Resumé

This discussion acknowledges that interest group - governmental relations deserve further study, but argues that the corporatist literature of the last decade has failed to give useful guidance to the research. One major complaint concerns the hostility in many of the sources to pluralist analysis. As critiques, these attacks have been too shallow to be defended as constructive debate and in dismissing pluralist writings important insights into group-Governmental relations have been neglected.

The article broadly distinguishes two types of pluralism. There is the version which dwells on open competition between groups; there is also, however, a tradition of writing about and discussing closed group-departmental relations and sectorized policy-making. It is clearly this second strain that overlaps considerably with contemporary corporatist interest. It is suggested that it is worth reserving the term corporate pluralism for this model of segmented policymaking. This model is more descriptive than explanatory and is less ordered and systematised than corporatism appears to be in current theoretical use.

It is further argued that in use many writers impute to corporatism little more than corporate
pluralism suggests — and that advocates of corporatism now use the term in a less ambitious
(an empirical) manner.

The Plurality of Corporatism

Neo-corporatism has spread like a disease through the footnotes of political science. This epidemic tells us more about the weakness of the population than the strength of the virus. In a recent paper Wynn Grant (1983) asked, 'Does Neo-Corporatism Tell us Anything We didn't Know Already?' Grant asserts that it does, but this article asserts that the merits of neo-corporatism are still hard to discern and that the theoretical investment has yielded scant reward. In particular this article seeks to disentangle neo-corporatism as developed

---

* This paper is a development of ideas published in the Journal of Public Policy, 1981, No. 1, in an article titled, 'Iron Triangles, Woolly Corporatism and Elastic Nets'. The paper, largely in this current form, was presented to the Department of Politics at Oslo in 1982. An earlier version of this was published as a Working Paper, 'Corporatism: the Unity of The Concept', Strathclyde Papers on Government and Politics, No. 13.

by Schmitter and his associates from corporate pluralism which has been
popularized by this journal.

As a preliminary to the discussion of corporatist theory, one can note that there is no consensus in the camp. This problem is not unique to corporatism, but the attempts to claim their failure to agree with each other as a strength is unconvincing (Lehmbruch 1979b, 299).'

There is self-acknowledged variation between the uses of various corporatist protagonists, but the inconsistencies are not admitted as weaknesses. For example, Lehmbruch (1982, 2) argues that the plurality of conceptualizations mirrors the high degree of interrelatedness of the 'pluri-dimensions' of the concept. Schmitter, equally optimistic, talks about 'productive confusion'. The acerbic literary style of Schmitter would not be so generous in describing inconsistencies among his critics.

Martin (1983) has drawn particular attention to the disunity of the corporatists on the question of the 'State'. Given that Schmitter had claimed that corporatism provided 'a different way of conceptualizing the role ... of the state', it would be expected that by now some clear corporatist position would have emerged. But Martin documents a gamut of opinion from the state as executor of policies of groups (Nedelman & Meier 1979), state control (Schmitter 1979b), to state as bargainer (Lehmbruch (1979a). Of course pro-corporatists such as Grant (1983, 13) also acknowledge the ambivalance of the state in the model (does the state dominate the interests, or the interests the state?) but he still commends the model to us. But the differences between some self categorised corporatists may be more fundamental than between some of these corporatists and the pluralists they apparently scorn.

Corporatism then is not a single theory but a range of theories which are not self evidently unified or consistent and which have not been demonstrated by any proponent as compatible. The most influential, and often quoted, definition comes from Schmitter:

'Corporatism can be defined as a system of interest representation in which the constituent units are organised into a limited number of singular, compulsory, non-competitive, hierarchically ordered and functionally differentiated categories, recognised or licensed (if not created) by the state and granted a deliberate representational monopoly within their respective categories in exchange for observing certain controls on their selection of leaders and articulation of demands and supports' (Schmitter 1979b, 13).

The definition is well cushioned with reservations, but he does claim detailed inquiry into the extent to which a given system of representation, limited in the number of component units, compulsory in membership, noncompetitive between compartmentalized sectors, etc., will help us distinguish the type of interest system that pertains. One can then stress Schmitter's own emphasis. He explicity noted 'a specific concrete set of institutional practices'

and empirically observable group interests (1979b, 9). His approach is intended to allow us to distinguish, by empirical inspection, corporatism from pluralism. But one can note that in practice regimes might be readier to allow effective representation to a limited number of units, with some disputed hierarchy of internal structure with quasi-_compulsory membership, etc. In other words the empirical examination is likely to be less rewarding, more difficult, than we were led to believe in 1979. While Schmitter confidently states the postulated components of corporatism 'can be easily assessed, if not immediately quantified' (Schmitter 1979b, 14), such assessment is in fact very difficult. Although Schmitter also suggested that 'corporatism' was an ideal type description (or actually a constructed type), this reservation went back to discussion about empirical referents and empirical cases (1979b, 45), and he found the ideal type almost perfectly reproduced in Brazil and Portugal (1979b, 14).

The Relationship between Corporatism and Pluralism

The main justification for the corporatist cause appears to be to offer a new theoretical paradigm to the study of pressure groups, lobbies, interest associations which had long been (it is claimed) an area of 'conceptual torpor and theoretical orthodoxy in the discipline of political science' (Schmitter 1979a, 5). Elsewhere he put the goal as to give an explicit alternative to the paradigm of interest politics which has until now completely dominated the discipline of North American political science: pluralism (Schmitter 1979b, 14). Were there any doubt about the function of the exercise he also suggested that the element of the corporatist definition constituted, 'a sort of paradigmatic revolution when juxtaposed to the long predominant pluralist way of describing and analysing the role of organised interests ..! (1982, 260).

Given the priority Schmitter accords to this purpose, there is a remarkable ambiguity in his presentation of the relationship between pluralist and corporatist ideas. One manifestation of the ambivalence about pluralism is his combination of claims that pluralism has 'completely dominated' North American political science and that, 'A considerable number and wide variety of scholars have discovered it to be deficient' (1979b, 14). This is a strange kind of dominance that is so widely rejected.

Having noted the considerable number and wide variety of scholars rejecting pluralism, Schmitter goes on to compress the American pluralist debate into six lines in the footnotes (25-27) citing only Lowi, Kariel and McConnell. Given the intention to replace the pluralist model with a superior offering, one would have expected more elaborate connections between Schmitter's criticisms and the views of the considerable number of other critics.

The one author in the corporatist coalition who has gone into some detail on these matters is Anderson (1979). He concludes that his ideas as a corporatist are, in some ways, similar to Lowi's conception of juridical democracy (Anderson 1979, 297). After his review of, and identification with, the existing American literature, Anderson is hardly in the paradigm revolution business.

Schmitter's attitude to pluralism seems generally hostile, but he does complicate interpretation by his acknowledgement that pluralism and corporatism share a number or basic assumptions, 'as would almost any realistic model of modern interest politics' (1979b, 15). These are:

(1) the growing importance of formal associational units of representation;

(2) the persistence and expansion of functionally differentiated and potentially
conflicting interests;

(3) the burgeoning role of permanent administrative staffs;

(4) the decline in the importance of territorial and partisan representation;
(5) the secular trend towards expansion in the scope of public policy; and
(6) interpenetration of private and public decision areas.

Both pluralism and corporatism, according to Schmitter, accept and attempt to analyse growing structural differentiation and interest diversity. But notwithstanding these basic similarities, they are as indicated above, at other times presented as radically different.

Schmitter claims it has been 'rather convincingly' shown that Sweden, Switzerland, the Netherlands, Norway, Denmark, Austria, Spain, Portugal, Brazil, Chile, Peru, Greece, Mexico, and Yugoslavia 'by and large' fit his corporatist definition (1979b, 18). For some cases it seems Schmitter is going some distance beyond the intention of the authors of the studies cited. For example his claim about Norway is based on Stein Rokkan's chapter in Dahl's Political Opposition in Western Democracies, 1966. We know that the article, in its tail piece, turned attention from electoral politics to corporate pluralism, but that seems little enough reason to claim it as validating Schmitter's particular corporatist thesis. A few phrases show that Rokkan's picture scarcely resembles identikit corporatism: 'a vast network of interest organisations' (p. 106);² 'what really counts is the capacity to ... halt a system of highly interdependent activities' (p. 106); The Cabinet has increasingly had to take on the role of mediator between the conflicting interests in the national community' (p. 107); 'an attempt to establish a Board of Economic Co-ordination on the lines of the Dutch Social-Economische Raad (SER) failed miserably in the early fifties; the partners felt that such a formal body made them hostages of the government' (p. 108). There is no clear cut evidence here of controlled emergence of groups, quantitative limitation, vertical stratification, etc. Schmitter's claim as to having shown a fit with his model is overambitious.

Disturbing examples of intellectual imperialism are those where he claims that on the basis of authors such as Lowi, Beer, Dahrendorf, Presthus and Berger, the USA, Britain, West Germany, Canada, France can be seen as partly, if not in substantial portions, 'corporatized'. This, in the extreme, is a case for citing the old saying 'Give a small boy a hammer and everything looks like a nail'. And while castigating the misuse of the pluralist paradigm he sees 'Something approaching the corporatist model in Poland, Czechoslovakia, Hungary, Romania, even the USSR itself (1979b, 46).

On the one hand he complains that politics have been labelled pluralist, 'for no better reason that the mere existence of a multitude of organised interests', but when on the other hand he cavalierly labels countries as corporatist, he appears to be setting no superior example. For a theory presented as empirically based, there are not a lot of data. For Schmitter to cite Sartori (1970) and complain about pluralism (as a concept) travelling too far, too easily, appears suicidal. Sartori complains that cases cannot be proved by transferring the same denomination from one context to another. He says this amounts to pure and simple terminological camouflage: things are declared alike by making them verbally identical.

What is at issue here is the form of argument used by Schmitter. He labels countries as corporatist, then he calls this purely verbal device 'a demonstration of broad structural identity (which has) ... the virtue of debunking, if not divesting, some of these polities of the pluralist labels they have acquired..! (1979b, 18). After so 'proving' that these many parties are 'corporatist', Schmitter lists the characteristics of pluralism that accordingly do not apply:

competitiveness within sectors, hence accountability to members;

cross-pressures and overlap and, hence, vacillation and moderation in
demands;

open competitiveness between interest sectors, hence, split-the-difference
solutions;

penetration and subordination of political parties, hence, broad aggregative
party goals;

low party discipline, absence of strong partisan ideologies, absence of stable
hierarchies of organisation influence, hence, irrelevance of class or ruling
elite as political categories;

low barriers of entry irito the policy process, hence, key roles assigned to
'potential groups' and absence of systematic bias or exclusion;

major importance attached to lobbying and hence, concentration of attention
on parliament;

policy initiatives from below and passive roles assumed for state executives
and administrative bureaucracies;

wide dispersion of political resources, hence, neither omnipotent veto groups
nor powerless marginal elements; sheer multiplicity of interests (1979b,
18-19).
'IAI

Two main comments need to be made. Firstly Schmitter suggests that these do not apply because the regimes are corporatist, whereas the criteria should determine the conclusion and not vice versa. Secondly, it is in fact at least an open question that policy making is now dominated by widely dispersed political resources. Policy making in Western Europe seems more like Schmitter's pluralism than anything else — i.e., interest representation with spontaneous formation, proliferation, horizontal extension and competitive interaction (1979b, 16).

Not all the dimensions of his so-called pluralist model look useful, but arguably these are deficiencies peculiar to his rather exaggerated version of pluralism. Would, in fact, a pluralist be surprised at the developments attracting Schmitter's attention? And, indeed, would any pluralist recognise, and admit parentage for, the version of pluralism used by Schmitter?

The fault, if there is one, is not wholly Schmitter's; the pluralist literature perhaps differs between what it says and the impression it gives of what it is saying. So many commentators before Schmitter have said that pluralism, is about open access, equal resources, competition etc., that one can only assume that the pluralist exposition is deficient. Schmitter himself (in Berger 1981, 286) states: 'the "pluralist" system will be both self-equilibrating and selflegitimating'. However, who has established the origin of such claims in the primary literature? (Dahl 1982, Appendix A). If one looks at the primary literature — and here the discussion is restricted to E.P. Herring Public Administration and the Public Interest, 1936, D. Truman The Governmental Process, 1951, and R. Dahl, A Preface to Democratic Theory, 1956 — what does one find?³

For Herring, if we take them in historical precedence, we discover in the preface that in theory our government should strike a balance among these conflicting forces so as to promote the welfare of all. In fact, he claims, some groups are placed more advantageously than others within our governmental structure and under our industrial system. The government draws its strength from the very elements it is supposed to regulate. Its officials both elective and appointive are subjected to constant pressure from these powerful interests'. Herring continued 'Is the scope and development of our administrative service to be determined by the urging of special groups ... How can interests that are socially important but politically weak be given a place in the federal administration?' (p. 5).

Truman, in turn, does not assume that groups are, by definition, harmful but at the same time it is a preoccupation of his to limit the undesirable aspects of group action. For example, he writes that we cannot hope to protect a governmental system from the results of group organisation unless we have an adequate understanding of the process of which they are a part (p. 12).

Schmitter's apparent reluctance to accept the sheer multiplicity of interests

rings untrue as judged by successive case studies. Truman repeatedly returns to complexity — the 'bewildering array of groups, multiple access points'. Hugh Heclo (1978) wrote about, 'issue networks' precisely because he noted fragmentation in policy making structures — a tendency towards atomisation. One's instinct is that the corporatist theory is not just under-developed but that its basic features are not approximate to known circumstances. When, in McKenzie's term, the 'first guess' (1983, 110) is off the mark, refinement will not significantly improve its value.

It is quite possible that Schrnitter (and others) looking at Truman's chapter 'The Web of Relationships in Administrative Process' find that the examples have dated and that the lengthy discussion of group access, the description of administration (implementation) by groups, observation of a tendency to the 'inflexibility of the established web, and the tendencies to closed political processes', were nonetheless inadequately underlined. But it is unhelpful to the development of the discipline to find these points totally ignored — and a unrealistic version of pluralism put up as a target.

Turning to Dahl, a superficial reading can give some basis for the Schmitter account. In A Preface to Democratic Theory (University of Chicago Press, 1956), Dahl had defined the 'normal' process, one in which there is a high probability that all active and legitimate groups in the population can make themselves heard effectively at some crucial stage in the process of decision. A group excluded may nonetheless often gain entry (p. 138). This assumption of widespread effective influence is a basic pluralist tenet. But Dahl continues:

'Clearly (the capacity to be "heard" does not mean that every group has equal control over the outcome. In American politics, as in all other societies, control over decisions is unevenly distributed; neither individuals or groups are political equals. When 1 say a group is heard 'effectively' I mean more than the simple fact that it makes a noise; I mean that one or more officials are not only ready to listen to the noise, but expect to suffer in some significant way if they do not placate the group..!

There is not then much justification in the pluralist literature itself — and particularly its critical commentaries — for assuming that all groups are equal, that access is not problematical, that there is not a tendency for closed arrangements. Elsewhere {Journal of Public Policy, No. 1, 1981), I have drawn attention to the iron-triangle type literature in the U.S. In this was a recognition of closed, regularised arrangements. Perhaps one of the problems with the pluralist literature is that there have been two ideas implicit and inadequately delineated. There is a concept of open conflict and ad hoc competition, but there is also this idea of regularised relations.

If we are interested in the current practices of groups entering closed
relations with government, in access to these arrangements being difficult, to
imbalances in political resources, to bargaining relationships, there is in existence

a lengthy literature. If corporatists find this deficient, the onus is nonetheless on them to set out the grounds for deciding to start afresh instead of building on the works of these earlier authors. This point has been effectively made by Almond (1983, 202) in his review of Berger (1981), where he remarks, 'The casualness of the search of the earlier literature and the distortion of its contents are serious weaknesses in an otherwise important contribution to the interest group literature!

In his 1979 (c) chapter Schmitter does pose a string of possibilities that
seem not implausible. For example:

'What if the spread of education ... began to produce a fully organised, aware and mobilised polity...?. What if this new 'tautness' resulted in an enormous increase in demands for public allocations, and a concomitant stalemate in the parallelogram of group influences? What if the expansion of public policy resulted not in the removal of items from the agenda, but ... a consequent reinforcement of mutually exclusive associated demands? What if the bureaucratic-technical personnel necessary to run the proliferating interest associations and specialised agencies of the state began increasingly to act on professional norms of their 0wn...? ...what might happen or already have happened to the pluralist mode of interest intermediation?' (1979c, 79).

As Schmitter no doubt intended, these questions suggest affirmative answers — there are these tendencies. But there is no argument that leads us from recognising problems in advanced pluralism to endorsing Schmitter-type corporatism. In fact most of his problems of pluralism seem to lead to growing fragmentation, disorder, unpredictability — and these are difficult to reconcile with corporatism.

Schmitter concedes that sub-types of pluralism are possible, but he does not delineate them, and so, while pluralism is presented in its rather unsatisfactory fashion of open and equal competition, there is an assymmetry of treatment for corporatism. Schmitter divides corporatism into two forms — state corporatism (as in Portugal, Spain, Brazil, etc.) and societal corporatism as found in Sweden, Switzerland, the Netherlands, Norway, Denmark and perhaps in emergent fashion in Great Britain, West Germany, France, Canada and the U.S. (1979b, 21). This social corporatism (which has more relevance in Western Europe) has voluntary, bargained, qualities that make it particularly difficult to distinguish from pluralism.

Where is the distinctive role of the State in this pattern? Relaxing the rigidity of the corporatist formula does make corporatism more plausible (we know few western systems have formal, imposed corporatism) but the cost in making corporatism more relevant is again to make it less distinctive.

Corporatism and Corporate Pluralism

A gap in the Schmitter et al. conspectus of the relevant literature is corporate
pluralism. Any material not assumed to be corporatist (Schmitter 1976b,
footnotes 30-40) is rejected or given limited prominence.

Schmitter complains, for example, (1979b, footnote 19) that Martin Heisler's account of corporatism is faulty as it expressly links zuilen, 'pillared', notions to corporatism. However, in the Heisler (and Kvavik) chapter (1974), corporatism is certainly not a major theme and indeed they are explicitly cautious in their use of the term. They prefer to discuss 'a scheme of sectoral representation akin to neo-corporatism, or perhaps more accurately, corporate pluralism', (1974, 42). What there is in the chapter is a description of practices that seem so close to Schmitter's own interests that some account is called for which either incorporates their work in the new model on offer or suggests why the new model cannot absorb that superficially related material.

Heisler and Kvavik discuss group participation in the decision-making
process on a continuing basis. Access is established and structured (1974, 43):

in Scandinavian politics we find economic sectors ... that are highly organised and enjoy substantial self-government. Within each sector, a centralised and bureaucratized network of interest groups serves as the principal means for the advancement and co-ordination of sector interests. Most groups are in a position to develop and implement policy, etc' (1974, 47);

'a decision-making structure characterised by continuous, regularised access for economically,
politically, ethically and/or subculturally based groups to the highest levels of the
political system..! (1974, 48);

'By being brought into the policy-making structure, the various sectors are given a vested
interest in the continued successful operation of the structure..! (1974, 54).

The Heisler/Kvavik discussion of 'structured co-optation' (a recurrent phrase) is manifestly relevant for a discussion of realistic models of modern interest politics. If Schmitter's corporatism was new and necessary to move us from the conceptual torpor of pressure group studies (1979a, 5), then it would have been more useful to contrast this new approach with the latest refinements and developments in State/interest studies than his version of naive pluralism.

Since — fairly or otherwise — the term pluralism appears to mean public competition between groups, it might be better to reserve the term corporate pluralism for regularized group-departmental relations. To date there has been a fatal imprecision in the interpretation of corporate pluralism.

If one looks at the 1979 special issue of Scandinavian Political Studies (Vol. 2, No. 3) which had as its theme 'Corporate Pluralism in Nordic Democracies', it was crammed with data on the relationship between interest organisations and government — and that there is a real need for work in this area. However if one reviews the definitions and terms used through the course

of the issue, one discovers that the claim of a 'coherent' analytical framework and Nordic tradition in the study of corporate pluralism is debatable (see p. 195). Arguably, the superficial homogeneity of the essays in the special issue concealed some contributors describing Nordic practice as corporatism and others as the rejection of corporatism.

The Buksti/Johansen paper is offered as a contribution to the scholarly debate on corporatism or corporate pluralism. Though the conclusions of decentralisation and complexity are correctly held not to be in accord with corporatism, there is no distinction between corporatism and corporate pluralism. Where they claim that one of the most crucial tenets in the literature of corporatism is institutionalised group access and direct group participation in public policy-making, administration and implementation (p. 199), it is not explicit if this distinguishes corporate pluralism from corporatism or if they are interchangeable.

While this conceptual imprecision exists, the data and interpretations are invaluable. Following Heisler (1974) and Olsen (1978) they examine and confirm the hypothesis that predominantly narrow technical, divisible and measurable questions fit best into the bargaining process and compromises of the corporate structure.

Directly addressing the Schmitter formulation, Buksti and Johansen conclude that direct organisational participation in government is not restricted to a small number of interest groups. They argue that corporate structures and practices go hand in hand with a system of myriad groups and a decentralised structure. As argued above, it would be preferable to leave corporatism depicting that orderly pattern presented by Schmitter and categorise this complex, empirically derived, pattern as corporate pluralism.

Helander's discussion of Finland similarly contradicts the editors' claim of a coherent use of corporate pluralism. The term is not prominent in the account of interest representatives in the Finnish committee system. Instead Helander claims to be writing about corporatization — which meant, 'that interest organisations continually participate in the making of authorative decisions' (p. 221). No attempt is made to differ between Schmitter and Lehmbruch's corporatisms and Kvavik's (1976) corporate pluralism (p. 222). The danger of using so diluted a definition of corporatism along the lines of 'continually participate in the making of authorative decisions' is that much traditional interest has been inspired precisely by the recognition of such participation.

The same issue of dilution arises in Hernes and Selvik's article on 'Local Corporatism in Norway'. There they recognise that Rokkan's focus was corporate pluralism, but this they compact with corporatism which they loosely define as, 'a system of interest intermediation between organised groups, particularly in the economic area, and the state apparatus' (p. 261). Christensen

and Egeberg's contribution of 'Organised Group — Government Relations in Norway' again does not use the term corporate pluralism, but they retain a stable sense of Schmitter's corporatism. While they see some corporatistlike tendencies — segmentation, some hierarchy — overall they see the pattern as too complex to be adequately captured by the corporatist model.

In Martin Heisler's 'Corporate Pluralism Revisited: Where is the Theory?' he starts from the position (also adopted here) that there is an issue which had originally been given insufficient prominence in pluralist writings, but had been brought out by Beer (1966), Rokkan (1966), McConnell (1966) and Lowi (1969) — the structured, regularised participation of organised interests in policy making (p. 277). Heisler certainly doesn't regard corporate pluralism as a cure for all our difficulties in making sense of modern trends, but he is very effective in arguing that whatever the final shape of the solution it won't look much like corporatism, (as used by Schmitter). He argues, for example,

'when their work (empirical scholars) is viewed as a whole (it has) ... shown that corporate pluralism — at least in the Nordic countries, for which large bodies of data have been accumulated — is immensely complex, multifarious and polycentric; its norms are characterised by heterodoxy; the actor's motives are often unarticulated (and sometimes perhaps inarticulable), as well as ad hoc or opportunistic ... Thus while it' would be an exaggeration to say that empirical studies of corporate pluralism in the aggregate depict a system near chaos, it is difficult to escape the conclusion that the relatively parsimonious model of societal corporatism provided by Schmitter et al. imputes a degree of order far greater than that uncovered at least by those who have intensively studied the Nordic cases in the past few years' (pp. 284-5).

Corporate pluralism is, then, a description of the problem rather than a theory of causation. It is a statement to the effect that there are patterns of regularity, although there are a growing number of active groups which introduce uncertainty and disorder to the system. But a mere statement of the problem is more useful than a theory which fails to connect to observable reality.

From Empirical Corporatism to the Ideal Type

It is argued here that as corporatism has been put to empirical use a different rationale for the concept has had to be developed. Two tendencies merge. Firstly there is the practice of diluting the criteria of corporatism to the extent that it is difficult to distinguish from pluralism — and particularly corporate pluralism. Secondly, Schmitter himself seems less inclined to offer corporatism as a label for empirical practice (above) than an ideal type to highlight empirical practice.

The first of the diluted corporatisms was Schmitter's own societal corporatism.

Lembruch in his first chapter in the 1979 collection prefers as his term 'liberal corporatism' (1979a, 55). This concept is differentiated from traditional preindustrial corporatism, but it is not explicitly linked to Schmitter's societal corporatism, with which it appears to have some similarities. Lembruch stresses the large measure of constitutional authonomy of the groups and the voluntary nature of the integration of conflicting social groups (1979a, 54). The distinguishing trait of 'liberal corporatism' is given as the high degree of corporatism among the groups themselves in shaping public policy and the existence of two levels of bargaining. Firstly bargaining occurs within the autonomous groups, then the bargaining shifts to exchanges between the government and the 'cartel' of organised groups (1979a, 54).

Not only is this concept of liberal corporatism less demanding in its criteria than Schmitter's corporatism, it is in Lehmbruch's hands less manic in its geographical scope and his 'future trends' for his weaker model are more pessimistic than Schmitter's forecasts for 'societal corporatism'. In short Lehmbruch's work — appearing along with Schmitter's — appears to give credence to the latter — but there is in fact little in the Austrian case to encourage Schmitter.

In Lehmbruch's second chapter in the collection, he makes one claim that is worth accentuating — 'it is precisely because of the intimate mutual penetration of state bureaux and large interest organisations that the traditional concept of 'interest representation' becomes quite inappropriate for a theoretical understanding of corporatism' (1979a, 150). Such a working use of corporatism is, arguably, not distinctively corporatist.

As another example of 'soft' corporatism we can look at Alan Cawson's model (Cawson 1982, 39-40). He contrasts pluralism with a corporatist model of policy-making where representation (of demands) and implementation (of policies) are fixed within a mutually dependent bargaining relationship in which favourable policy outcomes are traded for co-operation and expertise. He stresses the bargaining relationship (distinguishing it from incorporation and co-optation). If we return to Grant's question, 'Does Neo-Corporatism Tell Us Anything We Didn't Know Already?', one can answer that a work such as Cawson's could emerge from a close reading of the traditional literature — 'Policy-making in a corporate society is thus a complex process of bargaining and negotiation between the state and corporate groups' (Cawson 1982, 41). Other than his emphasis on bargaining, Cawson is also perhaps un-Schmitter like with his image of a 'fragmented state' (Chapter 4) and when he sees even individual companies as the effective units in government/industry relations (1982, 37). Precisely what is there here that is not corporate pluralism?

One of the clearest expositions of corporatism comes in Colin Crouch's
rejoinder to Ross Martin (Crouch 1983). For Crouch, the difference between
pluralism and (neo) corporatism hinges on control of memberships by organisations.

The organisation might be privileged in its access to policy making, but it performs a role of social control — what Cawson terms, 'responsibility for delegated enforcement of government policy' (1978, 184). Crouch convincingly catalogues a sufficiently wide range of corporatists to sustain his point that this is a theme in corporatisl: writing. Whether it is the theme that he claims is still not obvious to this reader — and Crouch generously concedes that 'several of the authors (he cites) have not placed the stress on discipline and control of the membership .... fundamental to establishing the value of using corporatism as a concept with pluralism! In this formulation Crouch appears to imply that until corporatism was resurrected, writers had not paid attention to group control of membership. This neglects references as in Truman (1951), Chapters 6 and 7 on 'Internal Polities'. Obviously a group which could not 'deliver' its membership was a poor bargainer in pluralist terms.

Crouch is severe on Martin for missing 'the crucial feature of corporatism', but it is difficult for legitimate criticism if the actors can't agree on a script. If we accept Crouch's corporatism rather than any other, he introduces a useful continuum:

There is much to be said for this continuum but one is still not so confident as Crouch that in practice one can distinguish cases. More relevantly for this juncture of the article is to note that in this criticism of Martin's review, a new argument suddenly develops: one cannot complain about the development of the model, but it is difficult to debate with a hydra. And what are we to make of self proclaimed corporatists such as Chubb (1983), who (citing Schmitter) reckons that corporatism is a bureaucratic strategy of political exchange which lures, 'the major beneficiary into a mutually beneficial association by offering it considerable control over policy implementation and substantial input into new legislative proposals! For Chubb, then, corporatism is essentially about bureaucracy-beneficiary group relations as opposed to bureaucracy-cost bearing group relations which he labels co-optive. This is probably a valuable distinction, but such a pluralism of corporatisms is confusing.

Looking for a term to label the trends they identify how many authors have then grasped at the term 'corporatism'. For example, in their discussion of public-private concergence, Nachmias & Greer (1982) find new partnerships and alliances organised by areas of functional interdependence. They self consciously, and at length, adopt Schmitter's corporatism to describe this

phenomenon. It is however difficult to imagine they seriously wish to lumber their description of governing boards with the baggage of Schmitter's formula. Lembruch (1982, 4) admits that it has often been difficult to see why some authors have employed the corporatist term.

If such low level definitions of corporatism are to be allowed, an awful lot of corporatism is going to be found — but without endorsing the Schmitter approach. While neo-corporatists once complained that ... 'some schemes have carelessly characterised virtually any and all intimate interest group state relations which have been accepted as legitimate as corporatist' (Panitch 1979, 123), it is difficult to see that a much more rigorous and technical use is now in force.

Thus Lehmbruch warned in 1979, '... liberal corporatism should not be confounded with simply more consultation and co-operation of government with organised interest groups ..! (1979, 150) yet, in the volume he edited in 1982, several of the case studies are less strict as the point. Streeck, for example (p. 32) presents the transformation of the pluralist mode of interest representation into a liberal-corporatist one as being 'largely matters of emphasis and degree'. Several of the contributors appear well disposed to Heisler — who was rejected by Schmitter — see Johansen and Kristensen (p. 193), Wassenberg [p. 93). Johansen and Kristensen conclude that the corporatist channel is characterized by a pronounced capacity for bargaining and making of compromises (p. 197). This is again difficult to distinguish from pluralism.

'What is neo-corporatism for?'. In the earlier work countries were by and large corporatist (Schmitter 1979b, 19). In the original use corporatism replaced pluralism as an effective summary description of political systems — as for example in Schmitter's portrayal of 'the decay of pluralism and its gradual displacement by societal corporatism' (1979b, 24). In the early Schmitter social corporatism is 'found' 1979b, 22). He argued that while no empirically extant system of interest intermediation may perfectly reproduce or replicate corporatism, sets of observable, institutionally descriptive traits tend to cohere making it possible to categorise historically specific systems (or parts of systems). Again this is broadly stating that there are largely corporatist systems to be found (1979c, 65).

By his 1982 article it is conspicuous that Schmitter labels no particular
and specific country as corporatist (in this the volume which he claims is
about an empirical focus on corporatism).

Corporatism has retreated (in the size of claims made) as it has advanced (in terms of widespread adoption). In this more guarded version corporatism no longer is, but, 'For definitional purposes it may be preferable to define concepts in terms of polar opposites ... but the real world is almost always located somewhere in between' (Schmitter 1982, 265). 'Corporatism, however defined and however preceded by adjectives, is clearly not something a polity

has or does not have' (1982, 264-5). Where now the confident categorisation
in the 1979 volume which found, by and large, corporatism from Sweden
to Yugoslavia to Peru?

Now corporatism is not presented as descriptive but the ideal type end of some continuum; this is more realistic but it cannot be disproved. But if it was only ever intended as some ideal type unrelated to the somewhere in betweens of the real world, the concept would not have generated so much excitement.

Another aspect of the retreat of the corporatists concerns the new emphasis on advancing corporatism as having mainly sectoral relevance. Grant now seems to wish to defend corporatism from those who wish to see it fitting at single country level, by airguing (a) that it is a concept for intra country comparison, (b) that it is found in sectors of single countries.

This latter argument — what might be termed curate's egg corporatism — seems to appeal to several authors with an instinct for academic intermediation. Diamant (1981) for one accepts this view and Schmitter himself now proposes that, 'Practices resembling the corporatist model are quite unevenly distributed across issue arenas' (1982, 265). One must note that this corporatism in sectors is much more limited than 'the whole gestalt or syndrome' discussed in 1979 (Schmitter 1979, 14).

Sectorised corporatism is somewhat more modest a label than the banner
of a system of interest intermediation that was repeatedly flown in the 1979
volume.

The sort of explanations that might be appropriate to account for the development of close interest group departmental relations in one sector might be very different from those hypothecated for a system of corporatism. As one of the many puzzles corporatist pseudo theory presents us, it is interesting to see which kind of sector might be especially prone to corporatisation. Thus Lehmbruch (1979, 152) sees organised agriculture as, iess frequently included (in corporatist schemes) and when participating apparently have no decisive voice. Hence they remain largely confined to the classical pluralist 'pressure polities'. Schmitter (at least by 1982) on the contrary found agriculture as showing a marked propensity for corporatism (1982, 265). Does one have to be an intellectual skinhead (Grant 1983) to find this all rather unsatisfactory?

Wyn Grant argues that lor all the expressions of scepticism the verdict on corporatism, 'isn't in yet' (1983, 39). By coincidence Alfred Diamant ends his review of corporatism and related issues by observing it was hardly the inexorable pattern suggested by some. Like Grant (but from the opposite point of view) he observed that, 'Not all the returns are in on the "century of corporatism". A few precincts are yet to be heard' (Diamant 1981). If the Nordic countries, to name but a few, have been considered corporatist, it is also time for a recount.

There is, in conclusion, a subject worthy of discussion. It has been termed

by Olsen (1981), 'Integrated organisational participation in government'. One suspects that this is what excites the interest of many of those in the corporatist cohort: it attracts, and has done so for many decades, many who manage to study the phenomenon from a basically pluralist position. It is not argued that the new corporatist concerns are totally foreshadowed in earlier analysis. For example the corporatist literature pays significantly more attention to economic tripartism or concertation than is to be discovered in pluralist sources. But, at the same time, policy sectorization, group integration and consultation are not themes in themselves distinctive enough to constitute a new, rather than extended, method of analysis.

As the corporatists in the more modern times of the 1980's now realise, integrated participation is hardly likely to match up to Schmitter style (1979 vintage) corporatism. The actual characteristics of these current practices might be usefully thrown in relief by the corporatist ideal type. To that extent the corporatist literature is to be welcomed, but it is a pity it has been so negative about other approaches. Criticism of the inadequate exposition of the corporatist model (and its promiscuous use) does not argue that there is not a subject deserving of study.

---

NOTES

1. Though Schmitter's main article dates back to 1974, citations here are to the more accessible 1979 source: Schmitter & Lehmbruch (1979).

2. Page numbers cited in italics refer to books/articles of authors mentioned in the text but not appearing in the reference list at the end of the article.

3. This section is drawn from a bibliographical essay for a book edited by G. Drewry and D. Englefield, Information Sources in Politics and Political Science, to be published by Butterworths in 1984.

REFERENCES

Almond, G. 1983. 'Corporation, Pluralism and Professional Memory', World Politics, Vol. xxv,
No. 2.

Anderson, C.W. 1979. 'Political Design and The Representation of Interests', in Schmitter and
Lehmbruch (1979).

Berger, S. 1981. Organising Interests in Western Europe, Cambridge. Cambridge U.P.

Chubb, J. 1983. Interest Groups and the Bureaucracy: The Politics of Energy, Standford U.P.

Diamant, A. 1981. 'Bureaucracy and Public Policy in Neo-Corporatist Settings! Comparative
Politics, Vo. 14, No. 1.

Grant, W. 1983. 'Studying Business Interest Associations: Does Neo-Corporatism Tell Us Anything
We Didn't Know Already', Paper for Political Studies Assoc. Conf., Newcastle.

Heclo, H. 1978. 'Issue Networks and the Executive Establishment' in A. King (ed.), The New
American Political System, A.E.I.

Heisler, M. & Kvavik, R. 1974. In M. Heisler (ed), Politics in Europe. David McKay Co Inc.

Lehmbruch, G. 1979a. 'Liberal Corporatism and Party Government', in Schmitter & Lehmbruch
(1979).

Lehmbruch, G. 1979b. 'Concluding Remarks', in Schmitter & Lehmbruch (1979).

Lehmbruch, G. 1982. 'Introduction', in G. Lehmbruch & P. Schmitter (eds) (1982), Patterns of
Corporatist Policy Making, Sage Modern Politics Series, Vol. 7.

Martin R. 1983. 'Pluralism and the New Corporatism', Political Studies, xxxi, No. 1, pp. 86-102.

McKenzie, W. 1983. 'Imagination and Acquaintance in Political Science', Scandinavian Political
Studies, Vol. 6.

Nachmias, & Greer, A. 1982. 'Governance Dilemmas in an Age of Ambiguous Authority', Policy
Sciences, Vol. 14, No. 2.

Nedelmann, B. & Meier, K.G. 1979. 'Theories of Contemporary Corporatism', in Schmitter
& Lehmbruch (1979).

Olsen, J.P. 1978. Pre-publication version of Olsen 1981.

Olsen, J.P. 1981. 'Integrated Organisational Participation in Government', published in P. C.
Nyström & W. H. Starbuch (eds.), Handbook of Organisational Design Vol. 2, 1981.

Panitch, L. 1979. 'The Development of Corporatism in Liberal Democracies', in Schmitter
& Lembruch (1979).

Sartori, G. 1970. 'Concept Misinformation in Comparative Polities', American Political Science
Review, LXIV.

Schmitter, P. 1979a. 'Introduction' to Schmitter & Lembruch (1979).

Schmitter, P. 1979b. 'Still the Century of Corporatism', in Schmitter & Lehmbruch (1979).

Schmitter, P. 1979c. 'Modes of Interest Intermediation and Models of Societal Change', in
Schmitter & Lehmbruch (1979).

Schmitter, P. 1982. 'Reflections on Where the Theory of Neo-Corporatism Has Gone and Where
the Praxis of Neo-Corporatism May be Going', in G. Lehmbruch & PC. Schmitter (1982).

Schmitter, P. & Lehmbruch, G. (eds), 1979. Trends Towards Corporatism Intermediation, London,
Sage Contemporary Political Sociology, Vol. 1, 1982.