Community based organization and bureaucratic resistance

Material Information

Community based organization and bureaucratic resistance four case studies in mexican municipalities
Arellano Gault, David
Place of Publication:
Denver, Colo.
University of Colorado Denver
Publication Date:
Physical Description:
1 online resource (15-296 pages) : ;

Thesis/Dissertation Information

Doctorate ( Doctor of Philosophy)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
School of Public Affairs, CU Denver
Degree Disciplines:
Public administration


Subjects / Keywords:
Organización burocrática
Industrial organization
Conflict management
Libros electronicos.
non-fiction ( marcgt )
Libros electronicos


General Note:
School of Public Affairs
Statement of Responsibility:
David Arellano Gault.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
846690341 ( OCLC )
HD38.4 A679 2000eb ( lcc )
316 ( udc )

Full Text
David Arellano Gault
B.A. Universidad Tecnologica de Mexico, 1985
M.A. Centro de Investigation y Docencia
Economicas A.C., 1986
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Public Administration

This thesis for the Doctor of Philosophy-
degree by
David Arellano Gault
has been approved
Michael Cortes
Linda de Leon
Peter de Leon
E. Sam Overman

To my wife, Laura, and my children Ana Laura and David
Ricardo, with thanks for their support and love.

My thanks to the Consejo Nacional de Ciencia v Tecnologia
in Mexico for financial support for my studies at the
University of Colorado at Denver.
Thanks also to the Human Subjects Research Protection
Committee of the University of Colorado for reviewing the
research proposal and advising me regarding protection of
the people I interviewed during my research.
To my committee, for their patience and advice, many
thanks. I am specially indebted to the late Dr. Sam
Overman, a great professor and human being, someone whom
I will never forget.
To the Centro de Investigacion y Docencia Economicas A.C.
(CIDE), particularly to Prof. Enrique Cabrero, Dr. Carlos
Elizondo, Dr. Blanca Heredia, and Prof. Juan Pablo
Guerrero, my thanks for their trust and support toward
the development of my academic career.
Finally, thanks to all the people in diverse
municipalities and states who allowed me to conduct this
study. They taught me many things, and gave me confidence
that the people of Mexico are working hard and humanely
to develop a great future.

Arellano Gault, David (PhD., Public Administration)
Community-Based Organizations and Bureaucratic
Resistance: Four Case Studies in Mexican Municipalities
Thesis directed by Assistant Professor Michael E. Cortes
Since 1989, the Federal Government of Mexico has been
encouraging the formation of community-based
organizations (CBOs) to help public agencies implement
federally-funded, locally administered social programs.
Specifically, federal resources allocated to social
programs are often tied to a requirement that members of
local communities being served participate in program
implementation. Thus, community-based organizations are
expected to coordinate their efforts with local municipal
government bureaucracies. Coordination is not easy,
because bureaucracies and CBOs have different
organizational characteristics. Municipal or state public
agencies sometimes resist cooperating with community
organizations, and CBOs sometimes find it difficult to
understand the needs and requirements of public
This research examines relations between governmental
bureaucracies, which have to manage social programs, and
community-based organizations, involved with the
implementation of social programs. Where CBOs have
emerged as an option for dealing with complex social
problems requiring flexibility and participation, and
where governmental agencies retain a dominant role in the

control of resources for those projects, the impact on
the CBOs ranges between two extremes. At one extreme, the
CBO might collapse due to increased bureaucratization
imposed by government agencies. At the other extreme, the
CBO might be legitimized by the bureaucratic and
political establishment, allowing the survival of the CBO
as an autonomous and flexible organization. Between these
two extremes, several intermediate outcomes are also
This dissertation presents four case studies that
demonstrate underlying causal dynamics that help
determine various outcomes between the two extremes. The
four cases from rural central Mexico were selected from a
larger set of Mexican communities, to help control for
geographic and other extraneous variables. Determinants
of outcomes in the bureaucracy-CBO relationship
identified by this study include the objectives, norms,
and authority structures of CBOs. Moreover, this research
discovered an evolutionary cycle among CBOs, and two
different types of relationship between bureaucracies and
CBOs (depending on the origins of the CBO) Practical
implications for promoting successful relationships
between bureaucracies and CBOs are discussed, along with
implications for future research.
This abstract accurately represents the content of the
candidate's thesis. I recommend its publication.
Michael Eduardo Cort§s

1. Introduction ................................. 1
2. Bureaucracy as Organization: Uncertainty,
Conflict, and Power ........................... 10
2.1 Weber's Bureaucracy ........................... 11
2.1.1 The Ideal Type ................................ 12
2.1.2 Bureaucracy as a Form of
Organization .................................. 15
2.2 Problems with the Ideal Type .................. 21
2.2.1 The Controversy with Weber .................... 21
2.3 Modernity and Organizations: the
Iron Cage ..................................... 31
2.4 The Bureaucratic Phenomenon: Crozier's
View ......................................... 3 7
2.5 The Actor and the System: Power, Strategy,
and Conflict in Organizations ................. 41
2.6 Bureaucracy: the contemporary discussion .... 45
2.6.1 Budget Maximizing Bureaucrat .................. 46
2.6.2 Postbureaucratic Paradigm ..................... 53
2.6.3 Democracy and Bureaucratic Action ............. 60
2.7 Final Remarks: Bureaucracy as a Complex
Organization .................................. 65
3. Methodology. The Nature of Case Studies and the
Design of the Comparative Case Study
Framework ..................................... 67

3.1 The Inductive Method .......................... 67
3.1.1 Some Definitions .............................. 67
3.1.2 Pragmatism: Induction as Cognitive
Systematization ............................... 70
3.1.3 Falsification (Popper) ........................ 73
3.2 Case Study Design ............................. 79
3.2.1 Case Study and the Problem of Validity
and Reliability ............................... 79
3.2.2 Research Problems and Questions ............... 84
3.2.3 Basic Definitions ........................... 87
3.2.4 Proposition ................................... 90
3.2.5 Argument ...................................... 91
3.2.6 Key Concepts .................................. 94
3.2.7 Why Case Studies? ............................. 95
3.2.8 Unit of Analysis .............................. 97
3.2.9 Criteria for Selecting Cases .................. 97
3.2.10 Research Protocol ............................ 104
3.2.11 Interviewing: a General Framework ............ 106
3.2.12 The Research Process ......................... 108
4. Case Studies ................................. 112
4.1 Basic and General Framework for the Mexican
Municipality ................................. 112
4.1.1 PRONASOL (National Solidarity
Program) ..................................... 116
4.1.2 Bureaucratic Structure of
Ayuntamientos in Rural Municipalities
in Mexico .................................... 122

4.1.3 Organization of Case Studies ............... 127
4.2 Case Study: Zaragoza, Veracruz: Community
Organization that Achieves
Institutionalization ........................ 129
4.2.1 The General Context ......................... 129
4.2.2 General View of the Municipality ............ 131
4.2.3 Recent History of the Relationship .......... 136
4.2.4 The First Step: Transforming CDPZ to
a Democratic and Formal Community-
based Organization (1992-1995) 140
4.2.5 The Organic Structure and Functions
of CDPZ ..................................... 144
4.2.6 The Second Step: Planning as a Way of
Institutionalization (1995-1996) 145
4.2.7 Key Elements for Discussion ................. 14 9
4.3 Case Study: Cuquio, Jalisco: a Community
Organization Looking for
Institutionalization ........................ 154
4.3.1 The General Context ......................... 154
4.3.2 General View of the Municipality ............ 156
4.3.3 Recent History of the Relationship .......... 160
4.3.4 The Electoral Alliance and the Use of
Communal Forms of Organization .............. 164
4.3.5 The New Administration and CBO: the
Need for Systematization..................... 166
4.3.6 The CODEMUC ................................. 171
4.3.7 Problems for the future in the
Relationship Between CODEMUC and
Municipal Authorities ....................... 176

4.3.8 Key Elements for Discussion .................. 179
4.4 Case Study: Tierra Blanca, Guanajuato: An
Organization Induced From the State
Level ............................................... 182
4.4.1 The General Context .......................... 182
4.4.2 General View of the Municipality ............. 184
4.4.3 Recent History of the Relationship ........... 187
4.4.4 The COPLADEM in Tierra Blanca: an
Organization Without Clear Rules,
Norms, or Objectives ......................... 200
4.4.5 Key Elements for Discussion .................. 205
4.5 Case Study: Vanegas, San Luis Potosi: Using
Indirect Incentives for Behavioral
Change ....................................... 208
4.5.1 The General Context .......................... 208
4.5.2 General View of the Municipality ............. 210
4.5.3 Recent History of the Relationship ........... 214
4.5.4 PAZA Program: Assistance with New
Objectives ................................... 216
4.5.5 PAZA in the Municipality of Vanegas:
Assistance by Imposition Under
Desperate Conditions ......................... 224
4.5.6 The Problem of Developing Self-
sustainable Economic Capacities .............. 231
4.5.7 Key Elements for Discussion................... 237
5. Comparative Analysis and Conclusions ......... 240
5.1 Conditions for the Comparison ................ 240
5.2 Comparison and Cross-analysis ................ 250

5.2.1 Cross-analysis Regarding the Origins
of the CBO ................................... 250
5.2.2 Regarding the Evolution of the CBO ........... 254
5.2.3 Regarding the Nature of the
Bureaucratic Organization .................... 257
5.2.4 Regarding Authority Framework and
Objectives/Norms definitions ................. 258
5.2.5 Regarding the Nature and Evolution
of the Interorganizational Relation-
ship (Bureaucracy-CBO) ....................... 261
5.2.6 Regarding the Solutions Developed to
Resolve Conflicts ............................ 267
5.3 General Conclusions .......................... 271
5.3.1 Conclusions Regarding Proposition and
Theory Involved in the Research .............. 272
5.3.2 Conclusions Regarding CBO-Bureaucracy
Relations .................................... 276
5.4 The Theoretical Discussion: Flexible
Bureaucracy, Adaptation, and Organizational
Evolution .................................... 283
5.4.1 Bureaucracy: Flexibility and
Adaptation ................................... 285
5.4.2 The New Organization: a New Paradigm or
a Redefinition of Bureaucratic
Principles? .................................. 286
5.4.3 Collective Action, Inter-organizational
Relations, and Democratic Strategies for
Participation ................................ 289
Bibliography ........................................ 295

1. Introduction
The idea that public bureaucracies are inefficient
organizational structures, unable to achieve results when
confronted with complex societal problems is commonly-
accepted among scholars. As a result, analysts have tried
to address difficult collective problems using flexible
non-governmental organizational structures rather than
rigid and formalized ones. Such an organization must
have enough flexibility to be capable of finding
innovative ways to resolve complex problems involving
several actors while being efficient enough to achieve
In authoritarian political systems, like the one that has
historically prevailed in Mexico, the organizational
capacity to resolve social problems can rarely be found
outside the governmental bureaucratic network. In recent
decades, Mexico's bureaucratic organizing and planning
capacities have been the main sources of solutions for
social problems. However, there is currently an effort by
social groups and diverse parts of the Mexican government
to increase the role of community-based organizations
(CBOs) in dealing with community-level issues such as
infrastructure, health, and social development. Most of
these CBOs are being developed, at least in Mexico,
within traditional, informal grassroots-like
organizations attached to religious or ancestral
traditions, which commonly lack formal structure or legal

existence. Furthermore, they lack prior experience in
working with bureaucratic organizations. The
incorporation of these organizations into the decision-
making process has created contradictions, uncertainties,
and conflicts with the public bureaucracies.
Mexican CBOs, as independent and autonomous social or
political organizations, with strong community or
religious traditions, often emerge to solve simple social
problems (i.e., to organize the annual community
religious party). However, poor communities often face
difficult social problems (i.e., the construction of a
large drainage system or community roads) that require
resources and technologies they do not possess. In these
cases, it is usually beyond the capacity of CBOs to
maintain organized action for long periods of time. Thus,
it is usually difficult or impossible for these CBOs to
solve social problems without help from federal, state or
municipal bureaucracies.
However, these bureaucracies usually attach several
administrative conditions in return for help. Basically
bureaucracies expect CBOs to endorse a clear definition
of responsible, transparent mechanisms for the use of
resources, and clear objectives and procedures to act.
Often, these requirements are so rigid that the autonomy
and independence of CBOs are restricted in such a way
that their ability in involve the community and innovate
is jeopardized. Moreover, guiding CBOs toward the
adoption of more formal organizational structures can be
a conflictual and complex process, usually requiring time

and resources bureaucracies do not want to commit.
The capacity for flexible action usually allows CBOs to
pursue diverse objectives, using various approaches to
achieve results under ambiguous conditions. Also, the
informal structure of CBOs permits the participation of
different actors in diverse authority frameworks. These
elements create several problems for the public
bureaucratic organizations, the most important of v/hich
is uncertainty regarding outcomes and accountability. And
uncertainty is exactly what CBOs amplify when they
confront difficult social and political issues through
informal, traditional and/or religious forms.
Greater centralized or bureaucratic involvement with
CBOs often leads to increased resistance by public
agencies to the CBOs' style of organization. However,
good relations with the CBOs are necessary, because the
informal organization of the CBOs, their ability to
resist developing more formalized structures, and their
capacity for social mobilization, make it difficult for
government Bureaucrats to direct social and political
In turn, survival and development of CBOs depend in an
important way upon their improving capacity to understand
the needs of public bureaucracies, overcome bureaucratic
resistance, and maintain their unique ability to address
community problems with flexible and ambiguous
organizational structures.

The aim of the following research is to examine how
public bureaucracies and CBOs have confronted this
dilemma; that is, how have the problems caused by
administrative and structural differences between
governmental bureaucracies and CBOs been worked out in
practice. Particularly, I concentrate on those CBOs that
have addressed or attempted to address important
community problems within a transitional but still
centralized political system like the one in Mexico.
The thesis of this dissertation is that, in order for the
CBOs to survive bureaucratic resistance, they must
develop the capacity to overcome it, by learning how to
negotiate with government bureaucracies and by
transforming their form of organization. The goal of this
improved capacity for negotiation and transformation is
to reduce uncertainty among bureaucracies, while
maintaining some degree of autonomy and flexibility ir
the CBO.
However, there are dangers involved in such
transformation. On the one hand, it is possible that this
"clash" between CBO and governmental agencies can result
in the disappearance of the CBO due to increased
bureaucratic rigidity and domination. On the other hand,
the result may be increased legitimization, in the eyes
of bureaucracy, of the CO form of organization. It is
possible that several other outcomes occur within this
The basic questions that arise from this examination of

Che possible outcomes of CBO-bureaucracy relationship
- Which factors appear to cause collapse of CBOs due to
increased government rigidity and bureaucratization?
Which factors lead to legitimization and survival of
- What other alternatives within the continuum exist, and
how are they generated?
This dissertation uses a multiple case study methodology
to generate plausible interpretations of the phenomena
described, in order to further understand this important
part of the modernization process of a country like
Mexico. Four case studies of CBOs-bureaucracy
relationships in four rural municipalities are developed
through this research. The general idea is to try to
maintain a high level of comparative capacity in order to
develop useful answers to the above questions.
In these case studies, I observe the crisis caused by
bureaucratic pressure to change the organizational
structure of the CBOs, in particularly the CBOs'
definition of rules, norms, and authority. As mentioned
above, this pressure is caused by bureaucratic
intolerance of the uncertainty caused by the difficulty
of controlling, managing, and supervising flexible
strategies and participant framework of actors. I then
link these factors to an outcome of transformation
efforts along a continuum that ranges from "collapse" to
"legitimization" of the CBO.

The thesis for this dissertation can be stated formally
as follows.
The tools (ambiguous and flexible objectives, unclear
norms and elastic authority framework) used by CBOs to
deal with complex social problems create resistance from
bureaucratic agencies when they have to be "co-
responsible" for the administration of a particular
government-sponsored social program, because the CBOs'
flexibility increases the uncertainty of the outcome. The
bureaucratic agency then struggles and pushes to increase
control over resources and strategies through the use of
bureaucratic rules (accountability, expertise, and
central authority) by the CBO. If the CBO fails to
negotiate, and reduce or resolve with new organizational
structures and values the crisis generated by the
uncertainty associated with the use of their traditional
tools, we cam expect the public bureaucracy to impose
strict norms, objectives, and authority frameworks,
increasing the rigidity of the CBOs' response to
community problems. The bureaucracy cam use
administrative and political regulations to force the
bureaucratization or disappearamce of the CBO. If the CBO
formulates am acceptable alternative which reduces,
controls or resolves the crisis generated by bureaucratic
intoleramce of uncertainty, the organization has a strong
chance of becoming a legitimate institution in the eyes
of the government bureaucracy, while maintaining the
autonomy it needs to maintain its flexibility and its
particular organizational structures.

The research is divided in four additional chapters. In
Chapter 2 I develop the discussion of bureaucracy as a
form of organization. The idea here is to find the basic
characteristics of the bureaucratic form of organization
in order to understand its need to reduce uncertainty.
Uncertainty, it is found, is a basic preoccupation for
bureaucracies. Moreover, we find a paradox in
bureaucratic behavior: although rules, procedures, and
clear authority frameworks are the basic instruments to
reduce uncertainty, they can be so rigid that they
reduce bureaucratic capacity for dealing with complex and
dynamic situations. According to the literature reviewed,
a split then takes place within the bureaucratic
hierarchy: bureaucratic agents react by finding some
"open" spaces to use discretion and to negotiate certain
values and resources in order to successfully deal with
the issue at hand, but this increased use of discretion
pushes bureaucratic authorities to impose more rigid
rules and procedures in order to control those "open"
spaces that have created new uncertainties that need to
be controlled (in the eyes of authority). This literature
review allows conceptualization of bureaucratic
structure as a complex form of organization that tries to
reduce uncertainty through formal procedures and
hierarchies of authority. However, rules and authority
can be so rigid that bureaucrats develop a capacity for
using discretion and negotiation in order to deal
successfully with complex problems, generating pressures
from bureaucratic authorities to reduce these
discretionary capacities, in the form of increased
bureaucratic procedures, rules, and fixed authority

With this interpretative framework, in the context of
bureaucracy-CBO collaboration and coordination in the
development of social programs, I argue that when CBOs
utilize flexible organizational structures, they are
incrementing uncertainties for bureaucracies. In this
sense, CBOs probably will face pressures from bureaucracy
to define more rigid procedures and authority frameworks.
However, given the fact that bureaucracies are also
complex organizations that have "open" spaces for
negotiation and discretion, a negotiation process is
possible in order to define these more formalized rules
and values while still maintaining a certain degree of
flexibility. In other words, CBOs can negotiate with
bureaucracies in order to maintain their flexible
organizational structure and at the same time to reduce
uncertainty for the bureaucracy.
In Chapter 3, I develop the methodological framework for
the research. The approach is to use multiple case
studies in order to apply a comparative analysis. The
first argument of this chapter is that the case study
uses inductive methods of analysis. Inductive methods
rely on systematicity and plausible interpretations, in
order to increase the probability that the factors and
causalities observed are real. Case studies are a special
application of inductive methods, used when the
researcher cannot control all contextual variables and
behaviors. More specifically, case studies are used when
the researcher wants to observe a complex setting to

answer questions like "why" and "how" in terms of
multiple and interrelated causalities. This chapter also
defines the research proposition and the operative
concepts used through out the research. A two-by-two
matrix is created in order to determine the
characteristics of each case studied using the two key
concepts defined to observe the CBO-bureaucracy
relationship: authority framework and objectives/norms.
Chapter 4 is divided in two parts. The first part
explains how Mexican municipal governments work, and
discusses their relationship with the state and federal
governments. This discussion helps place the study in its
political and administrative context. The second part
consists of the four case studies of the municipalities
chosen for this dissertation. Each case study follows a
basic structure: contextual data regarding the
municipality and the case, history of the conflict, and
negotiation of new procedures and values.
Chapter 5 develops the conclusions of the dissertation.
First, the limits and characteristics of the comparative
analysis are discussed. Then the comparative analysis is
developed, with the help of several tables to clarify
the cross-analysis. The last part develops three groups
of conclusions: (1) conclusions regarding the research
questions and proposition, (2) unexpected conclusions,
and (3) conclusions regarding theory review, plus a final
set of observations and suggestions for future research.

2. Bureaucracy as Organization: Uncertainty, Conflict,
and Power
This chapter has one general objective: to review the
concept of bureaucracy as a form of organization, in
other words, to discuss its organizational dynamics and
contradictions. To show that bureaucracy is more than a
static structure within the Weberian ideal type, and that
it remains a dynamic and powerful form of organization,
is an important sub-objective of this chapter. From this
discussion, this dissertation develops research
propositions and key concepts.
In particular, in order to support research propositions
and concepts in the following chapters, this chapter
- show the dynamics of the bureaucratic form of
- discuss the current dilemmas facing the bureaucracy,
- discuss the main alternatives developed in
organizational studies with reference to the extreme
rigidity of the bureaucratic form of organization,
- study the argument that gives bureaucracy, as a form of
organization, the maximum capacity for efficiency when
compared with other forms of organization,
- review theories related to bureaucratic resistance and
uncertainty caused by the complexity of the problems to
be resolved,

- define the vicious circle uncertainty-rigidity, and
- assert that even agents within the bureaucratic
apparatus have a certain degree of freedom and capacity
for negotiation. This flexibility is due to uncertainty,
which is a "normal" and (to some extent) unavoidable
element of social action.
In the last part, I offer a short review of more
contemporary authors regarding the issue of bureaucracy.
The idea is very simple: to show that the argument that
makes this form of organization a complex and dynamic
one, capable of adapting to new situations, is clearly an
important topic in today's specialized literature.
2.1 Weber's Bureaucracy
The following is a summary of ideas about bureaucracy
relevant to this dissertation that Max Weber built in his
posthumously published Economy and Society (1922 [1944]).
Bureaucracy, as an ideal type, intends to be a "rational"
conceptualization of social action, a mechanism that
allows comparison with and interpretation of reality. In
this sense, to observe bureaucracy as a theoretic
category implies to analyze rather than to search for
"dysfunctionalities." The study of bureaucracy, it is
argued here, implies studying bureaucracy as an important
reference for the dilemmas of modern organization and the
difficulties of finding radically different
organizational alternatives to it. In this sense, the
idea of bureaucracy is still an important general
category for comparing and searching for organizational

alternatives and for the understanding of contemporary
inter-organizational relationships.
2.1.1 The Ideal Type
The first discussion must address the term "ideal type."
By "ideal type" Weber was not expressing an ethical
judgment about the "goodness" of a social or historical
model. It is only a methodological artifact that he uses
to try build a construct that imposes the highest level
of rationality (in terms of means necessary to achieve
results) to the phenomenon under study. Therefore, the
model does not try to reduce the reality to the model,
because the role of the ideal type is not to explain
reality as is. The strength of the ideal type is revealed
when one historical or current reality is compared with
the ideal type, not to find congruencies, but
differences. Thus, the important things to study are not
the similarities but the differences between observed
reality and the extremely rational construct. Those
factors that escaped from the rationality defined by the
model (the ideal type) are an expression of the "real"
reality, of the "real" action of "real" actors.
Therefore, those disparities are likely to show the
important and transcendental subjects and objects that
should be studied in depth (Weber 1922, 10) In the logic
of the ideal type there is no place for reality as sn^h.
Weber dealt with bureaucracy as what he defined as an "ideal
type" This methodological concept does not represent an average
of the attributes of all existing bureaucracies (or other

structures), but a pure type, derived by abstracting the most
characteristic bureaucratic aspects of all organizations. Since
perfect bureaucracy is never realized, no empirical organization
corresponds exactly to this scientific construct. The criticism
has been made that Weber's analysis of an imaginary ideal type
does not provide understanding of concrete bureaucratic
structures. But this criticism obscures the fact that the ideal-
type construct is intended as a guide, as an empirical research,
not as a substitute for it. By indicating the characteristics of
bureaucracy in its pure form, it directs the researcher to those
aspects of organizations that he must examine in order to
determine the extent of their bureaucratization (Blau 1956, 24).
When Weber built an ideal type, he did not mean the
construct to be considered as the task to achieve.
Neither was he arguing that the ideal type can become a
"picture" of reality. The meaningful things after the
analysis are those contrasts between the reality and the
model. The intention is not to make a prescription to the
reality about the best way to become as "rational" as the
ideal type, but to observe the complex rationality that
is expressed by reality, building the parameters of the
social action. Weber assigned an heuristic function to
the ideal type.
In order to see how an ideal type differs from a theory or a
model, we must try to analyze in detail the various steps taken
for the construction of the former:
- The first step is the selection and conceptualization of
empirical data, this selection being determined by Weber's
interest in finding the typical aspects of certain types of
- The second step consists in exaggerating such selected gestures
to their logical extreme. For example, in ideal type of
bureaucracy, the hierarchical relations between bureaucrats are

one hundred per cent impersonal. In reality this is never the
-Finally, the selection and exaggeration of empirical elements
and their formation into an ideal type are not done in an
arbitrary way. These elements are interconnected in such a way
that they form a whole portraying an inner consistency and
logic... (Mouzelis 1967, 45).
The analysis above is important. Bureaucracy is seen by
Weber as an ideal type of rational-legal action form of
organization. In short, it is a concept that has logical
order and stable meaning (Aguilar 1989, 573) .
The basic problem for Weber is how to give form to unitary, non-
contradictory, historical concepts that recognize the peculiar
originality of a society or a historical epoch and that, on the
other hand, can reproduce without contradictions the inevitable
individual differences that are expressed in a particular society
or historical epoch (op cit., 575)
This means that the researcher chooses strategically only
those original components of an historical process that
can be elaborated as components of an action (as ends,
means, circumstances or consequences) and, second, that
can be ordered as components of a rigorous rational
action. Thus, the ideal type is an instrument to analyze
history through a process of rational conceptualization.
A common misunderstanding then is to take the ideal type
as a description or prescription regarding a current
phenomenon. Another common problem is to confound a
rational method with a rational perspective of reality.
Ideal types are rational because they are defined using a

definition of concrete means used to achieve specific
ends. However, these concepts or ideal types are not
meant to describe or prescribe how actors should behave
in particular circumstances. Rather they aim to propose a
scientific way to better understand reality. And here is
why Weber introduced the idea of probability. He very
carefully defined all social relationships as
probabilities. He defined probabilities in his historical
studies but did not intend to generate general rules to
define universal social probability structures. To do so
is the job of particular researchers confronting specific
2.1.2 Bureaucracy as a Form of Organization
Once the methodological role of the ideal type is
defined, it is possible to explain the characteristics of
a bureaucratic form of organization. Weber defined
bureaucracy as an ideal type that explains one form of
"association of domination." An association is a social
relationship where the order is guaranteed by specific
actors: a ruler and eventually the people that constitute
the administrative apparatus of the association. In this
dissertation, I use Weber's "association" as a synonym
for organization.
The order in social relationships was defined by Weber as
the basis of any association; the continuity of the
association depends on the capacity of particular
individuals to take responsibility for the maintenance of
the order. All associations have a ruler and eventually

an administrative apparatus. According to Weber, their
basic role is to maintain the order of the association.
Order is defined as the basic set of ordinances that
should be obeyed. Weber recognized two basic orders: the
administrative and the regulative order. The latter is
the assurance of the existence of benefits for all the
participants in the relationship. The former is the
assurance of obedience to the organizational rules. The
administrative apparatus has two, sometimes conflicting,
goals: to obey the rules while also making sure that the
rest of the members obey them (Weber 1922, 39-41).
For Weber, all organizations are formed and defined by
social actors engaged in relationships of domination.
Domination is defined as the probability of achieving
obedience given a specific command. Domination is a
special case of power and is thus much more specific: in
domination, the command is explicit. The main question
is, who commands and who obeys? The foundation of all
authority, and hence of all compliance with orders, is a
belief in prestige, which operates to the advantage of
the ruler or rulers. With different levels of belief
(from members of the association being dominated) in the
legitimacy of authority, different authority structures
and therefore organizational forms evolve.
Weber identified three kinds of such beliefs. The first
is that obedience is justified because the person giving
the order has some sacred or altogether outstanding
characteristic: a charismatic domination. Second, a

command might be obeyed out of reverence for old-
established models of order, due to tradition that
survives generation after generation: the traditional
domination. In the third kind of domination, the
obedience comes from the belief that the persons that
rule are invested by legal and technical authority: this
is the rational-legal ideal type of domination. This type
of domination can be further be broken down into
subgroups. The "purest" (meaning the most drastically
rational) among the different organizational subgroups,
is the bureaucracy (Weber 1922, 706).
Five related beliefs in the legitimacy of authority are
necessary for the ideal type of bureaucracy, according to
- that a legal code can be established which can
claim obedience from the members of the
- that the law is a system of abstract rules which
are applied to particular cases, and that the
administration looks after the interests of the
organization within the limits of the law.
- that the man/woman exercising authority also obeys
the law.
- that obedience is due not to the person who holds
authority but to the law which has granted him/her
this position (Albrow 1970, 43).
On the basis of these ideas, Weber proposed eight
elements of the rational-legal ideal type:

- official tasks are organized on a continuous,
regulated basis;
- these tasks are divided into functionally distinct
spheres, each furnished with the requisite authority
and sanctions;
- offices are arranged hierarchically, the rights of
control and complaint between them being specified;
- the rules according to which work is conducted may
be either technical or legal (in both cases trained
workers are necessary);
- the resources of the organization are quite
distinct from those of the members as private
- the office holder cannot appropriate his or her
- administration is based on written documents (this
tends to make the office the hub of the modern
organization); and
-legal authority systems can take many forms, but
are seen in their purest state in a
bureaucracy's administrative staff ( Albrow 1970,
43-44) .
The criteria for selection of personnel to belong to the
bureaucratic cadre are exposed in the famous ten Weberian
principles for the bureaucratic ideal type:
(1) They are personally free and subject to authority
only with respect to their impersonal official
(2) They are organized in a clearly defined hierarchy of

(3) Each office has a clearly defined sphere of
competence in the legal sense.
(4) The office is filled by a free contractual
relationship. Thus, in principle, there is free
(5) Candidates are selected on the basis of technical
qualifications. In the most rational case, this is
tested by examination or guaranteed by diplomas
certifying technical training, or both.
(6) They are remunerated by fixed salaries in money. The
salary is graded according to rank in the hierarchy; but
in addition to this criterion, the responsibility of the
position and the requirements of the incumbent's social
status may be taken into account.
(7) The office is treated as the sole, or at least the
primary, occupation of the incumbent.
(8) It constitutes a career. There is a system of
"promotion" according to seniority or to achievement, or
both. Promotion is dependent on the judgment of
(9) The official works entirely separated from ownership
of the means of administration and without appropriation
of his position.
(10) He is subject to strict and systematic discipline
and control in the conduct of the office (Weber 1922,
Weber saw bureaucracy as the most substantial
organizational arrangement that modern society could
create. Its technical superiority, based on its

impersonality and impartiality, would create not a world
of robots, but a world of organizations ruled by the
efficiency principle. However, Weber also saw several
problems stemming from this form of domination. Although
some people think that Weber did not see the possibility
that bureaucracies might fight to defend their interests,
he argued that any part of an organization (the ruler,
the administrative apparatus, and the ruled outside the
association) is interested in obtaining benefits from the
association of domination (following Weber's categories).
Moreover, in the case of bureaucracy, technical
superiority gives the ruler and the apparatus power to
centralize and interpret the orders.
Knowledge, one of the sources of the bureaucratic
omination, leads to a complex relationship of power,
where rules become the basic framework to interpret and
act upon a complex and dynamic world. Formally, the
impersonal rules and laws are the source of neutral and
efficient action. However, reality is so complex that
knowledge is also a source of power, due to the capacity
of actors to adapt rules to complex situations, therefore
making interpretations and taking decisions out from the
rigid framework of rules. In this sense, the rules and
laws could become inefficient. This is because
bureaucratic actors are supposed to take decisions
according to strict rules, but in practice the dynamics
of a complex world allow them some discretionary power.

2.2 Problems With the Ideal Type
Bureaucracy is more than an ideal type, it is also an
issue that affects important practical matters. The
actions of real bureaucracies appear for some people to
be very different from the efficient and "clean" Weberian
bureaucracy. Despite this, bureaucracies have expanded
rapidly, as Weber predicted.
To review some of the critics and further advances in the
theory of bureaucracy appears important in this context.
The goal of the following section is to take into account
several contributions by other theorists that could
enrich the methodological approach intended for this
2.2.1 The Controversy With Weber
This part introduces the ideas of Merton, Selznick,
Gouldner, and Crozier as important critics of the
Weberian model.
The critique of the idea of bureaucracy developed by
Merton (1940) is probably the best known. Merton asserts
that an emphasis on precision and reliability in
administration may well be self-defeating. Rules, planned
as means to ends, may well become ends in themselves. The
graded career structure of the bureaucrat may encourage
him or her to an excess of the virtues she or he is
supposed to embody: prudence, methods, and discipline.
Governed by similar work conditions, officials develop a

group solidarity which may result in opposition to
necessary change. In other words, the bureaucracy is a
static form of organization.
Merton emphasized that a structure that is rational, in
Weber's terms, can easily generate consequences which are
unexpected and detrimental to the attainment of an
organization's objectives (Albrow 1970).
Therefore, bureaucracies have several dysfunctions.
Actions based upon past training and skills might, under
changing conditions, become incompatible with the ends
that were supposed to achieve. Their emphasis on
discipline, as a way to obtain reliable behavior, can
create a process of displacement of goals where the
instrumental value becomes a terminal value. Conformity
to the rules can interfere with the fulfillment of the
purposes of the organization. One of the most important
aspects of this critique is that bureaucracy's inherent
nature (emphasis on reliable behavior and devotion to
rules) generates those dysfunctions. Perhaps a good term
to express this would be the term "paradox." It can be
said that the bureaucratic model of organization entails
a paradox. This dysfunction is paradoxical since it makes
no sense for bureaucracies to behave according to rules
if doing so means that the achievement of goals becomes
secondary. This happens, according to Merton, because
rules cannot entirely predict how reality will evolve, so
bureaucracy must adapt rules and laws in order to ensure
achievement of their goals.

The work of Selznick (1949), concentrating upon the
division of functions within an organization, shows how
sub-units set up goals of their own which may conflict
with the purposes of the organization as a whole.
Selznick studied the Tennessee Valley Authority (TVA),
which was created by the U.S. Congress in 1933 to plan
the proper use, conservation, and development of the
natural resources of the Tennessee River, its
tributaries, and the adjoining territory. He follows the
process whereby the TVA's administrators adjusted the
agency's policy and structure to satisfy the demands of
local interest groups. This case study shows the
evolution of a bureaucratic oligarchy behind a democratic
The core of Selznick's study lies in his analysis of the
sophisticated process of administrative and political
manipulation whereby the beliefs and practices associated
with the "grass roots" control over policy implementation
were reinterpreted so as to ensure policy results of a
much more orthodox nature. According to Selznick, this
metamorphosis was necessary for the survival of the
Bureaucratic pressures (i.e., pressures to keep the
decision-making process firmly attached to explicit rules
and laws) exist in any organization. The important issue
is to find how people seek and implement strategies to
control such pressures, even partially. Selznick
describes how specialization creates an organizational

capacity to respond to problems so narrow that technical
decisions are difficult to defend in a complex situation.
In other words, rules, norms, and procedures do not
ensure that unexpected outcomes will not be found.
Uncertainty regarding outcomes makes bureaucrats seek
political support in order to avoid bearing sole
responsibility for possible failures. However, this
process of seeking more political support leads to more
specialization due to the necessity of keeping a
framework of rules and laws to support any action or
decision. This dysfunction, observed in the TVA, was
tackled by the members of the organization in two
different ways: by co-opting different groups outside the
agency and by promoting TVA's "ideology" (in other words,
their values or mission) to secure a necessary minimum of
conformity and devotion to the organization within
bureaucracy as well as among persons outside the
organization, but affected by the agency (Selznick op
cit., 254-264). In other words, TVA sought to generate a
symbolic commitment from its cadres and the public
regarding the mission of the organization.
With this strategy, TVA co-opted several groups (the most
important being large-scale farmers) in order to shield
the organization from attacks coming from environmental
groups, small-scale farmers, and ethnic minorities within
the local agricultural community (Reed 1985).
Selznick explains that the TVA, entitled to develop a
federal effort of natural resource planning, had actually
little authority for engaging in large-scale regional

planning. Its delegated powers were very precise in
character and range, and were related to the fundamental
problems of navigation, flood control, power production,
and fertilizer use. Nonetheless, the organization was
given discretionary power to design a range of policies
and programs dealing with local people and institutions
reflecting the "grass-roots" ideology on which it was
founded. As a result, the agency's leadership developed
policies and programs adjusted to the demands of local
interest groups and institutions which were first hostile
to its existence as a mechanism of federal control over
regional business. The discretion given to officials to
make important decisions on their own was legitimated as
necessary to allow them to form a working partnership
with existing agents in the Valley.
The TVA created a network of supporting voluntary
associations that were arranged in such a way as to
legitimate the TVA.
In this sense, according to Selznick, bureaucracies have
a clear strategy for dealing with unanticipated
consequences: commitment (Selznick, op cit., 254-255).
Commitment then is a general tool of institutions:
In ocher words, the lack of effective control over the
tangential informal goals of individuals and subgroups within an
organization tends to divert it from its initial path. This holds
true whether delegation is to members and parts of a single
organization, or to other organizations.... The problems
indicated here are perennial because they reflect the interplay
of more or less irreconcilable commitments: to the goals and

needs of the organization and at the same time to the special
demands of the tools or means at hand (Selznick op cit.,258).
Once again, with Selznick, we are confronted with the
paradox of bureaucratic organization: the goals and needs
of the organization might require the use of non-legal
means in order to be achievable. Individuals cannot be
entirely controlled by rules and norms, so they will
enjoy a certain level of freedom. Such freedom allows
bureaucrats to seek commitment within themselves and
among other groups in the context of the organization in
order to achieve their objectives.
Gouldner (1948) analyses the inherent difficulties of
professional and bureaucratic organizational structures
to achieve goals. He distinguishes two major types of
bureaucracy: "punishment-centered" and "representative."
In the former, members of the organization conform
reluctantly to rules which they consider to be imposed on
them by an alien group. However, in the latter,
organizational members regard rules as necessary on
technical grounds and in their own interests. The two
attitudes have marked influence on the efficiency of the
In the first type of bureaucracy, Gouldner finds a
vicious circle of control and supervision. The impersonal
bureaucratic rules reduce tensions among bureaucrats
caused by subordination and control. However, at the same
time, those impersonal rules increase the discretionary
capacities of bureaucracies, thus pushing organizational

authorities to launch more actions to increase
organizational control. The need to regulate discretion
is achieved through expansion of rules (Gouldner 1950).
Thus, a paradox is also found by Gouldner: in order to
achieve their objectives, bureaucrats cannot be
supervised in every aspect of their work. To attempt that
would not only be impossible but would generate severe
pressure for bureaucrats. They need some space for
freedom in order to adapt to a dynamic reality. However,
this freedom generates discretionary powers among
bureaucrats that pushes organizational authorities to
launch more rules to control those powers, generating
then new pressures over the administrative apparatus.
Crozier (1964) tries to develop a new framework for
analyzing the bureaucracy. He observes that those
bureaucratic paradoxes found by Merton, Selznick and
Gouldner are seen as pathologies or "unfunctionalities"
within the bureaucratic organization. What he proposes is
that those paradoxes are not "illnesses" or pathologies
but are logical and "normal" contradictions within the
bureaucratic nature.
Here we summarize the basic ideas of Crozier's The
Bureaucratic Phenomenon (1964). In one of the case
studies of this work, Crozier examines the hierarchical
structure and operating routines found within the
Parisian branch of the French government accounting
bureaucracy. In the other case study, he develops an in-
depth analysis of three Parisian plants of a state-owned

tobacco monopoly. These organizations are part of a
centralized authority system.
The "Clerical Agency" case examines the formal structure
(i.e., hierarchies and formal chart of the organization),
the relationship and position of the different actors
within the structure, and their impact on the decision-
making process. This case tries to show how the power and
status interests of some work groups are protected by
formalized and routinized patterns of decision-making.
The "Industrial Monopoly" case analyzes the relationship
between the formal structure of administrative control
and the network of power relations that develops between
groups located at management and shop-floor levels. The
objective here is to show how a tight regulatory system
imposed through formal channels cannot cover every
eventuality, allowing some bureaucratic actors to obtain
high levels of power in some particular situations, even
within a tightly controlled formal framework of authority
and supervision.
Both case studies dealt with organizations with highly
structured rules, with management teams continually
subjected to control by senior officials located within
the central government bureaucracy. He found that while
this control by other governmental agencies replaces the
usual commercial pressures to which most private sector
organizations are subjected, it intensified the degree of
political uncertainty that bureaucratic agencies faced.
In both cases, the pattern of intergroup relations within
the bureaucratic apparatus was highly stable and centered

around those areas where uncertainty (regarding the
supervision in the first case and of control of engine
failures in the second) was not controlled by explicit
management policies. However, these rigidities made it
very difficult for the organization to adjust to
contextual changes, and in both cases, this rigidity
resulted in several crises.
The normal response to those crises, in both cases, was a
process of "mechanization," where the areas of
uncertainty were reduced by implementing a more specific
and rigid structure. The rigid bureaucratic structure
created more crises under conditions of environmental
change, but it provided managers with a more
sophisticated administrative system whereby they could
attempt to contain those conflicts and turn them to the
organization's long-term advantage (Reed 1985, 163). This
mechanistic model of organizing (approaching the Weberian
ideal type more and more) resolves some of the problems
related with bureaucracy (i.e., goal displacement) if it
is capable of accommodating internal power struggles
while encouraging the level of knowledge and sensitivity
required for top-level decision-makers to be able to
initiate and maintain incremental changes to resolve
severe crisis arising from internal and external
These works show the elements that must be studied when
comparing real bureaucracies and the ideal type. Weber
argued that the comparison between ideal type and reality
allows the observation of important elements, necessary

to understand the organizational reality. However, the
critiques of Weber show that the paradoxes (i.e.,
discipline that creates a process of displacement of
goals [Merton], or that the goals and needs of the
organization might need to use non-legal means in order
to be achievable [Selznick], or the vicious circle of
control and supervision [Gouldner]) that result from the
bureaucratic structure of action are complex phenomena.
The basic problem is how to obtain reliable and efficient
collective action within market economies and democratic
societies, and how to achieve a certain level of control
over human discretion in order to increase the
probability of obedience. The relationship between the
former (efficient collective action) and the latter
(ensuring obedience) is neither automatic nor without
contradictions. Often, the second factor displaces the
first. This might be an inevitable contradiction of the
bureaucratic form of organization.
The existence of power relationships within the formal
structure is a basic element of Crozier's study. The use
of power by some actors, sometimes to control and
sometimes to reduce "uncertainty areas," seems to be
fundamental to the study of the bureaucratic form of
organization and the actors involved. Bureaucracies,
rather than resembling a static organization, are complex
and dynamic forms of organization.
Perhaps it would be important to place bureaucracy as a
form of organization taking its dysfunctions as logical
contradictions of a particular form of collective action.

In the next part there is an attempt to extend our
understanding of bureaucracies beyond an analysis of
dysfunctions. The idea is to observe bureaucracies as
complex social responses to collective problems and then
connect this with the understanding of the role of power
and uncertainty in the bureaucratic organizational
pattern obtained from the previous analysis.
2.3 Modernity and Organization: The Iron Cage
Stewart Clegg (1990) has developed the idea of
bureaucracy as the main form of organization for and by
modernity. His assumption is that organizations are the
"creatures of modernity," the answer to society's need
for order and harmony in a turbulent context:
Organization, conceived in terms of its modernist antecedents,
implies a degree of legal and moral normative unity, a single
center of calculation and classification, 'a relative unambiguous
distribution of power and influence, and setting for action
sufficiently uniform for similar actions to be expected to bring
similar consequences for the whole and thus be interpreted in a
similar way' (Clegg 1990, 19)
Bureaucracies, following Weber, appear to be superior to
any other form of organization: they reduce uncertainty
through moral and legal homogeneity and impersonal
Rational action (according to Weber) consisted precisely in the

capacity to respond to the new uncertainties of a world without
meaning. The modern world was, by definition, an age of
uncertainty... Uncertainty both defined and limited freedom. It
defined freedom through posing the existential and environmental
conditions under which rational action was possible. It limited
freedom by imposing an ethic of calculation, as a totally
objective rationality, upon this freedom to act. (Clegg 1990, 32)
To master uncertainty implies paying that cost: freedom
is constrained by rationality in order to reduce
uncertainty. Bureaucratic rationality reduces complexity
by imposing a framework of rules and procedures through
an impersonal order. According to Clegg, a related cost
to this might be the loss of transcendence of meaning and
interpretation. In this sense, Clegg argues that
bureaucracies are a necessary feature of modernity (1990,
33) .
However, bureaucracies expand the calculability, as Weber
clearly noted, for only one part of reality: formal or
means-ends rationality. The other prominent part, what
Weber called substantive rationality, the rationality
that allows the mastering of the world through abstract
concepts (the part reserved for different forms of
cultural comprehension of the world like magic, myths,
and tradition), is abandoned in the pursuit of
calculability. The "iron cage" of bureaucratic
rationality arose as a nightmare for Weber, who imagined
a world of "little persons" defending their "little job
positions" with formal instrumental rationality.
However, Clegg argues, bureaucracy is also a human

project. Following the review of Merton, Selznick, and
Gouldner's critiques, we see that the "iron cage" is not
tightly closed. It would be difficult to find a case
where substantive rationality has completely disappeared.
The iron cage of bureaucracy -- in other words, the idea
that there is no alternative to manage uncertainty other
than by simplifying complexity throughout bureaucratic
organizations -- appears to be questionable. Clegg thus
prefers to see the ideal type characteristics of
bureaucracies as tendencies rather than closed
characteristics (1990, 38-40) .
Each characteristic of bureaucracy (as described by Weber
1922, 220-221) or each tendency (following Clegg) has
been seen as a variable for some scholars (Hall 1962,
1963, Hall, Haas and Johnson 1967, Pugh and Hickson 1976
and Hage 1965, Hage and Aiken 1970). Several authors have
tried to develop a natural theory of organizations and
bureaucracies, arguing that these variables are universal
and natural tendencies of all organizations (for example,
authors that use "open system" metaphor theory like Katz
and Kahn 1966 and Thompson 1967). The central idea is
that organizations attempt to create imposing 'organized'
mechanisms on themselves and their environments (Meadows
1963) .
In this sense, rational bureaucratization emerges in
response to problems of coordination, which are caused by
a variety of sources: (1) size: large numbers of actors
and/or transactions without any systematic rules for
sharing information (Dahl and Lindblom 1963; Blau and

Schoenherr 1971; Hickson et al. 1969; Blau 1970; Child
1973); (2) social-economic complexity or specialization
and differentiation of roles and organizations (Parsons
1964) ; (3) interdependence or complexity and
technological relations (Thompson 1967); (4) technology
and technological innovation (Woodward 1965; Lorsch 1965;
Aldrich 1972, Lawrence and Lorch 1967; Perrow 1968,
1986); (5) internalization of cost "externalities"
(Buchanan and Tullock 1962) ; and (6) war or external
threat becoming the source of state organizational
structure (Hintze 1968) .
Whatever variable, or tendency, or characteristic is
emphasized, these analysts argue that decision makers
develop large-scale bureaucracies to avoid the costly
impact of disorder or lack of coordination. The rational
bureaucracy is seen as a "natural" process essential for
the survival of the society and the organizations that
compose it. However, it is also a human instrument or
artifact, a social system that can suffer dysfunctions
and problems. However, following the same bureaucratic
mechanisms, these problems can be resolved by better
allocation of resources and greater efficiency.
From another point of view, but sharing the same
assumptions, the ecological view of organizations
proposes that the bureaucratic organization is the
consequence of a process of natural selection occurring
within a population or ecosystem of organizations (Hannan
and Freeman 1977; Freeman 1982; Aldrich and Muller 1982;
Astley 1985). Bureaucratization occurs as the result of

the environmental differentiation, selecting either of
the organizational variations discussed above depending
on which variation best assure its survival. Implicitly,
the selection process is still a mystery within this
theory, leading to some propositions about competition
and random variation.
Organizations in general and bureaucracies in particular,
conceived of as natural systems, are generally regarded
as oriented toward a certain type of rationality,
hierarchy, planning, goals or efficiency (Clegg op cit.,
52). In this sense, the "iron cage" conception of
bureaucracy failed to explain the diversity of
organizational structures and forms that can be found in
the world. In other words, organizational actors are able
to adapt bureaucratic structures.
The literature on this subject suggests two basic ways to
explain the ability of actors to create organizational
diversity. One of these, primarily sociological,
emphasizes the problems of structuring and allocating
power as the focus of decision-makers' concerns. The
other, dominated largely by economists, stressed the
problem of profit or efficiency maximization as the goal
of decision makers. This dissertation concentrates
primarily on the first set of ideas. The reason is that
the research and problem questions drafted in the
introduction are more related to the situation of group
negotiations in multi-value organizational settings
rather than homogeneous economic definition of
preferences in order to take decisions (to study some

important authors of the second set of problems see Blais
& Dion 1993; Buchanan & Tollson 1972; Downs 1967; Elster
1989, 1986, 1983, and 1979; Mises 1944; Niskanen 1971;
Ostrom 1973).
Power and conflict play an important role in terms of
sociological explanations for organizational diversity.
The basic argument is that decision-makers confront a
world where it is impossible to predict the consequences
of actions taken by different agents. Therefore, actors'
rationality is bounded (Simon 1947). Under conditions of
bounded rationality, actors attempt to maintain their
power by resorting to the utilization of rules of
decision making that are presented as being objective.
These rules require hierarchy of authority, knowledge,
well-defined roles, and obedience (Gouldner 1954; Crozier
1964; Selznick 1949; Clegg and Dunkerlay 1980; Pfeffer
1981). Expertise thus becomes the base of legitimate
authority, and the rules of a bureaucratic organization
become legitimate because they appear to be based on
scientific observation and analysis of large populations
of organizations or analogous systems. However, actors
are still able to mobilize resources to adapt or change
organizational conditions due to the permanent existence
of uncertainty stemming from a complex reality. In this
sense, it is important to explain how organizational
actors, within bureaucratic structures, are able to
mobilize resources using their freedom and power.
The work of Crozier (1964) and Crozier and Friedberg
(1977), deal with these issues in more detail. In the

next section I will summarize their ideas, on which part
of this research is based.
2.4 The Bureaucratic Phenomenon: Crozier's View
After this general review, it seems necessary to reach a
more precise position regarding the bureaucratic
phenomenon, where the "dysfunctions" of bureaucracy can
be seen as organic contradictions. In other words, it
seems necessary to study the bureaucracy as a structure
of collective action that uses contradictory mechanisms
to maintain order and achieve results, manipulating the
impersonal and technical instruments which give it
As discussed earlier, several authors identified a basic
paradox in the bureaucratic process and the ideal type of
legal-rational domination: the complex relationship
between impersonal authority, which leads to technical
decision-making, and the necessary but dangerous amount
of standardization that this relationship imposes on its
members (Mayer 1943). Among Weber's successors and
contemporaries, this contradiction continuously appears.
However, it is difficult to explain why this
contradictory process exists as well as its permanence,
adaptability, and universality as a form of organization.
Merton, and other American sociologists like Gouldner and
Selznick have made important contributions to the
development of the ideal type of bureaucracy and to the
discussion of whether it is possible to resolve the

opposition between organizational efficiency and human
freedom. Within the "vicious circle" of the bureaucracy
lies the resistance of the human being facing a
mechanical rationality where discretion tends to
disappear. This is one of the basic arguments stemming
from the work of these authors.
As we have seen, Merton, without contesting directly the
validity of the ideal type, suggested that there are
secondary consequences of the bureaucratic action, such
as ritualist attitudes and esprit de corps (Merton 1936).
Merton did not discuss why if the model is so static it
does not degenerate in time. After all, if the
dysfunctional consequences of the model demand the use of
more control, one should find more and more dysfunctions.
In other words, if dysfunctions generated by bureaucratic
action are so clear and general, why we do not see the
total and general collapse of bureaucratic organizations
everywhere? (Crozier 1964, 180).
Another good example that uses the approach of the ideal
type, is found in Selznick (1949) Selznick studied how
the organization deals with dysfunctions, without trying
to explain those dysfunctions (Crozier op cit., 181) .
According to Crozier, neither Merton or Selznick study
the possible interdependence between rationality and
dysfunction. This is the key argument that Crozier adds
to the analysis of bureaucracy.
Based on the analysis of two case studies, Crozier

identified four basic means of maintaining stability
within what he called a vicious circle of bureaucracy
(Crozier 1964, 187-193): (1) the development of
impersonal rules delimiting the general functions of
every individual within the organization; (2) the
centralization of decision making at the highest levels
of the hierarchy, protecting decision makers from
retaliatory actions by those adversely affected by the
decisions; (3) the insulation of individuals between
different organizational strata and the concomitant group
pressure on the individual coming from peers and
superiors (this pressure reinforces the barriers between
strata, leaving very little room for problem solving
across organizational categories); and (4) the
development of parallel power relationships (or informal
power structures) among groups and members, given the
impossibility to eliminate uncertainty just through the
definition of formal organizational rules and policies.
Crozier observes that individuals and groups are able to
control (through formal and informal means) certain
organizational activities. Several of these activities
deal with particular organizational uncertainties.
Crozier called these "areas of uncertainty" (1964, 190) .
Controlling some areas of uncertainties, individuals and
groups have the possibility of using conflicts (derived
from uncertainties) to their benefit.
The central idea in Crozier's analysis is that the
contradictions inherent in bureaucracies are part of
their organizational process. Crozier argues that

bureaucratic organizations are not capable of correcting
their behavior by learning from their errors, unless
crises arise. Consequently, they suffer a permanent
vicious circle of complexity: they try to control
uncertainties by creating procedures and norms; when they
fail because complexity and crises arise, the basic
response is to create more detailed norms and procedures.
This leads to increasingly rigid frameworks that allow
individuals and groups to struggle for different
organizational positions. Rigid norms often are unable to
deal with complex relationships among groups and
individuals (inside and outside the organization). To
resolve complex problems then, the bureaucracy usually
has recourse to discretionary powers created by the
incapacity of norms to define a complete set of solutions
for that problem. This bureaucratic discretion generates
ambiguity regarding the sources of legitimacy for the
decision-making process. Then, bureaucratic organizations
will seek to create more regulations to control
discretion and ambiguity in bureaucratic behavior,
according to Crozier (1964, 193) :
[T]he rigidity of task definition, task arrangements, and the
human relations network results in a lack of communication among
the groups. The resulting difficulties, instead of imposing a
readjustment of the model, are utilized by individuals and groups
for improving their position in the power struggle within the
organization. Thus, a new pressure is generated for
impersonality and centralization, the only solution to the
problem of personal privileges. (Crozier 1964, 194) .
Therefore, the dysfunctionalities of the bureaucracy

cannot be seen as pathologies or diseases. The complex
paradox of rationality and human discretion and power
struggles over strategic positions (involving the control
of uncertainty areas) are essential elements of action
within bureaucracies.
However, bureaucracies can adapt and change themselves
through different processes of mutual adaptation. In The
Bureaucratic Phenomenon, Crozier developed a very rigid
perception of bureaucracies, exaggerating their
resistance to change and their inability to adapt to
complex environments. In a later work with Friedberg
(1977), the bureaucratic phenomenon acquires more
flexibility through the study of uncertainty as a source
of liberty and transformation stemming from the
reallocation of actors' power structures.
2.5 The Actor and the System: Power, Strategy, and
Conflict in Organizations
There is a tradition in public administration studies of
analyzing the action of governments as a unitary and
rational decision making process (Model I as in Allison
1973). This view assumes that governmental programs are
defined through integral planning procedures and expects
bureaucratic behavior to be congruent with general
planned objectives.
However, as Crozier and Friedberg (1977) argue,
organizations are not "natural beasts" whose existence
and behavior can be transparently managed. They are human

"constructs," developed in part consciously, in part
unconsciously, to solve problems of collective action
when the participants are relatively independent social
actors pursuing diverse and conflicting interests
(Crozier and Friedberg 1977, 3-4).
Thus, it is very common for public actions implemented by
bureaucratic agencies to differ, in terms of outcomes,
from the expected results. The cause, however, is not
only due to possible improper use of rational mechanisms
to identify and resolve a problem. There is no social
endeavor that does not face uncertainty (regarding
values, objectives, or outcomes). And this uncertainty
leaves actors with a generous allowance of freedom and
discretion in the design and implementation of practical
Liberty and uncertainty allow the actors to define
different spaces of action that shape and direct
organizational performance. These spaces of uncertainty
and ambiguity are those that make it possible for
bureaucratic actors to take decisions, and permit the
definition and redefinition of power fields for different
Organizational games are built around logical, "objective"
uncertainties, those of the techniques, of the markets, of the
different constraints which are, at least for the short-run,
stable, unshakably given. Those who get the upper hand in the
game are those who control most of the crucial uncertainties. But
the way the game is structured and the stakes determined

will reduce Che possible gains and losses to acceptable
proportions. Artificial uncertainties such as authority
distribution, information channels, and legal constraints, can,
on the other hand, counterbalance the "objective" ones. In any
case this means that, to be processed, problems will have to be
redefined either to fit prevailing games or at least to
allow for the creation of some artificial uncertainties without
which no bargain can be struck" (Crozier and Friedberg 1977, 8).
These games, which define the areas in which uncertainty
exists, lead to the negotiation and configuration of
power relationships. Too much artificial regulation can
lead to extreme rigidity, but too little can lead to some
privileges, generating problems for rational action and
weakening the bases for cooperation. The counterintuitive
effects appear everywhere collective action is involved
(Crozier and Friedberg 1977, 9).
A basic tactic used by organizational actors to deal with
these problems, is called by Crozier and Friedberg "the
empowerment of a strategic position". They assume that
the actors rarely have clear objectives and that time and
events might change the meaning actors give to certain
processes. However, this does not mean that the actors
are not rational. The actor could be rational in terms of
opportunities or according to a different interpretation
of the predefined objectives. He or she behaves either
offensively (making use of opportunities to improve his
or her situation); or defensively (maintaining and
broadening of his/her margin of liberty and his/her
capacity to act) (Crozier and Friedberg 1977, 25).

Thus, the organization is defined as a social construct
that allows human beings to deal with uncertainty not
only with bounded rationality but also bounded legitimacy
(incomplete normative integration of members) and of
bounded interdependence (incomplete functional
integration of members) (1977, 6) .
Actors in organizational settings are free, they have
options, since decision-making processes require at least
some degree of freedom. However, this freedom is
contingent. Although they run the risk of being incapable
of observing irrational actions (accepted by the Crozier
and Friedberg 1977, 25), the above authors propose that
the freedom generated by the uncertainty elements
("natural" ones like the context, the markets, the
technology; and "artificial" ones like the authority
framework, the information channels, the legal
constraints) is ruled by power relationships. This
interpretation views power as a relationship, not as an
attribute, that, enables the actors to negotiate and
interact in a reciprocal but unbalanced relation (1977,
31) Moreover, organizational actors acquire conscience
regarding the formal expression of power, becoming part
of a system of organized regulation (Friedberg 1993) .
"The power of an individual or group, or social actor, is thus a
function of the size of the zone of uncertainty that the
unpredictability of the actor's conduct enables him to control
vis-a-vis his partners. We have already explained that not every
zone of uncertainty is important. It must be relevant to the
problem at hand and in relation to the parties involved. It must
be a zone of uncertainty whose existence and control condition

the capacity of these parties to act. The strategy of each of the
partners/adversaries will then naturally take the form of
manipulation of the predictability of his own and others'
behavior, either directly or indirectly, by means of judicious
modification of the structural conditions and 'rules' regulating
his interactions with others. In other words, each actor will
seek to widen his own margin of liberty as much as he can in
order to maintain as broad as possible a range of behavior, while
trying at the same time to limit his partner/adversary's range
and to shackle him with such constraints as to make his behavior
foreseeable" (Crozier and Friedberg 1977, 34) .
Crozier and Friedberg's analysis helps us to understand
bureaucracy not only as a formal structure of
organization, full of dysfunctions and pathological
responses to complexity. We find that bureaucracy
generates a vicious circle of uncertainty-generation of
rules-uncertainty. Moreover, that uncertainty generates
freedom for actors to control organizational structures
and to informally use human relationships to generate a
more stable order within the organization.
2.6 Bureaucracy: the contemporary discussion
In this subsection I present a quick overview of the
current status of the debate on bureaucracy. The aim is
to show that the issue of flexibility and the
adaptability of bureaucracy is an important topic in
current of public administration literature, which
understands bureaucracies as complex forms of
organization, and full of contradictions (basically rigid
yet capable of changing, for instance).

Since this discussion is not the primary objective of the
present research, the review is very focused and only
attempts to give some examples of the current debate,
rather than trying to give a complete overview of the
contemporary literature.
2.6.1 Budget Maximizing Bureaucrat
The economic analysis of bureaucracy is one of the most
important approaches in the literature today. The basic
idea is that the political and organizational decision
making process is similar to that which takes place in
open market transactions. Looking for a logic similar to
that of individual rational behavior in the market in the
collective or organizational arena (Arrow 1960), we find
a problematic relationship between equity and Pareto
optimum (Elster & Hylland 1986).
In the field of public administration, this type of
approach was proposed by Vincent Ostrom (1972) in an
attempt to redirect the paradigm that had predominated in
the U.S.
Ostrom establishes that since Woodrow Wilson, the
discipline attempts to be the obedient body of the law
and of the order from the government. However, Ostrom
argues that the Federalists Papers contain a different
interpretation regarding the nature of government.
The basic argument is then that bureaucrats in a
democratic society cannot be seen as neutral and obedient

machines. Ostrom argues that bureaucrats will generally
refuse to obey unlawful commands to exploit the citizens'
commonwealth or to use the coercive capabilities of the
government to adversely affect the rights of the people,
but he argues that they will use peaceful persuasion in
taking such stands (Ostrom 1972, 131).
Ostrom opens the door to a more complex study of
bureaucratic behavior using the theory of economic
rationality. He proposes that externalities, the logic of
the collective action and public goods, and public
enterprise will be the focus of discussions in public
administration research. Researchers will change their
focus from a preoccupation with the organization itself
to the opportunities that individuals can pursue in a
multi-organizational environment. A democratic theory of
the administration will be concerned not just with the
simplicity, equilibrium, and symmetry of the preferences
of the actors and organizations, but with their
diversity, variety, and responses. A system of democratic
administration relies on the ordered complexity in social
relations (Ostrom 1972, 132). These ideas are clearly
related to economic approaches that propose to find a
solution to the critiques of the Pareto optimum as the
inevitable outcome of efficient rational behavior. For
instance, Buchanan (1972) proposed to observe the
exceptions to the Pareto principle not as a meaningful
failure in the logic of the model, but as an inaccurate
interpretation when changing from "private behavior" to
"public behavior." This idea, defended by a school of
thought known as "public choice," says that the

assumption that the political system has the same process
of rationality as an economic market is wrong. Although
the political process also is an efficient regulator
toward optimum efficiency, the rules are distinct (McLean
1987) In other words, the rational instruments of the
economy are useful to study political behavior, as long
as the proper equations are built tc analyze the optimal
use of resources (also scarce) of the political system
(Broeck 1988).
Public Choice can be defined as the economic study of
decisions outside the market, or simply the application
of economic thought to the political science. The object
of study is the same as that in the political science:
the theory of the state, voting rules, behavior of the
voter, and government bureaucracies. The methodology,
however, is that used in economics (Mueller 1979, 1).
These ideas and methodologies were applied to
bureaucratic behavior by Niskanen (1971) with the
argument known as the "budget-maximizing bureaucrat" or
the economic explanation of the bureaucratic behavior.
Developing a rapid review of other approaches to the
concept of bureaucracy from the economic arena (Von Mises
1944, Tullock 1965, and Downs 1967), and assuming that
the concepts elaborated by Weber and successors do not
deal in detail with the characteristics of the economic
behavior of bureaucracy, Niskanen developed a basic
economic model of bureaucratic behavior in democratic
societies. He begins the analysis with a definition of

bureaucracy which, he argues, has two main
1. The owners and employees of these organizations do not
appropriate any part of the difference between revenues
and costs as personal income.
2. Some part of the earnings of the organization come
from a source other than the sale of output at a per-unit
This definition excludes clearly all private
organizations and includes the nonprofit organizations
that receive financial support from other agents. At the
same time, it excludes those public organizations whose
earnings obtained through sales are high (e.g.,
government-owned electricity and gas utilities).
In addition, Niskanen argues that the bureaucracies
specialize in the delivery of services that "some people
prefer be supplied in larger amounts than would be
supplied by their sale at a per-unit rate" (Niskanen
1971, 18) This means that one of the most important
differences between private organizations and the
government is the cost of transfer of membership from one
organization to another, and not, as is commonly
believed, the nature or degree of the individual's
influence on the organization's activities or the degree
to which coercion is used to obtain revenues.
The fact that some governmental bureaucracies deliver
products and services characterized by high fixed costs

of production or where it is difficult to collect
payments, means that the definition of the outcome that
must be obtained by the bureaucratic agencies is,
generally speaking, hard to establish. Therefore it is
extremely difficult to define homogenous and permanent
parameters of control over the bureaucracies. In
conclusion, it is difficult for the agency that controls
and finances the bureaucracies to clearly define the
outcomes it wants, and how to accomplish them (Niskanen
1971, 20).
Bureaucrats have an important advantage in negotiating
the outcomes that they are expected to reach, and the
necessary resources. Generally, the agency that finances
the activities of the bureaucracy is actually dependent
on it to generate services and the bureaucracy is usually
dependent on that agency for funds. In economic terms
this is known as bilateral monopoly.
The bureaucracy offers a group of activities and of
expected results in return for funding. However, because
of the asymmetry of information, the bureaucracy is able
to establish a series of conditions.
From the point of view of the bureaucrat, the
relationship with the sponsor can be summed up by means
of a budget-output function. In rational terms, the
sponsor of the resources is willing to increase the
assigned budget in exchange for a promise of a higher
level of accomplishment. However, given the asymmetry of
information in the relationship between bureaucracies and

a financial agency, the financial agency on some
occasions gives a larger budget and receives in exchange
the same or even less outcome.
Given this dependency on outside authority for financing,
the priority of the bureaucracy is to please the
authority, rather than the clients it is supposed to
serve (Niskanen op cit., 27). Usually this is reinforced
by an important disparity in the information available to
the bureaucrats, compared to sponsors. As a general rule,
the bureaucrats will have much more information. The
bilateral monopoly is thus transformed into a one-sided
In this sense, bureaucracy will act rationally if its
objective is to maximize the budget in terms of obtaining
more salaries, reputation, or power. The institutional
arrangement common in most democracies pushes bureaucrats
to maximize their budgets, generating a system that
enhances the satisfaction of the sponsor, not of the
In order to correct this situation, Niskanen proposes to
break the monopoly on information of the bureaucracy so
that market mechanisms lead to a more efficient outcome,
to increase the competition between bureaucracies
producing similar products, to transform the incentives
of the bureaucracy to encourage efficiency-driven
behavior, and to increase competition by permitting
private producers to enter the process (op cit, 228) .

Some authors have tried to test the arguments and
assumptions of this model. However, they have found
several problems when applying the model.
Lynn (1991) asks if cases actually exist where the
assertions of the model could be tested. Based on the
extensive analysis of diverse cases, he concluded that
although we can observe a tendency of the bureaucracy to
keep and maximize the budget, it cannot be concluded that
economic motivations are the predominant influence. For
example, in the U.S. it is common to find cases where the
bureaucracy prefers to maintain the same level of budget,
rather than trying to obtain large increases. To push for
large increments in budgets might well be politically and
organizationally difficult.
Moreover, given that bureaucracies have to deal with
complex social problems, it is easier and rational to be
oriented toward the "input" (or what is the same thing)
to be oriented to the activity rather than a concrete
outcome. Bureaucracies administer programs, something
that is easier than pursuing abstract and complex
objectives. This is not totally inconsistent with the
model of the maximizer bureaucrat, but the concept of
minimum cost, the relation of inputs to outputs, and the
incentives of the government to disseminate information
about costs are ambiguous in the practice. The
maximization of the budget is only one of the many ways
in which the bureaucracy can maximize its utility (Lynn
1991, 68) .

In any case, the argument of the "maximizing budget
bureaucrat" shows that the bureaucracy is an organization
where individuals have the power to take decisions and to
negotiate outcomes. As for this research is concerned, it
is very useful to note that bureaucracies form part of a
complex system of institutions and that they are able to
define outcomes and parameters of evaluation.
2.6.2 Postbureaucratic Paradigm
In this subsection I review a series of articles that
propose the replacement of the bureaucratic system of
organization. The arguments contained in these articles
have been widely diffused and have inspired concrete
government programs, such as the U.S. National
Performance Review (Gore, 1993). The aim of this section
is to point out that, according to these articles,
whether bureaucratic or postbureaucratic, governmental
action is showing more sensitivity in its relation with
other groups in society. This has become increasingly
important because it is difficult for government to
achieve its goals since it must act in a pluralistic
arena where various groups interfere and have access to
the decision-making process. The postbureaucratic
paradigm, regardless of the substantial debate that has
been generated, shows the current difficulty and
importance of the relationship between government and
society, and of the search for new ways of reaching a
mutual understanding, a principal topic of this research.
Thus, the idea that it is necessary to develop a new form
of governmental organization is also an important issue

in this research. Barzelay
Based on a case study of the State of Minnesota's
Striving Toward Excellence in Performance (STEP) program
implemented during the 1980s, Michael Barzelay (1990)
formalizes the characteristics of what he called the
postbureaucratic model.
The bureaucratic paradigm taught, according to Barzelay,
that the efficiency of the governmental action was due to
the specific delegation of authority, basically acting
according to established procedures and uniform rules
(Barzelay 1992, 5). Bureaucratic thinking takes place in
terms of expenses and procedures, and thus does not take
into account directly the objective nor does it try to
figure out the most efficient method to reach the goal.
Barzelay predicts that the bureaucratic paradigm will be
unable to function, given the complex society in which it
must act today and in the future.
Facing a more demanding society, the "postbureaucratic
paradigm" proposes that governmental agencies embrace
different principles.
These principles place a priority on the needs and
perspectives of the "customer." A postbureaucratic agency
is "customer-oriented," defining itself by the results
that it achieves for its customers. Therefore, a
"customer-driven" agency creates value net of cost, and

modifies its form of operation in response to the
changing demands for its services (Barzelay op cit., 8-
9) .
The values of the public agencies must be directed toward
new words: customer, quality, value, service, incentives,
innovation, and flexibility (Barzelay, 1994) . Osborne and Gaebler
The initial argument of Osborne (1988) is that a new
synthesis is being created in the relationship between
government and society in the U.S. If the traditional
liberalism was the thesis and the conservatism of Reagan
the antithesis, the innovative relationship that diverse
State governments have developed with their citizens
offers a new synthesis. The thesis was to observe the
private sector as the problem and the government as the
solution. The antithesis was to observe the government as
the problem and the private sector as the solution. The
synthesis redefines the nature of both --the problem, and
the solution. It defines the problem as the changing role
of government in the international market, and the
solution as a new relation between private and public
institutions. If the thesis was the government as the
solution and the antithesis was the government as the
problem, in the synthesis, the government is a partner
(Osborne, 1988) .
To constitute the new "partner," however, a radical
transformation of the government is necessary, converting

it into an efficient organization, with entrepreneurial
attitudes. In other words, the government must be
Osborne and Gaebler (1992) proposed ten measures to
achieve this:
1. To make the government a catalyst, in other words, to
have the government steer the rudder rather than to row.
This means that the government should be freed from the
heavy load of delivering services that private
organizations can do as well, or better.
2. The government belongs to the community. The
government must facilitate community action to resolve
the problems it faces.
3. When the government must deliver a service, it must
also be forced to compete. Competition forces
organizations and agencies to think about the customer,
and to reward innovation and not procedures.
4. The agencies of government must be guided by the
realization of a mission, not by the rules.
5. A government should be evaluated by the results it
achieves, not by the way in which spends its resources.
6. Government agencies must aim to satisfy the needs of
their customers, not of the bureaucracy.

7. The government can be entrepreneurial, generating
resources and not only spending them.
8. Public agencies must participate in the understanding
and the solution of existing problems.
9. The government must be decentralized, enhancing team
work over hierarchy.
10. The government must be market-oriented. The Limits of the Postbureaucratic Paradigm
Dilulio, Garvey, and Kettl (1993) propose to make a
dispassionate review of the possibilities of applying
theories of private administration to public
administration. The reinventing government approach, for
example, endorses ideas that nobody could reject (who can
be angry at the ideas that propose that the government
treat its citizens well?) However, good intentions are
not enough. First, the authors argue, it is important to
reduce the excessive simplification that the
postbureaucratic paradigm uses to define bureaucracy. It
is necessary to separate what the government must do and
what it does well. What critics say are programs that
encourage squandering the public's money, fraud, and
other forms of abuse are often programs well-administered
but aimed at achieving goals difficult to understand in
the private sector (such as increasing equity
generating capacities for organization, for example).
Moreover, technical solutions hide the competitive

differences of interest groups. As Wilson (1989) shows,
even bureaucracies are not a simple and uniform
phenomenon. Their culture and values sometimes change
from agency to agency. Nobody should be surprised then
when the promised results are not reached due to
conflictive programs where not every group agrees and
where different agencies participate with distinct
perspectives (Dilulio, Garvey, and Kettl 1993, 9).
Moreover, bureaucracies can be seen also as spaces of
organized anarchy (Cohen, March, and Olsen 1972; March
and Olsen 1976, 1983). Procedures and norms are just
general mechanisms to approach agreement, so
bureaucracies are also capable of managing ambiguity in
order to achieve results.
The implementation of governmental programs involves the
coordination and confrontation of many groups and
interests (Heclo, 1977). The trend of believing that
technical, neutral, market-oriented approaches will
better achieve social needs and thus of trying to
substitute the more traditional political and social
conflictive process is always a danger stemming from this
kind of proposals (Garcla-Pelayo 1974, 62).
The postbureaucratic paradigm and reinventing government
approaches can be seen as managerial approaches.
Managerialism is not only a technical approach to social
problems but can also be seen as an ideology, where any
deviation from the organizational (or social) goal is
seen as a pathology than can be corrected through the
adaptation of adequate values and managerial techniques

(Enteman 1993, 156). The temptation to think that a
correct and technically-driven value can substitute for
the conflictive and disordered social and political
processes is always present in this type of approaches,
and could be regarded as a threat to democracy (Enteman
1993, 159) .
Methodologically speaking, the postbureaucratic
literature is based on cases of successful
implementation. Using the philosophy of "best practices,"
they focus on cases where policies that have been defined
as adequate are applied and furthermore have been
successful (Overman and Boyd 1994). The use of the "best
practice" approach, implies that bureaucracies do not
learn from experience. Observing particular cases of
success does not allow scientific validation in the long
term. Today's innovation and success often become
tomorrow's problems. Moreover, the transferability of the
experiences is an illusion. Every case is observed under
specific conditions, where diverse factors combine in
unique forms. In other words, this kind of research
cannot be tested (Overman and Boyd 1994, 79).
Furthermore, Overman and Boyd continue, the short time
that advocates of this approach dedicate to the study of
other analysts that have dealt with bureaucracy causes
them to tend to reinvent theories rather than make
discoveries. Thus, the "use of market-like incentives to
achieve public results" has been considered already for
example in Dimock (1959) and in Schultz (1977). Osborne
and Gaebler rediscover the topic under the title "market-

oriented government." Even when bureaucracies are market-
oriented, management innovation in the public sector
requires a transformation in the structure of public
organizations (Thompson 1969, Britan 1981).
The postbureaucratic literature has important
implications for this research. Regardless of its
methodological limits, this proposal shows that
government and society are developing a new type of
relationship. This new relationship, by enhancing
efficiency and the quality of government service, is also
showing the need to generate new organizational forms to
better deal with social problems. Moreover, this debate
shows that changing bureaucracies and creating a new
paradigm is a difficult and complex endeavor.
2.6.3 Democracy and Bureaucratic Action
Another important set of articles that must be addressed
is the one that links bureaucracy and democracy. Opening
up a space where ordinary citizens can participate in the
resolution of public issues is difficult when there
exists a specialized corps of bureaucrats that deal
professionally and technically with most public issues.
Participation, in this case, can be a major problem, an
unnecessary burden where the urgent issues require quick
and professional action.
A review of some of the texts that deal with this problem
seems important for this dissertation, since it focusses
on the relationship between participatory community-based

organizations and the bureaucracy. The dilemma of
choosing between bureaucratic professional decision-
making or democratic participatory mechanisms appears in
the literature as one of the most problematic and
indeterminate issues in today's literature (Dahl 1989).
In this subsection, I choose just a few representative
books (using one as a guide for each subsection) from the
vast literature on the issue, in order to articulate the
arguments that most clearly apply to this research. Participation and Bureaucracy
Grises and Robert Kweit (1981) explain that the
participation of society in the solution of public
problems is not seen in the U.S. as the normal path to
follow in terms of the political system. Most U.S. people
consider that elected representatives with their
professional bureaucracy are responsible for making
political decisions, and that the participation of the
people (generally unprepared and unskilled for these
tasks) is unnecessary.
When public participation of society has been seen as a
panacea, the high expectations that accompany it have
tended to be disappointed.
In sum, there are two major sets of problems in implementing
citizen participation: first, there are the problems of
democratic expectations on government structures that were never
intended to function democratically, and second, there are the

unrealistic and conflicting expectations (Kweit and Kweit
1981, 8).
The experiences of the War on Poverty during the Johnson
administration in the U.S. placed the issue of
participation in the center of the debate. It was argued
that participation would educate citizens about
activities, problems, demands, and conflicts of
government. It was also argued that participation in
politics would increase the citizen's sense of efficacy
and would reduce, therefore, the citizen's alienation
from the government.
However, there are problems. The first problem is the
difficulty in defining objectives. This is one of the
basic reasons why the relationship between citizens and
bureaucrats are so conflictual when participatory
mechanisms are implemented. Discussion and dialogue do
not necessarily produce good solutions. The negotiation
between officials and citizens (the fight, for example,
between influence and social control) can be very
The necessity of balancing the expected benefits of
participation with the inevitable costs, is another
important preoccupation (Kweit and Kweit 1981, 37) .
Usually, participation involves time and effort to set up
a structure to allow participation. Negotiations are
time-consuming, and public problems usually need to be
resolved quickly.

In sum, the combination of citizen participation and
government bureaucracy generally leads to contradiction
and conflict. The bureaucracy has a logic: regularity,
professionalism, optimality, a political orientation;
while the citizen participation is non-professional,
based on negotiations, consensus, and socially-
orientated. Democracy and Bureaucracy
Yates (1982) analyzes what he calls the bureaucratic
democracy, the effect on modern societies of two
different but related necessities: pluralist democracy
and administrative efficiency.
Those who emphasize the importance of a pluralist
democracy, the pluralists, stress the "circularity" of
power, the limitations of bureaucratic power, and their
confidence that the pluralistic approach leads to the
"right" solution. Advocates of the need for
administrative efficiency emphasize rational decision-
making as the best way to make decisions. Yates (1982,
13-32), furthermore, argues that:
1. In the pluralist model, power is dispersed and divided; in
the efficiency model, power is concentrated...
2. In the pluralist model, there is a suspicion of executive
power,- in the efficiency model, great emphasis is placed on
centralizing power in the hands of the chief executive.
3. In the pluralist model, power is given to politicians,
interest groups, and citizens; in the efficiency model, much

power is given to experts and professional bureaucrats.
4. - In the pluralist model, political bargaining and
accommodation are considered to be at the heart of the democratic
process; in the efficiency model, there is a strong urge to keep
politics out of administration.
5. The pluralist model emphasizes individuals' and political
actors' own determination of interest and utility; the efficiency
model emphasizes technical or scientific rationality...
Bureaucracies are part of the political system, so they
cannot be disregarded when solutions to public problems
are necessary (Meier 1987, 3). Yates proposes that the
participation of citizens should be limited to the local
level and when there are very clear problems to resolve.
The rational implementation and long-range planning
should be in the hands of bureaucracy. The street-level
bureaucracy (Lipsky 1982) should maintain close links
with citizens, therefore, bureaucracies should enjoy a
larger scope for discretionary action. Representative Bureaucracy
Krislov and Rosenbloom (1981) ask whether a public
bureaucracy can be transformed into a representational
institution despite such organizational features as
hierarchy, specialization, and formalization.
When confronted with participation from citizens,
bureaucracies face a dilemma. Where participation is
confined to elites or groups that represent primarily
their own interests, it can be an effective device for

promoting individuals' interests. Bureaucracy in this
context does tend to be highly representative and to have
a good deal of legitimacy, at least among its clientele.
In such situations, participation can serve both the
needs of interest groups and those of the bureaucratic
agency. However, when attempts are made to involve
nonelites, bureaucracies are more likely to serve the
interests of the bureaucratic agencies than those of any
given target group (Krislov and Rosenbloom 1981, 3).
The contradiction between bureaucracy and participation
must be addressed by any attempts to reform public
administration. Despite their different approaches, the
above mentioned authors share the conviction that the
bureaucracy will nevertheless be able to find new avenues
of representation and participation.
2.7. Final Remarks: Bureaucracy as a Complex Organization
Through out this research, the bureaucracy is seen as an
organization that tries to control uncertainty. From
Crozier's theory, we assume that, when confronted with
uncertainty, bureaucrats will try to reduce it, defining
and pushing through more precise formal mechanisms.
However, uncertainty makes it possible for actors within
the organization to redefine strategic organizational
spaces. These organizational spaces are defined by
Crozier and Friedberg as uncertainty areas and are used
by organizational members to egotiate their informal and
formal position within the organization. What actors seek
is a position which allows them to redefine

organizational rules, norms, meanings, and symbols to
their own benefit. Thus, negotiation appears as an
important instrument that allows actors to rearrange
authority and power within the organization. As seen in
Crozier's case studies (1964), in an attempt to reduce
uncertainty, actors renegotiate or even increment
uncertainty on purpose, in order to negotiate new
organizational structures and rules. Particular actors
even can manipulate ambiguities in order to increment
their strategic space of freedom. Since it is a
contingent process, the negotiation between actors can be
categorized through a definition of formal mechanisms
(objectives, norms, and authority framework).
The next chapter uses this theoretical framework in order
to formalize and support the research proposal framed in
the introduction of this research.

3. Methodology. The Nature of Case Studies and the Design
of the Comparative Case Study Framework
This chapter develops the methodological framework of the
research project. The first part of this chapter contains
methodological discussion regarding the inductive method.
The aim here is to discuss the limits and potential of
inductive methods such as case studies, and to make clear
the difference between case studies and probabilistic
inductive methods.The second part of the chapter develops
the design of the methodological approach for this
research. It justifies the use of comparative case
studies, and specifies the elements that make this
research valid.
3.1 The Inductive Method
3.1.1 Some Definitions
One of the most widespread misconceptions in logic
analysis is the belief that deductive arguments proceed
from the general to the specific, and inductive arguments
proceed from the specific to the general. An argument is
deductively valid if and only if it is impossible that
its conclusion is false while its premises are true. An
argument is inductively strong if and only if it is
improbable that its conclusion is false while its
premises are true, and it is not deductively valid. The
degree of inductive strength depends on how improbable it

is that the conclusion is false while the premises are
true (Skyrms 1966, 7-13). One of the most difficult parts
of inductive logic is to define the probability, because
inductive arguments are not bad or good, but rather
simply are weak or strong.
The methodological discussion on case study analyses is
firmly attached to the more general discusion of the
problem of induction. A problematic process, where one
observes that the event A is followed by event B on one
occasion or several occasions, has been described by Hume
(1964 [1886]) as a situation where it is impossible to
logically conclude A will be followed by B on any or
every other occasion. Since Hume, this problem is known
as the problem of induction.
Several solutions have been sought for this problem,
although there is no consensus about the unique
scientific process that would guide any inductive
The next section summarizes the descriptions of these
solutions, and the following section concentrates on two
of them, which take two extreme positions in the current
debate on the induction problem. I will refer these two
solutions as the pragmatic and falsification solutions.
Following Rescher (1980, 187) we can classify the
different solutions to the induction problem as follows:
(I) Induction is dispensable; it is not needed, and

should be replaced by a fundamentally different process
of inquiry which proceeds simply by eliminating untenable
(II) Induction (while needed) cannot be justified.
(III) Induction does not need to be justified: it
requires no justification, either (A) on the ground that
induction is a perfectly natural process that does not
rest on any discursive considerations at all, but on a
purely instinctive or intuitive basis [INTUITIONISM
(HUME, ACADEMIC SKEPTICISM)]. Or (2) because all
justifications must stop somewhere and induction is part
AYER)] .
(IV) Induction can and should be justified, and such
justification is forthcoming through: (A) grounds of
logico-conceptual necessity by way of (l)
considerations of strictly demonstrative necessity
considerations of hypothetically demonstrative necessity
with regard to goal-attainment ("transcendental
arguments," "this-or-nothing argumentation") [CONDITIONAL
theoretical considerations regarding the modus operandi
of probability as a quasi-demonstrative device
(B) empirical-inductive grounds. INDUCTIVISM

(C) metaphysical grounds. UNIFORMITARIA DEDUCTIVISM
(D) methodological grounds. PRAGMATISM (RESCHER).
The different solutions to the problem of induction
basically take two extreme positions: induction is
useless and should be replaced by other methodological
instrument, or induction is still an important
methodological device for science, only if its
methodological procedures are carefully designed and
defined, and its limits are clearly stated. These two
different positions are well explained by the
falsification and the pragmatist approaches. In order to
define the position taken to develop this research, it is
important to make a deeper analysis of these two
scientific techniques. The next section explains the
discussion regarding pragmatism followed by a section
discussing falsification.
3.1.2 Pragmatism: Induction as Cognitive Systematization
(based on Rescher 1980)
The methodological problem of induction resides in those
processes where questions arise and the information on
hand is not sufficient, or where the situation is a very
complex and intricate one. The crucial thing about
induction is its movement beyond the evidence in hand,
from informatively lesser data to relatively larger
conclusions. Induction is an instrument for question-
resolution in the face of imperfect information. The goal

is not the best possible answer but the best available
answer, the best we can manage to secure in the existing
The term induction' is derived from the Latin rendering of
Aristotle's epagoge the process for moving to a generalization
from specific instances. Gradually extended over a wider and
wider range, it has ultimately come to embrace all non-
demonstrative argumentation in which the premises do (or are
purported to) build up a case of good supportive reasons for the
conclusion while yet falling short of yielding it with the
demonstrative force of logical deduction (seeing that it always
remains logically possible with inductive arguments to admit the
premisses and deny the conclusion) (Rescher 1980, 10).
Induction is not so much a process of inference as one of
estimation. Its conclusion is not so much extracted from
data as suggested by them. The task is to accomplish this
in the least risky and least problematic way, as
determined by plausible best-fit considerations.
Induction leaps to its conclusion through some extractive
process instead of deriving it from the given premises.
"[C]onclusions are not derived from the observed facts,
but invented in order to account for them" (Hempel 1966).
The necessity of induction arises when the body of
explicitly given information is insufficient for
determining which of the possible answers to our question
as correct. Induction is a family of methods (including
the analysis of cases) for arriving at a best estimate of
the correct answer to questions whose resolution
transcends the reach of the facts at hand.

The task is to provide an answer that is qualified to
serve as our truth-surrogate in factual contexts.
Induction must conform to the usual rules and
requirements for estimates in general, and this process
is called systematization:
1. - Character requirement. Estimates must have exactly
the same character as their estimanda.
2. - Uniformity requirement (reliability). A process of
estimation must be consistent in that it must yield
similar results in informationally similar circumstances.
3. - Coordination requirement (data sensitivity).
Estimates must correlate positively with the structure of
their data-base.
4. - Correctness-in-the-limit requirement (consistency).
That all arguments follow a sequence and are consistent
with each other.
5. - Accuracy requirement (validity). An estimation
process should in general yield estimates that are close
to the truth, insofar as this is verifiable.
The idea of pragmatism is that systematic analysis leads
to optimality and plausibility. Induction seeks to
present the best available answer to our questions. The
"best available" answer is formed by eliminating
implausible answers, and this can only be achieved by
systematization. The best estimate of the truth is equal
to the optimal systematization at given circumstances.
Systematicity becomes the test of truth, the guiding
standard of truth-estimation. The operative transition is
not from "systematic" to "correct" but rather from

"systematic" to "rationally claimed to be
correct".(Rescher 1980, 37)
Induction is certainly not a sure-fire device for getting
correct answers (as Hume has shown, this is an impossibility by
the very nature of the case.) Instead, the best we can hope to
show is that induction is a means for doing the job of truth-
estimation as well as it is possible to do in the epistemic
circumstances in which we live and work (Rescher 1980, 55)
Rescher gives two justifications for this argument: faute
de mieux (no more promising alternative lies at hand and
its use is relatively risk-free); and experiential
"retrojustification". How can we be certain that the
local and apparent regularities that these scanning
efforts detect in our observational neighborhood are
actually global and real regularities? The answer is
simple: we cannot be certain of this, nor indeed even
establish it with high probability. Hume's consideration
that we cannot demonstrate that induction yields reliable
results is related to the fact that we cannot either
demonstrate that memory or sensation do so. The
justification of induction is a posteriori. Any attempt
to justify induction is futile because the inductive
style of argumentation serves to define what rational
argumentation in inductive contexts is all about: the
rationality of induction is implicit in the very
conception of "rationality" as such.
3.1.3 Falsification (Popper)
For some authors, induction problem or "Hume's problem"

(i.e., that it is impossible to show that induction gets
accurate interpretations of reality) has shown that
assumptions concerning the regularity of reality cannot
be secured. Thus, they argue that empiricism is not a
sufficient basis for science. Bertrand Russell wrote that
induction is an independent logical principle, incapable
of being inferred either from experience or from other
logical principles, and that without it science is
impossible (Russell 1945). To find that science should
rest on foundations whose validity it is impossible to
demonstrate has been an embarrassing and problematic
discovery for some people (as in Russell 1961).
Magge (1973) claims that Popper's basic achievement has
been to offer an acceptable solution to the problem of
induction. Popper's solution begins by pointing to a
logical asymmetry between verification and falsification.
Empirical generalizations, though not verifiable, are
falsifiable; they can be tested by systematic attempts to
refute them.
We cannot prove that our arguments are true, but we can
justify our preference for one theory over another. The
popular notion that the sciences are bodies of
established facts is entirely mistaken. Nothing in
science is permanently established, nothing unalterable.
Popper's notion of "the truth" leads to a pursuit of
knowledge which brings us closer and closer to the truth,
even though we can never be sure if we have reached our
goal. Every theory can still be replace by a better

The induction process is a problematic one for Popper,
for three reasons. First, how one arrives at a particular
theory has no bearing on its scientific or logical
validity. Second, the observations and experiments are
partially derived from the theory itself, and are
designed to test it. Third, at no point does induction
come into the matter. So induction, Popper says, is a
dispensable concept, a myth. It does not exist, there is
no such a thing (Popper 1968), and when someone says that
by induction we can arrive at possible theories, this is
a psychological process, not a logical one. The creation
of theories has no logical process. There is no "clean"
process of observation. Observation cannot be prior to
theory as such, since some theory is presupposed.
Observation is always selective. It needs a chosen
object, a definite task, an interest, a point of view, a
problem. This means that observations, and even more so
statements on observed phenomenon, and on experimental
results, are always interpretations of the facts
observed, interpretations in the light of theories
(Popper 1968, 107).
According to the traditional inductive approach, what
scientists are looking for are statements about the world
which have the maximum degree of probability, given
certain evidence. Popper denies this. Highly informative
statements yield a lower probability of reoccurrence.
However, those statements with high level of information
are the important ones, since they are highly testable

and falsifiable (Popper 1957, 57) Our ignorance grows
with our knowledge, so we should always have more
questions than answers.
According to Popper, falsification in whole or in part is
the anticipated fate of all hypothesis, and we should
even rejoice in the falsification of an hypotheses.
Falsification is the criterion of demarcation between
science and nonscience. Verification, in the way
positivists handle it, is impossible. Moreover,
advocators of verification as the criterion observe
metaphysics as meaningless, when those metaphysics can be
often meaningful.
Another important concept in Popper's argument is
verisimilitude. The verisimilitude of a statement is
defined as its truth content minus its falsity content
(O'Hear 1980, 29) This definition enables us to make
comparisons of verisimilitude among different theories.
It is necessary to stress that neither falsification nor
verisimilitude are intended to have inductive overtones.
Reliability is not a possible concept for Popper, since
nothing in logic can assure future performance from past
performance (Popper op cit., 18).
The scientific method for Popper is:
1. -Problem (usually rejection of existing theory or
2. -Proposed solution, or new theory

3. - Deduction of testable propositions from the new
4. - Tests, attempted refutations, observation and
5. - Preference established between competing theories.
This can be expressed schematically as:
PI------> TS------> EE------> P2
where PI is the initial problem, TS the trial solution
proposed, EE the process of error elimination applied to
the trial solution and P2 the resulting event (Magee
1973, 61). It is essentially a feedback process. It is
not cyclic, for P2 is always different from PI: even
complete failure to solve the problem teaches us
something new about where the difficulties lie, and what
the minimum conditions are which any solution must meet.
This research takes in consideration these arguments of
Rescher and Popper, since it is clear that case study
method is part of inductive methods. I argue that the
general response to the induction problem is clear: there
is no absolute solution. There is no possibility or
ensuring that, in situation characterized by lack of
information or very complex circumstances where it is
difficult to isolate variables, we will find the same
conclusions over time. However, I do agree with Rescher
that inductive methods are useful to approach reality by
developing some plausible interpretations through
systematization. And I do agree with Popper that we must

develop our research design in such a way that we can
allow falsification by further scientific attempts. These
provisions do not ensure that our interpretations are
correct, but they at least allow us to argue for its
capacity to explain observed reality (through
systematization) and to follow the argument step by step,
even to falsify it.
Case studies' design should be developed in a systematic
way (as defined using Rescher's argument above) even to
allow falsification. In the next section I develop the
design for a systematic comparative case study for this
research, attempting to follow the requirements for

3.2 Case Study Design
3.2.1 Case Studies and the Problem o£ Validity and
Case-oriented researchers are always open to the charge
that their findings are specific to the few cases they
examine. Thus, when they make broad comparisons and
attempt to generalize, they often are accused of letting
their favorite cases shape or at least color their
Quantitative methods emphasizing probability statistics
and predictions based on falsifiability criteria have
created two basic dilemmas for research based on case
studies, according with Yin (1989). First, the
probability is high that bias or lack of objectivity
stemming from the study of few cases out of a
subjectively selected non-random sample will be a
problem. In this sense, neither reliability nor internal
validity can be ensured through the use of case study
approaches. Consequently, the case study method does not
allow generalization of findings.
Since these are important methodological dilemmas for all
inductive techniques of analysis, the solution boils down
to using methods appropriate to the research situation
and the goal pursued by the research (i.e., the research
process implies questions about the complexity of the
problem and the complexity of the relationships involved,
things that will affect the decision to choose a
methodological approach) .

All research techniques have advantages and disadvantages
in terms of accuracy and validity. In this sense a
researcher advocates also a particular position in the
epistemologic arena and in the kind of results sought.
The current research techniques, quantitative or
qualitative, depend upon researcher's epistemological
position and the kind of problem he/she wants to address.
As Yin (1989) observes, for the social sciences, the
difference in techniques used for quantitative and
qualitative approaches have increased in the last
decades. Some argue, however, that this differentiation
is more a matter of the capacity of the research
technique to deal with different types of research
questions rather than a different ability to avoid bias
and to approach reality more accurately.
For example, it is possible to say that the quantitative
approach does not ensure lack of bias, since every
problem begins with a question from the researcher, based
on different theoretical conceptions. Moreover,
quantitative analysis does not allow accurate
explanation of reality per se, because quantitative
analysis can show correlations, and correlation is not
the same as causation. Causation is a process that is
difficult to observe through partially correlated
variables. If the research question is talking about a
process, statistical methods are handicapped to study
those kind of questions (Stoecker 1991, 94).
Herre (1979) makes a distinction between intensive and

extensive research designs. The argument here is that it
is more a matter of different problems addressed "than
capacity to explain reality better." Extensive research,
which includes primarily large scale quantitative methods
such as samples and statistical analysis, is most
concerned with mapping common patterns and properties of
a population in a descriptive logic. Intensive research
uses primarily qualitative methods for the purpose of
holistic causal analysis. The problem with using
statistical methods for analysis of causal situations is
that they require the "taxonomization" of observations
into groups of variables that share common formal
attributes, but that are not necessarily connected or
interrelated with one another. In intensive research, the
causal relationships are based on observable concrete and
particular interconnections between actual properties and
people within an actual concrete setting that
interrelates in complex patterns difficult to taxonomize.
Case-oriented methods are basically intensive/qualitative
methods. The general characteristic of case studies is
that attempt to be holistic, to treat cases as whole
entities and not as collections of parts. These methods
treat causation as a contingent phenomenon. That is to
say that outcomes are analyzed in terms of intersections
of conditions, and it is usually assumed that any of
several combinations of conditions might produce a
certain outcome. (Ragin 1987, x)
Unlike quantitative multivariate statistical analysis,
which tends to be radically analytic (because it breaks

cases into parts --variables-- that are difficult to
reassemble into wholes), qualitative comparison allows
examination of constellations, configurations, and
circumstances. It is especially well suited for
addressing questions about outcomes resulting from
multiple causes --where different conditions combine in
different and sometimes contradictory ways to produce the
same or similar outcomes. Multivariate statistical
techniques start with simplifying assumptions about cases
and their interrelation as variables. The method of
qualitative comparison, by contrast, starts by assuming
maximum causal complexity and then builds upon that
complexity (Ragin, x). Complexity is accepted and then
analyzed in terms of multiple causation.
Lofland (1971, 13) argues that qualitative analysis
answers questions such as "what are the characteristics
of a social phenomenon, what forms does it assume, and
what variation does it display?" Qualitative analysis
addresses the task of delineating forms, kinds, and types
of social phenomena. Causality is assumed as multiple and
complex, and much importance is given to particular and
specific historical phenomenon that might explain it. On
the other hand, quantitative analysis answers questions
such as "what is the correlation between standardized
elements and the consequences of a social phenomenon?"
Finally, the researcher designing a case study analysis
must take into account the validity and reliability of
his/her data, and must develop clear criteria for
interpreting and analyzing these data. Validity and

reliability are necessary criteria to evaluate the
capacity of accurate analysis of causation and perhaps
the capacity of prediction of a particular methodology.
Even though case studies are not used to predict or to
identify specific correlations (as causations), these
criteria are important rules of scientific inquiry. While
no other criteria have been formulated so far, every
research project should at least justify the use or
dismissal of those criteria.
For case study research design, the first important thing
is to define its basic objective: analysis of complex
causation to propose a plausible interpretation of the
phenomenon. The possibility of case study analysis is to
understand causation in a real, specific setting, without
trying to generalize. The goal is not to predict, but to
understand. Increased understanding allows us to create
alternatives to current situations, to define positions,
to risk new concepts and possible relationships. From
this understanding there might appear a plausible but
speculative prediction or tentative generalization useful
for program design and professional practice. It is over
the concept of understanding that case studies should be
concerned with:
a) internal validity. That the concepts, categories and
relationships are actually useful and used by different
b) external validity. Case studies cannot define general
laws and cannot be used to obtain general patterns,
although the results and the interpretation can be

convincing enough to generate ideas and alternatives for
different settings; and
c) reliability. Case studies cannot always be repeated,
because we are analyzing complex phenomenons in time,
within social settings that change over time. In other
words, case studies are not experiments. Nevertheless,
the design should be explicit enough in order to allow
reconstruction of arguments by other persons and
falsifiability of interpretations.
In the following sections, I design the research process
attempting to deal with all these methodological issues.
Since the research is proposing a multiple case study,
the following discussion intends to develop a valid and
reliable framework for the field work and for the
interpretations to be developed throughout the study.
3.2.2 Research Problems and Questions
As part of the current process of decentralization, the
Mexican Federal government is proposing to transfer
responsibility for social programs to municipalities.
However, high numbers of Mexican municipalities are
administratively and financially incapable of dealing
with complex social problems (Cabrero 1996). Since 1989,
the Federal Government of Mexico deals with that lack of
capacity by encouraging involvement of traditional,
community-based organizations in the implementation of
federally-funded, locally administered social programs.
Specifically, federal resources allocated to social
programs are often tied to a requirement that members of

local communities being served also participate in
program implementation (as is the case of the National
Solidarity Program, or Pronasol). This process has
generally involved community-based organizations (CBOs),
which has led to the creation or strengthening of
independent community-based organizations.
These organizations have to coordinate their efforts with
local municipal government bureaucracies. Recent studies
of municipal capacities and innovative community
organizations in Mexico (Cabrero 1995) have identified
several problems resulting from the need for coordination
between government bureaucracies and CBOs. For example,
municipal public agencies sometimes resist cooperating
with community organizations, and CBOs sometimes find it
difficult to understand the needs and requirements of
public bureaucracies.
In Mexico, there are 2,412 municipalities. Most of them
have participated in governmental social programs like
Pronasol. I selected the four cases from municipalities
that have participated in the process of planning,
implementing, and coordinating social programs, requiring
cooperation between federal and state agencies, municipal
authorities and CBOs.
This research examines the relations between governmental
bureaucracies that have to coordinate social programs,
and community-based organizations, involved with the
implementation of social programs. Where CBOs have
emerged as an option for dealing with complex social

problems requiring flexibility and participation, and
where governmental agencies retain a dominant role in the
control of resources for those projects, the impact on
the CBOs ranges between two extremes: (1) the collapse of
the CBO due to increased bureaucratization imposed by
government agencies, or (2) legitimization of the CBO by
the bureaucratic and political establishment allowing the
survival of the CBO as an autonomous and flexible
organization. Between these two extremes, several
intermediate outcomes are also possible.
My analysis of the case studies focuses on the following
Which factors appear to cause collapse of CBOs due to
increased government rigidity and bureaucratization?
Which factors lead to legitimization and survival of
CBOs? What other alternatives within the continuum exist,
and how are they generated?
Although I do not rely entirely upon the theory of
bureaucracy developed by Michael Crozier, I do use his
concepts to help observe and explain a process by which
CBOs tend to create crises for the bureaucratic
establishment, leading in turn to increased
bureaucratization and rigidity within municipal
government agencies. The process by which this occurs
involves reaction by government bureaucracies to
uncertainties created by CBOs involved in implementation
of public programs.

More specifically, I propose to observe crises resulting
from the inability of government agencies in controlling,
managing, and supervising CBOs' flexible strategies and
participation frameworks. This inability creates
uncertainty within bureaucratic agencies. I propose then
to link these factors to a continuum of outcomes ranging
from "collapse" to "survival" of the CBO.
The crisis generated in this relationship causes a
redefinition of strategic spaces of diverse groups within
both organizations. I assume that the redefinition of
space takes the form of negotiation (between
bureaucracies and CBOs) concerning the formal
organizational values of the CBO. The government agencies
use the negotiation process to push the CBO toward a
clearer definition of procedures, objectives, and
authority structures, jeopardizing the flexibility and
ambiguity that allows a CBO to gather the community
together. The negotiation process also jeopardizes the
autonomy of the CBO, involving its ability to maintain
its base of authority within the community. Negotiation
between CBO and bureaucracy can induce the CBO to define
clearer norms, objectives, and authority frameworks,
while maintaining its autonomy and its flexibility.
3.2.3 Basic Definitions
A "community-based organization" (CBO) is an organization
that was formed within a particular community and is
characterized by the continuing existence of open and
free participation by community members as the principle

mean of resolving common problems. The fact that
participation is free and open is what gives the CBO its
high degree of legitimacy among community members.
In Mexico, CBOs are often formed around community
traditions and social relationships, which in turn are
based on local geography and history. Usually, in the
Mexican case, CBOs lack legal existence and a formal
configuration or structure. In general, these
organizations are rooted in pre-Columbian indigenous and
"mestizo" traditions formed during Colonial times.
The operational definition for CBOs in this project is:
an organization rooted in one or several communities,
that advocates, as a fundamental principle of its
existence, the participation of community members as the
main mechanism for resolving problems. In other words,
CBOs are defined as being rooted in the traditions of the
people of one or more communities, and as explicitly
justifying and practicing popular participation as its
basic approach to organizing.
"Bureaucratization" means the imposition, by established
governmental agencies, of specific and strict
regulations, norms and criteria of action on CBOs or
other organizations. Those criteria restrict action by
means of comprehensive rules, thereby reducing the CBOs'
capacity for flexible and open participation. The basic
elements of bureaucratization are identified by means of
organizational characteristics defined by Crozier:
extensive development of impersonal rules, centralization

of decisions, and isolation of internal organizational
"Legitimization" or "survival" are viewed as the
recognition by the bureaucratic establishment of the
existence and freedom of a particular CBO, expressed by
the acceptance of and non-interference with its
flexibility and autonomy.
"Collapse" is defined as the disappearance of the CBO,
due to its effective termination (Brewer and deLeon 1983,
383-416) as a participant in the implementation of a
public program.
These research argues that a CBO attempts to resolve
complex social problems with a program or activities that
require community participation, flexible norms and
objectives, and elastic authority frameworks.
Norms/objectives and authority framework are defined as
key concepts for this research. The operative definition
of these concepts are as follows:
NORMS/OBJECTIVES: Explicit rules, procedures and policies
that orient the action of individuals and groups within
the organization. Explicit set of outcomes that the
organization is seeking to achieve.
AUTHORITY FRAMEWORK: Implicit and explicit definitions of
responsibility, mechanisms for the enforcement of
obedience, and capacities that different individuals have