Societal influences and governmental responsiveness in policy communities


Material Information

Societal influences and governmental responsiveness in policy communities educational policy-making in Zimbabwe
Physical Description:
xv, 343 leaves : ; 29 cm.
Spear, Mary E
Publication Date:


Subjects / Keywords:
Education and state -- Zimbabwe   ( lcsh )
Educational sociology -- Zimbabwe   ( lcsh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1993.
Includes bibliographical references (leaves 335-342).
Statement of Responsibility:
by Mary E. Spear.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001923476
notis - AJZ9310
oclc - 30541259
System ID:

Full Text







Copyright 1993


Mary E. Spear

To my family--mom, dad, and sister Nova--who have

offered me endless love, support, kindness, and



I would like to thank a number of people who have made

this dissertation possible. I am grateful to members of my

doctoral committee who have encouraged me and offered me

constructive criticism. I want to express my particular

gratitude to Goran Hyden, my chair, for his excellent

advice, patience, encouragement, vision, and his ability to

challenge me. Goran Hyden has served as an excellent role

model of an academic committed to excellent research,

teaching, and service.

I am also thankful to Walter Rosenbaum, David Conradt,

and Rene Lemarchand whose courses and research have inspired

me to develop this dissertation. I want to also thank Steve

Sanderson and Ken Wald for the assistance they have provided

me at the University of Florida. I appreciate all of David

Hedge's helpful suggestions for my dissertation. Working

with both David Hedge and Richard Scher as a teaching and

research assistant has been a major highlight of my graduate

career. I am grateful to Doyle Casteel and Richard Renner

for helping me to better understand education policy. I

thank Doyle Casteel for his suggestions on my dissertation.

I am very appreciative of the support, friendship, and

encouragement of many Zimbabweans who have helped me to make


my dissertation vision a reality. Bernard Gatawa, Boniface

Chivore, and Betty Jo Dorsey, education specialists, have

offered tremendous support to me in the design and analysis

of this research task. Surveys by Betty Jo Dorsey and

Philip Foster in Zimbabwe and Ghana, respectively, have

stimulated my own research tremendously. Solomon Nkiwane

and Gordon Allan of the University of Zimbabwe served as

enlightening professors in 1989.

I am also grateful to the Minister of Education in

1991, Fay Chung, for agreeing to an interview with me, and

other Ministry of Education officials who asked to remain

nameless. My Zimbabwean friends deserve my appreciation for

their kindness and support during my research phase: Betty

Rukasha; Agnes and Nsingo; Matjaka; Rosemary and Veronica

Kalunga; Lilian, Godfrey, and Bongani Ndlovu; Lindie Dube;

Harry, Lynda, and Caroline Fincham; Jeffret Muzira; Pontso

and Mrs. Mafethe; Dr. Zengeni; Lilian and Tom de Chassart;

the Zderbergs; Lymon Betah; and Margarat Nyarota. I want to

thank Vice Chancellor of Africa University, Dr. Kurewa, for

his vision. I am grateful to all the students, teachers and

headmasters involved in my survey--masvita.

I am also deeply indebted to Rotary International and

the Ford Foundation for making my trips to Zimbabwe

possible. While they are in no way responsible for this

dissertation, both Rotary and the Ford Foundation have

offered me support. Peter Fry and Michael Chege of the Ford

Foundation, deserve my appreciation for their support and

friendship. Members of the Harare Central and Silver

Spring, Maryland Rotary Clubs have supported me as a Rotary

Graduate scholar. I am particularly appreciative for the

love and support I have received from Rotarians Tom Lawson

and Tom de Chassart.

I want to recognize Donald Gordon, Professor of

Political Science at Furman University for stimulating my

interest in Africa while I was on foreign study to Kenya.

Other Furman professors to whom I am indebted are James

Guth, Donald Aiesi, and Cleve Fraser. My Furman friends,

Jennie Smith, Kim Opperman, and Lea Alexander, continue to

be a source of inspiration for me.

To my mom, dad, and sister Nova, I am so fortunate to

have such a loving and supportive family who are also

experts on the computer. Kevin Hill has been my best friend

and boyfriend for seven years, I thank him for all the good

times we have had and for his love, friendship, sense of

humor, and insight. I would like to thank other individuals

who have inspired me such as Madeleine K. Albright, Eleanor

Roosevelt, Sissela Bok, Maya Angelou, Martin Luther King

Jr., Mohandas Gandhi, and Vaclav Havel. I want to

acknowledge the words of Martin Luther King Jr., "You ought

to believe something in life, believe that thing so

fervently that you will stand up with it till the end of

your days We have a power, a power as old as the

insights of Jesus of Nazareth and as modern as the

techniques of Mahatma Gandhi."




LIST OF TABLES . .. .. .xii

LIST OF FIGURES . .. .. .xiii

ABSTRACT . . ... .xiv



Overview . . 1
Importance of Studying Education Policy in
Zimbabwe . . 6
The Limitations of the State-Society
Approach . . 9
The Conceptual Building Blocks of the New Policy
Communities Approach . .. 17
Policy Community Approaches .. 28
The Policy Communities Approach .. 32
Explanation of the Model ............ 37
Policy Communities and Their Relationships to
Model and Theories . ... 40
Comparison of the New Policy Communities
Approach with Prior Perspectives .. 45
Governmental Responsiveness to Societal
Influence in Educational Policy-Making 47
Three Hypotheses ... .. 50
The Linkage Between Societal Preferences
(Public Opinion) and Policy .. 55
Methodology ................ 56
The Significance of This Study . .. 62
Organization of the Dissertation .. 66

ZIMBABWE . ... 68

The Important Role That Ethnicity and the Family
Has Played in the History of Education Policy 68


Zimbabwe Education Policy From a Historical
Perspective . .... .. 74
Munhumutapa Kingdom and Societal Pressures on
Education Policy . .... 75
British Missionary and Colonial Influence in
Zimbabwe--1850-1939--Paternalist Phase 78
Rise of African Consciousness and Conflicts
with Government--1940-61--Growing
Awareness Phase ..... 82
Liberation War--1962-79--Liberationist Phase 93
Societal Demands Influence Government Policy
in Independent Zimbabwe--1980-1987--
Populist Phase ... ..... 98
Government-Societal Relations-Mixed Results--
1988-present--Conflictual Phase 102
Conclusion . .... .104

ZIMBABWE . ... 107

Overview . . ... .. .107
Discussion of Components of the Model .. .108
Key Policy Actors and Their Preferences 108
Societal Efforts to Mobilize Their Resources 122
Degree of Openness . ... 133
Degree of Consultation ... .138
Government Responsiveness to Societal
Influence During Decision-Making 149
Conclusion . .... .153


Introduction to University Education Policy 156
Theoretical Perspective on University Autonomy 158
The Origin of the University of Zimbabwe and the
Development of Government-Societal Conflicts 162
A Period of Relative Calm in Societal-
Governmental Relations in University
Policy: 1980-1986 . ... 165
The University of Zimbabwe Act of 1982 167
An Important Year in Tracing the Closure of
the Policy Community to Societal
Involvement: 1987 ..... ... .. 168
Government-Societal Relations at the University of
Zimbabwe--1988-present: An Example of a
Closed Government Policy Community ... .170
Divergent Preferences and Mobilization of
Resources. . ...... 170
Exclusive Decision-Making Under a Crisis
Situation: The Case of the Closure of
the University of Zimbabwe ... .179

The Strong Government Role in the University
After the Closing of the University in
1989 . ... 181
The Closed Government Policy Community
Decides on the UZ Amendment Bill, NUST
Bill and the NCHE Bill .. .184
Societal Reaction to the 1990 Bills ... 192
The Precarious Position of the Vice
Chancellor in University Policy 196
University Demonstrations in 1992 .. .199
Explanations for a Closed Government Policy
Community During a Period of Democratization 201
Conclusion . ... 208


Overview . . ... .. 211
Educational Financing in Zimbabwe .. ... 212
Key Policy Actors . .. 216
Policy Issues in Educational Financing 217
The Emergence of an Open Conflictual Policy
Community . 219
Evidence of the Open Conflictual Policy
Community . ... 222


Overview . . ... .. 230
Preferences .................. 231
The Role of the Ministry of Education's Curriculum
Development Unit. . .... 233
Close Examination of Curriculum Policies ... 238
The Conflict Between Practical and Academic
Education .. . 238
The Case of Science Education and the Example
of Resource Mobilization ... .246
Discussion of Components of the Model .. .248
Degree of Consultation . 248
Government Responsiveness During Decision-
Making and the Policy Outcome .. 251
Conclusion . . 253


Overview .. . .. .. 255
Discussion of Teacher Training . .. 258
Benefits of Teacher Training .. 258
Different Teacher Training Programs .. 259
Discussion of Model . .. 264

Preferences . .
Hypothesis-Testing . .
Evidence of the Government Directed Policy
Community . . .


Overview . .
Conclusions . .
State-Society Relations .
Policy Processes .
Politics in Zimbabwe .
Political Democratization .


















Table 3-1.
Comparison of Societal and Governmental Greatest Needs 119

Table 3-2.
Teachers' Resources to Make Their Concerns Known to the
Government . . ... 123

Table 3-3.
Comparison of How Teachers Make Their Concerns Known to
Government in Rural and Urban Areas . .. .124

Table 3-4.
Differences Between Male and Female Teachers Regarding
Access to the Central Government . .. .125

Table 3-5.
Teachers' Access to the Central Government to Raise Their
Concerns . . ... 126

Table 3-6.
Comparison Between Students and Teachers on Community
Involvement in Education Policy . .. .134

Table 3-7.
Students' and Teachers' Perceptions of Whether the Central
Government Considers the Needs of Students, Teachers, and
Societal Groups When Deciding Education Policy 136



Figure 1-1. Indicators for Policy Community Models 36

Figure 1-2. Model of Governmental Responsiveness to
Societal Influences . ... 38

Figure 1-3. Comparison of Approaches Studying Policy-
Making . . ... 48

Figure 1-4. Policy Communities in Relation to Concepts
of Crisis/Non-Crisis and State/Society Approaches 49

Figure 1-5. Schools Included in Survey ... 60

Figure 5-1. A Summary of the School Fees Case and Its
Application to the Model . .. 224


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Mary E. Spear

August 1993

Chairman: Goran Hyden
Major Department: Political Science

This dissertation has examined the extent to which

education policy in Zimbabwe is being shaped by societal

interests and/or government initiatives. The policy

communities approach, which offers four policy communities

types, is used to assess the interaction between society,

government, and state actors, and it transcends the

limitations of the state-society relations approach.

Academics studying this topic in the developing world often

assume that the government is the principal actor in

education policy; however, this study reveals that societal

groups play an important role in policy-making and


A major finding in this dissertation was the wide range

of government responsiveness to education policies, from

very responsive to not responsive at all. To judge


responsiveness, 1,000 surveys from headmasters, teachers,

and students at 31 representative Zimbabwean secondary

schools were gathered, along with an interview with the

Minister of Education, and secondary source materials. This

information revealed that: 1) policy-making is not as open

and democratic as societal actors want, 2) societal actors

have some influence in policy-making through parent-teacher,

headmaster, and teacher groups, 3) the greatest needs of

government and society differ and conflict, and 4) the

societal groups perceive that the government's

responsiveness to them differs according to gender and

whether they work in urban or rural areas.

Another major finding of this dissertation was that the

following factors lead to democratic public policy-making:

1) the ability of strong societal groups to mobilize more

powerful resources than the government, 2) the issue and

political good provide a decisive advantage to societal

groups over the government, 3) the ability of societal

actors during a crisis situation to mobilize state actors

(such as the Parliament, District Councils, and/or the

courts) on their behalf, 4) the societal groups' ability to

mobilize the bureaucracy on its behalf during a non-crisis

situation, 5) the openness of policy communities, and 6) a

high degree of consultation in policy communities. These

factors are also responsible for linking societal

preferences with policy-making outcomes.



Education is important to parents, children, and

society at large because it signals a path to opportunity.

In Africa, parents invest in their children's education by

building schools, maintaining them, and paying school fees.

Consequently, parents expect the government to be responsive

to the needs of their children when it makes education

policy. Likewise, teachers and headmasters invest in their

schools and they want government to be open, democratic, and

responsive to their needs.

Governments in most African countries devote the

largest percentage of their national budgets to education

because they view education as a relatively inexpensive

means to socialize youths and families into the nation-state

(Fuller, 1991, p. xvii). For example, in Zimbabwe's 1991/92

budget, 16.3% of Zimbabwe's total expenditure was for

education, which was its largest expenditure, followed by

defense, which required 10.3% of the total expenditure

("Introductory Survey," 1992). In Botswana's 1991/92

budget, 21.7% of the expenditure went to education, while 8%

went to defense ("Introductory Survey," 1992). In Ghana,

25.5% of the government's total spending was for education

in 1990, while Rwanda spent 25.4% of its total expenditure

on education in 1989 ("Introductory Survey," 1992).

The government uses education to signal the provision

of opportunity and equity which it hopes will enhance its

legitimacy and authority (Fuller, 1991, p. xiii). Although

governmental and societal actors share some common education

objectives--the provision of opportunity--they are likely to

disagree about education policy. This research will examine

the extent to which societal needs and preferences are

incorporated in the education policy-making community, using

Zimbabwe as a case study.

Academics studying education and politics in the

developing world, such as Martin Carnoy and Joel Samoff,

often assume that the government is the principal actor in

education policy (Carnoy and Samoff, 1990), but they, like

others, do not provide adequate empirical evidence to verify

this proposition. Societal influence has been neglected.

Therefore, this dissertation will demonstrate the relevance

of examining the role of societal influence in education

policy in Zimbabwe.

This study will use insights from comparative politics,

policy process, and public administration to explain the

role of government, state, and societal actors

(participants) in education policy-making in the developing

world. This dissertation straddles the fields of

comparative politics and comparative policy and is derived

from three major literature bases in political science and

they are (1) state-society theory, (2) policy community

theory, and (3) open and closed system theory.

Specifically, this dissertation is concerned with education

policy but it also addresses many of the concerns of

comparativists, notably as they study governance and


In the current age of political liberalization and

democratization in Africa, much is being written about

societal protests, government responses, and reversals

(Huntington, 1991, Bienen & Herbst, 1991, Hyden & Bratton,

1992, and Bratton, 1989). Yet, this debate has been framed

almost entirely in either optimist or pessimist predictions.

Optimists examine the positive signs of liberalization and

pessimists identify signs of democratic reversals to

authoritarianism. Yet, at the heart of democracy is the

issue of whether society believes that its preferences are

represented in government policy and almost no research

currently exists which considers societal preferences and

whether those preferences are reflected in policy in the

developing world.

Political scientists have a long-term interest in the

democratic processes of public policy. Citizen

participation in public policy-making has been accepted as


an indicator of democracy. Political scientists Benjamin

Page and Robert Shapiro suggest, "The responsiveness of

government policies to citizens' preferences is a central

concern of various normative and empirical theories of

democracy" (Page and Shapiro, 1983, p. 175). One major

criticism of public opinion literature is that it is unable

to link opinion and preferences with policy outcome. Due to

this limitation, political scientists have been frustrated

when they try to assess how democratic the policy-making

process is in any particular country. This dissertation

will attempt to correct this limitation by offering a policy

communities approach in which government, state, and

societal preferences are linked with the policy decision in

order to determine if the policy is democratically made.

One inspiration for this dissertation was my

fascination with the following predictive question: under

what circumstances are societal preferences reflected in

policy? This has and will remain a key question of

democratic theory and public policy formation throughout my

study. The ability of societal actors to have a voice in

policy-making is critical to our notion of democracy.

Some Africanist scholars are presently concerned with

finding out what factors lead to democratization in Africa

(Hyden & Bratton, 1992). I will examine what factors are

present in constructive government-societal relations and

what factors are present in destructive relations. The


study of governance has begun to look at what factors are

necessary for constructive government-societal relations yet

more empirical work is needed. Governance, according to

Hyden, is concerned with how rules (or structures) affect

political action and the prospect of solving given societal

problems (Hyden, 1992, p. 14). Yet, there is a real need

for research to identify which factors are most likely to

produce constructive relations between state and society

(Bratton & Rothchild, 1992, p. 283). This dissertation will

identify the constructive and destructive factors that are

present in one sector, education policy. In chapter 8, I

will identify what factors lead to positive and negative

interactions between government and society in education

policy communities in Zimbabwe.

A policy community consists of a group of actors who

directly influence the choice of a public policy. A policy

that is democratically made is one in which societal actors

participate directly in the decision-making policy

community. I will test the usefulness of this policy

communities approach by examining four case studies of

education policy-making implemented in Zimbabwe in order to

discover how responsive the government is to societal

preferences regarding policy.

The first major question in my dissertation is as

follows: Does the government respond to societal influences

in making education policy? The second major question is,

what factors lead to democratic public policy-making?

Chapter Eight will address both questions in detail;

however, it is useful to introduce six factors that lead to

democratic and responsive public policy-making in the sub-

section of this chapter entitled, "Governmental

Responsiveness to Societal Influence in Education Policy-


The following issues will be discussed in this chapter:

the limitations of the state-society approach, the

conceptual building blocks of the new policy communities

approach, policy community approaches, the policy

communities approach, explanation of the model, policy

communities and their relationships to model and theories,

comparison of the new policy communities approach with prior

perspectives, governmental responsiveness to societal

influence in educational policy-making, three hypotheses,

the linkage between societal preferences and policy,

methodology, the significance of the study, and the

organization of the dissertation. Next, it is important to

discuss why Zimbabwean education policy was selected to test

the usefulness of the policy communities approach.

Importance of Studying Education Policy in Zimbabwe

Zimbabwean education policy was chosen for intensive

study for the following reasons. First, I was a student at

the University of Zimbabwe in October 1989 and I witnessed

the first major outbreaks of societal protests against the

government when students protested against government

corruption. While at the University of Zimbabwe, I began to

study education policy issues such as teacher training and

curriculum policy. Second, I returned to Zimbabwe in 1991

to work primarily for the Ford Foundation in its Education

and Culture Section, and I realized that many interesting

events were occurring in Zimbabwe that affected political

democratization and that they could--and should--be studied

more intensely. Third, I should mention that Jeffrey

Herbst's book from 1991 entitled State Politics in Zimbabwe,

which was one of the first major works on the policy-making

process in Zimbabwe was a main source of inspiration.

Building on his work, I decided that a sector-specific study

would be the best way of further enhancing policy-making in

Zimbabwe. For these reasons, I chose to study Zimbabwean

education policy.

The author has collected evidence from Zimbabwean

actors which indicates that societal actors do indeed play

important roles in shaping education policy. Yet, questions

remain about how influential societal groups actually are in

determining education policy-making, and under what

conditions they are. I will examine whether societal groups

are included in education policy communities. Further, I

will discover what factors are responsible for either

societal inclusion in or societal exclusion from policy


Based on a 1989 preliminary study conducted in

Zimbabwe, I developed the following main hypothesis for

empirical testing during field work in 1991: Education

policy is a function of competition between government and

societal interests over (1) divergent preferences, (2)

resources, and (3) control of the actual policy process;

however, the crucial variables in determining who controls

the actual policy-making process are found in three

hypotheses which are discussed later. The outcome of these

competitive interactions tends to be partly determined by

the extent to which the policy situation can be

characterized as "crisis" or "non-crisis." Beyond this, the

strength of the societal groups) vis-a-vis the government

is the crucial variable. If societal groups mobilize more

powerful resources than the government, they will have a

stronger influence in making policy. It may prove to be

accurate, though, that government and societal groups

collaborate instead of compete in developing education

policy. A large and original database was created to test

this hypothesis. This will be discussed later in the

methodology sub-section of this chapter.

The overall objective of this dissertation, then, is to

trace how education policies are shaped by relying on the

perceptions held by key actors, both governmental and


societal, and on secondary policy data. The following cases

will be used: teacher training, curriculum issues, school

fees, and university policy. The policy communities

approach allows one to conceptually isolate the various

actors involved in policy-making on a certain issue in order

to determine whether societal, state, and governmental

actors were involved in making the policy.

My study hopes to correct some of the limitations and

restrictions of the State-Society approach, which has been

used to study public policy. In order to stress the

weaknesses of the State-Society approach, it is instructive

to examine below the work of two of its leading proponents.

The Limitations of the State-Society Approach

Comparative scholars, such as Skocpol and Nordlinger,

have advocated a state-centered State-Society approach in

order to explain how public policies are made; however,

their approaches have limitations. According to Skocpol,

state autonomy exists when states "formulate and pursue

goals that are not simply reflective of the demands or

interests of social groups, classes, or society. This is

what is usually meant by state autonomy" (Skocpol, 1982, p.

4). Skocpol suggests that policies are either reflective of

society or that policies are reflective of the government's

view. However, in reality, the world is neither so black

nor so white. She does not account for the possibility that

the interaction among policy actors is more complex and

nuanced. Skocpol was correct in her interest in how

authoritarian a government is in making policy. However,

her notion of state autonomy, has major limitations. The

problem is the state, as defined by Skocpol, is too broad to

be useful. Further, Skocpol's definition of state autonomy

is so wide that every policy appears to have been the result

of an autonomous state decision--even if it was not. The

result is that Skocpol's state concept misleads people into

thinking that the state is almost always autonomous because,

according to her definition, any policy that "is not simply

reflective of the demands or interests of social groups,

classes, or society" (Skocpol, 1982, p. 4) is an example of

state autonomy.

One of the inspirations for my work is Eric

Nordlinger's On the Autonomy of the Democratic State.

Nordlinger wrote this book because he was frustrated with

the virtual unanimity with which "societal constraint"

models had been accepted by scholars studying democratic

politics. Societal constraint models, he suggests, are ones

in which "most or nearly all public policies are understood

as responses to the politically weightiest societal

expectations, demands, and pressures; the state is almost

invariably unwilling or unable to act upon its preferences

when these diverge from society's" (Nordlinger, 1981, p.

43). Nordlinger became disappointed with Society-Centered

models. He argued that in order to understand how public

policy is being made a State-Centered approach is more


Reading both Nordlinger and Skocpol's books I became

concerned with the limitations of having to work with either

the Society-Centered or State-Centered model, and the fact

that there was so little data to support either model.

Nordlinger's book in particular lacked empirical data. I

also realized that Nordlinger did not thoroughly account for

decision-making that involved both state and societal

actors; instead, he like Skocpol dichotomized state and

society centered approaches. Further, I became concerned

with Third World literature which suggested that public

policy was made almost exclusively from a State-Centered

approach. For example, Clive Thomas, Director of the

Institute of Development Studies at the University of

Guyana, even went so far as to suggest that internal

democratic practices are absent in newly independent African

states (Thomas, 1984). However, after independence in

Zimbabwe, the government was populist and responsive to

societal demands for a period of time as discussed in

Chapter Two.

I began to question the assumption in the literature

that public policies in the Third World are the result of

authoritarian practices, and that democratic practices are

absent. I knew from my experiences in Africa in 1989, 1990,

and 1991 that societal actors were not the helpless or

uninterested victims of state authority that the literature

suggested. Jeffrey Herbst in his book, State Politics in

Zimbabwe, also recognized that societal actors could be

influential in policy-making. Both he and I found the

prevailing conceptualizations of state-society relations


My work complements the work by Herbst, who examined

the concept of state autonomy in Zimbabwe (Herbst, 1990) in

relation to seven issues--two issues of land, agricultural

producer prices, foreign investment, minerals marketing,

health, and national minimum wage--though not education.

Herbst and I agree that an important weakness in the state

autonomy literature is that, "Most studies have not been

able to go beyond the affirmation of the potential of

autonomy to the far more important problem of predicting

when and under what conditions state leaders will actually

be free from outside pressure" (Herbst, 1990, p. 3). Herbst

notes that, "Twenty-five years after most African countries

received their independence, the extent to which government

decisions are made according to the preferences of leaders

and the extent to which the state has lost its autonomy to

societal groups in the political conflict over resources is

still unclear" (Herbst, 1990, p. 1).

Herbst's work offers the following two contributions.

First, as he suggests, it has become clear that global

judgments as to whether the state is or is not autonomous

are simplistic and misleading (Herbst, 1990, p. 251).

Instead, Herbst finds that a country such as Zimbabwe can

have authoritarian decision-making on one issue and still

have strong interest group influence in decision-making on

another issue. Second, Herbst acknowledges that to

understand policy-making in Zimbabwe one must pay attention

to the three i's: strength of interest groups, strength of

institutions, and the nature of the issue (Herbst, 1990, p.


My study differs in important ways from Herbst's work.

First, Herbst selects a broad range of issue areas, but he

does not have either depth or breadth concerning any one

issue. My study will closely examine one sector--education.

In order to provide a more concrete examination than Herbst,

I explore a number of sub-issues within education; sub-

issues relating to secondary and tertiary education.

Second, my work differs from Herbst in that I introduce the

concepts of "crisis" and "non-crisis" settings as important

variables affecting policy-making. Third, I look beyond

Herbst's notion that the state or society is autonomous, and

explore the possibility that the state, government, and

societal actors are all involved in the decision-making


Finally, I agree with Africanist scholar, James Hentz,

when he criticizes Herbst for not defining the state well.


Hentz suggests that Herbst's definition of the state leaves

room for ambiguities, and that Herbst, in his book,

represents the state as, "the civil servants, the

bureaucracy, the central bureaucracy, the middle and upper

levels of the bureaucracy, the technocratic class, the

judicial system, the cabinet, and the national leadership"

(Hentz, 1991, p. 713). If the state is one of the major

analytic variables, it must be defined more precisely.

Further, it needs to be distinguished from both the

government and the society. In the conceptual section of

this chapter, I will argue that the state is not unified in

Zimbabwe; therefore, it should not be repeatedly used and

ill-defined in the literature concerning state autonomy

versus societal autonomy. Instead, I will argue that the

government is analytically a much clearer and

distinguishable concept to use when looking at government-

societal relations.

Jack Hayward, a proponent of the policy community

approach, has offered persuasive criticisms of the state-

society approach as has Rene Lemarchand. I have found both

of their critiques of this approach very instructive. It is

important to examine, as they have, the weaknesses of the

state-society approach in order to justify the construction

of a new policy communities approach.

Lemarchand agrees with Timothy Mitchell that one should

not treat the state as an actual structure separate from


society. Lemarchand suggests, "When one treats the state as

an actual structure this creates an illusion of a more or

less elaborate institutional scaffolding separate from the

society upon which it rests" (Lemarchand, 1992, p. 180).

Lemarchand and Mitchell agree that state-society boundaries

should not be treated as a given but rather as problematic

(Lemarchand, 1992, p. 180).

While I agree to some extent with Mitchell when he

states that the boundary between the state and society

appears "porous, and mobile" (Mitchell, 1991, p. 77), I do

find that the effort to distinguish government (part of the

state) from society is analytically useful and justified in

order to determine whether policy is made by government

agents or in order to determine whether "societal" agents

play an important role in policy-making. Also, there is the

possibility that government and society cooperate with other

state actors, such as the bureaucracy in making policy. The

reason I think Mitchell and others believe the boundary is

not clear between state and society is because, until now,

the definition of the state has been so ill-defined and


Both Lemarchand and Hayward suggest that the proponents

of a dichotomy between state and society neglect the

phenomena of clientelistic relationships. Hayward says, "it

ignores the existence of a complex network of sectoral

clientelistic relationships between relatively autonomous

'segments of the state,' particular sub-governments or

public agencies, on the one hand, and the societal sectional

interest groups whose activities they seek to regulate on

the other" (Hayward, 1991, p. 382). Both acknowledge that

what can be observed are clientelistic links between public

and private actors in the policy process, whose informal

relationships are important to the way in which formal

arrangements work (Hayward, 1991, p. 382). Lemarchand and

Hayward, however, differ on one major point. Lemarchand

views the public/private relationship as "segmented,

syncretistic, and fluid" (Lemarchand, 1992, p. 190);

whereas, Hayward views the relationship between public and

private as stable. This difference may be due to the fact

that Lemarchand's studies have centered on African societies

which have more loose and segmented relationships with the

government than the more highly structured and stable

(similar to corporatist) relationships which Hayward finds

in France and Europe.

Hayward is a proponent of the policy community

approach. Before discussing the new policy communities

approach established in this dissertation and how it

compares with that of Hayward, it is necessary to start

piecing together the conceptual building blocks that are

essential to fully understand what will ultimately become

the new policy communities approach.

The Conceptual Building Blocks of the
New Policy Communities Approach

The following concepts and definitions provide the

operational foundations of this research task: governmental

actors, nongovernmental state actors, bureaucrats, and

societal actors, influence, preferences, mobilization of

resources, decision-making channels of communication,

responsiveness, open and closed, and crisis and non-crisis

situations. Policy communities will be discussed in the

next sub-section of this chapter.

First, a concise distinction between governmental and

non-governmental state actors is required. The major

"governmental actors" in this dissertation will be the

President of Zimbabwe, Mr. Mugabe, the Ministers of

Education (political appointments by President) and other

political appointees in the government. The government is

made up of the political leadership that was either elected,

appointed, or which came into power through forced means

(such as a coup). The government is most often made up of a

small group of party leadership and its major political

appointees (such as cabinet members in Ministries).

Therefore the President, Prime Minister and his/her chief

political appointees to the bureaucracy are government

actors. As is clear, I view the government to be only one

part of the state.

Governmental actors and the non-governmental state

actors combined make up the state. The following actors are

non-governmental state actors: bureaucrats, Supreme Court

officials, Parliamentarians, and local government officials.

The bureaucracy (with the exception of political appointees

such as the Minister), judiciary branch, Parliament, and

local government officials are not part of the government

but are non-governmental state actors.

The issue of exactly where the bureaucracy fits in this

dissertation deserves more elaboration. In this case,

according to Charles Jacob, bureaucrats, which are non-

governmental executive-branch state actors, are vocationally

oriented in that they have reached office by route of a

particular course of training in a specialized field (Jacob,

1966, p. 35). Yet, ministers, usually thought of as a part

of the government, have reached office due to their

political connections to the ruling group of the country.

Therefore, when bureaucrats are mentioned in this

dissertation, I am referring to all Ministry of Education

staff except the political appointees. Non-governmental

state actors, such as bureaucrats, may have their own

preferences, or they may agree with preferences of the

government, or they may agree with preferences of society.

We will see in Chapter Six on curriculum policy that most of

the Ministry of Education is made up of bureaucrats who

disagreed with the government Minister's attempt to

institute a political economy curriculum.

The idea of separating the concept of the state into

two major categories, governmental and non-governmental

state actors, is new and innovative, and it can be

attributed to the work of Roger Dale. His work, which is

derived from modern educational thought, successfully

distinguishes the boundary between the government and the

state. Dale's distinctions between government and state are

an integral part of this work; therefore, it is worthwhile

to expand on his distinctions. He considers the government

to represent, "the short-term interests of the temporarily

dominant coalition of forces within a social formation;

these coalitions are represented in political parties"

(Dale, 1989, p. 53). According to Dale's belief, the

Zimbabwean government in this case is a dominant coalition

of the following actors: President Mugabe, the ruling

national party (ZANU) leadership, and the political

appointments (such as Ministers). I concur with Dale's

assumption that the government is a dominant coalition of

forces or actors.

Dale believes that the state is made up of two parts--

government, the most important part, and "state

apparatuses." The Government, is one distinguishable part.

The second, "state apparatuses" (Dale, 1989, p. 54) are, as

Dale states, a set of publicly financed institutions; many

of which are accountable officially to the government.

Other parts of the state, such as the judiciary and the

legal system, however, are not accountable to the government

in the same way (Dale, 1989, p. 54). Dale finds that there

are often tensions between state institutions, for example,

the Ministry of Education, and the Supreme Court, and we

will discover this to be the case in Zimbabwe. Dale further

notes that there is also tension between local education

authorities and central government authorities (Dale, 1989,

p. 53-55).

Although governments may fall due to elections or

revolutions, Dale suggests that other state apparatuses,

such as publicly financed institutions, are able to continue

to operate perfectly well (Dale, 1989, p. 53). For

instance, the Ministry of Education, a state institution,

continued throughout the transition to Zimbabwean

independence from the white-minority government in Zimbabwe,

to the one-year transitional government, to the new black-

majority controlled government in 1980, and the Ministry

continues today.

Because the state is not unified and because there is

often internal conflict (except in the pure totalitarian

system), it is difficult to speak of a state as one unified

body. Instead, most often, what is important is to examine

societal interests versus the interests of the government

(ruling governing coalition). Accordingly, this

dissertation is most concerned with examining the

interaction between government actors and societal actors.

But, we must be aware that state institutions can and are

mobilized in order to aid either the government or the

society. What exactly is meant by societal groups? We now

turn to this issue.

"Societal groups" refer to headmasters, teachers,

students, parents, religious groups, citizens, and other

nonstate actors. Critics may argue that headmasters and

teachers should be considered government or state actors.

However, there are compelling theoretical reasons to

consider these people as part of the "societal groups."

First, headmasters and teachers in Zimbabwe organize their

own pressure groups which represent their preferences before

government. These preferences often diverge from those of

the Ministry of Education. Second, headmasters and teachers

are an integral part of their local communities. Thus,

teachers and headmasters will be considered part of societal

groups in this study.

Next, it is important to distinguish between two

groupings of societal associations examined in this study--

"contractual" and "communitarian." Associations based on

local residence or ethnicity, and which serve as informal

mechanisms in the sense that they are without a written

constitution legalizing their existence, are considered

communitarian. These are held together by the belief in

reciprocity. Associations which by contrast are formed

through contractual agreements, such as a written

organizational charter, are considered contractual. The

National Association of Secondary Headmasters (NASH), for

example, is a contractual association.

Since societal "influence" over education policy is the

dependent variable in this study, it is crucial to

understand what is meant by influence. According to Goran

Hyden, citizen influence and oversight refer "to the means

by which individual citizens can participate in the

political process and thereby express their preferences

about public policy and how well these preferences are

aggregated for effective policy-making" (Hyden, 1992, p.

15). Influence means to affect or sway a decision. This

study will ask what types of persons are the most

influential (Dahl, 1961, p. 331) in determining the outcome

of various educational decisions. We must, however,

understand that not only do societal groups influence

education policy but that the government often intervenes in

the classroom; thus, resulting in the role of influencing

teachers, headmasters, and students. Fuller finds that in

the case of Malawi, which is similar to Zimbabwe, the state

has a difficult time influencing the daily actions of

headmasters and classroom teachers. But, the state can

formulate, transmit, and sanction the knowledge presented in

the classroom through influence over curriculum, textbooks

and the national examinations (Fuller, 1991, p. 84).

If the school-based actors, such as teachers,

headmasters, and students are not satisfied with government

intervention they can mobilize their resources and attempt

to influence policy. Access to participation in the policy

process is necessary, but not sufficient. Actors also need

control over vital resources, plus the ability to act

effectively on their preferences and opinions. For persons

to act on their preferences, they must have open channels of

communication with all other actors and among themselves.

In order for a person or group to influence policy, the

actor must either be the policy decision-maker or have

access to the policy-makers. It is important to determine

if structures or procedures (channels of communication)

exist in order to allow societal actors access to decision-

makers. If they do exist, then societal actors can have

influence in the policy process. Herbst is also aware of

the importance of the concept of access. He found that a

communal farmers' organization had no access to officials in

regard to making a policy decision; and therefore, the state

acted autonomously (Herbst, 1990, p. 77-79). Herbst,

however, uses the concept "match-up" to mean access. He

says, "by match-up I mean an interest group directly

confronting the particular part of the state that is making


the allocation decision in such a way as to cause political

conflict" (Herbst, 1990, p. 253).

There are at least four different degrees of societal

influence which affect the government's role in decision-

making. First, when governmental and societal preferences

differ and the government prevents societal actors from

access to/or control over the policy process then it is

acting in an authoritarian manner. Second, in other

situations, the government may allow societal groups access

to the policy process but it could still, in actuality,

remain more influential. Third, the government may allow

societal groups access and some control over the policy

process, in which case the government is losing some of its

influence. Finally, societal groups may pressure the

government with powerful resources and the government has no

choice but to accept societal group pressure.

It is important to discuss "preferences" and

"mobilization of resources" in order to understand the major

components of this dissertation's model. "Preferences"

refer to the interests, attitudes, and motivations of

individuals or groups. Accordingly, the students'

preferences include their need for textbooks, qualified

teachers, papers, pens, and scientific equipment in order to

improve their education. Naturally, when their preferences

are unmet, they question why the government is not providing

them with these needs. Likewise, government actors have


preferences, such as a high quality practical education, as

will be discussed in Chapter Three.

Robert Dahl defines a political "resource" as "anything

that can be used to sway the specific choices or the

strategies of another individual" (Dahl, 1961). Two notable

political "resources" are the vote and persuasion. For

instance, an important resource for parents wanting to

influence education policy is the use of the voting threat.

According to Nordlinger, "votes constitute a highly

effective set of resources and sanctions" (Nordlinger, 1981,

p. 222). According to Fuller, in his study of the

relationship between the school and the state in southern

Africa, the most common means or "resources" for the state

to attempt influence over the schools include budgetary,

administrative, and symbolic actions (Fuller, 1991, p. 7).

Thus, the government must mobilize administrative resources

in order to train teachers, produce textbooks, and create

and enforce selection and allocation mechanisms, like

national exams.

In this study the most important resources for the

societal actors appear to be the vote; the ability to

protest and organize; and their access to the media;

whereas, the government's strongest resource is its control

over money and coercive weapons. Moreover, the implied

threat or actual use of force is a strong resource.

Additionally, the government has a legal resource in its

favor--the legal provision that gives it the power to

implement education policy. It is important to remember

that there is inevitably a tension between "preferences" and

"resources." Students at the University of Zimbabwe may

have strong "preferences" for academic freedom but few


Finally, governmental responsiveness is a key concept

linking societal influences to educational policy-making.

Government responsiveness to societal needs is significant.

According to Hyden, responsive and responsible leadership

refers to "the attitudes of political leaders toward their

role as public trustees" (Hyden, 1992, p. 15). President

Mugabe, a former teacher, said that, "Our people want

genuine democracy; They want a leadership responsive

and committed to their needs" ("People in, 1991). Thus it

follows that if a government is responsive to societal

needs, it is therefore responding to societal influences in


This dissertation will offer empirical evidence

concerning how responsive the Government of Zimbabwe is on

educational policy issues. Four levels of government

responsiveness to societal influence are conceptualized

below to examine the different degrees to which government

can respond to societal influence during the policy process.

First, the government can be unresponsive and not listen to

the needs of societal groups. Second, the government can

solicit and listen to views from societal groups. Third,

government can accommodate societal interests with its own

pre-defined, pre-determined interest in a policy. Fourth,

societal interests can supersede governmental interests. In

the last case, societal groups naturally have maximum


One factor that may affect policy-making styles is

whether a decision is made under a crisis or non-crisis

scenario, as Merilee Grindle and John Thomas have suggested.

They say that in crisis situations there is strong pressure

for reform, high stakes, high-level-decision-makers

involvement, chance for innovative change, and pressure to

act immediately (Grindle and Thomas, 1991, p. 161). The

case of government control of the University of Zimbabwe,

which will be discussed in Chapter Four, is an example of a

policy decided under a crisis situation. In crisis

situations, the government makes sure that the bureaucracy

is fully behind its policy. However, societal groups can,

and often do, mobilize state actors, such as

Parliamentarians and members of the judiciary, to support

societal claims. So, in crisis situations the Government

and the Ministry of Education (or Ministry of Higher

Education) officials are often pitted against a coalition of

societal actors, judiciary actors, and Parliamentarians.

A "non-crisis" situation has the following

characteristics: low political and economic stakes; middle

and lower level officials involved; incremental changes in

existing policy or institution; and little sense of urgency

(Grindle & Thomas, 1991, p. 107). In these situations

bureaucrats and interest groups are mostly involved in

policy-making. In the hypotheses section, toward the end of

this chapter, the "crisis" and "non-crisis" variables are

incorporated into the hypotheses. Now that the conceptual

building blocks of the new policy communities approach have

been presented it is important to discuss the development of

the approach.

Policy Community Approaches

In 1974, Hugh Heclo and Aaron Wildavsky were the first

political scientists to relate the notions of policy and

community in their book titled The Private Government of

Public Money: Community and Policy in British Political

Administration. They said a "community is the cohesive and

orienting bond underlying any particular issue" (Heclo &

Wildavsky, 1974). In 1978, Heclo discussed the role of the

administrative agency, interest groups, and Congress

(Parliament) in forming an iron triangle (Heclo, 1978).

Jeremy Richardson and A. Grant Jordan dropped the

parliamentary side of the iron triangle in their discussion

of a policy community in Britain. Richardson and Jordan

suggested, as Hayward states, that policy outcomes owed most

to "the policy community of departments and groups"

(Hayward, 1991, p. 392).

Zimbabwe, a former British colony, one would

hypothesize, would follow a similar pattern to Britain in

that the government departments and interest groups are the

two most important actors in the policy community, not the

Parliament. When one also considers the fact that the

Parliament in Zimbabwe is controlled heavily by the

President's ruling party and his executive branch political

appointees, more evidence is offered to conclude that it is

the Government, the Ministries, and interest groups that

will play the most important roles in policy communities,

not the Parliament. Still, on some issues such as school

fees and university policy, Parliament may still count.

Hayward advocates the policy community approach.

Hayward uses the notion of the policy community "to identify

a network of clustered 'insiders' engaged in an interactive

process to attain their policy aims. While they do not

share common objectives in terms of means or ends, they are

sufficiently committed to their established, closed policy

processes; thus, they accept some sacrifice of their short-

term particular interest" (Hayward, 1991, p. 395). He also

suggests that within these policy communities there is

general consensus among actors (although he does allow for

some relapses into conflict or domination by one of the

actors). He also seems to suggest that there is some

stability in these policy communities. Hayward, in his

policy community approach, suggests that policies can be

reflective of both society's interests and the state's

interests. Hayward's policy community approach suggests

that policies are mostly the result of interaction between

state and societal actors. This idea of policy communities

better represents the reality of policy-making than the

state-society approach.

While Hayward should be praised for discussing a

concept that allows for society and government cooperation

on an issue, his approach has some limitations. Hayward

says that the policy community approach has been tested in

cases mostly related to economic and industrial policy.

Yet, economic policy communities may have more cooperative

and stable relationships between government and societal

actors; a policy community on education, by contrast, may be

characterized by conflict. This may be because education

serves somewhat different purposes for the government and

for societal actors. While the government hopes to achieve

national development and unity through education often

students and parents hope to achieve individual benefits

from education. These differences may lead to inevitable

conflicts between government and society.

The way Hayward defines the policy community approach

limits its applicability to other countries. His portrayal

of a policy community as being stable and having a

consensus, while applicable to policies in France and

Austria perhaps, is too narrow a definition to apply to

policies in all countries. Also, Hayward contends that the

policy community approach may be less appropriate for Third

World political systems, but he agrees this should at least

be tested.

Herein lies a dilemma for comparativists. Many

comparativists seek a theoretical framework which is

universally applicable to countries in "developed" and

"underdeveloped" settings; however, present theoretical

formulations by Africanists and Europeanists may be narrow

and limited in their applicability to certain other

countries. I believe that a different conceptualization of

the policy community approach can, and should, be used to

assess whether the government, the society, and/or state

actors create education policy in Zimbabwe. Before looking

at the new conceptualization of the policy communities

approach, let us first look at Giandomenico Majone's notion

of a policy community.

One shortcoming in Hayward's approach is his suggestion

that a policy community is stable and closed. Majone

questions this and instead suggests that a policy community

is open and competitive. This expands the usefulness of the

policy community approach to different settings. He

indicates that a policy community is composed of specialists

who share an active interest in a certain policy or set of

related policies: academics, professionals, analysts,

policy planners, media, and interest group experts. He

suggests that the "policy community must be sufficiently

open and competitive" so that novel ideas may emerge

(Majone, 1989, p. 163). While Majone allowed for some

members of a policy community to also be political actors,

he viewed the two roles as separate and distinct (Majone,

1989, p. 161). He recommends that policy community members

must have a disciplinary and organizational base through

which access to the relevant political arena can be secured.

In my conceptualization of the policy community, it

represents the group of actors that actually make the policy

decisions. Majone's policy community includes too many

actors to be analytically useful. This is why I have

limited my definition of a policy community to only those

actors that are directly involved in making a policy

decision. My version of policy community does not include

all specialists with an active interest in policy.

The Policy Communities Approach

The policy communities approach involves a six step

process to measure how well the government responds to

societal influences during decision-making. First, it is

important to identify the preferences and resources of the

different actors. Second, it is important to consider

societal efforts to mobilize their resources and the

government's response. One resource that may be mobilized

by the government or by society is the assistance of the

bureaucracy or other state actors.

The third step is to find out to what degree the

decision-making policy community is open to outside

influence. Fourth, it is important to find out the degree

of consultation that occurs between government and society.

The fifth step in the policy communities approach is to

apply the case study and its degrees of consultation and

openness to how government responds during decision-making.

The sixth and final step in using the policy communities

approach is the examination of the policy outcome of each of

these four different types of policy communities.

The policy communities approach advances analysis

beyond the simple model of political demand and government

response. It looks at the interaction between key state,

governmental, and societal actors during the decision-making

process. The purpose of this study is to demonstrate that

the ways in which policies are made in developing countries

is much more varied than existing approaches suggest. In

adopting the concept of policy community, I am interested in

seeing how they are formed, what takes place within them,

and what comes out of them. The assumption here is that

policy communities, even within the same country, vary. As

this study attempts to show, such variation exists even

within a single sector. The issue that needs to be

discussed here concerns how one differentiates between them.

In order to develop the policy communities approach, I

have incorporated concepts from the public administration

field into the approach. In public administration an open

system represents a system that is permeable, changing, and

fluid in which members can enter and exit with ease. One

indicator of openness is whether societal groups can enter

the policy community, with ease, and influence a decision

for a period of time, and then exit the policy community,

with ease, when they feel their influence is no longer

necessary. A closed system represents a system which is

relatively stable, homogeneous, and non-permeable.

Therefore, one indicator of a closed system would be that

its membership is relatively constant and stable.

The two major indicators for the different policy

community models that I have created are (1) the degree of

consultation, and (2) the degree of openness. These two

indicators are examined in the third and fourth step of the

policy communities approach. The third step is to find out

to what degree the decision-making policy community is open

to outside influence. This can be done by assessing whether

the policy community is highly exclusive, medium exclusive,

little exclusive, or inclusive. Fourth, it is important to

find out the degree of consultation that occurs between

government and society. This can be done by assessing


whether the policy community involves high, medium, low, or

no consultation between the government and society.

The degree of openness measures the extent to which

societal actors participate in, or are shut out of, the

policy community. For the purpose of this study I am

differentiating between four types that vary in degree of

openness: (1) highly exclusive, (2) medium exclusive, (3)

little exclusive, and (4) inclusive.

The degree of consultation, the second indicator,

measures the amount of discussion, communication, and

deliberation that occurs before a decision is made. There

are four different degrees of consultation which correspond

to the four policy community types: a high level of

consultation found in the open conflictual policy community;

a medium level of consultation which is found in the closed

cooperative policy community; a little consultation is found

in the government directed policy community, and finally

there is a level of no consultation (none) in the closed

government policy community. In Chapters Four to Seven,

each one of these indicators will be discussed in relation

to their appropriate policy community. Figure 1-1

illustrates the relationship between the degrees of

consultation and openness to the four policy community


(1) Degree of Consultation Below
No Little Medium High
Consultation Consultation Consultation Consultation
Closed Government Closed Open
Government Directed Cooperative Conflictual
Policy Policy Policy Policy
Community Community Community Community
Highly Medium Little Inclusive
Exclusive Exclusive Exclusive
(2) Degree of Openness Above

Figure 1-1. Indicators for Policy Community Models

The fifth step in the policy communities approach is to

apply the case study and its degrees of consultation and

openness to how government responds during decision-making

which is found in Figure 1-2. The Closed Government Policy

Community, which is highly exclusive and involves no

consultation between government and society, corresponds to

number one on the model below titled, "Government Prevents

Societal Access." The Government Directed Policy Community,

which involves only a little consultation and yet it

maintains medium exclusivity, corresponds to number two on

the model below titled, "Government Allows Societal Access."

The Closed Cooperative Policy Community, which has a medium

amount of consultation and is only a little exclusive of

societal actors, corresponds to number three on the model

below titled, "Governmental Accommodation of Societal

Preferences." The Open Conflictual Policy Community,

involves a high degree of consultation and is inclusive, and

it relates to number four on the model on the next page

which is called, "Government Accepts Societal Preferences."

The model shown in Figure 1-2, connects the concepts

discussed earlier and incorporates a model one can estimate

from the gathered data. The sixth and final step in using

the policy communities approach is the examination of the

policy outcome of each of these four different types of

policy communities. The outcome is evident in Figure 1-2

under the category titled "Policy." Below is an explanation

of the model.

Explanation of the Model

In policy-making there are two broad categories of

"preferences"--governmental and societal preferences. As

Nordlinger suggests, when state and societal preferences are

the same, then the state's preferences become policies

(Nordlinger,1981, p.28). While this study measures

government (not state preferences), the same logic applies.

When government and societal preferences are the same, then

government preferences become policies. On the contrary, if

societal preferences and governmental preferences coincide

there is no problem; thus, there is no need to mobilize


o v-I


0 uI

El IO C .,

E ~*
0 c 0 r. 0 0 0 ~

40 (UU 0- I

0 41
0 0

0 1 0

0) 0

Sc Z 1 0 E
u u~ Y IQ
o ~ ~" u MP

> >0


If government mobilizes resources and it is

unchallenged by societal actors in the policy process, then

governmental preferences are reflected in the policy. But,

when societal preferences diverge from governmental

preferences then societal groups mobilize their resources as

do governmental actors. If government is responsive to

listening to societal concerns then during the policy

process the government allows societal groups access to

decision-makers. However, the government may prevent

societal actors from access to decision-makers during the

policy process. In both cases, governmental preferences

reflect policy. Yet, if the government decides that it must

accommodate societal preferences in the policy process then

a societal group can wield some influence. As a result,

governmental and societal preferences are always reflected

in policy. If societal influence is great in the policy

process, government may accept societal preferences over

governmental preferences in a policy.

Systems theory in political science has been attacked

for focusing on the inputs and outputs of the political

system and for neglecting what occurs in the middle--in the

black box. The model used here allows us to say something

meaningful about what is happening in this "black box."

What occurs in the policy communities affects the policy


Policy Communities and Their Relationships to Model and

Now that the indicators for the policy communities are

clear and the model has been explained it is important to

discuss each policy community. I label Hayward's concept of

policy community the Closed-Cooperative Policy Community

approach (which on Figure 1-2 would represent the category

numbered three which is labeled "Governmental Accommodation

of Societal Preferences"). This category represents a

little exclusivity in terms of the degree of openness. The

reason that this category is a little exclusive is because,

although, the decision-making community has at least some

societal actors involved, as Hayward calls them, a "network

of clustered 'insiders' engaged in an interactive process to

attain their policy aims," other societal actors, most

notably the outsiders, are excluded. A medium degree of

consultation is found in the Closed Cooperative Policy

Community. The result is a policy in which governmental and

societal preferences are both reflected in a policy. The

example I offer of this approach is Chapter Six of my

dissertation on the development of curriculum in Zimbabwe.

The Closed-Cooperative Policy Community resembles what

Philippe Schmitter calls societal corporatism. Societal

corporatism is, "a pattern of institutional relationships in

which the officially sanctioned sectoral interest

organizations, while collaborating with each other and state

policy-making elites in the pursuit of a commonly accepted

national interest, speak quite autonomously for their

socioeconomic sectors and actively engage the state in the

defense of their constituents" (Keller, 1991, p. 148).

Corporatism, according to Schmitter, is "a system of

interest representation in which the constituent units are

organized into a limited number of singular, compulsory,

noncompetitive, hierarchically ordered and functionally

differentiated categories, recognized or licensed (if not

created) by the state and granted a deliberate

representational monopoly within their respective categories

of leaders and articulation of demands and supports"

(Keller, 1991, p. 148).

In Chapter Six on curriculum policy, I find that while

teachers and business representatives work with government

and the state in making curriculum policy as is the case in

corporatism, they also speak quite autonomously on behalf of

their constituents.

The Closed-Cooperative Policy Community model and the

Government Directed Policy Community are both corporatist

forms of policy-making. The Government-Directed Policy

Community is the next policy community I wish to discuss.

It most closely resembles what Schmitter has called state

corporatism--that which is, "an institutional arrangement in

which the state seeks to co-opt or control major sectoral

interest organizations, usually by establishing rules that

govern their very creation as well as their behavior"

(Keller, 1991, p. 148). This policy community on Figure 1-2

is the category numbered two which is entitled "Government

Allows Societal Access." This category represents a medium

degree of exclusivity because most societal actors are

excluded from the policy community; however, a few

government selected societal representatives are, as we

shall see, involved in decision-making. There is a only a

little consultation in the Government Directed Policy


An example of the Government Directed Policy Community

is teacher training policy which is discussed in Chapter

Seven. In this case, the government allows some dialogue

and access to society, but in terms of the final decision-

making structure, society is virtually excluded and

governmental preferences result in the reflection of policy.

One indicator of the Government Directed Policy Community is

that the Minister of Education controls and directs the

participants in the final decision-making policy community.

As we will see in Chapter Seven, teacher education is

contained and controlled by a tripartite organization

controlled by the Ministry of Education, University of

Zimbabwe, and the Teachers' Colleges. The government co-

opts the organizations involved in teacher training.

I would also allow for two other types of policy

communities, which Hayward does not consider. The third

form of policy-making I discuss is the Closed Government

Policy Community which is represented in Figure 1-2, by

number one entitled "Government Prevents Societal Access."

In this case the government prevents societal access and as

a result governmental preferences reflect policy. This

policy community is highly exclusive in that societal actors

are excluded from input into the policy community. There is

no substantial consultation between government and society

in the Closed Government Policy Community. An example of

this is the crisis over academic freedom at the University

of Zimbabwe which makes up Chapter Four of my dissertation.

This approach is characteristic of a policy decision-making

group made up of select governmental actors which excludes

societal actors from a role in decision-making.

Accordingly, the main indicator of this Closed Government

Policy Community is that societal actors are excluded from

participation in the policy community.

The fourth and last policy-making community which I

have created is called the Open-Conflictual Policy

Community. It is represented, in Figure 1-2 by number four

called, "Government Accepts Societal Preferences." An

example of this type of decision-making is the case of

school fees, which will be examined in Chapter Five. This

resulting policy reflects societal preferences. This policy

community is not at all exclusive, instead it represents a

high degree of inclusivity in that societal actors are fully


included in the policy community. There is a high level of

consultation found in this Open Conflictual Policy


This final Community is found in pluralistic situations

in which the decision-making system is open, and competition

is strong between societal and governmental groups. Fuller

suggests that the state faces demands from groups and

institutions that question Western political-economic

ideals, such as the village chiefs, the urban middle class

that demand opportunity for their families first (and for

rural groups later) and for the post-colonial economic firms

that would not benefit from opening up liberal markets

(Fuller, 1991, p. xvii). Parents in the Open-Conflictual

Policy Community have demanded that the government allow

local control of school fees.

I believe that neither the society centered, state

centered, or corporatist approaches should be adopted until

data can be collected to provide evidence for a suitable

approach that will represent reality. The policy

communities approach that I have adopted may remedy this

problem because it allows the approaches to be tested as

data is collected.

The strongest benefit of the policy communities

approach is that it does not take any of the approaches as

given, but it recognizes the interdependence between public

and private actors in formulating policy. According to

Hayward, the dualisticc view of the state versus society

distorts realities in that it ignores the existence of a

complex network" of government and societal forces

collaborating in policy decisions (Hayward, 1991, p. 382).

The policy communities approach allows an analyst to

consider that key policy actors mobilize allies both within

and outside the processes of public decision-making

(Hayward, 1991, p. 390). The following sub-section of this

chapter will reveal how the new policy communities approach

relates to the other approaches discussed.

Comparison of the New Policy Communities Approach with Prior

The four policy community types introduced are the

result of an adaptation and refinement of the state-society

and policy community literature. I have examined the policy

literature and integrated useful concepts into a framework

that better represents the process of policy-making than the

previous state-society literature provides. Figure 1-3 is

introduced to demonstrate the difference between the

approaches discussed above. Perspective number one on the

next page is the theoretical basis that was considered at

the beginning of this dissertation research in which Skocpol

and Nordlinger implied that either the state or society was

responsible for making policy decisions. As scholars grew

dissatisfied with this crude dichotomy, new perspectives

were developed. Herbst in perspective two suggested that to

understand how policy decisions are reached one must

consider whether the issue extends state's capabilities and

the strength of state institutions compared to the strength

of interest groups. Hayward, in perspective number three,

argues that policy decisions are made by a policy community

of state and societal actors that interact with each other.

Perspective number four, the new policy communities

approach, goes beyond all these. It suggests that the

interaction between government, state, and societal actors

depends on (1) whether or not the decision was made during a

crisis or non-crisis situation, (2) the extent to which the

government permitted and encouraged openness and

consultation in the decision-making process, and (3) the

nature of the issue. The idea that interest groups,

institutions, and issues matter as suggested by Herbst has

been integrated into the policy communities approach. The

notion that state and societal actors interact in the

decision-making process, as suggested by Hayward, has also

been considered. For these reasons, it can be argued that

the policy communities approach allows for more probing

comparisons than the other approaches. Figure 1-3 offers a

comparison of approaches studying policy-making.

To provide the reader with a sense of how the data

chapters relate to the model in Figure 1-2, I have indicated

in Figure 1-4 how each of the four case studies illustrates

one of the particular policy community types. Numbers one


through four represent the four different policy communities

in which education decisions are made in Zimbabwe.

Governmental Responsiveness to Societal Influence in
Educational Policy-Making

Six factors affect the emergence of policy communities

that are responsive to societal influence: the strength of

societal groups in contrast to governmental actors, the

nature of the issue to be decided, the ability of societal

and governmental actors to mobilize other state actors (such

as the bureaucracy or the courts) on their behalf, the

existence of a crisis or non-crisis situation, openness of

policy communities, and the existence of policy communities

which have a high degree of consultation.

In situations where societal actors are very

influential in making policy, the following situations will

exist: societal groups mobilize more powerful resources

than the government, the issue to be decided is not being

viewed by the government as a threat to its legitimacy,

societal actors in a crisis situation mobilize non-

bureaucratic state actors on their behalf or in a non-crisis

situation societal actors mobilize bureaucratic state actors

on their behalf. Next, it is important to examine the main

hypotheses of this study.

1 2 3 4

Skocpol &
Nordlinger Herbst Hayward Spear






Interest Groups


During Crisis
A) Closed

B) Government
Bureaucracy and

C) Closed
Bureaucracy and
Interest Groups
are Most

D) Open
Government, and
State Important

Comparison of Approaches Studying Policy-Making

Policy Community
of State/Society

Figure 1-3.


Closed Government Policy
Result-Policy that
reflects governmental


1 Example-Chapter 4
University of


Government Directed
Policy Community

Result-Policy that
reflects governmental


2 Example-Chapter 7
Teacher Training

Open Conflictual
Policy Community
Result-Policy that
reflects societal


4 Example-Chapter 5
School Fees


Closed Cooperative
Policy Community

Result-Policy that
reflects governmental
and societal preferences


3 Example-Chapter 6


Figure 1-4. Policy Communities in Relation to Concepts of
Crisis/Non-Crisis and State/Society Approaches

Three Hypotheses

In order to predict if, when, and under what conditions

government leaders will be responsive to societal influence,

I have developed three hypotheses. Education policy is a

function of competition between government and societal

interests over (1) divergent preferences, (2) resources, and

(3) control of the actual policy-making process; however,

the crucial variables in determining who controls the actual

policy-making process are found in the following three


Hypothesis 1: Policy issues that require extended

capabilities beyond the government's capabilities and in

which the nature of the political good provides a decisive

advantage to societal groups results in societal control of

the policy process. Policy issues that do not require

extended capabilities beyond the government and in which the

nature of the political good provides a decisive advantage

to government results in governmental control of the policy

process. If policy issues require the capability of

government and society and the nature of the political good

does not provide either with a decisive advantage then

policy-making is controlled by both, such as in the case of

curriculum policy in Chapter 6.

The basic elements of this hypothesis are the (1)

ability of issue to extend beyond government's capabilities,

and (2) determination of whether the political good gives

advantage to society or the government or both, were

discussed as middle-level hypotheses by Herbst (Herbst,

1990, p. 255). Herbst is correct to be concerned about the

issue and whether the political good provides a decisive

advantage to the interest groups or government (Herbst,

1990, p. 255).

While I agree with Herbst on the general ideas in his

hypotheses, there are a couple important distinctions

between his hypotheses and my hypothesis. First, I believe

the political good and the nature of the issue are

interrelated and therefore should be combined into one

hypothesis, which I did. For example, the issue of teacher

training and the political good of training of teachers are

interrelated. Second, Herbst focuses on the state I instead

discuss the role of the government.

The outcome of the competitive interactions between

government and society are partly determined by the issue

and political good but also by the extent to which the

policy situation can be characterized as "crisis" or "non-

crisis." Grindle and Thomas (1991) predict that state

leaders act differently under conditions of crisis and non-

crisis decision-making. This notion that leaders act

differently under conditions of crisis is a very interesting

proposition that deserves further testing and discussion,

so, I incorporate it into my study.

The next hypothesis is derived from ideas that come

from Grindle and Thomas (1991). They state that in "non-

crisis" situations, "micropolitical concerns include

concerns about the more parochial demands of specific

interest groups, the use of policy resources to maintain

clientelistic relationships, the parceling out of policy

resources to ensure political control and to narrowly

defined groups in exchange for political support and more

short term interests of political elites" (Grindle and

Thomas, 1991, p. 105-106). One problem with Grindle and

Thomas' proposition is that they do not distinguish between

the state and the government. They do not account for

hypotheses that allow for non-autonomous policies; rather,

they suggest that the state usually acts autonomously. I,

however, believe that there should be a hypothesis that

allows for societal dominance in policy decisions;

therefore, I created hypothesis two.

Hypothesis 2: If in a "non-crisis" situation the

societal groups' capability is greater than the government's

capability then society controls the policy process. If the

government's capability is greater than societal groups'

capability then government controls the policy process. If

societal groups' capability is not greater than the

government's and the government's capability is not greater

than societal groups' then neither alone control the policy


Herbst's definition of strong societal group pressure

is used here to refer to situations where interest groups'

(or societal groups') capability is greater than the state's

institutional capability. This hypothesis was validated in

Herbst's work (1990). Herbst discusses whether the interest

group matches up with state institutions. Herbst describes

what he means by 'match-up,'

I mean an interest group directly confronting the
particular part of the state that is making the
allocation decision in such a way as to cause political
conflict. If the organizational structure of the
interest group cannot match up against the
institutional structure of the state, the state will be
autonomous (Herbst, 1990, p. 253).

There are other differences between Herbst and me.

While Herbst focuses on whether there is a political

conflict I focus on divergent preferences. While Herbst

focuses on the strength of the organizational structure of

the interest group, I focus on the capabilities of different

actors to mobilize resources. Finally, I focus on the

government's capability and Herbst focuses on the state's

capability. I discuss the strength of societal groups vis-

a-vis the government, by examining capabilities.

Hypothesis 3: In "crisis-ridden" situations, decision-

making tends to be dominated by concern about major issues

of political stability and control. Grindle and Thomas

argue that in such situations the state is the most

influential actor. In cases where the government believes

the government's legitimacy and very survival is in

question, they mobilize all possible resources in order to

stay in power and this results in the exclusion of

opposition societal actors from the policy process. The

societal groups are weak relative to the government.

However, in cases where government legitimacy and survival

are not in question, however the government believes high

stakes are involved in terms of political control and

stability, the government will mobilize resources but it

will not exclude societal actors from the policy-making


Technical analysis, bureaucratic interactions, and

international pressures often assume importance in crisis

decisions, but usually remain subordinate to concerns about

the stability or survival of the regime in power or to the

longevity of its incumbent leadership. The distinction

between "non-crisis" and "crisis" conditions seems to be the

first criteria for Grindle and Thomas to determine how

influential the state is.

Grindle and Thomas explain that in "crisis" situations

there is strong pressure for reform, high stakes, high-

level-decision-makers involvement, chance for innovative

change, and pressure to act immediately (Grindle and Thomas,

1991, p. 161). All these characteristics appear to have

been present in the case of the University of Zimbabwe in

which the government decided university policy.

A point of clarification must be made about societal

interests in Zimbabwe. There is not a single societal

interest in Zimbabwe as will be evident in the split between

societal actors that support practical education and others

that support academic education as discussed in Chapter 6.

However, in that case societal actors that support academic

education are more powerful. Therefore, when I discuss

societal interests I am referring to the views of the

predominant or most powerful societal interests. Next, it

is important to examine the linkage between public opinion

and policy.

The Linkage Between Societal Preferences (Public Opinion)
and Policy

According to V. O. Key, public opinion is important

only if the preferences, aspirations, and prejudices of the

public can be connected with the actual workings of the

governmental system (Key, 1961, p. 535). One major

criticism of public-opinion literature, such as in Page and

Shapiro (1983, p. 175), is that public opinion is unable to

link opinion and preferences with policy outcome. This

dissertation will attempt to make this linkage. Various

interviews and secondary source materials which discuss the

actual role of different people in education policy will

help form a clearer picture of the actual workings of

government. Also, to understand societal influence over

education policy, it was necessary to gather information on


influences of other key actors in the educational community,

including church and parent groups, groups of teachers,

headmasters, and others.

Key is correct in affirming the need for connecting

preferences with (what is) the workings of the governmental

system (Key, 1961, p. 535). Preferences are assessed by

asking people what they think should happen. By doing this,

we can connect opinion to policy processes and outcomes.

Surveys, interviews, and secondary source material shed

light on what happens in the policy process, and what should

happen in order to determine if societal preferences are

reflected in policy. The questions of what should happen

and what actually does happen are both critical elements of

my dissertation. In order to detect whether societal

preferences and/or governmental preferences (the should

question) are reflected in policy, we must know both what

actors prefer and what actually happens. Therefore, if

students and teachers' preferences match what actually

happens, then societal preferences are reflected in policy.

However, if societal preferences are not part of policy

outcome, then the government has been very influential, and

it has not incorporated societal preferences into policy.


I began preliminary research on education policy in

Zimbabwe during 1989 as a student at the University of

Zimbabwe. Then during a four month stay in Zimbabwe,

serving part of it as a Ford Foundation intern (May-August,

1991), the following data were collected: approximately

1,000 surveys of students, teachers, and headmasters;

interviews with key education actors such as the Minister of

Education; videotape interviews with teachers, headmasters,

and students at schools; my own diary entries from each

visit to schools; a tape of the National Association of

Secondary Headmasters Meeting with the Minister of

Education; and extensive secondary source materials

concerning education.

A variety of data sources was used to provide important

information. The surveys revealed societal perceptions of

their needs, preferences, resources, as well as their

perceptions of the government and the decision-making

process. The interviews offered detailed explanations of

preferences, resources, the education decision-making

process and the involvement of different actors in it.

Videotape interviews with teachers, headmasters, and

students at schools and my dairy entries provided me with

information on my perceptions of the schools and people I

met and they included some quotes I gathered from students,

teachers, headmasters, government officials, and state


A tape of the National Association of Secondary

Headmasters Meeting was made so that I could record the

proceedings and later I could document the important parts

of the meeting. Finally, the main secondary source

materials I used were newspapers, magazines, books, journal

articles, and Parliamentary debate reports. Parliamentary

debates provided extensive information on Parliamentarians

views of education policy which was discussed in Chapter 4.

The other secondary source materials had views of societal,

state, and governmental actors in them.

In order to obtain the attitudes of key societal

groups; teachers, headmasters, and students were surveyed.

Headmasters at the National Association of Secondary

Headmasters meeting in June 1991, and teachers in Zimbabwe

advised me on various representative schools to include in

the survey. As such, a representative sample of 31

secondary schools in Zimbabwe was selected to administer the

surveys. Figure 1-5 shows the location of these schools

throughout Zimbabwe.

It was determined that at least 30 schools were needed

because 30 is a large enough sample to use statistical

techniques. To insure representativeness of the sample,

additional steps were taken to select schools. At least

three schools in each of Zimbabwe's nine

geographical/administrative provinces were selected, one a

poorer quality school, one a medium quality school, and one

a high quality school. The quality of a school was

determined by the "0 level" (high school equivalent of SAT)

pass rate of the school's students. It was important to

have a representative sample in terms of school quality

because to represent the needs and preferences of all

students, teachers, and headmasters in Zimbabwe, I needed to

sample schools of differing quality. The needs of students

at a poorly equipped rural school are likely to differ from

students at a fully equipped urban school.

To make the sample as representative as possible, about

one half of the schools were chosen from rural areas and the

other half were chosen from urban areas because my sample

follows the national demographic breakdown of school

distribution. Of course, one important goal was to insure

that the different types of secondary schools in Zimbabwe

were represented. Therefore, the following types of schools

were selected: mine schools, commercial farm/company

schools, mission schools, rural local government

administered schools, former white only (Group A) government

schools, former black only (Group B) government schools,

overcrowded urban schools, and other private (non-mission)


Three different surveys composed by the author were

used. The first survey was the student survey which was

given to about 25-30 Form 4 (equivalent of llth grade in

USA) students from each school. There were 862 completed

student surveys from 31 schools.

1 Arundel
2 St. Georges
3 Prince Edward
4 Girls High School
5 Ellis Robins
6 Zengeza #1
7 Mbare #1

14 Pamushana Mission
15 Chatikobo
16 Rafomoyo

Matabeleland North
23 Christian Brothers
24 Mzilikazi
25 Wankie

Mashonaland East
8 Goromonzi
9 Rakodzi
10 Eagle Tanning

17 Fletcher
18 Mkoba #1
19 Zinatsa
20 Liebenberg

Mashonaland West
26 Rimuka #3
27 Chemukute
28 Munyaradzi

11 St. Augustines
12 Rusitu Mission
13 Nharira

Matabeleland South
21 Manama Mission
22 Mkhalipe

Mashonaland Central
29 Mazowe
30 Gwangwava
31 Rusambo

Figure 1-5. Schools Included in Survey


The second survey, the teacher survey, was submitted to

each headmaster who chose about four representative teachers

in each school to complete the survey--117 teacher surveys

were completed. Finally, the headmasters of all 31 schools

were given different surveys to fill out themselves but only

23 completed the survey. The response rate was over 90

percent because I administered and collected the majority of

surveys in person. Over 1,000 surveys have been entered

into the computer. Then, descriptive statistics were used

to analyze the data.

In addition to written surveys, extensive one-on-one

interviews were conducted. The surveys and interviews

revealed the amount and type of societal access to Central

Government officials through the actors' discussion of

resources they mobilized and channels of communication they

used. One of the most important interviews was with

Minister of Education, Fay Chung (personal communication,

August 14, 1991). National and local government education

administrators, teacher training college heads, a parent,

and university officials were also interviewed. The issue

of societal influence regarding university policy is very

timely because societal groups are challenging political

control of the university. Therefore, I consulted

university officials to find out their views of university


The variables in the model and hypothesis will be

measured by secondary source materials, interview

information, diary sources, and survey questions.

Governmental and societal interests in the model and

hypothesis will be measured from survey and interview

questions that asked students, teachers, headmasters, and

government officials what were their greatest needs and

suggestions. Interviews were used to measure societal and

governmental mobilization of "resources." Societal access

to central government officials is measured by what channels

of communication exist for societal actors to influence

government. The surveys asked the headmasters about how

much communication they have with central and local

government and further whether government listens to the


The Significance of This Study

By way of summary, it may be helpful to remind the

reader of the significance of this study. As I see it, this

research is important for six reasons. First, education is

an extremely important development issue; yet, it has

received little attention in recent years by academics

interested in public policy. Interest has focused on other

worthy issues such as environmental regulation. Education

signals opportunity to many but educational resources are

limited. Because of scarce resources in Africa, there is

often a trade-off between increasing the quantity of

education and improving its quality. Furthermore, the major

public policy studies on education have been conducted in

the United States, for example, by John Chubb of the

Brookings Institution and Terry Moe, a political scientist

at Stanford (1988).

Second, academics interested in the relationship

between education and politics in developing countries have

not paid adequate attention to societal interests in

education policy and government responsiveness to societal

preferences. Instead, research on education and politics in

the developing world has come from two predominant

paradigms: structural-functionalism and political economy.

Structural functionalists like James Coleman have focused on

education as a means of socializing and modernizing a

country (Coleman, 1965). The problem with the political

economy approach is that it assumes that the state is the

principal actor in education policy without testing to see

if this is the actual case. Unfortunately, the literature

we do have on education policy in developing countries

depicts societal groups as helpless, uninterested and

powerless victims of politics, economics and historical

structures. My research disagrees with this description.

This dissertation will test how interested and active

societal groups are in the policy-making process in

education. By testing this proposition, using a unique set


of data, I hope to make a contribution to our understanding

of public policy-making in developing countries. The policy

communities approach employed in this paper allows us to

examine the interaction between state, government, and

societal actors in educational policy-making in Zimbabwe.

Moreover, this approach attempts to transcend some of the

limits of the state-society relations literature. This

approach can be used in other countries and in the study of

different issues in order to clarify the policy decision-

making process.

The third reason why this study is important is that it

relates to the emerging literature on democratization in

Africa by providing hard data regarding the extent to which

governments act in response to societal influences.

Governance literature has sought to identify the potential

significance of societal actors, as opposed to government

actors. This study will assess the extent to which societal

preferences are included in education policy communities in

Zimbabwe. Further, I will identify what factors lead to

positive and negative interactions between government and

society in education policy communities.

The fourth reason this dissertation is important is

because it attempts to make a linkage between public opinion

(preferences) and policy. Page and Shapiro suggest, "The

responsiveness of government policies to citizens'

preferences is a central concern of various normative and


empirical theories of democracy" (Page and Shapiro, 1983, p.

175). By undertaking this research, one can find out what

the actors' preferences are, what occurs in the decision-

making process, and what actors are most influential in

policy results? The policy (government output) can be

determined from documents, interviews, surveys and other

available secondary source materials.

The fifth reason that this study is important is

because political scientists interested in Africa rarely

employ the survey method to collect data. The reason for

not collecting more survey data is that often researchers

are being denied research clearance to carry out the

research. This study will demonstrate that the survey

method, which is widely used in American political studies,

can be satisfactorily and usefully applied in Africa. Betty

Jo Dorsey's pre-independence survey of secondary schools in

Zimbabwe (Dorsey, 1975) and Philip Foster's study of schools

in Ghana (Foster, 1965) inspired the author.

Sixth, empirical testing of propositions is crucial in

comparative politics, "The central task in comparative

analysis is to identify concepts and variables of broad

relevance to the operation of social systems and then

establish procedures for "equivalent" measurement of these

variables within different systems in order to test

propositions about the relationships among variables"

(Graesser in Holm, 1978, p. 130). This study hopes to serve


as an example of how government-state-societal interactions

can be measured. Hopefully future research will utilize

this model and the policy communities approach.

Organization of the Dissertation

It is necessary to preview what will be found in the

rest of the dissertation. In the following chapter, the

interaction of actors in making Zimbabwean education policy

will be examined during the pre-colonial, colonial, and

post-colonial periods. Decision-making structures and

education policies created during the colonial period

influenced structures and policies made after independence.

Chapter Three will examine general findings from

surveys and interviews about how responsive the Zimbabwean

government is to society in policy communities. Chapters

Four to Seven will apply the four policy communities to

educational case studies in Zimbabwe. Chapter Four will

apply the Closed Government Policy Community to the case of

University of Zimbabwe policy. Chapter Five will apply the

Open Conflictual Policy Community to the case of school

fees. Chapter Six will apply the Closed Cooperative Policy

Community to the case of curriculum policy. Chapter Seven

will apply the Government Directed Policy Community to the

case of teacher training policy.

In Chapter Eight, the final chapter, the overall

utility of the policy communities approach to the study of


public policy will be evaluated in light of the research

findings. Chapter Eight will answer the question posed at

the beginning of this dissertation--what factors lead to

democratic public policy-making? Also, does the government

respond to societal influences in making policy?


The Important Role That Ethnicity and the Family
Has Played in the History of Education Policy

Before continuing to explore the colonial and post-

colonial periods, it is useful to understand the ethnic

composition of Zimbabwe and the role ethnicity has played in

history. Historically, Zimbabwe never had more than a small

percent of its population derived from European ethnic

groups. Seventy-five percent of Zimbabwean society is made

up of Shona people; while, nineteen percent of the

population is from the Ndebele ethnic group. Four percent

are members of either the Tonga, Venda, Shangaan or smaller

Black ethnic groups, and two percent are either of European,

Asian or Colored descent (Stoneman and Cliffe, 1989, p.


My sample of students from the survey reflects the

percentage of Shona in the population. I have 75.5% Shona

students in my sample. My sample under-represents the

Ndebele ethnic group in that eight percent of my student

sample are Ndebeles; yet they make up about 19% of the

population. One of the schools in an Ndebele area which I

was hoping to include in my sample was impossible to


incorporate. To make up for the under-representation of the

Ndebele group, I over-represented some of the other small

black ethnic groups in Zimbabwe. For example, although four

percent of the population is made up of Tonga, Venda,

Shangaan or other small black ethnic groups my sample

includes nine percent from these various small Black ethnic


Although only two percent of the Zimbabwean population

are of either European, Asian, or Colored descent my sample

includes almost seven percent from these groups. Part of

the difference between overall population of the country and

my form four school sample can be explained by the fact that

there is a higher percentage of European, Asian, and Colored

students in form four than in the population as a whole.

This is because these groups have more money and can afford

to attend form four at a higher rate than other ethnic


Ethnic struggles over values and power have been a

prominent aspect of the struggle between the government and

society for centuries. Documented ethnic struggles over

education policy date back to the period of the Munhumutapa

Kingdom in the 1500's. During this time, the Portuguese,

Arab traders, and Shona ethnic groups were in conflict over

who had the power to make policy.

Ethnicity continued to be an important issue throughout

Zimbabwe's history. For example, during the British


colonial times, the British wanted to replace ascriptive and

particularistic traditional norms with universal,

achievement-oriented-norms of the British (Fuller, 1991, p.

9-10). The newly independent government in Zimbabwe, like

other Third World governments, claims that it needs a

uniform national curriculum to incorporate disparate ethnic

and language groups into the modern polity and economy

(Fuller, 1991, p. 20). According to Bruce Fuller, the

action of the new Zimbabwe government is typical of a newly

independent Third World government, "Following national

independence, and anxious to lessen constraints of race,

class, and caste, centralized Third World states promptly

took over colonial school systems and pushed standardization

of curricula, teacher preparation, and school management"

(Fuller, 1991, p. 9).

The Zimbabwean government desires unity and

standardization in education as well as unity in the

political arena. In 1987, the two major political parties

in Zimbabwe merged into one which the President hoped would

lead to unity. ZANU (which represented people predominantly

of the Shona ethnic group) and ZAPU (which represented the

Ndebele ethnic group during the liberation war) merged into

a new ZANU, that Mugabe hoped would represent all ethnic

groups in Zimbabwe.

Ethnic groups have played important roles in education

policy-making. Zimbabwe's Minister of Education, Fay Chung,


confirmed in an interview with the author that ethnicity did

in fact affect education policy (personal communication,

August 14, 1991). Before 1987, the District Councils (local

governmental bodies), not the National Ministry of

Education, were given the power to hire teachers. Yet,

ethnic considerations biased the selection, said Minister


There was a bias in hiring, because someone's relative
or someone of the same ethnic group may be hired over
others who are more qualified. The teachers pressured
our Ministry to do something about these bias. So, in
1987, we gave into pressure from teachers and took the
power away from the District Councils and centralized
hiring teachers (F. Chung, personal communication,
August 14, 1991).

During this period of bias in teacher hiring some District

Council members served as patrons to teachers that they

hired. The teachers then became indebted to them and served

as clients. As a result an inefficient and detrimental

patron-client relationship was formed, which was only

destroyed by the actions of the National Ministry of


Some Zimbabweans have suggested that the Minister of

Education, Fay Chung, who is of Chinese heritage, is immune

to demands by ethnic groups, such as Shona or Ndebele

groups, because of her Chinese heritage. Chung suggested

that she did not mind being unpopular and she offered an

example (personal communication, August 14, 1991). In 1988

she would not allow parents to control the Ministry and

create more and more schools because she wanted to focus on

quality not quantity, "We slowed down the building of

schools, because we realized one-half that were built

quickly were bad anyway. The people began to realize that

it made sense to improve schools that were built, so we made

efforts to improve those that were built and we improved 430

of them and 320 are still bad." The previous Minister of

Education, Dzingai Mutumbuka, interestingly enough did

respond to more local demands to build schools--he was

Shona, the largest ethnic group in Zimbabwe.

Moreover, the national government has been accused of

ethnic bias in the allocation of resources. In a video

interview, a teacher from a Matabeleland school (minority

ethnic area with 19% of the population, mostly Ndebeles),

said the biggest problem in educational policy is "nepotism,

in which Matabeleland schools receive less than schools in

Mashonaland (majority area)." This interview revealed that

a teacher in a minority ethnic area feels as though his

school has received fewer and worst resources than areas in

the majority ethnic area--Mashonaland. This may be an

isolated comment or it may represent a more widespread

perception among non-Shona people in Zimbabwe.

There are two types of education discussed in this

chapter, formal and non-formal education. Formal education,

in this dissertation, refers to institutionalized education

in a school; whereas, non-formal education refers to

education outside of the school. Prior to the arrival of


the European and continuing today, is African indigenous or

non-formal education. This type of education, as described

by Betty Jo Dorsey, a Zimbabwean scholar, is controlled by

society, predominantly by the family group. According to

Dorsey, traditional (indigenous) education and socialization

were characterized by two types of knowledge passed onto the

young. One type is a body of knowledge that helped the

young carry out daily activities and occasional rituals in

the household and village. Dorsey describes the second type

of knowledge as the body of "traditional lore concerning

natural phenomena, custom and tribal history" (Dorsey, 1975,

p. 39). This education was often imparted by family members

at initiation ceremonies. Education was frequently imparted

through stories and riddles told by senior relatives, such

as the grandmother, around the fire after the evening meal

(Dorsey, 1975, p. 39).

Dorsey describes "traditional society" as one in which

status in society is based on ascriptive criteria such as

sex, age and lineage and not based on achievement.

"Traditional education, then tended to be an integrative

function for a relatively noncompetitive society whose main

aim was the transmission of a common culture from one

generation to the next and the maintenance of social

cohesion" (Dorsey, 1975, p. 39). This "traditional" form of

education which contributed to the social cohesion of a

society was based on a subsistence agricultural system.


Although aspects remain of this traditional education

system controlled by society, formal educational

institutions developed (particularly under British

colonialization) and replaced some aspects of traditional

non-formal education. Formal education developed as a

result of the complex forms of division of labor that have

formed with the spread of trade (Dorsey, 1975, p. 39). But,

some forms of traditional education continue and complement

formal institutional education.

Zimbabwe Education Policy From a Historical Perspective

Because historians of the "third world" have tended to

rely so heavily on government documents, the prevailing view

of these countries is highly government-centered. From

these accounts Zimbabweans are taught that the government

imposes its will on the people and that there is rarely any

societal resistance. This leaves one with the false

impression that the government is autonomous and that

societal groups are weak and helpless. As a counterpoint,

Zimbabwean historian Dickson Mungazi, has gathered extensive

data from societal groups (interviews, archives, etc.) as

well as from government documents which enable him to

present a more realistic and balanced picture of government

and societal conflict in Zimbabwe. His findings will be

discussed later in this chapter.

When we examine government-society relationships in

education policy in Zimbabwe, two distinctive historical

periods become apparent: (1) colonial and (2) post-

colonial. This chapter will focus extensively on the

colonial education policies to lay a foundation for the

post-colonial policies discussed extensively in later

chapters of the dissertation. However, it should be noted

that there was education before the colonial period, such as

the non-formal education discussed above, and such as

education under the Munhumutapa kingdom which is discussed

in the next section of this chapter. Interestingly enough,

when both societal and government information is gathered,

we find instances of strong societal influence on education

policy during all these periods. During some of these

periods, i.e. the colonial one, the government acts against

the interests of the majority of society. However, societal

pressure on government has remained an important feature

that affects education policy even during the periods when

the government appears to be very authoritarian. In the

next section, societal pressure on the government of the

Munhumutapa Kingdom will be discussed.

Munhumutapa Kingdom and Societal Pressures on Education

Beginning in the fourteenth century and remaining

strong through the early 1600's, the Munhumutapa Kingdom

ruled over much of Zimbabwe. One of the oldest documented


conflicts between society and the Munhumutapa government in

Zimbabwe over education policy was the result of

interactions between the Catholic church (Dominican) and the

government of Munhumutapa in 1561 (Mudenge, 1986, p. 3-10).

The powerful Munhumutapa kingdom in the 1500's was heavily

involved in trade with both the Arabs and the Portuguese.

With this powerful government, one may assume that education

policy was dominated by the government, but this was not the

case. The government of the Munhumutapa was not autonomous

in making education policy. Instead, societal groups were

very influential in shaping the government's education


Education policy during the Munhumutapa regime fits

into the broader context of a political struggle between

traditional leaders, Arabs, Portuguese traders, and

missionaries, for influence over the political kingdom.

Catholic church Friar Silveira from Portugal converted

Munhumutapa King, Negomo Mupunzagutu, to Christianity

through western education. Yet, Friar Silveira's influence

on the Munhumutapa government began to threaten certain

vested interests in the government and society, such as

traditional Shona religious leaders and Moslem traders, who

were close confidants of the Munhumatapa emperor (Mudenge,

1986, p. 3-10).

The Munhumutapa empire had to contend with two

competing societal groups with different interests; one

societal group made up of Friar Silveira and his Christian

converts and another group made up of Shona traditional

groups and Moslem traders (Mudenge, 1986, p. 3-10). The

groups had competing views of the world that affected their

education policies and religious practices.

While the Christian societal group influenced

government policy, societal pressure from Moslem traders and

Shona traditional groups convinced Munhumutapa King Negomo

that Friar Silveira was a spy for the Portuguese government.

King Negomo was told that the Portuguese would try to wrest

control of the Munhumutapa kingdom for the purposes of

Portuguese colonialization. Of course, the Arabs and

traditional leaders were concerned with their own political

influence within the regime. Muslims made the case against

Christian religious and educational influence, and the

societal pressures forced the King to act against Silveira.

The result was the execution of Silveira in 1561 (Mudenge,

1986, p. 3-10).

The western education of Munhumutapa Kings and Princes

also led to the cultural alienation of the leaders from

their followers (Mudenge, 1986, p. 30). Munhumutapa King

Mamvura, educated by Portuguese Friars, was considered a

puppet of Portugal and a sellout to his people (Mudenge,

1986, p. 20-21). This cultural alienation between leaders

and the people has remained an important theme in Zimbabwe

during British colonization and it continues in present

independent Zimbabwe.

British Missionary and Colonial Influence in Zimbabwe--1850-
1939--Paternalist Phase

British missionary activity in Zimbabwe dates back to

the 1850-1890 period in which Missionary David Livingstone

began a period of missionary domination of education.

British colonization soon followed. The British supported a

continuation of religious education by the missionaries in

order to pacify Africans into accepting European domination.

Mungazi uses Brazilian education specialist Paulo

Freire's three stages of cultural conflict to characterize

the different periods of British colonialization in

Zimbabwe. Freire argues that there are three stages of the

effect of education on the conflict between the people of

two different cultures. Mungazi finds that in the first

stage of colonialization (1890-1939) the British colonists

tried to justify colonization of the Africans by suggesting

that British culture was superior, and African culture was

primitive and inferior (Mungazi, 1992, pp. xv-xxvi). This

indeed characterizes the paternalism that was evident during

this period. The former tried to impose their cultural

values on the latter. Education policies by the British

were tools of the government in their broader plan to

dominate the Africans politically, economically, and

socially. The other two stages, the consciousness awakening


stage and the liberation war stage, will be considered later

in other sub-sections of this chapter.

Cecil John Rhodes initiated European settlement in

Zimbabwe in 1890 under the auspices of the British South

Africa Company (BSAC). He aimed to build an empire so that

he could reap the economic rewards of a mineral rich and

agriculture rich region. With the arrival of the BSAC, came

an extension of political control over education by the

BSAC. Under royal charter, the BSAC ruled over white

settlers and Africans from 1890 until 1923. During that

time, Company personnel, which served as the new

"government" in Zimbabwe, offered land grants to mission

personnel to set up mission schools (Dorsey, 1975, p. 40).

Company personnel offered land because they wanted Africans

to be literate for labor purposes, and they believed that

missionaries would be more effective in subduing the African

population than policemen (Hassing, 1960, p. 142). The cost

of providing education during the early period of European

settlement was borne by the missions and the African people

(Dorsey, 1975, p. 40), but it clearly benefitted the

political ambitions of Rhodes.

Academic Franklin Parker accurately characterizes the

1899 to 1927 educational period as one of Christian

missionary influence, industrial education bias (rather than

academic education given to whites), and paternalism.

Parker notes that the first Education Ordinance was offered


on December 15, 1899, and it created an Education Department

and an inspectorate, and set grant earning conditions for

separate and segregated white, Asian, Colored, and African

schools (Parker, 1959, p. 28). In 1920, a central-teacher-

training school was opened. The first government school for

Africans was also opened the same year; it served to teach

simple craft skills, agricultural techniques, and health

demonstrations. Whites did not want Africans to be educated

for anything but non-competitive labor positions in white

enterprises. The Europeans had strong influential trade

unions that prevented Africans from receiving higher

technical education from the government (Parker, 1959, p.

31). The Europeans did not want the Africans to receive the

same educational advantages because they feared that

Africans would then demand political and economic equality.

Zimbabwe (then known as Southern Rhodesia) became a

self-governing territory of Britain in 1923. This meant

that it was no longer under the BSAC. To some Southern

Rhodesians, the government appeared to be more like the

government of an independent country than of a dependent

colony (Sylvester, 1991, p. 31-32). According to this

source, the Southern Rhodesian government was gradually able

to gain control over colony finances. In 1927, a separate

Department of Native Education was created to administer

African education. The main purpose of African education,

according to the Europeans, was to create an African labor

pool for the Europeans. The creation of a separate

Department of Education for Africans was also significant

because it signalled that Africans and Europeans should

receive different education. The sentiment of one white in

1927, that blacks and whites should be treated differently,

represents the popular opinion of whites then,

We do not intend to hand over this country to the
native population or to admit them to the same society
or political position as we occupy ourselves, but we do
wish to do them justice and enable them to better their
position in every way; but we should make no pretense
of educating them in exactly the same way as we do
Europeans (Parker, 1959, p. 29).

Mungazi calls the period from 1924-1939 the adjustment

period in which Africans wanted western education for three

reasons: basic literacy, socioeconomic well-being, and the

need for political development, which would make their

adjustment to colonial conditions less painful (Mungazi,

1992, p. 19). Methodist Rev. Josiah Chimbadzwa offered an

example of this when he said,

I cried because I wanted to stay in school to learn to
read and to write. This was the only way for us to
survive in the society which the British controlled.
We had no choice, either we had to learn to read and to
write or the white man would continue to take advantage
of us. The ability to read and to write did not stop
the white man from exploiting us, but it made it
possible for us to become more aware of it than we had
been the past. Therefore, while the white man wanted
to use our ability to read and to write to force us to
become more productive laborers, we ourselves saw other
benefits which accrued to us and which the white man
could not see (Mungazi, 1992, p. 21).

A critical consciousness among the Africans became reality;

this was an unforeseen result of educational policy, says


Mungazi (1992, p. 22). This critical consciousness will be

discussed in detail in the next section.

Rise of African Consciousness and Conflicts with Government-
-1940-61--Growing Awareness Phase

The post-1935 period, as characterized by Parker, was

the "awakening" period in which Africans accepted western

schooling and demanded it with passion while outwardly

showing growing antipathy to white domination and missionary

guidance (Parker, 1959, p. 31). The awakening period, which

according to Mungazi, was from 1940-61, involved the rise of

African consciousness. Interestingly enough, World War II

served as a major catalyst to African nationalism, self

consciousness, and demands for better education. After

WWII, Africans who served in Europe returned to Southern

Rhodesia and were frustrated that they could not secure as

good of an education as the whites that they had fought


Africans who took part in the war in Europe learned

that Britain adopted educational policies that practically

eliminated them from participating in the political, social,

and economic activities of their countries (Mungazi, 1992,

p. 40). Africans fought in Europe in the Allied armies

because they were promised by the colonial governments that

their condition of life would be improved as a reward for

their service (Mungazi, 1992, p. 40).


Upon their return to Southern Rhodesia, Africans shared

their experiences in Europe and they stared to question the

legitimacy of the colonial leadership and the belief that

only the Europeans were capable of running a good government

(Mungazi, 1992, p. 41). Africans demanded better

educational opportunities. New societal groups developed to

meet the needs of Africans regarding education. These

groups posed a challenge to the colonial government. The

African Artisans Guild, the Council of African Chiefs, and

the National African War Fund helped Africans establish

educational objectives and this self-consciousness alarmed

the colonial government (Mungazi, 1992, p. 41). The

Europeans feared that the consciousness would lead Africans

to make political and economic demands for equality with the

Europeans (Mungazi, 1992, p. 69).

Although missions began secondary education for

Africans in 1939 in Penhalonga (near Mutare) at St.

Augustine's Mission, the Government did not open its first

African secondary school until 1946 at Goromonzi (Parker,

1959, p. 30). Besides being the first to offer secondary

education for the Africans, Missions also challenged the

colonial government to do more for African education. For

instance, in 1946, the Methodist Church issued a statement

that called for a new educational policy that would benefit

Africans (Mungazi, 1992).


Parker noted that after 1942, mounting African societal

pressures for more educational facilities prompted a

government inquiry into African education, which occurred in

1951. The result was 140 recommendations that would lead to

more trained teachers; five years of education for all; and

the expansion of secondary and vocational schools. In 1956,

a government teacher training school was opened. It was

desperately needed since 70% of the teachers, in 1950, were

untrained (Parker, 1959, p. 30). While some improvements

were made as a result of societal pressures (one result was

the 1959 African Education Bill that passed in the

legislature and provided a unified teaching service with

conditions to enhance the profession), there were still many

inequities in African education (Parker, 1959, p. 30). The

demands for better education and more land by Africans

helped rally them to fight white rule in Southern Rhodesia.

This led to the liberation war in 1966.

In 1953, Southern Rhodesia joined Northern Rhodesia

(now Zambia) and Nyasaland (now Malawi) in a federation of

territories, and as a result White, Asian, and Colored

education became a federal responsibility that same year.

However, African education remained the responsibility of

each territory. Godfrey Huggins, Prime Minister of Southern

Rhodesia from 1933 to 1953 and Prime Minister of the

Federation of Rhodesia and Nyasaland from 1953-56, expressed

his alarm at the growing consciousness among Africans,

We have the sad spectacle of many of our so-called
educated Africans wasting their energies on some
completely unattainable objectives such as self-
government. The African with one foot in his primitive
culture presents two problems for the government. The
first is how to deal with the immediate situation
caused by the misguided and so-called African
intellectuals. The second is how to develop the
African so that the educated ones do not waste their
time on sterile and futile nationalistic agitation.
Our task as a government is to reverse the
claimed consciousness which is believed to have come
about as a result of the war (Mungazi, 1992, p. 52).

Yet, the consciousness of African society only increased due

to unfulfilled needs. As a result, Africans put more

pressure on the government for better education.

A multiracial University College of Rhodesia and

Nyasaland, which opened in 1957, offered more equal terms

for all races (Parker, 1959, p. 28). Yet, pre-collegiate

education was still heavily in the control of missionaries.

In 1959, 95% of the schools were administered by

missionaries. Prime Minister Garfield Todd, of Southern

Rhodesia, coming from a missionary background, tried to

improve education policy for Africans. Yet, he was

pressured by white voters not to make changes. He was

ultimately removed from office, in 1958, after serving five

years as Prime Minister (Mungazi, 1992, p. xxi).

One government policy, the Native Councils, was created

by the government to curtail opportunities for the Africans.

Two factors explain how the Native Councils were meant to

curtail opportunities for the Africans. First, the Native

Councils Act of 1937, and amendments in 1943 and in 1957,