imagination seems a less imminent danger than sheer arbitrariness.

The consequence of what I have been pleading here is that politics must be understood by means of some sort of systems theory, not necessarily the Eastonian one, but some cohesive conceptual framework that permits some basic preliminary understanding of those aspects (normative as well as cognitive) of cooperation and conflict in society which concern in a more or less decisive way what most of the people abide by or acquiesce in most of the time.

Cooperation and conflict are the most crucial concepts of any empirically oriented political science. Without some measure of cooperation, however involuntary, there will be no society, nor a political society; without conflicting views and interests there will be no need for politics. To acknowledge that, and consequently to study how cooperation is brought about and how conflicts are solved (or not solved) by what is nowadays usually spoken of as the structures

and processes of a political system (part of which is a state apparatus), is not, as was often maintained some time ago, synonymous with accepting or supporting the Establishment (or whatever that system may be dubbed). It is simply a prerequisite of trying to understand what is going on around us and concerning us, however just or unjust we may judge that these goings-on are.

On the contrary, the meticulous value-consciousness of up-to-date political science is a constant reminder of the intricate interplay of the cognitive and the normative aspects in our discipline as well as in society, which is its subject matter. Value-consciousness offers as far as is humanly possible (but that is a grave warning) a fair consideration of Is as well as of Ought to Be and, thus, of evaluating the one and the other and their interdependence.

So far, I have advocated a modest revival of general theory as a framework (certainly not a strait jacket) suitable for summing up the present purview of our discipline, including its relations to other social sciences, and for triggering off still further research. The call for theoretical consciousness applies to lower levels of theory as well. In that respect there is no difference between macro- and microanalysis. But there are other differences in between them, which are difficult to overcome. Marxism certainly did not succeed in doing it. Maybe the two approaches ought to be considered complementary, macroanalysis focusing on the conditions and activities maintaining and changing whole systems, microanalysis on the single actor (an individual, a political party, and so on) in a concrete or a not completely abstract situation. Macroanalysis tends to minimize the importance of the human will, microanalysis to presuppose voluntarism to some not inconsiderable degree. The possibility of their being complementary ought not to serve as some sort of pulvinar diaboli. On the contrary, the seemingly insurmountable difficulties on the status of the will and how it is conditioned challenge the ingenuity of researchers who aim at covering some of the ground or, at least, mapping out promising strategies. That, perhaps, is the crux of political science, often evaded but puzzling our minds ever again, especially so if we pride ourselves on having disclosed the superficiality of crude empiricism as well as of shallow theorizing.

Neither Is nor Ought is at the centre, they are both peripheral in relation to
some hidden point where their radii, possibly, meet. They are like points of
departure, the point of arrival being so far unattained and perhaps inaccessible.

These last sentences may sound like the first flappings of an Icarian flight. That is not what I intend. I just wish to point to political science as being central to the interest of political scientists, but simultaneously peripheral to any full understanding of the whole of its own subject matter, because it is peripheral to other social sciences, to philosophy and a good many other subjects of human enquiry, just as they are peripheral to political science. John Locke — a favourite of mine — warned scientists aginst cantoning out to themselves a little Goshen in the intellectual world where light shines and day blesses them, but giving up to night and darkness the rest of the vast expansum, thus avoiding coming near

it (Yolton 1977, 319-24). That warning, uttered nigh on three hundred years ago by a down-to-earth English philosopher, still carries full weight. Besides, it illuminates — or so it seems to me — the advantages of being peripheral to several centres, geographical or non-geographical, within one's own discipline and in relation to mainstreams of human thought. Being peripheral adds to our chances of avoiding that Goshen business. That is the blessing of being peripheral, and at the same time a solace for being, for good reasons, ignored by the centres.

Danish political science can boast of no epoch-making contribution to our discipline. There is a lot of good research, mostly — as is natural for a late-comer into the field — concerning Danish politics and only part of it written in foreign languages, mostly English. I guess that some of my young colleagues are biding their time and may some day come to the fore.

Talking above of complementarity, I intended it to be understood in the concise way in which it was used by the physicist Niels Bohr. His was probably the most important contribution of Danish twentieth century science to what Locke meant by human understanding. Bohr argued, certainly with decent modesty, that his theory of complementarity might be useful in other fields as well, mentioning physiology, psychology and even the social sciences. My adding a warning against any too rash application of the concept of complementarity to our field does not imply a rejection of its possible fruitfulness. After all, what atomic research taught physicists about the frontiers of experimental science seems somehow familiar to us. Certainly, we have struggled with the problems of subject-object relations and the interference of observation with observed objects (i.e. the crux of the matter for atomic physicists) for a long time, a question one might say of what is centre and what is periphery. It remains to be seen if the concept of complementarity might help us to formulate a consistent theory of understanding out of our now rather unconnected ideas of scientific value relativism, of the role of the human will, of causal versus functional or teleological explanation, of macro-micro-relationship, etc. I incline to believe: that the trial ought to be ventured on, not knowing, of course, whether the results might be positive or mainly disappointing.

It is well worth remembering that the concept of complementarity did not undo the classic Newtonian physics of the macroscopic world. On the contrary, Bohr's so-called correspondence argument maintained close connection between traditional and atomic physics. If by a revolution is meant a turning upside-down of something, the acceptance of complementarity was not a revolution (nor was Einstein's theory of relativity). That is why Kuhn's choice of wording in his The Structure of Scientific Revolutions (1962) seems to me rather misleading or, at any rate, to have misled a number of social scientists. There has for centuries in our fields, and especially in political science, been an unhealthy' tendency towards revolutionary zeal, i.e. a crusader-like belief in own methods; and theories combined with contemptuous condemnation of those of the predecessors.

That seems to me a symptom of immaturity. The obvious imperfections of, say, Aristotelian teleology, natural rights theory, Marxian class struggle or classic behavioralism do not invalidate their assumptions altogether. To start afresh every thirty years may be refreshing for the initiators, but unhealthy for the discipline. That is why I have pleaded here for some measure of continuity.

REFERENCES

Brecht, A. 1959. Political Theory. Princeton, New Jersey: Princeton University Press.

Dahl, R.A. 1961. 'The Behavioral Approach in Political Science: Epitaph for a Monument to a
Successful Protest', American Political Science Review 55, 763-72.

Dahl, R.A. 1982. Dilemmas of Pluralist Democracy. New Haven: Yale University Press,

Easton, D. 1953. The Political System. New York: Alfred A. Knopf.

Easton, David 1965: A System Analysis of Political Life. New York: John Wiley & Sons.

Easton, D. 1985. 'Political Science in the United States. Past and Present', International Political
Science Review 6, 133-152.

Friedrich, C.J. 1963. Man and His Government. New York: McGraw-Hill Book Company.

March, J.G. & Olsen, J.P. 1984. 'The New Institutionalism: Organizational Factors in Political Life",
The American Political Science Review 78, 734-49.

Myrdal, G. 1958. Value in Social Theory. London: Routledge & Kegan Paul.

Rasmussen, E. 1970. 'Some Comments on the Concept of the Political System', Scandinavian
Political Studies 5, 11-19.

Yolton, J.W. 1977. The Locke Reader. Cambridge: Cambridge University Press.