A transformational view of the state

Up to this point, the term “state autonomy” has been identified with two things: first with a distinctive state agenda, one not simply derived from private interests of particular persons in society, and, second, with the ca­pacity of the state to pursue and execute its will.

According to this formulation, sources of causality are partitioned into two categories, those sources internal to the state and those external to it. State autonomy exists when the causal sources of its behavior are not external. While this view of autonomy is prevalent in the literature, we argue that it raises difficulties.

In defining state autonomy in terms of causal superiority of state over economy, the autonomous state is placed in an uncomfortable position in several ways. First, it is difficult to avoid the conclusion that the autonomous state violates certain principles of democracy. One core democratic principle involves a relation of responsiveness between ruler and ruled. To the extent that this responsiveness is damaged by autonomous state action, democracy is compromised. There are ways to circumvent this conclusion, for example, by arguing that the state is the only agent capable of identifying the interests of society as a whole. But, by and large, the less responsive the state is to private interests, the more suspect it is on grounds of democratic theory.

Second, the issue of state autonomy raises troubling questions of a general theoretical and methodological nature. What does it mean for the state to be autonomous in the limit? If the state forms its preferences and executes them apart from society, is it not autistic? How can the state behave purposively unless it gauges its own actions on the basis of societal conditions and modifies that behavior on the basis of feedback? A state that takes little account of society would have to present its own actions as pure willfulness, as pref­erences arbitrarily formed and not socially grounded.

These two difficulties suggest that more serious problems may exist in the conception of state autonomy and especially in the way it requires us to think about the relation between state and economy. In this section we explore an alternative conception of the state we refer to as transformational. While the idea of “state as transformer” is not often explicit, it finds implicit, partial expression in the work of Theda Skocpol, Peter Katzenstein, and Peter Gourevitch. Skocpol develops an historical-organizational understanding of the state, identifying it with “a set of administrative, policing and military organizations headed, and more or less well-coordinated by, an executive authority” (1979:29). These organizations are formed out of social and po­litical forces coming together under particular historical conditions. Since the state is a historically specific structure, it must affect policy in the following sense: Differences in organizational structure of states must affect policy outcomes even when initiatives come from the private sphere. At a minimum, the state refracts those initiatives as its organizational structure transforms them into policy. At a maximum, the state creatively transforms even the subtlest of private forces. Apart from these extremes, the state organization affects the agendas (conceptions) of groups and “state structures help to inspire the very demands that are pursued through politics” (Weir and Skoc­pol, 1985:118).

The preceding argument that state structures make a difference provides a different foundation for the conclusion that the state is autonomous. The state is not apart from the private sphere, nor does it necessarily exert leverage against it. But neither does the economy exist as a fully formed entity, as a fundamental given existing prior to the state. The state enters into the con­stitution of society just as society contributes to the constitution of the state. Also, since policy outcomes depend in important ways on the nature of state organization, this requires that we treat the state as “a structure with a logic and interests of its own, not necessarily equivalent to, or fused with, the interests of the dominant class in society or the full set of member groups in the polity” (Skocpol, 1979:27).

If this analogy has relevance, then we would need to adapt our sense of state autonomy to it. Persons are autonomous when they make the outcome of such external influences acting upon them unique to them (their own result). So too with state autonomy. In Skocpol’s sense, state autonomy cannot mean the inability of society to influence policy (and significantly so). It must mean (at least in part) that each state processes those influences in unique ways as it also contributes (just as persons do) to what sort of influences it experiences. Clearly Skocpol’s approach does not allow us to reduce state actions to personal preferences (either of officeholders or others). She ex­plicitly rejects the idea of the state as an “arena” for social conflict (1979:25). If the state has any organizational structure other than that adequately and correctly designed to aggregate-transmit preferences of interests (see public choice theory), then state structure affects policy. The state is not autonomous since it is influenced and formed by social forces; neither is it simply an arena or mechanism for the operation of those social forces since it has its own determinate structure, its own capacities that act upon and contribute to the making of the societal forces acting upon it.

In addition to Skocpol’s work, that of Peter Katzenstein and Peter Gour- evitch has relevance for this idea. Although Katzenstein does not typically use the vocabulary of state autonomy, his idea of policy networks fits with the general thrust of this alternative idea of autonomy. By a policy network, Katzenstein refers to groups and institutions in state and society that come together to engage in policy making (1978:19). The central question posed by Katzenstein in Between Power and Plenty is “Why does a common chal­lenge such as the oil crisis elicit different national responses in the interna­tional political economy?” (1978:3). The answer in very general terms is that domestic structure is the critical intervening variable between international economic stimuli and domestic strategies and policies.

The significance of Katzenstein’s placing of domestic structure between environmental input and policy result is clear. Domestic structure makes a difference and it does so not because policy networks triumph over society - indeed, parts of society are definitionally present in the concept of policy network - but because these networks energize and give creative direction to the demands and disturbances emanating from society (even if transna­tional society). Perhaps a part of the different policy outcomes across the countries examined can be attributed to differences in the substance of de­mands themselves - for example, West Germany’s “preference” for policies bringing low inflation rates versus Sweden’s “preference” for full employ­ment. But even with these factors taken into account, there is a hard-core residue attributable to political structure variables.

Peter Gourevitch (1986) continues this line of thought. Gourevitch is ex­plicit from the start that the autonomous state is not one that stands apart from or in opposition to society. First of all, the state participates in creating its own social base, indeed, “its interventions frequently require the com­plicity of forces it seeks to regulate or direct” (1986:230).

Does the notion of national interest play a part in this organizational- structural approach? Only with difficulty. As Rueschemeyer and Evans point out, when we tie state action to the national interest this “contradicts the state’s role as an autonomous corporate actor, since it presumes that the goals of state activities are not generated inside the state apparatus but are dictated by the general interests of civil society” (1985:47). If, of course, the state defines the national interest (as Krasner claims) this need not be the case. But, if to be effective, the state must define a national interest appropriate to the type of society it represents and defends then, a narrow notion of autonomy cannot be sustained.

What are the implications of the historical-structural approach for the conceptual distinction between state and economy, public and private? Here we enter somewhat speculative territory. The basic conceptual framework for thinking about the state derives from Max Weber. It emphasizes orga­nizational structure joined to the instruments, power, and agents of legitimate force. Under this view, organizations whose decisions carry the authority of law and the backing of legitimate force together with their coercive agencies constitute the state. The private sphere consists of agents and organizations whose decisions bear no such authority. In a strict juridical sense, “society” consists of voluntary associations.

The key to this whole construction is the idea of legitimacy. Can force be legitimate simply because it belongs to the state? Or is there a sense of what is legitimate that the state must embody and to which the state must adhere? The second alternative threatens the autonomy of the state since it subjects states to externally given canons of legitimate behavior. In this case, the concept of legitimacy disrupts the structural approach while in some sense being necessary to its concepts of the state. Although legitimacy has an historical aspect (carrying different connotations in different societies), it is not primarily historical. The term requires us to relate state action to an ideal. This ideal is the mainspring of legitimacy. It is neither (1) in society (in the sense of being a private interest) nor (2) reducible to historically evolved state structure. In other words, the concept of legitimacy marks out a dimension of the problem not clearly amenable to resolution within the historical-organizational methodology.

The idea of legitimacy connects the state to a sense of public purpose or interest and requires us to consider more closely the relation between the state and the economy in connection with the relation between public and private, a connection not explicit in the historical-structural approach. Within that approach, the state takes on a quasi-private character. Thus Skopcol observes: “Any state first and fundamentally extracts resources from society and deploys these to create and support coercive and administrative orga­nizations” (1979:29). This tends to make the state an actor within society (the system of competition for resources). Such a tendency develops more or less inevitably out of the identification of the state with specific historical- empirical institutions.

While states are historically formed organizations that (among other things) facilitate the pursuit of private interest, these qualities do not suffice to ground an adequate theory of the state. By themselves, they leave us without its vital differentia specifica. Treating the state as just another organization concerned with the struggle for resources blurs important distinctions between the state and civil society. We do not deny that the state participates in the struggle for resources, or that it is like interest groups in some respects. We do argue, however, that the state has a capacity for more than self-interested behavior with its role in preserving the structure and norms underlying both society and the state itself. To bring the state back in, we need to take seriously the work of the state in setting the ideal (or ideological) underpinnings of social order. We can see an awareness of this necessity in the various theories considered in this book: in the pluralist notion of “training in the norms,” in Arrow’s notion of “consensus,” in Lukacs’s notion of “class conscious­ness,” in Gramsci’s notion of “hegemony,” in Parsons’s emphasis on so­cialization and common values, and in Krasner’s notion of “national interest.”

The idea links with observations made by leading pluralist and utilitarian thinkers concerning the limits of application of their methodology. Kenneth Arrow insists that there must be “some sort of consensus on the ends of society, or no social welfare function can be formed” (1951:83). Robert Dahl claims that the “extent to which training is given in [the] norms [of democratic process] is not independent of the extent of agreement that exists on choices among policy alternatives” and that “polyarchy is a function of consensus” on the norms (1956:76-7). The needs of consensus limit the range of private interest. In utilitarian language they concern notions related to those ex­pressed in the Marxian theory under the heading of hegemony and relative autonomy.

All of these terms allude to work the state does that we cannot subsume under the heading of satisfying private needs (see Elkin, 1985). In a sense this constitutes a kind of theoretical work undertaken by the state.