Distributed leadership and core values as critical constructs in interorganizational networks

Material Information

Distributed leadership and core values as critical constructs in interorganizational networks
Schwartz, Teresa Petito
Publication Date:
Physical Description:
xii, 286 leaves : ; 29 cm

Thesis/Dissertation Information

Doctorate ( Doctor of Philosophy)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
School of Public Affairs, CU Denver
Degree Disciplines:
Public Administration
Committee Chair:
Gage, Robert W.
Committee Co-Chair:
deLeon, Peter
Committee Members:
Pogrebin, Mark
Fitzpatrick, Jody
Wurtele, Sandy K.


Subjects / Keywords:
Human services -- Management ( lcsh )
Interorganizational relations ( lcsh )
Leadership ( lcsh )
Human services -- Management ( fast )
Interorganizational relations ( fast )
Leadership ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaves 285-286).
General Note:
Submitted in partial fulfillment of the requirements for the degree, Doctor of Philosophy, Public Administration.
General Note:
School of Public Affairs
Statement of Responsibility:
by Teresa Petito Schwartz.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
39683614 ( OCLC )
LD1190.P86 1997d .S39 ( lcc )


This item has the following downloads:

Full Text
Teresa Petito Schwartz
BA, Oberlin College, 1972
MEcL, University of North Carolina, 1975
MPA, University of Colorado, 1991
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Public Administration

(§) 1997 by Teresa Petito Schwartz
All rights reserved

This thesis for the Doctor of Philosophy
degree by
Teresa Petito Schwartz
has been approved
Sandy K. Wurtele

Schwartz, Teresa Petito (PhJ), Public Administration)
Distributed Leadership and Core Values as Critical Constructs in
Interorganizational Networks
Thesis directed by Professor Robert W. Gage
The study of interorganizational networks has contemporary relevance
for public administration. The use of networked action to address
complex social problems at both systemic/policy and service delivery
levels has increased over the past decade. This study focuses upon
networks in human services. The current crises, related to increasing
needs in a time of decreasing resources, are documented, as is the
interrelated nature of social problems. The fragmented nature of
involved systems contributes to the problems and leads to the increasing
call for interagency collaboration.
Intergovernmental management (IGM) is the theoretical base of this
research. As described in the literature, IGM has a problem-solving
focus, provides a framework for understanding inter-sectoral exchanges,
and is a conceptual tool for action. Other literature relevant to this
research includes selected aspects of leadership literature, policy sciences
literature, and network theory.
Case studies of two human services networks in Colorado were
conducted. It was hypothesized that network participants would be
executive/senior management level on an ongoing basis. This was found
to be true early, but shifted over time, with negative impact on goal

achievement particularly systemic reform goals. The ability of the
networks to engage in meaningful strategic planning was also
negatively impacted. The second hypothesis suggests the presence of a
distributed model of leadership; this was confirmed (to certain extents
at certain times) in both cases. Lastly, the proposition that a shared
core value of commitment to the collaborative process would be present
was partially confirmed. A corollary stating that the above core value
and other discrepant values among members could both be maintained
could not be confirmed by this research. Unanticipated findings
emerged in three major areas the developmental aspects of coalitions,
the role and impact of diversity among participants, and training needs
in implementing network structures.
Conclusions point to: the importance of executive participation; the
importance of defining, implementing, and maintaining a model of
distributed leadership; and, the role of values and the importance of
clear articulation of shared and other values. Implications for both
theory and practice are identified and delineated, as are areas for
future research.
This abstract
I recommend
accurately represents the content of the candidate's thesis,
its publication. /PP

To Pere, who always knew that this was forthcoming.

The intellectual guidance and ongoing support and patience of my
thesis Director, Professor Robert W. Gage, are gratefully acknowledged,
as is the participation of my entire committee.
My sincere thanks are offered to the San Luis Valley Coalition for
Youth Services and the Boulder County Prevention Connection for their
participation in this research.
Thanks are extended to Sandra L. Adams for her diligent proofreading
and editorial assistance.
The help and support of my husband, son, parents, mother-in-law, and
friends in my pursuit of the Doctor of Philosophy degree are
acknowledged with gratitude and love.

1. INTRODUCTION______________________________________________1
Context of the Problem--------------------------------2
Problem Statement____________________________________10
Problem Formulation-----------------------------10
Operational Definitions______________________________17
Issues for Examination-------------------------------18
Research Hypotheses__-_______________________________18
Shared Leadership in Implementation Structures--19
The Importance of Shared Values-----------------22
Relevance of the Study--------------------------23
2. REVIEW OF THE LITERATURE_________________________________25
Intergovernmental Management-------------------------25
Selected Aspects of the Leadership Literature--------30
Creation and Communication of Vision------------32
The Definition of Leadership Tasks--------------34
The Role of Values..............................37
Power in Networks----------------------------40
Shared Leadership in Implementation Structures--43
The Policy Sciences Literature-----------------------48
Fragmentation and Complexity in the Policy
Environment.................................... 49

Issue and Policy Networks___________________________51
Implementation Theory and Networks__________________54
The Network Literature___________________________________57
The Structural Approach_____________________________57
The Resource Dependence and Political Economic
Networks and Their Environment______________________62
Understanding and Managing Interorganizational
a RESEARCH DESIGN AND METHODOLOGY________________________________77
The Case Study Approach----------------------------------79
Rationale for the Use of the Case Study_____________79
Description of the Case Study Process---------------82
Conduct of the Research: Data Collection-----------------84
Site Selection and Access---------------------------84
Field Procedures____________________________________87
Data Analysis____________________________________________89
Criteria of Soundness------------------------------------92

Contextual Variables Impacting the SLVC-----------------98
Initiation of the Coalition----------------------------103
The SLV Coalition 1990 1992__________________________107
Ongoing Successes and Emerging Challenges--------------116
Conflict, Cycles, the Need for Renewal:
1994 June 1995_______________________________________120
Research Questions and the SLV Case Study--------------124
Contextual Factors Impacting the BCC-------------------136
Political Factors---------------------------------138
Initiation of the Coalition----------------------------140
BCC: October 1991 through September of 1992------------143

Forging the Boulder County Prevention Connection:
Year 2 (October 1992 September 1993)-------------------148
Year 3: Solidifying Gains, Grappling with Ongoing
Issues, Moving Forward-----------------------------------152
November, 1994 through May, 1996: Questions of Meaning
and the Renewal or Termination Dispute-------------------156
Research Questions and the BCPC Case Study---------------168
Congruence with Definition of Interorganizational
Case Study Networks and Their Environments---------------181
Network Initiation: Antecedents and Barriers-------------183
Case Study Networks and the Intergovernmental
Management Context---------------------------------------187
Hypothesis 1--------------------------------------------.193
Hypothesis 2---------------------------------------------202
Hypothesis 2--------------------------------------------212
Unanticipated Findings.---------------------------------221
Developmental Aspects of Coalitions----------------221
Impact of Diversity--------------------------230
The Importance of Training------------------------235
7. CONCLUSIONS AND IMPLICATIONS----------------------------------237
Introduction------------------------------------------- 237
Summary of Chapters One through Six_---------------239
Conclusions: Hypothesized Findings....................242

Hypothesis 1
Hypothesis 2--------------------------------------246
Hypothesis 3--------------------------------------253
Interactions Among the Conclusions---------------------255
Non-Hypothesized Findings------------------------------256
Limitations and Implications for Practice--------------258
Implications for Theory Development and Suggestions
for Further Research-----------------------------------261
Relevance of Network Theory Development----------------262

This dissertation presents a study of leadership in
interorganizational networks (IONs) (alternately termed implementation
structures or collaborations [Gage & Mandell, 1990; Kagan, Rivera, &
Parker, 1991; Rossy & Mandell, 1992; Melaville & Blank, 199ID specific
to human services delivery across levels of government and including
the private not-for-profit sector and citizen stakeholders. It is the
intent of the research to increase understanding of the establishment
and functioning of interorganizational networks through study of
participants, leadership models, leadership tasks, and core values as they
are evidenced in the network context and relate to the process of
working toward achievement of network goals.
The dissertation is comprised of the following chapters. In this
first chapter, a statement of the problem, including context and specific
background, necessary working operational definitions, relevance and
scope of the study, and a general statement of issues to be examined
mare presented. The chapter concludes with the specific research
hypotheses to be investigated. The theory in which the study is
grounded and the relevant literature are reviewed in Chapter 2.
Chapter 3 describes methods and procedures, including research design,
data collection methods, data analysis methods, and the rationales for

each. In Chapters 4 and 5 each case is presented individually. Chapter
6 includes cross-case analysis, discussion of the findings related to each
hypothesis, and presentation and discussion of unanticipated findings.
Conclusions, limitations, recommendations, and relevance of the study
are discussed in Chapter 7.
Because, as Rogers and Mulford (1982) explain, "Different
philosophies and strategies of coordination are closely associated with
general environmental conditions present at a particular time" (p. 32),
this introductory chapter will continue by placing the problem in
Context of the Problem
The current decade has clearly been a time of shrinking
resources and expanded needs in the human services arena. Across the
board, child and family serving agencies are being not only asked, but
required, to do more with less (Sink, 1992). Increasingly, human
services of all types (e.g. health, mental health, child and adult welfare,
child and adult justice systems, vocational training systems) are
referred to as being in crisis (Agranoff, 1990, 1991; Goldberg & Stewart,
1996; Kirchoff, 1996; Melaville & Blank, 1991; National Commission on
Children, 1991; National League of Cities, 1991; Schorr, 1988). Agranoff
(1990) defines a crisis as "an unstable state of affairs in which change
is impending" (p. 61). Some of the pressures upon human service
delivery systems creating the present crises are documented as follows.

Beyond Rhetoric, the report of the National Commission on
Children (1991) found that thirty-two million Americans, including 8.3
million children, have no health insurance coverage. At present, "The
number of people lacking health insurance has increased substantially,
to approximately 42 million" (National Center for Children in Poverty,
1996). Nationally, 15.1% of all children birth to age 18 have no health
insurance (Childrens Defense Fund, 1997); this figure holds true for
Colorado as well, where 15% of children are uninsured (Shulman, 1997).
Also according to the Childrens Defense Fund,
Every day the number of children without private
health insurance grows by nearly 3,300. And some
children never get a healthy start in life, as at least
400,000 uninsured women give birth each year often
without access to good prenatal and postnatal care.
{CDF Reports, May, 1997, p. 2)
The problems of the uninsured continue past pregnancy and childbirth
throughout life. According to the Center for Studying Health System
Change (1997), the number of uninsured persons with unmet health
care needs is growing. In a recent survey, 45% of respondents stated
that they had unmet health care needs compared with 6% in a like
1987 survey. Beyond the direct impact on health,
Perhaps less obvious, but no less damaging, are
the educational, social, and economic costs to the
children who lack health insurance, and to the
nation. Children who are unnecessarily ill can
miss days, weeks, or even months of school and
their parents can miss significant periods of work.
(Childrens Defense Fund, 1997, p. 23)

The National Commission on Children also expressed concern
regarding poverty among children. In 1991, they found that
Children are the poorest AmericansNearly 13
million children live in poverty, more than 2
million more than a decade ago. Many of these
children are desperately poor, nearly five million
live in families with incomes less than half the
federal poverty level, (p. 24)
Children are still the poorest Americans. A 1996 report of the National
Center for Children in Poverty found that
*In 1994, as many as 45 percent of young children
were living in poverty or near poverty (Le, in
families with incomes below 185 percent of the
Federal poverty line);
*Between 1975 and 1994, the young child poverty
rate increased by 39 percent; and
*The proportion of young children in extreme poverty
(Le., in families with incomes below 50 percent of the
poverty line, doubled, from 6 to 12 percent
(P- 1)
The impact of child poverty is severe. The immediate and long-
lasting effects are captured by Weiss and Halperns (1990) statement
In early childhood, parents constitute the
primary environment for the child (Musick, 1987).
The devastating effect of poverty is that it
not only threatens infants physical well-being
from the moment of conception, but simultaneously
undermines the capacity of their parents to
protect, nurture, and guide them. (p. 3)

Cook and Brown (1993) confirm, stating that "Probably more
than any other factor, poverty limits the capacity of families, and
begins harming the young before they are bom" (p. 11). Weissbourd
(1991) speaks of "the litany of poor childrens distress" and "of losing a
wide band of poor children to early death, to lengthy incarceration,
to Joblessness, to one form of misery or another" (p. 2). These miseries
are further detailed in a study by the Kansas State Health
Department (State of Americas Children, 1994) which found that poor
are five times as likely as nonpoor children
to die from infections and parasitic disease,
four times as likely to die from drowning or
suffocation, three times as likely to die from
all causes combined, and twice as likely to die
from car accidents and fires. And for every
poor child who dies, others suffer needless and
often expensive health and learning problems.
(p. 2)
In Colorado, 130,000 children and youth (15%) live in poverty;
this percentage is unchanged since 1990, while the population has
increased 450,000, 81,000 are chronically poor (poor two or more
consecutive years) (Shulman, 1997). Brooks-Gunn (1995) states, "The
well-being of children is drastically reduced by being reared in poverty"
(p. 87) and Bailey (1997) asserts, "Poverty is a multidimensional problem
in which a number of potent risk factors interact to create the
likelihood of adverse outcomes" (p. 231).

Ways and Means for Children and Families (1991), a report of
the National League of Cities, states
Many of our children do not complete school,
and others that do are ill-prepared to fill the
challenging jobs of the 1990s. Many of our
children and families are homeless, needing
services far beyond those that are available.
Infant mortality rates and other health indicators
in some of our cities now lag behind those of
third world countries. Children are attempting to
learn in educational systems where the primary
concerns of teachers and obstacles to learning
are drug abuse, violence and teen pregnancy.
(p. 1)
As noted above, the high school dropout problem is serious^ furthermore,
"Every year the US. graduates 700,000 young people who are unable
to read their diplomas" (Kagan, Rivera & Parker, 1991, p. 2). In
Colorado, where the number of students has increased 11% since J.991,
there has been a 26% increase in the state dropout rate in the past
decade (Shulman, 1997). Nationally, Ingersoll and LeBouef (1997) found
that "In 1993, among 16-to-24-year-olds, approximately 3.4 million (11%)
had not completed high school and were not currently enrolled in
school" (p. 1).
Problems in schools continue to be consequential
Compared to past years, Colorado schools have
more children, homeless children, and children
whose problems outside of school affect the
classroom, their learning, and the learning of
other children (Shulman, 1997)

In the United States, "In 1994, courts formally processed
approximately 36.400 truancy cases, a 35% increase since 1990 and a
67% increase since 1985" and increases in aggressive behavior, violence,
and weapons possessions are also seen in national date (Ingersoll &
LeBouef, 1997). In Colorado,
An average of 10% of all students had been
threatened or injured with a weapon on school
property during the past year. Forty-two percent
of males and 26% of females reported they were
in a physical fight at least during the past 12
months; 4% of all students had to be treated by
a doctor.
(1995 Colorado Youth Risk Behavior Survey, p. 9)
As Ingersoll and LeBouef (1997) explain,
The costs of these problems, both for children and
for society, are prohibitively high_Each years class
of dropouts costs the Nation more than $240 billion
in lost earnings and foregone taxes over their
lifetimes: Billions more will be spent on crime control,
welfare, healthcare, and other social services, (p. 2)
Child abuse is another costly issue, both in human and fiscal
terms, which continues to worsen. In 1994, child abuse reporting
increased 45% from the previous year, exceeding 3.1 million reports; a
like increase in substantiated reports was noted as well; over one
million new cases entered the child protective services system.
Of those that were substantiated, approximately
72% received some form of service and 14% of
the cases involved the use of foster care.

(National Committee to Prevent Child Abuse
Memorandum, Volume II, #5, May 1995).
In Colorado, "There were 9,745 confirmed reports of child abuse
and neglect in 1995. Twenty-nine children died from abuse or neglect
the year before" (Shulman, 1997, p. 25). Moreover, Shulman (1997)
found that "Confirmed reports of child neglect have doubled over the
past decade" (p. 25).
Nor are the costly problems with which human services
agencies must deal limited to children and families. In addition to age-
based client groups, there are also what Agranoff (1991) has termed
impairment-based i.e, mentally ill, homeless, physically handicapped,
and developmentally disabled. He notes of the 1980s,
Study after study indicated that these primary
problems were bound up in webs of related
problems: large numbers of mentally ill lacking
care and basic living, families without shelter
and income, long-term unemployed without food,
and persons with AIDS who suffer a host of
discriminatory actions, (p. 538)
In Colorado, and throughout the nation, while many state and
local agencies are presently faced with additional budget cuts every
new fiscal year, the social problems with which human service agencies
are expected to assist and ameliorate are increasing in number and
intensity. Nationally,
On almost all measures poverty, declining
income, crime, infant mortality, out-of-wedlock
and teenage births, school achievement,
homelessness, family violence, and substance

abuse the condition of families has worsened
(Goldberg & Stewart, 1996)
At the same time, resources to meet these demands are
decreasing. This is most clearly seen in the dramatic decrease in
federal funds to states and localities for social programming through
the 1980s (Gray, 1989), but continues to be seen federally and in state
and local budgets as well, (Burlingame et. al, 1996). Additionally, "large
infusions of new funds for social programs are not forthcoming from
any level of government" (Hodgkinson, 1989, p. 25).
The crises are compounded even beyond inadequate resources by
deficient organizational forms and delivery systems, described by the
Child Welfare League of America (1993) as "immense, complex, and
cumbersome" (p. 6). OToole and Mont joy (1984) point out that
Many implementation problems seem to derive
from widely shared characteristics of people and
their patterns of organization; changing the
particular individuals involved is unlikely to
elicit dramatic solutions, (p. 499)
Hence considering a change in organizational form and pattern
logically follows. As demonstrated in the following citation, as well as
other information presented, the current "patterns of organization" are
not achieving their purposes.
A wide assortment of social service agencies has
been organized to service children and youth at
risk; but the services often overlap, agencies are
compartmentalized, and children are incorrectly

referrecL-the "bewildering array" of agencies has
become part of a large, unwieldy bureaucracy
where the emphasis is on self-preservationThose
families least able to navigate their way through
the maze of requirements are left out.
(Guthrie & Guthrie, 1991, p. 17)
The facts and figures cited above provide a contextual frame
within which the following problem statement may be understood
Problem Statement
"Put bluntly, the result of fragmented systems is inefficient use
of resources, lost opportunities, and useless conflict" (Chisholm, 1989,
p. 2).
Problem Formulation
As pointed out by Kingdon (1984), "There is a difference between
a condition and a problemConditions become defined as problems when
we come to believe that we should do something about them" (p. 115).
The social conditions outlined above meet Kingdons definition of a
problem; they have consequences for both individuals and organizations.
At an organizational level, the combined
results of this problem-oriented fragmentation
are bureaucracy and administrative inefficiency.
For families_.the consequences are spelled out in
more personal terms in the downward spiral of
school failure, underemployment, inadequate
health care, delinquency, and substance abuse.
(Melaville & Blank, 1991, p. 8)

Academic and popular literature, the newspapers, and electronic media
all reflect the demand that something be done about the current social
conditions of the country.
None seem to doubt that a bureaucratized,
categorized, fragmented policy and program
world that lacks capacity for holistic approaches
to whole individuals and whole families with
complex problems and needs cannot expect to
succeed (Kahn & Kamerman, 1992, p. 7)
One response to the increased demands in the face of decreasing
resources widely recommended by academicians and practitioners alike
is an increase in collaboration among human service agencies.
According to Schrage (1990), "Collaboration is a purposive relationship.
At the very heart of collaboration is a desire or need to solve a
problem, create, or discover something within a set of constraints" (p.
36). The National Association of School Boards (1991), in its Best
Practices series, defines collaboration as involving "cooperative efforts by
parties with similar or overlapping objectives to reach goals that cannot
be achieved by either party acting alone" (p. 3). Five elements of a
collaboration are specified*
1) . Sharing information and ideas
2) . Setting a common goal
3) . Sharing responsibility for meeting the goal
4) . Working together to achieve the goals, though
often performing different tasks and goals
5) . Sharing resources
(p. 4)

It is further clarified,
Although collaboration requires both communication
and coordination, it goes beyond both.
Communication involves providing more complete
information, but does not require any interaction
among agencies. Coordination implies joint activity,
but does not require participants to share a common
goal. (p. 4)
A collaborative group, therefore, may be described as
a group of community leaders who have agreed
to be partners in addressing shared problems
Partners using a collaborative strategy establish
common goals and agree to use their personal
and institutional power to achieve them.
(Melaville & Blank, 1991, p. 15)
Hanf and OToole (1992) further elaborate that it is the
interactions chosen by individuals
that lead us to hypothesize the existence of a
collectivity. The resulting "implementation
structures" are analytical constructs that reflect
the way in which individuals have self-selected
themselves and engaged in social interaction
with respect to the policy problem under
investigation, (p. 175)
In human services, as noted, the interest in collaboration is often
related to shrinking resources, the complexity of the problems to be
addressed, and the turbulent environment of todays social world.
Schopler (1987) explains that functioning and outcomes of
interorganizational groups are of particular importance and interest in

the health and human services domains. "Organizations in this arena
are under pressure to integrate and improve services while lowering
costs" (p. 702).
These concepts will be explored further in the review of
literature. The present purpose is to document the call and need for
collaboration, prior to stating the underlying questions and research
Goldberg and Stewart (1996) explain that,
Historically, we have attempted to find solutions
to complex social issues by focusing on a particular
sector to provide the solutions. If we have learned
anything over the years, it is that the responsibility
for social ills does not lie with any particular
sector and that it takes the cooperation and collaboration
of all sectors to effect change, (p. 95)
They add that, "Current services are too categorical, deficit-
oriented, fragmented, and bounded by rules to help families grow and
nurture their children" (p. 95). The stakes are high; Jennings and
Krane (1994) assert that, "Success of the welfare reform effort might
well turn on the degree to which the diverse organizations and
programs integrate their efforts" (p. 341).
Agranoff (1991) describes a family whose multiple needs for
services required action from fifteen public and non-profit agencies and
adds, "The best minds in social policy and administration agree that
comprehensive responses are needed for these and other problems"
(p. 534). Multiple services needs have continued to grow over the

decade and Agranoffs call for comprehensive responses continues to be
in evidence. The Carnegie Corporation, in its report Years of Promise
(1996), states,
The general lack of comprehensive, community-based
supports for children and insufficient integration of
existing services, however, represents one of the steepest
barriers to children's learning and healthy development
Fragmentation remains an immense problem, hampering
service delivery, information sharing, and follow-up.
(p. 45)
Allen and Mor (1997) agree, and say of fragmentation,
Current systems provide not a protective blanket of
health and social services but rather patchwork quilts,
with the size and adequacy of each individual quilt
depending on the number and size of the patches for
which one is eligible, (p. 154)
Nor are the problems of fragmentation limited to the service
delivery level. Knitzer and Page (1996), in analyzing issues related to
early childhood care and education systems, found that,
Administrative fragmentation has become increasingly
problematic. [It] takes many forms: different
eligibility criteria for similar programs; cumbersome
payment, administrative and reporting procedures;
arbitrary funding limits and eligibility criteria; lack
of access-to support services_and barriers to pooling
funds, (p. 37)
Gage (1990) has stated that "Problem-solving in public
management requires intergovernmental solutions" (p. 147) and Melaville
and Blank (1991) assert that

Real progress toward large-scale comprehensive
service delivery is possible only when communities
move beyond cooperation to genuinely collaborative
ventures at both the service delivery and system
level, (p. 4)
Alternately expressed, "The challenge is not simply to divide up
responsibilities, but to reconceptualize" (Guthrie & Guthrie, 1991, p. 17).
Hodgkinson (1989) adds,
We simply have to get more mileage out of
the resources and organizations we now have.
Interagency collaboration is one logical solution
to a number of issues-At the center of all our
social agencies sits a client who must be housed,
transported, educated, fed and kept healthy. For
every agency, it is the same person, the same
client (p. 25)
Provan and Milward (1991) summarize in stating,
The idea is that since no single agency provides
the entire package of related services often needed
by clients, multiple services are best delivered
interorganizationally through a coordinated and
integrated network of organizations offering
components of the complete service, (p. 394)
Given the acute nature of the social problems with which
governmental and not for profit agencies have to deal, it is important
that vehicles such as interorganizational collaborations, operationalized
as implementation structures, which might provide at least a partial
solution, be examined and understood. It should be acknowledged that

collaboration alone cannot provide answers to all the problems cited
above and that there are issues it cannot solve in and of itself, such as
limited resources, the desire of individuals to collaborate in the first
place (National School Boards Association, 1991). However, the
importance of studying potential solutions, even though they be partial,
is evident
The literature review will demonstrate that a great deal of
research on interorganizational networks has been conducted in the
areas of structural composition, antecedents of collaboration, and
establishment of linkages: There is a growing amount on mobilization
of networks for action, characteristics of network managers, and
voluntary versus mandated networks. Less work has been done on
outcomes of collaborations and the variables that may impact those
outcomes. As Kenis and Schneider (1987) explain,
Network thinking is not new__The abundance
and variety in which network concepts occur in
contemporary social sciences, however, indicates
a new qualityNetwork thinking conveys its own
picture of the world decentralized, dispersed
control and intelligence, multiple actions units
coordinated through purposeful interactions of
individual actors, (pp. 25-6)
Reviews of research on collaborations/IONs cite the need for
more work on the impact of shared leadership, contextual variables, and
the impact of shared resources on outcomes (Agranoff, 1991; Institute

for Educational Leadership, Inc, 1986; Kagan, 1991; Lambright, 1997;
Mulford, 1984; OToole, 1997; Rogers & Whetten, 1982).
Operational Definitions
The following operational definitions wall be utilized in the study:
Human services agencies: those public and not-for-
profit organizations whose mandate and mission relate to
provision of at least one of the followring: social services,
mental health, public health, job training and related areas
of self-sufficiency, education, law enforcement, judicial
services, probation, parole, detention or commitment to
juvenile or adult prison.
Interorganizational networks (IONs)/interagency
collaborations/implementation structures: These
terms are used interchangeably. For purposes of
this study, they are understood to be voluntary (but
formalized) associations of human services agencies
characterized by shared resources, joint planning,
shared power and authority, a stated mission, and stated
goals and objectives.
Collaboration outcomes: will refer to the results of the
activities of the network.
Contextual variables: will refer to those factors of
politics, history, geography, culture, and demography
specific to each network studied.
Network leadership: will refer to the model of
leadership utilized in each network studied.

Network leaders those individuals who are members
of the steering committee or otherwise named leadership
group (which has been given responsibility for forwarding
the objectives of the network and has been empowered
to make decisions on behalf of the network).
Issues for Examination
In considering interorganizational networks in light of identified
gaps in current research, the following issues underlie the research
hypotheses and study design to be proposed: What are the goals and
what have been the outcomes of the networks studied? What style of
leadership is being utilized in each of the networks? What are the
leadership tasks of a multiorganizational structure? How are leadership
tasks shared (or are they) in an interorganizational setting? What
values are important in such a setting, what role do they play, and to
what extent are they/should they be shared among participants? And,
most importantly, what relationships among the above can be identified
and what might they mean for improving the effectiveness of ongoing
collaborations and increasing the opportunities for success for new
Research Hypotheses
Implementation structures have become a reality in todays
organizational world; they are in use in addressing complex social
problems, as in the case of human services. Because "leadership is
widely considered to be critical for success" (Coe, 1992, p. 22), an

understanding of the leadership function in the network setting
becomes essential in addressing how collaborations are developed and
maintained, as well as in how collaborative outcomes may be improved.
The hypotheses to be advanced will address development of a model
which describes and explains how leadership is conceptualized and
implemented in a collaborative network.
Shared Leadership in Implementation Structures
Shared, or distributed, leadership in networks is an
underresearched phenomenon (Kagan et aL, 1991), although the
literature concedes that it appears to exist in some fashion in
implementation structures (Agranoff, 1991; Kagan et aL, 1991; MandeU,
1990, Rossy & MandeU, 1992).
Hosking and Morley (1987) propose a model of shared leadership
consisting of participants, processes and contexts. They suggest that
leadership is the construction of social order(s) that further the groups
interest and values and that assist in dealing with the complexities
with which the group is faced. They did not specify an organizational
form in which this model holds true, stating that leadership phenomena
have been and should not be "divorced from the organizational processes
of which they are a part" (p. 89). This description of leadership,
however, would appear to be highly relevant in the context of
implementation structures (Le, a specific organizational form) in human
services. Hosking and Morley (1987) state that "significant leadership

contributions may come from a minority, including a minority of one;
equally, they may be expected and contributed by the majority" (p. 90).
The first hypothesis to be advanced will suggest a definition of
participants and the second will suggest that, in a multiorganizational
implementation structure, leadership contributions are expected from
and imparted by all members. This approach is further supported by
Wildavskys (1989) statement regarding the dislike of equities for
hierarchy; that concept then may be extended to support that of shared
leadership. To continue this line of thought, the notions of participants,
processes and contexts of multiorganizational collaborations will be
combined with other relevant leadership literature reviewed, leading to
the following hypotheses.
Hypothesis 1: In successful implementation structures
(those which achieve stated goals and objectives), the
participants will be the leaders of the respective individual
member organizations (Le., CEO or senior management
level) and,
participation will remain at that level on an on-going
The implementation structures to be studied are voluntary, non-
hierarchical and have mobilized to address complex issues in human
services delivery and policy development and implementation, evidencing
the problem-solving focus cited by Wright (1983). It is suggested that
this goal cannot be achieved without leadership processes implemented

on a shared basis and that some of these processes in an
implementation structure are the same as those defined in the
literature for leaders of single organizations, leading to
Hypothesis 2: In successful implementation structures (as
defined in Hypothesis I), the processes/tasks of leadership
will be accomplished on a shared basis by participant
members. Leadership tasks (as explicated by Gardner
[1990] and Rossy and Mandell [1992D essential to
mobilization of a network and to goal achievement include:
a) . Building, maintenance and regeneration of a shared
vision of network purpose and goals.
b) . Defining and affirming values to be used in mission
and goal definition.
c) . Raising the level of trust among members through use
of "innovating behaviors"* as defined by Rossy and
d) . Achieving workable unity among members and building
commitment to mission and goals.
e) . Renewal of purpose and values.
1 These authors define "innovating behaviors" as creating an open environment,
challenging assumptions, searching for alternative solutions, learning from others,
encouraging reasonable risks and managing failures, and providing constructive

The Importance of Shared Values
The significance of common core values and organizational
effectiveness is widely acknowledged in studying traditional
organizations. Shared organizational values have been termed a
"bedrock" without which "principled leadership becomes nearly
impossible" (Gardner, 1990, p. 113). Common core values provide "a
mechanism to create a foundation of attitudes and practices that will
lead to enhanced long-term success-." (Osborne, 1991, p. 28). Further,
"Leaders promote persuasive scripts that help others to interpret actions
and events, in relation to the core values of their social order"
(Hosking & Morley, 1987, p. 97). Therefore, the area of core values, as
defined by Zander (1985) in the review of literature, is being discussed
as a separate issue to emphasize its importance, even though
"affirming values" appeared as part of Hypothesis 2 as well
In a traditional organization with centralized leadership, it falls
to that leader to establish core values and ensure that they direct the
behavior of followers (Zander, 1985). There is no reason to suppose that
values play a less important role in implementation structures than in
more traditional organizational forms. However, it has already been
hypothesized that centralized leadership is not a feature of successful
implementation structures. It has also been noted earlier that one of
the motives for establishing collaborative structures is to bring together
persons of diverse profession, training and background because the
particular problem to be addressed is too complex for any one

organizational entity (or discipline/profession) to resolve alone. What,
then, constitutes the base core value(s) of the implementation structure?
This question leads to
Hypothesis 3: The core, shared value of successful
implementation structures (that is, those that achieve
stated goals) is commitment to the collaborative process
(including shared information, authority, and resources) in
the service of identified and agreed upon mission, goals,
and objectives.
Corollary 1: Commitment to the collaborative process and
disparate values related to the purposes of a network
members own "home" organization can both be maintained
in the service of network goals.
The hypothesized relationship among the variables is correlational; they
are too complex and, as yet, insufficiently investigated, to propose
causal, or even predictive, hypotheses
Relevance of the Study
This issue has been addressed to a degree throughout much of
the discussion thus far. Schrage (1990) asserts, "Why collaboration now?
Not only because we dont really have a choice but because its the
best choice weve got" (p. 7). On the other hand, Kagan et aL (1991)
offer a reminder that research on collaborations is essential not only

because it might be "the" or "an" answer to the desperate social
problems cited earlier, but because it might not be.
Skeptics suggest that collaboration is like the
emperors new clothes, an illusory construct,
naively embraced by misguided true believers.
Such uncertainty, coupled with the rising
popularity of collaboration, demands that we take
honest stock of what collaborations are achieving
and that we better understand the conditions that
promote and inhibit collaborative efforts.
(p. 2)
Current work on networked systems continues to urge ongoing research
(OToole, 1997).

To understand interorganizational networks and the current state
of the literature regarding them and to provide further justification
for studying these implementation structures work across several
areas of public administration must be considered. Intergovernmental
management, a branch of intergovernmental relations, is the paradigm
in which this research resides and will be discussed first. As
leadership factors in implementation structures are the focus of the
study, relevant material from the leadership literature is reviewed,
followed by a section on applicable literature from the policy sciences.
Finally, the development of thought on networks is examined.
Intergovernmental Management
The overall paradigm under which this study falls is that of
intergovernmental management (IGM), a specific subset of
intergovernmental relations. IGM as a concept was first broadly used
in the 1970s and Wright (1983) identified the following factors which
forwarded its development and continued significance for public

1. An awareness emerged that policy/program
implementation encompassed more than merely financial
considerations; it covered management in its broadest
2. An immense increase occurred in the number of
recipients of federal assistance the intergovernmental
"game" and "plot" expanded enormously in terms of
governmental units, agencies, and associations.
3. A significant shift appeared in the character of federal
assistance the introduction of block grants, general
revenue sharing, and formula entitlement programs.
(p. 282)
It could be argued that these same factors have led to the
increased demand for, first coordination, and now collaboration among
human service providing agencies. The conceptual expansion from
intergovernmental relations to intergovernmental management, and its
application to implementation structures, underlies this study.
To explain further, Mandell (1989) summarizes IGM as going
beyond merely understanding relationships among
actors in the intergovernmental arena. Rather, it
emphasizes the problems of managing these
relations-whereas intergovernmental relations identifies
who the actors are in the system and how they relate,
intergovernmental management is the tool needed to
understand how and why these levels interrelate the
way they do and how we can cope in this
allows a perspective that looks at networks and
communications as positive ways to make things work in
the intergovernmental system, (p. 149)

Agranoff (1990) sees intergovernmental management as involving
"the daily transactional relationships or the working out of relationships
between governments" (p. 58). Again, the emphasis on ongoing
management needed to advance intergovernmental implementation of
policy and program is noted.
Wright (1983) described three qualities of intergovernmental
management First there is a problem-solving focus. The review of
network literature will show how human services implementation
structures evidence this quality in fact, it is a primary reason for
their existence. Second, IGM is a "useful and coping approach (p. 431).
That is, it provides a framework for understanding exchanges that take
place among sectors, it provides "guidance on how to cope with the
system" (pi 431), and it is a conceptual tool for action, all of which are
needed for complex structures, such as multiagency collaborations, to
achieve results. Thirdly, Wright states, IGM emphasizes
intergovernmental contacts and communication networks. Relationships
that "span organizational and governmental boundaries" (p. 442) are
foundational to human services implementation structures. In later
work, Wright noted "the utility of IGM for both analytical and applied
purposes" (1990, p. 155) and cited Schechters statement that IGM is
"both the natural extension and resuscitating element of the twin
commitment to federalism and managerialism in a time of scarcity -
both of resources and leadership" (p. 156).

Beyond the general qualities noted above, Agranoff and Lindsay
identified specific elements of IGM in 1983. These were: jurisdictional
respect the ability to recognize and heed both autonomy of involved
parties and the "common ground on which the specific problem at hand
can be resolved" (p. 230); the prevailing presence of political issues,
which they suggest be dealt with openly, in a solution-focused manner,
technical issues whose presence results in the utility of adaptive over
rational planning; and, task issues necessitating joint agreement on
the tasks to be pursued, decision-making at multiple levels, and
recognition and avoidance of the most sensitive turf issues if possible.
The continued presence and relevance of these elements in human
services implementation structures will emerge in the analysis of case
study data in later chapters.
Why does intergovernmental management continue to be
prominent in the public administration literature and a paradigm to be
augmented by continuing research? Anton (1988) states that
changes in relationships among governments
result from new national policies to address
perceived substantive problems rather than policies
that attempt to directly change relationships" (p. 1)
(that is, the problem-solving focus of IGM). And Hanf and OToole
(1992) argue that

Modern government is characterized by decision
systems in which territorial and functional
differentiation disaggregate effective problem-solving
capacity into a collection of subsytems of actors with
specialized tasks and limited competence and resources.
(p. 166)
As to the importance of IGM and implementation structures at
the local and regional levels, Gage (1988) contends that "Spontaneity
and innovativeness of local community life may be at the core of our
systems durability and its long term capacity to survive" (p. 17).
It is the goal of this research to augment this paradigm
through increased understanding of how implementation structures in
human services function and of how leadership is implemented in a
multiorganizational structure.

Selected Aspects of the Leadership Literature
The body of literature on leadership is truly vast and definitions
of the concept abound. According to Yukl (1989),
Leadership has been defined in terms of individual
traits, behavior, influence over other people, interaction
patterns, role relationships, occupation of an
administrative position, and perception by others
regarding legitimacy of influence, (p. 2)
The great majority of the literature is focused upon leadership in
single, traditional organizations and on leader/follower interactions.
The dominant theme in leadership research
has been an attempt to determine the characteristics
and behaviors most closely associated with effective
leadership_a second theme has been played,
in seeming counterpoint with the first; a concern
with group members beliefs about and perceptions
of leadership.
(Pavitt & Sackaroff, 1990, p. 374)
How then, may this literature be applied to multiorganizational
settings, implementation structures in which many members also lead
their own organizations and are equals in the usually non-hierarchical
network with its own goals and objectives?
According to Rossy and Mandell (1992),
Models of leadership which are useful in traditional
organizational settings often prove inadequate in
implementation structures. This does not mean,
however, that they cannot be useful in developing
leadership within implementation structures. There
is in fact much to be learned from these models:

The key is to apply them within the unique constraints
that characterize network program implementation.
(p. 2)
This was foreshadowed by Wildavsky (1989) who, after
decrying the "dead end" (p. 87) at which an overabundance of
leadership research has arrived, asserts, "Specifying forms of social
organization and relating each to different types of leadership would
reduce, instead of expand, the realm of relevant leadership behavior"
(p. 87). Gardner (1990) concurs that "the setting does much to
determine the kinds of leaders that emerge and how they play their
roles" (p. 6) and McElroy and Shrader (1986) cite "the increasing concern
for the social context of leader behavior-leadership cannot be considered
in a vacuum, but is deeply embedded in a social setting'1 (p. 359). The
intertwined nature of leadership and context is confirmed by Heclo
(1977), who states
Events also depend on how people choose to act;
political appointees can choose how to use their
personal attributes and how to respond to given
contexts_How political leaders choose to act -
their statecraft affects what gets done. (p. 155)
There are certain issues and concepts which appear to be most
strongly related to this study of leadership in implementation
structures. They are: the creation and communication of vision; the
definition of leadership tasks; the role of values, both shared and
disparate; the way power is defined and used in the network; and the

implementation of a shared leadership model. These will be discussed
in turn.
Creation and Communication of Vision
Multiple scholars and researchers have addressed the importance
of creating, communicating, and maintaining a vision of the desired end
state (Bennis & Nanus, 1985; Buell, 1992; Coe, 1992; Gardner, 1990;
Kouzes & Posner, 1987; Rossy & Mandell, 1992; Sergiovanni, 1987;
Sheive & Schoenheit, 1987; Tichy & Devanna, 1986). And Wildavsky
(1989) is critical of that leadership literature which "asks Who shall
lead? as if it didnt matter where" (p. 98).
Simply stated, "leadership begins with vision" (Coe, 1992, p. 6).
Why so?
Vision is the heart of leadership because vision
transcends political interests..Vision designs new
synergies. Vision challenges everyday, taken-for-
granted assumptions by offering new directions
and articulating what people feel but lack words to
say. (Terry, 1993, p. 38)
The definition of vision provided by Sheive and Schoenheit (1992)
is representative: "A vision is a blueprint of a desired state. It is an
image of a preferred condition that we work to achieve in the future"
(p. 94). More specifically, vision
focuses on a better future; encourages hopes
and dreams; appeals to common values; states
positive outcomes; emphasizes the strength of a

unified group; uses word pictures-communicates
enthusiasm and kindles excitement.
(Bryson & Crosby, 1992, p. 309)
And Tichy and Devanna (1986) add that "A successful vision has
a tension thats the result of its having been created from both
intuition (right-brain thinking) and logical analysis (left-brain thinking)"
(p. 126), thus providing both a conceptual framework, or road map, and
emotional appeal (p. 130).
However, Kouzes and Posner (1987) remind readers that
"Leadership vision is necessary but insufficient for an organization to
move forward with purpose toward a common destination. As
important, if not more so, is the ability to communicate that vision"
(p. 100). If vision is to inspire organizational (single or network)
direction and action, it must be shared by all parties involved. Bennis
(1992), speaking with regard to the private sector, insisted that shared
vision is a critical element of any joint venture; there is no reason to
believe this assertion does not apply to the public and non-profit sectors
as well Isaacson and Bam bury (1992), citing the work of Senge (1990),
define shared vision thusly
True shared vision is never imposed. It emerges
from people who truly care about one another and
their work, who possess a strong sense of
personal vision, and who see the collective vision
as one that can encompass the personal visions
of alL (p. 44)

Tichy and Devanna (1986) agree with the importance of shared
vision, but call communication of vision in complex organizations (as
multiorganizational implementation structures surely qualify) a
"Herculean challenge" (p. 150). Sooklal (1991) contends that "vision gains
clarity and social support through the leadership process" (p. 834) (as
opposed to conjecture, which is clarifed and supported through scientific
method). But, "As constituent diversity increases, visions must be
framed to appeal to a wider range of participants and organizations
must cope with more varied interests and concerns" (Brown, 1991,
p. 823), hence the challenge noted above.
Because collaborative structures are a newer organizational form,
creating and maintaining a vision of the structure and its purposes and
goals, along with generating commitment from members, is clearly a
relevant aspect of leadership in that setting.
The Definition of Leadership Tasks
The tasks set forth in the literature as the responsibility of
leaders again generally refer to traditional organizations. Nonetheless,
given the goals and objectives of collaborations, for example in human
services, it is again reasonable to assume their applicability to networks,
although perhaps with different facilitating and constraining factors
and different implementation styles. Gardners (1990) list is reasonably
representative of what is found in the literature. He suggests nine
tasks for leaders: envisioning goals, affirming values, regeneration of


values, motivating, managing, achieving workable unity, raising the
level of trust, serving as a symbol, explaining, representing the group,
and renewing. These are all encompassed in his charge: "The task of
leaders is to have a sense of where the whole system should be going
and to institutionalize the problem-solving that will get it there"
(p. 119).
Buell (1992) refers to "choreographing values" (p. 91) and Brown
and Hosking (1986) add the linking of participants capacities with the
demands of the organizational tasks.
Skilled actors help those with whom they are
interdependent to deal with the problems of
information, search, interpretation, influence,
and choice. In other words, they make their own
task no harder than it need be by making things
easier for others, (p. 71)
Korakunda and Hunt (1991) state that it is the leaders task to
be facilitative and catalytic and Bryson and Crosby (1992) add that
"issue framing is a prime leadership responsibility" (p. 66).
Writing about human services collaborations in pursuit of service
integration, Melaville and Blank (1993) state "Systems change demands
leaders who can hold fast to a collaboratives vision, battle
bureaucracies, share power, and provide consistent direction" (p. 79).
Rossy and Mandell (1992) suggest leadership tasks specifically for
implementation structures. The first is building commitment through
six behaviors: consensus building, focusing on priorities, leading by

example, recognizing contributions, managing disrespect, and removing
obstacles. They also perform what these authors term "innovating
behaviors [those that] create opportunities and the desire to exceed
initial expectations and goals" (p. 8): creating an open environment,
challenging assumptions, searching for alternative solutions, learning
from others, encouraging reasonable risks and managing failures and
providing constructive feedback.
Chrislip and Larson (1994) also speak to leadership in
collaboratives specifically, characterizing it as a process role whose tasks
incorporate creating shared ownership, inspiring confidence, and
maintaining a problem-solving focus. Badarraco and Ellsworth (1991)
state that "a central task of leadership is to prevent people from
becoming isolated and oriented solely toward their own needs, and
instead to open them to broader interests than their own" (p. 47). Were
one to substitute "agencies" for "people," this statement would apply to
leadership tasks in an implementation structure, as welL
However, it must be remembered that, however many tasks there
are related to process, facilitation, influence, and the like, there are
concrete leadership tasks as welL Sink (1991) suggests policy
entrepreneurship as a task for the leader.
He or she is often known for political connections
and negotiating skills-and preparedness to take
advantage of a policy window or opportunity. An
ability at linkage characterizes the actor, building
on a skill of linking solutions with problems and

agencies with each other. Such ability may be
applied to recruitment, networking, and
maintenance of the coalition, (p. 1182)
Brown and Hosking (1986) perceive the following to be essential:
skill in information search and interpretation, maximizing the capacities
of participants in concert with task demands, recognition of dilemmas
that emerge to threaten goal achievement, and possession of ability to
protect the interests and values at stake.
The -Role_of Values
"Shared values are the bedrock on which leaders build the edifice
of group achievement" asserts John Gardner (1990, p. xii), going on to
say later, "If leaders cannot find in their constituencies any base of
shared values, principled leadership becomes nearly impossible" (p. xii).
Alvin Zander (1985) defines values as follows:
A value is concerned with appropriate beliefs, not
with ends to be achieved. A value, for a group
member, is that individuals concept of an
ideal kind of behavior. The member uses this
notion to assess the goodness or badness, the
rightness or wrongness, or actions by participants
in the unit. A value describes things that should be
done or should be avoided without fail, not merely
behavior that is attractive or disliked. It delineates
moral efforts from immoral ones and is an
absolute good under all circumstances, (p. 94)
Fosler (1992) contends that values additionally "may encompass
clearly articulated legal principles as well as shared beliefs and more

subtle cultural traditions" (1992, p. 3). In implementation structures
where member organizations are diverse in statutory and regulatory
mandates, legal principles may well influence what is included in the
common ground of shared values and what conflicting legal rulings
dictate must remain outside of the collective effort. However, making
the effort to identify the common ground is essential if Gardner (1990)
is correct that "a society [collaboration?] in which pluralism is not
undergirded by some shared values and held together by some measure
of trust cannot survive. Pluralism that reflects no commitments
whatever to the common good is pluralism gone berserk" (p. 97).
Kouzes and Posner (1987), in a study that included over 2300
managers, investigated the relationship between organizational and
individual values. They were able to conclude that shared values foster
a sense of personal efficacy, promote high levels of company loyalty,
facilitate attainment of consensus with regard to organizational goals,
encourage ethical behavior, and reduce levels of job stress and tension.
Osborne (1991) emphasizes the importance of organizational
values being made explicit. His work focuses upon the concept of core
value statements, which he defines as "a mechanism to create a
foundation of attitudes and practices that will lead to the enhanced
long-term success of the firm" (p. 28). Researching publicly held firms,
he found three ways that they furthered their goals through core
value statements: improved crisis management, improved management
of strategic change, and improved growth management. At least the

first two have relevance for implementation structures in human
Also relevant for implementation structures is the caveat
advanced by Pratt and Kleiner (1989): "Perhaps the greatest risk
involved with a corporate value system is that of managerial hypocrisy
- the lip service disaster" (p. 11). The "lip service disaster" practiced by
a member of an implementation structure could be equally disastrous;
the acknowledgement and management of disparate values is a critical
issue for multiorganizational networks. As Goodsell (1989) points out,
multiple value orientations are often totally legitimate and to honor
one legitimate perspective over another does a disservice to both. He
states, "Administrators should attempt to balance competing value
orientations, instead of vehemently pursuing one at the expense of
another" (p. 576). After all, as Schrage (1990) indicates, "Collaborators
argue precisely because they come to the task with different
perspectives and backgrounds which is exactly why they are
collaborating (p. 159). Finally, the role of values for leaders, whether
in a single organization or an implementation structure, can include
being a force for innovation. "Leaders with specific values, principles,
and ideals that are not mirrored in the current state of affairs are
likely to be stimulated into transformational efforts" (Bass, 1985, p. 182).
How important are values to leadership? "Ultimately we judge
our leaders in a framework of values" (Kouzes & Posner, 1987, p. 67)

and, it may be suggested, multiple leaders in an implementation
structure may thusly judge each other.
Power in Networks
Power, as leadership, is a topic with a vast literature in its own
right Some sources of power in networks will emerge in subsequent
sections (eg, possession of resources other members are lacking).
Gardner (1990) sees power as interwoven with leadership and defines it
as "the capacity to ensure the outcomes one wishes and to prevent
those one does not wish" (p. 55). Meier (1989) observes that "The role
of leadership in organizations is to acquire power for the organization"
(p. 269), which translates, in his view, into the acquisition of resources
and autonomy. He suggests four strategic choices for leaders;
membership inducements, policy objectives, implementational strategy
and coalitional strategy (pp. 469-70). This literature on power
specifically interfaces with the idea of implementation structures in
several ways, explicated below.
Bass (1990) explains that
There are two opposing arguments about how power
should be distributed in organizations. In the
structuralist view, it is inherent in the nature of
organizations that power is distributed unevenly, and
as organizations develop, for power to accrue in
varying amounts among the units and supervisors of the
organizations_The opposing behavioralist view
suggests that the equalization of power is a basic tenet
of modern organization development (Bennis, 1965)

and the human potential movement (Maslow, 1965). It
has its roots in those aspects of democratic and
egalitarian values that stress the importance of informal
trusting relationships rather than a formal structure in
which relationships depend on the authority of the
position and the requirements of the role. (p. 264)
The latter view is supported by the work of Sergiovanni (1987),
who says that
Highly successful leaders practice the principle
of power investment: They distribute power among
others in an effort to get more power in return.
But_they know it is not power over people and
events that counts, but rather, power over
accomplishments and over the achievement of
organizational purposes_successful leaders are concerned
with the concept of power to. (p. 121)
Melaville and Blank (1993) concur; relating the issue specificially
to network leadership, they state: "Collaborative leadership requires
developing a new notion of power and learning that the more power
and control we share, the more we have to use" (p. 79). Gray (1989)
supports the importance of shared power for collaborations, stating that
such groups "essentially share the power to define a problem and
initiate action to solve it" (p. 112).
Also relevant to implementation structures is Gardners (1990)
notion of non-jurisdictional power. "The new leaders, dealing endlessly
and on many fronts, with groups over whom they have no jurisdiction,
find often that the power of their institutional position simply is not
decisive" (p. 119). He suggests development of other forms of power to

be effective those of ideas, knowledge of how systems work, public
opinion, and the media. Wright (1990) agrees that authority
relationships in an intergovernmental management context are
"preponderantly non-hierarchical" (p. 159) "There may be varying
dependency-autonomy power patterns among specific entities in a
network, but across the complete network there is no prime, single or
central source of guidance" (p. 159).
However, it is also important to remember that implementation
structures do not come into being from a void, but are a newer
organizational form built from existing organizations. Hence the
relevance of McCann and Grays (1986) caveat
Collaboration-can threaten the existing distribution
of power among the organizations and groups
involved in a human service delivery system.
Managing the distribution and use of power in a
human service system is therefore essential for
effective collaboration (p. 58)
Other troublesome issues that may emerge during the journey
toward the shared power ideal include behind the scenes decision-
making by an elite few who then ratify in public, lack of sufficient
power by would-be convenors to generate sufficient participation and/or
commitment, or lack of power to assure implementation of the
collaborations decisions (Gray & Hay, 1986).
A last power/implementation structure issue to be discussed that
is relevant to shared power and leadership in such structures is that of
network centrality. Power theories suggest that centrality of an

organization in a network enhances its power within that network
(Kanter, 1977; Boje & Whetten, 1981; Hallinger & Richardson, 1988).
That is, the more interconnectedness with other parts of the system an
organization has, the more power accrues to it. Management of
centrality in the service of shared power for maximum collaborative
effectiveness may become a network leadership task to the extent that
this theory holds true for a particular network.
In summary, power and leadership are inextricably linked. How
this plays out for implementation structures is judiciously stated by
Mandell (1990): "-power in an interagency network must be seen as
diverse, not decentralized, because each organization in the network
potentially has equal power" (p. 32).
Shared Leadership in Implementation Structures
Authors such as Kagan et aL (1991) have urged study of shared
leadership as implemented in an interorganizational field, while the
related concept of shared power and authority has been postulated as
critical to successful IONs by others (Agranoff, 1991; Mandell, 1990;
Whetten, 1981). As urged by Van Seters and Field (1990), "Leaders
must take on a more collective view of leadership" (p. 40). In 1990,
Barry conducted research on distributed leadership within Self
Managing Teams. The following remarks extrapolate to implementation
structures: "A significant question has gone unanswered: How should
leadership be exercisecLin settings where differences in formal

authority either do not exist or are downplayed?' (p. 32). His conclusion
is that such arrangements require even more leadership than
traditional settings.
Yet, "Shared governance is an elusive concept to grasp and put
into action" (Allen & Glickman, 1992, p. 80). Nor is it easy to
implement. "Hierarchies support leadership; equities suspect it The
hierarchy gives its leaders an extra push; an equity holds them
backMutual persuasion is what equities approve of (Wildavsky, 1989,
p. 105, 108). Nonetheless, "The status of managers in hierarchies will be
less important than the ability to influence people and programs based
on the strength of the vision and the power of the ideas" (Ryan, 1990,
p. 601).
Why the need to move from leadership centered in one
individual to a model of shared leadership?
In order to pursue long-range goals over time,
multi-sector collaborations must gradually cease
to depend greatly upon individual leaders and
must broaden the capacity and share legitimacy
among organizations and institutions for
assuming leadership and taking action to
solve problems.
(Institute for Educational Leadership, 1986, p. xiii)
In implementation structures, "Leaders may be formally
selected or may be (and often are) any members of the network;
different members may be leaders at different times for different
aspects (Coe, 1992, p. 8). In other words, "the distinction between

leaders and followers is not necessarily clear rather, at any given
time, any member may be serving the leadership function" (Coe, 1992,
p. 8). For, while
Most leadership theories adopt a person-centered
approach-these theories tend to ignore leadership
dynamics within a group context where the
development of the group almost always requires
frequent shifts in leadership behavior.
(Barry, 1990, pp. 32-33)
The social psychology approach to shared leadership, as
explicated by Hosking and Morley (1987), offers a model which appears
to be useful in considering leadership in IONs. They see leadership as
construction of "social orders that foster the values and interests of the
group to which group members belong in order to help group members
deal with complexities" (p. 87) and describe a three pronged model of
participants, processes and contexts.
Participants are leaders when they contribute to leadership
processes. Hosking and Morley (1987) state: "Our conceptualization
recognizes that significant leadership contributions may come from a
minority, including a minority of one; equally, they may be expected
and contributed by the majority" (p. 90). Processes in the model are
those means by which specific actions are perceived to contribute to
social order (therefore to leadership actions). Finally, the importance of
context (equally emphasized in ION and policy sciences literature) is
stressed, with the admonition that leadership must not be "abstracted
from the organizational processes of which it is a part" (p. 92). Brown

and Hosking (1986), in earlier work on this model, add that once
adopted by the group, a shared leadership model creates norms that
result in extreme resistance to any minority which attempts to
dominate such processes" (p. 76).
Another model representing a shift from single organization,
person-centered leadership models is that of Probst (1987), who sees
leadership as the management of social systems, be they
intraorganizational teams or larger social systems. His view is that,
while leadership fundamentally means leading people, seeing leaders or
followers as isolated beings is a reductionists approach. His systems
approach posits that leaders act, not only from personal goals, but in
light of maintaining viable systems. Therefore, leadership can only by
understood in light of the entire system and, not as an individual
actor/actions, but as attributes of actors in a complex network system.
Although currently underresearched and not thoroughly
understood, the concept of shared leadership is clearly emerging in the
literature. Van Seters and Field (1990) offer what might be a
description of this model when fully implemented:
Leadership rests not only on the shoulders of one
individual, but also on all who share the mission
and vision. In that sense, leadership becomes a
state of consciousness rather than a personality
trait or set of skills, (p. 38)

Gardner (1990) summarizes thusly: "The taking of responsibility
is at the heart of leadership. To the extent that leadership tasks are
shared, responsibility is shared" (p. 152).

The Policy Sciences Literature
The policy sciences are ".problem-oriented and contextual in
nature, multidisciplinary in approach, and explicitly normative in
perspective" (deLeon, 1988, p. 7). Additionally,
A major objective of policy analysis is to develop
recommendations regarding the ways in which
rules and institutions can be reformed to improve
the effectiveness of collective problem solving.
Such recommendations must be based on
systematic research into the ways in which
different structural arrangements affect patterns of
individual and collective choice.
(Hanf & OToole, 1992, p. 167)
Therefore, consideration of this academic context provides further
grounding and focus for a study of collaborative networks.
Public policy has been defined as concisely as ".whatever
governments choose to do or not to do" (Dye, 1972, p. 1) and more
expansively as "a generic shorthand for policies, plans, programs,
projects, budgets, and procedures that is, for all the concepts and
activities that are used to resolve problems" (Bryson & Crosby, 1992,
p. 57). Areas of this literature particularly relevant to a study of
collaborative networks include: the policy environment, particularly its
current complexity; issue and policy networks; and implementation
theory. Selected aspects of each, related to this research, will be

Fragmentation and Complexity in the Policy Environment
Complexity is now often treated as a given -
In academic and practical analyses-the basic
question of complexity has assumed a prominent
role-the complexity of policy problems is both
reflected in, and amplified by, the complexity and
polycentrism of the institutions through which
policy must be addressedOne of the most
significant factors involvecLis the inter-
organizational nature of the politico-administrative
world (Hanf & OToole, 1992, p. 165)
The admonition of Wittrock and deLeon in 1986 that
we should operate under the assumption that
alterations in the policy environment are to be
expected and, to a large extent, unpredictable,
or at least not particularly susceptible to confident
foresight (p. 55)
has proven to be a point well taken. Heclo (1977), as well, refers to the
"essential uncontrollability" (p. 166) of the environment This has
impacted both politicians and bureaucrats:
Increased complexity_produces_rapid and
unpredictable change with bewildering, hard-to-define
outcomes. This uncertainty and lack of clarity
increases risk for public leaders and managers,
because they are expected to make good decisions
and take effective actions in situations they do not
fully understand or controL
(Bryson & Crosby, 1992, p. 17)
As Chisholm (1989) quotes Douglas Yates,
Policy making takes place in a political and
administrative system that is fragmented to the

point of chaos. Too often the policies that result
are complex, contradictory, and ineffective.
Problems surpass the ability of any one agency or
governmental entity to solve, (p. 3)
Kingdon (1984) agrees, and also finds such an assessment highly
problematic, stating "The first consequence of system fragmentation is
policy fragmentation" (p. 125). This fragmentation results in policies
developed in isolation, hence producing unintended consequences for
those affected by, but not considered in, policy development It also
results in lack of common paradigms and instability of agendas for
action in a policy area It can be seen, therefore, that the
fragmentation and problems of systems in delivering services, as well as
the need for multiagency action cited earlier, are mirrored in the policy
environment Not only is the external environment complex, but so
also is the internal environment of implementation structures. Kagan
et al (1991) note that collaborations may be concerned only with the
delivery of services, only with policy and systemic impact and change,
or with both. Although members/leaders in human services
implementation structures are most often bureaucrats, as Meier (1989)
notes, "Although legislative bodies set the basic policy objectives of
governmental organizations, bureau leadership can still affect these
objectives" (p. 272). OToole (1989) concurs, stating that
The politically successful linking of policy problems
to appropriate alternatives, therefore, cannot be
assumed; it requires the participation of skillful
policy entrepreneurs, who are often administrators, (p.228)

And Heclo (1977) refers to a mixture of policy and
administration as "inevitable" (p. 240) at the higher levels.
Hence the goal of human services collaborations to affect policy
and systemic change, the potentiality of which is further validated by
Fosler (1992):
It is important to stress here that new institutional
approaches are needed not just to carry out new
policies, but to formulate them in the first
place_The organization for modern state policy
needs to be redesigned to cope with a very
different set of challenges, (p. 9)
Issue and Policy Networks
As Gage (1984) asserts, "Networks have an acknowledged value
for political problem-solving" (p. 136). One of the first applications of a
network concept to policy-making was the issue networks of Heclo
(1977). Issue networks were defined as "a shared-knowledge group
having to do with some aspect (or, as defined by the network, some
problem) of public policy" (p. 418) and replaced, according to Heclo, the
older notion of the iron triangle, comprised of a particular government
bureau, the applicable interest group, and congressional backers. Issue
networks addressed the "need to work together across agency lines on
policy problems that have become increasingly interdependent" (p. 236).
However, as Stillman (1984) notes, issue networks (as defined
by Heclo) also have an agenda of keeping attention focused on a

narrow issue and fomenting dissent to keep the issue alive rather than
seeking broad mandates for consensus. "They survive by talking,
debating, and arguing the alternatives, and not by finding common
grounds for agreement and getting down to make things happen"
(p. 415).
The policy making (or influencing) dimension of
multiorganizational collaborative structures is more accurately reflected
in the concept of the policy network as defined by Marin and Mayntz
Policy networks do not refer any longer to
networking* of individual personalities, to group
collusion, to the interlocking of cliques, elites,
party of class factions, as in older traditions,
but to the collective action of organized, corporate
actors, and consequently to interorganizational
relations in public policy making." (p. 14)
They further state that "Policy networks are explicitly defined
not only by their structure as interorganizational arrangements, but
also by their function "the formulation and implementation of policy"
(p. 16).
The policy network fits conceptually with the IGM paradigm
introduced earlier the hows and whys of interrelationships on multiple
levels and mechanisms to achieve effectiveness in complex systems.
This is seen in the explanation provided by Marin and Mayntz (1991).
The policy process is not completely and exclusively
structured by formal institutional arrangements.

It also emphasizes that the relationship among those
who de facto participate in the process is not hierarchical
The actors making up a policy network are
interdependent, but by and large formally autonomous.
This is not to say that policy is never developed within
hierarchical structures; the policy network concept simply
calls attention to the fact that the participants in a
collective decision process are often linked laterally (or
horizontally) rather than vertically, (p. 15)
The relationship of policy networks and IGM is also seen as
Kenis and Schneider (1991) describe these networks.
The coordination of these action units is no longer
the result of central steering7 or some kind of
prestabilized harmony but emerges through the
purposeful interactions of individual actors, who
themselves are enabled for parallel action by exchanging
information and other relevant resources, (pp. 25-26)
And Hanf and Scharpf (1978) found that the interdependence
which drives the formation and practice of many implementation
structures exists in intraorganizational policy making as welL They
The ability of individual decision units to
achieve their own objectives will depend not
only on their own choices and actions but also
on those of others; actions at any one level of
decision making will be influenced by the
relationships that exist between levels as well as
across functional boundaries, (p. 2)
Scharpf (1978) found the influence of intraorganizational,
intergovernmental structures upon policy making to be significant,
stating "it is unlikely, if not impossible, that public policy of any

significance could result from the choice process of any single unified
actor" (p. 347).
Finally, understanding of the relevance of these concepts to
understanding of leadership in human services implementation
structures is enhanced by Gages (1990) developmental model, in which
issue networks are seen as an arena for the formative stage of
potential activities. Policy networks are tighter and constitute the
policy formulation stage. What he refers to as implementation
networks, herein denoted as implementation structures, are concerned
with ".accomplishing a policy goal, a collective good, quite possibly for
different reasons" (p. 131). These structures, in turn, have substructures
for "policy making, planning and intelligence, resource provision, service
provision and evaluation-" (p. 131) all of which reemphasizes the
complexity of a given network, its environment, and their interaction.
However, as Heclo (1977) stresses: "The difficulty of serious
reform is only a symptom of its importance" (p. 264).
Implementation Theory and Networks
As with many other areas of literature cited, implementation
theory has a substantial literature devoted to it. There are specific
aspects particularly relevant to this study in that
Analysts who trace the relationship between
organizational structures and implementation-
have identified many connections between
administrative structures and processes and

implementation effectiveness.
(Ethridge & Percy, 1993, p. 343)
Even more directly related to this research, some studies have
confirmed that
Complicated, multiunit structures may encourage
success during implementation__Also, sometimes
programs need many kinds of specialties and
perspectives to be brought to bear during
(Goggin, Bowman, Lester, & OToole, 1990, p. 122)
Linder and Peters (1987) ascribe one type of policy failure to
failures of implementation. This is important to consideration of
networks as an organizational form to the extent that inappropriate
structure is the cause or one cause of implementation failure. For
example, in considering the evaluation of Wisconsins Learnfare
program, Ethridge and Percy (1993) found that there was a severe
underestimation of the interorganizational relations that would be
needed to implement the policy successfully. As noted by Calista (1986),
"the clarity of a policy intention depends upon the appropriateness of
the organizational form to implement it" (p. 263).
As complex as multiorganizational implementation structures are,
in a complex environment with multiple interdependencies (e.g., the
human services arena), such structures may well be the most
appropriate for implementing policies designed to address the problems
facing involved agencies. As OToole (1986) states,

Implementation usually requires action from
organizations_Multiorganizational implementation
is the rule in dense policy spaces, for complicated,
often cross-cutting, public problems; and in
political systems where power, and often
authority, are shared among several units, (p. 182)
Later in the same work, he cites research that
concludes that under certain conditions,
complexity may actually ease implementation
by making multi-actor compliance more attractive.
(p- 201)
Chisholm (1989) provides a compelling statement to summarize
the relationship between public policy and the need cited in Chapter
One for new organizational forms (interagency collaborations):
The need for coordination is a function of the
interdependence of the parts of an organizational
system: existing formal coordinative arrangements
are unable to manage interdependencies
effectively. In the face of this inadequacy,
coordination fails to occur, and irrational chaotic
public policy results, (p. 3)
Or, as concisely stated by Hanf and OToole (1992): "The inter-
organizational nature of the policy world is an endemic feature of the
modern era" (p. 167).

The Network Literature
The Structural Approach
The initial work on IONs is grounded in the structural
functional orientation of sociology and anthropology (Rogers & Whetten,
1982). Much of this early work focused on the structure of the
relationships and coordinative efforts. As defined by Zeitz (1980),
"Structure refers to that analytically distinguishable pattern that may
be perceived among a group of actors" (p. 77). The major thrust of this
work has been effectively summarized by Mulford (1984), who refers to
the aggregated organizations as interorganizational collectivities (ICs),
after the 1975 work of Van de Ven et aL The structural/functional
approach may be succinctly described by the assertion that "ICs, like
organizations, are distinguished not by their membership, but by their
functional goals, role structure, and behavior as a system" (p. 19). Or,
as stated by Tichy (1981), "The structural functional approach deals
primarily with the formal patterning of the relationships and the
causes and sources of such patterning (p. 226).
Whetten stated in 1982 that research to that time had been
"dominated by a structural-functional orientation inherited from
sociology" (p. 98), which means that it had been dominated by certain
This view of an interorganizational system assumes
that organizations voluntarily elect to form a union
to accomplish a mutually preferred objective_The

resulting functionally derived interdependencies
serve organizations as the basis for resource exchanges
that are intended to produce the maximum benefit
for the system as a whole, (pp. 98-99)
Early empirical work on interorganizational activity also reflects
the structural approach to its analysis. The use of adjacency matrices
to perform what is termed "smallest space analysis" was used by
Laumann and Pappi (1976) and Galaskiewicz (1979). In this technique,
adjacency matrices are ".manipulated in order to identify important
subnetwork units and cleavages between these units" (Mulford, 1984,
p. 152).
Block modeling, "a technique in which data in adjacency matrices
are manipulated in order to develop blocks of organizations that have
structurally equivalent positions in a network" (Mulford, 1984, p. 153) is
also found in many studies such as Knoke and Rogers (1978) and Van
de Ven et aL (1979). Other examples of early structural work include
the use of sociograms (e.g, Stentz, 1972) and structural typology
development (e.g, Warren, 1967).
As this work progressed, some researchers (e.g, Lauman & Pappi,
1976) began to express concern that the structural approach did not
address directly how collective decisions result from interorganizational
activity. Therefore, new directions for analysis began to look at the
collectivity as a total field rather than an assemblage of aggregated
dyadic relationships. These directions clustered into two approaches.

The Resource Dependence and Political Economic Models
Both of these models consider the entire interorganizational field,
or network, as the field began to be termed more frequently. Both are
concerned with resource acquisitions as critical to network development
and maintenance; however, there are distinguishing features between
the two models.
The resource dependence model has been articulated principally
by Aldrich and Pfeffer (1976), Aldrich (1979), Van de Ven (1980),
Aldrich and Whetten (1981), and Van de Ven and Walker (1984). This
model posits that dependencies develop between and among
organizations because of unequal access to resources. Zeitz (1980)
defines external resources as
any phenomena, external to the organization,
upon which the organization depends_The various
resources can be arrayed on a dimension from
those that are the most material and have a real
physical existence apart from organizations, to those
that are merely symbolic, (p. 74)
According to Aldrich and Whetten (1981),
The resource-dependence model of interorganizational
behavior treats dependence as a central concept,
emphasizing the extent to which an organization is
dependent on other organizations that control resources
and markets necessary to ensure its survival, (p. 393)
They see constraints on use of alternative resources as an
additional cause of dependence. According to this model, power is
achieved through control of "strategic interdependencies" (Boje &

Whetten, 1981, p. 378) and Zeitz (1980) argues that "organizations that
create economic resources are likely to dominate those that distribute
resources, such as social welfare agencies" (p. 78).
Van de Ven and Walker (1984) agree, and state that patterns of
coordination among interdependent agencies develop according to the
nature of the resources being coordinated. In their view, as well as
that of other resource dependence theorists, networks therefore become
hierarchical and network stability is dependent upon the dominant
Provan and Milward (1991) summarize and evaluate the resource
dependence model thusly:
Work in the area has generally concluded that
despite a loss of autonomy, organizations often
develop cooperative relationships with one
another as a way of reducing environmental
uncertainty and increasing their capacity to attract
the scarce resources needed to survive and grow
Despite the value of the resource-dependence
perspective, it generally neglects a consideration of
normative standards and legitimized procedures as
an explanation of why organizations develop and
maintain interorganizational relationships. Such
a consideration is particularly important when
focusing on organizations that have a strong, common
normative professional base and where the link
between resource acquisition and the performance
outcomes of individual organizations is tenuous,
as in the public and not-for-profit sectors, (pp. 391-92)

The political economic model of networks is also concerned with
resources, but attributes more importance to the environment of
linkages external to the specific network of organizations. This model
was used by Zald in 1970 and Galaskiewicz in 1979, with the seminal
article on this model generally considered to be that written by Benson
in 1975. Benson conceived networks as a political economy, concerned
with the acquisition and distribution of two scarce resources dollars
and authority which are essential to organizations ability to carry
out programs, to preserve their domain, and to uphold their methods of
operations, or paradigm. He saw resources as acquired from the
political economy, thus necessitating exchanges with the larger
environment bureaucracies, legislative bodies, publics, and other
authorities. This was supported in the 1977 work of Heclo, who stated
"Strategic resources are important because they provide the possibility
of exchanges with the bureaucracy that can create commitments to
mutual performance" (p. 194). In this model, power is based upon both
internal structure (centrality in the network) and external linkages,
placing much greater importance upon context and environment ("The
flow of resources into the network depends upon developments in this
larger environment" [Benson, 1975, p. 229D- The environmental elements
proposed to be most critical for networks are resource
concentration/dispersion, network autonomy/dependence, patterns of
dominance in the environment, abundance/scarcity of resources, and
control mechanisms affecting the environment-network.

In this model, as explicated by Benson (1975), balance in the
network C'interorganizational equilibrium'Tp. 229D is indicated by four
factors: domain consensus, or agreement among participants on the
appropriate role and scope of each organization; ideological consensus, or
agreement upon the nature of the tasks to be undertaken and
appropriate approaches to doing so; positive evaluation of the other
organizations in the network by each member, and, work coordination
among them. His conclusion: "The principal strength of this
perspective is that it establishes a genuinely interorganizational level of
analysis" (p. 248).
Networks and Their Environment
Bensons 1975 article was a springboard toward conceptualizing
networks themselves as units of analysis. Aldrich and Whetten (1981)
reemphasized Bensons conclusion, stating "Networks themselves may be
appropriate units of analysis" (p. 391). In determining this, they
maintain that, "The starting point for all studies of aggregates of
organizations is a relation or transaction between two organizations"
(p. 385). They conceptualized interorganizational behavior as falling
into one of three constructs: organization sets, action sets, or networks.
An organization set is defined as that group of organizations with
which a focal organization has direct ties. Action-sets are organizations
which have joined in a temporary alliance for a limited purpose. A
network is

defined as the totality of all the units connected
by a certain type of relationship (Jay, 1964:138)
and is constructed by finding the ties between all
the organizations in a population under study,
regardless of how the population is organized into
organization-sets or action-sets. (p. 393)
As networks began to be considered as a total field and unit of
analysis in themselves, the importance of the environment as a force
impacting network actions received new emphasis in the literature.
Mulford (1984) identified the following environmental conditions as
being particularly relevant to networks: technological, political,
economic, demographic, ecological, and cultural. And DiStefano (1984)
asserted: "domain, boundary, dependence, avoidance, and control all take
on full meaning only when viewed from an organizational-
environmental perspective" (p. 362).
The work of Emery and Trist (1965) began to be applied to the
study of network behavior. These authors state that "environmental
contexts in which organizations exist are themselves changing, at an
increasing rate, and towards increasing complexity" (p. 21). They refer
to environmental "textures" and propose four types: placid randomized,
placid clustered, disturbed-reactive, and turbulent. It is the last that
has been seen as most relevant to the development of
interorganizational networks and as having the most impact upon
network behavior. The turbulent environment is one in which
"dynamic processes, which create significant variances for the
component organizations, arise from the field itself (p. 26). This is

certainly the type of environment described in Chapter 1.
Interest in the impact of the environment developed in a branch
of the literature termed the population-ecology model of organization-
environment interaction.
The population ecology model of change implies
that a structure is only as complex as the
environment supporting it, and that in the absence
of environmental pressure to maintain complexity,
systems tend to slip back into simplicity.
(Aldrich & Whetten, 1981, p. 392)
Fields (1984) saw this model as a process of "mutual adjustment
between organism or organization and environment" (p. 55) and cited
the 1983 work of Astley and Fombrun, who proposed that this
adjustment can occur on a collective level as well as an individual level.
By joining with others in systems of mutual support,
organizations can produce a collectively managed
environment that is buffered, at least partially,
from the vagaries of the outside environment, (p. 577)
This analysis both helps to explain and to support the
development of IONs being proposed today, given the implication that
more complex structures are required in more complex environments.
As Gray (1989) explains, "Turbulence cannot be managed individually
because disruptions and their causes cannot be adequately anticipated
or averted by unilateral action" (p. 28).
Other theorists emphasized the interactional nature of
environmental/organizational relations, for example,

The study of organizational environments has
developed to recognize the environment not merely
as that which occurs outside the organization,
but as a system within which the organization
pursues purposes. (Leblebici & Salancik, 1982, p. 227)
Another view, that of Weiwal and Hunter (1985), suggests "the
environment may not really consist of open niches waiting to be filled
but rather of potential space that needs to be carved out, an enacted
domain' (p. 494).
The development of theory regarding the interaction of
organizations and their environments (also represented by the work of
Thompson, [1967]; Hall et aL, [1977]; Pfeffer & Salancik, [1978]; Aldrich,
[1979]; and Aldrich & Whetten, 1981] ) was foundational to the next
phase of the literature on interorganizational relations, that continues
the use of the network as unit of analysis and moves even further
beyond structure to process.
Understanding and Managing Interorganizational Networks
The literature on understanding and managing IONs is growing
rapidly. Mulford (1984) stated that the emerging theory building on
IONs must address three issues:
1. By what processes are interorganizational linkages
formed and maintained?
2. Through what mechanisms are partial or total
networks mobilized for collective action?

3. How do collective decisions and action result from
interorganizational activation? (p. 137)
The literature reflects these questions, but with decreasing
amount of work as one moves from the first question to the third.
A great deal has been done relative to the first question.
Antecedents of successful IONs and conditions favorable for ION
development have been studied by Whetten (1981), Mandell (1990),
Provan, (1982), Rogers and Whetten (1982), and Rosenthal (1984).
Whetten, in 1981, identified five antecedents of successful IONs. These
were a positive attitude toward coordination, a recognition of the need
and of partial interdependence, knowledge of potential partners,
assessment of compatibility and desirability (including roughly equal
status, common definition of the problems to be addressed, an
encompassing professional ideology, and lack of threat to domain
claims), and capacity to maintain coordination linkages.
The formation process of networks is clearly impacted by
whether or not the network is voluntary. Mandell (1990) describes
voluntary networks as occurring when the initiative comes from the
participating organizations. Other characteristics she cites include the
absence of governing legislation, decision-making by consensus ("Indeed,
the more the members are in agreement before the network is formed,
the more work will be achieved within the network" [p. 46]), and the
primacy of building linkages, particularly in the service of resource

She also describes two forms of mandated networks. One she
calls mediated based on legislative initiative in a vertical system
(meaning members are unequal) and mediated initiated by a third
party in a horizontal system.
It is suggested, therefore, that network formation may be
facilitated or hindered by the presence or absence of certain
antecedents and that etiology may be voluntary or mandated. Provan
(1982) and Rosenthal (1984) each theorize that IONs are grounded in a
need to maximize a static or declining resource base as well as to
reduce environmental uncertainty. In any case,
the design of an appropriate decision system
must be based on the deliberate integration of
the activities of individual decision units into a
multi-organizational arrangement for joint action.
(Hanf & OToole, 1992, p. 168)
Earlier work on IONs (e.g. Rogers & Whetten, 1982) stated that
there had been little attention given to the question of whether or not
IONs proceeded through developmental stages and what those states
might be. This has been more recently addressed in the work of
Kagan et aL (1991). Studying collaborations providing early care and
education in three states, they found that:
Collaborations pass through predictable developmental
states (formation, conceptualization, development,
implementation, evaluation, termination/reformation),
and while development is linear and sequential, its

pace and trajectory are dramatically affected by potent
mediating variables (goals, resources, power and authority,
flexibility), (p. 5)
Many authors have discussed the constraints upon successful
formulation and/or implementation of networks in intergovernmental
systems. Barriers identified by Mulford and Rogers (1982) included:
1. The tendency of organizations to maximize
their own autonomy
2. The ideological commitments of professionals
3. The varying concerns of client representatives
4. The division of interests of resource controllers
(p. 27)
Mandell (1990) identified compatibility of members, the
environment for mobilizing resources, and the political and social
environment of the network as potential constraints upon network
activity. Clayton, who reviewed collaborations specifically in human
services (1992) cited the following barriers: differences in decision-
making authority and/or processes, turf issues, perceived lack of
mechanisms for communication and information sharing, eligibility
issues, lack of motivation to collaborate, reward systems linked to single
institution goals with nothing comparable for broader goals, unclear
benefits, and resource inadequacies.
Another area pertinent to understanding network structures
discussed in this literature is the process of implementation and how
this is addressed from a network perspective. Implementation in a
network system relies upon what Porter and Warner (1979) refer to as

"mobilization behavior." As discussed by Mandell (1990), "Mobilization
behavior requires a view of the strategic whole and an ability to
develop and achieve a set of common objectives based on this whole" (p.
33). Porter and Warner (1979) further describe network characteristics
which managers who would implement programs must confront These
include a variety of public and private actors, an array of goals and
objectives, program oriented actors and subgroups of actors, a prevalence
of professional values and membership in multiple implementation
Agranoff and Lindsay (1983) studied intergovernmental networks
in the human services domain and found four dimensions to be critical
in framing implementation approaches. These were the legal-
constitutional-structural, the political, the technical, and a joint task
orientation. This and other work conveys the conclusion that
implementation of programs and policies through a network structure is
complex (appropriately matching, according to literature cited earlier,
the complexity of todays policy and management environments). As
urged by Bryson and Crosby (1992), it must be approached strategically
in order to "-achieve important public purposes in practice" (p. 298).
Campbell and Garnett (1989) assert that the strategic management
process is interactive rather than lines/ and suggest five key factors
for implementation strategy (all of which are highly relevant to
implementation through a network structure):

-strategy precision
-resource (human and financial) commitment
-an implementing agency suited to the program
-support of constituent groups and minimal
external dependencies
(pp. 269-272)
A newer branch of literature on implementation structures (to
move to the newer term for IONs as well) is that theory regarding the
management of interorganizational, intergovernmental networks.
Themes in this body of theoretical work include consensus building,
management of internal and external relations, jurisdictional respect,
conflict identification and management in the network setting,
management of interdependencies, resource acquisition, problem framing,
understanding and managing complexities, and the like. This literature
relates back to the fundamentals of IGM as outlined by Wright (1983):
problem solving, communications networks, and strategic and coping
These characteristics refer to the ability to achieve
coordination and control, to develop a pattern of
contacts within a system of networks, and to manage
interdependencies. (Mandell, 1990, p. 33)
As Mandell (1989) has also pointed out, it is essential to "manage
interdependencies so that both the goals of the individual organizations
and the goals of the network as a whole can be achieved" (p. 142). The
approaches and techniques used by managers of collaborative structures
are also significantly related to the success of network activities

(Agranoff & Lindsay, 1983; Aldrich & Whetten, 1981; Calista, 1986;
Mandell, 1984, 1989, 1990; Ring & Perry, 1985; Whetten, 1981). This
literature cites qualities seen as appropriate to public managers in
general and network managers, specifically. As OToole (1990) suggests,
".network managers or facilitators may be especially important
resources and may make the difference between success and
frustration" (p. 99).
Ring and Perry (1985) suggest five characteristics that are key to
the success of a public sector manager (which other scholars have
applied to the network manager, e.g., Mandell, 1990). These are
flexibility and adaptability, avoidance of immersion in detail, avoidance
of premature public commitments, avoidance of programmed responses,
and development and utilization of several attributes. The attributes
they specify are the ability to integrate competing viewpoints,
possession of a self-other orientation, a low level of dogmatism, open-
mindedness, ability for quick communication, ability to establish
productive relationships, the capacity to use symbols effectively, and the
capability to mobilize latent constituencies.
Mandell (1990) contends that "network management requires
managers to view their actions within a system of networks and to
make the best use of the relations forming out of these systems of
networks" (p. 32). Gardners (1990) discussion of attributes of leaders
also is relevant to network managers. He suggests that the following
are necessary: physical vitality and stamina, intelligence and judgment

in action, eagerness for responsibility, task competence, skill in dealing
with people, the need to achieve, capacity to motivate, courage,
steadiness, capacity to win and hold trust, capacity to manage,
confidence, assertiveness, and flexibility of approach.
Agranoff and Lindsay (1983) suggest other approaches for the
network manager. First, they state that it is essential to have
jurisdictional respect Secondly, they note that since politics are
unavoidable, they should be encouraged to a point, and conflict should
be addressed openly. Thirdly, they counsel reliance on adaptive rather
than rational planning, citing incrementalism as more appropriate to
networks. Finally, recognition of and respect for sensitive turf issues is
As Agranoff and Lindsay suggest, the recognition and
management of conflict is a significant task for implementation
structures and their managers. As early as 1982, Whetten contended
that the prevailing assumptions about the value of coordination were so
strong that
literature pays little attention to issues of coercion,
force bargaining, conflict of interests, and dissension
among members of the systemIndeed, there has been
such a pro-coordination emphasis in this field that
few authors have noted the potential benefits of conflict
in an IO system, (p. 99)
Buntz and Radin (1983) maintain that

the general failure of administrators at each level of
government to achieve a positive perspective on
conflict and its management is itself a barrier
to effective intergovernmental management (p. 403)
Their basic premise is that conflict neither can nor should be
avoided, but rather must be "actively managed" (p. 403). They theorize
that conflict in intergovernmental settings is generally due to either a
structural situation, a process event or environmental conditions. After
reviewing types of conflict and conflict management and possible
outcomes relative to human services, they conclude that
present intergovernmental managers would do
well to adopt a positive attitude toward conflict
and that they would be well served by seriously
considering application of the conflict management
perspective, (p. 409)
DiStefano, writing in 1984 on interorganizational conflict, also
expressed disagreement with others view that conflict is at odds with
good management practice, particularly in the interorganizational arena.
He called for increased research, stating that "conflict and change are
inevitable by-products of interorganizational growth and stability [that]
impact directly on management and operations" (p. 364). Alter (1990)
concurs, stating,
Conflict undoubtedly is an unavoidable process in
all interorganizational service delivery
systems_administrators and planners should be aware
of the possibility of high conflict levels and be prepared
to prevent dysfunctional conflict, (p. 497)

As well as positive outcomes from network activity, there may be
negative consequences, monitoring for which also falls to the network
manager. According to Whetten (1981), possible contrary outcomes
include a reduction in potential for adaptation due to too tight
integration, a loss of innovative potential due to joint programming, or
extensive coordination resulting in a reduction in the quality of services
provided by the network as a whole.
More recent literature on network management (Coe, 1992; Hanf
& OToole, 1992; Jennings & Krane, 1994; Rosenthal & Mizraki, 1992;
Rossy & Mandell, 1992) places a great deal of emphasis on building and
maintaining both common vision and focus, problem framing, sustaining
action, building and maintaining trust, and evaluation of network
activity and outcomes. Gage (1993) adds that "External networking
activities_are criticaLThey do this to build larger, more influential
networks of contacts" (pp. 9 -10). Resource acquisition remains as
important in the recent literature as in earlier work. Hanf and
OToole (1992) agree, pointing out that
A key element of interorganizational management
can be the need to work for appropriate levels and
types of resources, both for achieving network
objectives and also for managing or caring for the
network itself, (p. 173)
Rossy and Mandell (1992) summarize five management processes
they see as essential for implementation structures:

-building sponsorship and legitimacy
-articulating strategic issues and policy
-mobilization behavior
-measurement and outcome evaluation
(p. 14)
A last task for the network manager to be noted is that of
maintenance of the implementation structure. According to Gage
(1984), this process is "-moved by such factors as political intuition, goal
orientation, calculation of comparative advantage, and assessments of
capabilities of self and other actors" (p. 138). Gardner (1990) advises
Leaders of a coalition must ensure that each
member of the coalition becomes fully acquainted
with the constituencies of the others around what
they want, what they fear, and what assumptions
(misconceptions) they hold concerning the subject at
hand. (p. 106)
Ring and Perry (1985) remind that "-the skillful exercise of
influence is likely to be more critical than the wielding of authority"
(p. 283).
Network management, then, is understood to be a highly
complicated function taking place in a turbulent environment

The scope of literature with applicability for multiorganizational
networks verifies their complexity as well as the need for additional
work to extend existing theory.

The aim of this research is to increase understanding of the
constructs that are critical to the effective establishment and
functioning of human services networks as they work toward goal
achievement As has been seen, the state of the literature is such that
it was possible to generate research hypotheses in addition to general
issues for examination. However, it has also been demonstrated that
many questions regarding networks and their operations remain to be
answered, and the opportunity to generate information beyond that
relating only to specific hypotheses would also be highly desirable. The
research design for this study, then, must be suitable to explore and
describe the nature of the relationships between a dependent variable -
network outcomes and three major independent variables level of
participants, a shared leadership model, and a core value of
commitment to collaborative process. As well, it should allow for
emergence and analysis of additional data. Finally, given the
importance of context to the proposed research, the methodology utilized
must include the ability to identify and consider contextual factors.
Consistent with these purposes and needs, qualitative methodology was

selected. As noted by Harrison et aL (1990), "The qualitative approach
enables the evaluator to examine phenomenon more holistically and
provides detailed, in-depth information-in the context of the real world"
(p. 71). Denzin and Lincoln (1994) explain in more depth:
Qualitative research is multimethod in focus,
involving an interpretive, naturalistic approach to
its subject matter. This means that qualitative
researchers study things in their natural settings,
attempting to make sense of, or interpret, phenomena
in terms of the meanings people bring to them.
Qualitative research involves the studied use and
collection of a variety of empirical materials-that
describe routine and problematic moments and
meanings in individuals lives, (p. 2)
Bass (1990) advocates for use of qualitative methods particularly
in studying leadership, due to the complex and contextual nature of
These methods may be more suitable to providing
confidence in the results when complex hypotheses
are involved-qualitative research can uncover a wider
array of contextual variablesThere is much art in
leadership that is difficult or impossible to put into a
test tube. Nevertheless, there is much regularity in this
art that can be made understandable by detecting and
describing the patterns that appear, (p. 887)
Having determined that, in general, a qualitative approach would
be most appropriate for the problem at hand, the specific strategy of
the case study was chosen to proceed.

The Case Study Approach
Rationale for Use of the Case Study
According to Bailey (1992), "a case is a study of people, events or
organizational processes in situ, which incorporates a process of
discovery (Lofland, 1971, p.4) (p. 50). Thus the case study also fulfills
the requirements laid out in the introduction above, meeting Marshall
and Rossmans (1989) explanation that qualitative research
entails immersion in the everyday life of the setting
chosen for study, that values participants perspectives
on their work and seeks to discover those perspectives,
that views inquiry as an interactive process between
the researcher and the participants, and that_relies on
peoples words as the primary data (p. 11)
Ooms (1993), in exploring the need for more research on
programs and strategies designed to assist families, called for an
increase in case study research.
Case studies are used as an exploratory, descriptive,
and explanatory research strategy, and are
particularly useful when the focus is on a complex
new social event, institution, or phenomenon (such as
a reform initiative) over which the investigator has
no controL (p. 9)
The applicability of this description to study of networks is
evident. In the conduct of case study research, the investigator must
"concentrate on processes that are likely to lead to understanding -
Verstehen rather than on a search for causal explanations"
(Gummesson, 1991, p. 75). According to this author, case study research

has utility for both theory generation and development of tools for
Yin (1989) states,
In general, case studies are the preferred strategy
when "how" or "why" questions are being posed,
when the investigator has little control over events,
and when the focus is a contemporary phenomenon
within some real-life context, (p. 13)
This study substantively meets these criteria. It has been
demonstrated that human services interorganizational networks are a
contemporary phenomenon, occurring in a real-life context to deal with
extreme problems and issues facing people in their very real lives. Nor
can these phenomena be controlled by a researcher. Many questions
are being posed to the cases studied, including several how and why
questions how is leadership understood and implemented in the
network? How are values established and communicated? How do core
values impact network outcomes? Who is at the table and why or why
A case study strategy clearly met the research purposes at hand
and was selected for implementation. The next decision point was
single versus multiple case design. Although, as Gummesson (1991)
notes, "Case research seeks to obtain a holistic view of a specific
phenomenon or series of events. This is a time-consuming job and it
[may not be] possible to carry out more than one_in a research project"
(p. 76), single case studies are not always the most appropriate.

According to Yin (1989), there are three major reasons for selecting a
single case design. First would be the instance in which the single case
represents the "critical case in testing a well formulated theory" (p. 47).
It has already been demonstrated that development of "well formulated
theory" is a need not a condition of networks, so this reason does
not apply.
Secondly, "is where the case represents an extreme or unique
case" (p. 47). Again, as theory building and, secondarily, input for
practioners, are the goals, rarity was not a desirable feature, the
extreme or unique were not part of case selection, and this rationale
does not apply to this study.
A third rationale for a single-case study is the
revelatory case. This situation exists when an
investigator has an opportunity to observe and
analyze a phenomenon previously inaccessible to
scientific investigation, (p. 48)
While networks are underresearched, they do not meet the
criteria of previous inaccessibility. Thus it was concluded that a single
case study design was inappropriate for the defined research purpose of
this study.
On the other hand,
The evidence from multiple cases is often considered
more compelling, and the overall study is therefore
regarded as being more robust[However] the conduct
of multiple-case study can require extensive resources
and timeEvery case should serve a specific purpose
within the overall scope of inquiry. Here, a major

insight is to consider multiple cases sis one would
consider multiple experiments that is, to follow a
"replication" logic, (pp. 52-53)
Replication logic provides a sound basis for the selection of a
multiple case design for this study, hence, two human services networks
in Colorado were selected for study (to be described in the data
collection section of this chapter).
Description of the Case Study Process
According to Yin (1989), "The case study, like other research
strategies, is a way of investigating an empirical topic by following a
set of pre-specified procedures" (p. 25). As Yin (1989) has authored
what is generally acknowledged to be the definitive work to date on
the case study approach, the procedures recommended in his book are
those which were followed in constructing the design and
implementation of this research.
The first step in planning and conducting a case study is
specification of the constructs of major importance. In this research
the primary constructs are: achievement of network outcomes as
measured against the goals set by the network, as stated in one of the
guiding research questions; the nature (hierchical level in their
employing agency) of participants in the network leadership group as
stated in Hypothesis 1; shared leadership modeled as leadership
distributed among network members with the elements defined in
Hypothesis 2; and shared core values as specified in Hypothesis 3.

The second step to be taken is construction of a case study
protocol ("a major tactic in increasing the reliability of case study
research." [Yin, 1989]). The protocol developed for this study contained
the following elements recommended by Yin (1989): project overview,
field procedures, case study questions, establishing procedures for data
array, and a tentative plan for the case study report These will be
detailed in the Data Collection and Data Analysis sections of this
chapter and constitute what Yin designates as a "sufficient blueprint"
(p. 36) to proceed with research.
Third is the conduct of the case study, which "may be based on
six different sources of evidence: documentation, archival records,
interviews, direct observation, participant observation, and physical
artifacts" (p. 84).
Fourth is data analysis "examining, categorizing, tabulating, or
otherwise recombining the evidence to address the initial propositions of
a study_[which is] one of the least developed and most difficult aspects
of doing case studies" (p. 105). This procedure will be discussed further
in the Data Analysis section of this chapter.
The last of the "pre-specified procedures" (p. 25) is the reporting
phase "one of the most difficult to carry out" (p. 127). This
dissertation follows the recommended procedures for reporting on a
multiple case design:
This type of multiple-case report will contain
multiple narratives, usually presented as

separate chapters or sections, about each of the
cases singly. In addition to these individual
case narratives, the report also will contain a
chapter or section covering the cross-case analysis
and results, (p. 134)
Conduct of the Research: Data Collection
Site Selection and Access
The first tasks in the conduct of the research were site selection
and negotiation of access. According to Marshall and Rossman (1989),
the ideal site (rarely possible) is one where
(1) entry is possible; (2) there is a high probability
that a rich mix of many of the processes, people,
programs, interactions, and/or structures that may
be a part of the research question will be present;
(3) the researcher can devise an appropriate role to
maintain continuity of presence for as long as
necessary; and (4) data quality and credibility of
the study are reasonably assured by avoiding poor
sampling decisions, (p. 54)
Sites were not drawn randomly from a sample of implementation
structures. Rather, they were selected in accordance with the
operational definition of interorganizational networks set forth in
Chapter 1 (voluntary, but formalized, associations of human services
agencies characterized by shared resources, joint planning, shared power
and authority, a stated mission, and stated goals and objectives), while

concurrently striving to reach the ideal site characteristics described
above. As Eisenhardt (1989) explains,
Sampling of cases from the chosen population
is unusual when building theory from case studies.
Such research relies on theoretical sampling (Le_,
cases are chosen for theoretical, not statistical
reasons)Random selection is neither necessary,
nor even preferable_The goal of theoretical sampling
is to choose cases which are likely to replicate or
extend the emergent theory, (p. 537).
With consideration of all the factors named above and input
from human services professionals, other researchers and academics
regarding existing collaborations, ultimately two networks were selected
as research sites The San Luis Valley Coalition for Youth Services
and The Boulder County Prevention Connection. Both groups were
amenable to being research sites and both, upon preliminary review,
appeared to have probability of being rich data sources and to meet
the goal of theoretical sampling (likelihood to support extension of
emergent theory).
The next issue at hand was gaining access, shortly followed by
others typical of qualitative research: building trust, role management,
managing political issues, and awareness of reciprocity issues (Marshall
& Rossman, 1989).
In both cases, the initial link and introduction of the researcher
was made by a professional aware of the network and of the
researcher. The researcher attended a meeting of the leadership group,

presented an oral and written outline of the proposed project, and
requested their consideration of participating as a research site. In
both cases, the proposal was accepted It was agreed that the
researcher could attend, record, and/or take notes of any and all
meetings of the leadership group and other (entire network or
committee) meetings; have access to all network documents, including
grants, brochures, meeting minutes, evaluation and other reports, and
any other written products generated; and, request interviews from any
network member however, it was also agreed that members who did
not wish to be interviewed were free to decline. Overall, at those
initial meetings, there was a general feeling among participants that
they knew they were engaged in a new kind of process (interagency
collaboration), that there were new kinds of issues arising, and that
they were gratified to learn of academic interest in their venture. It
was clearly explained that the researcher would be a neutral observer
collecting data that would later be analyzed and reported according to
what emerged This was acceptable to all; reciprocity did not emerge as
a barrier, as both networks acknowledged being part of learning more
about collaborative endeavors as their reason to participate.
Trust and management of role and politics are ongoing process
issues which need to be addressed in the field The researcher was
readily accepted as an observer and, after a few initial jokes about the
tape recorder, appeared to have no impact upon the proceedings at alL
This perception will be supported by later presentation of case study

content, where it will be seen that many significant events unfolded
without regard to an observer being present. Trust was developed
through consistent attendence, neutrality, and respectful listening in
meetings and interviews. The researcher did not participate in any
way at meetings and remained neutral during interviews. Political
concerns were not an issue for the conduct of the research.
Field Procedures
As noted, the researcher enters the field with a case study
protocol to guide his or her efforts,.
The heart of the protocol is a set of substantive
questions reflecting the actual inquiry-The questions
are posed to the investigator, not to a respondent-
[and] are reminders to the investigator regarding the
information that needs to be collected, and why.
(Yin, 1989, p. 76)
Case study questions for this research, derived to investigate the
proposed hypotheses, had as their essence the following issues. What
are the goals of the network and what have been the actual outcomes?
How is leadership conceived and operationalized by the network? What
leadership tasks are seen to relate directly to outcomes and who
performs them? What values underlie the goals of the network and
how they are pursued? Are they shared? How important is it that
they are shared? How are value disparities managed?
With these questions in mind, data collection was initiated. Yin
(1989) notes six sources of evidence which "can be the focus of data

collection for case studies: documentation, archival records, interviews,
direct observation, participant-observation, and physical artifacts" (p. 85).
The first four were essential components of these case studies,
constituting the foundation for field procedures utilized. Extensive
taking of field notes and audio taping were the recording methods used.
Interviews of key participants in the network leadership groups
were a major component of data collection. As noted by Hanf and
OToole (1992),
It is rare that the perspective of the actor inside
the network is taken, that of someone who is trying
to work within and through a set of relationships
with other actors, in pursuit of both individual
and collective objectives, (p. 163)
To uncover that perspective, interviewing was a major technique
in striving for understanding of the dynamics of each network relative
to the hypotheses being investigated. As Easterby-Smith et al (1991)
explain, qualitative interviews are
the opportunity for the researcher to probe, to uncover
new clues, open up new dimensions of a problem, and
to secure accurate, inclusive accounts that are based on
personal experience__to understand how individuals
construct the meaning and significance of their
situations and [their] framework of beliefs and values.
(p. 73).
Interview questions were derived from the literature and focused
upon the identified constructs. All persons interviewed were asked the

Full Text


012 )03)#0)21+*#40/%2)5#3)1#1/2/#3 /%412/144)2%26#47#%4#34)8%291 *1$ #%/:.;< )3"4/:.;= *#"/:..: # "/0" 0* *# :..;


&>':..;*1$ #"


0* *1$ 2 !* ?$! 198 0


1$*&*@'*#' 03/5// $4! *286 # 12#/ $!" A"" "" !" "" "&6' #6(" "!( %" /"!/ $! "A" ""


"( B& '3 "" #"" C(" /D"B B" E 2& CF


0)0/#%4 *!


#/94%83)06))41 0*286! !135/ G1" /*"/ !13# (( 0* !"


/%4)41 /+#*)2 :42%0/%4HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH: /*(((((((((((((((((((((((((((((((((((((((((((((((((((< *1HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH:I *?(((((((((((((((((((((((((((((((((((((((((((((:I %0J(J((((((((((((((((((((((((((((((((((((((((((((:; )((((((((((((((((((((((((((((((((((((((((((((((((((:K 2+(HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH:K 131((((((((:. 15(((((((((((((((((((((((((((<< 2"1((((((((((((((((((((((((((((((((((((((((((

*4!HHHHHHHHHHHHHHHHHHHHHHHHH=: 4!HHHHHHHHHHHHHHH=M 4!3HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH=; 1#HHHHHHHHHHHHHHHHHHHHHHHHHH=; 20*) HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH=. 4!)"HHHHHHHHHHHHHHHHHH,< $ 4!HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH,= /HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH;, 2)1)#2/+0)164#40)+%0%3%6GHHHHHHHHHHHHHHHHH;; HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH;; /1#(((((((((((((((((((((((((((((((((((((((((((((((;. 2/1HHHHHHHHHHHH;. 0/1*(((((((((((((((((((((((K< /2D0/((((((((((((((((((((((((((KM 11#((((((((((((((((((((((((((((((((((((((((KM ?*HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHK; 0#HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHK. /1(((((((((((((((((((((((((((((((((((((((((((((((((((((.< /((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((.< ((((((((((((((((((((((((((((((((((((((((((((((((((((((.L 0((((((((((((((((((((((((((((((((((((((((((((((((((((((((.= /(((((((((((((((((((((((((((((((((((((((((((((((((((((((., 1(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((.,


M+)1#4315#33)G/%#3%4?%2G%+ 1)25/)1HHH.; J(((((((((((((((((((((((((((((((((((((((((((((((J.; /5135/(((((((((((((((((((((((((.K 6J((((((((((((((((((((((((((((((((((((((((((((((J.K +J((((((((((((((((((((((((((((((((((((((((((((((J:I: 0JJ(((((((((((((((((((((((((((((((((((((((((((((((:I: *J((((((((((((((((((((((((((((((((((((((((((((((J:I< 8:,,,,,*"1 /((((((((((((((((((((((((((((((((((((((((((J:IL 135/:..I(:..<((((((((((((((((((((((((((((((((((((((:I; %1)/(((((((((((((((((((((::, //42D :..M(@:..=(((((((((((((((((((((((((((((((((((((((((((((((((((((((((:

? /*"/D G<&%:..<(1:..L'((((((((((((((((((((((((((:MK GLD166% "?((((((((((((((((((((((((((((((((((((((((((((((((:=< 4":..M:..,DN 20((((((((((((((((((((((((:=, 2N /*//1(((((((((((((((((:,K ,/2%11/#1)#4#3G11#40/#1)10G?40461:;K /0$ 4!(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((:;. /14!)"(((((((((((((((((:K: 4!D# ((((((((((((((((((:KL /14!" /(((((((((((((((((((((((((((((((((((((((((((((((((((:K; +:((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((:.L +<((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((


/+#*)2: 42%0/%4 $!&%4'& O6F:..IB92"F *!:..:B2F:..

/M="/ ,( /" /; 2&:.K<'P0 "P&L<' /* 3#;2 "# "! &1!:..<' "& C"' &#:..I:..:B6F1 :..,B9:..,B"F !:..:B4/ /:..:B43/:..:B1:.KK'# &:..I'P P&,:'1"


4/ /&:..:'(#KL ""#P (1< &4//*" :..,'4:=:Q:K" &/E0?:..;'B /:=Q&1:..;' #/E0? )"" LLII# MIIIII" :..;<' #/1+1 /&:..;' "M=Q ,Q! :.K;" *" / !" &/E0?:..;

4 33 / ":..: /#J4:L ""< " 3 ""3&,*? ""( 8+E&:..I' "&!:.K;' "" E( &L' M


/! &:..L'P* P&::'8 &:..:'!PEPP ( @P&<' 91+ 0 !"#$%"&' :..M' #( % ! ! #" &<' /:LIIII&:=Q'""B :..I M=IIIIK:III& "'&1:..;' !(6&:..='P ("P &K;' &:..;'P*" !"P&

)""*"#'""%' &:..:' 43/ ( C:..I "" / &:' #"S P)"1;IIIII P&92"F*!:..:<' /::Q@..: <,Q &1:..;'43 &:..;' P:..L:,((

1P:..M L,MIIL=Q:..I ,;Q:.K=P""" &F 3 :..;'/ #":IQ C ?( <,Q :< BMQ &:..=/G2! "1".' #3 &:..;' "J)E 4T

&4/*"/# *%"+% 5U=:..=' /P.;M= :..=( P&1:..;<='"1&:..;' P/"" P&<=' 4" ( #&:..:' (( "+:.KI 1 D! ( #01 &=LK' / " 4 %(" ((! " )


( &6F1:..,' # :.KI&6:.K.' & :..,'#P ""P&+!:.K.<=' $" /83#&:..L'P P&,'%EC&:.KM' $B """! &M..' +$ # P$P #" $" !B"" $


, 3(PP ("J $ &6F6:..::;' "" *1 P* P&/:.K. <' >%8 #9&:.KM'P J/ "P&::=' "9E B""$ #$" ( ?H ( &"F !:..:K' *"


# 4$ $ -' -' &9F9:..<;' % #1&:..I'P/ +( #"" "P& L ,'4#1 &:..:' ""P" "C" "P&L'?" R :'1 <'1 L'1 M'8!" ='1 &M' **


# /""" /C" 3&M' #" J "+."''"/"("."/' %%."'"".+"' "+"'-"(%0 &"F !:..::=' +%E&:..<' $ "P P ""( "&:;=' "E 1&:.K;' $ *1


"P%$ "" P&;I<' 61&:..,' +" "" &.=' P/"( P&.='!B@ 9&:..M'P1 "$ P&LM:' #&:..:' "( P "P &=LM'""" *!


#" "// 1"#% &:..,' !"( "" V"J ? ""( &M=' #&:..;' /""! "! $" $ &:=M' 4" ""39$*&:..,'$ #" OW!D B "B B! ("H &L;' 6&:..I'P*(" "P&:M;'" !&:..:' *(


2(" "" "" """ "3&M' #P" $P&6F6:..::;' +!&:.K.' 8" $" (# !? "%"% &<=' *"&:..:'$ !" "" $ !$ "&L.M' 6" "" "$$ ""'


" "" &41 #:..:'+" " $! !D$ !! ""!3! #91&:.K;' 4!!HH "! J4!!" ($ "&<=(,' 2"A%4 !" &#:..:B :,


)3:.K,B9:..:B3:..;B :.KMB%E:..;B2F8:.K<' %0 $D +"D(( $ "D" C (C C" $!&%4'A AD ? "& $'" $C C" /D "! /"D 4!D $! X X


4!" &" C"! !!' 3383@ $! D8 "!Y8 $!Y8 !$Y+ !&'$Y8 AY# "" Y 2+ "E $B P P&/:..<<<'


! "" "" $ "! 131 1! &93:..:' &#:..:B93:..:B :..IB2F:..<' +!&:.K;' &'E $ "P"$ P&K.' "" &3$' "+!&:.K;'P


X B CP&.I' $ 8"!E&:.K.'! B $ "" +:D &"C"' "" $&3/)% "' "( "3 "( "$ """" ("8&:.KL' 1"


$ +

15 "$ "! $1$"" P!PP P&6:..I::L'/""P ((P&%:..:

$&A'"8 "&'Y +LD" &" '" &' C" /:D/" "! EPP$ "! $"B "" 2"1 1&:..I'P8Y 4E"(E E"P&;'%93&:..:'

PPPP 1!! E "" 1 !" &<' /!! &%E:..;'

/+#*)2< 2)5)8%?+)3)2#2) $! ("C (!" " # "" ?"! " "&6' "6 :.;I8&:.KL' <=


:#A B" <# %%" (" PPPP L# (! &

#&:..I'""" P! "P&=K'# "" 8&:.KL'" ?("" !" "( 16P-&ML:' "!! "P P&ML:' "86$ "!2 P$"P&MM<' !8P6 P&:..I:=='1E6 P ( P&:=,' <;


"#3 6:.KLDC ($"" P "P&

X "$ "(" $! &:,,' #6 "6&:.KK'P1 "" E""P&:;' $ <.


1#33 #G!&:.K.' 3" "" &<' C $A & "" H B E &*"F1!:..IL;M' +$ $( !C"Y #2&:..<' $" "" D LI


! $! &<' 8"!&:.K.' PP&K;'" "P1 $ ""P &K;'6&:..I'P P&,')1&:.K,'P "H "V:&L=.' + &:.;;' )"B (+( (&:==' D"B !B" B!B L:


3 //5 " & F4:.K=B :..

B( & F/:..

0"&:.K,' 686 "$& $' P+P&:=I'1!&:..:'P" P&KLM'& C P#"" & $ "P& :..: K

"""! "" DP! $("P &::.' &:..<'P"P&.:' +!&:.K,'!E $! 1! !! &;:' 9!+&:..:'E! /&:..<' PP&,,' 8"" !&:..L'P1 "E" "P&;.' 2&:..<'! "D L=


$" P" "(OW P&K'D" 6#22 ""! /3 =*77(? "$! 2 (" ) =*77*? P!" P&M;'8 EEPPP !3 +""! !31!&:..:' +! !H! "# !$ !! !+


1 &::K<' +!&:.K,'"D !$ "! #,BC83 P1"! "P@6&:..I' P "P&' #"7&:.K='"D #" "# ("'+ "E !" #" ""! &.M' ?&:..<'"P L;



"" #"" "*9&:.K.'DP*! """ ("P&::'P"P B !" $!#6&:.K.' """+ P#" P&=;,'#1&:..I'P/ "!( -&:=.'?" $ "P3" !P& :.K=:K<' +"YPC !"P&9$ *:.K;,;' L.


C *4! *" 1! &!' 6&:..I'" P" P&=='&:.K.'"P $$P &<,.'" +B C" &M,.(;I' &:..I' $ $" $" "" $H"" $ $"& :.,=' ZMI B


" =3' :.,=' &<,M' "!1"&:.K;' + "D H!" "" "" $H -0 &:<:' !&:..L'B !DP/" "P&;.'6&:.K.' P "P&::<' #"6E&:..I' (CP ""C "P&::.'+" M:


"(!! 8&:..I' P(P&:=.'(P" ( !! P&:=.' +" $$+ "/6E&:.K,'"D /H $ """" "&=K' %C [ !! ("A E&6F+:.K,' #A !* M<


29 !! &9:.;;B CF8:.K:B+F2:.KK' $ "" "!! !+ C &:..I'DP(! ("'4 $! "'' P&L<' 131 #93&:..:'" $ %4&#:..:B:..IB 8:.K:'#51?&:..I'P3 !""P&MI':..I 1 D 3$ D+ 3 ML


YV&L<'+ GP1"" P&# 6!:..

(" "&/:..< K'? ( ( " & :..IL<(LL' +!&:.K;' %4 P" P&K;' +!&:.K;'DP%$ ; B CP&.I'* &'? &$%4' P $P&.<' M=


+!&:.K,'! P P&;,' #$ (*&:.K;' $ 3 +" + " A! # 51?&:..I' D 3 " !&LK' M,


6&:..I'$DP! P&:=<' M;


*13 P( "P&3:.KK;'# #CC"" """ 1 "" &+F%E:..<:,;' "! *P" "P&0:.;<:' "P C( ""P& F/:..< =;'#" "!D" B!B 1 VMK


?/*)" /"( ( %'5 ( J% ""3( $(" &+ %E:..<:,=' 8!3:.K, &==' "!+&:.;;' PP&:,,'" D HH (( ! !" 3 & F/:..<:;' #/&:.K.'0G E*!! M.


E ""&L' 9&:.KM' P P&:<=' " "" "4" "9 3&:..:' "" #A" &:.K.' P#"C" "$ C"P&<;<'%E&:.K.' B! &<

#+&:.;;' P"P&

!" !P""! !P &M:=' !&' $" !$ &:..:'D *! E!R" ''("#."4" ."4""' '"+/''%"2.03 &:M' P*! ."4""' ( 3#%+'""%'%"#'3 &:,' !6 ( """ "$&:..:' =<


$ !! (( "" B! "!& $'"&:=' !6 91&:..:'! E;! E$E " "&<=(<,' #+1&:.;K' $!3 "C" B" &<' 1&:.;K'$ "! P! =L


P&LM;' ?" 6E&:..I'" !" "*! 8 P" P&:L:'" P!"" ""(P&:L:'($ "!" +"+&:.;;'DP P&<,M' 4! # " # $( " =M


" &)F*:..LLML' )"" / # &6 3F%E:..I:<<' 3*&:.K;' !$ ? "8E3 )*&:..L'" $ #/&:.K,' P $P&<,L' #$ "& "' ""#%&:.K,' ==


$H$ (B "&:K<' 3! !(" &

4!3 #8686# !%4 &2F8 :.K<'! "#7$&:.KI' P1 "P&;;'C !" 389& &:.KM' $$"&/' :.;=!553A P/! $ "P&:.'% &:.K:'P P&<<,' 8:.K< P( P&.K' "$ $" C"H =;


" "$ &.K(..' )!$" C PP 3*&:.;,'6!$&:.;.' CP !"P&:.KM :=<' !PC "!$" "!P&:.KM:=L' 9!2&:.;K'5 53&:.;.'%! &1$:.;<' "&8:.,;' #!&3F* :.;,' "$ "! =K : 555BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB


20*) $ !" B" #*&:.;,'#&:.;.'55&:.KI' #8&:.K:'558!&:.KM' $7$&:.KI' $ $H" $ &;M' ##8&:.K:' $ $$ $ !""&L.L' # "PP& CF =.


8:.K:L;K'7$&:.KI'P$ P&;K' 558!&:.KM' " ! $ *"&:..:'$" D 8! $ "" ""J 0"( "$ $" $1 $" "! "$ ((&L.:(.<' ,I BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB


! !!$ 7:.;I6!$:.;. :.;= "! ( ($E + "(" :.;;!+ P1" P&:.M' &!'! "&P !" "PO :.;=<<.0(" A!A "A "(! ,:


&:.;=' !/V$V<<.0 D $B !! B"" $!! +DP "$" P&

"/: "" ($( ( " &#F8:.K:L.<' ?&:.KM'PC $"P&==' :.KL!#? C""""3 C $" ""&=;;' "%4" #6&:.K.'P" "P&

6 $" "$" EE$ $ &3F1!:.K<<<;' #"8+&:.K='P "E -V&M.M' $"&! O:.,;WB+3O:.;;WB*F1!O:.;KWB# O:.;.WB#F8:.K:W' $ !"" $4! %4 &:.KM' %4D : $! Y < !$"Y ,=


L+" $"Y&:L;' !" #" #%4"%4 ""8&:.K:'&:..I' *"&:.K<'28&:.K<'2&:.KM' 8:.K:"%4 & '! !"&:..I' "!" $% "(!&P !"!POM,W' !" HHHHHHHHHHHHHHHHHHHHHHHHHHHHHH


1!% """ &' $ "*" &:.K<'2&:.KM'$%4 $ "" ($C &+F%E:..<:,K' )!%4&2F8:.K<' %4" 9&:..:'1" D /" &$" "A' ,;


C "& '&=' A!" 2&:.K<'D :$$ < L"" M" &<;' &:..I' "$ "!! "/" "&:..<'D[ !A"! !"! #! !" !*8&:.;.' +)


P$"P#&:..I'P$ "" ".-' ""C"P& LL'*8&:.;.'! "" C"" #3&:.KL'"! ( (C! !" & E"'# /&:..<' P("P&<.K' /6&:.K.' "( "! &" !'D ,.


( (&' ( ( (" &<,.(<;<' #& "%4' $"! C 8&3&2 !68&:.KL'D "! " &:..ILL' #&:.K.'P "$ !"P&:M<' !" ;I


&#F3:.KLB#F8:.K:B/:.K,B :.KM:.K.:..IB2F*:.K=B8:.K:' !#%E&:..I' P! P&..' 2*&:.K='"! &" !:..I' "" "$" ("( "" $ &:..I'P! 23 "! !P&L<'6E&:..I' "!+ D"C ;:


!! "" #3&:.KL' !?" C1 " !?" ##3 #:.K<8 "" J ( %&..' $2&:.KL' ;<


" """" ""&MIL' "P"P&MIL'$ ""# "" " "&MI.' 01:.KM$ E" $ +P "($OW P&L,M'#&:..I' /" $"" H "&M.;' ;L


#"!" %.# #8&:.K:' ""C "" "! !&/:..

(A" ( ( ($" (" &:M' #!! #6 &:.KM'P(" "" P&:LK'6&:..I'" 3 ( &'C &:I,' 2*&:.K='P(! !P &



/+#*)2L 2)1)#2/+0)164#40)+%0%3%6G "!! "# +" "( !(C"(" "# ?" $ /" ;;


#+3&:..I'P" "((P &;:'0$3&:..M'D N" """ C" N""" "( "E"&<' &:..I'"" ""("" "J 4" &KK;' +"" ;K


/1# 2/1 # &:..<'P" $E "E&3:.;:M'-&=I' 2E&:.K.'" "E" !!"" "" H E&::' %&:..L' /" "& "'"" 3&.' "" P!( 78 P &6:..:;='# ;.


" G&:.K.' PPPP """ (&:L' "$! ( ""4 ( !Y+"Y+ "!Y8 Y # "#6&:..:' P/!" "(C OWHCP &;,' KI


#G&:.K.'C ? PP&M;' "P P((! 1P 5% +9+ P&M;'# #( ('""0 "" $" "&MK' 8! % " JO+"W (" J)"" "+ "%" D K:


.%+''"-+' %+''5% ( PP&=<(=L' 2" "! /& 0/1* #G&:.K.'P! (P&<='#G&:.K.' !"! C D"! B&" '! +:B +

! &PC '"/' POG:.K.W'" G&:.K.'DC"" 0/0# GPP &L,' P "D" """ P&KM' ?(P$ JOW" P&:I=' 0# P(P&<=' (PP&:<;' D ( KL


" "( &:LM' /2D0/ 6&6633 #2&:.K.' &' &:'B&<' A B &L'" B&M' &=M' 1 2 $! /:&"$" $C C"' KM


" "#)&:.K.' 1 1&3H 'J2 "H &=L;' 8" (135/G1" /*"/ " &! "D & F2:.K.' ! K=




" "" "* ?* # ( (." ( OW" &G:.K.;,' /"" 8 !"Y +"$!Y8 Y8"! Y#Y+ Y+"Y 8G &:.K.'"P K;

PAGE 100

D"" "("P&K=' $)" "!! C#+ %E&:..<' !! "C"&:,L' """C 8&3&2 !" "#)(1&:..:' "" " OW!" &;L' "" #"! ))

PAGE 101

"$ %"!"C &'! PP&" '"PP#@&:.K.' E &:M' "# &/M='"" !" &' "" 0# #$ &):.K.=L.' K.

PAGE 102

8! "" "$ G&:.K.'"" D!" !"! !& '" 2 /M; N"" 0$&:..M' 4!/ !! &=II' #2&:.K.'!P P&::<' $BB B B .I

PAGE 103

+"+E&:.KM' P P&;.'#&' "P5 P&;.'?" D)"3/!D*# /!D* )&4! %C"01)'// 3!/!D)" ////% 5)" ( /,; +&:.KM'D "!A %"" %"(

PAGE 104

"" &):.K.=M:' "" / /, /1 #2&:.K.'P# C"P&:MM'? "PP& :MM'&!36O:.K=0 PO" "C"W "P&:M='D / P .<

PAGE 105

CP&:M=' "!" #" &:.K.'3+ "! "" ( "" &
PAGE 106

PP&;;' "&' +! $ &;;' ?G&:.K.'" $ $+P P&LK'2") "/ &=M<' 3E $ %4 )(1 &:..:' D .M

PAGE 107

" $ &' "" GE&:.K.' 3D "B B " 0 D HOW"A" .=

PAGE 108

&F2:.K.:M;' "" / PC"P &F2:.K.:M;' " P"!-&:M;' # ( 1 / "" .,

PAGE 109

/+#*)2M +)1#4315#33)G/%#3%4?%2G%+1)25/)1 135 (#/C/26 1( 1/1@ 1/1@ 5 /A46" "5 ( G 5 /6 "13 5/G1"&135/' $?:.K."" .;

PAGE 110

!PP (( #:," @2$/" @2135/ ?" /" /5135/ #/:" !B"/< $ 6 6" "13 5B5 $/:II&41',= &)8'%,<,L / .K

PAGE 111

&2:..<' ""5 0.: 58+ 1"'" "")"" &"0 11"26" '""! (& "5'[ / 85B" :IP"! "" "C E /#&1' 5 B Z.. BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB

PAGE 112

C" 2 :K;K# :.II "#1/ # /#?$&:.;,'P E((((E"! P&==' 5 5$B 2 &// $*'1 5&13(:K=:/C' 388&2 ) !P5P P5P '# "!" """ !2" 5"P3$2135 %%+ ?:..L !5" *""

PAGE 113

#?$&:.;,'P135( #@+"E +P&::' 3 5% 4"#&//' "1 :;,L&1:.;.'0 1! #:KII /:K,: :K;,+"1 "C5 E2#63 0&:..,/#/!1' #26/ :ML;K::M,M"/ "B,=M.KQ8 MKQ+ ""! )(8LQ !;Q *"*

PAGE 114

3.>66 :=Q%8=
PAGE 115

)"135/" / 4!" 4":M:.K.9P P*%1 &#/10'1 010@/*% "135/"+ /&+/'01"#/ 011"&011'#*/1 #/1)0 /"1"& %/1'" "+ :M((D A" """" "1+ *011! + :IL

PAGE 116

#E"" D :0Y
PAGE 117

" C" 38&2 !! &#"' #P+" 38222 "P 1! "" !2 / (@2 1 8+)2)#1" 13 5" 8+)2)#1135" "" "" &#"' 4"C "&"' !(& :I=

PAGE 118

# '+"" " B 135/ "#[ !! &'" A26/011 &5/01 1"@2 %:..I'&A260 '% &/:..<' !" "" B ""! *"+

PAGE 119

135/!@ :..L@:..= / ?:..<" 13G/:..I(:..< /"5 DPP "( :.K.@:..I &58+1"' /0"G1"12& 0"G/'011" //C1/ "D 5 1/ !B D :I;

PAGE 120

0/*%:<@ 0 04+01) %/1 3+0#011 #1$)"0+/ 1*#+1& "" +1"*1 $ "$ #+/ %/10G1* 0011#/C/1 /"B"T=II 135/?1/ & !'"( (( "&'A 135/ #" C"(""" !$ 135/":..I/ ?/E :IK

PAGE 121

" "%";I "/" +/+ ?*"*&C"' 5#*6 "# / "135/ !#:..I 5(C #/ +" / (%1# *"&%1#*/1#*(/1#*"' /0?/ /! %/1 C" # /( C""$" 135/$ *"7 BBBBB

PAGE 122

"( (! ( 135/( 0135/! A5 D 10G46 #11)1146" *3#4446F#05146" D "/%%204#%4" 1!)4+#4/)0%2 #00%4#3*2%62#46&135/' EC" ?:..I1%1#*/T:; "135/*"*CB/! *"*C0", "&135/" 'B5(B B*"*C **"

PAGE 123

"5 $"" %/1 *"*C 1%! @2 ;< &'D 135/"+/ :<@0*0 :<@0@ :<@03)# &0#' 135 /"1" 58+1" /1"/") #/011" 1/011" //011" /C/011" 135//1" /0"G1" 2!1 135#+)/ *1* / 0"+/135/ +/ "+3 5""!

PAGE 124

:..:CC "1 *"*C/ "" 135/" C& *"*C'+ (0*AA +/135/ 135/# *"*C 8C /DG* *"*C6+% 6/6+ 6// 5BG*E "B E" """/*" *CB%/ $ 135/B **1

PAGE 125

*"*C"#" *"*C/ 135/1/ 135/" "%1#*!:..: $ #:..
PAGE 126

@:..<1/ / 0"+/#11*&/E #1"1*CC4+ /1/ !" :..L ?" 1?2/B135/ ""A"C %135/G3/ 5E! 58+1" //?2/6"E% ?/C 135/ :..<$C& 'D :'#!"" $$ ::M

PAGE 127

6" 5 "6C" "" ""& '# <'/ "5 "# :..<" D135/*"*CB135 ?*"1"&+'B /A#+1B+ (0*CD#/B/ /?2/58/0?/ B(C *"*CC """D6+ 6/1 .M X::= 6 XHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH

PAGE 128

"135/ ? %1)/B:..L :..L$ // PP5 & A"'! 52,I(KI5 "$ #1/135/ 1// / C&1 510' #1/ "!1 //#1$ **+

PAGE 129

Z)"0135/" +/#"1 /04 + %/11$58+ \1"1&//' !( /4" 0*0" 011" /+"0 "1 .M! :..L" :"" 6+ !&)" 5 4 G(4 !G(' !*"*C C/"B#0#0 *"6! "#01"# %%+ &135/'?:..LP%"KIII """"" X::;

PAGE 130

"*" *CP&:' (C # 6/""( B/!#:..M% :..L135 /"+/T.IIIII )5C#?* 8 135/ !" //E/135/ =+'" /A#1+/ *"*CE5 #:..L/ /08 5(P #1$1// /!*"*C01/ /1#*/6"! #:..LB11/ ::K

PAGE 131

/L"" P8"(" ""38!1 /P&.AKA.L' 135/&"'D / +"" C 1/E"! """B 1E" %"/ 1/ 2"! 135/( / 1 / #( 1/:..L 135/4" /%+*"*C0 #0"01 ::.

PAGE 132

1"1/ "=01:..L 135/" "" *"*C" ( //42B:..M(@:..= :..M135/ L<$6+ ?/0(?/6 TLIIII135/*"*C 5"[ ( 6/ "135/!B +/ +"1135/! /1#* *1"

PAGE 133

/:..I"@ 1/ ""(" Y8 (Y%" Y # 6 135/(12" !? ""/" !" "%" 135/1 /12"!?135/ !(""(" ""1 / /606& +/'#1$/+" *1*

PAGE 134

1 2"!? /" "" 08#?*2 / "/ / 135 /MA:IA.= "6+ !M D/ #/ 10#/ G *"/ / *"/! @:..=" /1#*?38 /E *"*C 135/&" *11

PAGE 135

A:..I'1:..= )2.>. &*"*C' "P*"*C3!?P 5 8 135/G1" *"*C/ *C ) " (1 *"*C "" $ &:' 135/ "D1/ / "/B12" !? "" ""135 :
PAGE 136

2N135//1 1"$ " $" ?D8 "!Y8 "" &@2' &'1( A *@2135/ ("135/(C" ##0 C"!1 35B 5#"" C" *1(

PAGE 137

135 #" / B1 /" "$"" &!1 / "!' #" C" " PP !" +"C" "" C"" "" "" "135BC" :<=

PAGE 138

135"" 01*! ? "! /E# "("$ /E"! :..M" /EB *"*C"#" /"" "/1 /B"" /1/ /"#" 1/ "! E"8 C "" C" :<,

PAGE 139

(# "135/ C C" " C"(/ "" "C" #"C" A!/ "P *!P ":..LP8E9 0135P @:..M ""5 C"# 5&C #'( "135/"+" :<;

PAGE 140

"( # 135/ C"$ /" """ 4" @2P8! 0" !"E P "/ "E "!( E"" &" P*" (!" "135P *1)

PAGE 141

135/"" 135/ "135 &/1#*' 5! $A" &1"" !135/" (!' "$ 12"!? @2P) 135CP" %/ 1 /#1 / ? !8 ?/0 :<.

PAGE 142

0 !" 5 "& / '?! "C(" D8 89& !Y8 !$Y +< 135/"" /1/8 /& "" "' "PP/( "1/ :LI

PAGE 143

"""" B/ "$ "/)" $! 135/ ( *" *C2A8 / "" /C 1/"/ /A$" /&C'! 8 / "135/1 / :L:

PAGE 144

" /!&'# !" 1"!" !A "A &'D!"( "B" BB" !BB BCB B(B #D+! &' ,2932F GC' !" B &23 + 15 8"D A Y!135" H*!1

PAGE 145

"!4 "" ? + L #!8 "" Y /; :LL

PAGE 146

/+#*)2= +) %30)2/%4G*2)5)4%4/%44)/%4 &86 // L=0"B)(8 6*2! <<=LL.&:..I/'" ;=I / G"( /C"" 8%1#*"&%1#* (/1#*"/1#*'2 *C ":..I""C"C $( //& //' / *"/& /*/'8$ :LM

PAGE 147

" "" / 3! "C" //& "' 2" /E" /"" "(" 0"" " ( (8! &:..:,' // :L=

PAGE 148

/5 // #" 3I ?:K=KB 1:K=. / $8 /P#:K;I P&1 :.,;;'0:K;I:.K< ,
PAGE 149

" J2# /" /2! B /0""/ :K;KB:..I K,.,.
PAGE 150

:I:.KI:..I :IIQ#A* =MQ4"#L:Q ##<=Q +3 &:=Q'3&::Q'(" /
PAGE 151

"P2 / 683 + + 683 "P&:,'1$ ""D "!" # $ / "&=' 8/ "(
PAGE 152

/ // CD "( C"1" ""& /0?/O/0?/0 / 2?*%A //D 92/$0#? 3 3/ / 2? /0? / 6/#*/0" / 0C3/2 0 #+ /0# 6) /1 1* 510 # / / /+0 6 // 1" 8/ *0 @0 / / :MI HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH

PAGE 153

<(M@ :..:"!! C" 0#1TLIII // ( 8$ "E E(2" $ $ &:..:.' "$ "/ "& //' *C" /:: //1" "*C A" :M:

PAGE 154

&0'(" T:=C %::..: % //" !CC "&'! 2 B "4 $""! "/! 8 !! $# :M<

PAGE 155

( % 2" ($ 8"(" " !" 3P8PB $$ /*" /!4":..M:.., / :..M" //D%:..:1:..< 1" 51 0#"13@ 0// / :ML X

PAGE 156

*C0:IQ / *0"C (CC C #*"1% 1A)( /*//0 C" $!P P" //" "8 & //'C //* (( (# & :MM

PAGE 157

"CE"" #@:..< C"D :/"D/ (
PAGE 158

"!" & //' #" """+" $" C"C" "D$ BB "B !&/' #:..
PAGE 159

! 3;3 C# &34 //" CB ( 0! "/1#*#:..: "C $1 !( "& :..<'" "D :2 <2 L/"O!W M0""" = (( &#"' / A*(/

PAGE 160

22,#8&8>46IK1 &N@..
PAGE 161

* 3 //0" *)" !?G *" /*/ 2 /*/"8" C! ""B0:..< 2 &18"' !@:..L C$ "?%:..L 68 "B 3$ 3 / :M.

PAGE 162

" +" /*/ $" "3* 3"3 C"# 683' 4" /E+/"4 "*!*! /*/ B .<(.L # +"! /*/ :..<(.L:=,I "" /*/ )"D/(
PAGE 163

8" /*/ & '% !%0&:..<' P4! H"8!! (P "C "% &63,;2 0E"C """8 "& C" '!!D$ !CB B!"" B" :=:

PAGE 164

K!A&2J35J2#223383542 ? 1C &%:..L'3 4" G!?
PAGE 165

# /*/ :<"" "(( D1V0#E/+ 0 51015 5102" 3 $*"& "'B" 3 *""(( /*/ "BC C ""! ?*" /1#* :..L(.MK=II" "C:I=II C" B( C 0E"C *0!

PAGE 166

"1 /*/ B "" "& 3'%" "$* C "# 0 1" 68 " $ /" P P & /*/22:K:..M' :=M

PAGE 167

1 #/1+1"# /*// *+1"" "C // "" &'(" (" /*"/ "1 "C" "#"" 1"" ""3 #! "0:..M /*/ &' ? D :==

PAGE 168

4":..M:..,DN 20 8"" /" /*/ GM( 0 C" 8C @:..,$ !C" /1#*/1#* 0 /*/3"1G/ *" /+/ 6& #6'=I:&'L 3*# &! 3' :=,

PAGE 169

" Z C 2 "( Z 0 C"!+$ Z"""" ( /*/" ""!C X"" "C "!! C1" A"" 0"! #" *)"/ ( /*/?D :0" /*/Y Z8Y :=;

PAGE 170

<8 /*/C Y8 /*/ Y8Y L8" /*/ :.OW Y M8 /*/Y &+ /*/:
PAGE 171

Y1" Y"" 3*" /*/# /*/ ($ / /*/1 /*/ "& ;2 'P "P" % !&"' B0 */* B@M2E "&""' &/1#*' :=.

PAGE 172

("C /*/@3& *' /*/ "(P0" P /*/" "&T<=III' "% /*/ 5"!! /*/3*" /*/ "" "" CDP8E] !X-V 0( 683 C%! :,I

PAGE 173

/*/# P2" /*/P 2 C "0 "! ? *!? /B" !" (" 0E"(" /*/ /*/ /*/ &"' /*/E A5! PEE "/%! &34 ( &:..=M' #0A" /*/" :,:

PAGE 174

P8A YV!1* *8 P P 682 /*/1" /*/ B( */*Y PE /*/ E!P / C #""* )"/:1" "" "" P Y 0 !PP # *+1

PAGE 175

P $P& /*/MA:A.='8 Y BC /*"/C(! /(! /*/ +" 4 *" """ 4!" & /*/MA:A.=' 4 "! "( # ! :=C( #3&:.KL'(C :,L

PAGE 176

!( #" "!" /1#* 4 /*/" "/1#*4 TLIIII"C "D /! /*/""! !" " &8? "PPO /*/=ALA.=0Y 8" /*/ /1#*" @ :,M

PAGE 177

:..= ZC" Z /*/ 2' X X 45+1& /*/" V'( $"" B" "# "3"# "" %A !C ? "8 "!" C :,=

PAGE 178

3;2 /*/" 38' 68 !) 68 6835 8$ 683 ""8 /H" !" /*/CH8 "" (!" "" & /*/,A<=A.=' +( """ E /*/ 1 C+""/1#* :,,

PAGE 179

! /*/ /1#* "2 *"" !" C &#%0 "'0* 1 / 0"# #"B */1#* 0C "(" 5 *"/( "( "" "(( ?0E "P P&",A:,A.=' *+/

PAGE 180

2N /*//1 135/ /*/# $" ?D8 "!Y8 &' "&'% ( ? "CD C" !"%C":: C" B0"! / :,K

PAGE 181

*C C"BPP C !("# " / & C'C" 6 $*3 /0"" &#'+ "# "C #+B "EC A# $ :,.

PAGE 182

C"" "" 25 8& 'C" %""C "::III$" " 6835 # /" $ "# D/ 6G /*"" # """ "#" ( 6
PAGE 183

B // /*/ 68 "" !% "(( C" 68 #"C" 2 "! 8"" /*/ !"" "!"B C"" B! "" C" $ "" &' "2 "" :;:

PAGE 184

"&3+ 1(! DP" """ ""P C"" """" ( C"# /*/ "C )"" C" ""% ""!" 0E C"" ""C" P/P :;<

PAGE 185

328 ( /"" C"6L "" """ "( +( " /*/" 6MDP !" 68 "" -C" 68 E "" "C"( ( /*/ "C" " :;L

PAGE 186

PAGE 187

(" """" "&/1#* <' D8 89& !Y L#@B# 3868F # + @ 8"" (PP0 "C"D" 68 PP C 8 /*/ 135/$(! !4 /*/ "" "!( :;=

PAGE 188

1"!" !" // /*/" +" "C /*/ "( *( "C #+! &'$Y /*/ !" !&'" "" &2 C'% ""! $ +< 8" A :;,

PAGE 189

Y#135/" "! ""&"' ? +L #!8 "" Y /; :;;

PAGE 190

/+#*)2, /2%11/#1)#4#3G11#40/#1)10G?40461 135/G1" /*"/" /??"#+3&:..I' " !&;:' 1" P"P 4 /: ? +%E&:..<' $ "EEH ""( */)

PAGE 191

" &:;=' /0 ,29 4! /%$!& "$" $C C"135/ /*/"" (! "" C"&!'#! &'$ ( /*/135/& +/(" 135/"'+" & + ?5 @ :;.

PAGE 192

!B&!' +"E C E&:..I'" !D"$ "&" !' (! !)!$ B /*/P8"" (EPO",A:MA.=0 "" /*/ 135/ 8! !"! $$$ !! # /*/ *)"

PAGE 193

135/" $! /14!)" "! !" !" !E"!" ? /:B "#+!&:.K.' # !?" "%"%' &<=' ""! " PP&9F 9:..<;'!(" 4B" "$ :K:

PAGE 194

" "# (" """" "" #" "EOEW ("&+F%E:..<:,,' "" $P P&:,,'2! "135/ /*/ PP(P" &383 "P&6:.K.
PAGE 197

C""&8 "" /0?/ "' !!" P$P"" %" /135"" /*/" " 135/ "135 "" "$! /*/" / :K=

PAGE 198

"(" #" C""( # /*/"" #135/ /*/B"" / /*/ &8 /!" /*/ ""' &0 11"135/' &/:..<' 135 ( /*/ ""! "%! :K,

PAGE 199

/*/ ? !"C "C""&!' ( (" +"" /*/135/ $!"" /14!"/ #8&:.KL'" $(" !?""" 6! & '#!( "$"4 6" :K;

PAGE 200

8!(" B ""& '! 6"PP &8:.KLML:'! !!" "" &"' "#0 /*/! "&"' !1"P" EHEEC P&",A:,A.=' /*/135 / $"" "! :KK

PAGE 201

!#"( / "(+" !C" "$ 6 /*/ +"! A" B!" !"" !6 C"& '" C" #3&:.KL' 6DC" !#! !" /;+ *)7

PAGE 202

@$ ""$ ( ZC0 A" C"! ("" 135/"" & 12"!?'#135/ "C "&135/' (. " "$ /*/ /*/((! ($ &#$ !("" !=I:&'L %"("P PP((P' :.I

PAGE 203

135/ B& '! /*/ !( +"2 /*/! "!& ((' !(" & '!" 8 ""! 4 !( !(B"C ! !B % /*/P0 :.:

PAGE 204

"P&:AMA.=' 0"! & $ 2"6 !C" 2 #! Y #&:.K.' !P "$ !"P&:M<' "" A $ :.<

PAGE 205

#33* "&"'"! D & "C"' "$&/)% "' 135/ P /PPP 1 /*/E "4 $!" !"/)% "135/ DB0/ @B0#B1B+/ :.L

PAGE 206

)"0B01"#/ 011"B#*/B/* %@0B1)0 %/11 /*/D 0#1B#0+ 0B0 //1" 0B0 /// /B*"1 510B //0?/&$$' 3B/%/ 2/3B/0 #A*/+0"B //" !"$ "!! "" # 3 ( "" &"' A&" *7(

PAGE 207

$'B $ &1" 135/#10 1/' /*/ /*/B "" "" E"A "&"' 2.3 "" "135/( 0+11$(""" :..L1E :..M1/:..M 08//B! 1/" 8 A" /*/ /*0 / /1"0#*"1 :.=

PAGE 208

510#0+ 0% 510" """ /*/0 "" "" +"" "! 5## !&! ( /)%A "!! 1(C #:..L(.M1 510+ 135/1/ 88// # 135 135/51 0" 135/ :.,

PAGE 209

C/B C&" '! /;($ "135/1 "" 8 """ #10!5 10/ G/B1( 12"! ? #/)%135/ 81 135/&/' 1/B/ / /)%1/8#1$ 1/! +/E 1 3 8 /*// :.;

PAGE 210

"""" #C "8 &135/ /)%A 1 A "" "!( "! !(/ ( 4 (/)% ""# ("A :.K

PAGE 211

" "" /)% %!#8&:.K:' &/)%"' P !P&L.<'( # &135/'$& /*/'!( C !&135//' ( ("( /*/" E"# E (E"! 1 "E!"PE YPPYV *77

PAGE 212

!E!8 X""&",AKA.M' Z1"!G< /*/0! Z""!4 X : ZD ^8E"!0# XB18 XX8"Y Y+! ""Y88 "Y &*)"G<:,' (! "" 0E /*/DP "E"P&",A:,A.=' 135/ "/)%!( Z(12"!?# X /*/ "A Z""" 6 X 1""

PAGE 213

" %135/ C" "" PP!#! 1/ "! "( PE2?*HP8 !O" W/ !E !!&",AKA.M' 4"( ""/)% #!/)%A "# " $ &135/$ 12"!?!' 1"*

PAGE 214

+"!P 2MN &",AKA.M' /&:..<'! % /)% /)% !$ !" C" +"C"" $ +< D A! 3!&6 &:..I'2&:..<' 1"1

PAGE 215

$!"D "! '0" '2" P" "DP" ""!&2 :..<' '2" D8 "Y8 !Y 135/ /*/ (! """ B %1#* PC"
PAGE 216

P&:..I.' [ "B"!"# 4! "! 135/" B1 / -" /B" # / "4 )" !""" 1/"1/ ///&" "'"! (E/"1 / /P ( "($
PAGE 217

CP&+F%E:..<:,K') (( 3 */* &P //$!P O:..I.W'135/1/ &PE!EEPO# 1$135/1/:IA:LA.L0(+" ""(P8 !E( !P&!135/1 /:IA:LA.L' (" 3 135/EPPP/ "P&4!"' #"$ !" #" 135 /G1"P
PAGE 218

/P"&!"' P)"P @2P#" "135P&<'# P1 /135/ G1"!! 13 5/G1"P&<'%# 1$&1/:IA:LA.L'P4$ /P *1/ !/ "// %%+A P% 1/ C/P&
PAGE 219

&'" "P&M' /$ " 12"!? =(.*(.7(? P# ""3P# 83 *77! *77( 1/(" "" "8"!E =*7)7? P P& *"0? 4 135/E 08(+ /Y1P3 ( 1/1/ !P&"=A:IA.=' 8! Y# Y" !D 1"/

PAGE 220

""" "! ""! 3 C( 0"! !"1/ #" &'" "! / 135/1 / $ 1") HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH

PAGE 221

#" /*" /DP) ""H !!P&:..I<:'8 &A "'""" !"/ *:..M!! !" ?:..L" !0:..<(.L" "EP$ "(!P&:M' P4! P&:M' # """ C&
PAGE 222

"'%( ( & '5" """ 8135/1/$ /*/ .+"" .++ &! C ( +" /*/D 3 ( /*/ 8" 3/(B 1*"

PAGE 223

3E (E 3(" / / */* EE/ B #/E"@:..= /*/!!! ( "" 1 /$ 0 $ 1**

PAGE 224

&'" "C ? "! "" $ !8 !" # /&:..<'P#"" P&K'% "" "( #+'' "'' +L """ !6&:..I' "" !" 1*1

PAGE 225

(P1"! "P&'D &"' & '" C" /:D/" !EPP $ "! "!"! #! C""" !" "! "$" "!" 1*!

PAGE 226

&3' D Z"#" !&:..:' X Z 3"/'%%."'"".+ "'"+"'-"(%B &:=' %" Z X135/@2:..I D C C( PP""135 &135/"' %135/P" PDP EE&135/ "' Z! 135/# +<" // P /" Z"P&=' <:M

PAGE 227

!"P$C ("! P&,' /&#C (" "C( ( "! "( P!P(" "8! ""&/MO:
PAGE 228

$# /*/ 5O/W &",A:MA.=' !!E "?! "! !!& '"" # &3#& ""% "135/1/ "( ""&:IA:LA.L' 1" ""&" <:,

PAGE 229

!C"135/ "" P"P &",AKA.M' //" C &' C0E #"! ( E"!" !! !/ !&",A:=A.=' &! ""' "! "! 8" /*/ 135/%! "% 12"!?MA:MA.MD P <:;

PAGE 230

P!"! P" P"P! ""P&! "'#!& M/' # /4":..M& 'D /PP1/" !" (! & P#" P #"(" ( @:..= /' 3! /*/ "C B PP" <:K

PAGE 231

"" # (""/= "#08P1" "P&"=A:IA.=' " ?" C"! !P/"" P0 8&"=A:IA.='0 5E P+! YVP+! !YEP8YE P0 YEP YVE! &",A:,A.=' #"$ # <:.

PAGE 232

PAGE 233

+" Y /: 5 28 / !( +" CB (( ? "B ($ CB" !) 0"#/ :..:92"*! 11*

PAGE 234

# "&$ "" A'" C &' &=' $ D &:'" &<'*" &L'0" E &.' "8&:.K:' $! "" "" &93:..:.' 8 "! !" ?"93&:..:' ""! 111

PAGE 235

* P"P&._ "! P"P&.' (! "$ 135/[ "( !"! 1 /*/2?*E "" "!! &3! !! $((" !& 'CB" +"$ "P P&93:..:.'"# <
PAGE 236

13"P P"! " C"" PP" "4$A ( "" "! # "?" 135/ "AB" !( /*/!" C <
PAGE 237

'/" "( "C "! """ ? 135/"! !P#YP " %!A" !" "" "" ? 135/1/" % "" "" C&"'+" <<=

PAGE 238

"! # /*/ ""1 ":KC! B "%" " "" "" "! ("& !! '#3P0 P&",A:MA.=' <<,

PAGE 239

)"93&:..:' 135/& '" ( "" "D #EY4 V8( E"Y G!E (C E"! $J/ "! "E E EE!E !Y &41/:A:
PAGE 240

! 0 "! /1#* A P1Y 8YP # ""! # ""! "( 6O83 #!" "! !135/ !" ""135/ !"" 3"" 11)

PAGE 241

135/ /*/ 8 /*/"&('" "( ""P P"P EEC"P&"' """ $ /*/=I:&'L /*/ )"/1#* B PP&"'2([ "" P" P&+3:..I;:' <<.

PAGE 242

6E43 #($ !A "A! & $'! "& /*/' 135/" $+"" /*/ !($A """^ "" /; 135/" "1/#+ &08 A' ""!&,(K(.M' ""

  • PAGE 243

    ("6 !+
    PAGE 244

    H"HOW P &,:..:'#0P" /P &",A:,A.=' "" """ "8 "&"' "$ 3+"B" +" "" 8"" */*135/1/ # "+ /*/##4"# "! #" !"$
    PAGE 245

    683 ""! ?"" "B"" #0" /*/ "" ""! ""? ""( C" /*/"2E 2%" !! 1 @P P&' PP&"
    PAGE 246

    "" "" """ PP" """* " E"" "# "" 4 "!" C4 0"&,A:,A.='0P+ ( "" "YV8
    PAGE 247

    "" " !" ($( !/ "(("( !!! #"!"! !" *! %! EHHHHHHHH#"! 8 "1 "" ("!
    PAGE 248

    "^ E %" D ""(B B"! """"B X! B! (B "B # /*/P8E( E"( "P%( 2F + "Y 3!" "B !"D? /; X
    PAGE 249

    /+#*)2; /%4/31%41#40*3/#%41 &86 !" &35 #+3&:..I'P# O *3..(M=;W"P& ,.'%E&:..;'P &3 ": P"P&M,'+P/! "! &6< &M,' D &:'" !& /%!' &<'2A 24 &L'P* !!
    PAGE 250

    P&M,' 2&:..M' B &36833 &$'P "P&==L'P( ("C" P&===' P! P& :..
    PAGE 251

    8#3#82#@ "^" (!? ( "" !& """! # "" !" ? PP (!("
    PAGE 252

    " !"! D" !! "! !"! #" "/ @:..L@:..,
    PAGE 253

    $! ? ($($( C2$ "A" " !"" $!! $"# !3 " $!( ( 3" "" "E&'$ *1(*

    PAGE 254

    " ""!& 0($C D"!" !) " ($ /D+$? &86 !" """ !" ?" "!&
    PAGE 255

    !' &23 D " ? C #33* !!& "C"'" "" """ "&' "A"" #/,! "$ "%" "A
    PAGE 256

    B "" "" "! 8"A """! (( " 2"&P"PP" PO :.K.=IIW' D &3 "BBB B "A "" "!& '#+ &:..L'
    PAGE 257

    "' &=' /!P" P%(( "(" $"# A &!' C"" 4 R 135/E" $ /, "D &:')"A $ A
    PAGE 258

    6 &<'% .. 2 &L' ""B "" &M'" !"" #331 #! "(P P&/:..
    PAGE 259

    $"&6O:..IW2 O:..<0D "B"B" P""P&2F:..<::_ /, "E # +!&:.K,'0PE E "P &;,'" 8 P"P ""! )" !" 135/ /*/! "!! !" E"
    PAGE 260

    "!& '!! "4 +%E&:..<' !8" "" !! ""C"&:,L' "" ( $!#@9&:..M' @% 1P(" P&LM,'! "! G"" "3+
    PAGE 261

    $ !! "B" ""!E (? "" "! 3!C /*/ 2 !$" ($&P PO+ F%E:..<:,;W'? &' $&'P !P&%E:..;M;' E
    PAGE 262

    &' $%#&:..:' $# "!! $&=MI' 8"+< /," &:' !$A # 8 +:" "( "B E<=I X

    PAGE 263

    ! &<'! " !!* " !!! %"".%: '" #3&:..;'PH !P&ML' "! &L'! !$ ?" !" &M'" <=:

    PAGE 264

    !$?" 4 &='0 &,'! "( !"( #" !" /!E "# "( &;'0 <=<

    PAGE 265

    #33! !C" "" !# 135/ /*/" C"" "! "!!" "" &%:..:'( !!" "G" <=L

    PAGE 266

    "! #!" C" # "" &:'""! C" &<'6" !C! &L'?" !" "! &M'" <=M

    PAGE 267

    +"" !" !""(" #/ !# ?" "" "!$ !" !" <==

    PAGE 268

    1" "")& "' "" #"! "!C" 4(+$? "! /,! "! !* 10+

    PAGE 269

    8" P"P &93:..:.' ! "A /*/& "'! ""+"" "B ""* $" "# "" /*/ <=;

    PAGE 270

    ?135/ /*/ "! !"" 2!!" !" 3* $ ? 1 8 !! P !"P &3F*:.K;M;<'" ?3E&:..;'P(( <=K

    PAGE 271

    !"P&ML' /*/135/ G"" &:'0$ !"! &<'%" / !!B #8&:..M' #" &
    PAGE 272

    P&::KI' &M' "! "! !A! &='4!" &,'0"! "1 +&:..<' ( !( ( !"$&L<<' !3 &;' &K'/" !E"" 1+"

    PAGE 273

    8" EPEEP 6& PH"" P&1F/:..M
    PAGE 274

    ?" !( "#" !%D "!"& F+!:.K,B%E:..;'B" !!B !"!B 2"4!0" @ "!!" *! & !'8&:..M' 8 """ <,<

    PAGE 275

    (/ "$" "" !($ &<=:' /&:..<'D H" &::' /C &%E:..;'+3&:..L'" ( "OW &K' """ ?!" 3%E&:..;'" !" "" ?&:..<'!"" ( "" <,L

    PAGE 276

    &K' # $! &+F%E:..<:;K' <,M

    PAGE 277

    2)?)2)4/)1 #/31F/0&:.K<' '"0 /%D #"* #"$&:..:'1D # "%'!A>+"'#%" C+%"!(DE
    PAGE 278

    8+1!&)' C"/2#."4""'.7K; &LK=(MIK'%D%"* #3F6!/0&:..<'1"D" L$!!+''DM =M<'KI(K; #1F5&:..;' 68I 11F@9!&)'; %( "'""'";NNDA>/)+" "'. &:L<(:,I'1?/#@( $+."'K!"""''"#'"#"%'0 &:..L'80/D/83# #/&:..I'# $"" $"%# *"".%+"'II; L'M;K(=I< #@&:.KK' 6.(%"''"O!"A >5-"(#".0 80/C43 / #86F?/@&&:.KL'/"D $" $"%#*"".%(`=&M'=;,(=K; )&:.K.' >"#"'" G0F0 /#8*/ 1L&:.K.'%$D1 $"%#*"".%(-;H$F M.,(=:= @@3@F)22&:..:'3 +"'#J."4""'". MM'M,(== 0L&:..;')"" <,,

    PAGE 279

    "!@ 6!&)' >##(#"'( &<<;( ""%""."'"'" &L'4G!D? 86&:..<'D3<: *"".%IP M':,,(:,K 86F4 &:.K=' K"A>".#"2. ".0 4G!D+F2 9&:.;='$! $%"(!Q+"'EP <<.(MK A"-$%""."#'"#"%' &:..:'8 4// C0F80#&:.K:')$ $! $%"(! Q+"'E< L;K(L.= !(6@&:..='1 *3/(3F@ !(

    PAGE 280

    6&)' R"#%( &K;(:"2'.+/'/'%""--' 1? /#@( D 4#&:..<' "(E L$!!+'' ;,&=M<'KK(.< $/6F2 #&:.KL'"D +/'$%"(1A%MIL(M:I 0?0$8#8?F#&:..,' ""#".S>#-'".#('+0 4D"/* /#/!1&:..,' !+/2#+%."0 #5## /0@&:.K,'3!D $" $%""!;M &<'<,L(K, /28F6@a&:.K.'D @26@F8 +&)' C"/2".%"".% &<=;(<;.'4G!D 0!! 1+)

    PAGE 281

    /(3* !(6@&:..=' R"#%(0 4G!4GD/"* /E0?&:..M' >"#$%"&'0 80/C# /E0?&:..;' >"#$%"&'0 80/C# /0&:.K.' "-+"0 !/#D "/* /00F3/)&:..M' ''"/"('"AC4"('""%"2"##0 1?D @( /#&:..<'?" L-R.'"+"'#C+%"!(T6$F M(:L /"136" &:..<' 6%(.'"'(+..(%"'" "'"0 &1*4='**#D # / #&:..<#'3!D# !* #1*#/3 '"+"2A'"""0 &:..<'0"/%D /0) '"+2/"(+(0 &:..='0"/%D/ 0) /!@F @a&:..L' >-$%"A$'"(#++ #'(O!0 2*8!* <,.

    PAGE 282

    &/+*4*(8*4/**(I:::.L'#D" 14/+*"4* /!@'F/0&:.;.' Q+"85%"0 B +/ 3*&:.KK' $(">('%#' 4G!D21? 0$49&:..M'4/ 0$FG13&)' C"/2#9+"'"(" &=II(=:='%!/#D1#6)* 0$4@/F3G1 C"/2#9+"'"("0 %!/#D1#6)* 03(#&:.;I' R'""'4(-.0 )"3D 4"* 01&&:.KM'$D" C+%"'" `.;&='L=:(L,, 0&:.;<' O".+/''0 )/4@D *(+ )(12F3#&:..:' *"".% "0 4*!/#D1#6)* )8)&:.KL' C"/2#."4"%"".% 4 G!D0!! )9&:.K.' $"%#*"".%(-;H2 M'=L<(==I )?)F)a&:.,='$ C+%"'";M0 <:(L< <;I

    PAGE 283

    )F*1&:..L'#! D3 8 +/'$%"(-GIHF LMI(LM; ?)&:.KM' !"8'('''"/""%..(% +""("+#%('%L "'""JA"'."'(0 ##3 ?$3&:.;,' $'"%"A!"K+7"''&/.0 / 1/%D33* ?21&:..<'1D R%('%Q+"' ,&L'L(:L 628&:.KI#' $#.(%"''"# ;NMP *# 1*#1?/#(((((((((((((&:.KM'?D4!$ +/' $%"(-HH0 L':LM(:M=(((((((((((((&:.KK'"D#0"" 6.(%"'(!+%% :M(:;(((((((((((((&:..I'9"D# 286F*&)' !".#%""...(%"''"-2 &:<;(:=I'4G!D**(((((((((((((&:..<'1 23 D ":..I $%"(-# +/'$%"EE; L'
    PAGE 284

    !""K"' (%(-EG &:'.(:K 628F)&:..I' !".#%"".. .(%"''"-2 4G!D* 6!$@&:.;.' R5".-2"%%+' "+/#D1#6)* 6@8&:..I' J'"0 4G!D?* 6a #%E3@*F%Ea@@&:..I' 6%'%""" 6"D1 ? 6*F11&:..,'2 0?3&)' ""#".S>#-'".#('+0 4D"/* 6/&:.K.' "@a*&)' C"/2#+/'"%" &=;=(=KM'1?D @( 6!@&)'&:..;' >##(#"'(0 0D*+ !*/ 6 &:.K.' ''"/".#.%%.+#%+'" /'% 1?D@( 6 F+&:.K,'*$ >+"'#$'"("'$"' EEEF .=(::< <;<

    PAGE 285

    6)&:..:' Q+"'"(%%"".%"0 4*!/#D1#6)* 66*F63?&:..:'1 R+""'K"HI :':;(<< +*F20&:.KK'D )"" >O/"(-EPHF <<.("%'A%."# +""('(% 80/C )3A/0* +2#3?F @&:..L' %%+" L--"#.(/+ &3?/8!*' #D"3?/ <;L

    PAGE 286

    +!0F)&:.K;'!@6 + +*0F/#1&)' R%.. '"(" 3#3 +@6 0+* 1/#&)'&:.KK' R%..'"(" 3#3 )3&:.K,' *'2A('. ++.%'"''"/"0 &# 0?'80/(# 1RF9!@&)'&:..;' >%("'" "'";NNDA>/)+""'.0 1?/#@( 4F @&:..<'/ $Y R+""'K" =,&L'M<(MM X1F3 0&:..;'2 J+('++''
    PAGE 287

    5+(+%%"F0 4+"/ // 0"1*G" 9#F91&:..<' 6.".(."A$ ((-#"(+"/' 4G!D4 //*" 9*F15&:.K;'*!D 1$ F2$ &)' '-2 &<=(=.' /%D8"* 9@8&:.KM' $.""'"("+/'' # #3+/* 9@@&:..,' ''"/".%"2"##"'"A R"""0 *=;4 /#1*# #6# 9$@F*1&:..,' *"""2A!""(# +.'"#"%' 4G!4GD4/ /*"/"1*+ 9!#@3F+@6&:..:'*OW +"'#J."4""'".*"".% M&;':.(L: 9@*&:.K=' -"#'+ 4G!D?* 9$@F* V0 &:.K;' >'""''. 1 ?D@( 9!0F20a&:.;.'#! $! !'."!"'"
    PAGE 288

    9"1F20&:.K:' "(/+"+"" $%"'"'%0 4G!D** 38@3&:..;'D 1 +/'$%" (-GD;F L,(MM 3)30&:.;:' $8(-#' 4G!D #)"*/ G' )%F*?&:.;,' L-2#''("A$ (%%+#'+% 4G!D# 3+F1!6@3&:.K<'1$ D!/ $%"(!Q+"'ED <<;(
    PAGE 289

    /F26 &:.K.' ..9+"'"("0 4*!/#D1#6)* *8F &:..<' ''"/"A)"%"2 -2 1*4D#+8? $2&:.;K'"" 9+ ?81&)' 6."4""'' %"2. 0 6""'+"'#!'."!"' '< IF =K(,; /3&:..<' ('.("(."%+#"%' 80/C46"E# )@/F1/ &:.K,'# +"'#*"".%;E IF L=:(L,: 9@&:.K.' $ 0@&)' K""'A-('"' &<,;(.-"A$.+#"#." #"%'%#+""+%"( 8 0/D16"*% 2&:..<' >'"Q+2"2 !9D/* <;;

    PAGE 290

    F+#&:.KM' Q+"'"(""""'A$ +/2#-% 4*!/#D1#6)* @@F20a&:.;.'#" $ $%"(!Q+"'EH MIM(M<= 1#F/6&:.K.' !'#8..."4" K".-".#%"0 2#D#( 8*/ /a&:.KM' 6."4""''" 4G!D+ 1* /aF20a&:.K<'00a 2F0(#8&)' 6."4""'" &.(L:'##D1"* 2/F89a3&:..M'0""D 3"( +/'$%"(=M&L'
    PAGE 291

    %23&:..:'/"B +C4 1(%5"''#."4" 4G!D+F2 <;.

    PAGE 292

    *@a&)'&:.K.' C"/2#+/'"%"0 1 ?/#D@( *91#F%E 2&:.K.' >'#%+'8 "((..82+0 80/(/E 0? *0%F80/&:.;.' J."4""%'%" ++ &24A;.'8 ?26D *9+F9 @3&:.K.' K""J."4""'('% +"'; ;MF :I(:, *6@ &:.K;'/D3D# "@6+ @3 +*0F/H# 1&)' R%..'"(" &:=K(:,M' 3#D3 *"96&:.K<'$!" (! $"%#*"".%+"' EI MML(=: *"96F+ &:..:'(" $"""! +"'#+/'$%""">; HF L.:(M:; 2@6@F+8 &)'&:.K.' C"/2 ".%"".% 4G!D0!! 2*1F*@a&:.K='1 "$D" $"%#*"".%(-;P EF <;,(K, 22/&:..M'" 490$FG13&)' C"/2# 1)"

    PAGE 293

    9+"'"(" &=M=(==;'%!/#D1#6) 2*&:..<'?9!D "" L-R.'"+"'# C+%"!(TV &<'M(:, 20aF/a&:.K<'"0a 2F0#8&)' 6."4""'" &L<(=L'#3#1"* 20aF80#&)'&:.K<' 6."4""' "A""%'%"0 #3# 1"* 2 F$&:..I1'/D L1".% &4G!2# 6!':L(M 21@3&:.KM'4" +/'$%"(-< M,.(;, 26F*&:..<#'#! !* #1*#/a 21#&:..I'# L+.'# L$%" EII =.;(,IM 1?8&:.;K'$D "9+F?81&)' 6."4""''%"2. &LM=(L;I' "+/# 1#6)* 1)*

    PAGE 294

    1@ &:.,;' +'"-0 /%D* 1@+&:.K;'$D $"%#*"".%(-;EHF ;I<(;:L 13 &:.K.' )+"A"2.'# "("".0 4G!D# !&0' 1&:..I' !"%A>-'.# ''"/"0 4G!D2+ 1*&:..I' >#'0 4G!D 0A/ 1"&:.K;' 31"F 1&)' K"R5"%. '+( &::,(<.'#1"/ 0" 1"3F1 &)'&:.K;' K"AR5"%. R'+( #1"/0"(((((((((((((&:.K;'5! 31"F 1&)' K"AR5"%. '+( &.L(:IM'#1"/ 0" 11 =+'";NND0 0"/%D/ /E/ 15&:.;.' >!"K+7"''0 /%D* */ 1!08&:..:'$" C+%"'"HHT;F ::;.(::.=
    PAGE 295

    (((((((((((((&:..<'2! L-R.'"+"'#C+%" !(T6 IF <.(L= 1!3&:..:'! C+%" '"HH0MF KLL(K=, 1&:..,(:..;8'%#E L""'#'(L-"6+ :(M 1&:..;@:K&<'' "!$ :(<:M 1&:..;&:K&,'' :(M 12 1@+3+&:..<'/( D +"'# *"".%69+ ;&M'L<:(LLI 1/&:..I' $/"#(" 0"/%D# 1$ a&:.;<' 6""#"'."4" !+##'2+ E1"4G! 1 12@&)'&:.KM' +/'"%"A" &L' #D+/ !".'"#'"&#"%'"'A""##("%% %:..I 1#F/@&:..M'6D# ""4/0$ 1!0 3&)' C"/2# 9+"'"(" &<;L(
    PAGE 296

    * @&:..:' >+'+0 1%1#*" >#++#'0 &:..<'/?/ 3#/#D0"3*!? 28&:..=' $+'"0 1?D@( @0&:.,;' J."4""0 4G!D6( +3 4@5&:.K:'4!$*/4 8)1!&)' C"/2#."4""'. &53< <<=("#%""''"0 4G!D@8F1 55#+F8!6&:.KM' $ $%"(!Q+"' EN =.K(,<: 510# ?2+6&:..I'" +"'#% b1&L'<.(M= 8*&:..:'E LR.'"+"'#C+%"!(T &L'<(; 833&:.,;'" $%"(!Q+"';E L.,(M:. )""%"#'"#"%' &:..:'80/C 43/
    PAGE 297

    82&:..:' *"2.%-2#'0 4G!D// 8F+2&:..I' %%+8/"#"%'+" +"."%A!%.'%.-S 4G!D 4//*" 80&:.K:'$D" +"'#C.R+"GE &:':(<,(((((((((((((&:.K<'0a2F0# 8 6."4""'" &.;(:<:'#3# 1"*(((((((((((((&:.K.'8Y $"% #*"".%(
    PAGE 298

    1"#%A$%('".".#$%"& '0 &:..,'4G!4GD//4 G! G29&:.K.' "+" (0F0 4*!/#D 1#6)* G!6 $0 &:.K.' K"."4" )/4@( *+3 74&:.;I'*D#!" Xc47&)' -."4" &<<:(<,:' 4"4D5"*(((((((((((((&)'&:.;I' -."4" 4"4D 5"* 7#@d&:.K=' >+#.+"."4" 1 ?/#D@( 7$6&:.KI'$ $%"(! Q+"'EG ;<(KK