Politics and public administration Bureaucracy and pressure groups 185 Social movements and the bureaucracy 208 Bureaucracy and political parties 209… [600209]

1GUY PETERS
THE POLITICS OF BUREAUCRACY
Chapter 5
Politics and public administration
Bureaucracy and pressure groups 185
Social movements and the bureaucracy 208 Bureaucracy and political parties 209
Summary 210
Notes 211
Page 182
The preceding chapters were, in essence, th e foundation for this and the following three
chapters. We have discussed the relationship of public administrati on to its environment
through an elaboration of the social and economic surroundings of administration, the
cultural milieu in which administration func tions, attempts of public administration to
recruit personnel from that environment and the patterns of organizational structure
within the bureaucracy. In this chapter we be gin to examine the relationship of politics to
the conduct of public administra tion and to the policy decisions made by administrators.
It is sometimes appealing to think about publ ic administration as management, but this
would be misleading. The interaction of admi nistration with both formal and informal
political actors in the societ y has a profound impact on the beha vior of administrators and
on their decisions. The extent of this influence, and the manner in wh ich it is exerted, are
the subjects of investigat ion in this chapter and those that follow.
Perhaps the best place to begin this discus sion is by again noting the survival of the
ancient proverb of public life that politics and administra tion are separate enterprises and
that such a separation is valid both in the analysis of the institutions and behaviors of
government as well as in the ac tual conduct of public business.
1 Although any number of
authors have attempted to lay this proverb to rest, it has displayed amazing powers of
survival and reappears in any number of se ttings in any number of political systems.2 We
must therefore assume that this proverb, if not entirely or even partially valid from an
analytic perspective, serves some purpose for administrators and politicians. What does
the artificial separation of these two activities as sume to do that makes the survival of this
‘‘useful fiction’’ so desirable for both sets of actors?
For administrators, this presumed separation of administration and poli tics allows them to
engage in politics (organizational rather than partisan) without the bother of being held
accountable politically for the outcomes of their actions. Fu rther, they can engage in
policy making – presumably using technical or legal criteria for thei r decisions – without

2the interference of political ac tors who might otherwise recogn ize political or ideological
innfluences on policies and make demands upon them for the modification of those
policies.3 Thus, the actions of admi nistrators may be regarded by politicians, the public
and even by themselves as the result of the simple application of rational, legal, or
technical criteria to questions of policy. This apparent professional detachment may make
otherwise unacceptable decisions more palatable to the public.4 This appearance of
rational and technical decision making is heig htened when, as in the Anglo-American
democracies, great efforts are made to make the civil service politically neutral between
parties in office.
The separation of politics and administration also allows a certain latitude to politicians
which they might otherwise lack. In essen ce, the separation of these two types of
institutional choices facilitates many of th e difficult decisions of modern government
being made by individuals who will not have to face the public at a subsequent election.5
Thus it may allow politics to shape, or at l east innfluence, an important decision that will
be announced by a ‘‘non-political’’ institution th at will not be held publicly accountable.
Further, this conception that political and technical decisi on making can be separated in
public life has allowed political reformers to remove many important public decisions as
far as possible
Page 183 from the realm of ‘‘politics’’ – meaning largely corrupt machine politics and other
pejorative aspects of political life. Doing this results in many important governmental
functions being transferred from partisan political control to independent agencies,
bureaucracies and technocratic elites.
6
It is obviously assumed that th e administrators who make deci sions in these settings are,
in fact, insulated from political pres sures and are able to make decisions pro bono publico
because of that insulation. As we will show , however, these artificial separations of the
political and administrative functions, inst ead of removing decisions from political
innfluence, may actually subject them to diffe rent and more invidious types of political
innfluences. These innfluences are believed to be more invidious because, having already
been defined out of existence, they are diffi cult for the citizenry to identify and even
more difficult to control. Further, the capaci ty of the bureaucracy to make binding rules
for society may be hidden to all but the most astute members of society, given the
relatively arcane procedures used.7
Having still not completely exorcised the demon of the separation of political and
administrative choice, we are now at least in a position to understa nd why the actors in
the policy process may be willing to accept such a doctrine and why scholars may come
to believe them. Thus, although scholars ma y discount the dichotomy between politics
and administration in the abstract, when they confront the realities of how the actors
perceive their roles, they must accept at leas t the psychological realit y of the separation.
We will therefore go on to discuss the politic al environment of administrative decision

3making as well as the political innfluences on those decisions. In so doing, it is useful to
distinguish several basic dime nsions of the political activity of administrators.
The first of these dimensions is labele d ‘‘internal–external,’’ or perhaps more
appropriately ‘‘policy–survival.’’ On one end of this dimension is political activity within
the agency which seeks to take a variety of inputs from pressure groups, partisans, the
political executive, and any number of othe r sources and develop a policy. On the other
end of the continuum are politic al activities directed toward the maintenance and growth
of the organization – purposive and renf lexive goals in Mohr’s terminology.8 These two
forms of politics are rarely so neatly separated in real lif e, and each contributes to the
successful accomplishment of other goals. Howe ver, we can usefully distinguish the two
forms for analytic purposes and discuss the ty pes of innfluences likely to be brought and
the major loci of political connflict for each.
The second dimension of administrative politi cs is one of officialdom, or formality.
Administrators interact both with other govern mental officials (leg islators, the political
executive, other administrators, representativ es of sub-national governments) and with
unofficial political actors (lar gely the representatives of pressure groups). Again, these
interactions are not always clearly separable, for offici als often carry with them a
continuing commitment to the cause of par ticular interests, and pressure groups may
function in quasi-official capaci ties. However, it is useful to make such a distinction for
analytic reasons because the style of the interaction, its legitimacy and its probable
innfluences on policy will vary considerably as a function of the type of actor involved as
well as a function of the type of agency activity involved.
Page 184
The two dimensions of political activity by public administration, along with examples of
each category of activity, are presented in Figure 5.1. We show four categories based
upon a cross-classification of the two dime nsions. Thus we will be looking at
administrative politicized actions that have a characteristic of being both formal and
informal, and directed more toward policy form ation or survival. Our example of internal
(policy)–formal administrative politics is th e relationship between an upper echelon civil
servant and the cabinet minister he or she is designated to serve.9 Ministers, who are
charged with extensive political chores in addition to managing a large and complex
bureaucratic organization, cannot be reasonably expected to ha ve a sufficient grasp of the
issues involved or of the information availa ble for many policy decisions; such decisions
will, therefore, be produced through either cons ultation with, or delegation to, their senior
administrative officials.10 Consequently, interactions betw een ministers and civil servants
have become one of the dominant features of the policy-making process and must be
better understood in order for the analyst to pr edict the outcomes of the policy process in
contemporary political systems. Some progress has been made in formulating models to
assist in that understanding, but substantial refinement and conceptualization are still
required.11

4External–formal administrative politics are perh aps best identified in two ways. The first
is the process of public budgeti ng, in which administrative ag encies have to seek their
continued and expanded funding from other in stitutions of government. A number of
authors note that this is perhaps the most crucial locus of administrative politics because
of its pivotal role in the fu ture programs of the agency.12 It is certainly a political activity
that is the focus of an enormous amount of e ffort on the part of the agencies and one that
has received considerable attention in th e popular and scholarly literature. Budgeting
involves the mobilization of cons iderable political support for the agency, if it expects to
be successful in obtaining its desired funding, and consequently is an activity that will
involve considerable informal politics – l obbying by both interest groups and the agency
itself – as well.
The second important type of external–for mal politics is the politics of public
accountability through which other formal bodies may seek to curb the autonomy of the
public bureaucracy. Any number of institutions in the public

Figure 5.1 Types of bureaucr atic politics
Page 185
sector have some responsibility for seeing th at the public bureaucracy does not abuse its
discretion and acts in accordance with the laws that established its organizations and the
laws that it administers. Gi ven the importance of both of these types of ‘‘external–
formal’’ politics in contemporary politics, they will be discussed in chapters of their
own.13
Internal–informal administrative politics is probably best characterized by the
relationship of pressure groups and administration in the form ation of policy. In virtually
all political systems, attempts are made by interest (or pressure) groups to innfluence
public decisions.14 The openness of public administra tion to these innfluences and the
relative success of groups in obtaining the pol icies they seek are again a function of a
number of institutional, political and cultural factors that will require further discussion
and elaboration. However, except in the most totalitarian society, there is generally
considerable opportunity for group action a nd for group innfluence on the process of
policy formulation in bureaucracy.

5Finally, external–informal administrative politic s is best characterized by relationships
between interest groups, the public at large, and public bur eaucrats attempting to develop
support for their programs and for the contin ued success of the agency in the budgetary
process. Those bureaucrats have several means of trying to innfluence even the
inattentive public, including advertising and the promotion of a positive image among the
public, for example, ‘‘the Marine Corps needs a few good men.’’15 As was noted above,
this type of political activity is inextricably bound to the ability of pressure groups to
innfluence policy and the ability of agencies to survive in a competitive environment.
Now that we have some general picture of the scope and variety of administrative
politics, we begin our more intensive disc ussion of the politics of administration by
examining the informal side of these inter actions – that is, th e relationship between
administration and pressure groups, political pa rties and other unoffici al political groups
who are seeking to innfluence the course of publ ic policy or whom the administrators rely
upon in justifying their future programs and funding. Our discussion of the relationships
of political parties to administration is substant ially briefer than that of pressure groups.
This is so in large part because the major innfluence of party appears to be manifested
through official mechanisms, when members of the party occupy formal positions of
government and attempt to impose their vi ews on the bureaucracy. Since, almost by
definition, political parties are motivated pr incipally by the opportunity to hold public
office rather than the opportunity to i nnfluence policy through lobbying activities, it
makes more sense to look at the official ra ther than the unofficial side of partisan
activities. The major exception to that generalization would be small parties that know
they have little or no chance of holding o ffice and hence function in many ways as
interest groups.
Bureaucracy and pressure groups
We must suppose that the connflict between the demands of pressure groups and the role
of bureaucracy in decision making is, in mo st societies, one of the most basic in
government. On the one hand there is an institution of government,
Page 186
representing the authority of the state, impartiality, and in the Rechtstaat tradition of
Germany even a judicial temperament. On th e other side of the connflict are groups that
by their very nature represent only specialized narrow intere sts seeking some preferential
treatment from government. This type of division of the role of the state and the role of
interests is perceived differently in different political cultures. The connflict may not be
as intense in Norway as in, say, France, and may not be so intense in e ither of those as in
many less-developed societies. Interestingl y, however, the connflic t between bureaucracy
and interest groups has been sufficiently ameliora ted in most societies so that the two sets
of organizations are able not only to coexist but even to cooperate effectively.16 Further,
it is especially interesting that societ ies that have had among the most positive
conceptions of the public bureaucracy – Germ any, the Netherlands and the Scandinavian
countries – have been much more successful in accommodating the role of pressure

6groups into policy making than have political systems that have a less exalted conception
of their civil servants.17 In fact, a relatively positive eval uation of the civil service may be
required to allow the civil servants sufficient lat itude in dealing with the pressure groups
and in making accommodations to their demands for this relationship to be successful.
Given the apparent connflict in the roles of thes e two sets of political actors and
institutions, how are they able to cooperate so well, and so often in making policy? In the
first place, the stereotypical descriptions of the policy roles of these two sets of actors
obscure some of the reality of their interaction. The civil service rarely speaks or acts as a
unified entity. Rather, it is divided into numerous organizations, each with its own narrow
policy interests that happen to correspond to the interests of certain pressure groups.
Thus, agencies within the bureaucracy, especi ally a public bureaucr acy as decentralized
as that of the United States or the Scandi navian countries, may have a great deal in
common with the interest groups concerned wi th the same policy issues. Furthermore,
pressure groups in most Western democracies have found that acting in a less blatantly
self-interested manner in politics may produce greater policy benefits for them and their
members in the long run.
Second, connflict between interest groups and the bureaucracy is minimized, as both sets
of actors need each other to be successful. Administrators need the political support and
innfluence of pressure groups in their ex ternal relationships with other political
institutions, and they also need the inform ation supplied by pressure groups for making
and defending policies. Likewise, the pressure groups need access to the political process
and innfluence over the deci sions that are taken. This mutual need, given the
fragmentation of decision making in m odern governments, is the basic dynamic
explaining the frequent coope ration between public bureaucr acies and pressure groups.18
We are now left with the more formidable task of describing how the two partners in this
exchange interact, and what the effects are of differing pattern s of interaction.
We can classify the interactions between pressure groups and bureaucracies into four
basic types. This classifica tion, along with the presumed characteristics and effects of
each type, are illustrated in Table 5.1. This classificatory scheme places the interaction
between bureaucrats and pressure groups into four broad categories, constituting an
informal continuum from situations in which pressure-
Page 187
Table 5.1 Types of interaction between pressure groups and bureaucracy
Characteristics
Types Scope Innfluence Style Impact
Legitimate Broad Great Bargaini ng Redistribution/self-regulation
Clientela Narrow Moderate Symbiosis Self-regulation/distribution
Parantela Narrow Moderate Kinship Regulation/distribution
Illegitimate Variable None/great Confrontation None/redistribution

7group innfluence on policy is regarded as illegi timate to cases where it is regarded as
legitimate and necessary. Thes e four categories are themselves rather broad and may
contain substantial vari ation, but it is still useful to use such a scheme to begin to
understand the broad differences in relationshi ps between these actor s. Associated with
each of these patterns of in teraction are categories descri bing the manner in which the
relationship is carried on and its impact on policy.
Legitimate interactions
The first category of inter actions between bureaucrats a nd pressure group actions is
labeled ‘‘legitimate.’’ This denotes that in some political systems, not only are pressure
groups an accepted fact of political life, but th ey are also legally and officially involved
in the process of making and administering public policy. Indeed, in this conception of
government the involvement of interest gr oups with the public sector is not only
legitimate, it is almost nece ssary. The major examples of th is type are found in Germany,
the Low Countries and Scandinavia, but a number of other countries have adopted
legitimate roles for pressure groups in more limited forms.
Corporatism and public administration . One variant of the legitimate relationship
between interest groups and government ha s been described as ‘‘corporatism’’ or
‘‘neocorporatism.’’19 This is actually a rather extreme version of the legitimate
relationship in that it tends to restrict the num ber of interest groups involved in the policy
process and, to some extent, to incorporate that limited number into the state apparatus
rather directly. Phillipe Schmitter defines cor poratism as an arrangement characterized by
a ‘‘limited number of singular, compulsory, non-competitive, hierarchically ordered and
functionally differentiated’’ groups that are given a virtual license to represent their
particular area of competence.20 He further differentiates societal corporatism , in which
the private associations dominate the state in policy making, from state corporatism , in
which the state is the dominant actor.21 This definition of co rporatist arrangements
implies a monopolistic relationship of intere st groups in a particular policy area and
sanctioning of this relationship by some state or ganization. This pattern stands in contrast
to the more open bargaining arrangements thought to characterize pluralistic interest
group systems.
Page 188
Gerhard Lehmbruch develops a somewhat less restrictive version of corporatism, in
which the relationship between interest gr oups and government is less formalized and
there is greater bargaining among the groups themselves during the process of policy
making.22 This variant, which is more reminisc ent of pluralism, is termed ‘‘liberal
corporatism,’’ or ‘‘corporate pluralism.’’23 Subsequent research and writing on
corporatism has differentiated the concept ev en further, using terms such as ‘‘meso-
corporatism’’ and ‘‘the negotiated economy’’ to describe somewhat less restrictive
variations of the general patte rn of relationships between th e public sector and private
interests.24

8As initially formulated, none of these appro aches to the relationship between interest
groups and government had very much to sa y concerning the relati onship of groups and
the public bureaucracy. The major policy area discussed by the theorists of corporatism
was the setting of national economic policy, as in ‘‘Harpsund Democracy’’ in Sweden or
Konzertierte Aktion in the Federal Republic of Germ any. These meetings of the major
peak organizations of labor and management with the government involved negotiations
about the future course of wages, prices a nd profits. This was ‘‘hi gh politics,’’ directly
involving top government offici als, rather than the more mundane politics of policy
making by bureaucracy.25
But corporatism does have some relevance for this more ordinary type of policy making.
One of the major effects, similar to those that will be pointed out for the clientela
relationships between interest groups and the bureaucracy, is to rest rict the advice and
ideas coming into the bureaucracy.26 In the (arguably) corpor atist arrangements between
government and interest groups in Japan, fo r example, labor has been excluded, while
business groups dominate.27 If the strict corporatist definitions put forth by Schmitter do
indeed apply, then there will be a one-on- one relationship between interest group and
agency. In such an arrangement, only the posi tion of the official in terest groups will be
heard through this channel. Similarly, this form of relationship will likely produce even
greater incoherence in governme nt than might exist otherwis e. If agriculture groups are
talking to agriculture official s (be they political or admini strative), labor groups to labor
officials, and so on, then governmental prio rities and decision ma king will tend to be
highly fragmented. This is especially so if there is also a fragme ntation in the political
executive, for example, through cabinet committees able to trans-mit this individualized
priority setting into the highest reaches of government.28 A corporatist arrangement
would not be without benefits , however; the presence of a ca ptive interest group assisted
in legitimating actions and in implementing them certainly aids a governmental agency.
Further, as the relationships between peak interest groups and their members are deemed
to be hierarchical, there c ould be reasonably high levels of control exercised over
pressure groups that may present potential challenges to the legitimacy of policies.
Corporatist arrangements for making and th en delivering public policy have taken on
additional importance as governments have come under increased pressure to adopt
alternative mechanisms fo r service delivery. With ‘‘ government by proxy’’ or the
‘‘hollow state’’ being more important, servic es may be delivered by interest groups, by
other not-for-profit organi zations, or through private sector methods such as
contracting.29 Governments have been quite active in seeking
Page 189 the involvement of interest groups and in pr omoting partnerships between the public and
the private sector that can both deliver a service and provide le gitimacy for the policy.
30
The legitimating function becomes especially im portant as public tr ust and confidence in
government declines.

9In general, corporatism may be of limited importance for understanding how government
manages its relationships with interest groups. The entire pattern of relationships appears
to be under threat now, as the fiscal pre ssures that almost all developed economies
confront have made interest groups less coope rative with governmen t and less willing to
be co-opted.31 In addition, even at its high point, co rporatism describes only a relatively
small portion of the activities of interest groups – even those accorded some legitimate
rights of participation in govern ment – as they deal with government. Finally, for some of
the more well-developed corporatist systems in Western Europe, the movement into the
European Union means breaking up some of the cozy relationships that have existed at
home.
Networks and communities
Most forms of corporatism imply some lim itation on the number of groups involved in
interactions with government. Corporate plura lism is something of an exception to that
generalization, and several more recently developed conceptualizations about the
relationship of interest groups with government, as well as among public organizations
themselves, provide greater latitude for a range of interest gr oups to innfluence
government. There has been a growing body of literature focussing on ‘‘policy communities’’ and ‘‘issue networks ’’ as rather loose aggregations of groups involved in a
policy area and having (in general) legitimat e interactions with the public sector.
32
The basic concept behind network theory is that every policy area is populated by a large
number of interest groups. Each of these gr oups is attempting to have its own views of
policy adopted as law, and works with other groups to press their views. They do this
through making contacts with legislatures a nd the bureaucratic agen cies responsible for
the policy. They also interact wi th other interest groups that have similar orientations to
the policy in question in orde r to create coalitions to further those interests. These
‘‘networks’’ are also populated by other govern ment organizations that have connflicting
and complementary interests in the policy. As well as having an innfluence at the
formulation stage, these networks also beco me involved in implementing policies, much
as in the corporat ist arrangements.
The scholars who work in this field have engaged in a number of debates over the
different meanings of ‘‘networks’’ versus ‘‘ communities,’’ and even over different types
of communities (policy versus epistemic).33 We will not attempt to go very far into that
debate here, but will point out that ‘‘network ’’ tends to be a more neutral term, implying
only that groups interact. The term ‘‘commun ity,’’ on the other hand, tends to imply that
there is some sharing of common orientati ons to the policy area. For example, an
epistemic or interpretative community relies on a shared vision of the policy problem and
a relatively common scientific and
Page 190 professional basis for working with that problem. The increasing importance of the
European Community for public policy provides an especially interesting example of

10networks, given that there are at times mark edly different perspectives on policy among
the 12 member countries.34 The European case may be, howe ver, merely a special case of
the growing importance of international actors in domestic policy, with the international
‘‘regimes’’ surrounding policies having domestic impacts.35
This network literature has performed a numbe r of important services for students of
politics. Most importantly it has pointed out that even t hough governments may at times
attempt to limit the number of groups involved in a policy area, that is difficult to do in a
democratic political system. This difficulty is apparent even when the more important
groups also attempt to limit participation to preserve their own oligopolistic position.
There will always be groups with the resour ces to participate in some way, even if only
illegitimately. This does not mean that th e outcome of these interactions can be
predicted,36 but only that there is an enhanced capacity to pred ict who will be involved,
and to describe the pattern s of their interactions.
Another virtue of the network approach is that it focusses atte ntion on interactions
between government agencies and other ac tors – other government organizations,
politicians and private interest groups. Thes e interactions are impor tant whether or not
they occur within the context of formalized committees or other advisory structures.
There is an increasing body of evidence about who talks with whom in the process of
deciding about public policy. For example, in Sweden research from the ‘‘power project’’
has mapped patterns of interaction between departments and agencies and other
important actors in their environments.37 A good deal of empirical research in Norway
has also pointed to the extensive level of interaction between the public and private
sectors there.38 Knoke and Laumann in the United States mapped patterns of interaction
in two policy areas very thoroughly.39 In all these cases there was a very large number of
interactions, involving a wide variety of public and privat e organizations surrounding the
policy maker.
The empirical nature of this research approach enables us to compare interest group
innfluence over policy in a way that other ap proaches might not. For example, in general
we would expect interactions in less-develope d countries to contain a higher percentage
of contacts from institutional groups and other government departments than would be true in more developed systems. Likewise, there could be differen ces among policy areas
in the number, variety and type of interactions. For exampl e, a complex policy area such
as health care may have a wider range of interactions than would a somewhat simpler
policy area such as agriculture or defense.
40 For example, in the Swedish data mentioned
above, the defense department had on aver age many fewer contacts than did other
departments, with environment and energy ha ving a very high level of interaction.41
A final positive aspect of the network literat ure is that it does not necessarily assume
cooperation and agreement on policy as appears implied in the corporatist literature. If
groups are to be official in th e corporatist world they appear to have to accept much of
the existing policy framework. In a network or community
Page 191

11world, on the other hand, connflict is permitted and even expected. Sabatier, for example,
argues that policy change should be unde rstood through the connf licts of several
‘‘advocacy coalitions’’ construc ted around different interpreta tions of the policy issues
and problem solutions.42 In this view there are several networks that are involved in a
political struggle to control any issue. This a ppears a more realistic view of most issues
than does the more consensual one almo st inherent in corporatist models.43
Despite its analytic virtues, the network lite rature also has some problems as a means of
understanding the way in which interest groups innfluence policy.44 The most important
of these problems is the indeterminacy of a network. The approach argues, quite rightly,
that groups may be interested in making polic y, but does relatively little to explain how
one view wins and another loses.45 Are there not factors that explain the relative success
and failure of groups in pressing their demands ? The simpler corporatist and clientelistic
approaches are much better at predicting, albeit not always accurately. Further, given that
networks must be created and managed, th e real prediction of success may be that
government organizations will dominate, rather than the interests coming from the
society.
Other patterns of legitimate interaction
The two means of interaction described below, although they are involved to some extent
in corporatist and network arra ngements, are more generic and affect a much wider range
of behavior in the public sect or. They may be found in cor poratism but may also be found
in other systems that do not have such hi ghly formalized patterns of interaction.
Required consultation . The required-consultation mechan ism for legitimating pressure
group involvement in policy making is, as the name implies, the result of a variety of
rules that require administra tive bodies preparing new regulations to consult with the
relevant pressure groups for their opinions and to solicit advice and information from
them. In some cases this is done in the prepara tion of legislation to be sent to a legislative
body for enactment; in other cases required consu ltations are used for regulations that the
administrative body can issue as a result of dele gated legislative authority. In either case,
required consultation permits an interest group direct access to the making of
administrative policies.
One method of ensuring such input is through the use of remiss petitions as in Sweden
and Norway. In both these countries, when po licy modifications are being considered, the
administrative agencies are required to ask for remiss from interest groups. These
documents state the views of th e group as well as some of th e information that the group
considers relevant to the cas e. Originally the device was used only for pressure groups
directly affected by the new se t of regulations (for example, agricultural groups affected
by new regulations from an agricultural marketing board). More recently, the system has been extended to include virtually any orga nized group that wishes to submit an opinion
although there are still power differentials among the groups.
46 While such a system
provides no guarantee that the advice of th e group will be heeded, it does ensure

12Page 192
access to the relevant decision makers. Further, since it is customary for the remisser to
be passed on with any proposed legislation, th e system also provides some assurance that
those who must finally pass legislation are al so informed of interest group opinion. In
Germany, public agencies seeking to write ne w legislation or regul ations are required by
law to seek advice from interested pressure groups. In Germany the array of pressure
groups consulted on any one issue is not gene rally so wide as in the Scandinavian
countries, but the mechanism does allow direct and legitimate input of information and
opinions.47
While the remiss system relies on formal written co mmunications for pressure groups to
make their views known, other methods of re quired consultation employ more personal
approaches. Most notable in this regard is the use of advisory committees, especially in
Scandinavian administrations.48 The majority of public committees in these countries will
include representatives of some organized interests. For ex ample, in a study of Danish
committees, Johansen and Kristensen found that functional interest groups were
represented on the majority of committees and that local interests and institutional
interests within government itself were also frequently represented.49 A subsequent study
showed an increase in intere st group representation and a sl ight decline in the proportion
of official representation.50 In Norway each committee will contain a variety of
administrative and interest group personne l charged with advi sing the respective
administrative body on the proper course of public policy. These committees provide a
forum for the interest group to present th eir evidence and to make a case for their
particular views in an open forum. Intere stingly, however, most representatives on these
committees do not consider their role as that of an advocate of a particular position, but
rather as more that of a technical expert and manager.51 This role conception is obviously
useful in facilitating compromise in those committees on even difficult questions of policy. Further, the presence of all affected interests in the committee requires openness
and permits the analysis of evidence from a ll parties in a manner that can facilitate
compromise.
The use of advisory bodies is not confined to the Scandinavian countries, and in fact is a
relatively common featur e of administrative life around th e world. In Germany, interest
groups are represented on advisory boards for the ministries as a matter of legal right.
These advisory boards and commissions exist in virtually all count ries. France had some
3,700 in the 1960s, with more being added con tinually, and one more recent estimate put
the number at 15,000.
52 The biennial report on advisory committees in the United States
lists several thousand and, although less form alized, many also exist in the United
Kingdom.53 Richardson also argues that the use of advisory committees is a characteristic
feature of British public administration.54
What may distinguish the use of advisory committees in Ge rmany and Scandinavia is the
official sanction given to the role of the groups in the process of making policy.55 As
Kvavik notes, these systems move the role of pressure group innfluence from that of
input to that of ‘‘withinput’’; that is, in non-Eastonian la nguage, the political system

13recognizes pressure groups as an integral part of the decision-making process and
therefore accords them some of the same status accorded to other official participants in
the policy process.56 In Germany this situation may be viewed as a continuation of some
of the traditional corporate concep-
Page 193 tions of the state, but in all systems this type of innfluence provides an important
alternative to the usual liberal means of representation th rough elections and
legislatures.
57 Moreover, with the rapid growth of administrative policy making, it may
soon surpass the liberal means in its impact on public policy.
Implementation . The second major form of legitima te interest group involvement in
administration is the use of the groups as ag ents of implementati on for public policies.
Interest groups serve as quasi- official arms of the political system in implementing some
programs about which they are assumed to have expert knowledge and skills. Again, this
means of group involvement is particularly a pparent in Scandinavia and also frequently
occurs in the Low Countries.58 One of the most common areas in which administration of
this type occurs is agriculture, where e ither commodity groups or local farmers’
organizations administer regula tions, acreage allotments or c ontractual relationships with
the government. In Sweden the implementation of many portions of the labor law is left
to the individual groups most affected by the law.59 Similarly, in the Netherlands a large,
but declining, portion of that country’s comp lex system of economic regulation has been
administered by boards composed largely of interest group representatives charged with
the ‘‘self-policing’’ of a particular industry in order to maintain the delicate economic
balance within the country.60
In all of these cases of implementation with the aid of interest groups, the government
essentially allows the groups to engage in ac tivities in the name of the public, with only
indirect ‘‘political’’ control ove r their actions. They tend to be somewhat restrained by
having competing interests represented on the same administrative boards, but this is
obviously a manner in which interest groups can have a quite direct impact on the shape
of public policy and its execution. In most European societies this role is well
institutionalized although the theoretical separation of stat e and society in most Anglo-
American systems makes these implementation structures more suspect.
Again, we should point out the degree to wh ich the need of governments to implement
their policies in a cheaper or less obtrusive manner has given increased importance to the
legitimate role of interest groups as implementing agents.61 In such interactions the state,
of course, must be willing to trade some of its authority in return for ease of
implementation, and possibly even greater legitimacy among those who will consume the service. In some instances, such a trade-o ff might have been resi sted, but fiscal and
popular pressures in the beginning of the twen ty-first century are sufficient to make it
extremely palatable. In the language of state theory, states must b ecome weaker and less
autonomous as they become increasingly en meshed with, and dependent upon, private
interests.
62

14Although we usually think about the impact on government organizations, interest groups
also lose some autonomy when they become a part of the appa ratus of state policy
making and implementation. They run the risk of becoming coopted by government, with
maintaining their relationship to governmen t becoming as important a goal as changing
policy.63 The maintenance of the loyalty and commitment of interest group leaders to the
wishes of the rank and file members is an enduring problem in voluntary organizations.64
It is especially pronounced when leadership roles involve regular participation in
important
Page 194
negotiations over policy, and frequent contac ts with government officials. There is a
danger that interest groups become too much a part of the system of government to
represent adequately the wishes of their members to government.
Institutional groups . Institutional pressure groups appear to constitute a special class of
legitimate pressure groups. By definition, thes e groups are important social or political
institutions that are seeking to innfluence public policy.65 Some rather obvious examples
are the church, the army, and the public bureaucracy itself . Local governments, even in
unitary regimes, may also act as institutiona l groups. Institutional gr oups, like all interest
groups, seek to obtain benefits for themselves or their members, and their actions are
legitimated through the prestige of the institution, or perhaps the threat of extreme actions
in the case of the army in some societies. Johansen and Kristensen point out that these
groups are increasing more rapi dly than functional interest groups in their representation
on public committees in Denmark.66 Likewise, in France, To urnon argues that these are
the most effective groups in the system.67
Even in countries that tend toward legitimate patterns of interaction between interest
groups and the bureaucracy, institutional in terest groups may actually be better
conceptualized as a special class of clientela groups in that they have legitimate access
when a number of competing groups may not, a nd they tend to seek more particularized
special-interest outputs than tends to be true in legitimate pressure group systems. This is
often particularly true of local government organizations that lobby for support from the
central government, whether as a group or for their own particular community.68 In many
developing countries the military (and to some extent also the bureaucracy) is a special
case of the powerful institutional interest group, lobbying, or threaten ing, in order to get
what it wants from a relatively weak government.
As is noted in Figure 5.1, which characterizes the interactions of administrators and
pressure groups, the scope of interaction be tween the two in ‘‘leg itimate’’ situations
tends to be quite broad. A single pressure group may be consulted on a variety of
policies, and virtually all policy areas may be th e subject of inputs from interested parties.
Also, the innfluence of the legitimate groups on policy may be expected to be great
relative to other types of interaction patterns. The le gitimacy of the groups, their
frequency of interaction with ad ministrators and their official or quasi-official status all
make it possible for groups to have an impact that they would not have elsewhere. In part

15this is a function of not having to expend orga nizational resources simp ly to gain access,
and in part a function of the roles adopted by the interest group participants in the
process. These perceived roles contribute to th e bargaining style of their interactions. As
is noted by Kvavik, the domina nt-role type in these negot iations is the expert who
supplies information and opinion but who does not serve merely as an advocate of his
particularistic viewpoint.69 This bargaining activity was described by Stein Rokkan, and
the basic pattern has changed little since he wrote that:
The crucial decisions on economic policy are rare ly taken in the parties or Parliament: the
central area is the bargaining table where the government authorities meet directly with
the trade union leaders, the representatives of
Page 195 the farmers, the smallholders, and the fish ermen, and the delegates of the Employers’
Association. These yearly rounds of negotiations have in fact come to mean more in the
lives of rank-and-file citizens than have form al elections. In these processes of intensive
interaction the parliamentary notions of one member, one vote and majority rule make
little sense. Decisions are not made through the counting of heads, but through complex
considerations of short-term advantages in alternative lines of compromise.
70
When the innfluence of pressure groups must be more covert, this important bargaining
mode of interaction is, of necessity, elim inated through the politics of gaining access,
necessitating that few groups rather than ma ny will be involved in any one decision. This
characterization does not hold true, of course , in the corporatist arrangements described
above, in which there is a one-on-one re lationship between pressure group and
government. The relationship may not be as particularized as in clientela relationships
described below, but it lacks the broad bargaini ng style associated with most legitimate
patterns of interaction.
Finally, the policy consequences of this patt ern of interaction be tween pressure groups
and administrators are generally confined to two types (phrased in terms of the Lowi and
the Salisbury and Heinz typologies, which conti nue to have substantial relevance for the
political examination of public polic y): redistribution and self-regulation.71 That is, in
situations in which administrators are capab le of imposing the choices made by groups
through a bargaining and negotiation process, th e decisions taken are likely to take from
one group and give to another. This mean s of bargaining over policy is, however, a
relatively safe manner (politically) in whic h to adopt redistribu tive policy, since it
ensures the participation of both winners and lo sers as well as ensuring the application of
technical knowledge to the choice. These tw o characteristics – th e technical knowledge
of the participants and the presence of all competing sides – were in fact the criteria selected by respondents in Kvavik’s sample as most important in legitimating their
decisions.
72 Rather similar patterns of participa tion and involvement are found in Olsen’s
later work.73 Elvander and others likew ise note that, in Swede n, the inclusion of all
competing groups is important for the sm ooth implementation of policies adopted by
pressure group representatives cooperating with the government.74 In addition, Heisler

16and Kvavik point out that continued access to policy making may be sufficient
motivation in itself to produce compliance with the decisions taken, even in the face of
adverse decisions in the short term.75
In political situations in which the elite may lack the cohesion and consensus necessary to
implement a redistributive decision, these legiti mate interactions be tween interest groups
and administrators may result in policies of self-regulation.76 In these cases organizations
are generally allowed to manage their own affa irs and thereby essentially manage a sector
of public policy for the government. One example of this type of policy outcome is in the
area of agriculture, where the connf lict within the sector is re latively slight so that there
may be little need for directly redistributive decisions, and where some policies – such as
the allocation of acreage allotments – ma y have little effect on other groups.77 Even in
these cases there is pote ntial connflict, for exampl e, between agricultural
Page 196 groups wanting high subsidies, and therefore high food prices, and labor groups wanting
low prices and therefore lower subsidies; or agricultural groups may want to use
pesticides while environmentalists do not.
78 The choice between re distributive and self-
regulative policies may depend on the breadth of groups involved in any one decision as
well as the integration of the elites making and enforcing the decisi ons. In other policy
areas, one group may have such a monopoly of in formation and expertise that it is given
the responsibility of se lf-regulation on the basis of that expertise. This has been especially
true of medical and legal groups in the United States and is a prevalent finding for similar
groups in a variety of other political systems.79
The economic difficulties which most West ern democracies encountered during the
1970s and even into the 1990s have had contra dictory effects on legitimate patterns of
interaction between interest groups and the public bureaucracy. On the one hand, we have
already pointed to the increasi ng use of interest groups or ot her aspects of ‘‘third-party
government’’ to relieve some of the administ rative burden of government and to enhance
the legitimacy of programs.80 On the other hand, corporatism and other forms of
legitimate involvement of pressure groups appear much better suited to dividing a
growing economic pie in the publ ic sector than to decidi ng where the cuts must be
made.81 Access is important when there is more to be given out, but it becomes almost a
burden for the interest groups when there is little good news to distribute. For
government, however, there is a temptation, when faced with stress and a large number of
connflicting demands, to reduce its involvement with outside groups and make its own
decisions. The exact manner in which these co ntradictory pressures are manifested in
actual decisions will depend upon a number of cultural and situational factors, but
formalized relationships between governm ent and interest groups are undergoing
examination and some change. Also, alt hough the patterns are perhaps clearest in
Western countries, some of the same change s have been occurring in the less-developed
countries, especially the increased relian ce on the third sector for policy delivery.82

17Clientela relationships
The second type of interacti on between interest groups and administration is one of the
two major types discus sed by LaPalombara. A clientela relationship is said to exist when
an interest group, for whatever reasons, succeeds in becoming, in the eyes of a given administrative agency, the natural expression an d representative of a given social sector
which, in turn, constitutes the natural target or reference point for the activity of the administrative agency. What that rather long definition implies is th at a single interest
group becomes the exclusive representative of a sector, and there is a close connection
between the one organizati on and a single agency.
This type of interaction is characterize d by a perceived legitimacy on the part of
administrators of a single interest group rather than a form al statement of the legitimacy
of all or virtually all groups. The consequen ces of this seemingly slight difference are,
however, quite important. In the first place, the scope of
Page 197
interaction of pressure groups and administrati on tends to be rather severely constrained.
Each agency tends to select a single pressure group as the legitimate representative of its
particular social sector and to avoid most other groups seek ing to present information and
advice. Thus, whereas in the ‘‘legitimate’’ ar rangement mentioned previously the agency
might be able or required to c onsult a broad range of groups, in a clientela relationship it
will entertain a quite narrow range of information and advice. This narrowing is especially evident when two or more groups seek to organize a single sector of the
society but only one is accorded regular acce ss to decision making. This situation tends to
skew the sources of information, genera lly in the direction toward which the
administrators tended in the first place. Sule iman and others have noted that in France –
and certainly elsewhere – legi timate groups tend to be those whose economic strength is
undeniable and whose demands are in general accord with government policy.
83 The
pattern appears to persist after substantia l modernization of many aspects of French
government and administration.
The second consequence of this form of inte raction is that, while the innfluence of one
group may be increased, the overall innfluen ce of pressure groups on public policy will
be lessened. Not having legitimacy in any formal sense obviously reduces the
acceptability of special-inter est innfluence on policy for the general public, and the
ability of the bureaucracy to accept advice is also limited. Further, each pressure group
must expand relatively more of its organizational resources on the pursuit of access, so
that less is available for the information and innfluence functions. Moreover, in this
process of seeking access, any conception of th e ‘‘public intere st’’ – even as an aggregate
of pressure group interests – tends to be lost and replaced with a set of private interests,
each represented in government by a single agency.84 Associated with th e above is a need
to keep the negotiations and in teractions of interest groups and administrators private and
informal, thereby removing them even furt her from public scruti ny and accountability.
All these characteristics of the scope and manner of interaction indicate that the pressure

18group universe will tend to be less broadly i nnfluential over policies, with virtually any
innfluence that does occur having something of a taint of illegality among the general
public.
If we remove this taint of illegality from the idea of interactions between interest groups
and government, it is difficult to distinguish clientela relationships and corporatist
arrangements that have developed along sector al lines, especially in policy sectors such
as agriculture, with very close collabora tion between government and groups in almost
every country.85 That ‘‘taint’’ is, of course, importa nt in defining the dynamics of the
interactions and in innfluencing the legitimacy of the outcomes, but it is a very subtle
distinction.
The description of the interactions of in terest groups and public administration in
clientela politics leads to the characterization of these relationships as symbiotic. As in
biological symbiosis, this relationship implies a mutual de pendence of the two
participants. The administra tive agency depends upon the pr essure group for information,
advice, prior clearance of policy decisions a nd, most important, for political support in its
competition with other agencies for the scar ce resources within government. The pressure
groups, on the other hand, depend upon the agency for access to decision making and
ultimately for favorable decisions on
Page 198
certain policy choices. For both sides the existence of a clientela relationship serves to
regularize the political environment and to de velop friendships in what might otherwise
be a hostile political world.
This form of pressure group relationship w ith administration, noted by LaPalombara in
Italy, has been used by several authors as a means of describing much of the politics of
policy in the United States, and it would s eem to be prevalent in a number of other
political systems that have strong interest groups but where the interactions of these
groups and the government is at the margin of acceptability.86 Heclo and others have
argued, however, that this client elist pattern is now less descriptive of the United States
than it once was, and a much broader array of interests has come to be represented in
Washington.87 While this may appear virtuous on grounds of democracy, it does make
the policy-making process substantially less pr edictable than it had been. On the other
hand, the by now standard lament on the imp act of ‘‘single-issue politics’’ on American
government may lead one to continue to accept the clientelist description.88
Further, although the United States has been a principal example of clientela politics, it is
by no means the only example. Even countries th at have more legitimate interactions in
some areas may find certain sectors, or sometim es a broader array of sectors, behaving in
a clientelistic manner.89 As noted, agriculture often operate s in a clientelistic manner, and
financial interests have been able to pres erve this position even in countries being
governed by left-leaning governments.90 Likewise, given the plet hora of possible groups

19from which to choose, the European Union al so appears to relate to interest groups by
selecting a particular client thr ough which to channel innfluence.
Finally, the policy consequences of the clientela arrangement produce essentially self-
regulative and distributive outcomes. LaPalomb ara notes that regulation is one of the
defining characteristics of a clientela relationship but goes on to note that the regulative
activities undertaken are not necessarily those that would promote the ‘‘public
interest.’’91 Rather, they are activities that quite directly promote the interests of the
regulated. This pattern of regulation, as has been noted, has been referred to as ‘‘self-
regulation.’’ Again, this tendency toward self-regulat ion appears endemic to
administrative agencies and especially independent regulatory commissions in the United States. Lowi has argued that ‘‘interest gr oup liberalism,’’ or the appropriation of the
power of the state for private ends, is in f act the dominant characteristic of contemporary
public policy in the United States.
92 McConnell observes that th e ‘‘outstanding political
fact about independent regulatory commissions is that they have, in general, become the
protectors and promoters of the industries they have b een established to regulate.’’93
Evidence from other political systems, however, is that this phenomenon is not confined
to the United States but is a more general fe ature of industrial socie ties. In a variety of
settings the need for political support is su fficient to necessitate the replacement of
regulation with clientelism a nd self-regulation. Administrators may lack the resources
and the central political support to enfor ce programs of regulation in the face of
opposition of powerful and well-organized groups, so in essence they must gain support from those groups. However, as with Heclo’s analysis of clientela politics in the United
States, there is an interpretation that regul atory capture is less prevalent in the United
States than it once was, especially for the
Page 199 ‘‘New social regulation,’’ in which the regulatory bodies ha ve jurisdictions that cut
across a variety of industries.
94 Further, there is increasi ng attention to the design of
regulatory structures that can minimize the chances of capture.95
Besides self-regulation, clientela relationships also tend to be associated with
distributional outcomes, which may be merely more tangible manifestations of self-
regulative programs. In distribut ional politics, however, instea d of being allowed to make
its own regulations, a group is granted con tinuing benefits. In Lowi’s terminology,
distributional politics ‘‘create pr ivilege, and it is a type of pr ivilege which is particularly
hard to bear or combat because it is t ouched with the symbolism of the state.’’96 A further
feature of distributional politic s is its tendency to accord be nefits to all groups accepted
as legitimate rather than select some as worthy and others as less worthy. The close
relationship between the intere st group and the government agency can ensure that the
clients continue to receive something of va lue from government, and that questions of
redistribution and the need for adjustment of the relative benefits are rarely subjects of
discussion.

20Again, if we examine the impact both of a decline in the confidence of populations in
government’s ability to solve society’s problem s and of the declining budgetary resources
enjoyed by most governments, there ar e contradictory implications for clientela
relationships. On the one hand, the crisis ment ality that pervades government when faced
with declining legitimacy and less money te nds to produce more attempts at central
control. Central agencies con cerned with budgets and with the management of the public
sector as a whole attempt to impose their prio rities on the policy process, rather than
allowing the departments and agencies to wo rk so closely with private interests in
shaping policy.97 No longer are distribution and self-regulation so acceptable as policy
outcomes, but government is seen to need to set priorities and make hard decisions
among programs. As we will see in Chapter 7, a number of methods have been developed
to cope with these problems, but all depend upon the priorities set and the choices made.
On the other hand, the fiscal pressures on govern ment, and the crisis of confidence in the
system of governing in many countries, may ha ve forced public admi nistration closer to
organized interests and, therefore, may have accentuated tendencies to form clientele
relationships. If government need s an interest group to impl ement a policy, or to provide
advice and information which government orga nizations themselves are increasingly
hard-pressed to produce, then it is difficult fo r them to impose tighter controls on the self-
regulatory policies that have benefited those interests. As was noted above, this may be
the means by which clientele relationships lose their taint and become legitimate. Again,
how these pressures are played out will depe nd on cultural and situ ational factors.
The development of clientele politics is heav ily innfluenced by the peculiar politics and
economics of public bureaucracies. Despite the arguments that consider bureaucracy and
bureaucrats as integrated and homogeneous actors, seeking collectively to assume control
of the political system (at a minimum), ma ny bureaucratic systems are highly fragmented
institutions. By being so frag mented, they may be forced into competition simply because
there is rarely an effective central means of allocating resources according to the merits
of programs or the needs of society. This is true even if there are pressures for more
Page 200 priority setting arising from scarcity of resources. Bureaucratic competition for resources
can be overstated very easily, as it is in much of the ‘‘public choice’’ literature on
bureaucracy, and much of what competiti on does exist is a function of sincere
commitments to programs rather than of a desire to maximize budgets.
98 But
fragmentation within the public sector does contribute to the development of clientela
politics.
The budgetary process then tends to force public bureaucracies to seek public support and
make distributive accommodations in order to ga in that support. This outcome is further
magnified by the division among the other acto rs in the budgetary process and their needs
for other types of benefits and accommodations. The clientelism that extends between the pressure group and the administration may ex tend to a type of clientelism between
legislative committees and the administrative agencies they are ostensibly overseeing.
99

21This situation is, in pa rt, a function of the st ability of the actors i nvolved in the process,
as in the United States Congress, but more generally is related to the joint need of
administrators and legislators to serve a cons tituency. For example, legislators interested
in agricultural matters tend to come from predominantly agricultural districts; any
attempt to curtail the activities of an agri cultural program would not be well received in
their constituencies and would thereby threaten their chances for reelection.100
Institutional interest groups comprise a speci al class of group in th eir relationships with
the bureaucracy. In the case of clientele re lationships, there are a variety of important
patterns of relationship, although these too ar e under some pressure. In industrialized
societies these are local government groups, as well as interest groups within the
government itself.101 In less-developed countries, thes e groups might include the church,
the military, or the bureaucracy itself. These groups all have claims upon government that are difficult or impossible to deny, and, as with other clientela groups, these special
claims are pressed for special privileges.
Again, we would characterize the effects of clientela politics between the public
bureaucracy and interest groups as a tight in termeshing of interest groups, administration
and legislators, all of whom have something to gain by increasing certain types of public
expenditures. These interconnections produce patt erns of policy similar to those predicted
by Salisbury and Heinz in such situations: the parceling out of goods and services
available through the public budget in a manner that will provide each organized sector
with some portion of the benefits.
102 This, in turn, produces a lack of coordination and
coherence in the public sector as a whole.103
Parantela relationships
This is the second type of administrati ve pressure group relationship mentioned by
LaPalombara in his discussion of Italian interest groups. A parantela relationship
describes a situation of ‘‘kins hip’’ or close fraternal ties between a pressure group and the
government or the dominant political party.104 These relationships are generally
characteristic of preindustrial societ ies or of ideological regimes, but
Page 201 our discussion will show them occurring in a number of political systems in which there
is a single dominant party or faction. In th ese cases pressure groups must gain access and
legitimacy through their attachment to that particular party rath er than through their
ability effectively to represen t a sector of the society.
Parantela relationships between pressure groups and bureaucracies involve an indirect
linkage between those actors rather than the direct linkage discussed in the clientela
relationship. The important added linkage is the political party – most commonly a
hegemonic party – with which the pressure group must develo p some feeling of
consanguinity. In these cases, the pressure groups obtain their access to administrative
decision making through the willingness of the party to intercede on its behalf with the

22bureaucracy and therefore in essence to control bur eaucratic policy making. In
considering parantela relationships, we must be sure to think of organizations such as
think tanks and consulting firms as being re levant interest groups. With ideological
parties at the helm of government, there may be a need to ensure th at the advice given to
them is sufficiently ‘‘pure.’’ Of course, the pa rty itself may wish to provide such advice,
but they may also wish to endorse other or ganizations as providi ng acceptable ideas and
policy advice.105
There must, therefore, be a domination of policy making by a political party, something
that is not usually associated with Western de mocratic systems but that is still present in
those systems. LaPalombara, for example, fo und relationships of th is type existing in
Italy with the Christian Democratic Party.106 In France, the Gaullists during the first part
of the Fifth Republic involved themselves di rectly with the bureaucracy to favor one
interest group over another, especi ally in the area of agriculture.107 Likewise, when the
Socialists came to power they tended to press policy and the bureaucracy toward closer
alliance with a smaller but mo re leftist agricultural group.108 While this issue was
complicated by many issues internal to agricu ltural politics in Fr ance, the fundamental
point of the imposition of party control over ad ministration in order to favor one group is
clear. During periods of cohabitation the presidency and the prime minister may be
supporting different parantela groups.109
This type of interaction does occur in Western political system s, but it is more typical of
a number of political systems – the former S oviet system, a majority of African single-
party states and many Latin American countri es – in which one party or coalition is
dominant.110 As in the case of France, it is al so found where partisan competition may
exist but where there has been a tendency for one party to dominate government – for
example, Italy, Japan, India and Mexico.111 It is also common in more competitive
political systems in the relationship that exists between organized labor unions and
political parties, such as the relationshi p between the Trades Union Congress and the
Labour Party in the United Kingdom, or LO and the Social Democratic Party in Sweden.
Further, interest groups may be able to exer t an influence over th e parties by having their
members adopted into political party roles such as candidates for office.112
Another type of parantela interest group that has freque ntly existed in democratic
societies represents an attempt on the part of government to organize some aspect of
society traditionally difficult to organize. For example, in the United States, urban
renewal and model cities programs required the
Page 202
development of organizations in the aff ected neighborhoods that would represent the
interests of the residents. Such organizations have rarely been successf ul in that role and
more often have been co-opted by governme nt as a means of social control in the
neighborhoods.113 Also, in several Western European countries, government has taken an
active part in organizing consumers’ groups – also traditionally difficult to organize – and
these groups have frequently been critici zed for becoming co-opted by government. One

23very clear example of government’s forming organizations to organi ze difficult segments
of the society and using those organizations for its own benefit is the growth of para-
political organizations in Singapore.114 The People’s Action Party has formed a large
number of organizations that cut across trad itional lines in the society and form yet
another political base for that party.
The effects of parantela relationships tend to be quite pervasive. It is, in fact, one
tendency of these political systems that th e hegemonic party will seek to impose its
control over as much of the society and econom y as is possible. One principal means of
effecting this is the fostering of parantela relationships in a numb er of social sectors
through the co-optation of exis ting interest groups or thr ough the creation of new groups
directly allied with the party. The above example from France is a case of the party taking the side of an existing interest group in its struggle wi th other groups seeking to
represent the same social interests; and the previous Spanish regime ’s organization of the
workers into official or semi-official syndica tes is an example of a party creating its own
interest group structure.
115 In either case, this is an effective means through which the
party can extend its scope downw ard into the society to cont rol the nature of the inputs
being generated and regularize th e behavior of that social s ector in accordance with the
dictates of the party. This means of cont rol through organization has been especially
common in the former communist countries , challenged only when free organizations
such as Solidarity (in Poland) can gain some thing approaching legitimate status. It also
serves as a means of checking bureaucratic aut onomy within that part icular policy area.
The above would seem to imply that pressure groups involved in a parantela relationship
are little more than the pawns of a dominan t political party, and such an interpretation
would be justifiable in many instances. Weiner, for example, describing the relationship
of the Congress Party in India and its affiliated labor union, writes:
The Indian National Trade Union Congress – in reality the labor wing of the Congress
Party – is organized along these principles of political responsib ility and supports the
basic program of the present government. Its leaders proudly declare that their demands
are in the national interest, not in behalf of sectiona l interests. Their first loyalties are to
the Congress Party, then to the present governme nt, to the nation, and last of all to the
workers who belong to the union.116
While the Indian case is illustrative of many parantela relationships, it is by no means an
entirely general finding. Even in the case of pre-democratic Spain, Anderson can write:
Page 203
the conventional picture of unrepresentativene ss and ineffectiveness of the syndicates can
be greatly overdrawn. The government and th e syndicates did not speak with one voice
on public policy. The syndical leaders were expect ed by the system itself to play the role
of militant spokesmen for labor…. In their language and style of militancy many of the
syndical leaders were not unlike their counterpa rts in other Western nations. They were

24brokers, and they bargained for their clients, though in the last analysis they accepted the
judgment of the cons tituted authorities.117
The behavior of militant members of the Tr ades Union Congress in the United Kingdom
with respect to the policies of Labour governments represents the logical extremes of
independence of parantela partners.118 The pressure groups in these arrangements are
frequently capable of exerting substantial i nnfluence over the course of public policy, and
for many of the same reasons that motivate clientela pressure groups. The symbiosis
between a hegemonic party and a pressure gr oup is certainly not as important as that
between the clientela partners, but it is present. The pressure groups can be expected to
have some impact on bureaucratic choice beca use of their special relationship with the
dominant party. Typically in such a case the hegemonic party will have colonized the
bureaucracy as well as created it s own interest group partners. Further, both the party and
the bureaucracy gain the be nefit of the specialized know ledge of the group, thereby
reducing their own direct co sts for policy development and planning. Moreover, the
party’s direct costs of soci al control may be reduced by their developing subsidiary
organizations to perform functions that might otherwise have to be performed centrally.
The above-mentioned example of the rela tionship between the syndicates and the
government in Spain is one example of this type of control, as is the relationship between
communist parties and their unions in both hegemonic and competitive situations. The numerous party-sponsored organizations in China are perhaps th e epitome of this
relationship between part y and organizations.
Interestingly, the development of parantela organizations may also alter the behavior of
the hegemonic party and may pluralize (at least slightly) politics with in a one-party state.
The politics may begin to resemble those of clientela relationships, with greater
specificity by policy area. Dittmer writes that modernization in communist political
systems has been associated with the ri se of ‘‘quasi-interest groups’’ who practice
‘‘cryptopolitics.’’ Although th ey are not autonomous, and can not command the range of
resources available to interest groups in pluralist systems, they have been able to exert
increasing innfluence upon policy within the functionally specifi ed fields in which they
are assumed to have some professional competence. Such groups do not perceive
themselves to be representative of broadly popular constituencies, so this development is
hardly ‘‘democratic,’’ but it is consultative, better informed and meritocratic.
119
To some degree the policies adopted by the participants in parantela relationships are a
function of the ideology and program of the he gemonic party and as such may vary from
programs of the far left to the far right. In general, however, there is a tendency toward
distributive programs. This means there is a tendency toward dist ributing various goods
and services among the faithful and
Page 204
directing groups to develop claims on certain public goods and services as the appropriate
representatives of certain social sectors. In this sense, the party is acting as something of
a ‘‘canteen’’ for its adherent s and for official groups by essentially subsidizing their

25existence in the marketplace of pressure gr oups – and thereby essentially depriving any
competing, or potentially competing, groups.120 This may be especially important when
the dominant political party also has virt ually complete control over the economic
resources of the society and can distribut e both political and material rewards.
Thus, in some ways, the parantela arrangements closely resemble the corporatist pattern
of interest group relationships described a bove, especially the state corporatist model.121
Individuals would receive bene fits under such an arrangem ent as a function of their
membership in the appropriate corporate entit y, rather than as a ma tter of individual right.
Moreover, parantela relationships tend to be antithetical to the conception of modern
politics about the universalism of the distri bution of economic, so cial and political
benefits. Such benefits – even the most basic political benefits of the rights of
organization and participation – are, in parantela systems, most definitely the function of
having the proper political affiliations, and th e innfluence that any group may expect to
have over the outcomes of the decision-ma king process will be a function of this
consanguinity.
A second effect of parantela relationships is also obviously regulative. This is true not
only of the attempts of the party to re gulate the outputs of the bureaucracy through
regulating the advice that it recei ves, but more broadly to regu late the society as a whole
through the use of intermediary groups. Thes e intermediary groups not only structure
inputs but may also serve as means of impl ementing the programs of the regime. The
parantela relationships then serve as two-way streets; informati on – and to a lesser extent
power – can nflow in both dir ections. The extent to whic h power can nflow upward is,
however, ultimately determined by the willi ngness of the dominant political party to
entertain modifications and challenges.
Illegitimate group processes
The final category of interacti ons between administrators and pressure groups is labeled
‘‘illegitimate.’’ This term is used to describe a variety of political situations in which the
interaction of pressure groups with bureaucr acy may be defined as outside the pale of
normal political actions, but th ese interactions occur anyway . This style of interaction
may be a function of the nature of the politic al system as a whole, which may attempt to
suppress autonomous groups in society, or it may be a function of the nature of particular
groups, which are defined as being illegitimate as representatives of the social sector that
they purport to represent. In the first three types of interactions discussed, some or all
pressure groups were accepted as legitimate spokesmen for some social sector or another. In the case of the illegitimate pressure groups, neither the system as a whole nor
individual administrators may be willing to accept the legitimacy of the inputs of some or
all interest groups. It is rarely the case, how ever, that a political system will attempt
Page 205 to exclude all interest groups. Even the mo st ‘‘totalitarian’’ can find groups useful and
will attempt to form parantela style organizations.
122

26As might be expected, innfluence from pressu re groups of this type is not the normal
pattern of policy making. Such innfluences tend to be indicative of some rather
fundamental failures of the policy-making syst em in satisfying demands of one or more
sectors of the society. Thus th ese individuals feel constrained to go outside the bounds of
‘‘normal’’ politics to seek what they want fr om the political system. We may, therefore,
be discussing in large part the behavior of ‘‘anomic’’ pressure groups in their attempts to
exert innfluence through protest, demonstra tions and violence. It is not necessary,
however, to confine this discussion entirely to groups using violence as a means of
expression, for there are a numbe r of situations in which pr essure groups declared as
illegitimate in parantela or clientela arrangements may still seek innfluence and may
occasionally exert some innfluence on policies. The latter instances are rare. As one of
LaPalombara’s respondents in It aly noted on this topic:
I know of no policy within the Ministry of Industry and Commerce that says that there
are certain groups in Italian society whose repres entatives will not at l east be received. It
is true that once this is done we will assign different importance or give varying weight to
the proposals made to us by such groups , but they are free to approach us.123
Suleiman notes that in France groups defined as groupes de pression (in contrast to the
acceptable ‘‘professional’’ groups) may be received by the administrators but are
unlikely to be able to produce the results they desire.124
The illegitimate groups continue to play the political game by the administrator’s rules
and politely present their petitions and remonstrances, often knowing that their
probability of success is nearly zero. We may ask why these groups continue in these
seemingly irrational behaviors. There is th e odd chance that they may actually have an
innfluence. More commonly, however, they pers ist simply because this is what their
members expect them to do. This is the reas on the members pay their dues or give their
allegiance to the group, and the leaders must carry out a seemingly pointless exercise.125
Finally, having attempted to play by the ru les of the game, they may make future
extraordinary political activity appear more acceptable.
If normal politics cannot work for a group, then they appear more justified in using
violent or innflammatory means, even in societies that are usually very willing to
accommodate the demands of interest groups. Th is has been true, at least in part, of
‘‘social movements’’ in a numbe r of industrialized countries.126 These groups have not
had the permanent organizational structures that would be desirable for continuing
cooperation with government, and have tende d to eschew such relationships for
ideological reasons. Although usually associated with the political left – anti-nuclear
groups, the peace movement, tenants’ organizati ons, etc. – there also have been social
movements on the right. In all cases, they ha ve been willing to use confrontational
tactics, if not violence, to attempt to press home their political points.
The interaction between illegitimate pressure groups and administration Page 206

27tends to produce high levels of frustrati on and alienation for those groups. The rather
arbitrary categorization of pressure takes place even in political systems generally
receptive – if not partial – to group innfluences . For example, in the United States, with a
long history of pressure gr oup innfluence on agriculture pol icy, one group – the National
Farmers’ Organization – was for all practical purposes classified as illegitimate; members
reacted by descending on Washi ngton, DC, with their tractors.127 The level of frustration
may be even greater in political systems that attempt to suppress – rather than simply
ignore – the activities of interest groups declared to be illegitimate. The suppression of
Solidarity by the Polish government is an exam ple of this type of behavior which comes
most readily to mind.
We can better understand the characterizati on of the innfluences of these illegitimate
groups as coming essentially from extralegal activities – through some sort of connflict
with the system – and their innfluence as bei ng at best episodic. Li kewise, the impacts of
their activities on public policy are extremely di fficult to predict, if they occur at all.
Despite these limitations, it is important to understand that th ese innfluences may
occasionally be productive of important change s. The French student movement of May
1968, and its associated activitie s, have been used as an example of virtually every
political phenomenon known to hu mankind, but that should not re strain us from pointing
out that this is one example of an esse ntially illegitimate pressure group having a
substantial impact on a regime and on th e shape of subsequent public policy.128 The
American students and their protests agai nst the Vietnam War constitute another
example, while in some Latin American countries the argument can be made that
accomplishing almost any type of policy ch ange requires the kind of fundamental
challenge to the system that can be offered by illegitimate groups.129 Also, the early
successes of Solidarity in Poland produced some liberalization in the regime, although
the ultimate result was the imposition of martial law.
When illegitimate groups are successful, the impacts of their activities tend to be
redistributive, if for no other reason than they may force the system to recognize a set of
demands that it could previously declare as being outside its conc ern. Most illegitimate
pressure groups seek to transform the existi ng political system and its output distribution
in the direction of a redistribu tion of privilege, be it politic al, social or economic. Some
of the most obvious examples would be attemp ts of minority racial and ethnic groups to
have their claims for civil rights and equal treatment accepted when previously they had
been excluded from the political process.
In addition, there are other significant differences am ong these four classes of
interactions we have discussed. First, the activities of illegitimate pressure groups are
clearly the most distinctive. The other thr ee patterns accord some legitimacy to the
activities and innfluence of one or more groups, so that there are accep ted patterns of
interaction between the groups and the bureau cracy. In the case of illegitimate groups,
such interactions – at least if they are to have any significant e ffect on policy outcomes –
occurs almost by definition only in times of political crisis. Thus, the three more or less legitimate patterns imply a certain stability a nd institutionalization of innfluence, whereas
the illegitimate pattern implies episodic innfluence, or no innfluence at all.

28Page 207
Second, the legitimate pattern of interaction is the only one of the four in which there is
little or no politics of access. In this a rrangement, access exists for virtually any group
that seeks it – even for those that almost cer tainly would be declared illegitimate in other
settings. In one well-known example, organi zed street gangs in Amsterdam have been
given a represen tational role; organization appears to be the most crucial variable here.
By removing access from politics, such an in teraction pattern may in fact make the
pluralist’s dream of a self-re gulating universe of pressure groups formulating public
policy a possibility, if not a reality.130 As long as access remains a scarce and closely
regulated commodity, the possibi lity of finding the ‘‘public interest’’ among a set of
connflicting pressure groups is remote, if not non-existent. Having legitimate interactions
of pressure with administra tion – and, perhaps more importa ntly, open interactions of
pressure groups with each othe r in advising the ad ministrators – by no means ensures that
such a mystical entity as the ‘‘public interest ’’ will emerge, but it is more likely to appear
when interests are forced to bargain than when each interest is able to capture its own
portion of the administrative structure. This capture tends to conve rt public policy into
private policy. Likewise, unle ss one considers the hegemonic political party as an
accurate representation of the in terests of the population, the control of pressure groups
and bureaucracy by such a party is also likely to produce distortions of outputs from what
would emerge from a bargaining table, especi ally when many interests may be defined
out of existence by the dominant party. To put this in the terms of our original typology
of interactions, serious dist ortions of policy from what would emerge from a simple
bargaining process among competing groups are likely to occur when the politics of
policy making cannot be removed from the politics of organizational survival.
We should also note that we have not been able to argue clearly that any particular
pattern of interaction characterizes any one nation or a nother, although the examples tend
to point to some important patterns. In th e first place, political systems with hegemonic
political parties, be they ostensib ly democratic or not, tend toward parantela relationships
between interest groups and admi nistrators, if for no other reason than that the hegemonic
party is able to use these relationships as one means of social control and regulation.
Second, legitimate interactions tend to be ch aracteristic of the Northern European
countries, which have had long histories of th e involvement of organized groups in social
and political life, and whose l eaders have perceived a need to manage potentially divisive
connflicts within the society, either et hnic or socioeconomic in origin. Third, clientela
arrangements tend to be quite common in any nu mber of societies, especially when there
is a fragmentation of interests and a lack of overall coordinati ng mechanisms in the
political system (for example, a dominant political party or institution), that can regulate
the competition among interest groups or among the competing agencies within the
bureaucracy; the United States is a major, but by no means the sole , example. Finally,
illegitimate interest groups may arise in virtually any setting but tend to be most
important in settings where they are least lik ely, for example, in societies that seek to
suppress interest groups or at l east a wide variety of interest groups. That is to say, these
groups are most important in settings where th ey serve as a fundamental challenge to the

29regime. This means that their day-to-day interactions with administrators will be
unfruitful if they
Page 208
occur at all, but that they may produce substa ntial transformations of a political system.
Just as there is little patt ern of interest group–bureaucratic interaction by political system,
there is also little pattern by type of interest or policy area. There is some tendency for
interest groups that can be clearly defined ge ographically to be able to establish clientele
relationships with administration, perhaps b ecause of the ability to mobilize political
support more easily. The most obvious example of this pattern is agriculture, which has
been notoriously successful in utilizing clientele relationshi ps in almost all political
systems. Likewise, interest groups that may be vertically integrated with political parties
– frequently labor unions with labor parties – may develop parantela relationships, even
within the context of competitive political systems.131 Finally, groups that may be
regarded as outcasts in normal social affa irs, or that are not regarded as having
differentiated political view points by the dominant community – racial minorities,
students, women – may tend to act through illegitimate rela tionships with bureaucracies,
if they are able to form any relationship at all.
Social movements and the bureaucracy
Interest groups represent the majority of so cial and economic groups in society, but by no
means all. Changes in the structure of soci ety and in the issues being confronted by
government have spawned a variety of social movements.132 These groups differ from
conventional interest groups in several ways. First, they are often short-lived, created to
address a single issue and then disbanded when that issue is resolved or loses saliency.
Second, they are more commonly associated with non-economic issues such as the
environment, peace, human rights, etc., than with ‘‘bread and butter’’ economic issues.
Finally, their organization tends to be less stable and institutionalized than is true for
most interest groups or political parties.
As noted above, social movements tend to func tion most often as illegitimate interest
groups, often refusing on principle to c ooperate more directly with government.
Movements are often created because the ex isting close collaboration between certain
interest groups and the bureaucracy is pe rceived to exclude other equally important
segments of society from being heard. T hus, the members of social movements may
perceive most existing inte rest groups as being in parantela or clientela relationships
with government, even when those other groups themselves believe that the relationships
are more open and broadly legitimate. Thus, a good deal of the politics of social
movements revolves around defining issues a nd defining participation (whether with the
bureaucracy or more generally) in ways that have not been conventional in the political
system.

30As noted, any individual social movement tends not to pers ist long in the political life of
a country. They generally simply disband, but some also become more institutionalized
and become interest groups much like others, or even political parties. The environmental
movement is the clearest example of th ese changes over time. Although some more
fundamentalist environmental groups still beha ve like social movements, the majority
have become part of the normal
Page 209 interest group environment of government, esp ecially when there is broad legitimacy of
interest group activity. In many European countries these former social movements have
become institutionalized as ‘‘green’’ politic al parties and have become part of the
governing coalition, most notably in Germany.
Bureaucracy and political parties
In most contemporary political systems, th e direct impact of partisan concerns on
bureaucracy has been consciously limited by a number of structur al and procedural
devices. The most important of these, of cour se, is the institutionalization of the merit
system for appointment and retention of ad ministrators so parties can no longer force
large-scale changes of administrative pers onnel when there is a change in governing
parties. While some patronage arrangements certainly do exist in all political systems,
any widespread use of patronage is generally regarded in Western countries as evidence
of corruption and mismanagement. This self -serving view is somewhat less easy to
justify since the 1980s, as a num ber of Western countries ha ve begun to politicize their
civil services, and to recruit more sp ecialist political advi sors for ministers.133
A number of non-Western count ries, despite the tutelage of their former colonial
countries, have continued or reinstituted non-merit systems of appointment to
administrative posts – even the most routine a nd trivial of posts. This is justified largely
on the need for national unity and mobilizat ion in the face of the difficulties of
development. In such situations, loyalty to the nation – or more exactly to the current
regime – is considered more important than the possession of certain scores on objective
tests or the possession of requis ite diplomas. This practice is by no means universal in the
non-Western world, but a number of one-party re gimes tend to recruit their bureaucracies
in this fashion. As Kwame Nkrumah once said in relation to administration in Ghana:
It is our intention to tighten up the regulati ons and to wipe out the disloyal elements of
the civil service, even if by so doing we suffe r some temporary dislocation of the service.
For disloyal civil servants ar e no better than saboteurs.134
Also, in French-speaking Africa, a number of one-party regimes have attempted to use
partisan control to replace ‘‘selfish individualism’’ with ‘‘patriotic socialism.’’135 While
the language is less colorful there, political leaders in a number of Western democracies have stated their wishes to develop a ‘‘com mitted’’ civil service that would follow the
wishes of the dominant political party.
136 A smaller number of such countries have in fact

31developed the mechanisms for creating such a po litically loyal civil service. As yet, these
mechanisms have remained within the bounds of civil service law, if not always within
previously prevailing customs and understand ings. The potential for waste and abuse, and
the reinstitutionalization of the spoils system is, however, quite apparent. In this context,
we must remember that the pol itically neutral civil service that we in Anglo-American
democracies in particular consider to be normal is really only a century old, while the
desire for political loyalty is much older and perhaps much stronger.
Page 210 While the increasing politicization of the civil service is one option in the context of an
age of ‘‘conviction politics,’’ another option is the diminuti on of the political and policy-
making roles of the civil service.
137 If, no matter how loyal and committed they appear,
the civil service still has the ‘‘taint’’ of obj ectivity and neutrality, th en they may not be
perceived as being really tr ustworthy. Again, these doubts would be present even in
countries that have become fu lly accustomed to a civil servic e with substantial power in
policy making. Thus, one reaction to the percei ved need for greater political loyalty by
the civil service may be an even greater acceptance of the traditional politics–
administration dichotomy, at least among those who actually work in government.
The most obvious example of the utilization of partisan control ove r the state bureaucracy
occurred in the former Soviet Union and other European communist countries, and
continues in China and in some authoritaria n governments on the right. This control is
achieved to a great extent through dual hier archies – one party a nd one administrative –
used simultaneously to execute policies a nd to check for the political orthodoxy of
personnel.138 Such a system of duplication app ears redundant and inefficient to many
Western analysts of organizati ons, but it is deemed crucial in systems in which political
orthodoxy is so important. As with the non-Western systems of the underdeveloped
world, partisan control and th e use of the bureaucracy as a mechanism for fundamental
social and economic change appear to go hand- in-hand. Where political neutrality is not
really acceptable, much less valued, then many of the Western dogmas concerning non-
partisan merit appointment are simply not feasible as criteria for evaluating the
recruitment and executive actions of administrators.
Summary
We have developed a means of classifying and analyzing the politics of bureaucracy.
Beginning with the notion that it is not useful to separa te the political from the
administrative in either real life or analysis , we have attempted to provide some means of
better understanding how administration become s involved with politics and political
actors. This chapter has dealt primarily with administrative involve ment with pressure
groups, showing the extent to which these tw o political actors depend upon each other in
their attempts to shape public policy and to survive in what might otherwise be an
extremely hostile political envir onment. In three of the f our patterns of interaction
discussed, some type of legitim ating relationship was developed so that a stable pattern of
interaction between group and bureaucracy could be used in policy formation – the

32internal aspect of bureaucrat ic politics. These relationshi ps could, in turn, directly
(through clientelism) or indirectly (through parantela and legitimate interactions)
produce some support for the programs and th e continued existence of the specific
bureaucratic agency involved. These are then two political actors who need each other in
order to carry out their respective purposes in as efficient a manner as possible. Both
operate on the fringes of political respectabili ty and need friends in their battles. The
symbiosis that tends to develop between bureaucracy and pressure group is readily
explicable in terms of these needs for legitimation and support. The major
Page 211
question that remains is whether this symbiotic relationship is to be accepted – as with
the legitimate groups – or forced furt her into the gray areas of politics.
We turn now to bureaucratic politics, more directly con cerned with power and policy
than access. These are the politics arising fr om the public bureaucracy dealing with other
formal institutions of government. Each of these institutional actors has access to the
arenas of political connflict, and their positions in those arenas are more secure than that
of the bureaucracy. Here, then, the bureaucracy must engage in s ubstantially different
types of political behaviors, both to preserve its autonomy as an organization and to have
an impact on public policy. In some way the role of the bureaucr acy becomes that of
gaining access to legitimate pol itical power, just as the in terest groups had to do when
dealing with it.
Notes
1 This separation is usually attributed to American theorists of public administration
such as Woodrow Wilson, ‘‘The Study of Administration,’’ Political Science
Quarterly , 2 (1887), 209–13. The doctrine was el aborated by scholars such as
Willoughby, Pfiffner and Goodnow. The attack on it was led by Paul Appleby in
Policy and Administration (University, AL: University of Alabama Press, 1949).
2 See, for example, Colin Campbell and B. Guy Peters, ‘‘The Politics/Administration
Dichotomy: Death or Merely Change?,’’ Governance , 1 (1988), 79–100.
3 Richard Rose, ‘‘Giving Direction to Perman ent Officials,’’ in Jan-Erik Lane, ed.,
Bureaucracy and Public Choice (London: Sage, 1987).
4 Despite the general public’s dislike for bur eaucracy, the institution often does have
respect for being apolitical or for being the repository of tech nical expertise. See
Morris P. Fiorina, ‘‘Flagella ting the Federal Bureaucracy,’’ Society , 20 (1983), 66–
74. More recent attempts to make bureaucraci es more closely aligned with political
authority may therefore threaten their capaci ty to legitimate. See Joel F. Handler,
Down From Bureaucracy (Princeton: Princeton Un iversity Press, 1996).
5 B. Guy Peters, ‘‘Public Bureaucracy and P ublic Policy,’’ in Douglas E. Ashford, ed.,
History and Context in the Study of Comparative Public Policy (Pittsburgh:
University of Pittsburgh Press, 1992).
6 For a classic discussion of this phenomenon, see Karl Mannheim, Ideology and
Utopia (New York: Harcourt, Brace and World, 1946), pp. 105ff.
7 See Cornelius Kerwin, Rulemaking , 2nd edn (Washington, DC: CQ Press, 1999);

33Robert Baldwin, Rules and Government (Oxford: Oxford University Press, 1996).
8 Lawrence B. Mohr, Explaining Organizational Behavior (San Francisco: Jossey-
Bass, 1982).
9 See, for example, Colin Campbell, ‘‘Review Article: The Political Roles of Senior
Government Officials in Advanced Democracies,’’ British Journal of Political
Science , 18 (1988), 243–72.
10 Policy advice then becomes a crucial aspect of governance in modern societies. See
William Plowden, Advising the Rulers (Oxford: Blackwells, 1987) ; B. Guy Peters and
Anthony Barker, eds, Advising West European Governments (Pittsburgh: University
of Pittsburgh Press, 1993).
11 See Robert D. Putnam, ‘‘Po litical Attitudes of Senior Civil Servants in Britain,
Germany and Italy,’’ in Mattei Dogan, ed., The Mandarins of Western Europe (New
York: Halsted, 1975); B. Guy Peters, ‘‘Politic ians and Bureaucrats in the Politics of
Policymaking,’’ in Jan-Erik Lane, ed., Bureaucracy and Public Choice (London:
Sage, 1987); Donald J. Savoie, Governing From the Centre (Toronto: University of
Toronto Press, 1999).
Page 212
12 See Eduardo Zapico Goni, ‘‘Many Reform s, Little Learning: Budgeting, Auditing
and Evaluation in Spain,’’ in Andrew Gra y, Bill Jenkins and Bob Segsworth, eds,
Budgeting, Auditing and Eva luation: Functions and In tegration in Seven
Governments (New Brunswick, NJ: Transa ction, 1993); Dirk-Jan Kraan, Budgetary
Decisions: A Public Choice Approach (Cambridge: Cambridge University press,
1996).
13 See Chapters 7 and 8.
14 See Graham Wilson, Interest Groups (Oxford: Basil Blackwell, 1991); Jeremy
Richardson, ed., Pressure Groups (Oxford: Oxford University Press, 1993); David S.
Goodman, Groups and Politics in the People’s Republic of China (Armonk, NY: M.
E. Sharpe, 1984); Clive S. Thomas, First World Interest Groups (Westport, CT:
Greenwood Press, 1994).
15 In this case the advertising is directed as much to the members of the organization as
to the general public. Public service advert ising, e.g. about drug abuse, is targeted
more generally.
16 See Johan P. Olsen, Organized Democracy (Oslo: Universitets forlaget, 1983); Frank
L. Wilson, Interest Group Politics in France (Cambridge: Cambridge University
Press, 1987); Marcia Drezon-Tepler, Interest Groups and Political Change in Israel
(Albany: State University of New York Press, 1990).
17 See, for example, Olof Ruin, ‘‘Sweden in the 1970s: Police-making [sic] Becomes
Difficult,’’ in Jeremy Richardson, ed., Policy Styles in Western Europe (London:
George Allen and Unwin, 1982).
18 Edward C. Page, Bureaucratic Authority and Political Power , 2nd. edn (Brighton:
Wheatsheaf, 1992).
19 Grant Jordan, ‘‘Iron Triangles, Woolly Cor poratism and Elastic Nets: Images of the
Policy Process,’’ Journal of Public Policy , 1 (1981), 95–123; Phillipe C. Schmitter
and Gerhard Lehmbruch, Trends Toward Corporatist Intermediation (London: Sage,

341982).
20 Phillipe C. Schmitter, ‘‘Still A Century of Corporatism?,’’ Review of Politics , 36
(1974), 93.
21 Ibid.
22 Schmitter’s more restrictive view of corporatism may be a function of his initial work
in Latin American and Iberian corporatism.
23 Gerhard Lehmbruch, ‘‘Liberalism and Party Government,’’ Comparative Political
Studies , 10 (1977), 91–126; Martin O. Heisler, ‘‘Corporate Pluralism Revisited:
Where is the Theory?,’’ Scandinavian Political Studies , 2 (1979), 277–97.
24 Alan Cawson, Organized Interests and the State: Studies in Meso-Corporatism
(London: Sage, 1985).
25 Hugh Heclo and Henrick Madsen, Politics and Poli cy in Sweden (Philadelphia:
Temple University Press, 1987); H. Adam, Die konzierte Aktion in der
Bundesrepublik (Cologne: Bund Verlag, 1972).
26 See below, pp. 196–200.
27 Harmon Zeigler, Pluralism, Corporatism and Confuc ianism: Political Associations
and Connflict Regulation in the United States, Europe and Taiwan (Philadelphia:
Temple University Press, 1988), p. 164.
28 Thomas T. Mackie and Brian W. Hogwood, Unlocking the Cabinet: Cabinet
Structures in Comparative Perspective (London: Sage, 1985).
29 Donald F. Kettl, Government by Proxy: (Mis?)Managing the Federal Government
(Washington, DC: CQ Press, 1988); Brin ton Milward, ‘‘‘Symposium on the Hollow
State’, Capacity, Control and Performance in Interorganization Settings,’’ Journal of
Public Administration Research and Theory , 6 (1996), 193–5.
30 Donald F. Kettl, Sharing Power: Public Governance and Private Markets
(Washington, DC: The Brookings Institution, 1993); Jon Pierre, Partnerships in
Urban Governance (London: Macmillan, 1999).
31 Brigitta Nedelman and Kurt G. Meier, ‘‘Theories of Contemporary Corporatism:
Static or Dynamic?,’’ in Phillipe C. Schmitter and Gerhard Lehmbruch, Trends
Toward Corporatist Intermediation (London: Sage, 1982); Peter Gerlich, ‘‘A
Farewell to Corporatism,’’ West European Politics , 15 (1992), 132–46.
Page 213
32 Walter Kickert, Erik-Hans Klijn and Joop Kooppenjan, Managing Complex Networks
(London: Sage, 1997); Edward Laumann and David Knoke, The Organizational State
(Madison, WI: University of Wiscon sin Press, 1987); Olof Petersson, Maktens
Natverk (Stockholm: Carlssons, 1989).
33 Ernst B. Haas, When Knowledge is Power (Berkeley, CA: Univer sity of California
Press, 1990); Michael M. Atkinson and William D. Coleman, ‘‘Policy Networks,
Policy Communities and the Problems of Governance,’’ Governance , 5 (1992), 154–
80.
34 Sonia Mazey and Jeremy J. Richardson, Lobbying in the European Community
(Oxford: Oxford University Press, 1993).
35 Yves Meny, ‘‘The National and International Context of French Policy
Communities,’’ Political Studies , 37 (1989), 387–99.

3536 Some of the structuralist lit erature in sociology is atte mpting to produce just such
predictions but the results are as yet not promising. See David Knoke, Political
Networks: The Structural Pe rspective (Cambridge: Cambridge University Press,
1990).
37 Olof Petersson, Maktens Natverk (Stockholm: Carlssons, 1989).
38 Tom Christensen and Morten Egeber g, ‘‘Noen trekk ved forholdet mellom
organisasjionene go den offentlige forvaltningen,’’ in Christensen and Egeberg, eds,
Forvalt-ningskunnskap (Oslo: Tano, 1993).
39 Knoke and Laumann, The Organizational State .
40 These sectoral differences have been ar gued to be a more appropriate basis of
comparison than country. See Gary P. Freema n, ‘‘National Styles and Policy Sectors:
Explaining Structured Variation,’’ Journal of Public Policy , 5 (1985), 467–96.
41 Petersson, Maktens Natverk , p. 172.
42 Paul A. Sabatier, ‘‘An Advocacy Coalition Model for Policy Change and the Role of
Policy-oriented Learning Therein,’’ Policy Sciences , 21 (1988), 129–68.
43 The one case in which the corporatist mode l is more appropriate is the tripartite
bargaining over economic policy during the 1960s and 1970s. There was broad agreement on the economic model that be st described the relevant issues
(Keynesianism) and it was also clear that there were just the three relevant actors.
44 See Keith Dowding, ‘‘Model or Metaphor: A Cr itical Review of the Policy Network
Approach, Political Studies , 42 (1995), 136–58.
45 Walter J. M. Kickert, Erik-Hans Klijn and Joop F. M. Koopenjaan, Managing
Complex Networks (London: Sage, 1997).
46 Michele Micheletti, ‘‘Interesseorganis ationerna – i gar, i dag, i morgon,’’
Statsvetenskaplig tidskrift , 91 (1988), 41–54.
47 Nevil Johnson, State and Government in the Federal Republic of Germany , 2nd edn
(Oxford: Pergamon, 1983), pp. 105–43.
48 Neil Elder, Alastair H. Thomas and David Arter, The Consensual Democracies
(Oxford: Martin Robertson, 1982), pp. 105–43; Tom Christensen and Morten
Egeberg, ‘‘Noen trekk ved forholdet mello m organisasjoene og direktorat,’’ in
Christensen and Egeberg, eds, Forvaltningskunnskap (Oslo, Tano: 1997).
49 Lars Norby Johansen and Ole P. Kriste nsen, ‘‘Corporatist Tr aits in Denmark, 1946–
76,’’ in Gerhard Lehmbruch and Phillipe C. Schmitter, eds, Patterns of Corporatist
Policy-Making (Beverly Hills, CA: Sage, 1982), pp. 199–203.
50 Jorgen Gronnegard Christense n and Peter Munk Christiansen, Forvaltning og
omgivelser (Herning: Systime, 1992), p. 74.
51 Kvavik, Interest Groups in Norwegian Politics ; Micheletti argues that, in Sweden, the
groups have become more partisan and less expert. See Michele Micheletti, ‘‘Interest
Groups in Transition and Cris is,’’ in Clive S. Thomas, First World Interest Groups
(Westport, CT: Greenwood Press, 1993).
52 Yves Weber, L’administration consultative (Paris: Librarie Generale du Droit et
Jurisprudence, 1968); Frank L. Wilson, Interest Group Politics in France
(Cambridge: Cambridge University Press, 1987).
53 Office of the Federal Register, Federal Advisory Committees (Washington, DC:
Government Printing Office, 1998).

36Page 214
54 Jeremy J. Richardson, ‘‘Interest Group Behaviour in Britain: Continuity and
Change,’’ in Richardson, Pressure Groups .
55 Jakob Buksti, ‘‘Interest Groups in Denmark,’’ in Richardson, Pressure Groups .
56 Robert B. Kvavik, Interest Groups in Norwegian Politics (Oslo: Universitetsforlaget,
1976), pp. 68–73.
57 Klaus von Beyme, ‘‘West Germany and the New Germany: Centralization,
Expanding Pluralism, and New Cha llenges,’’ in Clive S. Thomas, First World
Interest Groups (Westport, CT: Greenwood, 1993), pp. 166–7.
58 See Jan Van den Bulck, ‘‘Pillars and Politic s: Neo-corporatism and Policy Networks
in Belgium,’’ West European Politics , 15 (1992), 35–55.
59 Jonas Pontusson, ‘‘Labor, Corporatism, a nd Industrial Policy: The Swedish Case,’’
Comparative Politics , 23 (1991), 163–79.
60 See M. P. C. M. van Schendelen and R.J. Jackson, eds, The Politicisation of Business
in Western Europe (London: Croom Helm, 1987): A 1999 report by the Audit
Commission in the Netherlands still identifi es dozens of regulatory bodies involving
interest groups as major players.
61 Lester M. Salamon, ‘‘Rethinking Public Management: Third Party Government and
the Changing Forms of Public Action,’’ Public Policy , 29 (1981), 255–75.
62 Much of the recent literature has been challenging the idea of autonomy and the
comparative strength of states , but these notions that stat es have more or less power
vis-á-vis organized interests is still an importa nt means of understa nding their actions.
See Michael M. Atkinson and William D. Coleman, ‘‘Strong States and Weak States:
Sectoral Policy Networks in Advanced Capitalist Nations,’’ British Journal of
Political Science , 19 (1989), 47–67; see also Jon Pierre and B. Guy Peters,
Governance, the State and Public Policy (Basingstoke: Macmillan, 2000).
63 See Martin O. Heisler with Robert B. Kv avik, ‘‘Patterns of European Politics: The
‘European Politics Model’,’’ in M. Heisler, ed., Politics in Europe (New York: David
McKay, 1974); Michael Saward, ‘‘Cooptation and Power: Who Gets What from
Formal Incorporation,’’ Political Studies , 38 (1990), pp. 588–602.
64 The classic case is that of political parties. See Robert Michels, Political Parties: A
Sociological Study of the Oligarchical Tendencies in Modern Democracies (London:
Collier-Macmillan, 1962).
65 Gabriel A. Almond and G. Bingham Powell, Comparative Politics: A Developmental
Analysis (Boston: Little, Brown, 1966), pp. 77–8.
66 Johansen and Kristensen, op. cit. , pp. 200–3.
67 Jean Tournon, ‘‘Les pressions publiques: les pouvoirs publics sont le premier lobby
de France,’’ in Maur ice Duverger, ed., Jean Meynaud ou l’utopie revisitee (Lausanne:
Universite de Lausanne, 1988).
68 R. A. W. Rhodes, Beyond Westminster and Whitehall (London: Unwin Hyman, 1988).
69 Robert B. Kvavik, ‘‘Interest Groups in a Cooptive Political System,’’ op. cit. , pp.
111–12.
70 Stein Rokkan, ‘‘Norway: Numeri cal Democracy and Corporat e Pluralism,’’ in Robert
A. Dahl, ed., Political Oppositions in Western Democracies (New Haven, CT: Yale
University Press, 1968), p. 107.

3771 Theodore J. Lowi, ‘‘The Public Phil osophy: Interest Group Liberalism,’’ American
Political Science Review , 61 (1967), 19; Robert H. Sa lisbury and John Heinz, ‘‘The
Analysis of Public Policies: A Search fo r Theories and Roles,’’ in Austin Ranney,
ed., Political Science and Public Policy (Chicago: Markham, 1968).
72 Kvavik, ‘‘Interest Groups in a Cooptive Political System,’’ op. cit. , 113.
73 Johan P. Olsen, Organized Democracy (Bergen: Universitetsforlaget, 1986).
74 Nils Elvander, Interessorganisationerna i dagens Sverige (Lund: CWK Gleerup,
1969); Lennart Lundquist, Forvaltning och demokrati (Stockholm: Norstedts, 1991),
pp. 154ff.
75 Heisler with Kvavik, op. cit.
76 See Leonardo Parri, ‘‘Neo-corporatist A rrangements, ‘Konkordanz’, and Direct
Democracy: The Swiss Experience,’’ in Ilja Scholten, Political Stability and Neo-
Corporatism (London: Sage, 1987).
Page 215
77 For a brief discussion see Salisbury and He inz, ‘‘Analysis of Public Policy,’’ op. cit. ,
55–9.
78 Peter J. May and Soren Winter, ‘‘Re gulatory Enforcement and Compliance:
Examining Danish Agro-Environmental Policy, Journal of Public Policy Analysis
and Management , 18 (1999), 625–51.
79 Robert Alford, Health Care Politics (Chicago: University of Chicago Press, 1975);
Ellen M. Immergut, ‘‘Institutions, Veto Points and Policy Results: A Comparative Analysis of Health Care,’’ Journal of Public Policy , 10 (1990), 391–416.
80 Donald Kettl, Government by Proxy , op. cit. ; Christopher Hood and Gunnar Folke
Schuppert, Delivering Services in Western Europe (London: Sage, 1988).
81 See Jorgen Hermansson, Torssten Svenss on and Per-Ola Oberg, ‘‘Vad blev det av
svenska korporativismen,’’ Politica , 29 (1997), 365–84.
82 E. A. Brett, ‘‘Adjustment and the State: The Problem of Administrative Reform,’’
IDS Bulletin , 19 (1988).
83 Suleiman, Politics, Power and Public Policy in France , op. cit. , pp. 338–9.
84 Wolfgang Wirth, ‘‘Control in Public Admi nistration: Plurality, Selectivity and
Redundancy,’’ in Franz-Xavier Kaufmann, Gi domenico Majone and Vincent Ostrom,
eds, Guidance, Control and Evaluat ion in the Public Sector (Berlin; deGruyter,
1986).
85 For Japan, see John Creighton Campbell, ‘‘B ureaucratic Primacy: Japanese Policy
Communities in an American Perspective,’’ Governance , 2 (1989), 5–22; Wayne
Moyer, Agricultural Policy Refo rm: Politics and Process in the EC and the USA
(Ames, Iowa: Iowa State University Press, 1990); Jeffrey Ira Herbst, State Politics in
Zimbabwe (Berkeley: University of California Press, 1990).
86 Theodore J. Lowi, The End of Liberalism , 2nd edn (New York: W. W. Norton, 1979);
John T. S. Keeler, The Politics of Neocorporatism in France (New York: Oxford
University Press, 1987); Ci nzia Dato Giurickovic, Senso dello Stato e anarchia delle
lobbies (Naples: Edizioni Scie ntifiche Italiane, 1990).
87 Hugh Heclo, ‘‘Issue Networks and the Executive Establishment,’’ in Anthony King,
ed., The New American Political System (Washington, DC: American Enterprise

38Institute, 1978); Jack L. Walker, Mobilizing Interest Groups in America: Patrons,
Professions and Social Movements (Ann Arbor: University of Michigan Press, 1991).
88 Robert H. Salisbury, ‘‘The Paradox of Inte rest Groups in Washington – More Groups,
Less Clout,’’ in Anthony King, ed., The New American Political System , 2nd edn
(Washington, DC: AEI Press, 1990).
89 Olsen, Organized Democracy , pp. 157ff; Ruin, ‘‘Sweden in the 1970s,’’ op. cit.
90 Jack Vowles, ‘‘New Zealand: Captur e the State,’’ in Clive S. Thomas, First World
Interest Groups (Westport, CT: Greenwood Press, 1993).
91 LaPalombara, Interest Group Politics , pp. 272–4.
92 Theodore J. Lowi, The End of Liberalism , 2nd edn (New York: W. W. Norton, 1979).
93 Grant McConnell, Private Power and American Democracy (New York: Knopf,
1966), p. 287.
94 William Lilley III and James C. Miller III, ‘‘The New Social Regulation,’’ The Public
Interest , 4 (1977), 49–61.
95 Jonathan R. Macey, ‘‘Organizational Design and Political Contro l of Administrative
Agencies,’’ Journal of Law, Economics and Organization , 8 (1992), 93–110; Mathew
D. McCubbins, Roger G. Noll and Barry R. Weingast, ‘‘Structure and Process,
Politics and Policy: Administrative Arra ngements and the Political Control of
Agencies,’’ Virginia Law Review , 75 (1989), 431–82.
96 Lowi, ‘‘The Public Philosophy…,’’ p. 19.
97 Peter Aucoin, ‘‘Organizational Change in the Machinery of Canadian Government:
From Rational Management to Brokerage Politics,’’ Canadian Journal of Political
Science , 14 (1986), 3–27; Donald Savoie, Governing from the Centre (Toronto:
University of Toronto Press, 1999).
98 See Andre Blais and Stephane Dion, eds, The Budget-maximizing Bureaucrat
(Pittsburgh: University of Pittsburgh Press, 1992).
Page 216
99 Von Beyme, for example, found that there was a concentration of legislators with
interest group connections on committees in the Bundestag that were concerned with
the same policy area. See Klaus von Beyme, ‘‘West Germany and the New Germany,’’ 170.
100 See B. Guy Peters, American Public Policy , 3rd edn (Chatham, NJ: Chatham House,
1993), p. 71.
101 For an example of a particularly stro ng government lobby see Ulrich Kloti,
‘‘Political Ideals, Financial Interests a nd Intergovernmental Relations: New Aspects
of Swiss Federalism,’’ Government and Opposition , 23 (1988), 91–102. The
Intergovernmental lobby can be important ev en in centralized regimes, e.g. Pierre
Gremion, La Pouvoir Peripherique (Paris: Seuil, 1976).
102 Salisbury and Heinz, ‘‘Analysi s of Public Policy,’’ p. 48.
103 See B. Guy Peters, ‘‘Managing Horiz ontal Government: The Politics of
Coordination,’’ Public Administration , 76 (1998), 295–312.
104 LaPalombara, Interest Group Politics , pp. 306–7.
105 This occurs even in competitive political systems. For example, the major political
parties in Germany each has a party foundati on that functions as a think-tank for the

39party as well as funding some academics that may be beneficial to the party. The
Adam Smith Institute in Britain was a sour ce of advice and advisors for the Thatcher
government, and the Heritage foundation serv ed something of the same function for
the Reagan administration. The close links between labor movements and social
democratic parties across much of the world is another example.
106 LaPalombara, Interest Group Politics , pp. 308–15; see Paul Furlong, Modern Italy
(London: Routledge, 1994), pp. 149–51.
107 Philip M. Williams and Martin Harrison, Politics and Society in De Gaulle’s France
(Garden City, NY: Doubleday, 1972), pp. 339–42.
108 Sarah Sokoloff, ‘‘Socialism and the Farmers,’’ in Philip G. Cerny and Martin A,
Schain, eds, Socialism, The State and Public Policy (London: Frances Pinter, 1985).
109 On cohabitation see Maurice Duverger, La Cohabitation des Francais (Paris:
Presses Universitaires de France, 1987).
110 See, for example, Friedrich-Ebert Stiftung, Organisationen und Verbande in der
DDR (Bonn: Neue Gesellschaft Verlag, 1987).
111 See the relevant chapters in Kay Lawson, ed., How Political Parties Work:
Perspectives from Within (Westport, CT: Praeger, 1994)
112 Muller-Rommel, for example, found that the major parties in (then) West Germany
all had numerous members of the Bundestag with direct ties to interest groups. See
F. Muller-Rommel, ‘‘Interessengruppenvert retung im Deutschen Bundestag,’’ in
Uwe Thaysen et al., US-Kongress und Deutscher Bundestag (Oplanden:
Westdeutscher Verlag, 1989).
113 Marilyn Gittel, Limits of Citizen Participati on: The Decline of Community
Organizations (Beverly Hills, CA: Sage, 1980).
114 Seah Chee Meow, ‘‘Parapolitical Organi zations,’’ in Jon T. S. Quah, ed.,
Government and Politics of Singapore (Singapore: Oxford University Press, 1985),
pp. 173–94.
115 Charles W. Anderson, The Political Economy of Modern Spain (Madison:
University of Wisconsin Press, 1970), pp. 30–4.
116 Myron Weiner, The Politics of Scarcity (Chicago: University of Chicago Press,
1962), p. 78. This quote is perhaps more illust rative of the theoretical case than the
relationship between the Congress Party and unions in contemporary India.
117 Anderson, The Political Economy of Modern Spain , p. 69. On the changes in the
relationship between unions and government si nce the Francoist regimes, see Victor
Perez-Diaz, El Retorno de la sociedad civil (Madrid: Institu to de Estudios
Economicos, 1987).
118 Colin Crouch, ‘‘The Peculiar Relationship: The Party and the Unions,’’ in Dennis
Kavanagh, ed., The Politics of the Labour Party (London: Allen and Unwin, 1982).
119 Lowell Dittmer, ‘‘Comparative Communist Political Culture,’’ Studies in
Comparative Communism , 16 (1983), 17.
Page 217
120 The term ‘‘canteen’’ to describe this part icularistic relationship between party and
interest groups was developed by Fred W. Riggs, Administration in Developing
Countries: The Theory of Prismatic Society (Boston: Houghton Mifnflin, 1964), pp.

40105–9.
121 Schmitter, ‘‘Still the Centur y of Corporatism?,’’ p. 93.
122 Victor C. Falkenheim, Citizens and Groups in Contemporary China (Ann Arbor:
University of Michigan Press, 1987).
123 LaPalombara, Interest Group Politics , p. 265.
124 Ezra N. Suleiman, Politics, Power , pp. 340–6; see also his Private Power and
Centralization in France: The Notaires and the State (Princeton: Princeton
University Press, 1987).
125 This connflicts rather obviously with Manc ur Olson’s view that individuals will tend
not to join groups if they receive no tangibl e benefits from that membership. See his
The Logic of Collective Action (Cambridge, MA: Harvard University Press, 1968).
126 J. Paluski, Social Movements: The Po litics of Moral Protest (Melbourne: Longman
Cheshire, 1991); Thomas R. Rochon, Mobilizing for Peace: The Antinuclear
Movements in Western Europe (Princeton: Princeton University Press, 1993); Chris
A. Rootes, ‘‘The New Politics and the New Social Movements: Accounting for
British Exceptionalism,’’ European Journal of Political Research , 22 (1992), 171–
91; Sidney Tarrow, Power in Movement: Social Movements and Contentious
Politics , 2nd edn (Cambridge: Cambridge University Press, 1998).
127 William P. Browne, Private Interests, Public Policy and American Agriculture
(Lawrence, KS: University Press of Kansas, 1988).
128 Michelle Salvati, ‘‘May 1968 and the Hot Autumn of 1969: The Responses of Two
Ruling Classes,’’ in Suzanne Berger, ed., Organizing Interests in Western Europe
(Cambridge: Cambridge University Press, 1981).
129 See Peter Snow and Luigi Manzetti, Political Forces in Argentina (Westport, CT:
Praeger, 1993).
130 Martin J. Smith, ‘‘Pluralism, Reformed Pl uralism and Neopluralism: The Role of
Pressure Groups in Policy-Making,’’ Political Studies , 38 (1990), 302–22.
131 Lewis Minkin, The Contentious Alliance: Trade Unions and the Labour Party
(Edinburgh: University of Edinburgh Press, 1991).
132 Tarrow, loc. cit .
133 Sir John Hoskyns, ‘‘Whitehall and West minster: An Outsider’s View,’’ Fiscal
Studies , 3 (1982), 162–72; Revue francaise d’administration publique , Special issue
on politicization, April–May, 1998.
134 Kwame Nkrumah, I Speak for Freedom: A Statement of African Ideology (New
York: Praeger, 1961), p. 173.
135 William Tordoff, Government and Politics in Africa (Bloomington: Indiana
University Press, 1993), Chapter 10.
136 Hoskyns, ‘‘Whitehall and Westminster…’’.
137 Richard Rose, ‘‘Loyalty, Voice or Exit?: Mr s. Thatcher’s Challenge in the Civil
Service,’’ in Thomas Ellwein et al., Yearbook on Government and Public
Administration , 1987/88 (Baden-Baden: Nomos, 1989).
138 Rolf H. W. Theen, ‘‘Party and Bure aucracy,’’ in Gordon B. Smith, ed., Public
Policy and Administration in the Soviet Union (New York: Praeger, 1980).