The task of political inquiry is to raise the level of political debate. In part, this is accomplished by describing, explaining and assessing power relationships and lines of action; in part by directing attention to similarities and differences in types of government and public policy. In this way, discussion is enriched by arguments which are grounded in systematic and empirical studies of political life, and less room is left for wishful thinking and downright speculation.

The question of how research in political science is able to influence political debate is one which I will attempt to answer in the following pages. In order to do this, I limit the scope of my concerns to three truly central ideas.

Resumé

Discussion about political research could be seen as dealing with three basic questions. The first being whether type of government or type of public policy should be considered as our main object. The second being whether constitutional law or the more realistic approach of political science should dominate. And the third being whether we should be arguing in or about politics. If we separate these three questions it is much more likely that discussion will improve.

1. Type of Government or Type of Public Policy?

The first of these fundamental ideas is the distinction which could be made
between type of government and type of public policy. For the most part,
political debate oscillates between these two concerns.

To the extent that one focuses upon type of government, the discussion

centers upon procedure and power relationships. Who has the right to
make decisions nationally? Who has the right in local government? Who
actually makes such decisions?

On the other hand, if one is more interested in type of public policy, one is primarily interested in the goals and methods for creating and distributing welfare services, as well as in who assumes formal responsibility and costs for the various programs. National and local governmental authority are used to achieve and apportion that which citizens want. What should we as citizens vote to allow ourselves? How should we vote collectively so that rights and duties are created and assigned within industry, housing, education, social and health care in a way which will help us to achieve our goals? How should we vote to distribute the economic sacrifices that must be made over time and among different social groups in order to support our various policies of welfare and decentralization?

A democratic type of government is not irrevocably bound to any one of the three ideologies inherited from the 19th century. These outline how welfare, responsibility and costs are to be created and apportioned. Conservatives, liberals and socialists can all support a democratic type of government with equal fervor and at the same time harbor different ideas about the nature of the particular policies which should be employed. These policies would thus be identified as socialist, liberal or conservative. This would be the case even under a dictatorship.

In reality, the form of government is an important precondition for the kind of public policies which are implemented. This relationship operates the same way in reverse. Policies influence the form of government. Therefore it is only natural that both of these problems are central to political debate. In fact, it is exactly for this reason that it is fruitful to work with two different concepts. There are political as well as scholarly reasons to support this.

As far as politics is concerned, we undermine the importance of a community of values around procedures if we count public policy in as a subordinate part of the type of government and vice versa. The danger is that we would not make a distinction between issues and methods. Opposition on issues would be incorrectly interpreted as mere differences of opinion about who has the right of decision for the nation as well as for the local community.

The scholarly reasons for making a clear distinction between type of government and public policy stem from the concern to reveal the real mechanisms behind the mutual influence of type of government and public policy upon each other. The most minimal attempt at defining such relationships,

i.e. one which attributes as little content as possible to these concepts and thereby leaves as much as possible in the form of hypotheses which should be proven, makes it possible to determine empirically in which way and to what extent the form of government and the public policies actually influence each other. With this approach, we do not limit our concerns to matters of definition and political valuation.

If we consciously separate the political reasons from the scholarly ones for making a distinction between type of government and type of public policy, we can avoid an unnecessary discussion of the political value that lies in how democracy is perceived as a superordinate ideology. In this way, those researchers who are not democrats in this sense can more easily speak of the distinction between type of government and type of public policy and use this as an aid in posing and solving political science research problems without experiencing psychological blocks.

The usefulness of this distinction becomes apparent when it is applied to the primary points of opposition between liberal and socialist standpoints in the current debate. Liberals claim that democracy will founder if we continue along the path of ever-increasing rights of co-determination and employees' co-ownership in the companies for which they work. The right of the individual both to possessions (whether accumulated through inheritance, work or business) and to the freedom of their disposition, is the essence of a democratic form of government. The socialist would say it is just the opposite. It is only if the ideal of democracy is applied in the widest social meaning that democracy has a chance of being realized and of prevailing in our political institutions.

Both sides in this trial of strength could formulate their respective standpoints and arguments far more clearly and forcefully if they were not so bound to nineteenth century ideologies. Despite whatever differences there may be between these two groups, they nevertheless have one thing in common. There is no fundamental ideology yet which has made a clear distinction between type of government and type of public policy.

In the opinion of political scientists, type of government can only be one thing. It is a concept which can and should be either a democracy or a dictatorship, regardless of whether the policies it practices are liberal or socialist. In other words, the matter of definition here is not an empirical question. Instead, it is a question of how we choose to delimit these two key concepts in relation to each other.

It is quite another issue to investigate empirically how government and
public policy influence each other, according to a liberal or a socialist
perspective. Does democracy gain by public policies which hold back

tendencies toward an equalization and a leveling out of welfare, responsibility and the shouldering of costs in society? Or is the case just the opposite, e.g. that an approach which takes into consideration the difference between social groups and which allows for a more differentiated distribution of welfare services, responsibility and costs according to need and ability is the precondition for and the consequence of a particular form of government, both on the national and the local level?

By maintaining a strict distinction between type of government and type of public policy, research in political science attempts to avoid having the answer to such questions take the form of delimitations and political desires. We construct our scholarly investigations to be as empirically oriented as possible. Comparative studies of the mutual influence of type of government and type of public policy upon each other are the kind of research endeavours which will be most fruitful in the long term.

2. Constitutional Law or Political Science?

The second of the three fundamental ideas which guide political science
research is a belief in the value of making a distinction between constitutional
law and political science.

'Government has two faces', according to an idea which has essentially been accepted since the end of the nineteenth century. The one side is turned toward the sociological and the historical, the other toward legal considerations. One issue is how types of government and types of public policy in different areas are constructed legally. To put it simply, it is then a question of which rules apply. Concentration is focused upon the content and significance of regulations which citizens and authorities are required to obey, as well as which regulations they have the right to invoke. Quite another issue is that which goes on informally, behind the scenes, but within the framework of the rules which formally and legally establish what political actors - citizens, businesses, political parties, organizations and public authorities - have the right to do. The question in the latter case would then be what type of government and what kind of public policies historians and social scientists can distinguish when they look at what really happens.

Many a penetrating investigation has been inspired by the question of whether it is both possible and appropriate to conceptualize political life within the framework of two different perceptions of government, the so-called two-sided theory (Jansson 1950). On the one hand it has been said that it is principally impossible for two completely different theories

of government to be true and correct at the same time. On the other hand, it has been argued that government must nearly always be viewed from one or the other of these two perspectives. At the current time it is commonly felt that there is greater cause to support the second of the two viewpoints above; that is, that it is plausible that two different conceptions of government do apply. But of course these are only conceptualizations. They are not things which exist in a physical sense. They are man-made constructions which are used to further our understanding of the various relationships which actually do exist. And it should be added that using a formal, legal conceptualization does not exclude the possibility of also choosing other approaches which might be more informal and realistic. Political science and constitutional law are not two competing hypotheses, of which the one can be proven better than the other in how it more accurately represents the facts. What we have here are two separate issues, the one dealing with legal aspects of political life and the other with more social aspects. Both are worth considering when looking at types of government and types of public policies.

This distinction between constitutional law and political science is even effective when used in social debate. The lack of clarity of thought and expression in this regard has been especially striking during recent years' debate about rights of co-determination and the rights of employees regarding co-ownership in the companies and businesses in which they work. The liberal point of view is that the trade unions want to give their members two votes. The socialist argues that the position of the union needs to be strengthened in order for them to carry their weight against the capitalists. Both liberals and socialists are prone to glide toward or away from constitutional law or political science, depending upon which of the two fits their party's position at the time.

In a formal, constitutional sense, one must always take the legal point of
view, regardless of whether or not corporative elements exist or might be
considered for the type of government in question.

The political science side of the issue is quite another thing. Realistically speaking, Sweden is not a democracy. In this respect, the socialist would be right. Citizens are not all in the same position when it comes to formulating their desires, to putting them forward and seeing them given equal weight in practice. 'The democratic breakthrough' is an expression from constitutional law.

From a political science perspective, the question is whether a stronger
position for the trade unions actually could redirect our type of government
away from or towards the democratic ideal. Will political democracy

be better or worse on the whole if employees are compensated for by
means of their organizations obtaining a stronger position in the world of
work?

One cannot answer that question by studying the various regulations which are currently in effect. In order to come up with a good answer, one would have to investigate how things really are: the actualities, not the formalities; how things are socially, not just legally; and what they are like from a political science point of view, not only from the perspective of constitutional law.

The distinction between constitutional law and political science should not be confused with the distinction between type of government and type of public policy which we analyzed earlier. This often happens when one, in a thoughtless and vague way, pits 'form' against 'content'. In principle, we are dealing with four possible positions here, accordingly:

(1) type of government and constitutional law
(2) public policy and constitutional law
(3) type of government and political science
(4) public policy and political science

For a long time, the conflict was mainly between (1) and (3) in this scheme. Then in the 1920'5, position (3) began to prevail, and this remained true to the present time. Positions (1) and (2) have been taken over by the researchers in our law faculties.

In Sweden, position (4) has long existed as a parallel, oppositional line. 'In these times of more expansive business activity, even the government has clothed itself in the garb of the businessman', proclaimed Rudolf Kjellén in the lecture he gave upon his installation as professor in 1916. 'The picture of the government which ... we see in reality is quite different from the old liberal ideal of a government which restricted itself to upholding the law while individuals took care of progress. Our modern government is itself a progressive power, the largest of all beyond comparison. The irrefutable conclusion is immediately apparent: political science must prepare a place for the social power of government alongside its capacity for legal power' (Kjellén 1916, 216 f.).

There are several reasons why position (4) above has had a difficult time in being accepted. The primary difficulty during the inter-war period must surely have been largely due to the fact that this new position was championed by Kjellén. Politically, it had wound up on the 'wrong' side of the conflict between democracy and authoritarianism. It is much harder to

find a suitable explanation for why Carl Arvid Hessler was met with such a cold reception when he appealed for the study of public policies during the 1940'5, 1950's and 1960'5. It is likely that this was due to the fact that the number of political scientists was so small until well into the 1970'5; and that position (3) continually played such a large role in the debate on democracy - guarding against Nazism during the war years, against the Cold War during the 1950's and appearing in the discussions about representative democracy and the lack of real political equality during the 1960'5.

However, one important thing should not necessarily exclude another of equal importance. That was the point which Hessler made in a paper he wrote in 1941 in which he attacked what he saw to be a shortcoming that dominated Swedish political science research during the inter-war period. He saw the ambition at that time to portray political life as realistically as possible as simply a movement away from constitutional law, toward political science. He maintained that it was every bit as important to take a step further and make a distinction between type of government and type of public policy.

'One experiences an attempt to be "realistic" in the political science of our time', Hessler wrote. 'The quest is for greater clarity regarding the forces which fulfill constitutional formulations and assume the task of governmental development... But this desire to reach insight into the various powers which lie behind governmental life gives birth to another one. Power gives rise to action on the part of the government. The government diminishes or increases its activities. Thus one feels a desire to increase one's knowledge about the state as an active and influential factor within social and cultural life. There is talk about a liberal state and about a cultural state. In more than one area we find that we have only very vague notions about what such terms really mean. But in order to obtain a totally realistic picture of government one should wish to have the clearest possible idea about what government has brought about in certain areas during certain times' (Hessler 1941, 49).

Thirty years later, in the beginning of the 1970'5, the same thought emerges in the US as something entirely new. 'Policy analysis', 'technology assessment' and 'evaluative research' took the lead then in a long-awaited and well-prepared restructuring of the subject in a manner which attached equal weight to the positions (3) and (4) above.

3. Are We Arguing 'in' or 'about' Politics?

With this said, I have arrived at the third and last of the fundamental ideas

which I have chosen to analyze.

In addition to the need for political science to investigate public policy, there is the need to differentiate between assertions which are made 'in' the political debate and those which are made 'about' the actions and statements of policians. Naturally, one finds the same need within studies of types of government, but experience shows that this need is more apparent within studies of public policy.

Why is it so important to make such a distinction between actor-problems
and research problems?

In part, it is a matter of how the field of political science, in contrast to many other disciplines, also includes actors which are prone to describe and give reasons for their own actions. Our duly elected officials and their experts often express viewpoints which are at least equally insightful as any that political scientists could make. Political actors also use the same sort of terminology which characterizes the academic research that is directed towards their activities.

Investigators of other social phenomena do not experience the same difficulties. Other scientists do not have to contend phenomena which are at the same time living beings which offer their own accounts for their actions. On the other hand, the person who studies public policies or types of government must always remember that assertions which appear to be worded the same can actually have been produced by two principally different sources in the field of analysis. Statements of the type 'The Soviet Union is the government of all the people' and 'Sweden is for a social housing policy' are both, in a sense, statements of solidarity which characterize the political culture in both of these countries. But these statements can also be interpreted as assertions of actual relationships, given the issue has been researched in line with explicitly established criteria (Brecht 1959).

The case would be different if it could be assumed that those who are politically active have mistaken conceptions about the context of which they are a part. Their views could then be disregarded once and for all. But this is obviously not the situation.lt is rather more likely that actors are occasionally right, occasionally wrong. Therefore it is crucial that political science research - whether it focuses upon forms of government or upon public policy - includes the delicate tasks of making analytical distinctions between these two levels of statements and of proving the validity of such opinions and arguments as political actors might profess and put into action.

Unfortunately, the practical difficulties one encounters in such an

analysis are not made much better when so many political scientists write
in such a way that their research is understood by the general public.

But in principle this should not be a stumbling block, as long as we
maintain a clear distinction between problems related to actors and problems
related to research.

There are primarily four problems related to actors: choice of goal,
measurement against goals, choice of means for obtaining goals, evaluation
of the effectiveness of the efficiency of the chosen means.

The approach which is taken in relation to each one above differs according to which of the actor-problems is to be considered. When it is a decisionmaker who chooses a goal, then the question is one of criteria. When it is a case of measuring against goals and/or assessing the effectiveness and efficiency of the means towards obtaining the goals, then one designs a hypothesis. The problem of choosing means for obtaining goals involves neither criteria nor hypotheses, but rather recommendations for action.

If we consider the kinds of arguments which can and should be offered for or against each one of the four actor-problems and their related issues above, then we see that these vary from descriptions of reality to value judgments. It is only when a politician chooses a goal that he may use both types of argments. In all other cases, only descriptions of reality can be counted. In other words, opportunities for arguing directly in question of forms of government and public policy are much more plentiful than what one might expect.

The problems of research and actor-problems can be one and the same

thing. But they do not necessarily have to be.

As researchers, we may choose to describe and take a stand concerning the same problems that our elected officials in national and local government deal with. As Herbert Tingsten would say, we would then engage in a criticism of ideas (Tingsten 1941, 1 ff.). We investigate what politicians do and say and try to ascertain the meaning behind it, and if things are actually what they are purported to be.

But we might also choose to describe and explain the positions that the politicians take as well as the arguments they use to back up their views. In this case, we do not try to investigate whether or not things are actually as the politicians say they are. Instead, we try to relate their positions and arguments to possible underlying motives and relationships. The problem that the researcher formulates for himself is then something quite different from the problems that are of primary interest for the politicians. What the researcher has to investigate are what kinds of considerations and circumstances actually determine what the politicians do, and not whether or not they are right in doing so.

It is only during recent years that the difference between actor-problems and the problems of research has been generally understood and considered in practice. Until rather recently, it was a harshly debated issue as to how one could determine the meaning, relevance and tenability of political statements in a scholarly way. In the Swedish debate, it has only been since Ingemar Hedenius made the distinction between 'authentic' and 'non-authentic' legal statements in the beginning of the 1940's that the problem has become easier to deal with (Hedenius 1941, 60 ff.).

According to our current understanding, one has the same formulation, modelled on the statement 'this is right', interpretable in two different ways. It could indicate one's perception of a particular position that should be taken concerning one of the actor-problems outlined above. In this case, we would be dealing with a statement which would be 'genuine' according to Hedenius. Actor-problems and research problems would be in agreement. This would be a study 'in' politics.

On the other hand, the statement 'this is right' could mean that a certain position has actually been taken by someone. 'Actor X considers that this is right' would then be the meaning conveyed. Such a 'non-authentic' statement as the latter is always theoretical in the sense that one can determine in which way it is right or not without too much difficulty, even if it would entail a goal or a value judgment supporting a particular goal. The research problem here centers around the question of which position the politician has actually taken and which arguments he or she maintains.

It is not a question of whether or not a particular position or argument
should be supported. That which needs to be determined here is a question
'about' politics.

Keeping this in mind, one may consider that the old question as to whether or not political science should be a science 'in' or 'about' politics has received a new answer. The view nowadays is that political science should encompass both approaches. This is just fine as long as we remember to make the distinction between actor-problems and research problems; as then we can clearly see whether or not we are arguing on the 'in' or the 'about' side of a problem, a viewpoint and/or argument that appears to be important for politicians themselves.

At the same time, it is clear that investigations 'in' the actual debate and 'in' the action that is taken, Jemand something of the research in political science that is similar to what natural scientists, agronomists, technicians and people in the medical profession practice. Even we political scientists must learn to work in a more effective and disciplined manner with representatives from other disciplines: first and foremost with economists, statisticians, sociologists, geographers and lawyers. Studies must be organized better into the so-called 'sectorial research', as the Swedish expression goes.

However, this is not to say that research in political science should renounce any of the fundamental ideas upon which it rests: the distinction between type of government and type of public policy, the distinction between a formal and a 'real' way of viewing political phenomena, or the distinction between actor-problems and research problems. On the contrary, cooperation is meaningful only to the extent that one has something to contribute oneself.

REFERENCES

Brecht, A. 1959. Political Theory: Foundations of Twentieth-Century Political Thought.
Princeton, N. J.: Princeton University Press.

Hedenius, I. 1941. Om rätt och moral (On Rights and Morals). Stockholm: Tiden.

Hessler, C. A. 1941. 'Debatt kring en lönestat' (Debate over a Budget Entry) in Festskrift
för Axel Brusewitz. Skrifter utgivna av Statsvetenskapliga föreningen i Uppsala, 12,49-60.

Jansson, J-M. 1950. Hans Kelsens statsteori mot bakgrunden av hans rättsfilosofiska åskådning (Hans Kelsens's Theory of Government Viewed against His Views on Legal Philosophy). Societas scientiarum fennica. Commentationes humanarum litterarum. XV. 5.

Kjellén, R. 1916. 'Statskunskapens objekt' (The Objective of Political Science).
Statsvetenskaplig tidskrift 19, 215-43.

Tingsten, H. 1941. Idékritik (Criticism of Ideas). Stockholm: Bonniers.