I have frequently remarked to those who wondered at what they termed the mysteries and complications of politics that public government is much like private government. If you wish to understand the political, do not try too hard. Look around you at the associations of which you are a member – how many, you know better than I.Footnote 1
Ever since political science became recognized as a separate discipline in the course of the twentieth century, two interconnected questions have affected its development. The first question is about what is politics; the second concerns the search for a comprehensive theory able to account dynamically for all political activities. Neither of these questions has so far been solved to the satisfaction of all concerned, the second proving even more difficult to handle than the first.
The matter of what is politics has been complicated from the start by the problem of the linkage of politics with the state and typically with the state alone. Although laymen, in ordinary language, probably refer to much of what goes on in organizations below the state as ‘political’, political scientists have been reluctant to go that far. One of those who tried hardest to define politics, in the middle of the twentieth century, David Easton, found it impossible in the end to separate politics altogether from the state. He accepted that politics was an activity and not what could be described as an institutional characteristic of the state; but he then stated that this activity takes place in society at large, or in what he called the ‘political system’ which is not truly different from the state.Footnote 2 Thus, although this is not stated, what seems implied is that, if there is politics in a party, trade union or university, for instance, this must be because these organizations are in some sense part of society. Unlike economists who, from the start, saw the activities which they studied as taking place by means of any exchange of goods or services among two or more individuals, political scientists have been reluctant, on grounds which remain unclear, to do what Merriam suggested in 1944, namely to see that politics exists, in the same manner, in any organization, indeed in any body composed of at least two individuals.
Yet, coincidentally, and somewhat unrealistically given the lack of an agreed answer to the previous question, there has been a search for a general theory which would help to explain how politics operates. Here, too, the image of economics plays a major part: political scientists have in reality been rather jealous of economists in this respect, since these have been able to discover a law, that of supply and demand, which has been deemed to account generally for the phenomena taking place in their discipline, even if the application of that basic law of economics has been found to have many warts. Warts or no warts, there is a law of some kind in economics; that law appears to account for many of the developments which take place, especially at the micro level. Political scientists have never been as lucky; they have simply no law at all, flawed or not flawed.
It will be argued at this point, at least by the political scientists of the rational choice persuasion, that it is no longer true that there is no general law but that one has now been found. That law is, to quote George Tsebelis, that ‘[a]long with the mainstream of contemporary political science, I assume that human activity is goal oriented and instrumental and that individual and institutional actors try to maximize their goal achievement. I call this fundamental assumption the rationality assumption’ (emphasis in original).Footnote 3
The trouble with such a law, if it is one, is that the question of proving or disproving it does not arise because it is an assumption; moreover, that law does not cover politics only but refers to all human activity. As political scientists have not so far claimed that all human activity is political, it seems difficult to use the notion of rationality to determine whether some matters are political.Footnote 4
Meanwhile, the definition of politics does not seem to progress markedly either. The New Handbook of Political Science, edited by R. E. Goodin and H. D. Klingemann and published in 1996, adopted the notion that politics has to do with ‘the constrained use of social power’,Footnote 5 except that this formula is preceded by the expression that politics might best be characterized in that way. This is obviously a way of saying that the formula remains rather vague and uncertain. In this respect, the formula presented in The Oxford Handbook of Political Science, edited by R. Goodin and published in 2009,Footnote 6 is on the same wavelength but, perhaps less judiciously, the limitation of the 1996 version is not added. As a matter of fact, it is surprising that the notion of ‘power’, which had been associated with politics primarily by Lasswell in the middle of the century, had not been abandoned as part of the definition,Footnote 7 although the authors claim that they distinguish themselves from Lasswell in this respect.Footnote 8
Why power should be adopted as the criterion is also surprising since the point seems to suggest that there cannot be any form of consensual politics or politics by amicable agreement, as has been suggested for Switzerland by J. Steiner,Footnote 9 or by E. Ostrom with respect to ‘governing the commons’ in California and elsewhere.Footnote 10 Indeed, David Easton was able to take the matter of consensus into account more than those who consider power to be fundamental, in the 1953 volume, when he referred to politics as being the means by which there was an authoritative allocation of values, that is to say a process of decision-making which applies collectively to a group; the only problem with that definition remains, as was noted earlier, that Easton restricted its application to what he called the ‘political system’, that is to say to society at large and in effect to the state. If, on the contrary, a more realistic definition of politics is to be found, it has to be developed around the same process of decision-making in groups and in the state and indeed it has to be applied both when those concerned are all in agreement and when they are not, some form of power having manifestly to play a part in the latter case.
The two matters which have just been discussed constitute jointly what might be termed the basic problem or even complex of political science as a discipline, namely that its scope has not been defined precisely and that the search for a general theory has been a never-ending unsuccessful endeavour. The object of this article is not to attempt to provide definite answers to what have been proved so far to be intractable problems, but to suggest that there is no chance of finding either a truly satisfactory definition or a general theory in the field of politics so long as the state is viewed as the object or even the principal object: since politics is now universally recognized in the discipline as being an activity, that activity cannot be defined by means of referring to an institution. Politics must therefore be defined in terms of a particular type of activity: merely as a suggestion and to circumvent the problem of the state, one might go as far as saying that what characterizes politics has to do with the process by which decisions applicable to a group are brought to a conclusion, whatever that group may be, whether it is the state or whether it is not.
Such a definition does also help with respect to the question of the discovery of a general theory, indirectly at least, in that it is manifestly highly improbable to think that one can find a general theory by concentrating one's attention on the most complex organization in existence, namely the state. Economists did not discover the law of supply and demand by concentrating on monopolistic or near-monopolistic multi-national companies, but by looking at the many little markets. The suggestion here is that, by examining relatively simple groups, one is markedly more likely to find the basis on which a general theory of politics can be found, if such a theory exists. If it is found, such a theory might then be tested at the level of larger organizations and perhaps even of the state.
Before returning in the last section to the simple groups which Merriam and later Lakoff called ‘private government’, however, two preliminary steps must be taken. First, we must explore somewhat more the question of the nature of politics by examining the relationship between individuals and the bodies in which these individuals operate: rational choice analyses have rightly insisted on the role of actors, but they have remained faced with a major difficulty in accounting for the existence of institutions in what seems to be a Sisyphus-like effort to attempt to see actors as the source of all political activity: this is for instance the case when an attempt is made to claim that institutions are endogeneous to the actors of these institutions, as is claimed in some forms of rational choice institutionalisms discussed by K. A. Shepsle in Chapter 2 of the 2008 Oxford Handbook of Political Institutions.Footnote 11
Second, we have to consider the inherently partial nature of what is conventionally known as theory in comparative government, that is to say of studies undertaken at the level of the state. The structural underpinnings of the state are so complex that the dilemma which specialists have to face in that relevant branch of the discipline, comparative government, is thus either to be rather vague in their generalizations or, if these specialists are to be precise, to concentrate on what can at best be described as middle-range theories. It will then become possible to return to the study of private governments in the third section and examine some of the problems which such a study must resolve if it is to help political science to advance in the direction of a general theory.
THREE KEY POLITICAL MATTERS IN NEED OF INVESTIGATION: COLLECTIVE ACTION, ACTORS AND NON-ACTORS, AND PRIVATE GOVERNMENTS
In his volume on Nested Games, published in 1990, already mentioned and perhaps his most important work, Tsebelis makes three distinct points at the outset. The first two are straightforward in a rational choice context. One of these is that actors are important, unlike in what the author refers to as holistic explanations of political life in which one would ‘disregard individual actions and argue that such issues are not important’.Footnote 12 The second point is that these actors are rational: this is said to be, as stated above, that ‘human activity is goal oriented and instrumental and that individual and institutional actors try to maximize their goal achievement’.Footnote 13 These two points are clear, although the second may be regarded by many as intrinsically unhelpful, as an inextricable problem is inevitably faced by anyone who claims that actors – any actors – try to maximize their goals: at the limit, every action of any individual becomes rational and the use of the adjective does not help to distinguish some acts from others.
Politics as Collective Action
Meanwhile, however, the third point constitutes a major restriction to the scope of the investigation. It is that ‘I treat institutional design less exhaustively than games in multiple arenas because institutional change by definition involves political innovation, and it is difficult (if not impossible) to know its rules, much less to have a complete theory about them’.Footnote 14 The author then adds that ‘RikerFootnote 15 considers the development of political innovation an art as opposed to a science … and argues that its laws are unknowable. Whether the laws of institutional design are unknowable or simply unknown, the issue of institutional design is too important to be left out of a book adopting a rational choice methodology. However, the current state of knowledge of institutions justifies the absence of theoretic rigor’.Footnote 16
The aim here is not to labour the point that rational choice analyses have difficulty in accounting for the development of institutions, groups, organizations or bodies: this difficulty was indeed mentioned in the Shepsle article on ‘Rational Choice Institutionalism’.Footnote 17 The key point is to see what relative place is to be occupied by such organizations, on the one hand, and by actors on the other. It is true that actors are very important and that they have typically not been given the importance which they deserved, in large part because much of general political analysis has tended to be, so to speak, over-structural. Yet it is also true that actors do in some sense inherit institutions, structures, organizations, groups and that this inheritance has to be recognized both because of the fact that these bodies have often been in existence long before current actors have been involved in them and because these bodies are collective and specifically impose their decisions, in many cases, well beyond those who are actors within them: such a point is true, not just of the state, but of all organizations, except the most ephemeral. This is why one can say that, while economics is about transactions, politics is about coalition-building in order to achieve what Easton referred to as ‘authoritative allocations of values’.
The fact that politics has to do with collective action is undeniable: some rational choice analyses at least appear in this respect to misunderstand the nature of political action. Moreover, by stressing the role of actors, such analyses tend very often, if not always, to stress competition rather than cooperation and in effect to seem occasionally to deny the fact that political actors have to act jointly (possibly against other actors, also acting jointly) in order to achieve some goals. The emphasis in rational choice analysis is on individual demands and pressures on the part of actors who know what they want for themselves: hence the part played for instance by bargaining operations during which individual actors present their demands and enter into negotiation with others. These operations are important but they do not take place necessarily, nor perhaps even frequently, in a climate of competition and tension. What seems occasionally to be an obsession to draw a parallel with economics may render those who adopt the rational choice approach blind to the fact that very large numbers of political actions, not merely among lower-level groups but at the level of the state as well, are based on cooperation, if not among all the possible actors at stake, at any rate among many of them.Footnote 18
Meanwhile, moreover, the political process does not stop when the decision is taken, consensually or otherwise: the political process continues and may continue for a very long time. There may indeed be conflicts after the decision has been taken, but that decision remains legally in force. The point, therefore, is that politics is a fundamentally collective activity and one which extends – or may extend – well beyond the tenure of those who participated in the decision itself. Thus institutions, organizations or groups are not peripheral elements in the political landscape: they are as much the stuff of politics as actors tend to be.
One has therefore to return to being holistic! As a first approximation, one may say that groups and organizations play a key part in politics because of the weight of history: at any rate in many countries, the political life of the state is shaped by structures which may have taken years, decades, even centuries to develop, the structures of the state being typically referred to as institutions.
Yet there is much more to these structures than just to be the consequence of the ‘weight of history’: this is so because political decisions are taken and can be meaningfully taken only if a body or structure exists within which these decisions will be able to apply. Structures have therefore to precede chronologically the ways in which actors come to take decisions.
There is a further aspect, which relates this time to the future: bodies or structures are indeed the mechanisms by which decisions taken by some (the actors) are applicable (in the future), not merely to the actors, but to what can be a very large number of non-actors. Politics is not a terrain in which decisions are taken in and for the instant: it is an activity which takes place within a body and for that body.
The body or structure to which the decisions apply has thus both to exist before these decisions are taken and to continue to exist afterwards: if it does not pre-exist, it cannot take any decisions unless and until it has first been decided to set up the body; if it does not continue to exist, a decision must have been taken to dissolve the body. Thus the actors in the decision cannot explain by themselves why the body exists; nor can they account for the reasons why the decision is also applicable to non-actors belonging to the body.Footnote 19 In terms of social science disciplines associated to politics, while politics depends on psychological attributes since actors are involved, it depends also (even if only in part) on such structural disciplines as sociology, law and history.
The Key Role of Actors and the Presence of Non-actors
Politics is thus a collective activity in which the decisions taken impinge on the members of the groups in which these decisions are taken. These decisions are taken by actors, and rational choice analyses have been entirely correct to stress the part played by actors in the process. Yet, precisely because organizations and groups are collective, actors in the strict sense of the word are not the only persons likely to be involved in the decisions taken. If actors are essential features of politics, non-actors must also be considered. Actors are those who have participated in the decisions taken; non-actors are not involved then, but they may well be involved in the consequences of these decisions. Being an actor means both that one is personally engaged in the decision which is being taken and that one believes one can have an effect on the decision to be arrived at.Footnote 20 One is not an actor if one of these two characteristics is missing, for instance when one argues about local, national or international politics (even if one is, for example, a voter), as is well summarized by the old French expression which refers to such exchanges as discussions ‘de café du commerce’.Footnote 21 Given that the decision taken by the actors applies to the non-actors by virtue of the fact that political activity takes place within a (pre-existing) body, non-actors (who may be called spectators) are often likely to be a large segment of the population concerned.
Politics in Private Governments: Micro-politics
Meanwhile, the stress placed on actors in rational choice analysis has indirectly had the effect of reducing the importance traditionally given to the state in the literature and to extol the part played by all (other) types of groups: it is as if there was no reason to privilege the state in political processes when the emphasis is on actors. As in the rational choice perspective there is no incentive, but on the contrary a considerable disincentive, to stress the role of institutions and of state institutions in particular, the emphasis placed on actors means that one can examine the role of these actors in any institutional context. As a matter of fact, examples provided in rational choice analyses are often drawn from within an environment which is not specified, as is the case for instance with such problems as the prisoner's and similar dilemmas. Without saying so, but because so little, if anything, is said about the contextual institutions in which the actors operate, it seems that, maybe by accident, rational choice has opened the way towards private government to an extent that perhaps no other approach previously did.
As a matter of fact, the private governments mentioned by Merriam and more recently by Lakoff might be better described as being micro-political, along the lines of the distinction drawn by economists between micro- and macro-economics. Admittedly, the expression micro-politics tends to be used with a different connotation, namely by reference to the behaviour of electors in national (or local) state elections: but these citizens are typically not actors.Footnote 22
The reference to micro-politics is neither esoteric nor original. It provides a useful distinction among the bodies within which political activity takes place: the complex role of the state is thus contrasted to the more limited range and forms of decision-making which tend to be found in lower-level organizations. As was noted in the introduction, ordinary language supports the point of view that politics occurs in a large number of bodies, whether these are associations, companies, trade unions, universities, churches or families.
It could of course be argued that only the state has the legal power to enforce its decisions authoritatively and that other bodies can merely suggest how members should act. This distinction may be legally correct but, in practice, there are marked differences, on the one hand, in the extent to which the state is able to enforce its decisions, and, on the other, in the extent to which other bodies experience difficulties in exercising pressure on members.Footnote 23 The three elements of Exit, Voice and Loyalty analysed by A. O. Hirschman,Footnote 24 apply to a variable extent to all types of organizations: the result is a continuous dimension on which the state is far from being always the body which enforces its decisions most easily and most commonly.
If there is politics in all these bodies, many of the persons concerned are actors, but many of those who are members are not. On the understanding that actors have to feel strongly affected by what the body is about to decide and have to believe that they can do something about the possible result, the proportion of actors is likely to be smaller in large bodies (including the state itself) than in smaller ones, although large bodies are typically sub-divided into semi-independent units: some are actors within one or more of these units without being actors at the overall level. In small bodies, on the contrary, the proportion of actors is likely to be large: at the limit, if the group is composed of two persons only, as is the nuclear family, both members are actors. Moreover, while, for very many, national (or even local) politics is distant and the problems under discussion may not be clear, the politics of the smaller groups to which these persons belong can count markedly, whether in the family, at work or in the context of leisure activities. Often at least, these persons are likely to be directly involved and thus be actors.
Given that political activity does appear to extend widely beyond the state, a key question which arises is why the discipline of political science has not felt willing – indeed obliged – to become involved in the analysis of the political processes taking place in these private bodies, while economics has felt more than willing to examine, indeed began its development by examining, the activities of individuals outside the state. The rather irrational consequence of that situation is that, by operating at the level of the state almost exclusively, political scientists complicated their task enormously, especially in terms of their ability to build a general empirical theory of the behaviour of political actors.
THE DILEMMA OF COMPARATIVE GOVERNMENT: UNREALISTIC GRAND MODELS OR LIMITED MIDDLE-RANGE THEORY?
Since political science came to be of age, broadly speaking, after the Second World War, comparative government (a more correct expression than ‘comparative politics’) has been the apex of empirical political analysis. Country studies have been widely regarded as limiting, although some specialists, more so in the past than currently, have felt that it was impossible to compare two countries or even the ‘same’ institution in two countries, precisely because of the structural differences arising from the historical past of these countries. Those objections are strictly speaking correct, but they go against the grain in that we all want to find out, would it be for practical purposes, what the experience of at least neighbouring countries (whatever this means) tends to be. In practice, no one is wholly a purist in this respect, given that, at the limit, one cannot even describe the institutions of a single country without referring to some conceptual vision of what such institutions might be in the abstract, whether one is concerned with governments, legislatures or parties. The very use of these words means that one is prepared to recognize that such entities are worth discussing in general.
In view of these difficulties about what its geographical scope should be, comparative government has naturally developed at many levels, ranging from two or a few countries, noted for the similarity of their structures, to broader presentations. Yet, despite some brave efforts undertaken in particular in the 1960s, and above all those of G. A. Almond,Footnote 25 these broader presentations did not succeed in producing realistic ‘grand models’ able to encompass all types of political systems, except at very high levels of generality. In the end, the only purpose of these models has been to make it plain that one could not really, except in a limited manner, analyse political life and political institutions on the basis of a single framework for the whole world. Thus, half a century after the idea of attempting to build truly universal comparative government was put forward, a number of broad geographical boundaries continue to play a major part, as if there was (and there may well be) some relationship between these regional divisions and a number of key political or socio-political (structural) characteristics of the countries concerned. Comparative government studies tend therefore to concentrate on Western countries or on Latin American countries or on African countries or indeed even, twenty years after the end of the Soviet Union, on post-Soviet studies.
This is to say that the tendency to undertake research at the middle-range level, as was already suggested by La Palombara at the end of the 1960s, is universally adopted, de facto, even if not necessarily in theory, by students of comparative government.Footnote 26 This is so for what might be termed the ‘classicists’, whose approach may not be so much descriptive as inductive–deductive, but also for those who adopt a rational choice approach and view themselves as more theoretical – at any rate to the extent that empirical analyses are developed on the basis of a pre-elaborated theory. This is the case specifically of the analyses undertaken by Tsebelis, not just in Nested Games but in his subsequent work, for instance on Veto Players.Footnote 27
The question which naturally arises is whether this situation is temporary or whether it is, deeper down, due to the inevitably complex, perhaps hyper-complex structural characteristics of the state. One could take what might be regarded as a reformist view of the future of comparative government: one might for instance claim that, indeed, progress has taken place to such an extent in the course of those 50 years during which political science has moved from what was truly its infancy to becoming a major social science, that further developments in the course of the next 50 years will make it possible to overcome the middle-range level, typically based on geographical divisions, in which it has remained up to now. It might indeed be argued that the rational choice approach is precisely the best instrument at the disposal of the discipline to overcome its middle-range complex, since it is based on what can be regarded as an optimistic view of the future rather than on the kind of pessimism in which the middle-range viewpoint is glued. Unlike the contributions on the subject in the New Handbook of Political Science of 1996, the chapter devoted to the ‘Overview of Comparative Politics’ by C. Boix and S. C. Stokes in the Oxford Handbook of Political Science of 2009 adopts a markedly more optimistic view (especially at pp. 563–4).
Two kinds of arguments can be put forward to support this optimistic approach. The first has to do with precision. Precision is not merely related to the fact that mathematical symbols are used widely and that these symbols can be used only if one is truly clear about what one is referring to. Thus veto players are not merely institutional structures which need to be taken into account in some circumstances (for instance on the party composition of the bodies concerned) but in all cases in which they can exercise their veto. This may reduce, at any rate ostensibly, the number of cases in which one is able to apply any adopted model, but, at least in these cases, there is precision.
The second argument has to do with simplification. A common criticism of classical comparative government is that it is overly complex and therefore, as a vehicle unable to get out of the muddy track in which it became stuck, comparative government analysis finds it impossible to progress. This simplicity may be acquired at a price, namely that relatively little can be achieved in one go; but this little achievement being a solid achievement as a result of the precision of the instruments adopted, the overall result can be regarded as progress.
It is of course impossible to know at what rate developments will take place during the coming decades. Yet there are serious problems about the nature of the findings which are part of the efforts made to move in the direction of systematic theory in the field of comparative government. Probably the most serious stems from the fact that the extent to which simplification does have to occur in specific cases is debatable – and at least must be debated. The problem seems then to be due principally to the fact that a sufficiently large number of cases have to be taken into account to render possible the mathematical treatment required to demonstrate the validity of the model. The consequence is that there is a tendency to lump in the same category situations which are in reality rather different. This is indeed an instance of the disease which seems to have always afflicted political science and has probably been inherited from legal analyses, namely that distinctions are too often viewed as dichotomous, perhaps at most as trichotomous, rather than as parts of continuous dimensions.
One clear example of such a difficulty arises in the field of governmental structures in the narrow sense, namely about what is meant when one describes the executive as presidential. Since the emergence of the French Fifth Republic in particular, it began to be felt that liberal (or at least moderately liberal) presidential systems might be divided into those which are fully presidential (on the US and Latin American models) and those which are semi-presidential. Yet difficulties arose about the determination of the characteristics of semi-presidential systems: further subdivisions were introduced, especially by M. S. Shugart and J. M. Carey in Presidents and Assemblies.Footnote 28 Even these distinctions turned out to be insufficient, however, for instance in order to categorize the post-Soviet presidencies which emerged in the 1990s, as well as a number of East-Central European and even Western European presidencies.Footnote 29
It is at this point that the complexity of even middle-range comparative government analyses clearly appears. For the problems posed by the classification of these regimes are almost insuperable. To begin with, the legal powers of the presidents vary appreciably. It is not only difficult, but truly impossible to elaborate distinctions which are both comprehensive and clear cut: continuous variables are much more realistic. Yet the most serious difficulty is different: it results from the fact that a stand has to be taken about the question of whether the powers given to these presidents in the constitution are used or not in practice. That problem is avoided if a particular power is never used and if this has been the case for long periods (as with the British monarchy with respect to the right to veto laws). One has then simply to note that a given power has in effect lapsed whatever the law or the constitution may state. Yet, in practice, the situation is rarely as clear cut: the power in question may in fact be used a little, in subtle ways or even under cover (for instance if a president points out to the prime minister that he or she might consider not signing a law or a regulation which is presented). As a matter of fact, this is so of presidents in parliamentary republics as of semi-presidential or parliamentary-presidential presidents.Footnote 30 Such a limited or undercover use of powers – indeed even of presidential interventions which are not strictly based on a power at all – is used more often than is stated in textbooks, almost certainly because not enough empirical work in this respect has been undertaken. This state of affairs renders the classification of presidencies not just difficult but, at the limit, arbitrary. Consequently, there is no hope of finding in this field a categorization which fits the characteristics of being both precise and simple. The variations are too many and too complex.
Such classificatory problems are in no way peculiar to the case of presidencies, however, even if one restricts the analysis to Western European presidencies. The problem is widespread and indeed arises with respect to every institution. It is endemic in comparative government because the relationship between law and reality is almost infinitely complex. It tends to be believed that a divorce between these two aspects of politics can be found primarily in authoritarian political systems, as an effort is undertaken by the leaders of these regimes to give the impression that their rule is liberal-democratic when it is not. Yet the distance between law and reality is also large in liberal-democratic systems in almost all aspects of the political system. Thus constitutions state that parliaments are sovereign, while they are in practice dominated to a different extent, in different ways and by means of different types of arrangements by the governments. Thus it is said that the cabinet is the collective decision-making organ of the government in parliamentary systems: yet it is clear that the prime minister, perhaps in combination with a very small number of key ministers, often dominates the government, but does so also to a different extent, in different ways and by means of different types of arrangements. The institutions of political systems within the state (including in local government) are modified by practices which it would seem so difficult and unnecessary to modify that it is simpler to let new habits develop and alter in effect what the constitution and other legal arrangements stipulate.
Since the key problem is that of the extent to which reality differs from the law, there is no hope of finding a solution by moving away from the word ‘government’ and adopting the ostensibly more flexible expression of ‘governance’, as has been increasingly the case among international organizations anxious to circumvent what is in effect the prohibition of the use of politically loaded expressions. Nor does the expression ‘core executive’, put forward by R. Rhodes and discussed in his article in the Oxford Handbook of Political Institutions Footnote 31 provide a mechanism of circumventing the difficulty: whether one is referring to prime ministers and ministers or to other members of such a core executive, for instance belonging to the public bureaucracy, in no way tells whether these bodies or their agents play in fact the part which they should be playing according to the rules or the arrangements.
All manners of subtle changes would have to be taken into account if one were to undertake a truly realistic assessment of the nature of political life in the state. This is already difficult to do when one considers a single state; if one attempts to undertake a comparative analysis of the ways in which decisions are taken in a number of states, the difficulties are multiplied. Thus it is in no way exaggerated to suggest that there is, at any rate currently, no hope of building true systematically theoretical analyses of the way in which political life develops at the level of the state; the complexity of the patterns of legal and customary arrangements is such that precision and simplicity cannot be expected to be achieved if a realistic picture is to be provided. Comparative government remains therefore more likely to progress by remaining in the hands of classicists eager to undertake truly realistic analyses. The complexity of the structures are such that a really general empirical theory cannot be expected to emerge at that level: does it not mean that the only solution is to search for such a general theory by undertaking studies at what might be regarded as a lower level, to be sure, but a level at which there are many – indeed large numbers of simpler examples from which to choose and on which to build at least elements of a general theory?
THE ANALYSIS OF MICRO POLITICS AND THE DEVELOPMENT OF A GENERAL POLITICAL THEORY
If one wishes to move in the direction of a general theory, there are thus reasons, indeed imperative reasons, connected to the complexity of state political activity, to try and find groups, at lower levels of generality, which are not as complex and are therefore easier to study. As a matter of fact, analyses at such lower levels are worth undertaking whether one aims at developing a general theory or not: as was stated earlier, it is somewhat surprising that political scientists should not have been interested in looking at the kind of politics which takes place at the level of groups below the state and that they should not be doing so as routinely as economists undertake inquiries at the micro level. The fact that this has not been the case seems to have to do with differences in the overall philosophy, so to speak, of the two disciplines: political scientists have always been concerned with the state because it was viewed (indeed with justification) as a leviathan in need of control; economists have never had to face such a problem.Footnote 32
There are indeed good reasons for studying politics at the lower level independently from the question of building a general theory. Politics at the level of the state is almost certainly viewed as distant and impenetrable by many, perhaps most, citizens: it is also regarded as too complicated to be grasped, this being possibly one of the reasons why the politics of the state is so often assessed in a negative manner. It does not follow that politics at the level of the groups which exist in society is universally simple; but it seems likely that, at any rate in many of these groups, politics tends to be less complex, largely because the scope of activity of the bodies concerned is more limited. It is therefore by investigating politics in groups which appear to be less complex that one could, as a preliminary inquiry, see whether one can hope to find a mechanism leading gradually to the discovery of a general theory. The remainder of this paper is thus concerned with some of the points which emerge if one is moving in that direction.
Such an inquiry cannot be effectively undertaken unless it is accompanied with, indeed based on, psychological analyses: yet psychological approaches have rarely been prominent in political science, which has, on the contrary, relied markedly on law, history and sociology.Footnote 33 This is possibly because at least social psychology was not markedly developed before the Second World War while it mushroomed since then.Footnote 34 Moreover, the importance given to legal considerations is unquestionably due to the fact that constitutions and laws have played a key part in the build-up of at least those political systems which were based on restraint. Meanwhile, sociology was largely devoted, since the early part of the nineteenth century, to the nature and development of the state. Thus markedly greater preference has been given, in endeavours designed to explain political development, to social variables (gender, race, class, as well as educational background and age) than to psychological variables such as personality traits. Thus, too, leadership has often been viewed as less important than the social structure, while it has also been regarded as brutal and evil and therefore as having to be reduced and perhaps even abolished altogether, while, from Rousseau onwards, despite Tocqueville and up to the group theory developments in the United States and indeed beyond, groups have been viewed with suspicion.Footnote 35 One of the purposes of the theoretical and empirical investigations of political scientists has therefore been to elaborate institutional instruments designed to reduce the importance of groups and of leaders and, according to the classical expression, to bring about ‘the government of laws and not of men’.Footnote 36 These developments clearly did not result in the study of psychology being given prominence in departments of political science; nor has it been at all common for students of these departments to embark in programmes combining psychology with political science. Such a tendency has to be altered: a different attitude must be adopted if political science is to undertake serious studies of the characteristics of groups and of their members.Footnote 37
At the risk of being criticized for proposing descriptive analyses rather than theoretical ones, three types of questions need to be examined before a general theory of politics can even begin to be considered. These entail: (1) inquiring into the broad ways in which groups can be classified on the basis of the type of goals which these groups pursue; (2) delineating the stages which political decisions are likely to go through; (3) coming to a clearer view of the distinction between actors and non-actors, while indeed introducing the important category of partial actors and thus looking at what may be in some circumstances a very difficult relationship between actors and non-actors. Let us examine these three points in turn and see how far they might indeed be able to advance the study of politics towards a more general theory.
Goal Variations Among Groups
States may vary in size, ideology and structure, but these variations are not regarded as affecting the fundamental raison d'être of the relationship between the state and the citizen, as this is assumed to be governed by what the law proclaims about that relationship in particular situations. This is not so in connection with groups, despite what has been typically stated, in particular by M. Olson in his well-known Logic of Collective Action,Footnote 38 on the basis of which it is typically pointed out that free riding is likely to be endemic in groups (at any rate in large groups). The matter is discussed in detail in R. S. Baron et al.Footnote 39
It is indeed questionable as to whether this standpoint is truly general within groups. The dichotomy was first proposed in the 1950s by S. E. Finer in Anonymous Empire Footnote 40 between protective and promotional groups; it was then to lead to the more fashionable subsequent division of these bodies into interest groups and movements. This suggests that there may well be many groups where the free rider problem is not likely to arise. While free riding may be endemic when what is at stake are the personal (typically financial) interests of members in protective groups, as with most trade unions, the decision to be a member or not does not arise for the same reasons in connection with those groups which individuals join to promote ideals or with movements in which they happen to believe.Footnote 41
Such a distinction between two broad types of groups on the basis of their goals – and in terms of the difference in the reasons why individuals come to join them – is one of the key ways in which the discipline of psychology is most useful for political science. For, in the course of the second half of the twentieth century, social psychology increasingly developed studies of ‘Altruism and Prosocial Behaviour’, which are currently a significant part of the development of the discipline, as was shown by C. D. Batson.Footnote 42 No doubt other classificatory bases will emerge gradually, also triggered by psychology, but there has already been a reflection as to whether the specific object of the group can result in a different approach and thus as to whether the relationship between the individual and the group will depend to an extent on the types of reasons which induce individuals to join. Such a question does not arise in the case of states, as the goals of states are very large, almost universal.Footnote 43
The forms taken by the decision-making process constitute an important part of the way in which the nature of political activity is characterized – and in particular whether the approach is cooperative or non-cooperative. This is to adopt the classical distinction drawn in rational choice analyses; to follow the more commonly-held descriptive distinction, this means to state whether the approach is based on consensus or on conflict. Advances made in analysing the stages of that process in social psychology help to determine which of these two forms prevails in a particular case (or what combination of these two forms does take place), a point which is also important in delineating elements of a general theory.
Groups and Processes of Decision-making
Groups vary markedly in terms of their processes of decision-making, partly because of the greater or smaller need to enter into a negotiation phase before the decision process itself begins. The forms which that negotiation phase takes, the characteristics of conflicts and the resolution of conflict disagreements have been studied by social psychologists in numerous articles. The general characteristics of coalitions are studied in this context: coalitions should therefore be examined by political scientists not just, as has been done so far, merely as they emerge in the context of parliamentary governments, but by taking into account, as social psychology does, the experience provided by the broader set of groups which social psychology studies, which is often markedly more informal.Footnote 44 For instance, social psychology emphasizes the point that there are individual differences among members of the groups and that these may well lead to differences in the way the negotiation takes place and comes to a conclusion, as is shown in R. S. Baron et al.Footnote 45 In this context, a distinction is drawn among four types of group members, individualists, cooperators, altruists and competitors. Specifically, ‘[i]t is not surprising that the greater value one tends to place on others' outcome, the more likely one is to make cooperative choices in social dilemmas’.Footnote 46 As Olson also suggested, group size does play a part. ‘This would suggest, for example, that smaller communities would be better able to provide needed public goods and preserve commonly held resources than larger communities…[p]erhaps because reducing group size affects each of the basic processes…maximizing self-interest, conforming to norms, trying to solve the group's problem effectively.’Footnote 47 Thus the analysis of the ways in which groups operate, even irrespective of the specific goals of these groups, suggests that there are here perhaps two poles of a dimension along which group decisions are taken, one pole being closely connected to self-interest while the other corresponds to the views of those who, for one reason or another, take seriously into account the interests of others.
Actors, Non-actors and Even Partial Actors
Especially in the case of larger groups, the question arises as to what is the relationship between the limited number of actors and the larger body of the non-actors. While the first two points which have just been discussed were concerned with the behaviour of actors only – and therefore could be regarded as constituting an horizontal analysis – one must consider a vertical aspect in decision-making as one extends over time, often significantly, the effect of the decisions: in this respect, it is not only necessary to examine the relationship among actors, but also the relationship between actors and non-actors. The first problem which arises is that of the fact of, and the reasons for, the division of members of groups into those who are at the centre and are willing to act and those who follow, the notion of membership being constructed informally and not necessarily in terms of persons having subscribed to the group and being for instance ready to pay dues. A general theory of politics needs to take into account such a distinction and to explore the reasons for the differences which can be found. These may be due to self-interest, to a general lack of interest, to a limited understanding of the problems as well as to more emotional aspects of the personality, for instance a degree of timidity which prevents some individuals from feeling able to participate. Characteristics of the self need to be taken into account in this respect – and social psychology has devoted considerable attention to the problem. ‘Possibly self-enhancing beliefs, even if unfounded, have strong benefits. For example, confidence may breed persistence, which may be helpful for achieving many forms of success.’Footnote 48 Whether or not one can ascribe rates or proportions to the extent to which various aspects of the personality lead to being or not being actors, it is manifestly mistaken not to place these elements in the equation.
Moreover, it would seem reasonable to believe that the two categories of actors and non-actors are not fixed over time, including with respect to a particular decision. Some actors may become less interested, or even change their minds about the outcome to which they may have participated; some non-actors may, on the contrary, become involved and may, for instance, want to see the decision altered. These variations in the part played by some members suggest once more that we are confronted with a dimension: over time, there may be a significant category of what might be referred to as partial actors. Only a careful study of variations of this type in the context of groups with limited goals and possibly of limited size, but probably not of truly small size, can provide a means of discovering the conditions under which some members are incited to change their position: such changes in classical state politics are frequent, but the reasons for the change are typically too complex (and too protracted) to allow for an adequate determination of the causes of the phenomenon.
The variations over time of the relationship between actors and non-actors, as well as the part played by partial actors in such a process need particularly to be explored, especially at the level of groups of a moderate size, as these relations are an important part of the background within which those who are, so to speak, permanent actors in the group have to operate. It is often pointed out in the classical political context of the state that, sometimes apparently suddenly and certainly without any clear premonition, a revolution occurs which can shatter the basis on which actors had hitherto operated and which these actors probably considered to be stable. It is plainly obvious that movements of this kind – especially those which are truly earth-shaking, such as the French or Russian revolutions and indeed the fall of the Soviet Union – can have such profound and varied roots (as well as being relatively speaking so rare) that it is wholly unrealistic to expect to be able to build a theory on the basis of these events. There are more limited revolutions, however, not just at the level of the state (such as the 1968 movements in France and a number of other countries) but within simpler groups and not necessarily ostensibly political ones. The probability is high that these more limited events can yield better results in terms of the reasons, structural or personal, which are likely to account for their occurrence: the probability is therefore also high that some clues will emerge in the process as to what shape theory may take, if based for instance on the idea that such movements need time to develop under the surface before they can break out openly and, perhaps, brutally.Footnote 49
It is therefore not claimed here that, by extending its interest, as it should have done long ago, to the groups which develop at lower levels than the state, political science will be able to solve the difficulties it has experienced in elaborating a truly satisfactory definition. It is unquestionably not claimed either that a comprehensive theory will be discovered which would account for the key relationships which emerge in political life. What is claimed is that political science has in a sense administered itself a major wound by effectively refusing to admit that there is politics below the state and that what goes on below the state is worthy of investigation. One could with some justification blame political science for being guilty of arrogance, perhaps somewhat surprisingly since that form of arrogance is not shared by other social science disciplines. The quicker that form of arrogance disappears, that is to say the quicker not just research but indeed teaching takes place in political science departments about politics within groups of all kinds, the greater the opportunity will be to see what general rules political developments obey in practice. From the discovery of such rules ideas about the shape of a general theory might emerge. Moreover, whether a comprehensive theory does emerge or not and whether a truly satisfactory definition comes to be discovered or not, the discipline will register one important side gain if and when political scientists become interested in politics within groups of all kinds: politics will no longer be viewed by so many citizens as an esoteric and perhaps even rather sinister activity, but as one in need to be studied and understood since it is practised by all of us, almost every day, in all types of situations.
