Edith Cowan University
Research Online
Theses: Doctorates and Masters
Theses
2017
Exploring ‘People’ as the key element in enterprise architecture
implementation: A Critical Realist Perspective
Edi Triono Nuryatno
Edith Cowan University
Follow this and additional works at: https://ro.ecu.edu.au/theses
Part of the Management Information Systems Commons
Recommended Citation
Nuryatno, E. T. (2017). Exploring ‘People’ as the key element in enterprise architecture implementation: A
Critical Realist Perspective. https://ro.ecu.edu.au/theses/1988
This Thesis is posted at Research Online.
https://ro.ecu.edu.au/theses/1988
Edith Cowan University
Copyright Warning
You may print or download ONE copy of this document for the purpose
of your own research or study.
The University does not authorize you to copy, communicate or
otherwise make available electronically to any other person any
copyright material contained on this site.
You are reminded of the following:
Copyright owners are entitled to take legal action against persons
who infringe their copyright.
A reproduction of material that is protected by copyright may be a
copyright infringement. Where the reproduction of such material is
done without attribution of authorship, with false attribution of
authorship or the authorship is treated in a derogatory manner,
this may be a breach of the author’s moral rights contained in Part
IX of the Copyright Act 1968 (Cth).
Courts have the power to impose a wide range of civil and criminal
sanctions for infringement of copyright, infringement of moral
rights and other offences under the Copyright Act 1968 (Cth).
Higher penalties may apply, and higher damages may be awarded,
for offences and infringements involving the conversion of material
into digital or electronic form.
Exploring ‘People’ as the Key Element in
Enterprise Architecture Implementation:
A Critical Realist Perspective
This thesis is presented for the degree of
Doctor of Philosophy
Edi Triono Nuryatno
Edith Cowan University
School of Business and Law
2017
USE OF THESIS
The Use of Thesis statement is not included in this version of the thesis.
ABSTRACT
TOGAF (2009) describes the purpose of Enterprise Architecture (EA) is to optimise enterprisewide systems - the often-fragmented legacy of data processes (both manual and automated)
- into an integrated environment that is responsive to change and supports the delivery of
the business strategy (The Open Group Architecture Framework [TOGAF], 2009). However,
for a number of reasons organisations still have difficulties establishing an effective EA (Raadt
& Vliet, 2008; Gartner, 2009; and Janssen & Klievink, 2012, among others) and various reports
suggest up to two thirds of implementations do not fulfil expectations (Roeleven, 2010). Being
organisation wide with a strong governance element EA has significant social implications and
social dependence, yet many implementations wrongly treat EA as solely a technical program.
This thesis argues that the lack of focus on the ‘people’ element of EA could be the reason
why many organisations still struggle with EA implementation. Recognising the importance of
people in EA implementation requires acceptance of implementation as a social program,
heavily influenced by the structural and cultural systems surrounding the architecture.
In order to address the need for greater recognition of the role of people and the social
aspects of EA implementation, this thesis adopts critical realism (CR) and its most recognised
methodology, the morphogenetic approach (MA). Realism emphasises ontology and strongly
argues that ontology, methodology and epistemology are closely linked – as Fleetwood (2005,
p. 197) suggests, ontology matters: “The way we think the world is (ontology) influences:
what we think can be known about it (epistemology); how we think it can be investigated
(methodology and research techniques); the kinds of theories we think can be constructed
about it; and the political and policy stances we are prepared to take”. In order to examine
the social implications of technology implementation it makes sense to adopt a wellrecognized social theory like critical realism.
This social realist approach proposes an analytical separation between structure, culture and
agency (people) in order to examine their interactions over time. The MA suggests three
important cycles – structural conditioning, social interaction and structural elaboration that
provide a platform for examining possible change. Archer also importantly suggests that the
emergent properties of collectivities and individuals differ. Such a model has clear value for
examining the “people” acceptance of the new impositions and opportunities provided by the
EA implementation. It acknowledges the sociocultural consequences of interactions between
the structure and the culture to provide particular situational logics that direct, but do not
determine the actions of people.
The MA emphasises strongly the role of time in situation examination suggesting that
structure and culture predate subsequent actions by involved agents. The thesis describes
particular situational logics or mechanisms emanating from the interaction between
structural and cultural systems that encourage particular behaviours in response to the EA
program. These actions are then further examined in the sequence of MA cycles. Since
mechanisms are only effective if people adopt them or not, another important element in this
ii
study is the part played by “reflexivity”. Reflexivity highlights the linkage between people
concerns, projects and practices as people act in order to promote their concerns, and form
projects to advance or to protect what they care about most. Reflexivity is an important
mechanism for explaining how people’s ultimate concerns impact on their approach to the
impositions of EA.
An Australian university implementing EA (termed UX for anonymity) has been used as a case
study in this research – this fortuitous timing allowed a careful and detailed examination of
implementation over a 3-year period from initial rollout to ultimate acceptance. The study
describes the challenging environment of university implementation where “academic
freedom” is paramount and individual and group autonomies are threatened by EA – the
study presents the important mechanisms and situational logics that direct people’s actions
within the complex social context of a university. Semi-structured in-depth interviews were
used as the primary method of data collection across UX stakeholders. A range of interviews
were held throughout the study period with the university IT Governance Committee, the
University Architecture Board, the CIO, and the Enterprise Business Group, as well as
individual end-users such as teaching staff, researchers, students, and administrative staff of
the faculties, schools and service centres. The MA provided a basic structure for unravelling
the social complexity and helped guide the interview questions to identify the generative
mechanisms hidden in the real domain, and to highlight the conditions that encourage
individual and collective acceptance of EA practices. The reflexivity indicator developed by
Archer –ICONI– is used throughout to explain how personal projects are formed and how they
mediate the exercise of structural/cultural constraints and enablement within EA
implementation. Passive participation in regular EA implementation meetings at UX was also
important and useful to unearth possible perceived causal possibilities emanating from within
the program itself and evident within the social context of implementation.
Underpinned by a critical realist perspective, the thesis demonstrates that the MA is a
powerful analytical tool to uncover the hidden mechanisms (the situational logics of
structures and cultures) and social responses that enable success of EA implementation. The
research examines the particular situational logics evident within the University under study
and how these provide opportunities and constraints to the acceptance of EA over time.
Equally important was reflexivity theory in attaining knowledge and understanding about
what it is about people’s internal relations that makes EA implementation succeed.
This thesis offers organisations a means to focus on the deeper issues of EA implementation
programs by understanding the social complexity surrounding the architecture. The
recognition of people as a key element in EA implementation provides a useful explanation
of how the key stakeholders (and their power, influence and interests) may constrain and
enable EA implementation. By including reflexivity as an important mechanism, organisations
will be in a better position to understand the role of people and their interactions with preexisting structures and cultures operating over different time periods – reflexivity suggesting
that “people” always have the possibility to do otherwise than expected, largely dependent
on their personal history and their current personal projects and ultimate concerns.
iii
DECLARATION
I certify that this thesis does not, to the best of my knowledge and belief:
i.
incorporate without acknowledgment any material previously submitted for a
degree or diploma in any institution of higher education;
ii.
contain any material previously published or written by another person except
where due reference is made in the text of this thesis; or
iii.
contain any defamatory material
31 January 2017
.................................
...........................
Edi Triono Nuryatno
Date
iv
ACKNOWLEDGMENTS
First and above all, I praise God, the almighty for providing me this opportunity and granting
me the capability to proceed successfully in this amazing journey. This thesis could not have
been completed without the support of so many people over the years. I would therefore like
to offer my sincere gratitude and appreciation to all of them.
To my principal supervisor, Dr Philip Dobson, thank you for your warm encouragement,
critical comments, corrections to the thesis, and especially for your thoughtful guidance
during the writing process. You were there for me every step of the way and I wholeheartedly
appreciate everything you’ve done for me.
To my associate supervisor, Dr Denise Gengatharen, thank you for the insightful discussion
and valuable advice, and especially for being so patient and helping me to improve. Your
perseverance, integrity and humanist nature are just a few of your qualities that continue to
inspire me.
To my wife, Anggraini Triono, children Satria Nuryatno and Bintang Nuryatno, thank you for
being such a wonderful wife and children during my PhD journey. To my lovely wife, you are
always my number one supporter. Thanks for listening to all my theoretical mumbo jumbo
every night and being the foundation of our family. To my children, thank you for your love
and understanding. You are a true gift from God.
To Bev Lurie, Research Support Coordinator and my thesis proofreader, we met during one
of the most difficult times of my life and you were there when I needed words of
encouragement. You are a blessing in my life. Thanks for all your support and useful advice
over the years.
I gratefully acknowledge the funding received towards my PhD from ECU’s Postgraduate
Research Scholarship-International (ECUPRS International). This scholarship has assisted me
immensely with university fees, expenses of living away from home, and provided me with
financial security. Receiving this scholarship has not only given me pride in my research, but
it has also given me the motivation to do well in the future and help me to accomplish my
v
current goal, move on to the next and eventually become a successful professional in my field
of study.
To my good friends, Jim Neilson, Danny Yonas, and Laurie Wozniuk – Jim, you are an icon of
integrity, kindness and a great philanthropist. My success is due to your support and
friendship. I really appreciate you so much and value everything I have learned from you. To
Danny, I don't know who I would be without a friend like you. You are someone who I can
always depend on. Thank you for letting me be myself. To Laurie - thanks for opening my eyes
to new stages of opportunity and strength. I have learned from you the value of tolerance,
patience and trust in life.
To Dr Bradly Augustson, Kerin Young, and all the nurses and staff at the Haematology
Department, Sir Charles Gairdner Hospital, cancer as a disease is ruthless, merciless,
frustrating and very tough. But a good doctor and caring nurses like you make this illness a
little less rough. Thanks for healing me from this disease so I could get back to the research I
love and finish my thesis.
To the organisation under study and people who willingly gave their time to answer my
interview questions - this thesis would not have been completed without your participation.
Thank you very much.
To my critical realist colleague, Layla, I have had so much fun learning with you. Wishing you
the very best of luck with your thesis!
Last but not least, a special thanks to my mother, Yetti Nuryatno, to my mother-in-law, Asiati,
and to my brother and sister for supporting me spiritually throughout the writing of this thesis
and my life in general.
This thesis is dedicated in loving memory to my father, Bambang Tri Adi Nuryatno. You never
saw me realise my dream, you may not be around in person, but your spirit will live on forever
in my heart. Your memory will always be treasured; you are loved beyond words and missed
beyond measure.
Perth, 31 January 2017
Edi Triono Nuryatno
vi
TABLE OF CONTENTS
CHAPTER ONE: INTRODUCTION............................................................................................ 1
1.1
Background ............................................................................................................. 1
1.2
The Gap in Enterprise Architecture Research ........................................................... 9
1.3
Significance of the Research .................................................................................. 12
1.4
Research Questions ............................................................................................... 16
1.5
Main Contributions of the Thesis ........................................................................... 17
1.6
Thesis Outline ........................................................................................................ 20
1.7
Summary ............................................................................................................... 22
CHAPTER TWO: ENTERPRISE ARCHITECTURE AND THE LITERATURE REVIEW .................... 23
2.1
Overview ............................................................................................................... 23
2.2
Enterprise Architecture Discipline.......................................................................... 25
2.3
Understanding Industry Perspectives on Enterprise Architecture Failure ............... 31
2.4
Enterprise Architecture Implementation Challenge and Governance ..................... 34
2.5
Enterprise Architecture Implementation in the Higher Education Sector ............... 39
2.6
Enterprise Architecture Adoption: Technical Capability versus Social
Intervention .......................................................................................................... 43
2.7
A Contingency View of People as the Key Element in EA Implementation .............. 44
2.8
TOGAF and the People Element ............................................................................. 47
2.9
Summary ............................................................................................................... 56
CHAPTER THREE: WHY CRITICAL REALISM? AN EXAMINATION OF ENTERPRISE
ARCHITECTURE FROM A SOCIAL REALIST PERSPECTIVES .................................................... 59
3.1
Overview ............................................................................................................... 59
vii
3.2
A Social Realist Perspective in Examining Enterprise Architecture ......................... 60
3.2.1
Bhaskar’s philosophy of critical realism........................................................... 61
3.2.2
The morphogenetic approach and reflexivity: Theories of critical realism ....... 63
3.2.3
Bunge’s realist philosophy of emergent systemism......................................... 65
3.2.4
Analytical sociology ........................................................................................ 66
3.2.5
The different under-labouring roles of social realism for enterprise
architecture examination................................................................................ 68
3.3
The Important Role of Mechanisms in Social Explanation ...................................... 70
3.3.1
Social mechanisms from various social realist perspectives ............................ 70
3.3.2
Generative mechanisms in social explanation under critical realist
perspective ..................................................................................................... 73
3.3.3
The role of ‘program mechanism’ from a critical realist perspective ............... 75
3.3.4
The role of ‘abduction’ in explaining the non-observable events
and non-events generated by mechanisms ..................................................... 76
3.4
Summary ............................................................................................................... 77
CHAPTER FOUR: THEORETICAL APPROACH –EXAMINING ENTERPRISE ARCHITECTURE
FROM A CRITICAL REALIST PERSPECTIVE ............................................................................ 79
4.1
Overview ............................................................................................................... 79
4.2
A Critical Realist Perspective in Information Systems Research .............................. 79
4.3
Morphogenetic Approach: A Social Investigation in Social Mechanisms ................ 83
4.3.1
The emergent properties: Structural – cultural formation, and
agency influence ............................................................................................. 84
4.3.2
Situational logics and enterprise architecture pathways ................................. 92
4.3.3
Possible contextual social mechanisms identified from the literature
review .......................................................................................................... 100
4.4
A Central Role for Reflexivity in the Implementation Program ............................. 101
viii
4.5
Contextual Influences: The Dynamic of Complex Social Program
Surrounding the Architecture .............................................................................. 108
4.6
Summary: Research Question Identified from Literature ..................................... 111
CHAPTER FIVE: RESEARCH APPROACH ............................................................................. 113
5.1
Overview ............................................................................................................. 113
5.2
The Underlying Philosophy and its Methodology Framework .............................. 113
5.2.1
The underlying philosophy ............................................................................ 113
5.2.2
The methodological framework .................................................................... 117
5.3
Explanatory Context of Research Questions ........................................................ 125
5.4
Research Design .................................................................................................. 129
5.4.1
Objective of the research .............................................................................. 129
5.4.2
Case selection ............................................................................................... 129
5.4.3
The case study .............................................................................................. 131
5.4.4
The target populations and participants ....................................................... 132
5.4.5
Materials, validity and reliability ................................................................... 133
5.4.6
Data collection procedure: Interview and observation protocol ................... 135
5.4.7
Data analysis and its methods....................................................................... 144
5.5
Limitations........................................................................................................... 163
5.6
Ethical Considerations ......................................................................................... 164
5.7
Summary ............................................................................................................. 166
CHAPTER SIX: CASE STUDY, DATA ANALYSIS AND RESEARCH FINDINGS .......................... 168
6.1
Overview ............................................................................................................. 168
6.2
Domain of Empirical ............................................................................................ 170
6.3
Domain of Real .................................................................................................... 174
ix
6.3.1
Analytical resolution stage ............................................................................ 175
6.3.2
Abduction stage............................................................................................ 209
6.3.3
Retroduction stage ....................................................................................... 213
6.3.4
‘Comparison with different theories/abstractions’ stage .............................. 221
6.4
Domain of Actual ................................................................................................. 231
6.5
Summary ............................................................................................................. 233
CHAPTER SEVEN: DISCUSSION .......................................................................................... 235
7.1
Overview ............................................................................................................. 235
7.2
Research Questions Revisited .............................................................................. 235
7.2.1
Sub-research question number 1: Macro – micro context of cultural
and structural system ................................................................................... 236
7.2.2
Sub-research questions number 2 and 3: Micro – micro context of
socio-cultural and social interaction ............................................................. 238
7.2.3
Sub-research questions number 4: Micro – macro context of
elaboration ................................................................................................... 243
7.2.4
Main research question ................................................................................ 245
7.3
Research Implications and Contributions ............................................................. 246
7.3.1
Methodological framework implications ....................................................... 247
7.3.2
Contributions to information systems theory ............................................... 254
7.3.3
Recommendations for enterprise architecture practice ................................ 257
7.4
Limitations and Future Research Directions ......................................................... 258
7.5
Summary ............................................................................................................. 260
CHAPTER 8: CONCLUSION................................................................................................. 261
REFERENCES ..................................................................................................................... 267
x
LIST OF TABLES
Table 2.1 EA Definitions in EA Contexts .............................................................................. 28
Table 2.2 Ten Common Failures in EA Implementation ....................................................... 32
Table 2.3 Main Challenges of EA Implementation in the Higher Education Sector
(adapted from Anderson and Backhouse (2009, p. 9) .......................................... 42
Table 2.4 ‘People’ Roles in EA Compliance within EA Implementation
(adapted from TOGAF, 2011, p. 565) .................................................................. 49
Table 2.5 ‘People’ Interaction and Action in EA Compliance during EA Implementation
(adapted from TOGAF, 2011, p. 566) ................................................................... 50
Table 2.6 Potential Success Program Mechanisms Identified from EA Literature Review .... 57
Table 3.1 Different under-labouring roles of social realist for EA examination
(Summarised around social realist conceptions of social reality and
Archer (2015) investigations on the processes of change) ................................... 68
Table 4.1 Various scope of the philosophy of science and meta-theory (adapted
form Fleetwood, 2013, p. 11) .............................................................................. 82
Table 4.2 The Cultural system and socio-cultural level within the morphogenetic/static
view of agency (adapted from Archer, 1996, p. 134) ........................................... 90
Table 4.3 Possible social mechanisms identified in the literature...................................... 101
Table 4.4 Sub-research questions and sub-research objectives ........................................ 112
Table 5.1 Validity in qualitative research (adapted from Zachariadis et al., 2013,
p. 860) ............................................................................................................... 134
Table 5.2 Basic conditions of people reflexivity, generative mechanisms and
situational logics (adapted from Archer, 1995; 1996; 2008; 2010a;
2013a; 2015) ..................................................................................................... 158
Table 6.1 Relation between particular EA event and EA context ....................................... 209
Table 6.2 EA management reflexivity ............................................................................... 215
xi
Table 6.3 EA end-users reflexivity ..................................................................................... 216
Table 6.4 Mapping the ‘program mechanisms’ to each of CR theories by which
social mechanisms are discovered ..................................................................... 224
Table 6.5 Events and non-events in the domain of actual ................................................. 232
xii
LIST OF FIGURES
Figure 1.1: Thesis structure .............................................................................................. 22
Figure 2.1: The Systems of Governance Mechanisms (adapted from Ross et al.,
2006, p. 120)................................................................................................... 37
Figure 2.2: Types of Linking Mechanisms (adapted from Ross et al., 2006, p. 128) ........... 37
Figure 2.3: Levels of Architecture Conformance (adapted from TOGAF, 2011, p. 560)...... 48
Figure 3.1: Morphogenetic/Morphostactic Approach (adapted from Archer, 1995,
p. 193) ............................................................................................................ 64
Figure 3.2: A Typology of Social Mechanisms (adapted from Hedström and Ylikoski,
2010, p. 59) .................................................................................................... 66
Figure 4.1: The three overlapping domains of reality in CR ontology (adapted from
Mingers, 2004, p. 94) ...................................................................................... 80
Figure 4.2: Analytical dualism in social theory: A stratified model of social structure
involving SEPs, CEPs and PEPs in EA implementation (adapted from
Archer, 1995, p. 190) ...................................................................................... 86
Figure 4.3: The morphogenetic/static of structure (adapted from Archer, 1995,
p. 193) ............................................................................................................ 88
Figure 4.4: The morphogenetic/static of culture (adapted from Archer, 1995, p. 193) ..... 88
Figure 4.5: The morphogenetic/static of agency (adapted from Archer, 1995, p. 194) ..... 88
Figure 4.6: Morphogenetic approach in explanatory study (adapted from
Fleetwood, 2013) ............................................................................................ 92
Figure 4.7: Morphostatis vs morphogenesis: Situational logics in social and system
integration in EA implementation (adapted from Archer, 1995, pp. 218
and 295) ......................................................................................................... 93
Figure 4.8: Situational Logics at different strata (adapted from Archer, 1995 p. 303) ....... 96
xiii
Figure 4.9: The analytical histories of emergence: Relationship between first- and
second order emergent properties, generative mechanisms, and
MA outcomes (adapted from Archer, 1995; Horrocks, 2009) .......................... 99
Figure 4.10: How reflexivity works in the basic morphogenetic sequence (adapted
from Archer, 2010a, p. 275) .......................................................................... 105
Figure 4.11: Morphostatis: Contextual continuity – low reflexivity (adapted from
Archer, 2010a, p. 281) .................................................................................. 106
Figure 4.12: Morphogenesis: Contextual discontinuity – predominant reflexive action
(adapted from Archer, 2010a, p. 284) ........................................................... 107
Figure 4.13: Picturing EA from the dynamic of social complex program surrounding
the architecture ............................................................................................ 110
Figure 5.1: Mapping the EA implementation on the critical realist ontology ................... 115
Figure 5.2: Mapping Sayer’s causal explanatory analyses of concrete and abstract
activities on the critical realist ontology ........................................................ 116
Figure 5.3: Research methodology framework (adapted from Danermark et al.,
2002) ............................................................................................................ 119
Figure 5.4: Mapping the morphogenetic approach on the critical realist ontology ......... 122
Figure 5.5: Mapping the methodology framework on the critical realist ontology .......... 124
Figure 5.6: The schematic diagram of research questions: The interplay within and
between the cycles of social change in morphogenetic approach
(adapted from Archer, 1995, pp. 193, 218, 295, and 303) ............................. 127
Figure 5.7: Mapping the empirical facts of EA events ..................................................... 145
Figure 5.8: The analytical histories of emergence ........................................................... 145
Figure 5.9: Second order emergent properties of MA cycles .......................................... 146
Figure 5.10: The interplay between and within structures, cultures and agency .............. 148
Figure 5.11: ‘Abduction’ process ...................................................................................... 151
Figure 6.1: The ontological map ..................................................................................... 168
Figure 6.2: The University cross-organisational architecture governance ....................... 174
xiv
Figure 6.3: The analytical resolution stage ..................................................................... 175
Figure 6.4: The analytical histories of emergence: Relationship between first- and
second order emergent properties, generative mechanisms,
situational logics, and MA outcomes (adapted from Archer, 1995) ............... 205
Figure 6.5: Reflexive mediation of mechanisms .............................................................. 218
Figure 6.6: Research findings: The relation of three overlapping CR ontological
domains in University ‘X’ EA implementation ............................................... 233
xv
CHAPTER ONE: INTRODUCTION
1.1 Background
The rapid increase in the use of information-related technology has changed the way in which
business is done in almost all organisations today. Technology-supported information and
communication has impacted organisations in numerous ways, particularly the way we do
business. The primary benefit of information and communication within organisations is to
keep workers connected. Therefore, to ensure that employees are productive, information
and collaboration technologies need to be in place and technology must support
collaboration. This has led to the use of Information Systems (IS) and Information Technology
(IT) as vital services for organisations who are increasingly looking towards the application of
IS and IT to not only underpin existing business operations and extend organisational
boundaries, but also to provide a strategic advantage by facilitating problem solving,
increasing productivity, quality and speed, improving customer service, and enhancing
information, communication and collaboration (Ward & Peppard, 2002; Turban, Leidner,
McLean, & Wetherbe, 2008).
From a technical point of view, enterprise architecture (EA) addresses IT demands by
providing the framework for technology, essential hardware, software and the
telecommunications network. From an IS standpoint, EA describes organisational plans,
visions, objectives and problems, and the information required to support organisational IS
goals in response to constantly changing needs in the business environment. The purpose of
EA is to optimise enterprise-wide systems, the often-fragmented legacy of data processes
(both manual and automated), into an integrated environment that is responsive to change
and supports the delivery of the business strategy (The Open Group Architecture Framework
[TOGAF], 2009). For this reason, many organisations use EA as part of their IT and IS
management along with their planning activities, as EA plays an important role in strategic
planning, alignment and prioritisation of IT and IS goals with the rest of the organisation (Ross,
Weill & Robertson, 2006). Such organisations view the right decisions as those driven or
guided by EA, and the right results as the use of EA in ensuring projects improve the bottom
line impact of IT and IS (Benson, Bugnitz & Walton, 2004).
1
Hence, EA can be seen as the strategic context for the evolution of organisational Information
Systems in response to ever-changing business environments. This was identified by the
JISC1.1 (2008) as:
Enterprise Architecture (EA) is about enabling organisations to adapt to
change by defining, in a generic way, how their business processes work in
tandem with their Information and Communications Technology (ICT)
systems. With this clarity of purpose, organisations are able to re-configure
or replace their systems with a clear understanding of how these changes
might impact on business processes across the organisation.
In describing the various possible interactions of the research components in
multiple ways, this study used text boxes to represent the emergent nature of the
research process. The aim of text boxes is to eliminate unnecessary confusion of
abbreviations, and terminology complexity within the IS-Social domain under
study.
Box 1.1: JISC
JISC is a Joint Information Systems Committee in United Kingdom, nondepartmental public body whose role is to support post-16 and higher education,
and research by providing leadership in the use of information and
communications technology in learning, teaching, research and administration
(http://jisc.ac.uk/)
Nevertheless, it is not simple for organisations to adapt to change triggered by EA programs,
since they are not only driven by forces outside the organisation to remain competitive and
customer focused, but also by forces inside the organisation, for example, the influence of
often unknown relationships between the social structural and cultural dimensions, as well
as key individuals and their functions. These social circumstances, which arise from actions
dictated by the interests and needs of key individuals, are likely to impact the way EA is
implemented. As identified by Lankhorst (2009):
2
…stakeholders will be influenced by their particular interest in the observed
enterprise, i.e., their concerns. Note that stakeholders, as well as their
concerns, may be regarded at an aggregated as well as at an individual level.
For example, a single business manager conceiving an information system is
a stakeholder. The collective business management, however, can also be
seen as a stakeholder of the information system. Yet concerns are not the
only factors that influence a stakeholder’s conception of a domain. Another
important factor is the preconceptions a stakeholder may harbour as they
are brought forward by his or her social, cultural, educational, and
professional background (pp. 55-56).
Over the years, a number of formalised EA frameworks and methodologies have been
developed to describe a detailed model of architectural work (JISC, 2008; Lange & Mendling,
2011) that can be used to develop, design and implement an EA program. According to
Urbaczewski and Mrdalj (2006, p. 18) an EA framework can describe “the underlying
infrastructure, thus providing the groundwork for the hardware, software, and networks to
work together”, [as EA] “relates organizational mission, goals and objectives to work
processes and to the technical or IT infrastructure required to execute them”. However, these
frameworks and methodologies tend to emphasise the technical aspects of implementation
and neglect the critical role of key individuals (Nuryatno & Dobson, 2016). It has been
suggested that the lack of focus on people aspects of EA could be the reason why
organisations still struggle with implementation.
As Bente, Bombosch and Langade (2012) suggested, EA “…deals with social elements such as
collaborative business processes, organisational leadership, political dynamics and work
culture...” (p. 36), and thus requires careful examination of the role of people as “the people
element brings complex behavioural attributes into the functioning of an enterprise…” (p. 35).
Such a perspective is the core of Janssen’s (2012) view of the socio-political dimension in his
description of EA as “…a means to inform, guide, direct, and constrain the decisions taken by
human beings within organizations” (p. 25). This thesis emphasises Bente et al. (2012) and
Janssen’s (2012) definition of EA.
3
In order to address the need for greater recognition of the role of people and the social
aspects of EA implementation, this thesis adopted critical realism and its most recognised
methodology, the morphogenetic approach (Archer, 1995; 1996; 2013a; 2015), as an
appropriate tool to research the topic. Critical realism distinguishes between three
overlapping ontological domains in the social world: the empirical, the actual, and the real.
As explained by Blom and Morén (2011):
The domain of the empirical consists of what we experience, directly or
indirectly. This domain is distinct from the domain of the actual where
events happen whether we experience them or not, because what happens
in the world is not the same as that which is observed. This domain is, in
turn, different from the domain of the real, where we also find the
mechanisms that can produce events in the world (p. 62).
According to Bhaskar (Searle, 1995, p. 25) critical realism proposes that events or phenomena
should not be the core focus of research. Rather the focus should be on:
…the structures and mechanisms that generate phenomena. These objects
are neither phenomena (empiricism) nor human constructs imposed upon
the phenomena (idealism), but real structures which endure and operate
independently of our knowledge, our experience and the conditions which
allow us access to them.
Critical realism not only focuses on a specific event observed, it also looks at what the event
says about the underlying causal relationships (or social mechanisms) that are enduring and
beyond the common experience (the empirical domain) (Mingers, 2004). Underpinned by a
critical realist perspective, Archer’s morphogenetic approach proposes an analytical
separation between structure, culture and people (agency) in order to examine their
interactions over time. It acknowledges the sociocultural consequences of interactions
between the structure and the culture to provide particular situational logics that direct, but
do not determine the actions of people (agents).
Although the morphogenetic approach has been widely accepted as an explanatory
framework for IS studies, there are still perceptions that the potential of this approach has
4
remained largely unrealised. However, a study conducted by Horrocks (2009) on the use of
the morphogenetic approach in a longitudinal case study of information systems
development and organisational change in British local government, provided significant
empirical evidence to address this perception. Horrocks (2009) argued that the
morphogenetic approach allowed detailed descriptions of the defining characteristics of
structures, cultures and people in his longitudinal case study. The focus of Horrocks’ article
was not directly related to the technology implemented, but more to the social and political
contexts within which the change occurred. As such, the article can provide guidance for an
examination of the complexities of enterprise-wide implementation of technical systems.
Equally, Mutch (2010) provided an example of using the morphogenetic approach in his
examination of the use of data warehouses by organisations. He suggested:
Three gains are seen to accrue from this approach: greater clarity about the
material properties of technology, links to broader structural conditions
arising from the conceptualisation of the relationship between agency and
structure, and the potential to explore the importance of reflexivity in
contemporary organizations, especially in conditions of the widespread use
of information and communication technology (p. 507).
Whilst Mutch was looking at a particular data-warehousing technology, his deep analysis of
the role of technology designers and the ultimate adopters provides guidance in the analysis
of EA, particularly the widespread implications for technology implementation within
organisations. Incorporating this perspective into a morphogenetic approach was an essential
element in the development of this study.
In the same way Dobson, Jackson and Gengatharen (2013) utilised the morphogenetic
approach to examine the adoption of broadband in rural regions. Whilst their study is more
focused on the societal implications of technology, the emphasis on situational logics has
relevance to this study on EA implementation. As argued by Archer (1995, p. 218) the
compatibility or incompatibility of a cultural system with its sociocultural interaction creates
a range of possible situational logics which create conditions for social reproduction
(morphostasis) or social transformation (morphogenesis). It will be interesting for future EA
studies to examine the extent to which situational logics define the mechanisms for change.
5
The stratified view of agency used by Archer (1995), separately describes people1.2 as human
beings, actors, and agents with particular institutional roles and positions, and provided fertile
ground for examining EA implementation.
Box 1.2: People Terminology
According to Archer (1995) people can be seen from different settings, referred to
as a stratified model of people: a) as a population with class, status and power; b)
as organised groups also referred to as corporate agency at institutional level; c)
as individual actors with roles and positions; and d) as collectivities or primary
agency with positions, places, functions, rules, duties, and rights. To appreciate
the stratified distinction of people, this research study used the different terms
above interchangeably. Detailed explanations of each of these terms are provided
in Chapter 4.
The importance of people presupposes EA implementation is a social program heavily
influenced by the structural and cultural systems surrounding the architecture. It is also
heavily impacted by belief structures underlying the program implementation – the theory or
built-in mechanisms. Astbury and Leeuw (2010) acknowledged the success and failure of
programs requires an understanding of both the underlying theories built into the program,
and the context within which it is implemented. EA methodologies and frameworks have
particular theories underlying their largely technical focus, and it is important to understand
such theories as well as their social contexts, since the latter is just as important as the former
in aiding EA implementation, particularly identification of the social (people-focused)
mechanisms that constrain or enable it. This thesis argues EA frameworks need to give more
consideration to the social aspects of their target audiences.
Astbury and Leeuw (2010) concluded, that only by understanding both the theories
underlying programs and the social contexts within which they are implemented, could they
determine “…how and why programs work (or fail to work) in different contexts and for
different program stakeholders” (p. 364). They described mechanisms as “…underlying
6
entities, processes, or structures which operate in particular contexts to generate outcomes
of interest” (p. 368). These authors viewed mechanisms as sensitive to variations in context,
since they may or may not be activated due to contingent conditions or possible
countervailing mechanisms in a particular context. Mechanisms play an important role in this
thesis and are best described as “triggerable causal powers” (Mason, Easton & Lenney, 2013).
Throughout the remainder of this Chapter and Chapter 2, mechanisms will continue to be
referred to in a general sense before a more detailed discussion in section 3.3, Chapter 3.
Since mechanisms are only effective if people (agents) adopt them, another important
element in this study is the part played by “reflexivity”. Archer (2007, p. 4) described
reflexivity as:
…the regular exercise of the mental ability, shared by all normal people, to
consider themselves in relation to their (social) contexts and vice versa. Such
deliberations are important since they form the basis upon which people
determine their future courses of action—always fallibly and always under
their own descriptions.
Such a social approach suggests new ways of looking at EA implementation that widens the
scope from a merely technical focus to the broader socio-technical1.3 aspects of
implementation, particularly in the IS domain (Carlsson, 2005; 2006).
An Australian university implementing an EA program (termed UX for anonymity) was used
as the case study in this research, over a period of 3 years from commencement to
completion. The case study describes the challenging environment and illustrates the
important mechanisms and situational logics that direct people’s actions within the complex
social context of a university. As identified by Gengatharen, Standing and Knight (2009), the
higher education sector in Australia, has over the past two decades, been operating in a
climate of uncertainty and change. The sector’s move towards corporatisation, marketisation
and rationalisation has introduced unique challenges in the form of a complex reality of social
interaction between structure, culture and individuals.
7
Box 1.3: Socio-technical
Detailed explanations have been presented by Eason (2014), which defines sociotechnical systems as "...heterogeneous systems, that is they are constituted of
components with very different characteristics" (p. 215).
According to Pasmore (1988) and Trist (1981) (as quoted in Carayon 2006, p. 529)
socio-technical systems consists of three components: a) social system; b)
technical system; and c) environment.
Furthermore (as quoted in Eason, 2013), “…in the social system, composed of
people, is completely different from the artifacts that make up the technical
system. The social system, for example, unlike the technical system, is made up of
sentient ‘components’ aware of their environment and capable of generating new
behaviour patterns in responses to the changes they perceive. And yet both kinds
of system component must work closely together if the system is to function well.
The interrelations that govern system behaviour extend in several directions: in
task interdependencies in the work to be done, between the people who share
the work, between the technical components in what might be more or less
integrated technical systems and between humans and technology at many
different levels. One way of expressing how socio-technical systems work is to say
that the work needed to undertake all the tasks needed to complete the collective
task is done by two kinds of resources: human resources and technical resources.
Human beings are thus components in the work performing system, taking
responsibility for particular tasks and turning their energy and skill into performing
those tasks” (p. 215).
JISC (2008) also expressed a similar opinion of the higher education sector (HEs):
Universities and colleges [HEs] are increasingly complex socio-technical
systems that are hard to change and yet they face enormous pressures to
increase operational efficiencies and to adapt to new challenges (p. 1).
It is in this climate that universities encounter strategic and operational challenges caused by
environmental factors (Anderson & Backhouse, 2009) such as global trends in HE, new future
themes, and the need for efficiencies, all of which require changes in technology capability.
Environmental factors in the HE sector have made it necessary for technology to respond
quickly to meet the new expectations of students, staff and other stakeholders, while at the
same time facing ever-increasing cost pressures. Increased use of new technologies has
changed the way business is done in almost all universities; and to add to the complexities,
8
different people within the university community view the technology and engage with it in
different ways.
Technology-supported information systems and communications in universities have various
organisational impacts; the impact on business being the most far-reaching. The primary
organisational objective of information and communications is to keep stakeholders
connected, so to ensure that students, staff and other stakeholders are productive,
information and collaboration technologies therefore need to be in place. This has led to the
use of IS as a vital service in universities, who increasingly look to the application of IT for not
only underpinning existing business operations and extending organisational boundaries, but
also for providing a strategic advantage by facilitating problem solving, increasing teaching
productivity, the quality and speed of learning, improving customer service and student
engagement, and enhancing information and research network collaboration. EA addresses
these challenges by providing a holistic view of the planning and development of an
organisation’s IS (JISC, 2008; JISC, 2009a; JISC, 2009b; Anderson & Backhouse, 2009; JISC,
2010).
This thesis deeply examined the important mechanisms, critical situational logics and their
social responses, the importance of reflexivity in the adoption process, and the EA program
mechanisms or theories that drive such large-scale architectural transformation in the
university sector. The case study provided an understanding of the contextual factors and a
unique opportunity for proposing amendments to existing frameworks to more clearly reflect
the critical role of people in the change process.
1.2 The Gap in Enterprise Architecture Research
Despite a considerable number of studies in the area of EA (particularly in the IS domain),
most have tended to emphasise the technical aspects of the program and neglected the social
aspects. For example, a number of researchers viewed EA in terms of framework comparisons
(Leist & Zellner, 2006; Urbaczewski & Mrdalj, 2006; McCharty, 2006; Zarvi´ci & Wieringa,
2006; Scheckkerman, 2007; Sessions, 2007), while others attempted to provide technical
guidance (Janssen & Madsen, 2007; Wegmann, Kotsalainen, Matthey, Regev, and
9
Giannattasio, 2008; Franke, Hook, Konig, Lagerstrom, Narman, Ulberg, Gustafsson, Ekstedt,
2009). Bass and Marbry (2004) focused on technical service-oriented architecture (SOA).
EA implementation with regard to the human domain (the role of people) has been
mentioned by several academics (Boh, Yellin, Dill & Herbsleb, 2003; Raadt, Schouten & Vliet,
2008; Raadt & Vliet, 2008; Martin, Purao & Robertson, 2009; Sembiring, Nuryatno &
Gondokaryono, 2011; Janssen, 2012; Lohe & Legner, 2013 among them), and by Gartner
(2009) and Gravesen (2012) from a practice standpoint, but none of these studies provide any
insights into EA as an organisation-wide program with strong social impacts. From a sociotechnical point of view there are still many unanswered questions.
Previous EA research from a critical realist perspective demonstrated the use of a
morphogenetic approach to explain EA evolution “as an interaction between the existing
structural setting (existing EA) and the action of introducing new business or IT capability into
an organisation [service-oriented architecture: SOA], which results in EA evolution outcomes
(SOA’s integration into EA outcomes)” (Alwadain, 2014, p. vi). Whilst Alwadain’s study is more
focused on the relationship between agency (agency seen as an ‘action’: Alwadain, 2014, p.
116) and structure to understand the nature of change, the emphasis on action and structure
has a different central point with this thesis, which has an emphasis on the interplay between
structure, culture and agency (agency seen as people, as demonstrated by Archer), of which
these are the fundamental components of the critical realist approach (Archer, 1995; 1996;
2013a; 2015). As Archer (1995, p. 195) suggested people are capable of resisting, repudiating,
suspending or circumventing not only the structural tendency, but also the cultural tendency
in unpredictable ways because of their creative powers as human beings in producing
tendencies towards change/elaboration (or reproduction) in the relational organisation of the
social order. These processes of change (or reproduction) are known as generative
mechanisms (Archer, 2015). Such a perspective is the core of Martin’s et al. (2009) view of
people role in EA evolution, they expressed the opinion that “Enterprise architecture, its
representation, and its uses all evolve; but they evolve in different phases and for different
reasons. Difficulty and sometimes even disaster ensue when these evolutions are misaligned.
Controlling these evolutions requires understanding the motivations and the mechanisms for
evolution (p. 1)”. They also pointed out that “architecture means different things to different
people (p. 2)” and “EA artifacts evolve through human-mediated transformations (p. 7)”.
10
In an enterprise-wide view of social context, people make up a key element in directing the
transformation of enterprises with technology. This enterprise-wide view is critical for
identifying the social phenomena in individuals’ desires, beliefs and actions within their social
and cultural environments. Using the philosophical lens and morphogenetic approach (MA)
of critical realism (CR) as a framework for studying the social phenomena provided a means
of representing how people (as individual and collective agents) reacted to the new
impositions and opportunities of EA implementation, both in terms of increased governance
and the potential impact on the way they were doing things.
As demonstrated by Archer (1995; 2015), people (agency) offer a stratified representation of
their role in organisational change as they can be seen to act as: a) primary agents in particular
positions who can generate important social consequences; b) corporate agents in
institutions who can organise themselves in pursuit of certain goals and articulate the changes
they seek; and c) individual actors in particular roles who acknowledge their vested interests
and weigh these interests against one another. In the same fashion, de Vaujany (2008) argued
people can be seen to act as: a) an individual/person with a personal and embedded history;
b) agents with cultural, economic and demographic features; and c) actors related to a social
group with specific interest and strategies. It is important to consider the social complexity of
the role of people in EA programs who are shaped by the interplay between contexts and
concerns.
This thesis argues, as a planning process with strong social impacts on an organisation’s
environment, the way in which EA is implemented will strongly impact and be impacted by
stakeholders’ acceptance of the program, as their actions ultimately enable or constrain
adoption. For the most part, the existing literature describes EA as a process of converting
strategy statements into plans to support organisational information systems and
applications, technology platforms, infrastructure, business processes and services. However,
the social nature of the process requires recognition of EA as a social program heavily
influenced by the interplay between structure, culture, and agency surrounding the
architecture and the mechanisms built into the program. CR will provide the “underlabourer”
for the study and the MA will provide the methodological grounding.
The MA emphasises strongly the role of time in situation examination suggesting that
structure and culture predate subsequent actions by involved agents. The goal of this study
11
was therefore to describe particular situational logics or mechanisms emanating from preexisting interaction between structural and cultural systems – it is suggested that these
situational logics encourage particular behaviours in response to the EA program. Since
mechanisms are only effective if people adopt them or not, another important element in this
study is the part played by “reflexivity”. Reflexivity highlights the linkage between people
concerns, projects and practices as people act in order to promote their concerns, and form
projects to advance or to protect what they care about most (Garcia-Ruiz & RodriguezLluesma, 2010). Reflexivity will also play an important role in the thesis as it is an important
mechanism for explaining how people’s projects and ultimate concerns impact on their
approach to the impositions of EA.
1.3 Significance of the Research
As an independent design discipline somewhere between business strategy and architecture
(Gravesen, 2012) EA has been widely adopted over the past 20 years in the commercial world
and public sector organisations as a tool for change (Anderson & Backhouse, 2009). There is
no doubt that EA benefits are real (Gravesen, 2012; Janssen, 2012), however, it appears that
EA has generally not achieved the desired results (Boh et al., 2003; Raadt & Vliet, 2008;
Gartner, 2009; Gaver, 2010; Roeleven, 2010; Gravesen, 2012; Čyras & Riedl, 2012; Janssen &
Klievink, 2012).
For a number of reasons organisations still have difficulty establishing an EA that is fully
integrated into their IT and IS management. Watkins (1998), for example, argues the lack of
a clear understanding of the distinction between the terms IT and IS has led to organisations
focusing on IT goals, without consideration for the purpose of IS, and this is one of the reasons
why EA has not achieved the desired results. IT refers specifically to technology, essentially
hardware, software and telecommunications networks, which in a narrow definition becomes
the technology component of IS. The UK Academy of Information Systems (UKAIS) (cited in
Ward & Peppard, 2003, p. 3) defined IS “as the means by which people and organisations,
utilising technology, gather, process, store, use and disseminate information”. UKAIS defined
IS studies as theories and practices related to the social and technological phenomena which
determine the development, use and effects of information systems on organisations and
12
society (Mingers, 1995). EA must recognize both aspects in its planning and management – IT
and IS.
TOGAF1.4 (2011, p. 52) describes some of the “people” roles in an EA managerial environment
thus:
There are many reasons to constrain (or restrict) the scope of the
architecture activity to be undertaken, most of which relate to limits in: the
organisational authority of the team producing the architecture, the
objective and stakeholders concerns to be addressed within the architecture
…
Box 1.4: TOGAF
TOGAF is a detailed method and set of supporting tools for developing enterprise
architecture (TOGAF, 2011). It is the most widely used EA framework (judged by
published certification numbers) and can be used freely by any organisation
wishing to develop and implement EA. A detailed explanation is provided in
Chapter 2.
This means the board and management within the organisation, both in business and IT, must
collaborate, interact and work together, so that EA becomes embedded in the organisation’s
management agenda. Raadt et al. (2008) viewed collaboration between architects and
stakeholders of IT as one of the key critical success factors for EA. According to TOGAF (2011,
p. 374) EA stakeholders are “…people who have key roles in, or concern about the system; for
example, as user, developers, or managers”. From a wider perspective, Bente et al. (2012, p.
32) argued that EA stakeholders are “…the people who deal with creation, evolution, and
operation of the system”. In other words they include owners, strategists, planners,
designers, subcontractors who provide constituent parts for it, and support staff who
maintain and operate the system. According to this viewpoint people play a critical role in
implementation.
13
Evidence suggests that EA implementation is not a single activity with a clear beginning and
end (Janssen, 2012), but influenced on an ongoing basis by users’ interpretations and
extensions, providing feedback for improvement and review (see Figure 4.3). It stands to
reason that EA will only be accepted if its implementation is aligned with people’s current
projects and interests, and people’s perceptions will determine how they respond to the new
environment provided by the EA. For these reasons EA implementation can be viewed not
only as a process for creating a technical IT architecture, but also as a social program.
Figure 4.13: Picturing EA from the dynamic of social complex program surrounding the
architecture (reproduced here for ease of reference)
In their interpretation of the organisation’s attitude towards change through enterprise
architecture, Bente et al. (2012) suggested:
…for whatever reason, people start behaving differently and things are
changing—not necessarily in the way you envisioned, and not always for the
better. What we see here is a classical misconception by people with an
analytical mindset. You would assume that a decision-making process works
in the sequence Analyze, Think, Change. You provide a thorough analysis,
the management thinks about it, and change is triggered. Indeed, this is the
kind of rational reasoning incorporated in numerous engineering disciplines
and process frameworks. The catch is, as Kotter and Cohen (2002) point out,
14
things do not work exactly that way. Change processes follow the pattern
See, Feel, Change. The issue needs to raise people’s attention (see). If they
develop a sense of urgency and importance about the topic (feel), they are
likely to trigger action (change) (p. 286).
Consistent with these arguments, the JISC’s Enterprise Architecture Group Pilot Projects in
the HE sector (JISC, 2009a; 2009b; 2010) indicated that implementation of EA is often more
social in nature than having a purely technical focus. The goals and scope of EA generally
support a more social approach, with all its associated implications in regard to structure,
culture, agents and its causal mechanisms. It is in this context that EA implementation has
been examined in this thesis, since EA is not merely a technology, but has wide social
implications for implementation in organisations.
The adoption of CR, which focused on the implementation of EA at a university over a 3-year
period from start-up to conclusion, has significant implications for the objects to be
investigated. It provides an important link between realist ontology and practical socialinformation systems outcomes, and forms a consistent foundation for ontology and
methodology by representing the deep and foundational role of people in explaining change.
Previous IS research has demonstrated that CR is suitable for studying IS in overcoming some
of the difficulties associated with the social contexts (Carlsson, 2006; Horrocks, 2009; Bygstad,
2010; Dobson et al., 2011; Carlsson, 2011; Dobson et al., 2013; Mingers, Mutch, & Willcocks,
2013; Nuryatno & Dobson, 2016, among others).
According to Fox (2009) CR can increase understanding of causal mechanisms and contexts in
IS research. Furthermore, the value and importance of CR as a philosophy in IS research was
also supported by Pettersen, McDonald and Engen (2009), who acknowledged that CR can
address social ontology as a necessary precursor to developing models and empirical accounts
of socio-technical systems. Carlsson (2011) expressed the opinion that CR can be useful as an
underpinning philosophy for behavioural IS research as well for IS design science research.
The use of case study is also well suited to CR-based attempts to find explicit causal
explanations for the complex social and organisational phenomena in the IS field (Wynn &
Williams, 2008). In addition, Bygstad and Munkvold (2011) indicated CR has an important role
15
in IS research in identifying causal structures of an ontological depth that is difficult to unveil
through other alternative approaches (i.e. positivist and interpretivist).
The university case study provided a unique environment for investigating the role played by
people in managing the implementation, as well as the critical role of end-users. In this way
this research potentially provides a useful explanation on how to examine the social
complexity of enterprise-wide implementation of technical architectural systems.
1.4 Research Questions
The research questions describe how the research mainly seeks to identify important
mechanisms that might constrain and enable the success of EA implementation. To
appreciate the explanatory context of the study the research questions mainly built upon
Archer’s morphogenetic approach, supported by Archer’s reflexivity theory: a) macro-micro
context (identifies the situational mechanisms of the cultural and structural systems that preexist people’s action); b) micro-micro (identifies the mechanisms of action and interaction
between structure, culture and agency); and c) micro-macro (identifies the transformation
mechanisms that constrain and enable people’s actions to adopt the EA program or not).
The main research question that this research sought to address was: “What are the key
implementation mechanisms and social responses triggered by EA implementation that might
constrain and enable the success of the EA program in University X and the sector in general?”
It was supported by four sub-research questions as follows:
1.
What are the important situational mechanisms, that by associated social structure and
culture, causally condition individuals’ actions?
2.
What are the consequent interaction mechanisms triggered by EA implementation at
University X?
3.
How does the existing culture and structure within University X impact the EA
implementation
and
shape
the
interaction
mechanisms
triggered
by
the
implementation?
4.
What are the necessary conditions to encourage individual and collective acceptance of
EA practices?
16
Adopting a multi-dimensional cycle of change in the MA, the research questions explore and
elaborate on the role of people in EA implementation in terms of structural, cultural and
agential change – leading to either morphostatic (reproduction: constraining the success of
the EA program) or morphogenetic (transformation/elaboration: enabling the success of the
EA program). A detailed explanation of each of the research questions will be presented in
section 4.6, 5.3 and 7.2.
1.5 Main Contributions of the Thesis
This thesis explores and builds a theoretical explanation of the role of people as the key
element in EA implementation, most notably within the university sector, where, it can be
argued, particularly complex social dynamics surround the architecture. To examine the
adoption of EA semi-structured in-depth interviews were held with members of the university
IT Governance Committee, members of the Architecture Review Board, the CIO, and
members of the Enterprise Business Group, as well as the end-users, such as teaching staff,
researchers, fellow students, and administrative staff of the faculties, schools and service
centres. The interview questions were developed based on a morphogenetic approach to
propose the generative mechanisms hidden in the real domain, and to highlight conditions
that encouraged individual and collective acceptance of EA practices. The reflexivity indicator
developed by Archer – ICONI – was completed by Interviewees and used throughout to
determine the likely reflexivity mode evident. Understanding reflexivity modes helps to
explain the role of personal projects and how ultimate concerns mediate the
structural/cultural constraints and enablements evident during EA implementation.
Attendance at regular EA implementation meetings at UX was also important in
understanding and explaining the major elements within the program.
Theoretical explanations are given for how initial opposition to the EA program was managed
by the university executive and lead to ultimate acceptance over a 3-year period of EA
implementation in UX. In particular, this thesis brings together two specific areas of study, EA
and social systems, and provides a deeper explanation of EA implementation programs by
understanding the social complexity surrounding the architecture. Recognition of people as
the key element in EA implementation provides a useful explanation of how the key
stakeholders (and their power, influence and interests) may constrain and enable EA
17
implementation. This thesis also highlights the linkage between people’s projects, priorities
and ultimate concerns shape their reaction to EA program, as people act in order to promote
their concerns, and form projects to advance or to protect what they care about most.
Several theoretical contributions can be advanced as follows:
1.
This thesis highlights the usefulness of the morphogenetic approach to identify the
hidden mechanisms and situational logics that drive acceptance of EA. This study makes
an important link between a realist ontology and practical social-IS outcomes, thus
establishing a fundamental consistency between ontology, methodology and practical
theory (see Chapter 5 & 6).
2.
This thesis identified gaps in EA research and highlights important program mechanisms
or theories identified in a review of IS literature (see Table 2.6), and describe how these
mechanisms built in to the program acted within situational logics to ultimately support
the implementation (see Chapter 6). Some of the mechanisms built in to the EA program
included:
• University EA Road Show: Individuals’ Engagement, Shared Vision, and Linkage
• University New IT Governance, management and regulation: EA Governance
• University Enterprise Business Group (EBG): Collaboration, Communication,
Stakeholders’ Role, and Compliance
• Distributed Leadership (by CIO): Leadership
These mechanisms reside in the program itself and were impacted by the social context.
These program mechanisms are shown to be important within the identified
opportunism situational logic that also acted as a transformation mechanism (generative
mechanism). By corporate agents presenting EA as a platform to promote their own
opportunistic projects acceptance was encouraged.
3.
This thesis identified the importance of situational logics in explaining social reactions to
change programs. It identifies and describes the impact of opportunism as a situational
logic in the university sector and how it leads to increasing sectionalism and
diversification. The thesis describes the importance of EA recognising this increased
sectionalism and diversification.
18
4.
This thesis identifies important generative mechanisms that constrain and enable success
of EA programs within the university (see Table 4.3 and Figure 6.6). This identification is
important – whilst the mechanisms within the social context are difficult to change, being
a consequence of the long-term interaction between culture and structure, their
recognition provides an important first step in being able to address their impact. A core
argument within social realism is that the first step in addressing inequitable structures
is recognition. In order to address inequities, one must be able to recognise their origins
and features. This argument can be similarly made for examining the cultural and
structural conditions necessary for successful EA.
5.
This thesis identified two key implementation mechanisms for encouraging successful EA
adoption.
Collaboration and Communication are necessary key implementation
mechanisms required to accommodate the opportunism situational logic. It is the
condition under which EA operates and EA must be in line with it for EA to be successful.
6.
This thesis advances the application of reflexivity theory to people’s responses to the
embedded mechanisms that encourage individual and collective acceptance of EA
practices by: a) Identifying how the different reflexivity modes reacted to the inbuilt
mechanisms, and; b) Identifying the important collective reflexivity modes. The research
findings also show that the reflexivity investigation tool, the ICONI, can successfully be
used to better identify the conditions necessary for encouraging individual and collective
acceptance of EA practices. The thesis thus provides a unique theoretical contribution to
IS theory by advancing our understanding of the role of people’s internal conversations
(reflexivity theory) in enterprise-wide IT change; it describes how people’s projects,
priorities and ultimate concerns shape their reaction to EA (see Table 5.2, 6.2 & 6.3).
7.
This thesis advances the application of abduction or theoretical redescription in
identifying events and non-events within the actual domain (not directly observable),
which can be applied to other IS research from a CR perspective. Creative abduction was
used as the frame of interpretation; and by focusing on abduction in two domains (the
empirical and the real) this study provides fresh insights (see Figure 5.11 & Table 6.5).
19
From a practical point of view this study makes the following contributions:
1.
The thesis provides an example of the application of CR as underlabourer examining the
social aspects of EA implementation and management.
2.
The thesis provides a detailed example of the application of MA as a useful tool for
examining large scale enterprise wide change.
3.
The thesis applies Archer’s newly developed reflexivity theory to understand and explain
how people reflexively react to large scale IT based change programs.
4.
The application of MA and reflexivity in combination provide a powerful explanatory
framework for examining the critical role of people in the change process.
5.
The long term 4-year case study followed the implementation from initial start-up to
ultimate acceptance – this unique opportunity has particular benefits for other
universities as the importance of EA is increasingly becoming recognised in the sector.
The study allows a detailed understanding of the contextual factors and implementation
issues related to such large-scale transformation (see Chapter 6 & 7).
1.6 Thesis Outline
The first chapter of the thesis provides a brief description of enterprise architecture as an
independent design discipline and as a model-based management and planning approach. In
addition, the social contexts of higher education information systems are discussed. Chapter
one also includes a review of current influences, previous studies on the topic, the gap in
enterprise architecture research, the significance and main contributions of the thesis.
The next chapter provides an analysis of existing literature related to the importance of
people in enterprise architecture implementation from the perspective of a social program
heavily influenced by the structural and cultural systems surrounding the architecture and
the mechanisms built into the program. Program mechanisms, specifically the success
mechanisms identified in the literature, are reviewed and summarised.
Chapter three examines the various versions of social realist theories; their underlying focus
on ontology and the methodological and theoretical implications of their underlying premise.
The chapter provides a deep understanding of the implications of particular philosophical
positions and allows an understanding as to the benefit that critical realism can provide. The
20
chapter can only provide a limited comparison but emphasises the underlying argument of
this thesis that philosophy matters (Dobson, 2001a).
Chapter four discusses the key concepts of critical realism. It presents a literature review
associated with a critical realist perspective of information systems research and a
morphogenetic approach as an explanatory framework, and describes the central role of
reflexivity theory in EA implementation. An analysis of the morphogenetic literature provides
the grounding for a causal explanation for the study as an exploration of the interplay
between structure, culture and agency. On the basis of the MA and critical realism the
research questions are defined.
Chapter five outlines the research approach and methodology, including the underlying
philosophy and its methodological framework, the context of the research questions, the
research design and its limitations, as well as the ethical considerations of the research. This
chapter explains the rationale for using critical realism as the underlying philosophy for a
social investigation in the IS domain, and for using a single university case example. The use
of a critical realist perspective for the methodology of the study is also discussed to provide
an overview of the target population, materials, the data collection procedures and the
morphogenetic approach as the main analytical tool.
The case study, data analysis and research findings are presented in Chapter six which
describes the analysis by means of a critical realist view of the causal factors and mechanisms.
It presents the patterns of results from each of the critical realist domains to reveal a range
of important generative mechanisms, critical situational logics, analytical histories, key
implementation mechanisms, and social responses.
The findings for each research question are discussed in Chapter seven in relation to
information systems theory and the wider body of knowledge. The implications of using a
critical realist philosophy to gain further understanding of the research problems are
explored, along with implications for practice. The limitations of the research, along with
future research directions, are also discussed. The final chapter contains the concluding
remarks. The structure of the thesis is illustrated in Figure 1.1 below.
21
Figure 1.1: Thesis structure
1.7 Summary
This thesis was prompted by the observation that some organisations appear to succeed with
EA implementation and others don’t. It proposes that a lack of recognition of the role of
people may provide an explanation for this anomaly, and uses critical realism to gain an
understanding of the complex social dynamics amongst EA stakeholders and their key issues
and concerns related to the EA implementation.
The importance of people in EA implementation presupposes acceptance of EA
implementation as a social program, heavily influenced by inherent structural and cultural
systems and mechanisms built into the program. Together with the structure and levels of
rigour and formality within the organisation, these factors influence implementation by
preventing or supporting change and effective operation.
22
CHAPTER TWO: ENTERPRISE ARCHITECTURE AND
THE LITERATURE REVIEW
2.1 Overview
For many years enterprise architecture (EA) programs have encapsulated a long-term
perspective of companies’ business processes, information systems and technologies (Ross et
al., 2006). EA is widely regarded as the starting point for a process of change because it shows
up the gaps between the current situation and the ideal situation, and facilitates alignment
between organisational information systems (IS) and business goals (Anderson & Backhouse,
2009). However, the implications of EA programs and their implementation are still varied
and complex (Boh et al., 2003; Raadt et al., 2008; Gartner, 2009; Roeleven, 2010; Gravesen,
2012; Janssen, 2012; Janssen & Klievink, 2012; Lohe & Legner, 2013). To date research has
focused predominantly on the IT capability of organisations and the technical aspects of
implementation and has neglected the role of people (Nuryatno & Dobson, 2016). Benson et
al., (2004) pointed out that many companies describe EA programs in “as-is” and “to-be”
technical contexts. For instance, the “as-is” context focuses more on technology issues, less
on business processes, and even less on the social context; whereas the “to-be” context
focuses predominantly on the technology to be implemented. On the one hand, these
programs are primarily built on a resource-based view of EA methodologies and frameworks2.1
that have particular social theories underlying their largely technical focus; yet on the other
hand, implementation of EA programs is often driven by technical aspects of IT, regardless of
the complex social contexts in which they exist (Bente et al., 2012; Janssen, 2012).
The social context of EA is just as important as the technical context in facilitating the
implementation of EA programs, particularly identification of the social (people-focused)
mechanisms that constrain or enable them. A social perspective is necessary to allow for such
examination. A review of the academic literature and so-called grey literature revealed a
focus on the TOGAF framework, the most referenced EA implementation framework in the
higher education sector (Anderson & Backhouse 2009; JISC, 2009b). Coupled with evidence
23
of gaps in the existing body of knowledge, the research questions in Chapter 4 were
formulated.
Box 2.1: EA Methodology and Framework
EA methodology is “…a defined, repeatable series of steps to address a particular
type of problem, which typically centres on a defined process, but may also include
definition of content.” (TOGAF, 2011, p. 7)
EA framework is “…a foundational structure, or set of structures, which can be
used for developing a broad range of different architectures. It should describe a
method for designing a target state of the enterprise in terms of a set of building
blocks, and for showing how the building blocks fit together. It should contain a
set of tools and provide a common vocabulary. It should also include a list of
recommended standards and compliant products that can be used to implement
the building blocks.” (TOGAF, 2011, p. 27)
The first section of this chapter examines the extent to which the role of people has been
identified in existing EA research and EA frameworks. Since the importance of people is also
impacted by the theory or mechanisms built into the program (Astbury & Leeuw, 2010), it is
important to understand such theories in order to identify program mechanisms and issues
that influence the success and failure of EA implementation. Program mechanisms,
specifically the success mechanisms identified in the literature, are reviewed and
summarised.
Next, the chapter provides an analysis of the existing literature related to an industry
perspective on EA failure. The challenges of EA implementation and governance are
summarised, and the literature on EA implementation in higher education is discussed in
more detail to illustrate the naturally conservative nature of universities impacting the
implementation program. The processes operating at social, structural and cultural levels are
also examined, followed by a summary of the most significant challenges and program
mechanisms that facilitate EA implementation in the higher education sector. Social context
versus technical capability is discussed from the viewpoint of people as a key element in EA
24
implementation. Finally, the use of programs with embedded social and cultural norms,
aligned to central themes in the TOGAF literature, are recommended to integrate the
important role of people and relationships, program mechanisms and guidance in the form
of a TOGAF-based implementation framework for the higher education sector.
2.2 Enterprise Architecture Discipline
EA has been used in industry for over twenty years (Gravesen, 2012) as an independent
design discipline that combines strategy and architecture. Today there is little doubt that the
benefits of EA are real (Janssen, 2012). Spragg (2015) expressed the opinion that EA is about
preparing a business for change, and the flexibility and agility of EA programs are therefore
becoming strategic necessities (Doucet, Gøtze, Saha, & Bernard, 2008) as they enable
transformation. Ross et al. (2006) concluded “greater globalization, increasing regulation, and
faster cycle times all demand an ability to quickly change organizational processes” (p. 12).
According to these authors the IT engagement model2.2 allows organisations to achieve the
right balance between IT efficiency and business innovation.
Box 2.2: IT Engagement Model
The IT engagement model “…is the system of governance mechanisms that ensure
business and IT projects achieve both local and companywide objectives. The IT
engagement model influences project decisions so that individual solutions are
guided by the enterprise architecture. The engagement model provides for
alignment between the IT and business objectives of projects, and coordinates the
IT and business process decisions made at multiple organizational levels (e.g.,
companywide, business unit, project). To do so, the model establishes linkages
between senior-level IT decisions, such as project prioritization and companywide
process design, and project-level implementation decisions.” (Ross et al., 2006, p.
9)
Despite its wide adoption as a strategic management technique for aligning business with IS
(Ross et al., 2006) and adapting to change (Anderson & Backhouse 2009), EA implementation
25
appears not to have achieved the desired results. This is referred to in the literature as EA
phenomena (Boh et al., 2003; Raadt & Vliet, 2008; Gartner, 2009; Gaver, 2010; Roeleven,
2010; Gravesen, 2012; Janssen & Klievink, 2012; Čyras & Riedl, 2012). Research related to an
industry perspective on EA failure (representing 161 respondents from 89 organisations
across a wide range of industries) stated that 66% of EA programs did not fulfil organisations’
expectations (Roeleven, 2010, pp. 2-3). For a number of reasons organisations still have
difficulties establishing an EA that is fully integrated with their IS management. More recent
research has identified people, as individuals or social groups (Bente et al., 2012; Janssen,
2012; Nuryatno & Dobson, 2016), as a key factor in EA implementation and a major cause of
the failure of EA programs to meet organisations’ expectations (Boh et al., 2003; Raadt et al.,
2008; Gravesen, 2012; Janssen, 2012; Janssen & Klievink, 2012; Lohe & Legner, 2013). A
comprehensive review of the EA framework, such as TOGAF’s (2011) key ideas about peoplefocused mechanisms, highlighted:
Typically, an enterprise architecture is developed because key people have
concerns that need to be addressed by the IT systems within the
organization. Such people are commonly referred to as the ‘‘stakeholders’’
in the system. The role of the architect is to address these concerns, by
identifying and refining the requirements that the stakeholders have,
developing views of the architecture that show how the concerns and the
requirements are going to be addressed, and by showing the trade-offs that
are going to be made in reconciling the potentially conflicting concerns of
different stakeholders (p. 7).
Another interesting EA framework that can be used to highlight the people element is the
Zachman framework (Zachman, 1987). As the oldest and best-known of the EA frameworks,
Zachman prescribes people as “who is doing what” (Zachman, 1987, p. 292). Zachman’s
people element or so-called the “people dimension” (the “WHO” column, see Zachman,
2008), is focused on identifying people participation in the organisational activities. According
to Ertaul and Rathod (2012), the “WHO” column defined the roles, privileges and
responsibilities attached to each person within the business management perspective.
However, they also highlighted:
26
Although the Zachman Framework provides a perfect tool for classification
of artifacts and delegation of responsibilities, it fails to provide any step-bystep process for building the reference model and Enterprise Architecture.
[Thus] Enterprises in general have become global now. This led to rapid
dynamic changes in organization. The Zachman Framework lacks the agility
to handle these rapid changes. (p. 5).
Whilst the Ertaul and Rathod (2012) study is more focused on the Owner’s perspective of the
Zachman Framework and its security requirements, the suggestion that a framework must
recognise of the ever-increasing size and complexities of EA is important. Janssen and
Madsen (2007) indicated that “...the Zachman framework is too complex to support
communication…. it is too abstract to capture our architectural problems (p.6)”. While the
framework is widely discussed and is a foundational contribution to enterprise architecture,
its practical value has been questioned (Ylimäki and Halttunen, 2006; Janssen & Madsen,
2007; Scherer & Wimmer, 2011; Ertaul & Rathod, 2012). The framework does however
specifically highlight the “people” dimension and places it equally alongside the other
important dimensions. The framework is an important foundational framework that supports
the basic suggestion that a recognition of the “people” role is essential for a complete
description of EA.
Amongst numerous goals (e.g. Benson et al., 2004; Ross et al., 2006; Urbaczewski & Mrdalj,
2006; Zarvic & Wieringa, 2006), EA can be understood as a means “…to optimise across the
enterprise-wide systems the often fragmented legacy of data processes (both manual and
automated) into an integrated environment that is responsive to change and supportive of
the delivery of the business strategy (TOGAF, 2011)” or as “…a guide for an enterprise’s
business processes and the associated IS towards a common goal to integrate business, data,
information, and technology (Madsen, 2006, p. 2)”. While these definitions deal primarily
with technical systems, the goals and scope of EA are often more social in nature. Bente et al.
(2012) argued that “the people element brings complex behavioural attributes into the
functioning of an enterprise…” (p. 35), thus EA “…deals with social elements such as
collaborative business processes, organisational leadership, political dynamics and work
culture...” (p. 36). This definition was supported by Janssen’s (2012) notion of socio-political
27
factors within EA implementation, which he claimed could be seen as “…a means to inform,
guide, direct, and constrain the decisions taken by human beings within organizations” (p.25).
Based on the findings of the EA literature review (Table 2.1), most EA definitions (EA theories,
methodologies and frameworks) prior to 2009 tended to emphasise the technological and
business aspects and did not discuss the specific social contexts of implementation. Anderson
and Backhouse (2009), in their JISC report (Joint Information Systems Committee in United
Kingdom), wrote:
[EA] provides an evolving, dynamic way of describing and aligning the
functional aspects of an organisation, its people, activities, tools, resources
and data/information, so that they work more effectively together to
achieve its business goals. EA is also about achieving desired future change
through design. It holds that by understanding existing information assets,
business processes, organisational structures, information and application
infrastructure (the 'as is' state) it is possible to 'do something different' (the
'to be' state) (p. 8).
Implementation of an effective EA program involves dealing with a number of pitfalls (Van
Den Berg & Van Steenbergen, 2006). It becomes even more difficult when an organisation
does not consider the social context until it becomes an issue (Janssen, 2012; Anderson &
Backhouse, 2009). Misconceptions continue to swirl around definitions of EA and what it
delivers in terms of business benefits. As declared by Fairhead and Good (cited in Saha, 2009)
“…there is no generally accepted definition on architecture (p. 266).
Table 2.1 EA Definitions in EA Contexts
Literature
From EA Theory/Methodology/Existing Research:
A holistic representation of all the components of
the enterprise (or organization), and the use of
graphics and schematics are used to emphasize all
Benson, Bugnitz and Walton
(2004)
Social
Definitions
Technology &
Business
EA Context
✓
28
the parts of the enterprise and how they are
interrelated.
A complete expression of the enterprise; a master
plan which "acts as a collaboration force" between
aspects of business planning such as goals, visions,
strategies and governance principles; aspects of
business operation such as business terms,
organisation structures, processes and data;
aspects of automation such as information systems
and database; and the enabling technological
infrastructure of the business such as computers,
operating systems and networks.
The organizing logic for business processes and IT
infrastructure, reflecting the integration and
standardization requirements of the company’s
operating model.
Relates organizational mission, goals & objectives
to work processes and to the technical or IT
infrastructure required to execute them.
A guide for an enterprise’s business processes and
the associated IS towards a common goal and to
integrate business, data, information, and
technology.
The structure of an enterprise, consisting of the
relationships among its ICT systems, the external
properties of those ICT systems, and the way these
create emergent properties with added value for
the enterprise.
A coherent set of descriptions, covering a
regulations-oriented, design-oriented and
patterns-oriented perspective on an enterprise,
which provides indicators and controls that enable
the informed governance of the enterprise’s
evolution and success.
The (usually recursive) structural and functional
composition of components of a collection of
organisations, where the organisations have a
common set of (essentially functional) goals. This
definition therefore positions structure as a
function-enabler and as such it’s subservient, while
placing clear emphasis on functional objectives of
the enterprise and its constituents that are geared
to meeting the objectives (or a coherent related
set of them) of the enterprise.
A coherent whole of principles, methods, and
models that are used in the design and realisation
of an enterprise’s organisational structure,
Literature
Schekkerman (2004)
✓
Ross, Weill and Robertson
(2006)
✓
Urbaczewski and Mrdalj
(2006)
✓
Madsen (2006)
✓
Zarvi´c and Wieringa (2006)
✓
Land, Proper, Waage Cloo,
and Steghuis (2009)
✓
Bhagwat cited in Saha
(2009)
✓
Lankhorst (2009)
✓
Social
Definitions
Technology &
Business
EA Context
29
business processes, information systems, and
infrastructure.
A high-level, strategic technique designed to help
senior managers achieve business and
organisational change. It provides an evolving,
dynamic way of describing and aligning the
functional aspects of an organisation, its people,
activities, tools, resources and data/information,
so that they work more effectively together to
achieve its business goals.
A framework for the business to add new
applications, infrastructure, and systems for
managing the lifecycle and the value of current and
future environments. Enterprise Architecture
provides alignment across business strategy, IT
strategy, and IT implementation. It tightly
integrates the business and IT strategies to create
an ongoing way to use IT to sustain and grow the
business.
EA is a means to inform, guide, direct, and
constrain the decisions taken by human beings
within organizations. Architecture is the
description and prescription of a set of elements
[people, processes, systems, and technology] and
the relationships between them.
The representation of the structure and behaviour
of an enterprise’s IT landscape in relation to its
business environment. It reflects the current and
future use of IT in the enterprise and provides a
roadmap to reach a future state. EA deals with a
socio-technical system. The people element brings
complex behavioural attributes to the functioning
of an enterprise.
From EA Framework:
A logical structure for classifying and organizing the
descriptive representations of an enterprise that
are significant to the management of the
enterprise, as well as to the development of the
enterprise's systems (Zachman Framework)
A normative restriction of design freedom and
operationally a set of design principles
(The Netherlands Architecture Forum-NAF)
A set of principles, rules, standards, and guidelines,
visualising and expressing a vision and
implementing concepts, containing a mixture of
style, engineering, and construction principles
(Capgemini Framework).
Literature
Social
Definitions
Technology &
Business
EA Context
Anderson and Backhouse
(2009)
✓
✓
Godinez, Hechler, Koenig,
Lockwood, Oberhofer, and
Schroeck (2010)
✓
Janssen (2012)
✓
✓
Bente, Bombosch and
Langade (2012)
✓
✓
Zachman (1987)
✓
xAF working group:
Extensible Architecture
Framework version 1.1 cited
in Land et al. (2009)
Capgemini EA framework
cited in Land et al. (2009)
✓
✓
30
An integrated framework for evolving or
maintaining existing information technology and
acquiring new information technology to achieve
the agency’s strategic goals and information
resources management goals
(The Clinger–Cohen Acts)
Encompassing all of its information and technology
services, processes, and infrastructure – to
optimise across the enterprise wide systems, the
often fragmented legacy of data processes (both
manual and automated) into an integrated
environment that is responsive to change and
supportive of the delivery of the business strategy
(TOGAF)
A discipline for proactively and holistically leading
enterprise responses to disruptive forces by
identifying and analysing the execution of change
toward desired business vision and outcomes. EA
delivers value by presenting business and IT
leaders with signature-ready recommendations for
adjusting policies and projects to achieve target
business outcomes that capitalize on relevant
business disruptions. (Gartner Framework)
Literature
US Federal Enterprise
Architecture Framework
cited in Land et al. (2009)
✓
TOGAF (2011)
✓
Gartner Framework cited in
Saha (2014)
✓
Social
Definitions
Technology &
Business
EA Context
2.3 Understanding Industry Perspectives on Enterprise
Architecture Failure
In this section, resistance to EA and the failure of EA implementation are discussed from an
operational viewpoint (focusing on the tasks and activities within architecture operational
elements, and the management required to conduct architectural operations), followed by
an academic perspective. Table 2.2 shows Gartner’s (2009) ranking of the ten most common
failures in EA implementation in order of importance. The final column gives an indication of
the focus of this study, as it provides a starting point for investigating potential success
mechanisms to counteract common failures.
31
Table 2.2 Ten Common Failures in EA Implementation
No.
Common Failures
1
Chief architect is an
ineffective leader.
2
Insufficient
stakeholder
understanding and
support.
3
EA not engaging the
business people.
4
EA group focuses only
on technical-domain
architecture.
5
EA group does currentstate EA first.
6
The EA group does
most of the
architecting.
Gartner’s Arguments
The lead architect may
understand EA well, but has
ineffective leadership skills that
even a good organisational
structure and staffing levels
cannot overcome.
This happens when employees
outside the EA team do not
participate in the EA program,
EA content is not used in
projects, and management
questions its value.
When IT and business goals are
not aligned, the resultant
problems will include nontechnical people trying to make
technical decisions while
enterprise architects become
too reactionary and tactical in
response to projects.
This dated EA approach is still in
use in some organisations and is
even narrower in scope than
technical architecture.
Successful EA provides
prescriptive guidance but
current-state EA does not, so it
delays delivery of EA value and
hinders the creation of good
future-state EA. It does not
establish the business context
and then focus first on futurestate EA.
This is a pitfall because the EA
content is typically off the mark
as it was not informed by those
on the business side. As a
consequence, there is no buy-in.
The primary job of architects is
to lead the EA process rather
than impose EA content on the
organisation. They should form
Possible questions for this
study
How can leaders engage
individuals at all levels in
leading the success of EA
implementation?
To what extent can social
communication secure
executive-team support?
What role do these
stakeholders play in EA
implementation?
Is a necessary mechanism for
success the involvement of
enterprise architects in the
development of the business
context? How should EA
management engage jointly
with the ultimate end-users?
To what extent can program
mechanisms provide a muchneeded broader focus than
solely the technical
architecture?
To what extent does currentstate analysis hinder
successful EA
implementation?
What structures,
mechanisms and tools can
encourage user involvement
and shared understanding
and commitment?
32
No.
Common Failures
Gartner’s Arguments
Possible questions for this
study
virtual teams to create content
and seek consensus on the
content.
7
Not measuring and not
communicating the
impact.
The value of EA is often indirect,
so it may not be obvious to
everyone in the organisation.
This exposes the EA program to
the risk of failure. The
enterprise architect has to
determine ways to effectively
communicate the value of EA in
a simple way.
8
Architecting the
‘boxes’ only.
9
EA group does not
establish effective EA
governance early.
10
EA group does not
focus sufficiently on
communication.
Enabling better business agility
and integration is key but
architecting standards for the
‘boxes’ (business units) in
process, information, technical
and solution models doesn’t
address this. Integration and
interoperability standards are
high EA priorities and must
account for more than just
technical architecture.
Enterprise architects must resist
the temptation to wait for more
architecture content before
setting governance processes,
and instead develop content
and governance in parallel.
Key messages about EA are not
intuitively obvious, so enterprise
architects must work to educate
the business.
How to measure and
communicate the impact of
EA on ultimate end-users?
How do people reflexively
judge the relevance of EA for
themselves? Do different
reflexive types assess value
differently and if so, what
mechanisms can help their
acceptance?
How should architects allow
for the particular technical
and social characteristics of
impacted business units?
How can governance be best
represented and applied to
different stakeholder groups?
What communication
mechanisms need to be put
in place?
Gravesen (2012) found that EA implementation is still considered immature, and many
organisations remain ambivalent or sceptical. This is because the term EA means little to most
executives and many view it as simply another IT-related acronym that rarely delivers all of
the promised benefits. Even amongst IT practitioners and enterprise architects,
misconceptions prevail around the definition of EA and what it delivers in terms of business
benefits, which causes further ambiguity in attempts to explain its value. Bloomberg (2014),
33
also reported that “…responsiveness benefits that EA has purported to deliver have been few
and far between. Stories of stalled or misdirected EA initiatives vastly outnumber bona fide
examples of EA efforts leading to measurable business value.”
From an academic perspective, Boh et al. (2003) argued that most EA research does not focus
on standards management within organisations and suggested this lack of attention could be
the cause of EA failure. Boh et al. (2003) hypothesises that effective management of
integration architecture standards should govern the interactions of architects, IT personnel
and line management, and should be driven by the business goals. It recommends a matrix
structure for architecture teams, with different levels of information exchange and
collaboration between all stakeholders to produce positive outcomes. Raadt et al. (2008, p.
19) concluded that “efficient collaboration between architects and EA stakeholders is one of
the main critical success factors for EA. The basis for efficient collaboration between
architects and EA stakeholders is mutual understanding”. These authors went on to explain
that most EA research focuses on the role of the architect, with little attention paid to the EA
stakeholder, which could explain why effective EA implementation may appear to be a
daunting task and potentially fail. Another study of EA compliance by Čyras and Riedl (2012)
found that organisations encountered problems with IT compliance issues which were too
large and too complex for any one organisation to solve. This occurred because most
organisations, through their enterprise architects, adopted the simplest of EA frameworks for
the sake of usability, and as a result, focused on few real-world IS issues.
2.4 Enterprise Architecture Implementation Challenge and
Governance
Several challenges related to governing change arise throughout the development, evolution
and implementation of EA. A top-down approach is common, since management support and
financial resources are vital for bringing about sustainable change. However, more recent
investigations concluded that the trend in EA implementation to attack from the top down
(the CxO perspective) does not have a significant impact on the success of the program. For
example, Bente et al. (2012) found that:
34
Any optimization for EA needs to make sure that all stakeholders are on
board and that they remain involved along the way—the business, the IT
crowd on the ground, management, and the enterprise architects
themselves. This idea can be summed up in two main challenges:
•
How can we structure EA activities on a day-to-day basis in order to
master a demand driven workflow at all levels of operation and
achieve a holistic result?
•
How can we elicit the participation of all, in particular the ground-level
stakeholders, to balance the helicopter view and the ground-level
perspective?
Our answer to these questions is to introduce lean, agile, and participation
concepts into EA (pp. 26 – 27).
Bente et al.’s (2012) investigation is strongly linked to Land et al.’s (2009) concept of EA
governance as a bottom-up approach in order to understand the consequences at a lower
level and translate them into effects at a higher level. Land et al. (2009, p. 1) defined
enterprise as “…a goal-oriented cooperative to be implemented by people and means” and
“…the key drivers for these means therefore are the enabling of informed decision making on
these changes, as well as ensuring compliance to these decisions.” These authors viewed the
EA program as a means of governing the changes.
The TOGAF (2011) concept of exponential growth in governance challenges focused on the
scope of the enterprise as one of the main challenges of EA:
The scope of the enterprise, and whether it is federated, will determine
those stakeholders who will derive most benefit from the enterprise
Architecture Capability. It is imperative that a sponsor is appointed at this
stage to ensure that the resultant activity has resources to proceed and the
clear support of the business management. The enterprise may encompass
many organizations and the duties of the sponsor are to ensure that all
stakeholders are included in defining, establishing, and using the
Architecture Capability (p. 59).
35
Janssen’s (2012) assumptions are useful for evaluating the TOGAF (2011) extrapolation of
governance challenges. He suggested EA approaches demand capabilities that go beyond
methodology and framework, although these are necessary, and concluded:
The key to relational capability is to ensure voluntary and collaborative
behaviour based on mutual trust and goodwill. Stakeholder participation
should balance the IT and business involvement and ensure the resolving
divergent perspectives and stakeholders’ conflicts. Relational capabilities
should stress the shared learning and dialogues among all stakeholders. This
should ensure that EA are understood and will be used.” [Thus], “There is a
need for governance structures and mechanisms through which stakeholder
influence can be tunnelled and understood (p. 34).
Ross et al. (2006) described the important role of program mechanisms in EA implementation,
and suggested a focus on mechanisms can provide a basis for understanding the contextual
factors that drive such large-scale architectural transformation. These researchers
recommended using different management mechanisms, as stakeholders might view EA
implementation as an unnecessary constraint on their current projects.
As described in Figure 2.1, Ross et al. (2006) argued the three different levels of governance
within the program: the companywide level, the business unit level, and the project team
level should all be coordinated. At the company level the system establishes high-level goals
and incentives and might share certain mechanisms, such as executive commitment and
budgeting. They proposed decision-making processes at this level should be driven by the
program; at the business unit level the linking mechanisms will ensure that the architecture
reflects and informs the goals and priorities of all parties (see Figure 2.2); while at the project
team level, the architecture will ensure that all projects align with the program.
36
Figure 2.1: The Systems of Governance Mechanisms (adapted from Ross et al., 2006, p. 120)
Figure 2.2: Types of Linking Mechanisms (adapted from Ross et al., 2006, p. 128)
37
Figure 2.2 shows the three important types of linking mechanisms for any system of
governance, referred to by Ross et al. (2006, p. 128) as the IT Engagement Model. These are
a) the architecture linkage; b) the business linkage; and c) the alignment linkage. Ross et al.
proposed these linking mechanisms should also be applied to companywide governance and
business-unit projects:
These three types of linking mechanisms address the key alignment and
coordination concerns of the company as long as key stakeholders take
responsibility for them – and IT governance and project management are
effective (p. 128).
The architecture linkage encompasses multiple mechanisms for connecting IT governance
decisions about architecture goals and the project design. These include a) the architecture
review board; b) the architect training program; and c) the architecture exception processes.
In the same way, the business linkage also incorporates two mechanisms for coordinating unit
projects and architecture transformation efforts: a) the process of designing and updating the
business architecture; and b) incentive programs to guide behaviour as new projects demand
new ways of thinking. Business linkage ensures that business goals are effectively translated
into project goals. The alignment linkage ensures ongoing communication and negotiation
between IT and business concerns – why the changes are needed – the motivation, the
anticipated benefits of the change, where changes are expected to be made, and what the
expected changes may be (Spragg, 2015). This alignment linkage is comprised of multiple
mechanisms as follows: a) project management office; b) training and certification; and c)
metrics for assessing the project.
In summary, these program mechanisms (or theories) play an important role, not only in
terms of the implementation itself, but also in terms of obtaining consensus from the
stakeholders touched by it. Some program mechanisms may be considered an instrument of
the technical context, while others are focused on the particular context of EA. In terms of
this study which views EA as a social program, understanding such theories within the social
context of EA implementation is helpful to determine the conditions under which program
mechanisms may or may not work.
38
2.5 Enterprise Architecture Implementation in the Higher
Education Sector
There are limited empirical studies on EA implementation in the higher education (HE) sector.
One extensive EA research project was undertaken by the Joint Information Systems
Committee (JISC, 2008) in the United Kingdom, they provided funds for the project since the
beginning of 2008, to allow early adopters (education institutions: i.e. universities, colleges)
to explore the use of EA in the HE sector. In this JISC research project, Anderson and
Backhouse (2009, p. 26) conducted an in-depth investigation into the extent to which HE
differs from industry “and whether or not the differences make it a ‘special case’ in terms of
EA”. They concluded that “…the vision for EA in HE needs to be brave and bold. Making a
success of EA within the sector will require an uncompromising focus on the big picture vision,
both within institutions and across the sector (p. 1).” According to Anderson and Backhouse
(2009) the major concern is whether HE as a sector “is ready to listen to the benefits of EA.”
They identified that:
Various project staff expressed concern that senior management are
worried about a general ‘initiative overload’ and EA would simply be one
more acronym to deal with. Related to this were concerns about the
naturally conservative nature of universities, which find institutional change
difficult. This can lead to problems over getting decisions on and articulating
the 'to be', big picture vision. Also, TOGAF [the implementation framework
used in this EA research project] assumes that some of the key strategic and
business vision cornerstones are in place prior to inception of an EA
programme and sometimes this is not the case in HE (p. 26).
With respect to the naturally conservative nature of universities highlighted above, it could
be argued that EA implementation in the higher education sector is unique. As explained by
Anderson and Backhouse (2009) “…the collegiate system, common among older university,
follows a highly federated ‘community of scholars’ model” (p. 8), which would find
institutional change difficult. They also wrote:
39
…evidence from the commercial world indicates that the kind of business
change that EA sets out to generate can only happen when senior
management are fully engaged with EA work and are driving it from the top.
However, this type of approach does not sit easily within the often highly
decentralised HE environment. Indeed, there is often a cultural tension in
HEIs [higher education institutions] between top-down and bottom-up that
does not exist in industry to anything like the same extent and this poses a
particular challenge to the implementation of EA in HE (p. 9).
Another interesting state of affairs to emerge from the JISC research project is the social
interaction at the structural level. Anderson and Backhouse (2009, p. 9) found:
One point that did seem to get common agreement was to start EA work by
focusing on 'central' processes that everyone uses, most likely key
administrative systems such as Human Resources and Accounts. This was
partly a pragmatic decision that came out of the rejection of areas such as
teaching and learning where it was generally felt that there was so much
diversity, for example in pedagogy models, that getting a consensus that
could be acted upon would be difficult.
It appears EA implementation in the higher education sector presents particular challenges
that need to be addressed, most notably issues associated with participation of stakeholders
(people), and often unknown social circumstances (structure and culture). A different way of
implementing EA may therefore be appropriate for the HE sector as outlined by Anderson and
Backhouse (2009):
There was general consensus that identifying stakeholders is of primary
importance to the start-up of EA. Identifying key actors and decision makers
has been shown to generate debate within an institution and this is clearly
one of the benefits of embarking on EA work. However, a particular problem
is that some roles within universities vary so widely. What can we say is a
researcher? Many participants thought it was important to focus less on
labels within organisations, since roles and responsibilities do not always
40
follow the label, and to look more carefully at who makes what decisions (p.
28).
These factors contribute to how key TOGAF concepts have been implemented within various
departments in a university. As identified by Anderson and Backhouse (2009), the industry
cases were so dissimilar to the HE setting in the UK that it sparked debate about scope and
what constitutes an enterprise. For instance, the enterprise scope within a university and
“whether that should be across the entire institution, at departmental level or even at the
level of a single, cross-campus project” was a matter of contention. As a result, JISC put extra
work into refining the scale and complexity of work involved, specifically in applying TOGAF
as the implementation framework. Table 2.3 presents a number of key questions raised by
JISC, while the final column gives an indication of the focus of the current research study from
TOGAF EA continuum2.3 perspectives. This EA continuum provides recommendations for
investigating potential success mechanisms to answer key JISC questions.
Box 2.3: Enterprise Architecture Continuum
Architecture continuum: A repository of architectural elements with increasing
detail and specialization. This Continuum begins with foundational definitions like
reference models, core strategies, and basic building blocks. From there it expands
to Industry Architectures and all the way to an organization’s specific architecture.
TOGAF (2011, p. 21)
Enterprise continuum: A categorization mechanism useful for classifying
architecture and solution artifacts, both internal and external to the Architecture
Repository, as they evolve from generic Foundation Architectures to OrganizationSpecific Architectures. TOGAF (2011, p. 25)
41
Table 2.3 Main Challenges of EA Implementation in the Higher Education Sector
(adapted from Anderson and Backhouse (2009, p. 9)
No
Key Question
EA Continuum Views of
TOGAF
1
What is the correct scope to initially work at: the
whole institution, an individual department or single
project?
Who should be involved in the process as
stakeholders and what level of senior management
commitment is required?
Should an institution start with the ‘as-is’ aspect of
EA or the ‘to-be’?
Architecture vision, principles,
and strategic goals.
2
3
4
Does HE have the level of business planning and
strategic vision required by the ‘to-be’ phases of EA?
Architecture governance,
communication and
stakeholder management.
Architecture compliance and
conformance within its
domains: business, data,
application, and technology.
Architecture compliance –
target architecture.
Below is a summary of key EA lessons identified by the JISC research project (Anderson &
Backhouse, 2009, p. 34):
• Governance of the EA process is extremely important.
• Architecture work should not be the sole preserve of the IS department.
• There is a tendency to focus EA work on systems and technologies, but
actually the big areas of work involve people and their day-to-day work
processes.
• Keep EA work and those doing it tightly integrated and communicating
with the rest of the organisation: it is easy to develop an 'ivory tower'
mentality.
• Approach EA work with a view that it is an enabling and empowering tool
rather than a management ‘control’ tool.
• Understand and articulate the institution’s core operating model.
• It is very important to get senior management engaged with EA work.
• All institutions have an architecture, but without formal EA work it is not
modelled, mapped or understood.
42
• Avoid agonising over a governance infrastructure, select something and
then fine-tune as things go along.
2.6 Enterprise Architecture Adoption: Technical Capability versus
Social Intervention
EA programs can be understood as the implementation of changes within certain contextual
guidelines. As explained by Van den Berg and Van Steenbergen (2006), most of the
implementation will be a reflection of management’s belief that the program can only realise
its promise when the architectural thinking is adopted by those who initiate and execute the
changes, such as sponsors, business managers, IT managers, project managers, information
analysts, designers, builders and administrators. However, as identified by Bente et al. (2012),
the evidence in EA adoption to date is contradictory. Among stakeholders there will almost
certainly be a group that thinks EA is worth the trouble in view of the IT complexity and
continuous changes in technology, since people must relearn processes and will influence
current projects and affect existing solutions. Ross et al. (2006) found commitment to
technical capability did not translate into the IT application with the best fit in terms of
functionality, because it did not align with stakeholders’ personal projects. It therefore comes
as no surprise that for EA to be used as an instrument to steer change, the social context must
be embedded into the relevant change processes.
In view of the challenges posed by social intervention, it can be argued that EA doesn’t only
deal with technical aspects such as IT systems and infrastructure, since the enterprise context
is represented by people within an organisation who, amongst other things, use IT to do
business. Some of the social interventions identified in the literature, such as the social
structures and cultural attributes of the people element in the functioning of an enterprise
are: stakeholders’ collaborative processes (Ross et al., 2006; Boh et al., 2003; Raadt et al.,
2008), political dynamics, (Janssen, 2012), effective communication (Gravesen, 2012), and
work culture (Bente et al., 2012).
For these reasons it is logical to assume that the social context of EA can help an enterprise
to understand the real challenges associated with people issues. It can provide a strategic
approach to evolving IT systems that can deal with the social complexity of the environment
43
and manage the change of its IT systems. EA programs need to find their proper place in
complex and unique environments through collaborative development that enables the
organisation to better deal with the social interventions of the program. As suggested by
Bente et al. (2006, p. 9):
The IT systems [in EA program] must support new ways of doing business
collaboratively with partners and customers. The result is a “multi-entity”
ecosystem that allows interaction at more touchpoints and a depth not
previously attempted.
Janssen’s (2012, p. 28) analysis of organisations’ social structure in EA implementation also
looks at collaborative development with a socio-political focus:
…the socio-political perspective on EA looks at other elements like
collaborating among stakeholders including aspects such as trust, goodwill,
power, and mutual interests. EA is an activity in which many, diverse
stakeholders are involved, all having their own objectives. Alignment and
integration require understanding of each other’s needs and requirements
that go beyond the definition of models at various levels.
Both Bente et al. (2012) and Janssen (2012) explored the relationship between the technical
aspects and social context of EA programs, but Janssen’s interpretations includes a sociopolitical perspective on EA, and other crucial aspects such as creation of a shared vision,
communication among stakeholders, and evaluation of the impact as the crucial aspects.
2.7 A Contingency View of People as the Key Element in EA
Implementation
According to Ross et al. (2006) a key program mechanism in most organisations is the
engagement of senior executives who deal with IT. Their involvement in IT decision making is
crucial to establish principles and priorities for IT investment, as they will ultimately motivate
use of the program and ensure everyone understands the benefits it will bring. While most
people involved in the program attempt to do what they think is right, without a clear
direction, shared vision, collaborative process and effective communication, some of their
44
actions will do as much harm as good (Ross et al., 2006, Boh et al., 2003; Raadt et al., 2008;
Gravesen, 2012). Ross et al. (2006, p. ix) found that:
…people make [a] difference. Good people design the operating model,
build the foundation, execute, and innovate. But good people need
direction, leadership, and incentive to perform at their best.
Janssen (2012) expressed the opinion that EA implementation is not a single activity with a
clear beginning and end. He argued EA is influenced by ongoing use and users continually
interpret and extend the program and provide feedback for improvement and review,
thereby influencing its shape. Similarly, Van Den Berg and Van Steenbergen (2006) stated:
In most cases, the people directly involved have their own personal view of
the importance of architecture…” (p. 14), thus, “…the various people
involved [in the EA program] have different objectives in mind, expectations
may diverge. If this happens, support and approval for architectural
practices may disappear (p. 17).
The above arguments are strongly linked to Fairhead and Good’s (cited in Saha, 2009) notion
of people-led enterprise architecture:
[EA program should] highlights the need to ensure that enterprise
architecture, as both a discipline and set of deliverables, recognises the need
to focus on people before technology. (p. 285)" Thus, "Architecture is about
change, which means that it is about people. It is often stated that people
do not like change. This is not necessarily true but it is certain that they hate
having change forced upon them without their understanding and
involvement (p. 287).
It is evident that implementing EA programs is a hot spot with regard to people issues. As
claimed by Janssen and Klievink (2012, p. 29), among the reasons for the failure of EA
programs can be grouped into people category such as: undermined motivation, individual
capabilities, working relationships, absence of collaboration mechanism, failure to deal with
45
problems with employees, adding people to a late project, no sponsorship, and change in
stakeholders (contractors).
In their EA post-implementation review, Bente et al. (2012, p. 3) concluded: “after exhaustive
analysis and many discussions we found that each of the previously described situations
boiled down to people issues”. They ascribed this to EA programs on the borderline between
business and IT, critically dependent on collaboration between the two (Bente et al., 2012).
In a sense it sets the stage for structural and cultural clashes, a meeting point of business
people on the one hand and IT people on the other. In this context, the people in EA
management (senior executives) play a role as intermediaries in the decision-making
processes and in elaborating on drafts from the dialogue between business and IT. Since the
scope of EA is larger than enterprise architects can handle, extensive coordination and
collaboration is required between the business people and the IT people (Ross et al., 2006;
Bente et al., 2012).
In the HE sector, structural and cultural clashes are not only likely between IT and business
people, but also between service centres, faculties and schools. These tensions pose
particular challenges to the implementation of EA in the higher education sector (Anderson
& Backhouse, 2009). As explained by Anderson and Backhouse (2009):
Universities are organisations in which new ideas and technologies are often
being developed or experimented with. Individuals in departments are
frequently introducing new ways of working [...]. This is clearly not unique
to the university sector, but the scale and decentralised nature of it might
be, given that one of the primary roles of the sector is to act as 'ideas
communities'. [EA] Pilot participants were keen to debate whether and how
EA could be dynamic enough to incorporate new developments and 'leftfield thinking' (p. 27).
The idea of people as the key element in EA implementation provides a useful vocabulary for
examining the overlap between people, processes and technology, particularly in explaining
EA adoption in the HE sector. It is therefore important to examine people’s perceptions of the
46
processes and technologies, as their perceptions ultimately lead them to change their habits
and routines.
If people do not like the proposed transformation they may reject its implementation. Hence,
EA implementation will only be accepted if the program is aligned with their projects and
preferred programs. Their perception of the program will determine how they respond to it
and consequently, their adoption of EA. It will be interesting to examine the role of internal
relationships in the way that programs impact their ultimate decisions about EA
implementation.
2.8 TOGAF and the People Element
TOGAF is a detailed method and set of supporting tools for developing enterprise architecture
(TOGAF, 2011). It is the most widely used EA framework (judged by published certification
numbers) and can be used freely by any organisation wishing to develop and implement EA.
Lange and Mendling (2011) endorsed TOGAF as the most referenced EA framework, and more
recently, Mueller et al. (2013, p. 3) claimed that TOGAF is “…the widest and most extensive
framework for Enterprise Architecture Management (EAM)…” that provides “…methods and
supporting resources to improve business efficiency by building suitable EA”. TOGAF has also
become the most referenced EA implementation framework for the HE sector (Anderson &
Backhouse 2009; JISC 2009b).
One of the challenges of EA is compliance issues (Čyras & Riedl, 2012). The TOGAF framework
emphasises the importance of architectural compliance between the architectural
specification and ultimate outcomes. Compliance can be seen to vary between nonconformant, fully conformant, conformant, compliant, consistent and irrelevant. Figure 2.3
illustrates the detail of each level of architecture conformance.
47
Figure 2.3: Levels of Architecture Conformance (adapted from TOGAF, 2011, p. 560)
Such a view suggests the various levels of non-conformance equate to people’s acceptance
of the inbuilt structures and mechanisms. As detailed in Table 2.4 below, TOGAF defined
various important people elements together with associated responsibilities for achieving
conformance. This table provides guidance for the important roles, relationships and
mechanisms involved in a TOGAF based implementation.
48
Table 2.4. ‘People’ Roles in EA Compliance within EA Implementation
(adapted from TOGAF, 2011, p. 565)
No.
1
People Role
Architecture Board
Responsibilities
TOGAF Notes
To ensure that IT architecture is
Sponsor and
consistent and support overall
monitor architecture
business needs
activities
2
Project Leader
Responsible for the whole project
3
Architecture Review
To administer the whole
More likely to be
Co-ordinator
architecture development and
business oriented
review process
than technology
oriented
4
Lead Enterprise
To ensure that the architecture is
An IT architecture
Architect
technically coherent and future-
specialist
proof
5
Architect
One of the Lead Enterprise
Architect’s technical assistants
6
Customer
To ensure that business
Manages that part
requirements are clearly expressed
of the organisation
and understood
that will depend on
the success of the IT
described in the
architecture
7
Business Domain
To ensure that the processes to
Knows how the
Expert
satisfy the business requirements
business domain
are justified and understood
operates; may also
be the customer
8
Project Principals
To ensure that the architects have
Members of the
a sufficiently detailed
customer’s
understanding of the customer
organisation who
department’s processes. They can
have input to the
provide input to the business
business
domain expert or to the architects
requirements that
the architecture is to
address
49
Similarly, Table 2.5 describes the possible milestones to ensure compliance over the period
of EA implementation and provides guidance for the important milestones and core events
that need to be considered. By observing any anomalous outcomes from these core events,
the program can focus on the deep underlying mechanisms for failure or success.
Table 2.5 ‘People’ Interaction and Action in EA Compliance during EA Implementation
(adapted from TOGAF, 2011, p. 566)
No.
People Action
1 Request architecture review
2
4
Identify responsible part of
organisation and relevant
project principals
Identify Lead Enterprise
Architect and other
architects
Determine scope of review
5
Tailor checklists
6
Schedule Architecture
Review Meeting
7
Interview project principals
3
TOGAF Notes
As mandated by IT
governance policies and
procedures
Architecture Review Coordinator
Identify which other
business
units/departments are
involved. Understand
where the system fits into
the corporate
architecture framework
To address the business
requirements
To get background and
technical information:
• For internal project: in
person
• For procurement
activities: in person
8
Analyse completed
checklists
Who (Interaction)
Anyone, whether IT or
business oriented, with
an interest in or
responsibility for the
business area affected
Architecture Review Coordinator
Review against corporate
standards. Identify and
Architecture Review Coordinator
Lead Enterprise
Architect
Architecture Review Coordinator with
collaboration of Lead
Enterprise Architect
Lead Enterprise
Architect and/or
Architect, Project
Leader, and Customers
Lead Enterprise
Architect
50
No.
9
People Action
10
Prepare Architecture
Compliance review report
Present review findings
11
Accept review and sign off
12
Send assessment
report/summary to
Architecture Review
Coordinator
TOGAF Notes
resolve issues. Determine
recommendations
May involve supporting
staff
To Customer
To Architecture Board
Who (Interaction)
Lead Enterprise
Architect
Lead Enterprise
Architect
Architecture Board and
Customer
Lead Enterprise
Architect
TOGAF does not currently provide a formal maturity model in its documentation, but refers
to a number of external maturity models for guidance. One of these is the US Department of
Commerce (DoC) Architecture Capability Maturity Model (ACMM) 2.4 designed to aid internal
assessments of EA readiness. The goal of this model is “to enhance the overall odds for
success of enterprise architecture by identifying weak areas and providing a defined
evolutionary path to improving the overall architecture process” (p. 596). It models the
cultural and technical readiness for adopting EA practices by proposing 6 maturity levels
focused on 9 elements. While many of the elements require people involvement and
acceptance, the role of people is not explicitly defined within the framework and the focus is
clearly quite technical. Such a limited framework underscores the lack of recognition of the
role of people beyond purely technical capacities. Given that TOGAF is the most popular EA
framework, it is disconcerting to see so much emphasis placed on powerful management
elements and so little on affected stakeholders.
51
Box 2.4: TOGAF ACMM – Architecture Capability Maturity Model
The DoC ACMM (TOGAF, 2011, pp. 596 – 600) consists of six maturity levels and nine
architecture elements. The six levels are:
0.
1.
2.
3.
4.
5.
None
Initial
Under development
Defined
Managed
Measured
The nine enterprise architecture elements are:
1.
2.
3.
4.
5.
6.
7.
8.
9.
Architecture process
Architecture development
Business linkage
Senior management involvement
Operating unit participation
Architecture communication
IT security
Architecture governance
IT investment and acquisition strategy
Level 0: None
No enterprise architecture program. No enterprise architecture to speak of.
Level 1: Initial
Informal enterprise architecture process underway.
1. Processes are ad hoc and localized. Some enterprise architecture processes are
defined. There is no unified architecture process across technologies or business
processes. Success depends on individual efforts.
2. Enterprise architecture processes, documentation, and standards are established
by a variety of ad hoc means and are localized or informal.
3. Minimal, or implicit linkage to business strategies or business drivers.
4. Limited management team awareness or involvement in the architecture process.
5. Limited operating unit acceptance of the enterprise architecture process.
6. The latest version of the operating unit’s enterprise architecture documentation
is on the web. Little communication exists about the enterprise architecture
process and possible process improvements.
7. IT security considerations are ad hoc and localized.
8. No explicit governance of architectural standards.
9. Little or no involvement of strategic planning and acquisition personnel in the
enterprise architecture process. Little or no adherence to existing standards.
52
Box 2.4 cont.: TOGAF ACMM – Architecture Capability Maturity Model
Level 2: Under Development
Enterprise architecture process is under development.
1. Basic enterprise architecture process is documented based on OMB Circular A-130
and Department of Commerce Enterprise Architecture Guidance. The architecture
process has developed clear roles and responsibilities.
2. IT vision, principles, business linkages, Baseline, and Target Architecture are
identified. Architecture standards exist, but not necessarily linked to Target
Architecture. Technical Reference Model (TRM) and Standards Profile framework
established.
3. Explicit linkage to business strategies.
4. Management awareness of architecture effort.
5. Responsibilities are assigned and work is underway.
6. The DoC and operating unit enterprise architecture web pages are updated
periodically and are used to document architecture deliverables.
7. IT security architecture has defined clear roles and responsibilities.
8. Governance of a few architectural standards and some adherence to existing
Standards Profile.
9. Little or no formal governance of IT investment and acquisition strategy. Operating
unit demonstrates some adherence to existing Standards Profile.
Level 3: Defined
Defined enterprise architecture including detailed written procedures and TRM.
1. The architecture is well defined and communicated to IT staff and business
management with operating unit IT responsibilities. The process is largely
followed.
2. Gap analysis and Migration Plan are completed. Fully developed TRM and
Standards Profile. IT goals and methods are identified.
3. Enterprise architecture is integrated with capital planning and investment control.
4. Senior management team aware of and supportive of the enterprise-wide
architecture process. Management actively supports architectural standards.
5. Most elements of operating unit show acceptance of or are actively participating
in the enterprise architecture process.
6. Architecture documents updated regularly on DoC enterprise architecture web
page.
7. IT security architecture Standards Profile is fully developed and is integrated with
enterprise architecture.
8. Explicit documented governance of majority of IT investments.
9. IT acquisition strategy exists and includes compliance measures to IT enterprise
architecture. Cost benefits are considered in identifying projects.
53
Box 2.4 cont.: TOGAF ACMM – Architecture Capability Maturity Model
Level 4: Managed
Managed and measured enterprise architecture process.
1. Enterprise architecture process is part of the culture. Quality metrics associated
with the architecture process are captured.
2. Enterprise architecture documentation is updated on a regular cycle to reflect the
updated enterprise architecture. Business, Data, Application, and Technology
Architectures defined by appropriate de jure and de facto standards.
3. Capital planning and investment control are adjusted based on the feedback
received and lessons learned from updated enterprise architecture. Per iodic reexamination of business drivers.
4. Senior management team directly involved in the architecture review process.
5. The entire operating unit accepts and actively participates in the enterprise
architecture process.
6. Architecture documents are updated regularly, and frequently reviewed for latest
architecture developments/standards.
7. Performance metrics associated with IT security architecture are captured.
8. Explicit governance of all IT investments. For mal processes for managing variances
feed back into enterprise architecture.
9. All planned IT acquisitions and purchases are guided and governed by the
enterprise architecture.
Level 5: Optimizing
Continuous improvement of enterprise architecture process.
1. Concerted efforts to optimize and continuously improve architecture process.
2. A standards and waivers process is used to improve architecture development
process.
3. Architecture process metrics are used to optimize and drive business linkages.
Business involved in the continuous process improvements of enterprise
architecture.
4. Senior management involvement in optimizing process improvements in
architecture development and governance.
5. Feedback on architecture process from all operating unit elements is used to drive
architecture process improvements.
6. Architecture documents are used by every decision-maker in the organization for
every IT related business decision.
7. Feedback from IT security architecture metrics are used to drive architecture
process improvements.
8. Explicit governance of all IT investments. A standards and waivers process is used
to make governance-process improvements.
9. No unplanned IT investment or acquisition activity.
Source: TOGAF (2011, pp. 596 – 600)
54
Despite these issues, many of TOGAF’s overall goals offer potential synergies to guide EA
development of technology and standards, particularly in the HE sector (Anderson &
Backhouse, 2009). For example the ADM2.5; a methodology that provides a step-by-step
process for developing and implementing an EA program.
Box 2.5: TOGAF ADM – Architecture Development Method
The TOGAF ADM provides a tested and repeatable process for developing
architectures. The ADM includes establishing an architecture framework,
developing architecture content, transitioning, and governing the realization of
architectures.
All of these activities are carried out within an iterative cycle of continuous
architecture definition and realization that allows organizations to transform their
enterprises in a controlled manner in response to business goals and
opportunities. (TOGAF, 2011, p. 10)
The ADM attempts to increase usability and provide additional help with the process of EA
within the HE sector. However, consideration should also be given to how ADM is introduced
to HE staff, as described below (Anderson & Backhouse, 2009, p. 20):
• Avoid 'analysis paralysis' – there is a certain amount of ‘learning by doing’
involved in using TOGAF and project teams should not be overly worried
about getting it catastrophically wrong;
• Select somewhere to start and get going;
• Develop a basic skeleton of the architecture before filling in too much
detail (it is common for beginners to try to put in too much detail);
• Avoid the temptation to try and fully complete each step of the ADM
before starting others;
• Make use of a suitable standards-compliance and tool;
• Acknowledge that TOGAF can be adapted to fit the specific situation
within an organisation.
55
Anderson and Backhouse (2009, pp. 20-21) also reported the following specific benefits from
using TOGAF for architectural compliance:
• Communication and governance-related benefits of having a structured
method in place for architectural work. Conflicts between stakeholders
will inevitably arise and TOGAF, to a degree, offers a way of reaching a
compromise across the institution;
• A focus on the requirements of users and stakeholders, by, for example,
placing this at the centre of its framework model, which helps to enable
communication across the campus. These user requirements tend to
have permanence over and above technical solutions and allow some
continuity;
• Architectural principles and attendant templates provided by TOGAF
were very useful. The templates provided in the documentation forced
projects to express and communicate the rationale behind institutional
business and technology decisions at a high level.
2.9 Summary
This study views people as a key element of EA implementation, and accordingly people are
recognised as subjects, not objects. The literature review does look at other important
mechanisms not specifically related to “people” since they will also play a role in the
explanatory target of the study. However, people-focused mechanisms have been
emphasised and people-focused theory is used to explain how EA was examined. In doing so,
the study has theoretically placed mechanisms at the centre of the research, to understand
the theories and social context of EA implementation and to define program mechanisms and
the contexts for success and failure in the higher education sector.
As a social program, EA requires investigation of the program mechanisms built into
implementation and the social mechanisms constituting the context within which the
program operates. This helps to understand how and why the program works or fails to work.
A review of the current literature highlighted the centrality of program mechanisms and
suggested a focus on mechanisms to provide a basis for understanding the contextual factors
56
that drive large-scale architectural transformation. It is widely recognised that EA
encompasses numerous goals, various products and dimensions (Janssen, 2012), which could
explain the many views on program mechanisms. The diversity of possible program
mechanisms is also reflected in the terminology used in the EA literature. Potential success
mechanisms identified in the EA literature can be seen in Table 2.6 below.
Table 2.6 Potential Success Program Mechanisms Identified from EA Literature Review
Program
Mechanism
Individuals’
engagement
Stakeholders’
role
Role as a Mechanism
To ensure that all stakeholders are on board
and remain involved along the way - the
business people, the IT people on the
ground, management and the end-users
themselves.
To change stakeholders’ perceptions of the
architecture by addressing their concerns
and requirements; and by identifying the
trade-offs that will need to be made to
reconcile their potentially conflicting
concerns.
Literature
1. Gartner (2009)
2. Bente et al. (2012)
1. Ross et al. (2006)
2. Gartner (2009)
3. Anderson and
Backhouse (2009)
4. TOGAF (2011)
5. Bente et al. (2012)
6. Janssen (2012)
1. Ross et al. (2006)
2. Anderson and
Backhouse (2009)
3. TOGAF (2011)
4. Bente et al. (2012
5. Janssen (2012)
1. Ross et al. (2006)
Governance
To ensure business and IT projects achieve
objectives at: a) at companywide level, b) at
business unit level, and c) at project team
level.
Linkage
To ensure that the architecture reflects and
informs the goals and priorities of all parties
through: a) architecture linkage; b) business
linkage; and c) alignment linkage.
Collaboration
To ensure the EA program supports new
ways of doing business collaboratively with
partners and customers so that the result is
a “multi-entity” ecosystem that allows
interaction at more touchpoints and in more
depth.
1.
2.
3.
4.
Communication
To ensure ongoing communication
1. Gartner (2009)
Ross et al. (2006)
Gartner (2009)
Bente et al. (2012)
Janssen and Klievink,
(2012)
57
Program
Mechanism
Compliance
Conformance
Shared vision
Sociopolitical
Leadership
Role as a Mechanism
and negotiation between IT and business
centres – why the changes are needed (the
motivation), the anticipated benefits, where
the changes are expected to be made and
what the expected changes may be. This
mechanism also assures sufficient
stakeholder understanding and support.
To ensure various important people and
their associated responsibilities adhere to
architectural compliance over the period of
EA implementation.
To ensure all features in the architecture are
implemented in accordance with the
specifications.
To ensure that EA principles, business
linkages, baseline, and target architecture
are identified.
To ensure all political aspects such as trust,
goodwill, power, and mutual interests of
stakeholders are embodied in the EA
implementation, from the creation of a
shared vision to communication with
stakeholders and impact evaluation.
To understand the importance of
communication and team building as key
critical success factors. A combination of
social skills, business focus, IT literacy, and
an ability to lead are crucial to the success of
EA programs.
Literature
2. Anderson and
Backhouse (2009)
3. Gravesen (2012)
4. Bente et al. (2012)
1. TOGAF (2011)
2. Čyras and Riedl (2012)
1. TOGAF (2011)
1. Janssen (2012)
1. Janssen (2012)
1. Ross et al. (2006)
2. Gartner (2009)
3. TOGAF (2011)
A review of the EA literature revealed gaps in existing knowledge and led to the formulation
of the research questions as a basis for exploring the interplay between structures, cultures
and agency (Chapter 4). The following chapter examines EA from various social critical realist
perspectives, and solidifies the use of critical realism as the underlying philosophy for this
examination of the social factors and their impact on EA implementation.
58
CHAPTER THREE: WHY CRITICAL REALISM?
AN EXAMINATION OF ENTERPRISE
ARCHITECTURE FROM A SOCIAL REALIST
PERSPECTIVES
3.1 Overview
The previous chapter provided reasoned evidence from the literature to support the premise
that successful implementation of enterprise architecture (EA) requires organisation-wide
collaboration with a strong focus on social impacts. This chapter discusses EA and its
implementation from the social science perspective. The first section examines the different
versions of social realism theories, their underlying focus on ontology (the nature of its
theoretical perspective) and the methodological/theoretical implications that should be
considered. It describes the fundamental implications of the research process and the
ultimate outcomes. As Archer (1995) argues:
In any field of study, the nature of what exists cannot be unrelated to how it
is studied. This is a strong realist statement, which I endorse, but cannot
explore here. Instead, I want to examine the more modest proposition that
what is held to exist must influence considerations about how it should be
explained. In other words, what social reality is deemed to consist of (and
what is deemed non-existent) do affect how its explanation is approached
(p. 16).
The maintenance of a consistency between ontology and methodology is important in
understanding the targets and goals of research:
…the social ontology endorsed does play a powerful regulatory role vis-a-vis
the explanatory methodology for the basic reason that it conceptualizes
social reality in certain terms, thus identifying what there is to be explained
and also ruling out explanations in terms of entities or properties which are
deemed non-existent. Conversely, regulation is mutual, for what is held to
exist cannot remain immune from what is really, actually or factually found
59
to be the case. Such consistency is a general requirement and it usually
requires continuous two-way adjustments between ontology and
methodology to achieve and to sustain it as such (p. 17).
The next section examines the distinction between the terms used for mechanisms in various
social realism theories, and describes which definition is most suitable for each purpose.
Underpinned by a critical realist perspective, this section also provides a social definition of
generative mechanisms and describes the role of program mechanisms, and abduction in
explaining non-observable events and non-events generated by mechanisms. The final
section of this chapter examines the contextual influences of the ontological, theoretical and
mechanistic frameworks, and substantiates the use of critical realism as the underlying
philosophy of the study and why it is appropriate for the objectives of the research.
3.2 A Social Realist Perspective in Examining Enterprise
Architecture
The aim of this thesis was to examine deeply the sociological aspects of EA implementation
at the university under study. Such examination requires looking beyond everyday events to
find the deep causal mechanisms involved. Nash (1999, p. 449) suggested:
The purpose of sociology is to explain social events and processes. If
explanations are tied to ontology, as the argument has suggested then it is
necessary to decide what social entities are real and how that reality can be
described and demonstrated.
The following section examines a number of realist approaches, all fundamentally in
agreement with the critical realist approach proposed by Bhaskar, with a focus on the
implications of the different arguments for EA examination.
60
3.2.1 Bhaskar’s philosophy of critical realism
Bhaskar’s concept of critical realism (CR) distinguishes three ontological domains in reference
to social reality: the empirical, the actual and the real. Morén and Blom (2003) suggested:
The empirical domain consists of what we experience, directly or indirectly.
This domain is distinct from the actual domain where events happen
whether we experience them or not, because what happens in the world is
not the same as that which is observed. This domain is in turn different from
the real domain, where we also find the forces, mechanisms, which can
produce events in the world. [Thus] …abstractions are not to do with actual
events, but deal with what produces them (p. 44).
It is useful to examine TOGAF from a critical realist perspective. TOGAF (2011) was also the
implementation framework used by the organisation under study. The Open Group
Architectural Forum has developed TOGAF over many years by disseminating and
encouraging its use and providing certification opportunities. The certification program
involves examination following self-directed study or attendance at certification courses.
Such an approach to learning follows the traditional, teacher-centred [objectivist]
instructional learning model in that it sees learning as concerned with transmission of
knowledge, as being individual and involving a process of information reception, storage,
retrieval and comparison with others (Brown, 2009, p. 11). It differs from the student-centred
[social constructivist] learning approach which is “concerned with meaning-making, is social
and involves a process of internal and social negotiation (dialogue) and sharing with others”
(Jonassen & Land, 2000, as quoted in Brown, 2009, p. 11). Brown (2009) compared these two
common approaches to learning using a critical realist approach and argued for an
“ontological turn” in education. Brown’s (2009) argument for a critical realist approach to
learning has synergies with the traditional objectivist use in enterprise architecture, which
can be seen as an accumulated repository of knowledge that provides knowledge acquisition
opportunities. TOGAF (2011) assigned two meanings to “architecture” depending upon the
context: "1) A formal description of a system, or a detailed plan of the system at component
level to guide its implementation; and 2) The structure of components, their inter-
61
relationships, and the principles and guidelines governing their design and evolution over
time” (p. 9).
Brown (2009) agreed that such an objectivist representation has benefits for education:
The strengths of objectivism are that it accounts for the objective character
of public knowledge and for direct transmission as a teaching/learning
process; it does not account for variations in students’ construction of
meaning. Because constructivism does account for this variation, it would
seem to provide a preferable account of knowledge that is fluid, nonpropositional and subjective, that is, based on personal rather than external
(objective) criteria. However, constructivism does not provide a coherent
account of objective knowledge; indeed, strong versions of constructivism
deny it (p. 13).
In contrast to these positions Brown proposed a shift to a critical realism ontology which he
termed an “ontological turn” in educational theorising, concluding “...in critical realism it is
the ontology that enables and constrains the acquisition of knowledge, that is, learning” (p.
14).
In a similar fashion, it could be argued that a focus on the ontology of EA, its meaning and the
context of its application can provide a new, less-objectivist perspective. Seen from this
viewpoint the enterprise environment is not predefined; it provides a set of conditions that
enable and constrain adoption. The adoption environment is real and intransitive and
therefore “a) it exists whether or not we have (fallible) knowledge of it, and (b) its elements
have causal powers or susceptibilities or tendencies” (Brown, 2009, p. 17). CR thus
emphasises the context in which EA is developed and used, requiring a clear representation
of the EA “knowledge” or artefact repository, the EA “knower” or stakeholder, and the
process of “knowing” or developing and applying EA knowledge in the context of the knower.
Consequently, the focus in such a representation is contextual, and seeks to understand the
social elements that constrain and enable the effective transmission and use of EA
knowledge.
62
3.2.2 The morphogenetic approach and reflexivity: Theories of critical
realism
Archer (2013a, p. 9) described the morphogenetic approach (MA) as a methodological
complement for CR:
All theories have a social ontology, whether implicit or explicit, which
effectively defines the constituents of the social world. Therefore, the SO
[social ontology] performs a role of conceptual regulation because it governs
those concepts that are deemed admissible in description as in
explanation—just as an atheist cannot attribute his well-being to divine
providence. In itself, a social ontology explains nothing, although it may
exclude certain explanations, cast in ‘improper’ terms. In itself, an SO tells
no one how to go about explaining anything. For this an explanatory
programme is needed. That is what the Morphogenetic Approach is; the
methodological complement of Critical Realism, which is its metatheoretical social ontology.
To emphasise the contribution of MA, Archer described the “morpho” element as an
acknowledgement that society has no pre-set form or preferred state, while the “genetic”
part is recognition that it takes its shape from, and is formed by, agents, originating from the
intended and unintended consequences of their activities. MA provides direction for critical
realist interpretation in that it reflects the fundamental CR assumption that social systems
require an analytical separation between macro (structural) and micro (agency) dimensions
and that: (a) structure necessarily pre-dates the action(s) leading to its reproduction or
transformation, and (b) structural elaboration necessarily post-dates the action sequences
that gave rise to it (Archer, 1995; Dobson, Jackson & Gengatharen, 2011). This analytical
separation of structure and agency in order to explain is a consequence of the different nature
of the two: “people and society ...do not constitute two moments of the same process. Rather
they refer to radically different things” (Bhaskar 1989, p. 76). Thus, an educational system can
be centralized, while a person cannot, and humans are emotional, which cannot be the case
for structures” (Archer 2010a, p. 275). Through analytical dualism the study can separate
“structure” (social systems, institutional structures, roles, and positions) and “agency”
63
(groups, collectives and individuals) and examine their interplay to account for the structuring
and re-structuring of the social order. This is possible because structure and agency are
different kinds of entities, with different emergent properties and powers, despite the fact
that they are crucial for each other’s formation, continuation and development.
In her MA, Archer assigned a central role to mechanisms and emphasised the recognition of
their role over time by introducing a temporal element: the important macro-micro, micromicro and micro-macro mechanisms are placed in context over time.
Alwadain et al. (2013) recommended the use of MA as an explanatory program for examining
the evolution of EA because its analytical separation of structure and agency over time
provides a powerful framework for examining the social complexity of EA implementation as
the various macro-micro situational mechanisms, the micro-micro interaction mechanisms
and the micro-macro transformational (or reproduction) mechanisms play out over time
(Figure 3.1). This socio-cultural focus provides clear opportunities for examining the interplay
of the various mechanisms and their roles in structural change.
Archer’s MA can also be used more broadly to examine the interplay between the cultural
(people ideas, thoughts, beliefs and values that influence their work) and the socio-cultural
practices (habitual actions, established routines, traditional preserves or conventional
divisions of activities: Archer, 1988). For Archer, the cultural and the socio-cultural form two
interacting morphogenetic cycles (Figure 3.1) that have profound and ongoing impacts on
each other, and thus on the social outcomes. The extent to which the socio-cultural and
cultural are in harmony define various situational logics that “predispose agents towards
specific courses of action for the promotion of their interests” (Archer 1995, p. 216). The
environment within which the university operates (that is, the organisation under study) can
be seen as a microcosm of the discontinuity evident in the greater society that Archer
described.
Figure 3.1: Morphogenetic/Morphostactic Approach (Archer, 1995, p. 193)
64
In similar fashion, Archer’s examination of the role of individual reflexivity has relevance to
an examination of EA adoption. It is interesting to speculate about the various reflexive
modalities prominent in universities today, each with its own different responses to change.
For example, her conclusions concerning the increased role of the autonomous reflexives
within today’s morphogenetic society can be extended to the university context, as
autonomous reflexives adopt “…a strategic stance towards constraints and enablements;
they seek to avoid society’s ‘snakes’ and to climb its ‘ladders’. They thus aim to improve upon
their social positioning and, if successful, become upwardly socially mobile”. In the same way
it could also be a meta-reflexive: if the academic staff of a university seek personal advocation
rather than a career, they will be compelled by the need to make a difference in their field of
study rather than simply meeting their immediate needs and desires. As Archer’s most recent
research indicates (Archer, 2014; Archer, 2015), the adoption of such a “people reflexivity”
approach encourages a focus on how EA needs to support the personal projects of staff in
order to have the best chance of success. It is now useful to examine other forms of social
realism to better understand the strength of CR. The next section briefly examines prominent
social realist approaches Bunge’s emergent systemism and analytical sociology.
3.2.3 Bunge’s realist philosophy of emergent systemism
According to Wan (2012), Bunge’s emergent systemism describes “...the ontological status of
supra-individual actors (cohesive social groups) as concrete systems with novel causal
powers, a bonding structure, and specific mechanisms that make it behave as a unit in certain
respects...” (p. 1555). The aim of Bunge’s ontology is to provide "...a completely general
model of concrete systems of any kind, living or non-living’’ which constitutes a general
framework for understanding supra-individual actors as concrete systems with emergent
properties and causal powers. Bunge argued a social system is “...ultimately the aggregate
outcome of individual actions”, and thus “…the features of a social system depend upon the
nature, strength, and variability of social relations, which in turn are reducible to social
actions” (Wan, 2012, p. 1548). Bunge made it clear that binding social relations are concrete
connections among people, in relations of competition, cooperation, trade and employment
(Kaidesoja, 2009, p. 310).
65
Kaidesoja (2009, p. 309) further demonstrated that Bunge’s social systems possess emergent
properties such as “…structure, cohesiveness, stability, stratification, norms, coordination,
division of labor, and distribution of wealth, that are not properties of the individual agents”,
but properties of a social group (a collection of individuals that share certain features, held
together by connections, bonds, or forces of some sort in cohesive social groups in an
organisation). The organisation is conceived as a structured system that behaves in certain
respects as a whole and interacts with other social organisations. From this perspective,
Bunge provides valuable insights into the human social systems evident within the university.
Such a perspective provides useful guidance for the examination of the emergent properties
of particular faculties, schools, and service centres as social groups. Based on this view
structure constitutes the properties of a particular group or entity – it is the properties of that
material group rather than a separate thing. Structure is to do with being a property of a
system, and as described by Wan (2012), the analytical separation is more related to system
and agent rather than to structure and agent evident in CR. The emergent properties of
systems have causal powers in the social world, and this view seemed most appropriate for
analysing the animosity of the various centres, schools and faculties towards the
implementation of EA.
3.2.4 Analytical sociology
According to Hedström and Ylikoski (2010) analytical sociology emphasises the importance of
mechanism-based explanations to help social scientists avoid some philosophical pitfalls –
“the mere adoption of mechanism talk will not suffice” (p. 58).
Figure 3.2: A Typology of Social Mechanisms (adapted from Hedström and Ylikoski 2010, p.
59)
66
Figure 3.2 shows Hedström and Ylikoski’s (2010, p. 59) mechanism-based explanation that
can never include solely macro-level mechanisms. They suggested:
A basic point of the mechanism perspective is that explanations that simply
relate macro properties to each other (arrow 4) are unsatisfactory. These
explanations do not specify the causal mechanisms by which macro
properties are related to each other. Deeper explanatory understanding
requires opening up the black box and finding the causal mechanisms that
have generated the macro-level observation. Rather than analysing
relationships between phenomena exclusively on the macro level, one
should identify the situational mechanisms by which social structures
constrain individuals’ action and cultural environments shape their desires
and beliefs (arrow 1), describe the action-formation mechanisms linking
individuals’ desires, beliefs, etc., to their actions (arrow 2), and specify the
transformational mechanisms by which individuals, through their actions
and interactions, generate various in-tended and unintended social
outcomes (arrow 3). Only by understanding the whole chain of situational,
action-formation, and transformational mechanisms have we made sense of
the observed macro-level relationship.
The aim of analytical sociology is to explain “…complex social processes by dissecting them,
accentuating their most important constituent parts, and constructing appropriate models to
understand the emergence of what is observed” (Wan 2012, p. 1545). Wan suggested that
the commitment of analytical sociology to the importance of micro and individuals in the
causal reconstruction of the processes that give rise to given collective phenomena, has
encouraged efforts to bring to the fore mechanism-based explanation grounded in a realist
approach to causality (p. 1546). Whilst the approach has been criticised for its exclusive focus
on the causal role of the micro and individuals in social mechanisms; it can nevertheless
provide a useful way of representing the role of individuals and examining their interactions
with existing structures.
67
3.2.5 The different under-labouring roles of social realism for enterprise
architecture examination
Foundational realist ontology is crucial for building a theoretical explanation of the key role
of people in EA implementation. The different under-labouring roles of social realism in EA
examination present different arguments about the basic conditions to test for the existence
of the phenomena studied. Since the role of theory in research is so decisive (Danermark,
Ekström, Jakobsen, & Karlsson, 2002) it has a significant impact on how EA is examined. As
Archer (2015) argues that social theories and approaches should be examined as to their
commitment to structure (S), agency (A) and culture (C). Ideally approaches should not be
“Lite” in any of these areas. Table 3.1 shows the different under-labouring roles for such
examination.
Table 3.1 Different under-labouring roles of social realist for EA examination
(Summarised around social realist conceptions of social reality and Archer (2015)
investigations on the processes of change)
Focus
Theory
Roles
Agency
Structure
Culture
(People)
Morphogenetic Approach
Heavy
Heavy
Heavy
Macro and Microscopic theme
Emergent Systemism
Medium
Heavy
Lite
Microscopic theme
Analytical Sociology
Lite
Heavy
Lite
Microscopic theme
The analytical sociology approach (Figure 3.2) appears similar to Archer's MA at a schematic
level (Figure 3.1). However, unlike Archer’s argument (morphogenetic approach) for
examining the macro-micro level between the structural, cultural and agency in creating
change, analytical sociology focuses on the micro elements of the change process. The “macro
level” of analytical sociology is only perceived as a constraint on individual action, and thus
rejects strong ontological version of structural-cultural emergence. Archer (2015) expressed
the opinion that within analytical sociology “…the resemblance is only superficial, not only
because social structures are treated as weakly emergent, but also because human persons
are treated as “rational actors” [that is, methodologically individualistic] (p. 30)”. Equally
important, Archer (2015) explained:
68
…the analytical sociology is epistemologically realist but not ontologically
realist, [thus], [Analytical sociology] admits the existence of non-observable
sub-individual level entities and process (e.g., conflicting desires and rational
choices) while denying the existence of supra-individual entities and
processes on the grounds that they are not observable. In this regard,
[analytical sociology] is still empiricist and not fully realist (p. 31).
In the same way, Bunge’s Emergent Systemism denies the power of culture to operate within
the social reality. Bunge focused on supra-individual actors (cohesive social groups) and
viewed social structure as a property of social groups, where culture is a part of the structure
itself. As identified by Archer (2015): “Bunge’s oscillation between the Cultural System and its
logical relations, and Socio-Cultural interaction and its use of ideas to influence social
relations, means he denies culture the power to operate as a generative mechanism…” (p.
10). By affording supra-individual actors the central (microscopic) theme, he claims that the
analytical separation is more related to social system (in which the social structure is also
located, along with other social components (e.g. cohesiveness, stability, stratification,
norms, coordination, division of labour, and distribution of wealth)) and agent, rather than
the structure and agent in CR.
To theorise about the role of people in an explanatory study, it is inadequate to focus only on
one or two part(s) of a social reality in which people are at its core. This is the “structure”
(people situations in their social systems, institutional structures, roles, and positions); the
“culture” (people ideas, thoughts, beliefs and values that influence their work); and the
“people” themselves (groups, collectives and individuals), which are intertwined with each
other (Archer, 1995). In researching the role of people, this study emphasises the interplay
between structure, culture and people (agency) as the basis for explaining people’s adoption
of EA practices, by identifying the mechanisms which hinder and encourage architecture
transformation. Accordingly, this study adopted a viewpoint that treats mechanisms as
ontological in providing an account of how the underlying mechanisms work in the three
overlapping ontological domains that define social reality (Archer, 1995). This representation
in Archer’s MA was useful for building a theoretical explanation in this study of the role of
people as a key element in EA implementation.
69
3.3 The Important Role of Mechanisms in Social Explanation
The aim of social science to provide a social explanation for the role of mechanisms has a long
history “…but only in recent decades has this idea been an object of more systematic study”
(Bechtel, 2006, as cited in Hedström & Ylikoski, 2010, p. 50). Hedström and Ylikoski (2010)
identified many definitions of “mechanism” conceived by researchers in different fields of
study “because the entities and processes studied by different sciences are quite
heterogeneous, it is difficult to propose a mechanism definition that would both be
informative and cover all examples of mechanisms” (p. 50). Similarly, Archer (2015, p. 3)
stated: “…there are now at least 24 definitions of mechanisms in social theory (Mahoney,
2001)”, largely within a ‘realist mechanismic framework’”. Hence, it is important to clarify
which is most usable for a specific study, and which definition is most suitable for that purpose
(Van Den Berg & Van Steenbergen, 2006). The following section examines a number of realist
approaches, all fundamentally in agreement with the critical realist approach, and their
different definitions of “mechanism”, as well as the associated implications of each definition.
3.3.1 Social mechanisms from various social realist perspectives
Bhaskar (1978; 1986, as cited in Demetriou, 2009, pp. 4-5) defined “mechanism” as
“ontologically linked to structure, within a view of reality that features structural pluralism at
various levels, thus holds mechanism to be that aspect of the inner and environmental
structure of a thing by virtue of which the thing has a certain power”. According to Bhaskar a
mechanism operates when triggered and normally endures longer than any pattern of events
it triggers.
In contrast, Bunge defined a mechanism as “…the collection of processes that occur within a
system (and often among systems in the case of the social world), bring about (or block) its
transformations, and alter (or maintain) its structure” (Wan 2012, p. 1557), whereby the
ontological concepts of system and mechanism are defined in terms of each other. According
to Demetriou (2009, p. 5), “…Bunge sees mechanisms essentially as constituent events, while
Bhaskar sees it as power-affiliated reality that is over and above any pattern of events it
generates”. He added: “...despite any differences of the definition of mechanism […], Bunge
and Bhaskar are on the same page when describing the complexity in which mechanisms
70
operate” (p. 6). Unlike Bhaskar’s (1978) deep realist scheme of structure and its complexity
in relation to the ontology of causality, Bunge’s (2003) realist concept of emergent systemism,
as described in Wan (2012), provides a contrasting perspective to that of the critical realist by
suggesting that “structure” is a property of a system rather than a “thing” in its own right.
Seen from this viewpoint “social structure” stands for “the set of relations among members
of a given social system and among items in the system’s environment, and the total social
structure of a society is defined as “the union of its biological, economic, political, and cultural
structures” (Bunge 1998a: 66). Since (social) structure is “a set (or collection) of relations, it is
a concept, not a concrete thing such as an organism, a person, or a group” (p. 1550).
Differentiating entities from structures allowed this study to handle distinctions between the
causal powers of unions, for example qua-entities, and the norms and regulations or quastructures around them. There was found to be value in this differentiation between entities
and structures in examining the resistance of certain faculties and service centres towards EA.
The custodian of IT Services in University X (the organisation under study) spoke about the
likelihood of a different response from Teaching and Learning to the rules and regulations
imposed by the EA program, mainly due to this entity’s different properties from other
centres. As such, their adoption of EA structuring processes and regulations may prove more
problematic. This is supported by Anderson and Backhouse’s (2009, p. 9) conclusion that
implementation of an EA program in the higher education sector would be difficult.
While such discussion may strengthen the argument that various “entities” will respond
differently to the EA structure imposed, the lack of focus of Bunge’s Emergent Systemism on
the causal powers in cultural/socio-cultural to operate as a generative mechanism, can cause
difficulty in identifying the cultural environments that influence people (agency) by
constraining and enabling certain habits and routines. Kaidesoja (2009, p. 308) argued that
Emergent Systemism focuses only on two levels of an organisation’s complex system at one
time:
…in studying social organizations it is often useful to methodologically
distinguish the following two levels of analysis: (i) the members of the
organization and their interactions, and (ii) the organization conceived as a
structured system that behaves in certain respects as a whole and interacts
71
with other social organizations. This model, then, is abstracted both from
the subsystems of the individual members of the organization (e.g. mental
systems and personality) and the supersystems (e.g. economy or state) of
which the organization forms only one part. Due to these abstractions, the
explanatory power of this kind of model might be rather limited, since
subsystemic and supersystemic processes may also be important in
explaining events and processes that take place in social organizations
(Bunge 1998, 73–79).
Hedström’s (2005) focus on mechanisms was instrumental in developing a new approach to
social investigation titled analytical sociology. Whilst this approach has been criticised for its
exclusive focus on the causal role of micro and individuals in social mechanisms, it
nevertheless provides a useful way of representing the role of individuals and examining their
interactions with existing structures. As Hedström and Bearman (2009) argued, analytical
sociology can be used as a strategy for understanding the social world, and can explain the
important social facts, such as network structures, patterns of residential segregation, typical
beliefs, cultural tastes, and common ways of acting, through social mechanisms. Manzo
(2010, p. 139) described it thus:
...a mechanism can legitimately constitute the basis for a general theoretical
proposition provided that it is shown to be at work behind different
phenomena and, for each of them, appears under various spatial-temporal
conditions involving actors with heterogeneous identities, beliefs and action
logics.
Such subtly different perspectives on mechanisms can offer significantly different explanatory
foci, because the seemingly small differences between the various ontological
representations provide different avenues for explanation. No single platform is sufficient –
all have their own benefits and disadvantages. EA implementation in universities has its own
uniqueness given the social complexity of the university environment. Analysis of such a
complex social environment is difficult and should include an investigation of the interplay
between structures, cultures and agency (people), all fundamental components of social
reality. The goal of this section is therefore to define CR and the theories (in particular Archer’s
72
Morphogenetic Approach) that can be used to build a theoretical explanation of the role of
people as a key element in EA implementation.
3.3.2 Generative mechanisms in social explanation under critical realist
perspective
According to Bhaskar’s definition, a generative mechanism (Bhaskar, 1979; 2008 as cited in
Archer, 2015, p. 3) is “a way of generative complexes”, therefore the mechanism provides the
real basis of causal laws and hence causal explanation. Detailed arguments, presented by
Archer (2014), suggest that “generative mechanisms are a key concept in the realist ontology
on which the present analysis of normativity is based […]”. She defines them as “…the
emergent causal powers of related entities within a system” (p. 195). Archer clearly
demonstrates the terminology of generative mechanisms as follows:
Firstly, there is nothing ‘mechanistic’, in the sense of deterministic, about
the notion of ‘mechanisms’. Indeed, while mechanisms acting in closed
systems will generate event regularities, mechanisms active in open systems
will generate tendencies against which other (often unknown or even
unknowable) mechanisms may countervail. Thus, the use of the word
mechanism does not preclude the use of ‘reflexivity’ or the attribution of
creative agency to people. Secondly, mechanisms play a central role in
explanations, as distinct from mere descriptions. Indeed, while analyses
confined to the level of description will comprise (typically interesting)
successions of or association between events, explanations must also
include both an identification and a discussion of those mechanisms held to
be minimally sufficient to account for the situation described to be the way
it is and not otherwise (see Lawson 2003, pp. 86–109 on contrast
explanations). Thirdly, mechanisms are typically nested into one another. To
my knowledge, there is no way of proving ontologically that this nesting has
an end and does not lead to infinite regress. From an epistemological
perspective, we know that we have reached a satisfactory explanation not
when we have identified all the mechanisms at play (while excluding other
hypotheses) but rather when we have outlined sufficient mechanisms (and
73
excluded alternative hypotheses) to answer the question at hand. In other
words: every mechanism can be used both as explicans (that which explains)
and as explicandum (that which is being explained) (pp. 195 – 196).
In similar fashion, Mayntz (2004) expressed the opinion that:
“The term “mechanism” is used both to designate a certain class of real
phenomena (mechanisms are such and such, they do such and such) and to
designate a class of (causal) propositions referring to such phenomena. (p.
239)” Thus, “…the search for mechanisms starts not with a correlation but
with the identification of an explanandum. The term “generative
mechanism” underlines this explanatory strategy. Processes generally do
not come as discernable, “given” units; they have no naturally given
beginning and end. (p. 244)”
Another essential point about generative mechanisms also articulated by Archer (2015, p. 2):
… ‘generative mechanisms’ are required to (a) explain such associations (i.e.
how they arose and work) and (b) are robust enough to account for cases
and times when no such ‘constant conjuncture’ can be found (i.e. Y is not
significantly correlated with Z) but do not entail scrapping the mechanism
itself.
From a critical realist perspective, generative mechanisms always exist in several diverse
elements (i.e. structure and culture) and thus are in interplay with one another (Archer, 2015,
p. 3).
Based on the concept of Bhaskar’s generative mechanism, Blom and Morén (2011, p. 63)
argued:
…the generative mechanisms that explain how and why the events
happened are only accessible indirectly by developing theory in relation to
those mechanisms. Clearly, the mechanisms are not less real for not being
directly observable, but exist (in the domain of the real) whether we
conceptualize them or not. Accordingly, this type of explanatory knowledge
74
demands a theoretical language that penetrates the empirical surface and
forges contact with the reality that exists beneath the level of events.
In the same way, Archer (2015, p. 70) pointed out that generative mechanisms are located in
the domain of real:
…the realist generative mechanisms do not operate at the level of events
but at the ‘real’ and unobservable level of the factors (structure, agency,
culture) working together in that relation.
EA is regarded as the starting point for the process of change, and is conceived of as
generative mechanisms that produce tendencies towards change in the relational
organisation of the social order. As explained by Blom and Morén (2011) “…generative
mechanisms actually exist in the social world, but they are to be regarded as potential or
tendencial (p. 63).” Obviously generative mechanisms underline the explanatory focus of the
CR philosophy and can be used to examine the social mechanisms surrounding the
architecture, as the identified generative mechanisms leading to architecture transformation
(or not) will be traceable from the architecture events. The mutual connection between the
generative mechanisms can explain the underlying success of large architecture
transformation, or known in MA terms as a “synthetic” picture of social transformation (or
reproduction) (Archer, 2015).
3.3.3 The role of ‘program mechanism’ from a critical realist perspective
Astbury and Leeuw (2010) described the important role that mechanisms play in social
programs and policy and suggested a focus on mechanisms can explain “…how and why
programs work (or fail to work) in different contexts and for different program stakeholders.
This is where the explicit use of mechanisms can play an important role in assisting theoryoriented evaluators to articulate more precisely the causal linkages between programs and
their desired effects” (Astbury & Leeuw, 2010, p. 364). These authors defined mechanisms as
“…underlying entities, processes, or structures which operate in particular contexts to
generate outcomes of interest” (p. 368).
75
To critical realists, mechanisms are seen as sensitive to variations in context, since a
mechanism may or may not be activated in a particular context due to contingent conditions
or possible countervailing mechanism(s). The important role of mechanisms for evaluating
social programs was studied by Pawson and Tilley (1997), who differentiated between the
mechanisms built into the program design (also referred to as the “program measures” or the
“program mechanisms”) and the social mechanisms, constituting the context within which
the program operates. They recommended an important focus on understanding the theory
underlying the program design, such theory being important in understanding and explaining
program outcomes. The bottom line of program evaluation is to determine under what
conditions the program theory might work or not – that is “what works”, “for whom” and “in
what circumstance”. Understanding the theory underpinning program implementation is
helpful in deriving the mechanisms by which change is achieved. Mechanisms are a
fundamental part of a critical realist explanation – they provide the causal possibilities within
any social program.
These objectives help to understand some of the mechanisms built into the EA
implementation program, such as the various consultation groups and support mechanisms
to facilitate frequent input from stakeholders. Given such aims, it is important that the
underlying philosophy and associated social focus allow a clear representation of such
elements in the EA program. It is also important to properly examine the success of these
program mechanisms to fully appreciate the success of the program.
3.3.4 The role of ‘abduction’ in explaining the non-observable events and
non-events generated by mechanisms
In critical realist studies, abduction plays a vital role in explaining and proposing the creative
use of theory for deriving suitable explanation. For example, for non-observable events and
non-events generated by mechanisms abduction requires creativity and imagination. Mingers
(2011, p. 4) suggested:
Abduction is the point where novelty, innovation and creativity enter the
scientific method, as indeed they must. With deduction, we get nothing
more than the consequences of the premises – but where did they come
76
from? With induction, we just get a generalisation from the observations we
have made – but how do we know they are all that matters? However, with
abduction we get explanation and the possibility of new knowledge.
Abduction is consistent with a depth of realism “…where explanation is not about prediction,
but about the steady unearthing of deeper levels of structures and mechanisms” (Dobson,
Jackson & Gengatharen, 2013, p. 7). Similarly, Wad (2001, p. 2) suggested:
If we take explanation to be the core purpose of science, critical realism
seems to emphasise thinking instead of experiencing, and especially the
process of abstraction from the domains of the actual and the empirical
world to the transfactual mechanisms of the real world.
In short, abduction plays a major role in critical realist analysis, since such an approach often
requires transcending, or speculating, which may be used to identify non-observable events
and non-events (in the domain of actual) generated by the mechanisms (in the domain of
real) to explain the architecture events that have occurred (in the domain of empirical).
3.4 Summary
Having briefly described EA examination from a social realist perspective and how the
different arguments have associated implications, this researcher came to the conclusion that
a critical realist perspective was suitable for achieving the objectives of the research. There
were two main reasons for underpinning the study with a CR philosophy.
Firstly, as mentioned earlier, in the past two decades, the higher education sector in Australia
has moved towards corporatisation, marketisation and rationalisation (Gengatharen et al.,
2009) as universities faced significant challenges due to market change, new teaching models
and the need for efficiencies, all of which required changes in technology capability. EA
effectively addresses these challenges by providing a holistic view of the planning and
development of an organisation’s business, application and technology architecture.
Recognising the importance of people in EA implementation requires acceptance of EA as a
social program, heavily influenced by the structural and cultural systems surrounding the
architecture. Thus, if this study was to follow other realists who do not endorse the value of
77
structure, culture and agency (people) to the same extent, then its explanatory context may
not be fully realised. The morphogenetic approach (MA), as a complementary methodology
for critical realism, deals with the components of structure, culture and agency and seeks to
make these components methodologically traceable. In terms of this study, MA provided an
explanation of the generative mechanisms operating between structure, culture and people
over periods of time, where the structural and cultural influences were mediated by people
shaping the situations in which they subjectively defined a particular course of action
(“concern” – people, personal) in relation to their objective social circumstances (“context” the EA program).
Secondly, and more compellingly, MA is well-suited to the mechanism-based explanation
proposed by this research, since its focus is to understand the people-focused mechanisms
by identifying the important mechanisms of EA, and building a theoretical explanation of the
role of people as a key element in EA implementation. MA provides an account of how the
underlying mechanisms work, and was therefore appropriate for this study which required a
definition that treats mechanisms as ontological in providing an account of how the
underlying mechanisms work. MA can also be used to examine the increasing intensity of
social change. For instance, MA allows an examination of social reorientation, perhaps a
radical social change/transformation, which involves macro-micro-macro explanations to
explain transformation (morphogenesis) rather than stability (morphostatis), and vice versa.
Moreover, the premise that people play a key role in successful EA implementation has the
potential to change the way it is approached in future.
The following chapter presents the key concepts and theories of a critical realist perspective,
most notably the morphogenetic approach and reflexivity theory, and describes how the
adoption of each has fundamental implications for the research objectives and ultimate
outcomes. Chapter four also looks at the dynamics of the complex social program surrounding
the architecture, and presents the research questions identified from the literature review.
78
CHAPTER FOUR: THEORETICAL APPROACH –
EXAMINING ENTERPRISE ARCHITECTURE FROM A
CRITICAL REALIST PERSPECTIVE
4.1 Overview
The aim of this chapter is to explain the key concepts and theories identified in the previous
chapter, and describe how each has a bearing on the research objectives and ultimate
outcomes. The first section of this chapter discusses the key concepts of critical realism, and
is followed by a literature review of critical realist perspectives on information systems
research. The next section discusses the morphogenetic approach (MA); the role of situational
logics and enterprise architecture (EA) pathways; the possible social mechanisms identified
from the situational logics; and describes the central role of reflexivity theory in EA
implementation programs. The final sections respectively discuss the dynamics of the
complex social programs surrounding the architecture and the research questions formulated
from the literature review.
4.2 A Critical Realist Perspective in Information Systems Research
The adoption of critical realism has significant implications for the objects to be investigated,
the progress of the research and the outcomes that can be expected. In applying such a focus
to EA implementation, this study required a deep understanding of the structural and cultural
systems, the mechanisms currently in place, and how the stakeholders react to the new
impositions, both in terms of increased governance and its impact on the way things are
currently done.
As a complementary methodology for CR, Archer’s Morphogenetic Approach (MA) provides
direction for critical realist interpretation in that it reflects the fundamental critical realist
assumption that social systems require an analytical separation between macro (structure)
and micro (agency) dimensions. Such an approach provides an important link between realist
ontology and practical social-information systems outcomes, and forms a consistent
79
foundation for ontology and methodology. In order to better understand the research
process, it is important to understand critical realism (CR).
Bhaskar’s (1978) concept of CR distinguishes among three ontological domains in reference
to social reality: the empirical, the actual and the real. The empirical domain consists of events
that are actually perceived or experienced directly or indirectly, whilst the actual domain
includes events which are experienced or not. Both are encompassed by the real domain,
which is made up of structures and mechanisms that are relatively enduring, with potential
powers and properties that are activated or triggered in particular contexts or by agency
(people) action, and thus may be causal in generating perceived or non-perceived events.
Bhaskar (1978, p. 25) proposed that events or phenomena should not be the core focus of
research; instead the focus should be on the structures and mechanisms that generate the
phenomena. In the same way, Mingers (2004) argued that CR not only focuses on a specific
event observed, but also on what the event can say about the underlying causal relationships
(or social mechanisms) that are enduring and lie beyond the common experience (the
empirical domain). Figure 4.1 below depicts Bhaskar’s three overlapping CR ontological
domains.
Figure 4.1: The three overlapping domains of reality in CR ontology (adapted from Mingers,
2004, p. 94)
80
Previous research underpinned by Bhaskar’s concept demonstrated that CR is suitable for
studying information systems in overcoming some of the difficulties associated with the social
contexts (Carlsson, 2006; Horrocks, 2009; Bygstad, 2010; Dobson et al., 2011; Carlsson, 2011;
Dobson et al., 2013; Mingers, Mutch, & Willcocks, 2013; Nuryatno & Dobson, 2016, among
others).
The use of case study is also well suited to CR-based attempts to find explicit causal
explanations for the complex social and organisational phenomena in the IS field (Wynn &
Williams, 2008). According to Fox (2009) CR can increase understanding of causal mechanisms
and contexts in IS research. Furthermore, the value and importance of CR as a philosophy in
IS research was also supported by Pettersen, McDonald and Engen (2009), who acknowledged
that CR can address social ontology as a necessary precursor to developing models and
empirical accounts of socio-technical systems. Carlsson (2011) expressed the opinion that CR
can be useful as an underpinning philosophy for behavioural IS research as well for IS design
science research. In addition, Bygstad and Munkvold (2011) indicated CR has an important
role in IS research in identifying causal structures of an ontological depth that is difficult to
unveil through other alternative approaches (i.e. positivist and interpretivist).
In order to appreciate the importance of CR as a philosophy in identifying possible “peoplefocused” mechanisms in EA implementation, Table 4.1 below provides an overview of the
various philosophies adapted from Fleetwood (2013). The table addresses the dearth of
literature on theories underlying the social context within the domain of information systems,
and opens up interesting avenues for the study of social elements in EA.
Fleetwood’s table demonstrates the tripartite connections between ontology, methodology
and practical theory, which is pivotal within the explanatory context of this study. As
explained by Archer (1995, p. 5) “…some social theorists have returned to work exclusively on
the reconceptualization of social reality. As such they may be playing a useful role in the
division of sociological labour, but if they suggest that their ontological exertions suffice, the
theoretical enterprise simply cannot be resumed on this unfinished basis. The practical
analyst of society needs to know not only what social reality is, but also how to begin to
explain it, before addressing the particular problem under investigation. In short,
methodology, broadly conceived of as an explanatory programme, is the necessary link
between social ontology and practical theory.”
81
Table 4.1 Various scope of the philosophy of science and meta-theory (adapted form Fleetwood, 2013, p. 11)
Associated meta-theory
Ontology
Scope of philosophy & meta-theory
Epistemology
Aetiology
Methodology
Research technique
Objective
Explanation
Prediction
Theory
Mode
Empirical realist ontology:
Entities are observed, atomistic events
Positivism or ‘scientism’
Atomistic, observable, events
No recognition of social construction
No agency-structure approach, only rational
agents as individuals
Avoids virtually all discussion of meta-theory
Gets on with applying its method and ‘doing’
O&M science
Idealist ontology:
Entities are constituted entirely by discourse
Various
Entities cannot exist independently of their identification
because all entities are constructed from discourse (etc.).
‘Reality’ is entirely socially constructed.
‘Reality’ is problematized, doubted & sometimes denied.
‘Reality´ is multiple.
‘Reality´ is becoming & processual.
Agents: decentred subjects constructed via discourse.
No agency-structure approach
Replaces philosophy of science with socio-politics of science.
Offers a socio-political critique of meta-theory.
As yet little engagement with CR.
Knowledge derives from (a) observing (b) event
regularities.
Truth established via testing hypotheses.
Not relativist at all.
Primacy of epistemology over ontology
Fudges or denies ontology-epistemology divide.
Recognises the fragility of knowledge – for ontological
reasons.
‘Truth’ (with capital ‘T´) is impossible for ontological reasons:
it is socially constructed.
Pragmatic notion of ‘truth’.
Epistemically & judgementally relativist.
Humean: causality as event regularity.
Reduces causality to Humean causality, rejects the latter,
Laws, law-like relations & functional relations. thereby rejecting the notion of causality.
Covering law method.
Mainly deconstruction, genealogy, but other methods used.
Explanation = prediction
Laws or event regularities = closed systems.
Maths, stats & quantitative data.
Permissive.
Regression, analysis of variance, correlation,
Avoids quantitative analysis.
structural equation modelling, factor analysis
Prediction.
To construct & test predictions & hypotheses
to establish whether claims are true or false.
Explanation is `thin´.
Explanation = prediction.
Socio-political not meta-theoretical.
Attempts to uncover power-knowledge & socio-political
agendas & lend voice to relatively powerless.
What is to be explained shifts from entity to its social
construction.
To explain is to provide a socio-political account of how
‘reality´ is socially constructed.
Prediction confused with explanation.
Rejected as a naïve idea sought by positivists who accept the
Explanation based on inductive generalisations. modernist idea that we can predict & control ‘reality’.
Spurious precision.
Vehicle for delivering predictions
Unclear.
Sceptical of the very idea of theory.
Deduction & induction
Unclear
Critical realist ontology
Entities are stratified, emergent, transformational + relational & processual
Critical realism
Some entities exist independently of their identification because not all are
constructed from discourse – i.e. extra-discursive
Single reality but multiple interpretations.
Four modes of reality; materially, artefactually, ideally & socially (essentially
builds on the three domains, see Figure 4.1):
Reality is stratified, emergent, transformational, systemically open,
becoming, processual & often relational (see section 4.3.1)
Agents & structures: distinct but related: TMSA – Morphogenetic Approach
Explicitly reflects upon meta-theory.
Engages with the other ontologies.
Accepts socio-political critique of meta-theory.
Retains both philosophy of science & socio-politics of science
Subordination of epistemology to ontology.
Recognises the fragility of knowledge - for epistemological reasons.
Knowledge derives from uncovering causal mechanisms.
Truth (not with capital ‘T´) is difficult but not impossible.
Epistemically but not judgementally relativist.
Separates Humean causality from causality as powers & tendencies.
Powers & tendencies replace laws, law-like & functional relations
Causal-explanatory.
Explanation via uncovering & understanding causal mechanisms
Deconstruction & genealogy accepted.
Permissive
Critical discourse analysis, action research, archaeology
Mainly uses qualitative techniques: role of (some) quantitative techniques is
debated.
Explanation.
Accepts attempts to uncover power-knowledge & socio-political agendas &
lend voice to relatively powerless.
Explanation is ‘thick´ - operation of causal mechanisms
Not confused with prediction.
Accepts a role for socio-political account.
Tendencial prediction based on knowledge of causal mechanisms.
Tendencial prediction is not precise, but not spurious either.
Vehicle for delivering causal-explanatory accounts.
Abduction & retroduction
82
4.3 Morphogenetic Approach: A Social Investigation in Social
Mechanisms
In adopting CR, a number of philosophical assumptions must be met. One of the most
important is the analytical separation of structure and agency over time. Termed “analytical
dualism” by Archer (1989), this is a fundamental component of the critical realist approach.
The aim of analytical dualism is to enable examination of the complex duality of structure and
agency, an artificial separation or dualism, and investigate the complex duality between
structures (macro) and agents (micro). It suggests:
a) Structure necessarily pre-dates the action(s) leading to its reproduction or
transformation; and
b) Structural elaboration necessarily post-dates the action sequences which gave rise to
it (Archer, 1995, p. 15).
The morphogenetic approach, also referred to as Morphogenetic/Morphostatic (M/M)
(Archer, 2015) is a meta-theoretical social ontology developed by Archer (1979; 1989; 1995;
2013a) as a methodological complement to CR by applying an analytical dualism over time.
Morphogenesis “refers to the complex interchanges that produce change in a system's given
form, structure or state (‘Morphostatis’ is the reverse)” (Archer 1989, p. xxii). MA analysis
takes place in three cycles: a) structural conditioning, which refers to pre-existing structures
that condition but do not determine; b) social interaction, which arises from actions oriented
towards the realisation of interests and needs emanating from current agents and may lead
to c) structural elaboration or modification, that is, a change in the relations between parts
of the social system.
Archer (1995, p. 195) suggested people are capable of resisting, repudiating, suspending or
circumventing not only the structural tendency, but also the cultural tendency in
unpredictable ways because of their creative powers as human beings in producing
tendencies towards change/elaboration (or reproduction) in the relational organisation of the
social order. These processes of change (or reproduction) are conceived as “generative
mechanisms” (Archer, 2015). To elaborate, Archer (1995, p. 175) described it thus:
83
…morphogenetic approach makes no leap from the real to the actual, but
rather dwells on the ground between them by analysing the generative
mechanisms potentially emanating from structures (and cultures) as
emergent properties and their reception by people, with their own
emergent powers of self and social reflection. Outcomes never simply mirror
one or the other, but are the products of their interplay.” Hence, “In society
there are a variety of emergent properties – structural, cultural and agential,
each of which is irreducible to the others, has relative autonomy, and is also
relatively enduring.
4.3.1 The emergent properties: Structural – cultural formation, and agency
influence
Structure, culture and agency3.1 are viewed by Archer as analytically distinct strata of social
reality in which social structures are viewed as “…relatively enduring, anterior social objects
that possess causal powers and are neither observable nor reducible to social interaction”
(Luckett, 2012, p. 340). Archer (1995) portrayed the social world as a stratified model
involving: a) different structural emergent properties (SEPs): roles, institutional structures,
social systems and positions; b) different cultural emergent properties (CEPs): ideas, theories,
beliefs, values and ideologies; and c) different people emergent properties (PEPs): dependent
on a stratified model of human beings, agents, and actors.
Archer (1995) expressed the opinion that:
Structures (as emergent entities) are not only irreducible to people, they
pre-exist them, and people are not puppets of structures because they have
their own emergent properties which mean they either reproduce or
transform social structure, rather than creating it. To explain which occurs
the realist examines the interplay between the two (endorsing and utilizing
separability) and in both cases, reproduction and transformation necessarily
refer to maintaining or changing something which is temporally prior to
these activities (p. 71).
84
Box 3.1: Structure, Culture, and Agency
Structure is a relational property that has the generative capacity to modify the
powers of its objects and to exercise causal influences (Raduescu & Vessey, 2008,
p. 9). Structures (at any given time) are the results of human interaction,
including the results of the results of that interaction - any of which may be
unintended, unwanted and unacknowledged (Archer, 1995, p. 196).
Culture is the product of human agency (Archer, 1996, p. 77), and presented as a
resource which agents draw upon within action contexts but is never something
which shapes these contexts for them. (Archer, 1996, p. 304).
Agency is used as a generic term which stands for the 'people' which constitute
the 'parts' of society. Agency stands as the middle element linking Persons
Agents Actors and is needed to account for who occupies which roles - and
why they do what they do when the role does not require them to do it: the
genealogy, Human Being-Agent-Actor (Archer, 1995, p. 256).
Archer also argued that culture is approached analytically in exactly the same way as
structure:
Like structure, culture is a human product but it too escapes its makers to
act back upon them. The CS [cultural system] contains constraints (like the
things that can and cannot be said in a particular natural language), it
embodies new possibilities (such as technical applications undreamed of in
the pure theory on which they are based), and it introduces new problems
through the relationships between the emergent entities themselves (the
clash of theories), between these and the physical environment (mastery or
ruin), and between these and human agents (makers and openers of
Pandora's box). Consequently, as CEPs, ideational contradictions exist
independently of people noticing them or caring about them — indeed since
there are an infinite number of situations upon which any theory may bear,
it might well contain contradictions of which no one is aware. Similarly, the
85
relationship between a problem and a solution, which is one of
compatibility, is ultimately divorced from whether anyone does understand
it, though not from the ability of someone to do so. Thus, as a CEP, a soufflé
recipe might not have been used by anyone living, but would still work for
the cook who eventually tried it (1995, p. 181).
Archer suggested the emergent properties of collectivities and individuals differ from the
emergent properties of organised groups, which differ yet again from those pertaining to
populations (p. 190). Yet, as detailed in Figure 4.2 “these different levels of ‘social integration’
are not discrete from the powers of ‘system integration’, despite their capacity for
independent variation at any given time” (p. 190). Archer described a double morphogenesis:
…where agency undergoes transformation, acquiring new emergent powers
in the very process of seeking to reproduce and transform structures. For in
such structural and cultural struggles, consciousness is raised as collectivities
are transformed from primary agents into promotive interest groups; social
selves are re-constituted as actors personify roles in particular ways to
further their self-defined ends; and corporate agency is re-defined as
institutional interests promote reorganization and re-articulation of goals in
the course of strategic action for their promotion or defence (p. 190-191).
Figure 4.2: Analytical dualism in social theory: A stratified model of social structure involving
SEPs, CEPs and PEPs in EA implementation (adapted from Archer, 1995, p. 190)
86
This stratification of agents described on the right-hand side of Figure 4.2 allows a rich
representation of people’s roles in organisational change as they can be seen as primary
agents in particular positions, corporate agents in institutions, or as individual actors in
particular roles. Structural influences (the generative powers of SEPs and CEPs) are mediated
towards people through shaping the situations in which they find themselves.
Archer’s stratified model of people in Figure 4.2 (the different emergent properties of people)
suggests that: a) primary agents are collectivities of people who share the same position
(places, functions, rules, duties and rights); b) individual actors are those who fill particular
social roles that further their self-defined ends; and c) corporate agents are organised groups
who institutionally play the major part in institutional decision-making. Archer (1995)
explained:
For in such structural and cultural struggles, consciousness is raised as
collectivities are transformed from primary agents into promotive interest
groups; social selves are re constituted as actors personify roles in particular
ways to further their self-defined ends; and corporate agency is re-defined
as institutional interests promote reorganization and re articulation of goals
in the course of strategic action for their promotion or defence. All the above
processes are reinforced or repressed by the overall state of systemic
integration, whose incompatibilities foster their actualization and whose
coherence serves to contain this transformative potential of agency (p. 191).
In short, MA deals with emergent properties in the analyses of structure, culture and agency
as an attempt to bridge the gap between the explanatory power of the practical social theory
and the ontological strength of the realist philosophy. MA analysis therefore works in three
part cycles; each has relative autonomy yet interacts with the others (Archer, 1995). As
described by Luckett (2012, pp. 341–342): at Time 1 [T1] the “…structural and cultural
conditioning is already set up, before human actors with particular intentions, concerns and
projects located in particular roles and positions in institutions begin interacting with each
other at Time 2 – Time 3 [T2 – T3] (social and socio-cultural interaction). It is here, at the
second stage of the morphogenetic cycle that human agency, in the form of personal
emergent properties (PEPs) is exercised. Some institutional roles are necessarily related to
87
each other […], whilst others are contingent to the context. Exactly what emerges from a
particular period of social/socio-cultural interaction (Time 2 – Time 3) is contingent on the
context of [the] situation and cannot be predicted.” Thus, the realisation of interests and
needs emanating from current agents may lead to structural and cultural elaboration
(morphogenesis) or reproduction (morphostatis) at Time 4 [T4]. The overall MA diagrammatic
analyses are shown in Figures 4.3 – 4.5.
Figure 4.3: The morphogenetic/static of structure (adapted from Archer, 1995, p. 193)
Figure 4.4: The morphogenetic/static of culture (adapted from Archer, 1995, p. 193)
Figure 4.5: The morphogenetic/static of agency (adapted from Archer, 1995, p. 194)
88
In review, the practical application of morphogenetic/morphostatic analysis of structure
requires a social system with four basic propositions (Archer, 1995, pp. 168-169):
a) There are internal and necessary relations within and between Social Structures (SS);
b) Causal influences are exerted by Social Structure(s) (SS) on Social Interaction (SI);
c) There are causal relationships between groups and individuals at the level of Social
Interaction (SI);
d) Social Interaction (SI) elaborates upon the composition of Social Structure(s) (SS) by
modifying current internal and necessary structural relationships and introducing new
ones where morphogenesis is concerned. Alternatively, Social Interaction (SI)
reproduces existing internal and necessary structural relations when morphostasis
applies.
In the same way, culture also presupposes a social system with four basic propositions
(Archer, 1995, p. 169):
a) There are internal and necessary logical relationships between components of the
Cultural System (CS);
b) Causal influences are exerted by the Cultural System (CS) on Socio-Cultural interaction
(the S-C level);
c) There are causal relationships between groups and individuals at the Socio- Cultural
(S-C) level;
d) There is elaboration of the Cultural System (CS) due to Socio-Cultural Interaction (S-C)
modifying current logical relationships and introducing new ones, where
morphogenesis is concerned. Alternatively, Socio-Cultural Interaction (S-C)
reproduces existing internal and necessary cultural relations when morphostasis
applies.
To summarise, the context of interpretation of the cultural levels (CS and S-C) in MA is
supported by sociologically sound methodology. Archer has made a clear argument about the
distinction
between
the
Cultural
System
and
Socio-Cultural
level
within
the
morphogenetic/morphostatic agency, claiming that: “…if we clearly distinguish between the
two cultural levels, the Systemic and the Socio-Cultural, then we can also differentiate
89
between the aspects of context - 'other ideas' and 'other people' - on which the former and
latter depend respectively (1996, p. 134).” This implies that in a cultural context of agency,
the logical and the causal are systematically separated. As detailed in Table 4.2 below, Archer
observes that the MA separates the macroscopic cultural interactions or textual ideas (CS)
from people's meanings (S-C) and that this turns out to be of value to explain the microconcerns of those whose first reaction was to flinch away from such textual definition (1996,
p. 136).
Table 4.2 The Cultural system and socio-cultural level within the morphogenetic/static view
of agency (adapted from Archer, 1996, p. 134)
Cultural level
Context
Relation between them
dependent on
Cultural system (CS)
Other ideas
Logical: Logical relationships between
component of CS
Socio-cultural (S-C)
Other people
Causal: Causal influences are exerted by the
Cultural System (CS) on Socio-Cultural
interaction (the S-C level); Causal relationships
between groups and individuals at the SocioCultural (S-C) level;
Morphogenesis:
The S-C level elaborates on the composition of the CS by modifying
current logical relationships and introducing new ones
Morphostatis:
S-C reproduces existing internal and necessary cultural relations
Nevertheless, Archer’s micro-concerns about agency still had a central dilemma, particularly
from a CR perspective. According to the author “CR cannot be charged with smallism. It has
been committed to strong emergence and ontological stratification since its inception (2015,
p. 35).” By smallism Archer describes the tendency to privilege the small over the large in
scientific discourse (for example methodological individualism). Archer concluded it is
possible for MA to work in a macro-micro-macro context of the social world:
…it [CR] has not entirely freed itself of the smallest prejudice. For example,
the recurring trope of “underlying mechanisms” carries the unfortunate
90
connotation that mechanisms operate at the micro-scale. There is, as well,
a small remnant of causal deflation. The MM approach [morphogenetic
approach – MA] does draw a clear distinction between “macro” and “micro”
causation, to be sure. But macro-to-micro causation is often represented in
terms of efficient causation: structure at T1 impacts agency at T2. No doubt!
But not only. Structure also influences agency synchronically by constraining
and enabling certain agentic powers. What is needed, then, is: (1) an
understanding of social mechanisms that is fully shorn of the mechanistic
metaphysics of the physicalist imaginary; and (2) an understanding of social
causation that is more attentive to: (a) different forms of social causation;
and (b) specific types of causal powers in the social world (2015, p. 35).
In this EA examination, it can be seen that an explanation (not prediction) of generative
mechanisms, and their interplay between structural, cultural and people (agency), becomes
the objective of the study. As a complementary methodology, MA is ‘causal-explanatory’
(Fleetwood, 2013), whereby “explanatory” refers to its objective to explain, and “causal”
because it explains in terms of providing a causal account. Hence, if an explanation of peoplefocused mechanisms and their interplay between and within the structure, the culture, and
the agency (people) was found, then this study would have presented a theoretical
explanation of the role of people as a key element in EA implementation. Figure 4.6 illustrates
the use of MA in identifying the hidden generative mechanisms within social contexts, and
their interplay between structure, culture and people (agency) in EA implementation
programs.
In Figure 4.6 people (agents) can be seen to be acting purposefully through conscious and
unconscious interaction, thereby reproducing or transforming the structures and cultures
that enable and constrain their actions. Structures and cultures are the ever-present
conditions and the continually altered or sustained outcomes of human agency (adapted from
Fleetwood, 2013). This causal-explanatory framework of MA directed this study towards
developing the research questions and consequently formulating the interview questions (see
Chapter 5).
91
Figure 4.6: Morphogenetic approach in explanatory study (adapted from Fleetwood, 2013)
4.3.2 Situational logics and enterprise architecture pathways
In understanding the structural and cultural conditions from the viewpoint of agency, Archer
considered situations are:
…shaped very differently for agents according to whether such emergent
properties are characterized by tensions between their component
elements or by coherence between them (1995, p. 214).
According to Archer (1995) emergent properties (see Figure 4.9) have two roles: a) the role
of first-order emergent properties (that is, the results of past interaction); and b) the role of
second-order emergent properties (that is, the relations between the results of the results of
past actions/first-order emergent properties). First-order emergent properties is the shape of
past distributions: systemic, institutional, roles, and positions that play in the strategic
directional guidance of an organisation at the (macro) institutional level and affects large
segments of the population. In this study the two roles of emergent properties were
understood to determine: a) the relationship between a time prior to EA implementation and
at the time of EA implementation (the first- and second-order emergent properties); and b)
the situational logics of the interplay between and within structure, culture and agency at the
time of EA implementation (second-order emergent properties).
92
Archer (1988, 1995) argued that, at institutional level, interactions between emergent
structural and cultural properties create different modes of conditioning and interaction in
situational logics, which predispose agents to follow particular courses of action to promote
their important personal projects (that is, in the second-order emergent properties). Archer
also concluded there may be consistency or contradiction between ideas within the cultural
system. Consistency and contradiction can be necessary (internally related as the ideas
depend on each other and cannot operate apart) or contingent (externally related and
contextual). Figure 4.7 describes Archer’s causal influences exerted by the cultural system on
the socio-cultural and defines four possible interactions between the cultural (A) and the
structural (B) systems (Archer 1995; 1996):
Figure 4.7: Morphostatis vs morphogenesis: Situational logics in social and system
integration in EA implementation (adapted from Archer, 1995, pp. 218 and 295)
1.
Necessary complementarities – The ideational compatibilities between A and B lead to
an environment of mutual support. “In other words, invoking A also ineluctably evokes
B, but since the B upon which this A depends is consistent with it, then B buttresses
adherence to A.” (p. 234). The structural and the ideational are in harmony, and such a
position has causal possibilities at a structural level, creating a situational logic of
protection at the systems level. Archer suggests that increasing the depth of
systemization at the structural level increasingly blocks change because of its threat of
disruption. At the personal level, corporate agents see the increasing barrier to their
93
advancement and may thus seek unpredictable avenues to break out of the constraining
systemic limitations.
Necessary incompatibilities – These are the reverse of necessary complementarities;
2.
components of the socio-cultural system that contain some particular belief or theory
which is internally inconsistent with ideas at the cultural level. “When the constitution of
the social system is marked by incompatibilities between institutions which are none the
less internally and necessarily related, this has rightly been seen as containing a potential
for change which is entirely lacking in the complementary configurations. Generally,
when two or more institutions are necessarily and internally related to one another yet
the effects of their operations are to threaten the endurance of the relationship itself,
this has been referred to as a state of 'contradiction'” (p. 222). Such incompatibility, or
contradiction, provides a situational logic of correction as these ideas must ultimately
accommodate each other. At the structural level the need for accommodation suggests
the emergence of properties directed by compromise as parties struggle to remain in
power. Unification is a consequence at the socio-cultural level as compromise becomes
essential and emergent. This holding state is inherently unstable and suggests a period
of instability as participants’ jockey for position with accommodation as the focus,
seeking to survive amongst the incompatible cultural ideas.
3.
Contingent incompatibilities – Occur when the material world produces situations which
are incompatible with the prevailing social and cultural properties “because partisans of
A and B are unconstrained by any dependence between these items, there is nothing
which restrains their combativeness for they have everything to gain from inflicting
maximum damage on one another’s ideas in the course of competition” (Archer, 1988,
p. 240). A “battleground of ideas” emerges providing a situational logic of elimination;
the pluralism at a cultural level promotes the creation of distinct loyalties at a sociocultural level – such cleavage encouraging competition at a structural level and
polarization at a social level as groups and individuals struggle to remain in the game.
Cooperation on the acceptance of change is discouraged and diversity reduced as
elimination of alternatives is attempted and progressed.
94
4.
Contingent complementarities – Occur when material opportunities arise that resonate
with the social and cultural properties, stimulating opportunism - “Only the contingent
complementarity simultaneously holds out choices to the adherents of A but leaves them
free to make what they will (if anything) of B…only the contingent compatibility is free
from sociocultural manipulation, designed to induce avoidance or adoption or aversion.
Certainly, distracting sociocultural practices – habitual preoccupations, established
routines, traditional preserves or conventional divisions of subjects – may well reduce
subjective willingness to explore new and congruent possibilities, but these will usually
coexist with various sticks and carrots which stimulate originality, innovation and
experimentation (as in the derived sequence). The actors concerned have substantial
freedom to survey or to ignore the broader horizon which has come in to view…” (Archer
1988, p. 243). The situational logic of opportunism has a net systemic result of great
cultural variety. It “breaks down artificial knowledge barriers, stimulates new departures
and bold syntheses”. At the cultural level wild ideas and daring proposals can ensue
unchecked by the socio-cultural. At the socio-cultural diversification, specialization and
recombination can ensue as “marginals disengage themselves to recoalesce in a group
with a novel brief” (p. 244).
MA can be applied at lower levels than those proposed by Archer (2013a). In the university
environment (for example Luckett, 2012), it allows examination of the roles of the different
centres within the overall university system. At this level one could argue that the internal
operation of the sandstone (traditional) universities can be framed as an environment of
contingent incompatibilities over the initial period of EA implementation, in that the cultural
requirement of academic freedom conflicts with a systemic need for increased managerialism
and greater control, with EA implementation providing the battleground for such
contradictions by reflecting the university’s requirement for greater control and IT
centralisation.
Seeking to evangelise the corporate commitment to control, impinges on the centres’ desire
for freedom to do what they wish. As detailed in Figure 4.8, the situational logic of correction
suggests an outcome at the cultural ideational level of syncretism (the attempt to sink
differences and effect union between the contradictory elements concerned) (p. 233). This
situational logic must lead to compromise at a systemic level, as one or both parties lead to a
95
re-definition of the EA to accommodate the irreconcilable ideational differences. The main
thrust of this situational logic is the “sinking of differences”, a unification and compromise.
Archer does not include time in her detailed figure, but it is implicit in the development of the
argument in her seminal book, Realist Social Theory: The Morphogenetic Approach. Figure
4.8 illustrates the situational logics of both structural competition and cultural opportunity,
thus redefining the relation between cultures and structures in EA implementation.
Figure 4.8: Situational Logics at different strata (adapted from Archer, 1995 p. 303)
A state of contradiction would hopefully be short-lived as the benefits achievable through
compromise become clear to the parties upon whom change is being imposed. An alternative
pathway for contingent incompatibilities may ensue when parties deem unification
impossible and unnecessary. Under this situational logic a battle will rage as logic moves
towards elimination, and accepting the divergence at a cultural level in turn leads to cleavage
at the socio-cultural level and competition at a systemic level. While the battle rages, centres
will seek advantage, power and politics playing a crucial and important role. Cleavage is not a
desirable option given the immediate threats evident within the university environment and
cannot be allowed to prevail, as universities do not have the luxury of time due to the urgency
of external forces.
Yet as Archer suggests, the acceptance or otherwise of these logics depend on their ultimate
social reception. Primary agents are defined by Archer as “collectivities sharing the same life
chances”, in other words everyone is inescapably an agent in some of their doings by being
part of a collective, intentionally or otherwise. Archer distinguishes thus between corporate
96
agents and primary agents: primary agents are those who have no say in possible cultural or
systemic re-modelling. They “neither express interests nor organize for their strategic pursuit”
(p. 259). This is not to say that the aggregated effects of primary agents can have no impact
at a cultural or systemic level; they can well generate aggregated and powerful impacts at a
systemic level, but the outcomes are a consequence of uncoordinated action and without
stated aim.
Social interaction in programs like EA implementation plays out in an environment of
corporate agents promoting the systemic state in question. Corporate agency thus shapes the
context for primary agency. Yet primary agency also has aggregate effects, as it unleashes a
range of environmental pressures and problems which may impact the aims the corporate
agent seeks. This is what Archer referred to as double morphogenesis, “...where agency
undergoes transformation, acquiring new emergent powers in the very process of seeking to
reproduce and transform structures” (p. 190). Corporate agency thus has two tasks with
respect to the promotion of their goals: “the pursuit of its self-declared goals as defined in a
prior social context; and their continued pursuit in an environment modified by the responses
of primary agency to the context which they confront” (p. 260).
The acceptance of EA is ultimately dependent on primary agents’ acceptance of its basic
premise given the context within which they reside. Horrocks (2009) suggested the
recognition of a distinction between corporate and primary agents is useful in examining
program implementation. There are many cases where agents are ambivalent to EA and its
underlying premise, yet their aggregate effects may well be significant to its ultimate rejection
or acceptance. Those with little understanding of IT and its strategic role will need to be
convinced of the ultimate benefits of restrictions to their current ways of doing things.
Presenting EA to such a group is a challenge. The program mechanisms must place a major
focus on communication of the benefits in context, clearly recognising the cultural and
ideational elements involved.
Figure 4.9 illustrates the causal explanation between the first- and second-order emergent
properties; the different modes of conditioning and interaction in situational logics; and the
MA outcomes. As an explanatory framework in the examination of people’s role and the social
aspects of EA implementation, MA assists in providing analytical histories of emergence that
97
can uncover the interplay between and within social structures, socio-cultural systems and
agency (the analytical dualism). Archer (1995) explained:
…the kind of explanation which the morphogenetic approach proffers takes
the form of analytical histories of emergence for the practical issue under
investigation. It does so by examining the interplay within and between the
three cycles, for the ultimate benefit of analytical dualism is that it is not a
static method of differentiation but a tool for examining the dynamics by
which the 'parts' and the 'people' shape and re-shape one another through
their reciprocal interaction over time (p. 194).
The explanatory format consists in providing analytical histories of
emergence. At every level the tendential powers of generative mechanisms
are complemented and supplemented by a historical analysis of the
concrete contingencies which intervened to produce particular outcomes.
The format itself is none other than the three-phase morphogenetic/static
cycle, with the phases delineated according to the problems in hand. The
three parts of the analytical narrative consist of 'structural conditioning' by
the prior distribution of resources, of life chances, of vested interests and of
bargaining power which are mediated to agents situationally; 'social
interaction' as conditioned by the former, by other structural factors which
also impinge on agents, by social affinities and antagonisms between them,
and ultimately by the reflexive monitoring of an inalienably innovative
agency; 'structural elaboration' is quintessentially dependent upon how (or
whether), in the precise combination of conditioning and contingency,
bargaining power is converted into negotiating strength between corporate
agents. But neither combination nor conversion are mechanical processes
compelling or propelling agents: on the contrary they are the situated
products of self-conscious agents which is what makes their strategic use of
power and exchange that which actually mediates elaboration (p. 327).
98
Analytical histories represent agency with different modes of conditioning and interaction in
situational logics for their attainment (see Figure 4.8); the connective generative mechanisms
between structural and cultural conditioning (T1) and structural and socio-cultural interaction
(T2-T3), which constitutes the possible transformation or reproduction mechanisms at
structural/cultural elaboration or reproduction (T4). These mechanisms are complemented
and supplemented by the historical analysis of the concrete contingencies of the structural
and cultural conditioning, the structural and socio-cultural interaction, and the structural and
cultural elaboration (morphogenesis) or the structural and cultural reproduction
(morphostatis). Either of morphogenesis or morphostatis will generate the key
implementation mechanisms as the results of MA outcomes.
Figure 4.9: The analytical histories of emergence: Relationship between first- and second
order emergent properties, generative mechanisms, and MA outcomes (adapted from
Archer, 1995; Horrocks, 2009)
99
4.3.3 Possible contextual social mechanisms identified from the literature
review
A review of the current literature highlighted the importance of possible social mechanisms
for EA implementation programs, and affirmed Archer’s MA as the most critical in defining
situational and interactive social mechanism by making clear the situational logics proposed
by Archer (1995). Archer (2015, p. 171) emphasised that:
…one always has to take the structural conditioning and the related
situational logic into account – the particular resources actors can count on,
and the reflexive agency, individual and social, acting upon the given
situation. No real ‘situation’ exists without these factors…
Although Archer never identified the different modes of situational logics as generative
mechanisms, based on the present theoretical context in the literature, this study assumes
that these different modes (see Figures 4.8 and 4.9) have definite relations to the “generative
complexes” because they can explain: (a) such associations (i.e. how they arose and work)
and (b) are robust enough to account for cases and times when no such ‘constant
conjuncture’ can be found (i.e. Y is not significantly correlated with Z) but do not entail
scrapping the mechanism itself since tendencies may be possessed yet unexercised, exercised
yet unrealised, and realised yet unperceived (or undetected) (Archer, 2015, p. 2). To conclude,
it is logical to assume that Archer’s generative mechanisms fit into a broader view of the aims
and standards of critical realist explanatory approaches, and more importantly, supplies a
valid basis for causal explanation.
Table 4.3 is derived from Figure 4.8 and provides an overview of the possible social
mechanisms suggested by Archer. It forms the basis for identifying key implementation
mechanisms and social responses triggered by the EA implementation.
All the possible mechanisms identified in the literature (program mechanisms in Chapter 2
and social mechanisms in Chapter 4) provide different explanations for implementation. They
provided the basis for examining specific key implementation mechanisms to which people
responded in the new EA program, as provided by the process and technology.
100
Table 4.3 Possible social mechanisms identified in the literature
Social component
Level
Syncretism
Compromise
Unification
Containment
Correction
Pluralism
Competition
Cleavage
Polarisation
Elimination
Systematization
Integration
Reproduction
Solidarity
Protection
Specialization
Differentiation
Sectionalism
Diversification
Opportunism
Situational mechanism
Situational mechanism
Interaction mechanism
Interaction mechanism
Reproduction mechanism
Situational mechanism
Situational mechanism
Interaction mechanism
Interaction mechanism
Transformation mechanism
Situational mechanism
Situational mechanism
Interaction mechanism
Interaction mechanism
Reproduction mechanism
Situational mechanism
Situational mechanism
Interaction mechanism
Interaction mechanism
Transformation mechanism
Morphogenesis
Morphostatis
Literature
Social interaction
Structural system
Generative
mechanism
Cultural systems
Situational
Logic
Socio-cultural interaction
Tendency
Power
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
4.4 A Central Role for Reflexivity in the Implementation Program
Although MA is advantageous for examining the complex duality of social structure (macro)
and agency (micro), it does not answer how social structure (and culture) and agency are
linked. Archer (2008) suggested:
101
…agents’ use of their reflexivity constitutes the missing and mediating link.”
Thus, “Through their reflexive deliberations (‘internal conversations’) agents
subjectively define particular courses of action (‘projects’) in relation to their
objective social circumstances (‘contexts’). Without this, we have no
explanatory purchase upon exactly what agents do because in the same
circumstances they do not act in uniform ways. Conversely, a proper grasp
of reflexive deliberations accounts for subjects’ evaluations of their
situations in the light of their personal concerns, and their (re-) evaluation
of their projects in the light of their situations. Without such an account,
sociology has to settle for empirical generalizations about what ‘most people
do most of the time’ (Archer, 2008, p. 1).
In examining the role of agency (people) and its interplay with structures and cultures in EA
implementation, it is necessary to examine the mechanisms or causalities that determine the
state of interplay. As mechanisms are only effective if people adopt them reflexively, this
study used reflexivity to explain how people responded to the embedded mechanisms of EA
implementation. Archer (2007, p. 4) described reflexivity as:
…the regular exercise of the mental ability, shared by all normal people, to
consider themselves in relation to their (social) contexts and vice versa. Such
deliberations are important since they form the basis upon which people
determine their future courses of action—always fallibly and always under
their own descriptions.
In other words, reflexivity refers to people’s thoughts and ideas that tend to consider
themselves (the personal concern: ‘concern’) in relation to their social situation (the social
situational: ‘contexts’) in determining their future courses of action. People’s thoughts and
ideas tend to be inherently biased and bidirectional, with both cause and effect influencing
one another’s actions – i.e. there is a circular relationship between cause and effect.
Archer (2010b) expressed the opinion that reflexivity is not a ‘routinized action’ and people
do cope with change through conscious or unconscious decision-making (p. 7). Thus
“…[reflexivity] is the process through which reasons become causes of the courses of action
adopted by social subjects. Their subjective internal deliberations – internal conversations –
102
are responsible for mediating the conditional influence of objective structural and cultural
factors upon social action… (p. 5)”. Reflexivity is held to be a prerequisite for all forms of
social life (Archer, 2010b, p. 8) and reflexive deliberations form an important “mechanism”
for explaining how structures constrain and enable agents.
There are two essential components of reflexivity that need to be considered and understood
in this study. First are the personal (individual) reflexivity styles to explain different individual
responses to the situations people find themselves in; and second is the collective reflexivity
which also reflects the people role in EA implementation. Archer (1995) suggested that only
corporate agents have morphogenetic possibilities, and their power to change is only
achieved via their involvement in some collective. A collective concern for the role of people
in EA implementation entails reflexive deliberation about collective responses to situations
where a common focus in EA implementation is shared through intrinsic commitment. This
communal experience provides a new way of building commitment for the successful
implementation of EA.
Archer (2007) suggested individual reflexivity is particularly important in understanding the
role played by individuals in responding to change proposals. Individual motives and
properties are necessary for understanding the mechanisms by which structures impact
agents and the following need to be considered:
(1) Why do people act at all? What motivates them and what are they
(fallibly) trying to achieve by endorsing given courses of action? This entails
an examination of their personal concerns and inner reflexive deliberations
about how to go about realizing them. (2) How do social properties influence
the courses of action that people adopt? This involves a specification of how
objective structural or cultural powers are reflexively mediated. (3) What
exactly do people do? This requires an examination of the variability in the
actions of those similarly socially situated and the differences in their
processes of reflexivity (Archer 2007, p. 6).
To determine the collective reflexivity Archer (2010b; 2013b) recommended looking at the
four primary modes of individual reflexivity: a) communicative reflexivity, meaning that
internal conversations need to be confirmed and completed by others before they lead to
103
action; b) autonomous reflexivity, where internal conversations are self-contained, lead
directly to action and are characterised by instrumental rationality; c) meta-reflexivity occurs
when internal conversations critically evaluate previous inner dialogues and are critical about
effective action in society in promoting value rational action; and d) fractured reflexivity,
where internal conversations cannot lead to purposeful courses of action and only intensify
personal distress and disorientation, leading (temporarily) to passive agents. Archer (2010b;
2013b, p. 11) argued there is no reason why collective reflexivity should not be characterised
in the same manner as modes of personal reflexivity: a) collective communicative reflexivity
is practised among family and closest friends; b) collective autonomous reflexivity as the
pragmatics of group well-being; c) collective meta-reflexivity when considering relational
enhancement or protection; and d) collective fractured reflexivity if the contingencies of life
in an open system, including openness to the creativity and destructiveness of others, will
deprive us of its emergent fruits by destroying the bonds upon which that of great worth is
relationally dependent.
Collective reflexivity is not about people thinking in the same way, or people sharing external
commitments, or people having a mutual intention; rather it is about people being in a special
relation and that relation is what makes them reflexive in a social, instead of a personal way
(Donati, personal communication, citied in Archer, 2010b; 2013b, p. 11). This special relation
is the collective reflexivity, and was used in this study to identify collective reflexivity in EA
implementation.
In identifying the important mechanisms triggered by EA implementation programs in the HE
sector and University X in particular, the central role of reflexivity provides the basic condition
for explaining the people-focused mechanisms identified from the MA cycles. Archer
concluded that reflexivity acts as a key mediatory mechanism: “…reflexivity is crucial for
bridging the gap between formal expectations and actual eventualities in the open social
system” (Archer, 2010a, p. 281) and suggested:
…it is necessary to look more closely at the interconnections between the
relationships summarized in the basic diagram [Figure 4.10]. Within any
cycle this is to clarify {relation a}, that is, how structural/cultural conditioning
effectively influences sociocultural interaction. Without such clarification,
the term "conditioning" merely rules out any form of determinism, but does
104
not arbitrate between two possible answers: conditional influences are
exerted largely through […] the exercise of reflexivity, entailing deliberation
about the appropriate course of action in a given social context (Archer,
2010a, p.276).
Figure 4.10: How reflexivity works in the basic morphogenetic sequence (adapted from
Archer, 2010a, p. 275)
Mutch (2007) expressed the opinion that MA is focused on the structure and culture shaping
the context in which agency is exercised, thus agency reflexivity is shaped by the interplay
between context (the social situation) and concerns (the personal concerns of agents). At this
level it can be argued that by applying reflexivity to EA as an organisation-wide program with
a strong social context the study can examine the role of people’s reflexivity by the way the
program impacts people’s ultimate decisions concerning (‘concerns’) EA implementation.
Hence, the reflexive mediation of mechanisms ensures that people have the necessary
mechanisms in place to enable architecture transformation.
In Figure 4.11 below, Archer (2010a, pp. 280-281) argued the continuity of morphostatic
societies make a substantial contribution to low reflexivity, since agencies possess the
situational logics of morphostatis (see Figure 4.8):
…the co-existence of cultural and structural morphostasis together
generated a high and lasting degree of everyday "contextual continuity" for
the populations in question: repetitive situations, stable expectations, and
durable relations…”, as the, “…continuity of morphostatic societies
(generative of contextual continuity) was underpinned by a low level of
structural differentiation and an equally low degree of ideational
diversification –the two being mutually rein forcing. Thus, the structural elite
105
was trapped in the only form of cultural discourse in parlance, given the
absence of an alternative fund of ideas; similarly, the cultural elite was
enmeshed in existing leadership roles, given the lack of any other form of
social
differentiation.
Cultural
morphostasis,
through
the
stable
reproduction of ideas among a unified population, generated an ideational
environment that was highly conducive to structural maintenance. Equally,
structural morphostasis, through perpetuating subordination and thus
controlling differentiation, made a substantial contribution to cultural
maintenance.
Figure 4.11: Morphostatis: Contextual continuity – low reflexivity (adapted from Archer,
2010a, p. 281)
In contrast, Figure 4.12 below illustrates Archer’s (2010a, p. 284) contention that increases in
reflexivity result from discontinuity of morphostatic societies, and has made a substantial
contribution to predominant reflexive action as the agencies possesses the situational logics
of morphogenesis (see Figure 4.8):
From the 1980s onwards, the synergy between multinational production
and information technology resulted in unprecedented morphogenesis,
whose generative mechanism is for variety to spawn more variety. With it,
the situational logic of opportunity began to emerge at both corporate and
106
individual levels for the first time in human history, at variance with
modernity's zero-sum "situational logic of competition." This is what
Thévenot terms the "imperative of innovation" (2006, 2008:14) and it
constitutes the condition for "the reflexive imperative." On the one hand,
exercising personal reflexivity in order to make choices in uncharted
territory means that the previous guidelines, embedded in "contextual
continuity," are fast vanishing as they become increasingly misleading. On
the other hand, the prizes in work and employment start going to those who
detect, manipulate, and find applications for links between previously
unrelated bits of knowledge; ones whose contingent complementarity could
be exploited to advantage. The "winners" become such by extruding their
skills to match the fast shifting array of opportunities or making their own
opportunities by innovating upon contingency. All of this fosters the
"reflexive imperative" because the old routine guidelines [Figure 4.11] are
no longer applicable and new ones cannot be forged because (even) nascent
morphogenesis [Figure 4.12] is inhospitable to routinization.
Figure 4.12: Morphogenesis: Contextual discontinuity – predominant reflexive action
(adapted from Archer, 2010a, p. 284)
Understanding reflexivity modes and mechanisms for acceptance helps to guide suggestions
for conditions necessary for successful EA implementation. This is predicated on Archer’s
107
suggestion that individuals will react differently to EA impositions depending on their current
projects and ultimate life-concerns.
4.5 Contextual Influences: The Dynamic of Complex Social Program
Surrounding the Architecture
EA is a means of enabling informed decision-making on IT-business transformation as well as
ensuring compliance with EA governance (Op ’t Land et al., 2009). The social complexity
surrounding the architecture is a function of the number of stakeholders involved; the variety
of concerns, socio-political dimension, diversity in their backgrounds, and work culture that
brings complex behavioural attributes into the governing of EA program and its
implementation (Op ’t Land et al., 2009; Bente et al., 2012; Janssen, 2012). The increasing
complexity of the social issues facing enterprises as well as the growing diversity and
heterogeneity of the concerns and interests of the stakeholders involved renders pre-existing
approaches less adequate.
Seen from the perspective of a complex social program surrounding the architecture, a useful
point of departure is Pawson, Wong and Owen’s (2011) realist review, which focuses on the
dynamics of complex social programs. These authors suggested:
1.
Programs are active, not passive. Interventions do not work in and of themselves; they
only have affect through the reasoning and reactions of their recipients.
2.
Programs have long implementation chains and multiple stakeholders. Recipients are
many and varied; reactions to programs thus differ; outcomes are thus generally mixed.
3.
Programs are embedded in complex social systems. Recipients are rooted in different
localities, institutions, cultures, histories, all of which shape the fortunes of a program.
4.
Programs are implemented amid the turbulence of other interventions. The policy
agenda is delivered through a multitude of interventions, each one interfering with the
reception of another.
5.
Programs beg, steal, borrow, and adapt. Practitioners work constantly to improve the
delivery of interventions rather than preserving uniformity to meet evaluation and trial
requirements.
108
6.
Programs are the offspring of previous interventions. Social problems are longstanding;
interventions evolve to try to combat them; the success of a current scheme depends on
its history.
7.
Programs change the conditions that make them work in the first place. An intervention’s
success is always time limited since alleviating a problem always involves changing its
concomitant causes (p. 519).
Prevailing organisational structures, regulations, etc., have become ingrained in agency
(people), structure, and culture of an enterprise. Due to the complexity of social programs as
interrelated business processes, people and technology, stakeholders are keen to find a way
of harnessing this complexity when judging the impact on their concerns (Op ’t Land et al,
2009).
During implementation, EA programs deal with social complexities and pre-existing social
structures within organisations. Janssen (2012) argued:
…there is a need for reconceptualising EA. We plea[d] for a broader look at
EA that includes both capabilities and a governance structure and
mechanisms. Doing EA should be incorporated in every day’s processes […].
The use and acceptance is determined by the social processes surrounding
the architecture. As such, there is a need to swap from the dominating
blueprint focus to a relational and governance focus. The use of effective
governance widens the scope from having merely a technical focus and
being an artefact toward viewing EA from a broader socio-political
perspective (p. 34).
Unfortunately, organisations with effective IT governance can still encounter an ineffective
EA implementation program. As the starting point for the change process in IT-business
transformation, EA programs face a number of significant challenges. Fundamental to this is
that EA identifies the contextual influences of the real-world social complexity surrounding
the architecture. As identified by Ross et al. (2006, p. 80) “optimized core [of the EA program]
means that local managers lose discretion over core business process and some times over
the people and the systems that execute them”. As demonstrated by Archer (1995; 2015),
109
people offer a stratified representation of their role in organisational change as they can be
seen to act as: (a) primary agents in particular positions who can generate important social
consequences; (b) corporate agents in institutions who can organise themselves in pursuit of
certain goals and articulate the changes they seek; and (c) individual actors in particular roles
who acknowledge their vested interests and weigh these interests against one another. In the
same fashion, de Vaujany (2008) argued people can also be seen to act as: a) an
individual/person with a personal and embedded history; b) agents with cultural, economic
and demographic features; and c) actors related to a social group with specific interest and
strategies. Considering the social complexity of the role of people in EA programs we see that
they are shaped by the interplay between contexts (the social situation) and concerns (the
personal concerns). In such situations people are likely to suggest that the program be more
socially dynamic with varying levels of complexity and thus influence the architecture
transformation. Figure 4.13 below illustrates how contextual influences impact EA programs.
Figure 4.13 below illustrates how EA programs feature in the contextual influences of a
dynamic social program.
Figure 4.13: Picturing EA from the dynamic of social complex program surrounding the
architecture
110
4.6 Summary: Research Question Identified from Literature
This study identified a number of research questions pursuant to a literature review of
disciplines associated with EA from a critical realist perspective. Since the EA implementation
program under study was unique given the social complexity of the university environment,
the critical realist literature primarily assisted in identifying central elements of the research
questions. To appreciate the explanatory context of the study the research questions mainly
built upon Archer’s morphogenetic approach, supported by Archer’s reflexivity theory: a)
macro-micro context (identifies the situational mechanisms of the cultural and structural
systems that pre-exist people’s action); b) micro-micro (identifies the mechanisms of action
and interaction between structure and culture); and c) micro-macro (identifies the
transformation mechanisms that constrain and enable people’s actions to adopt the EA
program or not).
Significant issues and gaps in the EA literature are used to identify the specific objectives of
the study. The findings of the literature review showed that EA programs have mainly been
viewed from a technical perspective, regardless of any real influence of the complex social
context on EA. There is limited research to date that uses the social complexity of EA programs
as a starting point from which to investigate the social responses triggered by implementation
that might constrain or enable success in the university environment.
Therefore, to explore the interplay between structure, culture and people, the main research
question of this study sought to answer: “What are the key implementation mechanisms and
social responses triggered by EA implementation that might constrain and enable the success
of the EA program in University X and the sector in general?” The main research question was
supported by four sub-questions as shown below.
111
Table 4.4 Sub-research questions and sub-research objectives
No
1
Sub-research question
Sub-research objective
What are the important situational
Identify the situational mechanisms by
mechanisms, that by virtue of associated
which social structure conditions
social structure and culture, causally
individuals’ actions and how cultural
condition individuals’ actions?
environments shape their social
situations.
2
What are the consequent interaction
Describe the interactive mechanisms
mechanisms triggered by EA
linking the social situation (context) and
implementation at University X?
the people’s personal concerns
(people’s thoughts and ideas that tend
to consider themselves – their desires,
beliefs, values, acquaintances and
interests) to influence their actions.
3
How does the existing culture and
Identify the cultural and structural
structure within UX impact the EA
dimensions of an action context, both
implementation and shape the
how they shape and are shaped by
interaction mechanisms triggered by the
groups of individuals.
implementation?
4
What are the necessary conditions to
Specify the transformational
encourage individual and collective
mechanisms by which individuals,
acceptance of EA practices?
through their actions and interactions,
generate various in-tended and
unintended social outcomes
A detailed examination of the interrelationships of the research questions will be presented
in section 5.3 and 7.2
112
CHAPTER FIVE: RESEARCH APPROACH
5.1 Overview
This chapter discusses the methodology of the research; the underlying philosophy and its
methodological framework; the explanatory context of the research questions; the research
design used for the study; its limiting factors; and the ethical considerations of the research.
Given that enterprise architecture (EA) is an organisation-wide initiative with a strong social
impact, and in order to address the need for greater recognition of the role of people and the
social aspects of EA implementation, this study used a critical realist framework and its most
recognised methodology, the morphogenetic approach (MA), as a useful tool for theorising
and examining EA and its implementation from a social perspective. The explanatory method
provided an important link between the technological and social phenomena in the domain
of information systems (IS) with critical realism (CR) as the underlying philosophy. The
research approach provides a consistent linkage of realist ontology and the methodology for
practical social outcomes, and brought together two areas of study, namely EA and social
systems.
5.2 The Underlying Philosophy and its Methodology Framework
5.2.1 The underlying philosophy
An exploration of people, who are socially transformed in the course of IT-business
transformation during EA implementation, requires a deep understanding of the structural
and cultural systems in which they operate, as well as the mechanisms of social change.
Archer (2015) argued the enterprise is “…the site of an extraordinarily powerful set of
mechanisms of social change in the modern world…”, [and] “…the structures [and cultures]
at the heart of its workings appear to be relatively unexplored” (p. 214).
It is difficult to identify inherent structures, cultures and mechanisms with traditional EA
methodologies and frameworks, particularly in knowledge societies with strong social
hierarchies such as universities. Implementing IT-business transformation provokes tension
between structure and culture, not only between IT and business people, but also between
113
departments (faculties, schools and service centres) that do not exist in industry to the same
extent, and this poses a particular challenge to implementation of EA in higher education
(Anderson & Backhouse, 2009).
This study therefore required a deep understanding of the structural and cultural systems and
the mechanisms in place to appreciate how stakeholders reacted to the new impositions;
both in terms of increased governance and the way they did things. Based on the findings in
the previous chapter, this study used CR as the underlying philosophy to incorporate a strong
social element in its analysis. According to Danermark et al. (2002) critical realism sees “…the
world as structured, differentiated, stratified and changing”, thus “within philosophy, critical
realism involves a switch from epistemology to ontology, and within ontology, a switch from
events to mechanisms” (p. 5).
CR ontology views the social world as three overlapping domains: the empirical, the actual
and the real (Bhaskar, 1978). Mingers (2004) elaborated on Bhaskar’s (1978) and Archer et
al.’s (1998) arguments to provide a more detailed understanding of CR ontology in reference
to social reality:
For Bhaskar, reality is both intransitive (existing independently of humans)
and stratified (Archer et al., 1998: p. 41). The first form of stratification is
between mechanisms; the events that they generate; and the subset of
events that are actually experienced. These are known as the domains of the
real, the actual, and the empirical […]. The real contains mechanisms, events,
and experiences—i.e., the whole of reality; the actual consists of events that
do (or do not) occur and includes the empirical, those events that are
observed or experienced (p. 93).
Bhaskar (1978) proposed events or phenomena should not be the core focus of research;
instead the focus should be on the structures (and cultures) and mechanisms that generate
phenomena. Similarly, Danermark et al. (2002) stated:
To switch from events to mechanisms means switching the attention to what
produces the events – not just to the events themselves. Reality is here
assumed to consist of several domains [the real, the actual, and the
114
empirical]. One of these is that of mechanisms. These mechanisms
sometimes generate an event. When they are experienced they become an
empirical fact. If we are to attain knowledge about underlying causal
mechanisms we must focus on these mechanisms, not only on the
empirically observable events (p. 5).
Figure 5.1 below illustrates how EA implementation uses critical realist ontology to
understand how and why EA programs work or fail to work.
Figure 5.1: Mapping the EA implementation on the critical realist ontology
Sayer (1992) supported CR ontology in reference to social reality with his conclusion that CR
generally works with causal explanatory analyses involving both abstract and concrete
studies. Figure 5.2 shows Sayer’s (cited in Danermark et al., 2002) set of events (concrete
activities), mechanisms and structures/cultures (abstract activities) as they exist in a complex
and compound whole, for example in human society. Danermark et al. (2002) explained:
When the structural mechanisms are activated, they produce certain effects,
depending on what other mechanisms they at the time happen to combine
with. A particular mechanism can produce completely different actions at
115
different times, and inversely the same event can have completely different
causes (p. 58).
Figure 5.2: Mapping Sayer’s causal explanatory analyses of concrete and abstract activities
on the critical realist ontology
Danermark et al. (2002) expressed the view that reality (in abstract activities) is not a series
of events, and a set of events is part of concrete activities:
Against empiricism and objectivism, critical realism further claims that the
method of obtaining knowledge cannot be reduced to observation of events.
Reality is not a series of events, where one thing follows on another with
empirically observable regularity. The relation between reality and our
knowledge about it comprises, as we have seen, three distinct ontological
domains: the empirical (our experiences), the actual (events), and the real,
where the mechanisms are what produce the events in the world (p. 203).
Given the heavy social emphasis of this study, it is safe to assume that applying CR as the
underlying philosophy to examine the role of people and the social aspects of EA
implementation can provide a useful perspective for theorising and examining the peoplefocused mechanisms within a IS domain. Mingers (2004) explained:
116
Critical realism is important for IS because: (i) CR enables us to take a
basically realist stance whilst accepting the major critiques of naive realism;
(ii) it addresses both natural and social science and thus encompasses the
main domains of IS; and (iii) does potentially fit well with the reality of IS as
an applied discipline (p. 97).
Dobson (2009) also stated:
In researching the social context within which information technology (IT)
and IS operate, a modern social philosophy such as critical realism has
considerable potential. It can provide useful insight into the type of
(retroductive) questions that may be asked and also the means by which an
examination can progress. The integrated nature of the philosophy
encourages a consistency in research in that it recognizes the tripartite
connections between ontology, methodology, and practical theory (p. 808).
5.2.2 The methodological framework
In order to appreciate the importance of CR as a philosophy in identifying the key mechanisms
for explaining EA success or failure, the framework proposed by Danermark et al. (2002, pp.
109 – 110) was adopted to guide the methodology for the study. This framework allows the
researcher iterative movement between concrete and abstract activities in order to refine
explanations by retroduction (Blundell, 2007). According to Sayer (cited in Blundell, 2007),
retroduction is a form of scientific inference that discerns underlying structures and
mechanisms capable of causing events in the social world.
The framework describes six stages. The first explains the often complex and composite
event, situation or phenomenon, by making use of a number of sources like theoretical
perspectives, existing research, observation and actors’ own accounts, among others.
Analytical resolution is the second stage, which separates or dissolves the composite and the
complex by distinguishing the various components, aspects or dimensions. In the third stage
this explanatory framework interprets and redescribes the different components or aspects
from hypothetical conceptual frameworks and theories about structures and relation. This
stage is referred to as “abduction” or theoretical redescription, where the original ideas of
117
the object of study are placed in a new context of ideas. It describes several different
theoretical interpretations and explanations for later comparison and integration, and is
followed by retroduction.
The fifth stage entails a comparison between different theories and abstractions and,
elaborates and estimates the relative explanatory power of the mechanisms and structures
in the problem being investigated. According to Danermark et al. (2002), this stage might
conclude with one theory describing the necessary conditions for what is to be explained, and
therefore has greater explanatory power.
Finally, concretisation and contextualisation make up the sixth stage, which examines how
different structures and mechanisms manifest themselves in alternate concrete situations,
and tests the applicability of proposed mechanisms in specific contexts of the problem being
investigated. The framework can therefore be seen as a guide for an explanatory study based
on CR.
In this study, the methodological framework (Figure 5.3) only described the initial five stages
of the explanatory framework, since the focus of the research was on uncovering the peoplefocused mechanisms by identifying the important mechanisms of EA (Stages 1 and 2) and
building a theoretical explanation of the role of people as a key element in EA implementation
(Stages 3 to 5). As suggested by Danermark et al. (2002, p. 109):
…this model (containing six different stages) should be seen as a guideline
and not as a template to be followed to the letter. Research processes can
and should be structured in different ways.
118
Figure 5.3: Research methodology framework (adapted from Danermark et al., 2002)
The first stage of the methodology framework, Description, determined a) the possible
people-focused mechanisms to explain the context for success or failure; b) the program
mechanisms that were in place at the time of EA implementation; c) the empirical facts of EA
events at University X; and d) the actors’ own accounts. This stage explains the often complex
and composite events, situations or phenomena by making use of a number of sources, such
as a) theoretical perspectives and existing research; and b) interviews and observations.
The first material source, theoretical perspectives and existing research, examined the extent
to which people are identified in existing EA research, the TOGAF framework, and existing CR
research that defines possible people-focused mechanisms and program mechanisms (EA
theory). It provided the following:
1.
From existing EA studies and theories (Chapter 2):
a) Identification of EA implementation challenges and governance; and
b) Identification of possible program mechanisms and contexts for success and failure
of EA implementation.
119
2.
From the TOGAF-based implementation framework (Chapter 2):
a) Identification of the important role of people and their relationships;
b) Identification of the important milestones and core events that need to be
considered over the implementation period;
c)
Identification of the role of people beyond purely technical capacities within the EA
maturity model (ACMM);
d) Identification of the program mechanisms involved in a TOGAF framework; and
e) Identification of the TOGAF-based implementation framework in the higher
education sector.
3.
From an industry perspective (Chapter 2): Identification of EA failures that can to propose
possible success mechanisms.
4.
From EA implementation in the higher education (HE) sector (Chapter 2):
a) Identification of the naturally conservative nature of universities;
b) Identification of the social interaction operating at the structural and cultural levels;
c)
Identification of important implementation challenges which need to be addressed;
and
d) Identification of possible program mechanisms are used at the time of EA
implementation.
5.
From a social theoretical and CR-research lens (Chapter 3 and 4):
a) Examination of various versions of social realism and a description of how the
adoption of each has fundamental implications for the research process and ultimate
outcomes;
b) Determination of critical realism as the underlying philosophy of the research, a
morphogenetic approach as the main analytical tool, and a central role for reflexivity
in the implementation program; and
c)
Identification of the contextual influences on a dynamic and complex social program
of the surrounding social context.
6.
Specifying the research questions identified from the literature review and potential
contextual mechanisms – both for the program mechanisms (Chapter 2) as well as the
social mechanisms (Chapter 4).
120
The second material source, observations and interviews, examined the empirical facts of EA
events at University X using materials obtained from observations (directly and indirectly) and
a review of archival data (secondary data). During this stage, primary data of people’s
personal accounts of EA implementation in University X were collected using interviews and
the Internal Conversation Indicator (ICONI: explained later in the following section). These
empirical materials were analysed in the second stage of the methodological framework of
the study – the Analytical Resolution.
In Stage 2 (the Analytical Resolution) the stage one representation was resolved by using
Archer’s MA as the main defining tool. This involved primary and secondary data collection
from the actual case study (University X) and provided the following:
1.
Identification of the possible underlying causal relationships between people (agency),
structure and culture in University X to gain an understanding of how stakeholders affect
and are affected by EA implementation;
2.
Examination of the interplay between agency, structure and culture over time (from Time
1 – Time 4 in MA) in University X and how they constrained and enabled EA
implementation;
3.
Identification of the different modes of conditioning/interaction in situational logics
(generative mechanisms); and
4.
Identification of the key implementation mechanisms and social responses triggered by
EA implementation and generated by the transformation mechanisms to describe the
important mechanisms and social responses that drove such large-scale architectural
transformation.
Figure 5.4 below illustrates how Archer’s MA was used in the analytical resolution stage.
Stages three, four and five also involved collection of primary and secondary data from
University X. In the case study (stage three) abduction was used to seek a connection or
relation, not directly observable, between the EA events (in the empirical domain) and the
identified EA context (mechanisms and situational logics in the real domain – MA) during EA
implementation. The abduction stage provided useful insights into the actual domain, which
are not directly observable during EA implementation. This stage identified:
121
1.
The social situations influenced by activities associated with EA implementation; and
2.
The social situations influenced by the willingness of members of University X to
cooperate with each other in order to realise EA implementation.
Figure 5.4: Mapping the morphogenetic approach on the critical realist ontology
Stage four, retroduction, described a mode of inference by which this study arrived at what is
basically characteristic and constitutive of people adopting EA, since MA (Stage 2) still posed
the central problem of how social structure, culture and agency were actually linked (their
social relationship: Archer, 2008). This stage clarified the basic conditions for social
relationships: people’s actions and interactions; identifying people-focused mechanisms;
reasoning and knowledge of EA implementation. In clarifying these basic conditions which are
not directly observable, retroduction went beyond the empirical (something that can be
observed and experienced) by using reflexivity theory (Archer, 2008; 2010a) to attain
knowledge about what internal relations made EA implementation at University X succeed or
not. This stage:
1.
Identified how people subjectively defined a particular course of action (people’s
personal concerns) in relation to their social circumstances (context of EA) by using the
reflexivity investigation tool, the ICONI (obtained from the Description stage);
122
2.
Explained how structures and cultures constrained and enabled agents (obtained from
the Analytical Resolution stage) to adopt EA programs, or in critical realist terms, the
“reflexive mediation of mechanisms”; and
3.
Identified the “collective reflexivity” or social relationships to explain the special relations
amongst people and what made them reflexive in a social, instead of a personal way with
regard to EA implementation.
Finally, the fifth stage elaborated and estimated the relative merits of each
abstraction/theory in explaining the observed happenings:
…the relative explanatory power of the mechanisms and structures (and
cultures] which have been described by means of abduction and
retroduction within the frame of stages three and four [in the methodology
framework of the study – see Figure 5.3]. (Danermark et al., 2002, p. 110)”
Typically, the purpose of this stage is to bring to light the theory of the program mechanisms
(obtained in the Description stage) to determine the conditions under which the program
theory might work or not: “what works”, “for whom” and “in what circumstance”. Pawson
and Tilley (1997; 2004) recommended a focus on understanding the theory underlying
program mechanisms to understand and explain the impact of program outcomes that might
be creating change. Understanding the theory underlying program implementation is helpful
in deriving the mechanisms by which architecture transformation is achieved or not. This
stage provided the following:
1.
Identification of the program mechanisms (obtained in the Description stage) which
contributed to CR theories that propose social mechanisms (obtained from the Analytical
Resolution stage); and
2.
Examination of how different program mechanisms (obtained in the Description stage)
supported different parts of: a) the social mechanisms proposed by MA (obtained from
the Analytical Resolution stage); and b) the reflexive mediation of people-focused
mechanisms (obtained from the Retroduction stage).
123
5.5 below illustrates how the study methodology framework plays on the ontological of CR.
Figure 5.5: Mapping the methodology framework on the critical realist ontology
In Figure 5.5 it is evident that the stages of the methodological framework of the study, when
applied to the ontology of critical realism, do not follow each other in a strictly chronological
order. For instance, the abduction stage was used to identify events and non-events in the
actual domain. However, the framework will be intertwined with the ontology, since the focus
of the study is to fit into the ontology by attempting to obtain knowledge of constitutive
qualities and causal mechanisms generating events. As explained by Danermark et al. (2002):
Such a model [the methodology framework] should be guiding the research
that is trying to attain knowledge of constitutive qualities and causal
mechanisms generating events, but also knowledge of how different
mechanisms cooperate and, under specific circumstances, contribute to the
production of concrete events and processes. (p. 108)” [Thus], “The separate
stages can also be intertwined and need not follow each other in a strictly
chronological order. In research practice it can often be necessary to switch
between the different stages. In a concrete study there may also be reasons
124
for concentrating on certain stages and touching upon the others more
lightly (p. 109).
This methodological framework guided the research process with a critical realist foundation
and brought consistency to ontology, methodology, and practical theory related to IS
research. Danermark et al. (2002) stated:
The model [the methodology framework] describes the research process as
a way from the concrete (stage 1) to the abstract (stages 2–5) […]. Every
stage (except the first) in itself involves such a swing between different levels
of abstraction. Abstraction and concretization provide two different types of
knowledge about reality, both important but not to be confused or reduced
to one another […] (p. 109).
To summarise, the methodological framework of this study steered the research process
towards an understanding of the people-focused mechanisms in the EA implementation, and
its interplay with structures, cultures, and agency (people), by building upon a theoretical
explanation of the key role of people in EA implementation, most notably in the university
sector.
5.3 Explanatory Context of Research Questions
The social realist literature primarily assisted in identifying the central element of the research
questions of the study (see Chapter 4). Significant issues and gaps identified in the EA
literature were used to define the specific objectives of the study, that is to define the
program mechanisms and social responses built into EA implementation (see Chapter 2).
Using an explanatory context with a CR-philosophy foundation, the research questions mainly
seek to identify important mechanisms that might constrain and enable the success of EA
implementation. The explanatory approach was determined by complex and often difficult
mechanisms covering causal laws (Archer, 2015), however, the approach was aimed to
provide a real basis for causal explanation, particularly social explanation. As explained by
Tilly (2001, cited in Archer, 2015, p. 161): “…explanation by mechanisms must be regarded as
one of the main explanatory strategies adopted in the social sciences…”, thus “…a deep
scientific explanation is an answer to a question of the form, “How does it work, that is, what
125
makes it tick – what are its mechanisms?”” (Gorski, 2004a, cited in Archer, 2015, p. 2). Equally
important, Archer (2015) suggested: “It is the generative mechanism that supplies the real
basis for causal laws. Nevertheless, no law simply expresses the universal manifestation of
causal powers operating in the open system that is the social world. Explanation will be realist
rather than dependent upon empiricism (p. 3).”
The research questions provided the mechanism-based explanations and involved theoryoriented accounts of episodes in the macro–micro–macro contexts from: a) Archer’s MA
schematic stage (structure, culture and agency/people: SCA); and b) reflexivity theory (only
effective as mechanisms if people adopt them) to identify mechanisms constituting the
context within which the program operated. The macroscopic context explored the interplay
between structure, culture and agency; and uncovered the social mechanisms under which
the EA program was implemented. Context also explored the program mechanisms built into
EA implementation associated with theory gathered in the EA literature review. The
microscopic context explained how people reflexively responded to the embedded
mechanisms of EA implementation. People’s reflexive deliberations also formed an important
mechanism for explaining how structures/cultures constrained and enabled agents to adopt
the program.
As discussed in chapter 4, the main research question that this research sought to address
was: “What are the key implementation mechanisms and social responses triggered by EA
implementation that might constrain and enable the success of the EA program in University
X and the sector in general?” It was supported by four sub-research questions as follows:
1.
What are the important situational mechanisms, that by associated social structure and
culture, causally condition individuals’ actions?
2.
What are the consequent interaction mechanisms triggered by EA implementation at
University X?
3.
How does the existing culture and structure within University X impact the EA
implementation
and
shape
the
interaction
mechanisms
triggered
by
the
implementation?
4.
What are the necessary conditions to encourage individual and collective acceptance of
EA practices?
126
Figure 5.6: The schematic diagram of research questions: The interplay within and between
the cycles of social change in morphogenetic approach (adapted from Archer, 1995, pp. 193,
218, 295, and 303)
Figure 5.6 presents a multi-dimensional cycle of change in the MA stage associated with the
research questions. The schematic diagram also shows the connection between the macro–
micro–macro contexts in Archer’s MA (T1 to T4). The research questions were operationalised
under Archer’s meta-theory of analytical dualism, and thus bridged the social realist theory
between the explanatory power of MA (T1 to T4) and reflexivity theory (T2 – T3) to strengthen
the study’s objectives and its outcomes, since MA still presented a fundamental problem in
terms of how social structure (and culture) and agency are linked (their social relationship)
(Archer, 2008). The schematic diagram presents a three-stage morphogenetic/morphostatic
cycle (time, from T1 to T4, is incorporated as sequential tracts) and covers social structure,
culture and agency. Through analytical dualism, the study separated social structure (and
culture) and agency, and examined their interplay to account for structuring and restructuring of the social order. This was possible because social structure, culture and agency
are different kinds of emergent entities, with different properties and powers, despite the
fact that they are crucial for each other’s formation, continuation and development. Horrocks
(2009) described it thus:
127
Although analytically and temporally separable, the three interrelated cycles
of emergence – interplay – outcome are continuous and, therefore, ‘when
studying any given problem and accompanying periodisation, the projection
of the three backwards and forwards would connect up with anterior and
posterior morphogenetic cycles’. The delineation of the cycles is according
to ‘the scope of the problem in hand’ with each cycle containing the same
three core stages. The result is that it becomes possible to set out the
conditions under which change or reproduction is likely to occur in
social/structural/cultural contexts and produce an analytical history of this
without having to resort to a determinist approach (p. 40).
In brief, through a multi-dimensional cycle of change in MA, the research questions explored
and elaborated on the role of people in EA implementation in terms of social structural,
cultural and agential change – either morphostatic (reproduction: constrained the success of
the EA program) or morphogenetic (transformation/elaboration: enabled the success of the
EA program). As explained by Archer (1995, p. 193): “the three [-stage MA cycle] are
continuously operative in society and are always interrelated because they intersect in their
middle element – since all generative mechanisms are only influential through people”. She
stressed:
…the social system is open, open because peopled, and therefore of no fixed
form due to human powers of unpredictable innovation”. Thus “…by
examining the interplay within and between the three cycles, for the
ultimate benefit of analytical dualism is that it is not a static method of
differentiation but a tool for examining the dynamics by which the 'parts'
and the 'people' shape and re-shape one another through their reciprocal
interaction over time (p. 194).
Such a focus uncovered the key implementation mechanisms and social responses triggered
by EA implementation that enabled and constrained the success of the program in the
university sector, and University X in particular.
128
5.4 Research Design
5.4.1 Objective of the research
The main objective of this study was to understand people-focused mechanisms and their
interplay between and within the structure, culture and agency (people), by building a
theoretical explanation of the key role of people in EA implementation, most notably in the
university sector. To understand how and why EA programs work or fail to work from a social
perspective, four sub-research objectives were introduced (Figure 5.6) in order to:
1.
Identify the situational mechanisms by which social structure conditions individuals’
actions and how cultural environments shape their social situations;
2.
Describe the interactive mechanisms linking the social situation (context) and people’s
personal concerns (people’s thoughts and ideas that tend to consider themselves – their
desires, beliefs, values, acquaintances and interests) to influence their actions;
3.
Identify the cultural and structural dimensions of an action context, both how they shape
and are shaped by groups of individuals; and
4.
Specify the transformational mechanisms by which individuals, through their actions and
interactions, generate various intended and unintended social outcomes.
Such an examination requires scrutiny beyond everyday events to uncover the deep causal
mechanisms involved. MA emphasises the need for recognising structure, culture, and agency
(SCA) over time; and the introduction of a temporal element placed the important macromicro-macro mechanisms in context. Reflexivity theory (micro-micro context) was used to
strengthen the study’s objectives and outcomes. These foci provided direction for examining
the interplay between the various mechanisms and their roles in structural/cultural change
for understanding the research objectives.
5.4.2 Case selection
The case study was conducted in University X (for reasons of anonymity) – a large multicampus institution serving local communities as well as a significant cohort of international
students. The university recognised that its future depended on the institution’s
implementation of EA to deliver its mission and strategic priorities. Although the study
129
focused on a single case example, the adoption of a focus on mechanisms suggests the
arguments can be applied more generally to other universities. As argued by Stake (1994,
cited in Dobson, 2001b), this approach aligns with the argument that a case study is not a
methodological choice, but a choice about the object to be studied.
Yin (2003, cited in Wynn & Williams, 2012, p. 9) also argued: “A case study is an empirical
inquiry that investigates a contemporary phenomenon in depth and within its real-life
context, especially when the boundaries between phenomenon and context are not clearly
evident”. Thus “a case study involves investigating one or a small number of social entities or
situations about which data are collected using multiple sources of data” (Easton, 2010, cited
in Wynn & Williams, 2012, p. 9).
As the main objective of the study was to gain understanding of the people-focused
mechanisms in EA implementation and its interplay with social structure, culture and agency,
Walsham’s (1993, cited in Dobson, 2003) concept of in-depth case studies as the only means
to understand human nature in context, was deemed appropriate. Similarly, Gustafsson
(2017) stated: “A case study can be defined as an intensive study about a person, a group of
people or a unit, which is aimed to generalize over several units (p. 2)”. Thus, “Single case
studies are better when the writer wants to create a high-quality theory because this type
produces extra and better theory. A single case study also makes the writer to have a deeper
understanding of the exploring subject. Other benefits are that single case studies richly can
describe the existence of phenomenon and it is better to make a single case study than a
multiple case study when the writer wants to study, for example, a person or a group of
people (p. 11)”.
In respect to the critical realist case study, Wynn and Williams (2012) stated: “…to uncover
the causal mechanisms and contextual factors that combined to generate them, case study
research is well-suited to conduct critical realist research” and “…the case study method is
the best approach to explore the interaction of structure, events, actions, and context to
identify and explicate causal mechanisms” (p. 9).
130
5.4.3 The case study
In the past two decades, the Australian higher education (university) sector has been
operating in a climate of uncertainty and change in a move towards corporatisation,
marketisation and rationalisation (Gengatharen et al., 2009). This has introduced unique
challenges in the form of a complex reality of social interaction between organisational social
structures, cultures and individuals. It is in this climate that University X (abbreviated as UX)
adopted a “One-IT” approach to all its systems, requiring EA for implementation. UX’s current
technology environment, built over many years and evolved in a random manner, had
become increasingly complex, inflexible and difficult to manage. To deliver the required
technology capability, UX tasked the Chief Information Officer (CIO) with responsibility for all
of its IS, through the establishment, maintenance and custodianship of the architecture and
governance structures.
Moreover, as stated by the CIO of UX, “global trends in higher education are resulting in
pressure on IT to respond quickly to meet new expectations of students, staff and other
stakeholders, whilst facing ever-increasing cost pressures. Therefore, the right EA
implementation will enhance the customer experience, address the need to improve
productivity and, at the same time, manage risk across UX. Since EA implementation at UX
took a university-wide view, encompassing all aspects of IS including socio-cultural change
within the university and all its complexity, UX needed a more adaptive EA that could translate
strategic directives into actionable implementation aligned with its purpose, vision, values
and goals. It will be interesting to examine the role of people’s reflexivity in the way programs
impacted their ultimate decisions concerning EA implementation, and the effect this had on
the structures, mechanisms and culture of UX.
It is widely known that the Australian education system has become a stand-out success as a
service exporter over the past decades. Large numbers of overseas students are educated in
Australia and many universities now rely on overseas markets for a large part of their revenue.
The significant numbers of overseas students enrolled in Australian universities have led the
Australian government to change its policies regarding the education system. One of the most
significant changes was a requirement for education providers to record class attendance for
international students undertaking certain levels of qualification, and to report on students
131
who had exceeded a threshold of allowable absences (Wright, 2013; DET, 2007). This means
every university needs to provide data and information electronically, requiring IT
standardisation.
The importance of EA implementation in universities is also reflected in the Council of
Australian University Directors of IT (CAUDIT). CAUDIT has been the major sponsor of the EA
Symposium in Higher Education since 2006. The CAUDIT Symposium brings together EA
stakeholders from Australian and New Zealand higher-education institutions, and provides
attendees with opportunities to: showcase and view good practice and progress in enterprise
architecture in the higher education sector; gain a snapshot of enterprise architecture in
Australasian higher education; share and debate tactics and programs that have proven to be
effective; network with people sharing similar interests from different perspectives; and
collaborate and support each other with respect to enterprise architecture into the future.
This illustrates the value of EA implementation to universities (CAUDIT, 2013).
5.4.4 The target populations and participants
The population targeted by this study comprised EA stakeholders within UX. They included
any person, group, organisation, member or system which affected or could be affected by
EA implementation. Accordingly, this study proposed two forms of agency as the population
target: a) people who affect and are affected by EA implementation – these people are
referred to as the top-down population sample; and b) people who can be affected by the EA
implementation – these people are referred to as the bottom-up sample.
Top-down participants are also referred to as EA management, whose institutional structures
and positions range from involvement in the university IT Governance Committee, the
University Architecture Board, and the Enterprise Business Group to Lead Enterprise
Architect. Bottom-up participants are referred to as the end-users and include teaching staff,
researchers, students, and administrative staff of the faculties, schools and service centres.
In alignment with Archer’s MA, EA management participants were categorised as corporate
agents and primary agents who affect and are affected by EA implementation, while the endusers were categorised as individual actors who are affected by EA implementation. Six
participants from the EA management sample played a major role in the implementation, and
132
were selected for interview based on the level of dependence of their business unit on the
university core systems, and more importantly, their role in EA implementation. Six
participants were from the EA end-user, representing the UX environment, and were
randomly selected for interview.
5.4.5 Materials, validity and reliability
The study used materials obtained from interviews, UX official documents, historical data, the
UX IT masterplan and EA implementation program, materials posted on complementary
websites, materials distributed at various meetings, presentation materials, and other
material distributed via electronic mail, as well as scrutiny of other contemporaneous
materials.
Careful consideration was given to issues of validity and reliability in the collection and
analysis of the data to provide a CR view of causality. According to Johnston and Smith (2008)
there are four principles of reliability for CR when discussing research validity:
1.
CR allows a distinction between theory and the social mechanisms (causal influences)
that the theory describes;
2.
CR allows a distinction between social mechanisms and the particular events they cause
in particular circumstances;
3.
CR allows a distinction between the actual events we would like to explain and the
empirical traces of these events that we can observe; and
4.
In CR, theory testing shows that the generative mechanism is the theory that describes
and produces the actual events and constitutes the research domain to which the theory
applies.
Zachariadis, Scott and Barrett (2013, p. 858) compared different types of conventional validity
with CR to provide a summary of the types of validity in CR qualitative research.
133
Table 5.1 Validity in qualitative research
(Adapted from Zachariadis et al., 2013, p. 860)
Validity Type
Design Validity
Conventional Description
Critical Realism
Descriptive validity: Accuracy of
Explanations of mechanisms in action
events, objects, behaviors, and
and the conditions with which they
settings reported
are interacting;
Credibility: Results are believable
appreciation of the field by
from the participants of the
identifying, prioritizing, and scoping
research.
boundaries of the study.
Transferability: Results can be
The idea that similar or related events
generalized and transferred to
that occur (or might occur) in other
other settings.
settings are caused by the generative
mechanism that caused the actual
events in the field.
Analytical
Theoretical validity: Theoretical
Theory is used to help hypothesize
Validity
explanation developed fits the
about the mechanisms and provide
data.
explanations for the events that have
occurred.
Dependability: Researchers
This is an essential part of the
describe changes in the research
retroductive process and
setting and its effects on the
identification of contingent factors.
research approach of the study.
Consistency: Verifying the steps
Challenge and inform the terms of
of qualitative research process.
(quasi-) closure and process of
ongoing inquiry in retroductive
analysis.
Plausbility: Findings of the study
Whether data that is empirically
fit the data from which they are
available gives valid knowledge about
derived.
the actual manifestation of the
alleged generative mechanism in the
field.
134
Validity Type
Conventional Description
Critical Realism
Inferential
Interpretive validity:
Findings from qualitative research can
Validity
Interpretation of participants’
provide information about the
views are accurate.
mechanisms that cause the events at
Confirmability: The results are
the empirical level.
confirmed by others.
This study was guided by the principles of reliability and validity for CR-based studies as
outlined above.
5.4.6 Data collection procedure: Interview and observation protocol
The data collection procedures emphasised the clarification of major EA events, UX’s social
structures (and cultures) and the contextual environment – the methodological principles of
CR (Wynn & Williams, 2012). The explanation of major EA events was based on TOGAF highlevel models (TOGAF, 2011) as UX’s EA implementation was based on TOGAF (i.e. the
architecture compliance models and its principles, the cross-organisational architecture
governance, and the stakeholder management). It emphasised UX stakeholders at the time
of EA implementation in the contextual environment, along with the relationships among
them, and included details of the key actions and outcomes, and the specific structural and
cultural components that were involved.
Firstly, with respect to architecture compliance models and its principles, this study set out to
determine the involvement of the various “people” elements and associated mechanisms for
conformance, and describe the action mechanisms and possible milestones to ensure
compliance over the period of EA implementation. Secondly, from the perspective of crossorganisational architecture governance and stakeholder management, this study examined
the structural and cultural readiness for adopting EA practices, reflected by the awareness of
senior management involved.
However, these major EA events (TOGAF-based implementation) may only emphasise the
powerful management elements or corporate agents with little attention on the affected
stakeholders (primary agents and individual actors), which would therefore neglect the
important actual events not recognised by TOGAF models. It was hoped that these “actual”
135
events or people-focused mechanisms would emerge to enhance our understanding of the
evolution of important mechanisms. Wynn and Williams (2012, pp. 10-11) argued that the
principle of explication of major events will “…describe the necessity to identify the detailed
aspects of events being studied, usually through the abstraction of experiences, as the
foundation of causal analysis (mechanisms).”
To identify the important mechanisms that could potentially evolve this study developed a
causal, transitive explanation of the complex socio-technical system, through empirical
observation, of the experiences perceived by UX stakeholders, together with the various
outcomes identified and measured empirically. These observations were based mainly on
casual conversations (interviews) with the two sample population groups: EA management
and EA end-users, as mentioned in section 5.4.4 above.
The interview materials were divided in two. The first was for EA management participants,
and the second, for EA end-users. Each consisted of two parts as follows:
1.
The ‘in-depth’ interview (discussion/opinion questions).
Semi-structured in-depth interviews were used as the primary method of data collection
for UX stakeholders. The questions were developed based on Archer’s MA (Archer, 1995;
1996) to identify the generative mechanisms hidden in the real domain, and to highlight
the conditions that encourage individual and collective acceptance of EA practices.
Semi-structured interviews provided flexibility and allowed new questions to be raised
during the interview as a result of what the interviewees said (Lindlof & Taylor, 2002). As
suggested by Wynn and Williams (2012) “a primary objective of scientific research
conducted under CR is to develop explanations for the way things act and how they are
capable of so doing (p. 9).” Thus “in information systems, the greatest potential
contribution of CR-based research comes from developing context-specific causal
explanations of socio-technical phenomena by explicating the specific mechanisms which
generate them (p. 9).” In addition, the interview questions emphasised the explication of
major events amongst UX stakeholders at the time of EA implementation, based on
TOGAF models (TOGAF, 2011). Examples of the interview questions for EA
management5.1 and EA end-users5.2 can be seen in textboxes 5.1 and 5.2.
136
2.
The ‘internal conversation indicator’ (ICONI).
ICONI (Archer, 2008) consists of thirteen items, assigning nearly all subjects
unambiguously to a dominant mode of reflexivity. The aim of ICONI, as suggested by
Archer (2008), is “…to identify clear practitioners of a dominant mode of reflexivity [as
the mediatory mechanisms] for in-depth interviews” (p. 4).
According to Archer (2008), reflexivity is responsible for a) “…a subject’s adoption of
projects [or context], that is, courses of action intended to realise their personal
concerns; b) for the reflexive mediation of structural and cultural properties that shape
the situations (contexts) they confront, facilitating some actions and hindering others;
and c) determining precisely what different subjects do with what intentions and with
what consequences” (pp. 4-5). ICONI is divided into 3 parts:
Part 1.
Part 1. Sets the scene by comparing and contrasting “Reflexivity’s Biographies”
(examines the distinctive nature of internal conversation – the four dominant
modes).
Part 2.
Outlines how different combinations of (natal) contexts and personal concerns
hold the key to the dominant mode of reflexivity practiced, unrelated to gender
or socio-economic background, though significantly correlated with length of
education. Qualitative features of the natal background, summarised as
contextual continuity, contextual discontinuity and contextual incongruity,
were found to be related to the dominant practice of communicative,
autonomous and meta-reflexivity respectively. This depended equally upon
the concerns of the Communicative being capable of realisation within their
natal context, those of the Autonomous being capable of fulfilment in a
different but accessible context, and those of Meta-reflexives having the
continuous promise of realisation in a succession of available contexts.
Part 3. Uses life and work histories to show how Communicatives work at staying put,
how Autonomous subjects are upward and outward bound, and how Metareflexives are continuously moving-on, thus generating their three
characteristic patterns of social immobility, upward mobility and social
volatility (Archer, 2008, p. 5).
137
This reflexivity indicator is useful in explaining how personal projects are formed and
mediate the exercise of structural/cultural constraints and enablements within EA
implementation. ICONI identifies how people subjectively define particular courses of
action (personal concerns) in relation to their objective social circumstances (the context
of EA) and explains how people’s reflexives respond to the embedded mechanisms of EA
implementation. The ‘internal conversation indicator’ (ICONI)5.3 and Scoring Subjects on
ICONI5.4 can be seen in textboxes 5.3 and 5.4.
138
Box 5.1: Example of the interview questions for EA Management
Exploring ‘people’ as the key element in Enterprise Architecture implementation:
A critical realist perspective
Case study: University X
EA Management Participant Interview
Date:
………
Interviewee Code:
………
Participant background
1. No. of Years at UX:
<2
2-4
5-10
2. Office/Department:
………………………………………
3. Position in EA Management Team:
………………………………………
4. Role in EA Management Team:
………………………………………
11-15
>15
MA Time: T1 to T4
Discussion/opinion questions:
1. UX expressed that new governance arrangements will play a key role in EA program.
What do you feel are the main changes in your department as a consequent from new
governance arrangements? How have these been accepted?
2. What do you see as the main risks in the EA implementation and associated
governance? What major “people” impacts will these have? What challenges or
concerns have you encountered?
3. The CIO expressed that there is a communication program in place to communicate the
value of EA. How important do you see this? Do you think the acceptance of the
stakeholders improved since the program started? From your point of view, are there
any departments better or worse than others in accepting the changes? Do you think
stakeholders are understanding the value of EA program better now? What do you see
in stakeholder engagement so far?
4. How is the value of EA being communicated to prospective users?
5. Are there any local “politics” impacting the EA implementation from new governance
arrangements? Do you find that individuals react to the changes differently? If so why
do you think this is so and how is it demonstrated
6. New policy requirements from government continually impact the university sector. Do
you see any particular governmental requirements/policy affecting the current EA
arrangements significantly?
7. The CIO and Lead Architect see organisational culture as very important in EA success.
Do you think UX culture is generally compatible with EA goals? How do you think UX
culture will affect EA adoption? Do you think the “culture” of individual departments
make adoption more difficult?
8. What would you define as a successful EA implementation? What can be done to
encourage the adoption of EA at UX?
9. Is there anything else that has come to mind as we talked today that you would like to
add?
N.B.:
The interview questions will be slightly different for each to appreciate the nature of semistructured in-depth interviews: allow new questions to be brought up during the interview as a
result of what the interviewee says
139
Box 5.2: Example of the interview questions for EA End-users
Exploring ‘people’ as the key element in Enterprise Architecture implementation:
A critical realist perspective
Case study: University X
EA End-users Participant Interview
Date:
………
Interviewee Code:
………
Participant background
1. No. of Years at UX:
<2
2-4
2. Faculty/School/Centre:
………………………………………
3. Position:
………………………………………
5-10
11-15
>15
MA Time: T1 to T4
Discussion/opinion questions:
1. UX expressed that new governance arrangements (e.g., policies, rules, regulatory, etc.)
will play a key role in the implementation of new IT program. Did you notice any change
in the way you use of IT over the last few months – do you think this has to do with the
new governance? Are there any considerable concerns about the new governance?
2. What projects do you currently have which require IT elements? How important is the
quality of IT to your overall satisfaction as a staff at UX? What doesn’t work – what
needs to be improved?
3. What do you see as the most effective and exciting use of IT at UX? What have you
learned – how does IT help your life at UX?
4. What do you understand by the term of enterprise architecture (EA)? Are you aware
that UX is implementing new architecture to meet the expectations of students, staff
and other stakeholders? Do you think this new architecture will effect on what you are
currently doing?
5. UX expressed that there is a communication program in place to communicate the
value of new architecture. Are you aware of this? How important do you see this?
6. This new architecture will address the needs of individuals making it easy to engage and
connect with UX through preferred channels. How important do you see this compared
to the current system? What is the possibility for learning within the new system? What
are the conditions necessary for you to accept the new system?
7. What do you think about the transformational implications of this new architecture? Do
you think this will has an implication to current UX culture (society and environment),
particularly on what you are currently doing with your project?
8. To what extent do you think this new architecture implementation will bring about
change in structure that might influence the users of IT? What do you think about the
implications of this change? What are the implications for you?
9. For you, what is the most important factor in the implementation of future technology
at UX? What circumstances that might motivate you to accept the future technology
implementations?
10. Is there anything else that has come to mind as we talked today that you would like to
add?
N.B.:
The interview questions will be slightly different for each to appreciate the nature of semistructured in-depth interviews: allow new questions to be brought up during the interview as a
result of what the interviewee says
140
Box 5.3: ICONI (Archer, 2008)
PART 1:
Some of us are aware that we are having a conversation with ourselves, silently in our heads. We might just call this ‘thinking
things over’. Is this the case for you?
YES
ON THE WHOLE
NO
Strongly
Agree
7
6
5
4
3
2
Strongly
Disagree
1
2. I think about work a great deal, even
when I am away from it.
3. I dwell long and hard on moral questions.
4. I blot difficulties out of my mind, rather
than trying to think them through.
5. My only reason for wanting to work is to be
able to pay for the things that matter to me.
6. Being decisive does not come easily to me.
7. I try to live up to an ideal, even if it
costs me a lot to do so.
8. When I consider my problems, I
just get overwhelmed by emotion.
9. So long as I know those I care about are OK,
nothing else really matters to me at all.
10. I just dither, because nothing I do can really
make a difference to how things turn out.
11. I’m dissatisfied with myself and my way of
life - both could be better than they are.
12. I know that I should play an active role
in reducing social injustice.
13. I feel helpless and powerless to deal with my
problems, however hard I try to sort them out.
1. I do daydream about winning the lottery.
PART 2:
X.
In general, what are the three most important areas of your life now - those that you care about deeply? (Please
give the most important first).
1.____________________________________
2.____________________________________
3.____________________________________
PART 3:
Y.
Your (current or last) occupation________________________________________
Z.
Age _____ Gender _______
141
Box 5.4: Scoring Subjects on ICONI (Archer, 2008)
Part 1:
1. The questions are divided into 4 categories, that is there are 3 questions indicative of
‘Communicative reflexivity’, 3 questions indicative of ‘Autonomous reflexivity’, 3 questions
indicative of ‘Meta-reflexivity and 4 questions indicative of ‘Fractured reflexivity’.
2. The scores for the four modes of reflexivity are calculated as follows:
➢ Communicative reflexive score = (Q1 + Q5 + Q9)/3
➢ Autonomous reflexive score = (Q2 + Q6* + Q11*)/3 (*= inverted)
N.B. Please note carefully that for Question 6 and Question 11, the numerical scores should
be INVERTED when calculating an individual’s score.
➢ Meta-reflexive score = (Q3 + Q7 + Q12)/3
➢ Fractured reflexive score = (Q4 + Q8 + Q10 + Q13)/4
Modes of reflexivity:
➢ Communicative reflexives: whose internal conversations need completion and confirmation
by others before leading to action.
➢ Autonomous reflexives: who complete their internal deliberations alone and act upon them.
➢ Meta-reflexives: who scrutinize and criticize their own inner dialogues, their chosen actions
and their social contexts.
➢ Fractured reflexives: who are (temporarily) unable to conduct purposeful self-talk but,
instead, augment their own distress and disorientation
3. A score of 4 and above on any of the four categories of questions assigns a subject to the C, A,
M, F category, as their dominant mode of reflexivity - whichever is their highest score over 4.
4. F scores of over 4 are held to ‘trump’ other scores. Such subjects are registered as ‘F’ regardless
of their other scores - even if these are higher.
Part 2:
5. Question ‘X’ is an open-ended question about subjects’ ultimate concerns:
Reflexivity Concerns:
➢ “C” concerns: interpersonal relationship with family and friends
➢ “A” concerns: work, career, performative achievements, financial success
➢ “M” concerns: intrinsic interests, socio-ethical pre-occupations, spiritually
➢ “F” concerns: resolving problems, establish a better way of life, overcoming present
difficulties
Part 3:
6. In addition, the background data collected used in order to test for statistical association
between the EA management participants and the EA end-users’ participants.
142
Even though ICONI uses numeric scoring subjects, it is not registered in standard quantitative
studies, since ICONI qualitatively integrates people’s subjective course of action (personal
concerns) and objective social circumstances (personal context) in a workable method of
operation that can be considered as the living link between structure and agency. Archer
(2008) suggested:
Particularly interesting is the fact that many of the trajectories of mobility
documented here would not have registered as such in standard
quantitative studies because they entailed small but very tenacious
occupational shifts. Thus, the aim was accomplished of contributing to a
theory of social stratification that gives due weight to both its voluntary and
involuntary aspects, which intersect through the medium of the internal
conversation (pp. 3 – 4).
Another essential point is that passive observations were undertaken in regular EA meetings
at UX, and this was triangulated with secondary data collected as per section 5.4.5 above.
These observations were useful to: a) examine how EA management participants (the
corporate and primary agents who affect and are affected by EA implementation)
communicate with each other; b) check definitions of terms that participants use; c) identify
how things are organised and prioritised; d) identify how people interrelate, and e) grasp the
cultural parameters (i.e. leadership, politics, social interaction) (Kawulich, 2005).
Interview questions and protocols were constructed to unearth perceived causal inferences,
while the direct observations covered contextual events in real time (Johansson, 2005). All
data were classified according to Archer’s MA tool, and all data sets were stored and coded
using qualitative software tools (Nvivo10, 2012).
The results of the data collected from the study were critically redescribed from people’s
viewpoints into theoretical perspectives (see 5.4.7 Analysis Methods). Explanations of events
(usually from experiences) identified them as a foundation for understanding the underlying
phenomena. Next, the explication of structure, culture and context identified the components
of social reality, along with agency relationships (Wynn & Williams, 2012, p. 10).
143
5.4.7 Data analysis and its methods
Data analysis identified the fundamental properties and tendencies of structural and cultural
entities to do certain things, and their relationships, which through a process of abstraction
combined to produce the emergent properties of the structure and culture as a whole. The
abstraction can be extended by redescribing the component parts of structures and cultures,
and their relationships, in terms of existing theories and frameworks to provide leverage for
potential explanation (Wynn & Williams, 2012).
To appreciate the methodological framework of the study in providing a CR ontological view
between causal explanatory analyses and mechanism-based explanation, in other words
between concrete and abstract activities, the analysis was undertaken in five stages: a)
description; b) analytical resolution; c) abduction; d) retroduction; and e) comparison with
different abstractions/theories.
1.
Description: The explanatory social science analysis
Description is the first stage in the analysis process of describing EA phenomena. The
analysis focused on the empirical domain to obtain knowledge of the empirical facts
related to major EA events at UX which are directly or indirectly observable. In terms of
direct observation, the empirical evidence was analysed qualitatively using the
stakeholders’ interpretations and perceptions (primary data: interviews). Indirect
observation comprised a review of archival data (secondary data: ranging from the
theoretical/existing research to UX’s official EA documents). The interpretations of the
people involved in EA events and their way of describing the current situation (primary
data) together with the secondary data were grouped into TOGAF high-level models to
describe the often complex and composite EA events. Coupled with review of archival
data, these stakeholders’ interpretations and perceptions are what this research study
refers to as empirical facts.
144
Figure 5.7: Mapping the empirical facts of EA events
2.
Morphogenetic approach as an analytical tool in ‘analytical resolution’ stage
The empirical facts gathered and described in the first stage of this study were analysed
using the morphogenetic approach (MA) in an attempt to identify the interplay between
cultures, structures and agency over the time of EA implementation. This determined two
important relationships: a) the relationship between a time prior to EA implementation
and the time of EA implementation (the first- and second-order emergent properties)
(see Figure 5.8); and b) the interplay between and within social structures, cultures and
agency at the time of EA implementation (second-order emergent properties) (see Figure
5.9). This evaluated the so-called analytical histories of emergence (Archer, 1995). The
following discussion explains how MA is perceived as an analytical tool to uncover
important mechanisms hidden in the real domain.
Time prior to the EA implementation
Figure 5.8: The analytical histories of emergence
145
Figure 5.8 above describes the relationship between a time prior to EA implementation
and the time of EA implementation (the first- and second-order emergent properties). In
other words, the study identified the kinds of mechanisms involved in structural and
cultural transformation and reproduction in previous MA cycles (prior to EA
implementation). This analytical history of previous MA cycles is important in describing
and identifying the situational mechanisms within the structure and culture in the second
emergent property (time of EA implementation). The situational mechanisms at the time
of EA implementation were guided by the cultural opportunities and structural
competition in previous MA cycles (Archer, 1995, 2015).
Time of the EA implementation
Figure 5.9 represents the second-order emergent properties of the MA cycle, whereby
this study attempted to uncover important mechanisms and social responses hidden in
the real domain. MA allowed for deep examination of the key implementation
mechanisms resultant from the tendency powers of generative mechanisms and the
social processes that drive such large-scale architectural transformation in the university
sector, and in UX in particular.
Figure 5.9: Second order emergent properties of MA cycles
146
Figure 5.9 above shows different SEPs and CEPs existed at Time 1 (T 1). These emergent
organisational properties or powers of the structural system (SS) and cultural system (CS)
define different modes of conditioning in situational logics, which predispose agents
towards specific courses of action to promote their own interests. The different modes
of conditioning in situational logics represent possible situational mechanisms
(generative mechanisms) that shape people’s opportunities and orientations. A stratified
model of people (PEPs) helped define how the EA program engaged with the levels in
different ways. Analysis at T1 identified the situational mechanisms by which social
structures constrained individuals’ actions and cultural environments shaped their
desires.
From Time 2 (T2) to Time 3 (T3), examination of social and socio-cultural interaction
(SI/SC) occurred to understand how the different strata pursued different courses of
action within the EA program. During the interaction stage, people developed their PEPs
based on: a) the alternatives available to the people involved; b) the restrictions that
governed the choice of alternatives; and c) their evaluation of the possible consequences
of their choice (Archer, 2010a). SI/SC interactions were viewed as being structurally
conditioned but not structurally determined, and agents themselves were seen to
possess their own irreducible emergent powers. The different modes of interaction in
situational logics represent the possible interaction mechanisms (generative
mechanisms) that shaped individuals’ and collectives’ action with regard to EA
implementation. Individuals and collectives with particular intentions, concerns and
projects located in particular roles and positions in the university began interacting with
each other. Examination of T2 to T3 constructively describes: a) the interaction
mechanisms linking the social situation (context) and the personal concerns of agents
(concerns: individuals’ desires, beliefs, etc.,) to their actions; and b) identifies the cultural
and structural dimensions of an action context, both shape and are shaped by groups of
individuals.
At Time 4 (T4) and as described by Luckett (2012, p. 324), analytical histories of particular
contexts are required to explain the outcomes of social interaction which may involve
structural and cultural change [morphogenesis: when the majority of the university
environment adopts the EA program, positive feedback predominates to elaborate or
147
change the social system’s given form/structure] or reproduction [morphostatis: when
the majority of the university environment rejects the EA program this is governed
fundamentally by negative feedback, thus preserving the social system’s given
form/structure]. Detailed explanations in regard to the occurrence of morphogenesis or
morphostatis can be seen in Figure 5.10. The T4 outcomes specify the transformational
(or reproductive) mechanisms by which individuals, through their actions and
interactions, generate various intended and unintended social outcomes.
In addition to the research objectives above, the results of Archer’s MA are also expected
to: a) identify any socio-cultural change that could result from some autonomous
processes operating at the structural level; b) identify social activities between people
(micro), who represent UX stakeholders, and the organisational (macro) features of
systems are either reproduced or transformed; c) explain how organisational agents or
actors react to the EA implementation; d) identify the key implementation mechanisms
and social responses triggered by EA implementation in the university sector and UX in
particular.
Morphostatis or morphogenesis?
Figure 5.10: The interplay between and within structures, cultures and agency
Figure 5.10 illustrates the occurrence of Morphogenesis and Morphostatis. According to
Archer (1995; 1996), the social or sociocultural integration consists of people emergent
148
properties (PEPs), while the structural/cultural integration consists of structural- and
cultural emergent properties (SEPs and CEPs). In other words, the social or socio-cultural
integration describes the level of people’s tendency in response to social change, while
the structural and cultural integration describes the level of an organisation’s tendency
with regard to social change.
Transformation will occur if people foster values of social reorientation, which can
predominate the existing social system towards social transformation. Where adoption
of the EA program by the majority of the university environment is characterised by
positive feedback, this is known as Morphogenesis:
•
People’s resistance to change is Low (PEPs: social or sociocultural integration) &
structural and cultural integration is High (SEPs and CEPs): referred to as
contingent complementary (Archer 1995; 1996).
•
People’s resistance to change is Low (PEPs: social or sociocultural integration) &
structural and cultural integration is Low (SEPs and CEPs): referred to as
contingent incompatibility (Archer 1995; 1996).
Reproduction will occur if people are reluctant to change, whereby the existing social
system remains unchanged. Social reproduction occurs when the majority of the
university environment rejects the EA program, characterised by negative feedback and
known as Morphostatis:
•
People’s resistance to change is High (PEPs: social or sociocultural integration) &
structural and cultural integration is High (SEPs and CEPs): referred to as
necessary complementary (Archer 1995; 1996).
•
People’s resistance to change is High (PEPs: social or sociocultural integration) &
structural and cultural integration is Low (SEPs and CEPs): referred to as necessary
incompatibility (Archer 1995; 1996).
3.
Abduction: Theoretical redescription/recontextualisation
Danermark et al. (2002) described abduction, or theoretical redescription or
recontextualisation thus:
149
…to observe, describe, interpret and explain something within the frame of
a new context, [as] a central element in scientific practice.” […] “The
revolution of recontextualizations is that they give a new meaning to already
known phenomena [or event]. Social science discoveries are to a large
extent associated with recontextualization. Social scientists do not discover
new events that nobody knew about before. What is discovered is
connections and relations, not directly observable, by which we can
understand and explain already known occurrences in a novel way (p. 91).
Danermark et al. (2002) emphasised that the various events in the empirical domain can
be part of and explained in relation to structures, cultures, internal relationships and
contexts which are not directly observable in the real domain. This is what these authors
called abduction – a theoretical redescription or recontextualisation whereby the original
conceptual framework or theory of an object under study, is placed in a new context of
ideas. Abduction is to move from the original concept of something to a different,
possibly better developed, deeper concept.
Umberto Eco (1984, cited in Danermark et al., 2002, p. 93) distinguished between three
different types of abduction: a) overcoded; b) undercoded; and c) creative abduction.
Overcoded abduction is a mode of inference characterised by uncontrived naturalness. It
is a matter of spontaneous interpretations which are made from an innate cultural and
social prejudging. Undercoded abduction implies a choice from a number of possible
frames of interpretations or theories. Creative abduction is characterised by uniqueness
and innovation. In the context of social science, it is a form of creative abduction when a
researcher observes something from a frame of interpretation that nobody has used
before, or which at least opposes conventional interpretations.
In view of the above it can be concluded that if an EA event has been identified in the
empirical domain as a theoretical explanation in the real domain, then the results will
lead us to a new supposition about the perceived or unperceived event in the actual
domain. Therefore, in this study, abduction is used to seek connections or relations, not
directly observable, between the EA events (in the empirical domain) and the identified
EA context (mechanisms and situational logics in the real domain – MA). This gives new
meaning to already known EA events by identifying the events and non-events in the
150
actual domain. The type of abduction used in this study is creative abduction. Figure 5.11
depicts the interconnections of EA events and contexts located in the actual domain (not
directly observable) that generated the already known EA events (the observable EA
events in the empirical domain).
Figure 5.11: ‘Abduction’ process
4.
Retroduction: The transcendental argumentation
According to Danermark et al. (2002, p. 96), the core of retroduction is transcendental
argumentation:
…to clarify the basic prerequisites or conditions for social relationships,
people’s actions, reasoning and knowledge. The term ‘conditions’ here
means the circumstances without which something can’t exist.
The essential characteristic of retroduction is its capability to provide knowledge of
transfactual conditions, which can be identified by going beyond something that can be
observed and experienced. As suggested by Danermark et al. (2002):
…realist metatheory emphasizes the difference between observable events
[in the domain of empirical] and the domains of structure and mechanisms
[domain of real]. Social reality consists of structures and internally related
151
objects containing causally operating properties. Knowledge of this social
reality can only be attained if we go beyond what is empirically observable
[beyond the domain of empirical] by asking questions about and developing
concepts of the more fundamental, transfactual conditions for the events
and phenomena under study. Retroduction is about advancing from one
thing (empirical observation of events) and arriving at something different
(a conceptualization of transfactual conditions) (p. 96).
EA implementation programs can be extremely hard to manage, not only in terms of the
architecture implementation itself, but also in terms of buy-in from large numbers of
stakeholders. EA programs contain a powerful set of mechanisms for social change with
people (stakeholders) at its heart. This study placed mechanisms at the centre of its
investigation and required identification of people’s responses to the mechanisms that
encouraged individual and collective acceptance of EA practices. These people responses,
which cannot easily be seen, can be explained through retroduction, which examines
“what is basically characteristic and constitutive (that is, people’s internal relations) of
people in adopting the EA”. While this study was able to obtain the required data in
regard to people’s interpretations and perceptions of EA adoption, the internal relations
of EA adoption were still not known. Within its conceptualisation of transfactual
conditions retroduction regards people as relational, which means this stage can
reconstruct people’s internal relations towards adopting the EA. In identifying people’s
internal relations it was also necessary to identify the basic conditions for people’s social
relationships, including the mechanisms identified (that is, the people-focused
mechanisms in MA), their reasoning and knowledge.
To clarify, these basic conditions are not directly observable within EA implementation.
Retroduction, also known as transfactual argumentation (Danermark et al., 2002), goes
beyond the empirical (something that can be observed and experienced). Reflexivity
theory was adopted in this stage (Archer, 2008; 2010a; 2010b; 2013b – see Chapter 4) as:
…the responses of individuals are obviously extremely diversified because of
their different modes and degrees of reflexivity. Social processes carry out
the selection of the influence exercised by individuals on the basis of their
different capacities and objective opportunities for responding with suitable
152
reflexivity to the challenge of having to find a relation appropriate to
themselves and their world (Donati, cited in Archer, 2015, p. 70).
People’s internal relations can be described by using reflexivity theory’s “internal
conversations” to identify how they subjectively define particular courses of action
(personal concerns) in relation to their objective social circumstances (context of EA).
This is done by means of the reflexivity investigation tool, the ICONI. These “internal
conversations” explain how people respond to embedded mechanisms (interaction
mechanisms). Questions can also be answered by attempting to reconstruct people’s
reflexive deliberation (internal conversations), thereby forming an important mechanism
for explaining how structures and cultures constrain and enable agents (reflexive
mediation of mechanisms). As suggested by Archer (2008; 2010a; 2014), the significance
of the distinctive modes of reflexivity lies in their different relationships to structural and
cultural constraints and enablements. To elaborate, Archer (2008) identified that:
[The] Communicative reflexives work at remaining embedded in their
original social ‘context’ by evading both the objective costs that would be
incurred by resisting constraints and repudiating the objective bonuses
associated with enablements. They do so in order to promote family wellbeing, which usually entails occupational self-sacrifice on their part.
Autonomous reflexives adopt a strategic stance towards constraints and
enablements, fallibly seeking to avoid society’s ‘snakes’ and to ride its
‘ladders’. They progress up the occupational hierarchy through selfdisciplined dedication to work and subordination of all other relationships
to this ‘ultimate concern’. Meta-reflexives are subversive towards
constraints and enablements because they are willing to pay the price of the
former and to forfeit the benefits of the latter in attempting to live out their
vocational concerns. All of the above are active agents who succeed in
achieving some governance over their own lives, in contrast to Fractured
reflexives who are (temporarily) passive agents to whom things happen (p.
5).
153
These modes of reflexivity can be useful in detecting the implicit knowledge of various
people’s internal relations with regard to structural and cultural constraints and
enablements (the social tendency power, the situational logics and its generative
mechanisms):
•
Communicative reflexives are extroverted chatterers […] whose internal
conversations are almost immediately relayed by interpersonal ones.
They think and talk. […] they are in fact [medians] who mistrust their lone
internal conversations and turn to significant others in their immediate
environment to talk things through and dialogically resolve their
questions. Their priorities are clear; what they care most about is […]
family and friends. […]. Shunning objective enablements to social
advancement, all of them reproduced their familial backgrounds and
showed contentment with their lot. [They] are guided by [the] traditional
action […] (Vandenberghe, 2008, p.7). In terms of structural and cultural
constraints and enablements (Archer, 1995; 1996), the social tendency of
communicative reflexives is guided by morphostatis (Archer, 2008;
2010a), as they tend to reproduce the current social system (reluctant to
change). The institutional configuration of social responses of
morphostatis is either necessary complementary or necessary
incompatibility (Archer, 1995; 1996) which leads to reproduction
mechanisms (see Table 4.3) and situational logics (see Figure 4.8) of
protection or correction. From an EA adoption point of view,
communicative reflexives will likely reject the EA program as they are
fundamentally governed by the situational logic of protection. From the
situational logic of correction they tend to refine the EA target to
reproduce the current social system that also negatively affects the EA
program.
•
Autonomous reflexives are lone thinkers […] with independent minds
whose internal reflections are primarily goal-oriented. They think and act.
Work seems to be their primary concern and, unlike communicative
154
reflexives, they subordinate their interpersonal relations to work and are
not afraid to move away from their initial context. In fact, it appears that,
early on in […] life, they had articulated life projects [personal concerns]
that burst the bounds of their social environment [personal context: e.g.
the EA program]. Keen to act on social enablements, they also know how
to circumvent anticipated constraints to accomplish their own ends. [The
autonomous] invest their lives in performative accomplishments and
[their] instrumental […] actions benefit the system and strengthen the
integration of its components (Vandenberghe, 2008, p.7). From a
structural and cultural constraints and enablements (Archer, 1995; 1996)
point of view, the social tendency of autonomous reflexives is guided by
morphogenesis (Archer, 2008; 2010a), as they tend to foster values of
social reorientation which can predominate the existing social system
towards social transformation. The institutional configuration of social
responses of morphogenesis is either contingent complementary or
contingent incompatibility (Archer, 1995; 1996) which leads to
transformation mechanisms (see Table 4.3) and situational logics (see
Figure 4.8) of opportunism or elimination. In terms of EA adoption,
autonomous reflexives are likely to use the opportunity mechanism to
gain more flexible architecture and an open architecture solution to the
EA program (opportunism situational logic), while from a situational logic
of elimination autonomous reflexives tend to eliminate the functional
areas that do not align with the EA program to meet the organisation
goals.
•
Meta-reflexives are idealists who critically reflect on their reflections
(hence “meta”) and seem genuinely concerned about their concerns,
which don’t quite mesh with their ultimate concern and which they
cannot dovetail to their own satisfaction. They think and think. Their
internal conversation is directed towards [them] selves. Preoccupied with
their selves […], they seek self-knowledge and practice self-critique for
the sake of self-improvement and self-realization. Driven by a personal
155
mission, they also criticize their environment and find [it] invariably
wanting. As no available context ever satisfies their requirements, they
are contextually unsettled and continuously on the move (even across
continents), searching for a new job, a new career, a new life, a new self.
As they cannot be bought off by inducements and are willing to pay the
price of downwards mobility to realize their ideals, they are immune to
constraints and enablements (Vandenberghe, 2008, p.7). Meta-reflexives
are unique as their social stratification can be social immobility, upward
mobility or even social volatility (Archer, 2008; 2010a). Unlike
communicatives and autonomous reflexives, the social tendency of metareflexives within structure and culture could be guided by morphogenesis
if they foster values for social re-orientation and develop organisations
that gesture towards social transformation (high morphogenesis and low
morphostatis: Archer, 2008; 2010a) or by morphostatis if they are willing
to pay the price of the constraints and forfeit the benefits of the
enablements in attempting to live out their vocational concerns (high
morphostatis and low morphogenesis: Archer, 2008; 2010a). The
institutional configuration of social responses and situational logics
depends on the personal mission (or project) which can generate the
transformation or reproduction mechanism (Archer, 2008; 2010a). In
terms of EA adoption, they can be adopters if the program does not
directly impact the way they do things (the necessary conditions that
encourage people’s acceptance of EA practices) or non-adopters if they
see the EA program negatively impacting their personal mission
(conditions that discourage people’s acceptance of EA practices).
•
Fractured reflexives are individuals with broken lives whose powers of
reflexivity have been either temporarily suspended as they are moving
from one mode of reflexivity to another or even impeded all together,
[…]. Either way, reflexivity does not work for them. The more they think
and talk to themselves, the more they get emotionally distressed and
cognitively disorientated. Unlike [other] reflexives, fractured reflexives
156
have no real [personal concerns] and no strict personal [context] either.
As their internal deliberations, do [not] allow them to deal successfully
with their situations, they are passive agents who are at the mercy of
their social environment, which affects them from without as a pseudonatural one. […] Alienated and reified into things, they are […] people to
[whom] things simply happen (Vandenberghe, 2008, p.7). In terms of
structural and cultural constraints and enablements (Archer, 1995; 1996),
the social tendency of fractured reflexives is undefined (Archer, 2008;
2010a) as it could be morphogenesis or morphostatis since they have no
real personal mission (or project). In resolving problems or overcoming
difficulties, fractured reflexives find themselves on the side of the
majority to establish a better way of life (Archer, 2008; 2010a). In terms
of EA adoption, fractured reflexives become adopters if most of their
colleagues adopt the EA program and vice versa.
Table 5.2 below provides a summary of the basic conditions for reflexivity
(people’s internal relations) to link with generative mechanisms and situational
logics resulting from MA analyses. The table provides an important overview of
how to link people’s context and personal concerns with the generative
mechanisms and social situations, both theoretical (abstract activities in the real
domain) and empirical (concrete activities in the empirical domain: i.e. the indepth interview data and ICONI scoring). Such a table can provide evidence of
how personal projects are formed (basic condition of reflexivity) and how they
mediate structural and cultural constraints and enablements (reflexive
mediation of mechanisms).
157
Table 5.2 Basic conditions of people reflexivity, generative mechanisms and situational logics
(Adapted from Archer, 1995; 1996; 2008; 2010a; 2013a; 2015)
Basic conditions of people reflexivity
Mode of
reflexivity
Communicative
reflexives
Autonomous
reflexives
Reflexivity of
personal
concern
(‘concern’)
Interpersonal
relationship with
family and friends
Work, career,
performative
achievements,
financial success
Qualitative
feature of
social
situational
(‘context’)
Contextual
continuity
Contextual
discontinuity
Social
stratification
Social
Immobility
Upward
mobility
Relationship to
structural:
constraints and
enablements
Resisting constraints and
repudiating the objective
bonuses associated with
enablements
Adopt a strategic stance
towards constraints and
enablements
Reflexive
mediation of
mechanisms:
The modes of
reflexivity for
social order
Invest themselves in
the family, thus
making a huge
contribution to
social cohesion and
to intergenerational
solidarity through
the dense microworlds they sustain
Devote themselves
strenuously to the
market and
contribute most to
economic growth
and development
Tendency power
Morphostatis
Morphogenesis
Generative mechanism:
The modes of
conditioning and
interaction in situational
logics
CEP’s
C.S: Syncretism
S-C: Unification
SEP’s
S.S: Compromise
S-I: Containment
CEP’s
C.S: Systematization
S-C: Reproduction
SEP’s
S.S: Integration
S-I: Solidarity
CEP’s
C.S: Pluralism
S-C: Cleavage
SEP’s
S.S: Competition
S-I: Polarisation
CEP’s
C.S: Specialization
S-C: Sectionalism
SEP’s
S.S: Differentiation
S-I: Diversification
Situational
Logic
Correction
Protection
Elimination
Opportunism
158
Basic conditions of people reflexivity
Mode of
reflexivity
Meta-reflexives
Fractured
reflexives
Reflexivity of
personal
concern
(‘concern’)
Intrinsic interests,
socio-ethical preoccupations,
spiritually
Resolving
problems,
establish a better
way of life,
Qualitative
feature of
social
situational
(‘context’)
Contextual
incongruity
Undefined
qualitative
feature: have
no real projects
Social
stratification
a) Social
immobility
b) Upward
mobility
c) Social
volatility
Undefined
social
stratification
Relationship to
structural:
constraints and
enablements
Reflexive
mediation of
mechanisms:
The modes of
reflexivity for
social order
a) Foster values for
social reorientation and
develop
organizations
that gesture
towards social
transformation
Tendency power
High Morphogenesis
Low Morphostatis
Subversive towards
constraints and
enablements
b) Willing to pay
the price of the
constraints and
to forfeit the
benefits of the
enablements in
attempting to
live out their
vocational
concerns
(Temporarily) passive
agents to whom things
happen
Passive – no action:
who are at the
mercy of their social
environment
High Morphostatis
Low Morphogenesis
Could be Morphostatis
or Morphogenesis
Generative mechanism:
The modes of
conditioning and
interaction in situational
logics
CEP’s
C.S: Pluralism
S-C: Cleavage
SEP’s
S.S: Competition
S-I: Polarisation
CEP’s
C.S: Specialization
S-C: Sectionalism
SEP’s
S.S: Differentiation
S-I: Diversification
CEP’s
C.S: Syncretism
S-C: Unification
SEP’s
S.S: Compromise
S-I: Containment
CEP’s
C.S: Systematization
S-C: Reproduction
SEP’s
S.S: Integration
S-I: Solidarity
No specific generative
mechanisms: Depending on
the interplay situation
between the systemic
Situational
Logic
Elimination
Opportunism
Correction
Protection
No specific
situational
logics:
Depending on
159
Basic conditions of people reflexivity
Mode of
reflexivity
Reflexivity of
personal
concern
(‘concern’)
overcoming
present
difficulties but
their internal
deliberations do
not allow them to
deal successfully
with their
situations
Qualitative
feature of
social
situational
(‘context’)
Social
stratification
Relationship to
structural:
constraints and
enablements
Reflexive
mediation of
mechanisms:
The modes of
reflexivity for
social order
Tendency power
Generative mechanism:
The modes of
conditioning and
interaction in situational
logics
integration (SEP’s + CEP’S)
and the social integration
(PEP’s) in the organisation
under study.
Situational
Logic
the interplay
situation
between the
systemic
integration
(SEP’s + CEP’S)
and the social
integration
(PEP’s) in the
organisation
under study
160
5.
Comparison with different abstractions stage: An understanding of the EA
implementation phenomena through abstractions/theories
As discussed above, the final stage of analysis is expected to elaborate and estimate the
relative merits of each abstraction/theory in explaining the observed happenings: “…the
relative explanatory power of the mechanisms and structures [and cultures] which have
been described by means of abduction and retroduction within the frame of stages three
and four [in the methodological framework of the study]” (See Figure 4.3) (Danermark et
al., 2002, p. 110).
Danermark et al. (2002) concluded:
Abduction can be redescribing and giving meaning to events, taking one’s
starting point in a theory, a coherent system of ideas or concepts. Through
retroduction, concepts and theories are developed which can provide
answers to such questions as: What characteristics make X what X is? (p. 120).
Importantly, Danermark et al. (2002) suggested theories are abstractions and
indispensable when it comes to explanation, since they conceptualise causal
mechanisms. Therefore, the main objective of this stage (see Figures 5.3; 5.4) was to
bring the theory of the program mechanisms to the fore, which had been found in the
empirical domain (description stage) to determine the conditions under which the
program theory might support: a) the social mechanisms that generated the events and
non-events; and b) to support the people-focused mechanisms (reflexive mediation of
mechanisms) in underpinning the stakeholders’ interactions.
Pawson and Tilley (1997) proposed an important focus on understanding the nature of
programs and how they work. They concluded programs are “… theories, they are
embedded, they are active, and they are parts of open systems (p. 3). These authors also
identified mechanisms as a basic concept in explaining and understanding programs:
Mechanisms describe what it is about programmes and interventions that
bring about any effects. Mechanisms are often hidden, rather as the
workings of a clock cannot be seen but drive the patterned movements of
161
the hands. This realist concept tries to break the lazy linguistic habit of basing
evaluation on the question of whether ‘programmes work’. In fact, it is not
programmes that work but the resources they offer to enable their subjects
to make them work. This process of how subjects interpret and act upon the
intervention stratagem is known as the programme ‘mechanism’ and it is
the pivot around which realist research revolves. Realist evaluation begins
with the researcher positing the potential processes through which a
programme may work as a prelude to testing them.
Based on Pawson and Tilley’s (1997) description, this study defined program
mechanisms, as discussed in section 1.1 – Chapter 1, as the underlying entities and
processes which operate in the context of success in EA implementation programs to
generate outcomes of interest in respect of architecture transformation. Thus, program
mechanisms are the EA theories underlying the implementation, and can also be referred
to as the embedded EA theories built into the EA implementation program (see Chapter
2 for a list of program mechanisms identified in EA theory). Understanding the theory
underlying program implementation is therefore helpful in deriving the mechanisms by
which architecture transformation is achieved or not.
The relative merits of each EA theory (abstraction) also provides an important overview
of the relationship between theory (abstraction, i.e. the social mechanisms) and
observation/data (concrete, i.e. the ICONI – reflexive mediation of mechanisms) in
studies with a CR foundation. Danermark et al. (2002) suggested:
We can never understand, analyse or categorize reality without using a
theoretical language of concepts. (p. 116)” Thus “What is important is that
we conceptualize events, mechanisms and internal relations in a certain
way, with the help of theories. Conceptualizing is the same thing as
abstracting and isolating fundamental qualities […]. The concepts provide an
abstract language enabling us to speak about qualitative properties,
structures [and cultures] and mechanisms (p. 120).
162
Equally important, Danermark et al. (2002) argued:
To a large extent the objects of social science are such social situations,
relations, processes and structures that never appear as given facts or as
something directly observable. Social relations and structures can only be
understood by means of concepts. It is a fairly common notion that our
knowledge increases primarily through data collection and surveys. With
Bhaskar’s terminology we can say that such a notion is based on a double
reduction. Our knowledge of social reality is reduced to knowledge of
events, which in turn is reduced to empirical observations of these events
(p. 117).
5.5 Limitations
As with most research projects, this study also has its limitations. These are associated with
the characteristics of the research approach that might impact or influence the interpretation
of the findings. The following limitations are acknowledged:
1.
In regard to the research approach, the methodological principles of CR require that:
“…explanation via mechanisms must specify the [structural] powers and [cultural]
propensities of that particular mechanism and identify the [people] causal tendencies
produced by it” (Archer, 2015, p. 53). As a result, those methodological processes will
require the researcher to formulate a new idea about the interconnection of the EA
phenomena resulting from the theories (abstraction) and the empirical case under study
(concrete). In view of the time limit for completion of this study this may not be fully
realised. However, it was anticipated that a new concept, resulting from the research
methodology and findings, will provide a unique opportunity for proposing amendments
to existing EA methodologies and frameworks to more clearly reflect the critical role of
people in the change process of EA implementation.
2.
Given that CR is particularly useful for examining the social aspects of the environment
in which IS resides (Mingers, 2004; Wynn & Williams, 2008; Dobson, 2009; Bygstad &
Munkvold, 2011; Carlsson, 2011), the technical aspects of IS could be considered not to
have significant benefits. Nonetheless, as EA is a means of enabling informed decision163
making in IT-business transformation as well as ensuring compliance with EA governance
(decisions made), viewing EA as a social program will reduce ineffective EA
implementation with regard to the technical aspects of IS – the non-technical aspects will
affect the technical aspects. Fundamentally EA provides the contextual influences of realworld social complexity surrounding the architecture, and CR can increase our
understanding of causal mechanisms and contexts needed to achieve positive outcomes
in EA implementation.
3.
The limited sample size may make it difficult to identify significant relationships from the
data, since samples normally require larger numbers to ensure representative
distribution of the population. Nevertheless, sample size is less relevant in qualitative
research – as Morse (2000) suggests, there are no specific rules when determining an
appropriate sample size in qualitative research, and qualitative sample size may best be
determined by the time allotted, resources available, and study objectives (Patton, 1990).
This qualitative study used semi-structured, in-depth interviews with a case study to
emphasise the need for understanding human nature in the social context of EA
implementation. It was designed to identify the important mechanisms hidden in the real
domain and highlight the conditions that encourage individual and collective acceptance
of EA practices.
4.
Finally, due to the difficulty of engaging an organisation that is implementing EA as a
condition for investigating EA implementation, this study was limited to a single
organisation case study. However, it was anticipated that the rich data obtained from this
study would be useful to other researchers seeking to develop a better understanding of
people as a key element in EA implementation, particularly in the higher education
sector.
5.6 Ethical Considerations
This study was guided by the “Australian Code for the Responsible Conduct of Research” with
regard to the ethical oversight of research involving humans: National Statement on Ethical
Conduct in Human Research (2007) - Updated May 2015. Semi-structured, in-depth
interviews were conducted to gain an understanding of how organisational stakeholders
affect and are affected by EA implementation. It was essential to obtain genuine opinions,
164
even if they were opposed to the EA program. The following statements and ethical guidelines
were contained in the information letter to participants in order to gain their trust and
confidence for revealing their true opinions:
1.
Potential Harms/Benefits: There are no identifiable risks associated with participation in
this research. The benefits to participants are to gain a deeper knowledge related to the
people roles of EA, along with a greater understanding of the dynamics of the complex
social situation surrounding the architectural program. The researcher will share with the
participants the deep understanding and recommendations as to the possible
mechanisms that most likely lead to desirable outcomes. The sharing of academic
knowledge should assist the organisation to achieve greater positive outcomes. The
results are also expected to be applicable to other organisations in different disciplines
within Australian universities and elsewhere.
2.
Participation: Participation is completely voluntary. Confidential interviews should take
about one hour, at a location convenient to the participant. The interview questions were
provided before the interview was conducted. Participants were free to ask questions
and review their answers. Also, the decision to participate was not binding, and
participants could withdraw at any time and end the meeting without reason and without
negative consequences. Participants could request copies of the written paper for their
own purposes if so desired.
3.
Confidentiality: Participants and his/her organisation were not able to be identified
individually so that responses remained anonymous. The interviews were recorded for
the purpose of transcribing the content. Transcripts were provided to the participants
upon request. Any recording stored on computer has been protected by password. The
researcher will erase digital recordings when all the reports for the research have been
accepted. All written notes by the researcher were immediately coded in order to protect
the participants’ personal information. Notes and transcripts have been stored in a
locked cabinet at the researcher’s university. The data may also be shared with the
researcher’s supervisors who are guiding the work of the student. Any information or
details given for this study will be kept confidential, will be used for the purposes of this
research, and may be used in related future research. If a participant has any concerns or
complaints about the research project and wishes to talk to an independent person, they
165
were provided with the contact details of the Research Ethics Officer at the researcher’s
university.
4.
Consent: Consent related to making an informed, voluntary decision to participate was
sought from individuals participating in the study. By signing the letter of consent as a
participant and acknowledging that he/she had read and understood the information,
consent to participate in the research was obtained with the proviso that he/she could
withdraw at any time.
5.7 Summary
This chapter focused on the research approach for investigating social mechanisms in EA
implementation. The study used realism, particularly CR, as the underlying philosophy to
investigate the social responses triggered by EA implementation that constrained or enabled
the success of the EA program in the university environment, thus providing an account of
how the underlying mechanisms work. Such an investigation requires a focus beyond
everyday events to uncover the deep causal mechanisms involved. Given the nature of the
explanatory context in identifying the mechanisms, the research approach required a
definition that treats mechanisms as ontological in its methodological framework (Archer,
2015). Danermark et al.’s (2002) model of explanatory research, based on CR, was adopted
as a framework to guide the investigation into mechanisms. The framework allowed the
researcher iterative movement between the concrete activities (the empirical and actual
domains) and the abstract activities of the real social world (the generative mechanisms
hidden in the real domain that generated the key implementation mechanisms) in order to
explain the observed events.
Within this framework the research design used Archer’s MA (Archer, 1995; 1996; 2013a;
2015) as a useful tool for theorising and examining EA implementation and to identify the
generative mechanisms and its key implementation mechanisms. Archer’s reflexivity theory
(Archer, 2003; 2008; 2010a; 2010b; 2013b) was adopted to identify how people subjectively
define particular courses of action (personal concerns) in relation to their objective social
circumstances (their context of EA). The data collection was mainly guided by Archer’s MA
and reflexivity theory to identify the important mechanisms hidden in the real domain and
highlight the conditions that encourage individual and collective acceptance of EA practices.
166
Although this study involved a single case example, the focus on mechanisms allows the
arguments to be extended more generally to other universities.
The central theme of the research design formed an important link between the technological
and social phenomena in the study and CR as the underlying philosophy. The following
chapter presents the detailed findings of the contextual success mechanisms identified from
the case study, along with a discussion of the potential consequences of mechanisms in
different time settings (prior to EA implementation and at the time of EA implementation)
within the university under study. The theoretical and practical implications identified from
the case study are also explained and illustrated.
167
CHAPTER SIX: CASE STUDY, DATA ANALYSIS AND
RESEARCH FINDINGS
6.1 Overview
This chapter presents the data analysis and research findings from the enterprise architecture
(EA) implementation at University X (UX). An analysis of the abstract and the concrete in
providing a critical realist (CR) ontological view of causal and mechanism-based explanations
is provided. A brief summary of UX EA Program6.1 from initiation to implementation is also
described. The findings are presented in three sections (see Figure 6.1). The first section
discusses the EA events already experienced by the UX community (people); i.e. the empirical
facts in the empirical domain. This domain encompasses EA events that are perceived or
experienced directly or indirectly.
Figure 6.1: The ontological map
The following sections discusses the real domain and identifies the hidden mechanisms and
social responses (the situational logics of structure and culture) that endure the potential
powers and properties activated by people’s actions and triggered by EA implementation. It
is in this domain that mechanisms are identified. When mechanisms produce factual EA
168
events and non-events they fall into the actual domain, whether people experience them or
not. This domain describes how each mechanism has fundamental implications for how EA
implementation generates perceived or unperceived EA events.
Box 6.1: UX EA Program: Initiation to Implementation (2012 – 2016)
Background:
The current technology environment at the UX has been built over many years and evolved
in an unplanned manner. It has become increasingly complex, difficult to manage and
inflexible. The IT complexity problems mainly stems from: a) silo-based organisational
partitions in the IT investment; b) complex business operations arising from the inherent
complexity of business itself; c) global emphasis – changes in technology and HE trends; d)
no architectural initiatives – leading to improvisation instead of following an architecture
roadmap; and e) poor governance and decision making due to locally organised IT
governance.
The executive decision:
In August 2012, the board of university executive had decided that governance structures
and a new enterprise architecture (EA) program will play a key role in delivery of IT at UX
in future. The program will take a university-wide view encompassing all aspects of IT
including people, processes and technology. Since then the EA program was developed
and, at the end of June 2013 the planned EA strategy and governance was implemented.
The executive direction for EA program:
In early 2014, a Steering Committee, chaired by a high ranking UX executive, appointed
various Enterprise Business Groups (EBGs) to represent different cross-sections of
Faculties, Schools and Service Centres to reflect those centres expected to play a key part
in the EA program. Subsequently, a UX IT Governance Committee (ITGC) was also formed
and chaired by the Chief Information Officer (CIO). This was then followed by the
establishment of the University Architecture Board (UAB). The objective of each EBG is to
make certain that initiatives for the relative business area have sound business oversight
and follow appropriate UX wide technology governance requirements. The EBG meeting
frequency was to be determined by the volume of work and projects, and reporting
requirements to the ITGC. The objective of the ITGC is to ensure that the University wide
compliance to technology governance principles was addressed in a structured manner
with recommendations made to the university are based on sound governance practices
that have been designed to deliver the University’s objectives for Information Technology.
The role of the UAB is to make sure UX wide compliance to the UX enterprise architecture
and roadmap and to make recommendations to the EBG on technology investments.
Since mid-2014, the EBGs, ITGC, and UAB have played an important role incrossorganisational strategy and IT governance. For instance, the EA program can oversee
implementation and obtain appropriate political backing, focusing on the rights, roles, and
equitable treatment of shareholders, the program can also identify those investments that
best support the achievement of UX’s strategic goals.
169
6.2 Domain of Empirical
The empirical domain incorporates the empirical facts of EA implementation events which are
directly or indirectly observable. The empirical facts were obtained from:
1.
Primary data (directly observable – interviews): Interpretations of the people involved in
EA events and their descriptions of the current situation. Six participants from the EA
management sample played a major role in the implementation, and were selected for
interview based on the level of dependence of their business unit on the university core
systems, and more importantly, their role in EA implementation (referred to as the
corporate and primary agents who affect and are affected by EA implementation). Six
participants from the EA end-user, representing the UX environment, were randomly
selected for interview (referred to as the individual actors), see section 5.4.4.
2.
Secondary data (indirectly observable):
a) A review of archival data. Materials obtained from UX official documents, historical
data, the UX IT masterplan and EA implementation program, materials posted on
complementary websites, materials distributed at various meetings, presentation
materials, and other material distributed via electronic mail, as well as scrutiny of
other contemporaneous materials, see section 5.4.5.
b) Passive observations in regular EA meetings. These observations were useful to:
examine how EA management participants (the corporate and primary agents who
affect and are affected by EA implementation) communicate with each other; check
definitions of terms that participants use; identify how things are organised and
prioritised; identify how people interrelate, and grasp the cultural parameters (i.e.
leadership, politics, social interaction) (Kawulich, 2005), see section 5.4.6.
These primary and secondary data were grouped into TOGAF high-level models to describe
the complex and composite EA events (see Figure 5.7). Members of the corporate groups (the
corporate agents: members of the university’s IT Governance Committee, University
Architecture Board, and Enterprise Business Groups) (see Chapter 5) emphasised the
importance of TOGAF high-level models within the UX EA implementation program:
170
Corporate agent A6.2:
We probably don’t dive down sometimes to as low a level as what TOGAF in
a purist form would do, but we sometimes keep it upper level [TOGAF highlevel models] because we think that’s as far as we need to go. You know
we’re not
doing software
development;
we’re outsourcing our
infrastructure development and all of that sort of stuff, so there’s some
things where we don’t need to go down as low level. We focused more on
the “principles” around TOGAF than all the diagrams and all the... we’ll do
that when we need to. For example, in research administration we’ve
mapped their processes. The next piece of work is data modelling in their
area and we’re getting a lot of resistance there because they don’t see the
benefit of that. So now we’ve got to explain to them “well, you’ve got
research administration systems now and when you need to do your
‘compliance’ reporting and all of that sort of thing there’s about three
months of work for about five people to do this because you’re not even
capturing half the data”. And that’s why data modelling is very important.
So we’ll go down to where we need to go down to, to do what we need to
do, and a data model for research administration has definitely got to be
done because their lack of understanding... they’ll want to go and buy this
system without any thought to the day-to-day need to capture and the
processes they need to support.
Box 6.2: Interview Quotation
The interview quotations are used to add analytical depth to the data analysis and
findings: i.e. the roles and the original descriptive phrases highlight the subject
discussed and provide support and credibility.
171
Corporate agent B:
TOGAF is a very good… well it’s the primary-used EA framework. It’s quite
pragmatic which suits me down to a tee. From our perspective we’re still
very early on in our maturity, so we’re adopting components of TOGAF that
make sense [TOGAF high-level models], primarily from the artefact
repository, so where it makes sense we’re utilising those things.
One of the principles around TOGAF is tailoring TOGAF to make sense, so if
it doesn’t make sense for the organisational level of importance it puts on
communication, then if you’re applying TOGAF correctly, you should be
modifying it or adapting that process to make sense of your organisation.
The technical jargon of things like TOGAF throwing out about particular
artefacts [low-level models], you know, it doesn’t help the situation when
you’re calling things and you think IT acronyms or you think techno jargon
to explain stuff – the biggest barrier is actually technology in itself and the
way that we as technology professionals communicate. A lot of the stuff that
we’ve been trying to do within our team is trying to avoid utilising catalogue
items that make those statements, so firstly we’re talking about current
state. Don’t worry about a particular TOGAF artefact, it talks about process
flow or whatever else, we’re talking quite simply about documenting how
you do things currently. It’s taking the technology jargon out [low-level
models], I think personally.
Data analysis and findings:
Four major EA implementation events, based on TOGAF high-level models, describe the
important category of mechanisms underlying successful large-scale architecture
transformation in the university sector (see Figure 5.7), the particular events within each
overall category are listed below each category:
1.
Events related to the architecture compliance:
a) Service Level Agreements (SLAs) – to ensure support of the IT services (operational
systems) of the university’s functional areas at desired service levels.
172
b) Operational Level Agreements (OLAs) – to ensure all architecture artefacts, contracts,
and principles of the university’s functional areas are at desired operational levels.
c) Technology standards catalogue – to ensure the application portfolio of the
university’s functional areas are in place to gain a baseline view of EA compliance and
technology standards. These standards are justifiably credited with reducing the
university’s cost, enhancing technological interoperability and leveraging innovation.
d) Communication guidelines – to support effective communication and decisionmaking, and manage the execution of communication between the EA program and
stakeholders, and between the EA program and the consumers of its services (the
university’s functional areas).
e) Regulatory – the university’s decisions with regard to the EA program, such as which
IT policies are negotiable and which must be enforced for regulatory or statutory
reasons.
2.
Events related to the architecture principles:
a) Events that govern the architecture process, affect implementation and use of the EA.
b) Events that govern implementation of the architecture, establishing the first tenets
and related guidance for designing and developing information systems.
3.
Events related to the cross-organisational architecture governance: Events that oversee
implementation and obtain appropriate political backing, focusing on the rights, roles,
and equitable treatment of shareholders. These events produce several outcomes
related to the Strategic EA Governance Process (see Figure 6.2) which encompass all
areas of UX and support the strategic, operational and technical decision-making
processes required to ensure IT enables the university to deliver its objectives. The
Strategic EA Governance process outlines the decision-making rights and controls that
take place between formulation of the business concept and decision-making to fund and
initiate the work as a project. These events are part of the EA collaboration program.
173
Figure 6.2: The University cross-organisational architecture governance
4.
Events related to the EA roadshows and workshops for the university community:
These events are part of the EA communication program to explain the architecture
roadmaps and strategies to the university community. They are a key part of EA
implementation designed to:
a) establish a fact base and prepare the university for the Future Scenario Workshops,
b) communicate issues to and from the program, and
c) be actively involved in analysis and design of program outcomes
In brief, these four major categories of EA implementation events were guided by the EA
communication and collaboration program led by university executives. The next section goes
deeper to explain the reality of related objects, their powers and mechanisms, such objects
often not directly observable within the empirical domain.
6.3 Domain of Real
The real domain is the deep dimension where explanation is sought to explain university X’s
(UX) social structure and cultural influences (the generative powers of SEPs and CEPs - see
Chapter 4) and how they are mediated by people shaping the situations (the situational logics)
in which they subjectively choose particular courses of action in relation to their objective
social circumstances. In other words, this domain is where generative mechanisms that
enable the success of the EA implementation (often not directly observable) are to be found.
174
In order to identify these generative mechanisms an analysis was undertaken in four stages
(see Figure 5.5, Chapter 5): a) analytical resolution; b) abduction; c) retroduction; and d) a
comparison of different abstractions/theories.
6.3.1 Analytical resolution stage
Using Archer’s morphogenetic approach (MA), empirical material gathered (see chapter 5)
was elaborated and analysed in terms of the three defined part cycles (periods of time), each
with relative autonomy yet interacting with each other. In terms of the MA, this analysis, also
called the analytical histories of emergence (Archer, 1995) is designed to uncover: a) the
relationship between a time prior to EA implementation and the time of EA implementation
(the first- and second-order emergent properties; see Chapter 4); and b) the interplay
between and within social structures, culture and agency at the time of EA implementation
(second-order emergent properties).
Figure 6.3: The analytical resolution stage
175
Data analysis and findings:
This subsection presents the analysis of the data and discusses the research findings, based
on the research method and theoretical framework that have been outlined earlier in the
thesis (see Chapter 5). The subsection also provides the supporting evidence (data: interviews
quotation) for building greater credibility of the findings.
1.
Time prior to EA implementation: First-order emergent properties
University X (UX) is a young Australian public university formed from an amalgamation of
Australian colleges, with over a hundred years of experience in higher education. As a
young modern university, almost thirty years after amalgamation, UX possesses a unique
cultural and social structure. Its history shows UX demonstrates openness and tolerance
towards cultural and structural changes, as indicated by a member of the corporate
group:
A university is a university and we all do similar things, but I think the
implementation’s very likely to be different, and I think that’s because of the
cultural aspects. And [University X] has a very different culture to some of
the more traditional universities – we’re much more ready to look
holistically at our business model rather than have different parts of the
University look at just their bit. I’m not saying that that’s not a problem, it
still is a problem here, but it’s much less of a problem than it is at other
universities. And if we look at say [for example, the University Y], one of the
sandstone universities, there’s a lot of tradition in those universities as to
how things are done and a lot less willingness to change the way things are
done. What I can see of [UX], it’s an extremely adaptive university and I think
that’s a lot to do with the youth at the University. It’s a young university who
just doesn’t have that hundreds of years of tradition and I think the culture
in the University is very different to some of those types of universities.
From a critical realist viewpoint, amalgamation of Australian colleges into a university
(UX) created the necessary conditions for transforming the social system into a higher
order, based on the need for a sustainable future. This first social-system transformation
176
at UX is what Archer called the first morphogenesis (1995; 2015), or the first structural
and cultural transformation within the MA cycle – the first-order emergent property.
First-order relationships determine the potential bargaining power of primary agent
collectivities (members of UX business units/functional areas that are also members of
UX Enterprise Business Groups - see Chapter 5) on second-order emergent properties.
2.
Time of EA implementation: Second-order emergent properties
In the decades after its formation university stakeholders have gradually adopted various
socio-cultural practices (habitual actions, established routines, traditional preserves or
conventional divisions of activities). With the commencement of the EA program
implementation, many of the established forms of socio-cultural practices are
threatened, owing to the fact that EA is widely regarded as the starting point for a
significant process of change. To some university stakeholders the EA program also
meant that UX’s business units and functional areas (faculties, schools, and service
centres) may lose discretion over core business processes, even over the people and
systems that execute them. Some of the primary agents articulated the main changes
impacting their departments as a consequence of EA implementation thus:
Primary agent A:
…I guess at UX, instead of everyone managing their own IT like we’ve been
doing, it’s really to put it together as a corporate framework and a corporate
process. I’m very much in favour of that and why I say that is I think we had
[xxx, number of] people who really didn’t have a career path because they
just looked after our systems, so where do they go after that? So I actually
lost the money for that and the money went across to IT department and it
went into the bigger pool. Now I probably lost some level of service and
everyone says ‘we need our own resources’ …
Primary agent B:
I had one committee called [‘z’ committee] which was the [Department of X
Systems Reference Group] – a committee that came together and we talked
177
to the library and included IT and all the service centres, to talk about IT and
governance. Now that committee which was called [‘z’ committee], looks
like it probably will no longer have a role because that role really has been
superseded by [the IT Governance Committee, the Enterprise Business
Groups, and the University Architecture Board]. So that would be one of the
first things. The second major change from my perspective as well is that
everyone really now needs to think very long-term. People can no longer say
‘oh, I just need this little tweak done to my system in order to make an
improvement’ because that’s not really feasible anymore. So people [from
UX Business Unit] now have to fill out the forms that are submitted to the
[University Architecture Board], […], if I may be perfectly honest, […] any
programs which come through [my department], improvements around IT,
[the Enterprise Business Groups] then just get slotted into the full list of all
of the IT opportunities. And so while something may come through [my
department] as being extremely important, must be done now, category five
rating, in the overall scheme of things the big [Enterprise Business Groups],
the top governance, they might say ‘well no, we need to do these other
things first’. So the changes then really are about [the ‘z’ committee], having
to think long-term and realising that while we see something as a priority,
potentially the University might not see it as a priority.
Another primary agent indicated the main risks of EA implementation impacting the
associated governance as follows:
Primary agent C:
I think when you get complexity and lack of understanding [of EA program]
you get all sorts of people feeling ‘I don’t trust this, I don’t…’ You know, you
have a lack of acceptance by the business in there. And also too, a lot of it’s
not visible and therefore it’s technical as well as being invisible, and people
are just innately suspicious and therefore hostile to what these things are.
178
Consistent with these arguments, a corporate group member indicated:
… [Some people] are possibly the ones that have the most to lose out of
enterprise architecture. Maybe previously they had a bit more free rein to
spend their IT budget on the things they thought were the most important,
which wouldn’t necessarily be the things that will deliver the most value to
the University or didn’t align with the strategic direction of the University,
whereas now they might view it as an enterprise architecture roadmap for
their business area or it might constrain what they would want to do because
it would give the University reason to say ‘no, you can’t invest in that
because it’s not on your roadmap’.
Nevertheless, the empirical domain shows that the EA implementation, after 4 years,
finally achieved what was expected without significant difficulties in transitioning the IT
architecture. One corporate group member indicated the current status of the EA
implementation:
…we’ve certainly got our critical strategies in place and the roadmaps for our
critical systems in place. We have an ad vocation portfolio management
piece that’s been done, so that we know with our existing applications
whether we’re going to maintain them, enhance them, retire them,
whatever we’re going to do. […] …it’s to the stage we can say that all of the
enterprise architecture work [has] been done...
What was not known were the key implementation mechanisms and social responses
triggered by EA implementation that enabled the success of the program at University X.
Using Archer’s MA as the main defining tool, this study tried to find the important
mechanisms and social responses in the real domain. The following discussion describes
how MA supports findings related to second-order emergent properties (at the time of
EA implementation) and helps explain the key mechanisms and social responses that
propelled architectural transformation in UX (see Figure 5.8 in Chapter 5).
179
Time 1: Modes of conditioning of the cultural system (CS):
Macro (cultural) - Micro (people) context of cultural system
Analytical history6.3 of cultural system:
Time 1 is the initial stage of EA implementation at UX (see Figure 6.3) when modes of
conditioning in the cultural system predisposed agents’ (people) to use IT for promoting
their important projects. An analytical history shows that in the past, an agreement was
made with the university stakeholders whereby the university’s functional areas have the
freedom to organise their own IT environment in the way most suited to their natural
competence, capability, skill, and qualification in respect of their specific business unit
activities. Each business unit or functional area (faculties, schools, and service centres)
has been shaped by cultural opportunities as a result of past socio-cultural interaction
(previous MA cycle outcomes: first-order emergent properties).
Box 6.3: Analytical History
The features of analytical history are strategically important to provide the history
of emergence within which particular situations (situational logics) can occur. The
analytical history can help to explain the nature of the situations and how they
affect conduct. As explained by Archer (1995, p. 167): “Since the existence of
effects cannot serve to explain origins (a prime error of functionalism) then the
task of social theory cannot be restricted to the mere identification of social
structures [and cultures] as emergent properties, it must also supply an analytical
history of their emergence which accounts for why matters are so and not
otherwise. Equally, once they are so, they constitute part of the social
environment, and, as with any other environmental influence, we can neither
assume that agents are determined by them nor are immune from them, but can
only examine the interplay between the powers of the two”. Thus: “The end-point
and the whole point of examining any particular cycle is that we will then have
provided an analytical history of emergence of the problematic properties under
investigation” (p. 91).
180
A report from UX’s EA project team reveals: “Over the years, technology use within [UX]
seems to have stagnated to the use of data defined by the functional area that consumes
the data produced by the provision of the service. …Most of the [IT] services provided at
the university are determined by the university functional areas responsible for the
service provided. Most of the [IT] services were manual in nature.” Another EA report
revealed that most of the IT products and services in UX functional areas focused on
specializations to suit their own interests rather than university-wide interests.
Information gathered from UX’s EA project team indicated the cultural milieu at Time 1
was characterised by a silo approach to IT. For instance: a) there were more than 200
unregistered applications/technologies in use by business areas to support their specific
business functions; b) business processes were not end-to-end within the university; c)
there was limited use of common technologies, e.g. no collaboration with other business
units; and d) user experiences were inconsistent, i.e. there was no single vision for all
students’ interactions with the university to deliver a consistent experience.
A member of the corporate group indicated that IT products and services, at the initial
stage of EA implementation, were defined by the respective functional area:
There certainly is politics between the service centres and the academics, so
[called] general and academic staff areas. Again, I think in the past, certainly
for the academics and the faculties and schools, [they] have probably been
a lot freer to do what they want to do, whereas [EA] … is going to again, in
their eyes, … essentially constrain them.
Modes of situational logic in the cultural system:
It is fair to conclude that at the institutional level most of the university’s business units
(functional areas) were culturally guided by past history in tailoring IT products and
services to suit their own interests. According to Archer (1995), this mode of situational
logic, or so-called specialization (see Figure 4.8), creates a situation that aligns with
people’s specific needs (Archer, 2015). In this study, specialization is viewed as the act of
a business unit making the IT environment suitable for their own purposes.
181
Figure 4.8: Situational Logics at different strata (adapted from Archer, 1995 p. 303)
(reproduced here for ease of reference)
Situational mechanism in the cultural system:
This study views situational mechanisms in the cultural system as an instrument of
cultural environments (the generative power of cultural emergent properties, see
Chapter 4) in shaping people’s ideas, thoughts, beliefs and values that influence their
work.
Specialization can be seen as the situational mechanism that conditioned the cultural
environment at Time 1 of EA implementation (see Table 4.3). The situational logic of
specialization represents a causal power of culture in the social world, and embodies the
underlying mechanisms that influence people (agency) by constraining and enabling
certain agential habits and routines. As a situational mechanism, specialization describes
the cultural phenomena shaping people’s reflexive deliberation in forming ideas about
the IT environments suitable for their own purposes. To the members of a business unit,
specialization is a natural way of getting something done within a particular cultural
system, since to them, specialization is an instrument or established way of practicing
their habits and routine. Seemingly, specialization was the legacy mechanism within the
functional areas at UX prior to the EA program. It generated a specific socio-cultural
interaction mechanism over the initial period of the Time 2 to Time 3 of EA
implementation, since the situational mechanism is the fundamental reason for the
socio-cultural shaping of individuals’ desires and beliefs.
182
Time 1: Modes of conditioning in the structural system (SS):
Macro (structural) - Micro (people) context of structural system
Analytical history of the structural system:
The structural system at the initial stages of EA implementation (see Figure 6.3) will
predispose agents’ (people) to using IT environments to promote their important
personal projects.
An analytical history shows that the past tendency of the university’s IT environment
from one to many (faculties, schools, and service centres), from simple to complex IT
activities (activities based on competence and capability of the units), and from
homogeneous to heterogeneous qualifications and skills, had changed the business units’
arrangements for running IT environments from being relatively generalised to being
functionalised or business-unit owned (governed by the needs of the individual business
units).
According to UX official documents, this has lead to more than 20 different database
technologies in use across the wider business areas at the commencement of EA
implementation, with numerous disparate tools and technologies to support backup,
portal and collaboration services, media streaming, email, telephony, networking,
monitoring etc., thus making the IT environment complex, inflexible and difficult to
manage. Investment in technology tended to be siloed. The structural conditions at Time
1 were not only complex and costly to maintain, but also limited UX’s agility in responding
to trends in the higher education sector to address the broader business directions of the
university. Most of the IT services were not easy to integrate and automate across system
boundaries, resulting in inefficiencies. Differentiation of business unit activities had
manifested as: a) fragmented data with no “single view”; b) some sharing of
infrastructure but no shared resources; and c) inaccessibility from all devices due to
different technology platforms between business units.
A corporate group member indicated the structural conditioning at UX actually impacted
the way EA was implemented:
183
…I think a university is different to other organisations, because each
school’s driven by the industry that they’re teaching future employees for,
so they each have their different type of technologies they’re using. It could
be absolutely, totally different and the cost of maintaining and managing
that centrally is really not viable I don’t think, or sustainable, so really what
we do is we look at something and say “will this have an impact on the
enterprise, the University as a whole, or is this a need for the University
across the board? […] So we are taking a bit of a layered approach, but in
saying that, there’s probably not a lot that shouldn’t be looked at from an
enterprise perspective, because the majority of stuff will come onto our
network and I’ve got to manage the network to support that.
Modes of situational logic in the structural system:
It is safe to conclude that at the institutional level a state of differentiation applied to IT
arrangements (standard, application, investments, etc.) that provided exclusively for
specific business unit activities. This situational logic of the structural systems during the
initial stages of EA implementation, or differentiation (Archer, 1995) (see Figure 4.8),
creates diversity through unity (Archer, 2013a). This study views differentiation as the act
of creating divergent IT environments for the sake of convergence in IT self-organisation.
Figure 4.8: Situational Logics at different strata (adapted from Archer, 1995 p. 303)
(reproduced here for ease of reference)
184
Situational mechanism in the structural system:
A situational mechanism at the structural level can be seen as the generative power of
systems integration (SEPs) (see Chapter 4) in shaping people’s situations in their social
systems, institutional structures, roles and positions. Differentiation was the situational
mechanism (see Table 4.3) that conditioned the structure at Time 1 and initial EA
implementation, since it carries the underlying mechanism that structurally influences
people (agency) synchronically by constraining and enabling agential powers.
Differentiation can be viewed as an established process by which members of a business
unit manage IT arrangements independently within a particular structural system. It was
an instrument and a natural way of deriving IT arrangements at Time 1. As a causal power,
differentiation describes the structural environment shaping agents’ (people) reflexive
deliberations in making IT arrangements for their exclusive purposes.
The situational mechanism of the structure at Time 1 generated a specific social
interaction from Time 2 to Time 3 in view of the fact that this mechanism causally
impacted people’s actions and interactions with regard to the EA program.
Time 2 to Time 3: Modes of interaction at Socio-cultural (S-C):
Micro (people) - Micro (people) context of socio-cultural
Analytical history of socio-cultural:
Time 2 to Time 3 is the EA implementation interaction period (see Figure 6.3) when the
interaction between social and cultural elements both shaped and were shaped by
groups of individuals within the UX population. A population is part of a stratified model
of people. In terms of MA (Archer, 1995), a stratified model of people (see Chapter 4) or
so-called People Emergence Properties (PEPs) consists of: a) population (class, status, and
power); b) organised groups (corporate agency); c) individual actors; and d) collectivities
(primary agency).
The analytical history shows that during Time 1 to Time 2, the cultural system shaped
people to use IT suitable for their own purposes. Specialization was the primary
mechanism for managing the IT culture of UX business units. Culturally, this mechanism
created a gap between expectation and reality, caused by poor governance and decision
185
making due to locally organised IT governance. During Time 2 to Time 3 corporate agents
looked for a solution to address the lack of architectures initiatives and governance, and
the PEPs were influenced by corporate agents’ actions and concerns about the interests
of particular collectivities (primary agents) and individual actors (the EA end-users i.e.
lecturers, students, administration, researchers). For corporate agents, understanding
the business units and end-users, whether they were students, academics or others, was
paramount in providing a superior, targeted architecture and roadmap for guiding
decisions.
Corporate group members indicated the main target of EA was involvement of UX
stakeholders, gaining their support, and improving the quality of the targeted
architecture:
Corporate agent A:
In the original program we tried to get involvement from right across,
representation – not about the technology but about how they wanted to
learn or teach or administer - whatever the role was.
…I don’t think that there’s any difference between those different
stakeholder groups. It should all be… and I hope our enterprise architecture
reflects that, we should all be focusing on achieving the same goal.
If we don’t have governance to get to the endpoint, people will still be
making decisions in silos and you’ll never get to the endpoint, because part
of the EA is to have that roadmap to… well, you’ve got that “this is what we
want to look like and these are things we’ve got to do to get there. If we
don’t totally focus on those things and if we have other people making
decisions that actually could undermine that, then we’ll just be going round
in circles.
Corporate agent B:
If you called the biggest stakeholders [the Enterprise Business Groups], so
that’s the heads of the service centres, the heads of the faculties, schools
186
and so on, I think because we’ve talked about [EA program] quite a lot in the
various meetings that we’ve had, I think they do have an understanding of
[EA program] somewhat. Whether they think it’s positive or negative…
I think by the involvement of people outside of IT in enterprise architecture,
it encourages the [EA] adoption over time.
Corporate agent C:
…it means that we can focus on the things that matter to the University, so
[focus on] the stakeholder management relationship, management of the
quality control, the prioritisation of activities... would be the best of both
worlds [the EA program and the stakeholders].
Consistent with these arguments, the primary agent also recognised the importance of
stakeholder involvement in the EA program:
I’d be involved in workshops where they’re gathering information but I’m
not involved in any implementation or development or framing of the
strategy or implementation at all. I’m just mainly providing input.”
…so people who run these systems [EA] are getting the buy-in and they’re
being listened to. It’s a consultant-led program…
Some individual actors felt their involvement in the EA communication program would
give considerable merit to their personal projects, while others felt the EA program would
not directly impact the way they were doing things:
Individual actor A:
They keep [updating]... yes, and they are very good in terms of informing if
one of the systems I’m using, like [X application, removed for reasons of
anonymity] or whatever, [has] specific data-related.
Look, people try to be helpful, however their role is limited because of the
compatibility, so I found them helpful and they generally make my life easier.
Don’t forget twelve or twenty years ago we had to type in a lot of things into
187
student databases and now everything is a click on the computer or the
keyboard. So it does, because it makes our lives more straight forward. I
don’t have to liaise with so much paperwork because most of the things are
done on computers. We generate invoices in the computer now, we don’t
have to ask in the written form, ask for invoices to be generated. I just simply
go on Apps and I can generate an invoice to private or to somebody else
inside UX – it goes, and it’s also escalated to the relevant people.
…if the IT doesn’t work I can’t perform my job. It’s simple. Because our job
is now we’re relying on technology…
Individual actor B:
…I was involved in those discussions [EA forum] as far as to see what we
need in order to improve the IT services here, as I said, because we do not
rely fully on the IT service but we use it to improve the results of our
research.
…we are running smoothly and I believe that we have no problem with the
IT. That means the system has been implemented properly.
Individual actor C:
Definitely – it [EA] will make it better. Of course it will make it better. Any
new initiative, especially from the feedback of different stakeholders at UX,
it will make it a better infrastructure and a better service that will benefit
everybody.
Modes of situational logic at socio-cultural:
The actions of corporate agents to obtain appropriate political backing focuses on the
rights, roles, and equitable treatment of shareholders in the EA program and is referred
to as the situational logic of sectionalism (Archer, 1995) (see Figure 4.8). In the English
dictionary, sectionalism is defined as “undue concern with local interests” (Oxford
Dictionaries, 2015a). In the case of UX, the lack of appropriate stakeholder involvement
in the EA program in the initial stages of EA implementation was the underlying cause of
188
sectionalism, and explains their observed lack of collaboration. Archer (1995) claims
corporate groups necessarily promote sectionalism in the cultural domain – an example
being when they seek to legitimise their advancement of EA implementation by
espousing newly elaborated ideas.
Figure 4.8: Situational Logics at different strata (adapted from Archer, 1995 p. 303)
(reproduced here for ease of reference)
The Enterprise Business Groups (EBGs) (see Text Box 6.1) were formed by UX’s EA
management (UX executive) in response to the observed sectionalism within the various
business centres – it was a necessary response to encourage EA support. These groups
underpinned the technology governance framework and reported to the university’s
Architecture Board, which in turn assessed the impact of EA and made decisions about
the governance approach. The EBGs were responsible for quantifying, reviewing and
prioritising any proposed technology initiatives prior to submission to the IT Governance
Committee. They provided a forum for learning about ongoing and planned technology
programs across UX, and provided an opportunity for the university’s leadership to
discuss pertinent IT operational reports.
Interaction mechanism at socio-cultural:
At this time (Time 2 to Time 3) causal influences exerted by the cultural system (Time 1)
were expected to impact the socio-cultural interaction. The interaction mechanism
describes the causal relationships between the corporate groups and UX stakeholders.
Sectionalism can be seen as the interaction mechanism (see Table 4.3) at a socio-cultural
level in relation to IT-business transformation at UX.
189
The recognition of sectionalism served as a useful way to explain elements of the
observed EA resistance. The response to sectionalism, i.e. creation of the EBGs, was
important for the program’s ultimate success. By involving the EBGs resistance from
stakeholders can be identified early, and their input sought to shape the architecture thus
garnering their support and improving the quality of the program. This mechanism
encourages university-wide participation rather than focusing on local objectives.
Recognising sectionalism and responding appropriately with the EA implementation have
the capacity to limit resistance and create a culture of understanding across the
university.
Long termer responding to this mechanism in the cultural domain is critical for achieving
measurable and sustainable outcomes from the IT-business transformation at UX as
stakeholders culturally must move towards a commitment to global goals rather than
local optimization. Stakeholders will be expected to contribute actively and seriously by
quantifying their IT demands relevant to global requirements in submissions to the
University Architecture Board (UAB). This takes the form of proposals which are discussed
at the UAB and prioritised for consideration at the IT Governance Committee (ITGC) so
that ITGC was able to make informed decisions with regard to IT requirements at UX.
Time 2 to Time 3: Modes of interaction at Social Interaction (SI):
Micro (people) - Micro (people) context of social interaction
Analytical history at social interaction:
The structural history prior to Time 1 revealed that, over the years, IT use within UX was
not easily integrated and automated across systems boundaries, resulting in business
inefficiencies. The different systems had different entry points, user interfaces and, in
some cases, separate logins, leading to inconsistent and fragmented user experiences.
Users were not able to access the full range of systems and services on their devices of
choice in a consistent manner. Each system had its own data stores and data models,
making a “single view” of key data entities difficult, particularly for students and their
interactions with UX. Differentiation, principally of business unit activity, was the
situational mechanism that preserved the structural status quo at Time 1 of EA
190
implementation. It was also apparent at Time 1 that the members of the business units
were not ready to change direction and embrace the EA program.
Subsequently, at Time 2 to Time 3 (see Figure 6.3), the PEPs were influenced by the
corporate agents’ endorsement of the EA program to enable UX’s decisions on
technology investment for the best outcomes.
The EA program supported initiatives for different UX stakeholders (ranging from primary
agents to individual actors) through various EA events (i.e. EA roadshows for the
individual actors, and specific forums, such as cross-organisational business groups for
the primary agents – the Enterprise Business Groups), where they could discuss and
clarify IT requirements. IT demand versus IT supply became a transparent process,
actively contributing towards UX’s strategic goals. The outcomes were published on UX’s
website where they were visible to all stakeholders.
Some corporate agents suggested aligning the processes, committee meetings and
controls with university-wide strategic planning and the annual planning cycle as a key
factor for successful EA implementation.
Corporate agent A:
…there are some pieces of work that are happening in the enterprise
architecture area over the next eight months that will produce some
enterprise architecture artefacts, some strategies and roadmaps. Those
strategies and roadmaps will be presented to the [Enterprise Business
Groups] which are the groups that are providing the governance around the
technology decisions within the University. So I guess in terms of the
governance arrangements it will be that the [Enterprise Business Groups]
and the business areas will have visibility of those enterprise architecture
artefacts. The idea then would be that they would hopefully submit
initiatives that would be aligned to those architecture strategies and
roadmaps. They haven’t been put in place yet because it’s too early but they
are planned to be implemented. Some of the things we’ll do will be to
publish them up on the UX website so they’ll be visible to all staff, not just
191
the [Enterprise Business Groups]. I believe there will be some kind of
roadshow that’s going to take place in the University, so I guess there’s going
to be demonstrations or information sessions.
Corporate agent B:
…the concept of enterprise architecture at the University is extremely new. […] It
involved a number of stakeholder engagement activities […]
The governance process is around providing visibility of the work that’s being done,
making sure that the right people with the right skills are making the right decisions
at the right time.
Primary agents viewed endorsement initiatives for the EA program to different UX
stakeholders as critical:
Primary Agent A:
Certainly I think it’ll hit the ground for us once [my Department] roadmap
up, so then we’ll have a plan for the next five years. And I think that’s a great
part of the strategy – that they’ve looked at the organisation. And I don’t
think everyone’s getting a roadmap, but I think they picked five or eight key
areas and we’re one, so we’ll have our own roadmap of where to go in the
future. And I think that process to get us to there is very well thought
through.
Primary Agent B:
There has been a significant amount of engagement because we recognise
that we need to be a part of the process, we need to have a strong voice in
order for the projects from [my Department] perspective to get put onto
that large agenda. …– yes, we have to be engaged
However, the EA events in the university’s functional areas in particular, were not as
effective as expected. It appears some (primary agents) misconstrued the purpose of the
cross-organisational business forums (the Enterprise Business Groups) and viewed them
192
as a potential source of funding for what they wanted to do. Some thought EA was a
waste of time and not profitable for their business areas, and subsequently became
passive participants. To overcome the ineffectiveness of the EA forums due to the
misconception and lack of engagement, the corporate agents employed a different
model of engagement to get people’s buy-in and acceptance. This model engaged with
people in each of the university’s functional areas through diversification. In the English
dictionary, diversification is defined as the act or process of reducing risk or volatility by
investing in a variety of assets (Oxford Dictionaries, 2015b). One influential corporate
agent articulated the lack of engagement in EA forums as follows:
We had the [Enterprise Business Groups] that fed into the [University
Architecture Board and] IT Governance Committee, and what we found was
that people I guess misconstrued the purpose of the [Enterprise Business
Groups] and really thought it was about getting some funding to do what
they wanted to do. And when they didn’t get the funding to do what they
wanted to do there was a lot of “we think they’re [EA program] not working”,
“they’re [EA program] a waste of time” and all that sort of thing, and a lack
of engagement by the members of those groups. So instead of coming along
and talking about “this is what we’re trying to achieve in our business area
or our school” or whatever it might be, they’d just basically be passive
participants. So we recognised that that wasn’t going to work, and we found
that where things really work in the University is when you have
conversations with people. We need to get people’s buy-in and acceptance
of this and to get that’s not through formal structures. The University’s very
good at putting in formal structures, the committees and things like that, but
how effective they are I’m not quite sure. They certainly weren’t very
effective from an enterprise architecture perspective, so now we’ve actually
turned that totally around and we talk now about our engagement model.
So we’ve now got a new model whereby we engage with individuals. Of
course this has been made much easier now since we’ve only got [xxx,
number of] schools to do it with, with each of those [xxx, number of] schools
and with each of the service centres. So each of them had basically two
193
account managers, for want of another word. One that understands the
strategic stuff, the enterprise architecture and all of that sort of thing, and
the other one that has more an operational focus, like what are the problems
you’re experiencing right now? So we will meet on a regular basis with each
of those schools and service centres around the University and that’s [one
of] the model we’re going forward with now.
Consistent with this argument, a primary agent also highlighted similar issues with EA
forum engagement:
Whether or not they actually understand the [EA] content – I don’t think
they [the corporate groups – corporate agents] do– because simply from the
point of view… you know, they [the corporate groups – corporate agents]
say it simply takes forever and a day. They [the corporate groups – corporate
agents] actually don’t understand the rationale. They [the corporate groups
– corporate agents] are trying with stakeholder [the primary agents]
engagement but I think they [the corporate groups – corporate agents] need
to know a lot more about it, what components are sitting where. […] we’ve
got all these massively beautiful [enterprise architecture] diagrams in there
to be prepared by consultants at an outrageous cost, and that’s alienated
people. But I think that the cost [to develop the enterprise architecture] and
the way that it’s been done [enterprise architecture implementation] is
really, really…[unreasonable] particularly when all the other centres and
faculties are having to have some financial constraints just to deliver
effective teaching and learning and research and support those activities.
And I think that’s really impeded stakeholder [primary agents] engagement
and coloured their understanding for it. And I think the thing is that they [the
corporate groups – corporate agents] are in one way better off, they’re [the
corporate groups – corporate agents] accepting of the concepts in there
which has been better, but worse in the sense that for all this we’re [primary
agents] not seeing anything. It’s a hidden thing, the enterprise architecture.
194
Modes of situational logic at social interaction:
In terms of MA (Archer, 1995), the actions of corporate agents in engaging with varied
stakeholder interests is the situational logic of diversification (see Figure 4.8). This study
viewed diversification as a process of identifying those who would gain and those who
would lose from EA implementation, and developing a strategy for dealing with them.
Diversification describes the social situation at social interaction level: a) between the
corporate agents and the collectivities; and b) between the corporate agents and
individuals.
Figure 4.8: Situational Logics at different strata (adapted from Archer, 1995 p. 303)
(reproduced here for ease of reference)
As members of the Enterprise Business Groups (EBGs) and key stakeholders in the future
success of the university, primary agents were given the opportunity to discuss and
decide, within their EBG, what value their initiatives would bring to UX. Chairpersons of
the EBGs were also voting members of the IT Governance Committee, where the full
university-wide portfolio of work was discussed. With improved governance,
transparency in the capturing, prioritising, funding and managing of initiatives was
significantly enhanced.
A number of trusted IT personnel from each business area attended all the EBGs – two of
them were also regular attendees at IT Governance Committee meetings. The role of
these stakeholders was to ensure the corporate agents communicated relevant
information at the EBGs. In addition, these new forums improved the level of
transparency in quantifying and prioritising IT demand and supply at UX, and served as a
195
forum for sharing information between EBGs via reports. Put simply, each EBG was privy
to the broader EA portfolio of work at UX. This level of transparency offered opportunities
for leveraging synergies and for sharing information with all participants on how their
areas might be impacted.
Interaction mechanism at social interaction:
This study viewed social interaction as a social process whereby people act and react with
each other at different social levels in society. Diversification was the interaction
mechanism (see Table 4.3) at a social interaction level at the time of architecture
implementation at UX. Recognising and responding to diversification prompt corporate
agents to engage with primary agents and individual actors to ensure alignment of the
EA program with the university’s goals, whether these are to be adopted locally by
business units or globally by the university.
Responding to this mechanism can lead to identifying: a) who gains and who loses from
EA implementation; b) who will make the decisions; c) who controls resources; and d)
who has influence. As a mechanism diversification provides a solution for the diversity of
university business activities and its collegial management (see Chapter 2). The varied
perceptions of the technology environment and information systems on the part of
stakeholders will necessarily result in differences in understanding the EA program. The
value of recognising diversification therefore lies in identifying key stakeholders and their
power, influence and interests, and defining the key business requirements to be
addressed in the EA implementation. Responding to diversification in a positive manner
is vital for successful EA.
From a technical point of view, diversification explores the fitness of core applications for
purpose – to reuse where possible, avoiding customisation and reducing the system’s
complexity. From a business point of view, diversification focuses on business outcomes,
i.e. sustainability, flexibility and agility, given that the university’s business requires its
members to operate in more than one information domain.
196
Time 4: Structural and cultural elaboration of agents (Structure, Culture, and Agency):
Micro (people) - Macro (structure and culture) context of elaboration
Analytical history at structural and cultural elaboration of agents:
In Time 2 to Time 3, people’s (micro context) actions and interactions generated various
intended and unintended social outcomes at Time 4, which influenced structural and
cultural change (macro context) as follows: a) people’s actions and social interactions (SI)
elaborated upon the composition of social structure (SS) by modifying internal and
necessary structural relationships and introducing new modes of situational logic; and b)
the cultural system (CS) was elaborated due to people’s actions and interactions at a
socio-cultural level (S-C), which modified current logical relationships and introduced
new modes of situational logic (Archer, 1995).
The MA helps to explain that the university environment (the state of its structure and
culture) was modified by the corporate agents during Time 2 to Time 3 in the interest of
EA implementation. From a structural point of view (social interaction or SI), the
governance process had been modified to a new integrated structure (a rationally
balanced structure: diversification mechanism) that emphasised accountability and
identified all the processes and controls required to align technology solutions with UX
business strategies. This new state of rational inquiry and sound evaluation of all IT
activity decisions supported university performance, financial objectives and risk
management. From a cultural point of view (socio-cultural or S-C) the existing sociocultural practices (habitual actions, established routines, traditional preserves, and
conventional divisions of activities) had been modified into new socio-cultural practices
(addressing the sectionalism mechanism) to enable the right people with the right skills
and experience to make technology decisions through the formation of committees and
delegations. This ensured that technology, as a strategic resource, delivers value to the
university over the long term.
A number of EA events which occurred during the period Time 2 to Time 3 resulted in
Time 4 being approached in a tactical opportunistic way; that is the tendency of taking
advantage of social events (and non-events). For instance, the IT projects of primary
agents and individual actors outside the control of corporate agents are influenced to
197
achieve EA opportunity (IT-business opportunity). The opportunism tendency generated
from SI and S-C is an expedient action, guided primarily by the EA events of architecture
principles. These opportunistic actions embodied the following principles for EA
implementation at UX: a) transparency: a clear EA governance structure and process.
How decisions were made in EA forums, who had input and who made decisions were
visible and transparent to the UX community; b) accountability: IT Governance
Committees and task forces (University Architecture Board and Enterprise Business
Groups) were held accountable for delivering on their responsibilities; c) responsibility:
EA governance structures and processes engaged people with the right skills and
experience to support the process at all levels; and d) appropriate representation:
faculties, schools and centres across UX were represented at appropriate levels.
Situational logic of aggregation:
The policies and practice of corporate agents in ensuring EA compliance with
stakeholders’ requirements to provide the best outcomes for the university is referred to
as opportunism (see Figure 4.8]. Opportunism is the situational logic of aggregation
generated in the past period (T2 – T3) between social interaction and the socio-cultural
level (Archer, 1995).
Figure 4.8: Situational Logics at different strata (adapted from Archer, 1995 p. 303)
(reproduced here for ease of reference)
At Time 4, corporate agents appeared to understand what was real and possible, and
seemed able to deal with problems in an effective and practical way through
198
opportunistic action. One corporate agent indicated how to deal with problems and
opportunity:
And the other thing that we do too, is we identify champions. We’ve just
finished a business case for collaboration [EA] capability, we’ve had it
endorsed, we’ve actually had it approved for funding, and we’re finding
champions around the University, [Mr X] is one of those. So I went to a
meeting last week and it was with the associate deans of Teaching and
Learning, and there was one particular associate dean who just wanted to
use it as a “I’m gonna have a big whinge about IT” and “what’s wrong with
IT?” and “how do we know that this solution that ITs come up with is right?”
and all this sort of thing. And [Mr X] basically took the mantle, I didn’t have
to do anything, [Mr X] took the mantle and said “we’ve all had a look at it. It
might not be perfect but it’s a lot better than what we’ve got now”. But what
it means is he sees the biggest part of this won’t be technology that we’re
going to use, it’ll be the cultural change, and the way we have to change the
way we do things. And he said “that’s our problem, not IT’s problem”. So
getting those sorts of champions; getting the support of the Deans and then
getting the champions.
Opportunism increases cultural fairness and is free from socio-cultural manipulation, as
the opportunistic tends to take up a compatible practice (Archer, 1996). Existing sociocultural practices, such as habitual preoccupations and established routines, may well
diminish and be replaced by an increased willingness to explore new architecture and
congruent possibilities. Stakeholders, particularly the primary agents and individual
actors, have substantial freedom to adopt the EA program which has come into view.
Through the accommodation of opportunism, EA can be designed to induce adoption in
accordance with stakeholders’ interests.
Primary agents in particular understand that EA programs needs to make more efficient
use of technology platforms and its services to drive down costs and improve the
reliability of core business services. A number of opportunities exist to leverage
technology platforms more widely and create reusable enterprise technology services
199
wherever possible. The situational logic of opportunism requires a basis for building
collaborative opportunities across faculties, schools and centres. EA is in line with the
evident opportunities that IT can provide. It could be argued that without EA benefits of
opportunism cannot be properly developed.
The university moves from competition/elimination towards opportunism as systemic
integration increases with the new Vice Chancellor in charge – opportunism is still the
basic situational logic as the university faces a common threat of new entrants and the
global trends in HE (i.e. Massive Open Online Courses: MOOCs) – the challenges facing
university require novel ideas and new alliances and UX has shown itself to be good at
that “when material opportunities arise that resonate with the social and cultural
properties, stimulating opportunism (Archer 1988).” Universities must be innovative in
their offerings in order to gain enrolments. As explained by Archer (1998, 243) “habitual
preoccupations, established routines, traditional preserves or conventional divisions of
subjects – may well reduce subjective willingness to explore new and congruent
possibilities, but these will usually coexist with various sticks and carrots which stimulate
originality, innovation and experimentation (as in the derived sequence). The actors
concerned have substantial freedom to survey or to ignore the broader horizon which
has come in to view…”. The sticks and carrots are enrolments - staff are rewarded if they
are good. The habitual and routine elements of course approvals requiring a 2-year time
line constrains this freedom - universities have to be careful to not have systemic controls
dominate freedom to act (the province of protection as a situational logic).
Transformation mechanism:
Opportunism as the situational logic at the time of transformation (Time 4) motivates
different forms of strategic action by predisposing different sections of the UX population
to supporting their own interests: the process of tailoring stakeholders and their power,
influence, and interests provides the opportunity to remove obstacles that are operating
elsewhere in the university. The EA support of opportunism allows EA to be accepted and
to provide transformation. It can be safely concluded that opportunism represents the
generative mechanism of morphogenesis and is the transformation mechanism at Time
4 (see Table 4.3). As it has a net systemic situation of great cultural variety, that
200
generates: a) a mutually beneficial synergism of opportunities at structural level; b) a
socio-cultural reintegration as harmony at cultural level; and c) a gradual shift of people
from relationship of EA rivalry towards to ones of EA supporter and then to synergy which
produced the EA co-operative relationship. (Archer, 1995; 1996; 2015).
The accommodation of opportunity with EA can be seen as a mechanism to engage the
involvement and support of all stakeholders with an interest in or responsibility for EA
implementation and the objective of ensuring that the university’s interests are served
and EA objectives achieved. The recognition of EA supporting opportunism served as a
useful way to identifies stakeholders’ needs with a clear sense of EA being in their
interests in and then having responsibility for EA implementation. Through this
mechanism UX strategic business directions and its architecture are aimed at: a) a culture
that encourages EA participation in university-wide rather than local objectives; b) an
integrated structure that manages EA activities across interest areas; c) socio-cultural
practices that foster meaningful, as opposed to symbolic, participation in EA
management processes; and d) commitment, through social interaction, to ongoing EA
implementation and challenges, and openness to other parties’ advice.
As a mechanism, opportunism also underpins EA initiatives by justifying the risks
associated with the architectural vision, and assesses the readiness of IT-business
transformation and the potential risks associated with it. It shows a clear line of sight
between technology decisions and business outcomes which provide UX with: a) a
unified, targeted IT architecture to engage business stakeholders and deliver stronger
business outcomes; b) a tangible link between the core strategies of the university and
enabling IT plans; and c) an objective method for assessing planned technology
investment.
The opportunism evident within UX needs to avoid: a) the act of a business unit making
the IT environment suitable solely for their own purposes (specialization: C.S. level) and;
b) the act of a business unit creating divergent IT environments for the sake of
convergence in IT self-organisation (differentiation: S.S. level).
201
Opportunism is supported via a communication strategy intended to achieve a successful
implementation of EA program (see Text Box 6.1). Such a strategy, namely through: a)
addressing sectionalism (S-C. level: to provide a forum for learning about ongoing and
planned technology programs across UX and, provided an opportunity for the university’s
leadership to discuss pertinent IT operational reports) and; b) addressing diversification
(S-I. level: to ensure the corporate agents communicated relevant information at the
forum – a level of transparency offered opportunities for leveraging synergies and for
sharing information with all participants on how their areas might be impacted). As
explained by Archer (1995, pp. 303-304):
For both culture and structure, the systemic level (presented on the top line
for each) [see Figure 4.8 below] shows the full range of developments which
can be generated if the respective situational logics are all successfully
followed (and each of the relevant contingencies materializes). As we have
seen, those Corporate Agents whose interests are vested in any one of these
four states of affairs, in either the cultural or social system, have a
corresponding ideal at the level of social or socio-cultural interaction which
is most conducive to securing the systemic status quo desired by them. To
this end various forms of structural and cultural power will be deployed by
them as containment strategies4 intended to preclude deviant social
developments; such preferred social states are presented on the bottom line
for both culture and structure. However, as always, these are conditional
effects (of the C.S. on the S-C and of the SS on SI) and their success is no
foregone conclusion. Everything in fact depends upon their social reception.
And that is determined by the relational negotiating strength between the
Corporate Agent promoting the systemic state in question and the array of
PEPs which have now disengaged in society, whose goals may be at variance
with those of the former.
202
Figure 4.8: Situational Logics at different strata (adapted from Archer, 1995 p. 303)
(reproduced here for ease of reference)
From these findings, it can be seen that: a) the opportunism evident within the university
needs to avoid the specialization and differentiation via an effective communication
strategy concerning meeting stakeholders needs, ranging from the initial EA initiative to
ongoing governance. Effective communication of targeted architecture to the right
stakeholders at the right time is critical for EA implementation; b) opportunism presents
a loose situational logic requiring socio-cultural practices to take advantage of it. Rather
than addressing challenges of cultural complexity, corporate agents led the university
community’s focus on issues such as sourcing and sharing business capability, expanding
business capability with the new architecture, and the university’s future vision.
Developing a collaboration plan to address the social complex situation presented an
opportunity to achieve a more flexible architecture solution.
The relationship between first- and second order emergent – the morphogenesis:
Based on Archer’s (2015) MA paradigm it can be argued that the emergent phenomena
of EA programs originate from the time between contextual conditioning of EA
implementation (downward causation) and people’s responses to the implementation
(upward causation), which are empirically distinct and temporally sequential. The
structural and cultural elaboration begins in the period of Time 2 to Time 3, and emerges
at Time 4, in a form based on the relations between the agents (people) in the interactive
network during Time 2 to Time 3. Mechanisms at Time 2 to Time 3 arise from a process
of morphogenesis in the second-order emergent properties, while mechanisms in Time
203
1 arise from previous morphogenetic cycles in the first-order emergent system (the
results of past socio-cultural and social interaction prior the EA implementation). Such
conditions determine the potential bargaining power of people in adopting the EA during
Time 2 to Time 3.
The second-order emergent system shows the structure and culture also experienced
elaboration and transformation into new structural and cultural practices. Figure 6.4
describes the relationship between first- and second-order emergent properties, the
identified generative mechanisms, and the MA outcomes.
204
Figure 6.4: The analytical histories of emergence: Relationship between first- and second order emergent properties, generative mechanisms,
situational logics, and MA outcomes (adapted from Archer, 1995)
205
Key implementation mechanisms:
Based on the research findings it is evident that several different mechanisms were
operating simultaneously (i.e. several different mechanisms at Time 2 to Time 3), and
that the relation between the generative mechanisms (i.e. mechanisms at Time 2 to Time
3 generated transformation mechanisms at Time 4) is contingent. These contingent
effects are what generated the key mechanisms or key implementation mechanisms.
Therefore it can be safely concluded that: a) effective communication and collaboration
are key implementation mechanisms for effective accommodation of the prevailing
transformation mechanism (opportunism) at Time 4; and b) opportunism as the
prevailing SL and transformation mechanism was fundamental in supporting the premise
of EA. As indicated by the corporate agents, a key focus of EA was communicating the
benefits to stakeholders. In addition, they viewed communication and collaboration as
the primary factor in widespread EA adoption:
Corporate agent A:
My understanding of the communication program is probably these [EA]
roadshows that apparently are going to happen at some point. I think there
is a real risk though, of it being presented at too high a level within the
University and people just, again, not understanding it and then the
perception is ‘oh, that’s something that IT does’, and not understanding that
they need to be involved in it, and if they’re not involved in it there’s going
to be a disconnect between what potentially IT people or consultants think
is the enterprise architecture and what the business actually needs. You’re
not going to have that joined. So it’s incredibly important that there’s a
communication program in place and that the stakeholders here accept
enterprise architecture and that they understand the importance of it”;
…you’re looking at strategies to get to an endpoint, so you’re saying where
is the business now? Where does the business want to be? And what do we
need to do to get there? But what do we need to get there are potentially
all projects or pieces of work, programs of work that have to be undertaken
206
to get there. And the technical side of things, well they’re delivering those
projects or they’re feeding into those projects, so if we don’t have a
collaboration with them then it just ends up with this siloed environment
which is what we’re trying to come from.
Corporate agent B:
I think the main change from EA will be about governance and investment in
technology - who makes the decisions? … [The purpose of the EA program]
is really educating people about the value of EA and getting them to
understand it and then communicating and keeping the linkages between
the EA and the business of the University. […] …enterprise architecture can
become a very theoretical thing that’s done, and if you don’t get the
communication right and if you don’t present it in the right way to an
organisation it becomes a heap of artefacts that gets stored away here in a
repository and nobody ever looks at again. So it’s about continually giving
the messages and communicating that and creating those linkages”;
We’ve just finished a business case for collaboration capability, we’ve had it
endorsed, [and] we’ve actually had it approved for funding… [For instance],
the Deans of Schools sit at the University executive... we still have the IT
Governance Committee but anything that comes out of the IT Governance
Committee gets endorsed. When it gets endorsed by the IT Governance
Committee it then goes in to University Executive for approval, and that has
a lot more power because now the Deans of Schools are actually part of
making decisions about what technology we invest in, which is what it was
always meant to be, but they weren’t taking that opportunity to do that.
They weren’t feeding in from the bottom up, so now that they’re sitting up
here they can take that accountability from up here. They are the ones that
then need to deal with pushing it down through their schools.
In review, identification of the key implementation mechanisms and social responses
triggered by EA implementation were a central goal of applying the morphogenetic approach
207
in this study. The findings revealed that each of the key implementation mechanisms defined
theoretical explanation (abstraction) of the generative mechanisms (from Time 1 to Time 4)
representing the structures and cultures of University X during EA implementation, each
aimed at a particular context of stakeholders (people) and addressing a particular set of
concerns. In this study this is called social responses.
The social responses triggered by EA implementation were dependent on the interplay
between social integration (PEPs) and the structural and cultural (SEPs and CEPs) integration
(see Chapter 5). In terms of this study people’s resistance to change (social integration) was
low and structural and cultural integration was high, which led the organisation to social
transformation by supporting the mechanisms identified in encouraging individuals’
acceptance of EA practices. The social responses triggered by EA implementation at UX is
known as contingent complementarity (Archer, 1995) (see Figure 4.7), which required
cultural variety within the EA program and meant that the architecture needed to be
structured to accommodate the business units’ cultural differences.
Figure 4.7: Morphostatis vs morphogenesis: Situational logics in social and system
integration in EA implementation (adapted from Archer, 1995, pp. 218 and 295)
(reproduced here for ease of reference)
208
6.3.2 Abduction stage
The stage referred to as abduction or theoretical redescription or recontextualisation is where
original ideas (a conceptual framework or theory) about the object of study are placed in a
new context (Danermark et al., 2002). In other words, abduction seeks connections or
relations, not directly observable, between the EA events (in the empirical domain) and the
EA context identified (mechanisms and situational logics in the real domain – MA). This stage
gives new meaning to already known EA events.
Since the mechanisms at Time 1 arose as a result of past interactions (a time prior to EA
implementation – the first-order emergent), only the mechanisms arising from Time 2 to Time
3 and Time 4 (the second-order emergent) were used in the abduction analysis to formulate
new ideas about the interconnection of EA events and contexts.
Data analysis and findings:
1.
Relation between particular EA event and EA context
Table 6.1 below shows the relation between particular EA events and EA contexts
(mechanisms and situational logics within cultures and structures) by redescribing or
recontextualising EA events from a set of ideas (in the form of a conceptual framework
or a theory).
Table 6.1 Relation between particular EA event and EA context
EA event
EA context
T – T , S-C (sectionalism mechanism):
Changes in habitual action by means of
encouraging participation towards
university-wide rather than local objectives.
Creating social circumstance of synergy
management.
T2 – T3, SI (diversification mechanism):
Changes in social formation by means of
ensuring alignment of the EA program that
UX wishes to be performed, whether these
are to be performed locally to the units or
globally to the University. Creating social
circumstance of linkage management.
2
Events of the architecture compliance:
3
209
EA event
Events of the architecture principles
Events of the cross-organisational
architecture board
EA context
T4, Transformation (opportunism
mechanism):
Opportunity by means of supporting more
detailed definition of a consolidated, crossbusiness unit roadmap within the EA
program. Creating social cohesion of
strategic alliance.
T2 – T3, S-C (sectionalism mechanism):
Changes in established routines by means
of focusing on the requirements of users
and stakeholders. Creating social
circumstance of diversification
management.
T2 – T3, SI (diversification mechanism):
Changes in social composition by means of
engaging in varied stakeholders’ interest of
EA program. Creating social cohesion of
distributed leadership.
T4, Transformation (opportunism
mechanism):
Opportunity by means of removing
obstacles that are already carried out
elsewhere in the University, add EA
program specific tasks to align the key
business requirements. Creating social
circumstance of linkage management.
T2 – T3, S-C (sectionalism mechanism):
Changes in conventional divisions of
activities by means of placing the
stakeholders’ input/feedback to shape the
architecture and ensure their support to
improve the quality of the EA program.
Creating social circumstance of synergy
management.
T2 – T3, SI (diversification mechanism):
Changes in social formation by means of
introducing a good EA governance through
transparency in the way EA initiatives are
captured, prioritised, funded and managed
to completion. Creating social cohesion of
strategic alliance.
T4, Transformation (opportunism
mechanism):
Opportunity by means of focusing on EA
issues such as sharing of business
capability, sourcing of business capability,
210
EA event
Events of the EA roadshows and workshops
for the University society
2.
EA context
and exposure of business capabilities to
new architecture and university future
theme. Creating social cohesion of
distributed leadership.
T2 – T3, S-C (sectionalism mechanism):
Changes in established routines by means
of building culture of understanding across
UX and provide useful insights for the EA
Implementation. Creating social cohesion
of strategic alliance.
T2 – T3, SI (diversification mechanism):
Changes in social composition by means of
diversifying to identify those that will gain
and those that will lose from EA program,
and then develop a strategy for dealing
with them. Creating social circumstance of
diversification management.
T4, Transformation (opportunism
mechanism):
Opportunity by means underpinning any EA
initiatives in identifying the risks associated
with the architecture vision and assess the
readiness level of IT-Business
transformation and the potential risk
associated with it. Creating social
circumstance of linkage management.
Placing a new context of EA within key ‘implementation’ mechanism
In this analysis of theoretical redescription or recontextualisation, a new concept of ideas
(EA context) has two meanings depending upon the social situations:
a) Social situations that are influenced by activities associated with EA implementation.
This can be regarded as a social circumstance or event that happens whether we
experience them or not. It can be seen that social circumstances or events not
directly observable within EA implementation have been generated by different
mechanisms.
b) Social situations that are influenced by the willingness of UX members to cooperate
with one another in order to realise the EA implementation. They can be regarded
as a social cohesion or non-events that happen whether we experience them or not.
211
Non-events not directly observable within the EA implementation have also been
generated by different mechanisms.
Building on Danermark et al.’s (2002) methodological model, it can be argued that
theoretical redescription or recontextualisation provides deeper knowledge of particular
EA events. In this study, the social situations (social circumstance and cohesion) that
influenced the university community as events and non-events located in the actual
domain (see Figure 5.11) were identified to detect relations between the new contexts
of EA (Table 6.1) and the key implementation mechanisms (identified at analytical
resolution stage – MA analysis) as follows:
Figure 5.11: ‘Abduction’ process (reproduced here for ease of reference)
a) Distributed leadership (non-event): Successful EA implementation at UX depended
on leadership that was both designated (i.e., someone was formally in charge of the
program) and distributed (i.e., professionals, partner organisations and teams shared
responsibility for mobilising efforts and delivering the program components) (Best et
al., 2102). Distributed leadership engages individuals at all levels in leading the ITBusiness transformation (change) to keep everyone (primary agents and individual
actors) on board and focused on the program. EA endorsement was sponsored by
corporate agents (executives) who were appropriately aligned to provide leadership
and able to articulate and defend the needs of the endorsement at the executive
212
level. These executive sponsors were engaged through communication
mechanisms.
b) Synergy management (event): Where business objectives of the units (the
university’s functional areas) deviate from the EA program, significant effort is
required to avoid wide-ranging impact. This activity, flagged as the right set of
circumstances, was addressed by the collaboration mechanism.
c)
Linkage management (event): The process of tailoring stakeholders and their power,
influence, and interests provided an opportunity to remove obstacles elsewhere in
the university, thereby adding to EA program-specific tasks of aligning the key
business requirements, and was addressed by the collaboration mechanism.
d) Strategic alliance (non-event): Social situations supported a more detailed definition
of the consolidated roadmap (the university’s functional areas) within the EA
program, was addressed by the collaboration mechanism. Through the
communication mechanism, widespread communication and understanding of EA
by all UX stakeholders was enabled, and reassured them the EA program would
address their concerns.
e) Diversification management (event): In making decisions about diversification of
the business-unit roadmap, technology environment, and socio-cultural practices,
communication acted as the mechanism to derive a series of transitional changes for
delivering continuous business value to realise the EA implementation.
6.3.3 Retroduction stage
Based on the ideas of Danermark et al. (2002), retroduction describes a mode of inference by
which this study arrived at what is basically characteristic and constitutive of people adopting
the EA (see Chapters 4 and 5), since MA still posed the central problem of how social
structure, culture and agency are actually linked (their social relationship: Archer, 2008). In
other words, this stage clarified the basic conditions for a social relationship: people’s actions
and interactions; people-focused mechanisms; reasoning and knowledge within EA
implementation. To clarify these basic conditions which are not directly observable within the
EA implementation, the retroduction stage went beyond the empirical (something that can
213
be observed and experienced). It is also called transfactual argumentation (Danermark et al.,
2002).
This stage used reflexivity theory (Archer, 2008; 2010a) (see Chapters 4 and 5) to clarify the
basic conditions outlined above and attain knowledge about people’s internal relations that
would make EA implementation at UX succeed. Exactly how people’s internal relations are
viewed was made possible by reflexivity theory of internal conversations and use of the
reflexivity tool, the ICONI (see Chapter 5), to identify how people subjectively defined
particular courses of action (personal concerns) in relation to their objective social
circumstances (context of EA). These internal conversations explained how people responded
to the embedded mechanisms (interaction mechanisms) of EA implementation. The question
can also be answered by reconstructing people’s reflexive deliberations (their internal
conversations) in regard to adoption of EA, as these reflexive deliberations form an important
mechanism for explaining how structures and cultures constrain and enable agents to adopt
the program (reflexive mediation of mechanisms).
As mediatory mechanisms (see Chapter 4), reflexivity theory described people’s adoption of
EA (their actions and interactions), shaped by the interplay between the context (mode of
people reflexivity in social relationships at the time of EA implementation – see Chapter 5)
and concern (people’s personal reflexivity concerns at the time of EA implementation – see
Chapter 5). Since individual responses are diverse because of their different modes and
degrees of concern, their collective reflexivity (social relationships) was also described.
Collective reflexivity is not about people thinking in the same way, or people sharing external
commitments, or people having a mutual intention; but rather about people being in a special
relationship which makes them reflexive in a social instead of a personal way (Donati, 2008,
cited in Archer, 2010, p. 11) (see Chapter 4).
Data analysis and findings:
1.
The Internal Conversation Indicator (ICONI)
Tables 6.2 and 6.3 provide the ICONI data of people’s interplay between context and
concern at the time of EA implementation. ICONI was used to identify dominant modes
of reflexivity.
214
In tables 6.2 and 6.3 below, EA management terms refer to people who affect (corporate
agents) and are affected (primary agents) by EA implementation, and whose institutional
structures and positions range from IT Governance Committee, University Architecture
Board, and Enterprise Business Groups to Lead Enterprise Architect. End-users (individual
actors) are people who can be affected by EA implementation and comprised teaching
staff, researchers, students, and administrative staff of the faculties, schools and service
centres.
Table 6.2 EA management reflexivity
Dominant modes of
Stratified model
Reflexivity concern
reflexivity
Subject’s ultimate
Subject’s dominant
(Context)
concern
concern
Corporate agent
Autonomous Reflexive
Communicative concern
Autonomous concern
Corporate agent
Autonomous Reflexive
Communicative concern
Autonomous concern
Corporate agent
Autonomous Reflexive
Communicative concern
No dominant concern
Primary agent
Autonomous Reflexive
Communicative concern
Fractured concern
Primary agent
Autonomous Reflexive
Communicative concern
No dominant concern
Fractured concern
No dominant concern
of people
and Meta-Reflexive
Primary agent
Autonomous Reflexive
Table 6.2 shows the dominant mode of reflexivity of people within EA management as
autonomous reflexive, which means they have their internal deliberations alone and act
upon them. In other words, these people had the freedom to govern or control their own
affairs with regard to the EA program, which reassured them that the program would
address their concerns. As a result, the autonomous reflexives counted on EA
implementation.
Interestingly, one primary agent possessed dual dominant modes of reflexivity
(autonomous reflexive and meta-reflexive), yet her internal relation in regard to her
position (place, functions, rules, duties, and rights) (see Chapter 4) and career inclined
her towards autonomous reflexive, indicating a reliance on the program. A meta-reflexive
mode refers to a tendency in people to seek a personal vocation rather than a career,
compelled by the need to make a difference, rather than simply meeting their immediate
needs and desires.
215
In Reflexivity Concern, people’s “ultimate concerns” are mostly related to communicative
concerns. This means their ultimate concerns are about family responsibility and family
wellbeing, and are likely to secure positions and careers rather than seeking personal
vocations. This communicative concern is dependent on the success of EA
implementation for retaining their current positions at work and continuity of their
careers. Only one primary agent showed fractured concern, which means siding with the
majority to establish a better way of life. In this case, like most of his colleagues, the
fractured concern endorsed the EA program.
There were no dominant concerns amongst the EA management population. The
majority of corporate agents’ concerns were autonomous concerns, which means they
were more concerned about their work, career, performance achievements, and financial
success. Only one primary agent had a fractured concern. The non-dominant concern
condition is likely due to the influence of the EA program being stronger than their
personal projects.
Given that the dominant modes of reflexivity context within the EA management
population was autonomous reflexive, it can be concluded that the dominant mode of
collective reflexivity was collective autonomous reflexive, which means the collective of
people in this mode were realistically dealing with the EA program in a way that is
consistent with group well-being.
Table 6.3 EA end-users’ reflexivity
Dominant modes of
Stratified model
Reflexivity concern
reflexivity
Subject’s ultimate
Subject’s dominant
(Context)
concern
concern
of people
Individual actor
Meta-reflexive
Communicative concern
Communicative concern
Individual actor
Meta-reflexive
n/a for personal reason
n/a for personal reason
Individual actor
Meta-reflexive
Fractured concern
Autonomous concern
Individual actor
Autonomous reflexive
Communicative concern
No dominant concern
Individual actor
Meta-reflexive
Fractured concern
Fractured concern
Individual actor
Autonomous reflexive
Autonomous concern
No dominant concern
216
Table 6.3 shows the majority of end-users’ dominant modes of reflexivity as metareflexive. These people felt EA implementation was “somewhat important” as
implementation would not directly impact the way they did things. The other individual
actors – autonomous reflexives – viewed the program as “quite important” to their
personal projects. This does not mean they refused to adopt the EA program; they just
accepted it. The subjects’ ultimate concern, which acknowledged EA implementation,
shows that: a) communicative concern subjects gave credit to the program; b) like most
of their colleagues, the fractured concern subjects agreed with the EA programs; and c)
the autonomous concern subjects counted on the program.
Furthermore, the dominant concern for EA end-users was the same as for EA
management – non-dominant concern. However, in the case of EA end-users, the
condition of non-dominant concern was in all likelihood due to the fruitfulness of EA
engagement during implementation, indicative of a successful strategy for engaging with
them.
To summarise, since the majority of EA end-users’ reflexivity context was meta-reflexive,
it would appear that the dominant mode of collective reflexivity is collective metareflexive, which means this collective of people considered EA a relational enhancement
to the way they were doing things.
2.
Reflexive mediation of mechanisms: Placing reflexivity theory within people-focused
mechanism
According to Archer (2008), people do not act in uniform ways in the same circumstances,
and therefore there was no clear explanation of exactly how people responded to peoplefocused mechanisms, that is mechanisms that encouraged individual and collective
acceptance of EA practices, for instance, how people addressed dealing with these
mechanisms, what situations they encountered, what relationships they dealt with, etc.
Reflexivity theory provided an explanation of how personal projects were formed, and
how they mediated the structural and cultural constraints and enablements within EA
implementation. Hence, it could be argued that reflexivity theory can be used to describe
the people-focused mechanisms identified in the MA stage. In clarifying the basic
217
conditions for identifying people-focused mechanisms, reflexivity theory focused on the
micro-micro context of sociocultural (SC) and interactions and social interaction (SI)
during Time 2 to Time 3. Figure 6.5 below explains how reflexivity theory mediated the
basic conditions of people reflexivity and the social interaction and sociocultural
properties that shaped the context and concerns of the people-focused mechanisms.
Figure 6.5: Reflexive mediation of mechanisms
At the time of people’s interactions, when UX addressed governance of the EA, key
factors were the basic conditions of people’s reflexivity (Figure 6.5) and the level of detail
of people’s responses to sociocultural and social interaction. These factors assist in
identifying modes of reflexivity for social order or the reflexive mediation of mechanisms.
Reflexive mediation ensures that people have the necessary mechanisms in place to
enable the architecture transformation. The process for identifying reflexive mediation
was as follows:
a) The EA management:
Autonomous reflexive was the dominant mode of reflexivity for corporate agents and
primary agents. Corporate agents viewed EA as the foundation for executing
architecture transformation at the sociocultural level. Having the freedom to govern
218
or control their own affairs with regard to the program, the first step at this level was
to culturally identify the key players (primary agents) in EA engagement, during
which different primary agents (part of stakeholders) may have been uncovered as
the engagement progressed through the sectionalism mechanism. Corporate agents
viewed EA as a mandatory program to support future university strategies, and this
was the condition at the social interaction level. To ensure continuation of the
program at UX, corporate agents avoided concentrating too heavily on the formal
structure of the university functional areas as the basis of identification through
diversification mechanisms, since informal stakeholder groups may be just as
powerful and influential as formal ones. The primary agents acted according to their
autonomous reflexive mode to realistically deal with the EA program, which allowed
them some assurance that the program would address their concerns.
The qualitative feature of the corporate and primary agents’ social context is referred
to as contextual discontinuity (see Table 5.2), since their actions would transform the
cultural and social structure. This is predominantly a reflexive action. Upward
mobility was the social stratification of the corporate and primary agents (see Table
5.2), an agential objective to improve upon their social position, and in doing so, to
become upwardly socially mobile. Their relationship to structure and culture was a
strategic stance towards constraints and enablements although they had the
freedom to govern autonomously. Nevertheless, their ultimate personal concerns
made them act in accordance with their moral duty rather than their desires. The
corporate and primary agents’ ultimate concerns were communicative.
To review, the above discussion suggests the mode of reflexivity for the social order
was for people to devote themselves to the program and contribute to architecture
initiatives and implementation (see chapter 5). This was the reflexive mediation of
mechanisms for EA management (corporate and primary agents). According to
Archer (2008), this reflexive mediation of mechanisms leads to morphogenesis (see
Table 5.2). Sectionalism and diversification were the morphogenetic generative
mechanisms.
219
b) The EA End-users:
Meta-reflexive was the dominant mode of reflexivity for individual actors. These
individuals belonged to more than one stakeholder group (i.e. members of faculties,
schools or service centres) and tended to arise from specific EA events. As discussed
earlier, they viewed EA implementation as “somewhat important” because it did not
directly impact the way they were doing things. Some were interested in details of
the implementation while others didn’t care. Certain individuals thought the
program was too complex for a layperson to understand. For them, EA was not like
other systems where they had prior knowledge, and could agree or disagree or make
suggestions. Culturally, they were accepting of EA and would not interfere with the
program unless implementation conflicted with their interests. Structurally, they
considered EA a relational enhancement to the way they were doing things, and they
engaged with sectionalism and diversification mechanisms to ensure consistent
governance of their interest and concerns.
The qualitative feature of individual actors in the social context is referred to as
contextual incongruity (see Table 5.2). While this social situation may conflict with
the EA program, the continuous promise of the program’s benefits in enabling them
to realise their context, will cause them to acknowledge and trust the EA program.
As meta-reflexives they subscribe to one of three social stratifications (see Table 5.2)
depending on the social situation: a) social immobility; b) upward mobility; or c)
social volatility. However, due to the fruitfulness of EA engagement during
implementation, the social stratification of the individual actors was upward
mobility. Their relationship to structure and culture was subversive towards
constraints and enablements since they tended to seek personal vocations rather
than careers, compelled by a need to make a difference rather than simply meeting
their immediate needs and desires. However, their ultimate personal concerns made
them credit the program (communicative concern) and agree to embrace it like most
of their colleagues (fractured concern).
To summarise, the discussion above suggests the modes of reflexivity in the social
order was to foster values for social reorientation and develop the organisation
220
towards social transformation (see chapter 5). This was the reflexive mediation of
mechanisms for EA end-users (individual actors). According to Archer (2008), this
reflexive mediation of mechanisms leads to high morphogenesis and low
morphostatis (see Table 5.2), conducive for encouraging individuals’ acceptance of
EA practices.
6.3.4 ‘Comparison with different theories/abstractions’ stage
Danermark et al. (2002) credited this stage with elaborating and estimating the relative
explanatory powers of the mechanisms, structures and cultures described by means of
abduction and retroduction. In other words, this stage assesses the relative explanatory
powers of the mechanisms and structures in the problem investigated. According to
Danermark et al. (2002), one particular theory might describe the necessary conditions for
what is to be explained and therefore has greater explanatory power.
In terms of abduction, this study identified the connections and relations between the
observable EA events and the identified EA context (the mechanisms and situational logics),
and gave new meaning to already known EA events. The interconnections between the EA
events and contexts were the events and non-events (see Figure 5.11) located in the actual
domain (not directly observable) that generated the already known EA events (the observable
EA events). This study explains that those events and non-events were impacted by: a) the
social mechanisms that drive large-scale architectural transformation (found in the analysis
of MA – analytical resolution stage); and b) the theory or mechanisms built into the EA
program (found in the description stage, the empirical domain). However the role of theory
and program mechanisms in supporting the social mechanisms proposed by MA was still
unknown.
The retroduction stage of the study clarified the basic conditions for social relationships: a)
people’s actions and interactions; b) people-focused mechanisms; c) reasoning; and d)
knowledge that lead to identification of reflexive mediation of mechanisms. Retroduction was
instrumental in explaining people’s responses to the people-focused mechanisms
(mechanisms that encourage individual and collective acceptance of EA practices), and also
for clarifying the basic conditions for people-focused mechanisms. Nevertheless, the role of
221
the program mechanisms, referred to as program measures by Pawson and Tilley (1997), in
supporting the people-focused mechanisms underpinning the stakeholders’ interactions,
were still unknown.
In this study, the purpose of comparing different theories was to uncover the theory of the
program mechanisms in order to determine the conditions under which they might work or
not, i.e. “what works”, “for whom” and “in what circumstance”. Pawson and Tilley (1997)
suggested an important focus on understanding the theory underlying the program
mechanisms to understand and explain EA outcomes. Understanding the theory underlying
the program implementation is helpful for deriving the mechanisms by which architecture
transformation is achieved or not. Mechanisms are a fundamental part of social realist
explanation as they provide the causal possibilities within social programs.
These aims helped to understand some of the mechanisms built into the implementation
program, such as the various consultation group mechanisms (i.e. TOGAF: people’s roles and
actions in the EA program) and the support mechanisms for encouraging frequent input from
stakeholders. Given such aims and their associated social focus, the underlying philosophy
should clearly represent these elements. It was also important to properly examine the EA
program mechanisms in order to fully appreciate the success of the program.
In some EA studies, particularly larger-scale architecture transformations that purely examine
technical or business aspects, the program mechanisms were deeply involved in, and perhaps
led the EA implementation. In this study, which explored the role of people and the social
aspects of EA implementation, the program mechanisms did not substitute the main critical
realist theories by which social mechanisms are proposed, but rather supplemented them.
Data analysis and findings:
Eleven program mechanisms (theories) of successful EA implementation were identified in
the EA literature (see Chapter 2) as follows:
1.
Individual engagement: A mechanism that ensures all stakeholders are on board and
remain involved along the way, including business people, the IT people on the ground,
management, and the end-users (Gartner, 2009; Bente et al., 2012).
222
2.
Stakeholders’ role: A mechanism that addresses stakeholders’ concerns by identifying
and refining their requirements through promoting views of the architecture that show
how their concerns and requirements will be addressed, as well as showing the tradeoffs that will be necessary to reconcile potentially conflicting concerns of different
stakeholders (Ross et al., 2006; Gartner, 2009; Anderson & Backhouse, 2009; TOGAF,
2011; Bente et al., 2012; Janssen, 2012).
3.
Governance: A mechanism that ensures business and IT projects achieve their objectives
at: a) companywide level; b) business unit level; and c) project team level (Ross et al.,
2006; Anderson & Backhouse, 2009; TOGAF, 2011; Bente et al., 2012; Janssen, 2012).
4.
Linkage: A mechanism that ensures the architecture reflects and informs the goals and
priorities of all parties through: a) architecture linkage; b) business linkage; and c)
alignment linkage (Ross et al., 2006).
5.
Collaboration: A mechanism that ensures the EA program supports new ways of doing
business collaboratively with partners and customers. The result is a “multi-entity”
ecosystem that allows interaction at more touchpoints and additional depth (Ross et al.,
2006; Gartner, 2009; Bente et al., 2012, Janssen & Klievink, 2012).
6.
Communication: A mechanism that ensures ongoing communication and negotiation
between IT and business – why the changes are needed (the motivation); the benefits to
be expected from the changes; where the changes are expected to be made and what
the expected changes may be. This mechanism also ensures sufficient stakeholder
understanding and support (Gartner, 2009; Anderson & Backhouse, 2009; Gravesen,
2012; Bente et al., 2012).
7.
EA compliance: A mechanism that ensures various important people comply with their
associated responsibilities; including people’s actions and their relationship to
architectural compliance over the period of EA implementation (TOGAF, 2011; Čyras &
Riedl, 2012).
8.
EA conformance: A mechanism that ensures all the features of the EA program are
implemented in accordance with the specifications (TOGAF, 2011).
9.
Shared vision: A mechanism that ensures EA principles, business linkages, baseline and
targeted architecture are identified (Janssen, 2012).
10. Socio-political: A mechanism that ensures all political aspects such as trust, goodwill,
power and mutual interests of stakeholders are embodied in the EA implementation; the
223
creation of a shared vision, communication amongst stakeholders, and evaluation of the
EA impact (Janssen, 2012).
11. EA Leadership: A mechanism that ensures the key success factors of communication and
team-building are incorporated into the EA program. A good mix of social skills, business
focus, IT literacy, and an ability to lead are crucial (Ross et al., 2006; Gartner, 2009;
TOGAF, 2011).
Comparison with different theories: What works, for whom, and in what circumstances?
Table 6.4 describes how the program mechanisms map to each CR theory according to the
social mechanisms uncovered. This table shows the variety of possible objectives and
emphases, and how EA objectives might overlap or conflict with some mechanisms. The
diversity of program mechanisms is also reflected in the terminology in the EA literature
(Chapter 2).
Table 6.4 Mapping the ‘program mechanisms’ to each of CR theories by which social
mechanisms are discovered
Social mechanisms proposed by:
No
Program Mechanism
Meta-theory of MA
Reflexivity Theory
1
Individuals’ engagement
Diversification; Opportunism
Sectionalism
2
Stakeholders’ role
Sectionalism; Diversification
Diversification
3
Governance
Sectionalism; Opportunism
Sectionalism; Diversification
4
Linkage
Opportunism; Collaboration
Sectionalism; Diversification
5
Collaboration
Opportunism; Collaboration
Sectionalism
6
Communication
Opportunism; Communication
Sectionalism
7
EA compliance
Opportunism; Collaboration
Sectionalism; Diversification
8
EA conformance
Opportunism; Collaboration
Diversification
9
Shared vision
Opportunism; Communication
Diversification
10
Socio-political
Opportunism; Communication
Sectionalism; Diversification
11
EA Leadership
Opportunism; Communication
Sectionalism; Diversification
224
Table descriptions – what works, for whom, and in what circumstances:
Table 6.4 elaborates and estimates the EA theories underlying the program implementation
(the role of program mechanisms) (see Chapter 2) to support (a) the social mechanisms
proposed by MA, and (b) the people-focused mechanisms underpinning the frequent
stakeholders’ interaction (reflexive mediation of mechanisms). What this verifies is that for
EA to work it must be compatible with the observed situational logic. This powerfully supports
Archer’s argument about what situational logics are important and it also supports the benefit
of knowing what situational logics might be in place. The reflexivity mechanism identifies the
mechanisms shaping their ultimate concerns – it defines the people focused basis for
individual and collective responses.
1.
Individuals engagement:
a) Viewed from MA Meta-theory: This theory fits the diversification mechanism
proposed by MA as it acts as an instrument for corporate agents to engage with
primary agents and individual actors, which, under social interaction circumstances,
ensures alignment of the EA program, whether operationalised locally by the
business units or globally by the university. The theory is also suited to the
opportunism mechanism as it essentially serves as a mechanism to engage the
support of all stakeholders with an interest in or responsibility for EA
implementation, to ensure that the university’s interests are served and the EA
objectives are achieved.
b) Viewed from reflexivity theory: This theory fits into the reflexivity mediation of
people-focused mechanisms (sectionalism) as it describes people’s adoption of EA
(their actions and interactions) which is culturally shaped by the interplay between
people’s context and concerns. In other words, people’ s ultimate concerns are
related to sectionalism – the macro-level opportunism encourages the belief that
they can benefit their own sectional interests.
2.
Stakeholders’ role:
a) Viewed from MA Meta-theory: This theory fits with the sectionalism mechanisms
proposed by MA as it focuses on the requirements of users and stakeholders, which
225
in socio-cultural circumstances, creates a culture of understanding across the
university and provides useful insights for EA implementation. This theory is also
compatible with diversification mechanisms as it fundamentally focuses on business
outcomes, particularly considering that the business of universities structurally
requires their members to operate in more than one information domain.
b) Viewed from reflexivity theory: This theory suggests that different stakeholders with
different roles in the system will have different concerns, which in reflexivity
mediation of people-focused mechanisms (diversification), is one of the key interests
of crucial importance to stakeholders in the EA implementation, and determines the
acceptability of the EA.
3.
EA governance:
a) Viewed from MA Meta-theory: This theory works for sectionalism and opportunism
mechanisms proposed by MA. In both mechanisms, the theory encourages
university-wide participation rather than local objectives. Stakeholders are expected
to actively contribute by quantifying their IT demands for the EA program.
b) Viewed from reflexivity theory: From the reflexivity mediation of people-focused
mechanisms (sectionalism and diversification), this theory engages with
stakeholders and ensures consistent governance of their interests (context and
concerns).
4.
Linkage:
a) Viewed from MA Meta-theory: By nature this theory works on opportunism and
collaboration mechanisms proposed by MA. As the theory shows a tangible link
between the core strategies of the university and enablement of the EA program.
Linkage gives stakeholders opportunities to collaborate in order to achieve synergies
across multiple architecture priorities.
b) Viewed from reflexivity theory: This theory fits with reflexive mediation of
mechanisms (sectionalism and diversification), as the linkage mechanisms serve to
commit stakeholders to the program and contribute most to architecture initiatives
and implementation.
226
5.
Collaboration:
a) Viewed from MA Meta-theory: Fundamentally, the theory fits with opportunism and
the collaboration mechanisms proposed by MA. This theory suggests a “multi-entity”
ecosystem that allows interaction by stakeholders at more touchpoints and in more
depth, compatible with collaboration mechanisms in dealing with social complex
situations as an opportunity to achieve more flexible architecture and an open
architecture solution.
b) Viewed from reflexivity theory: This theory is appropriate for reflexive mediation of
mechanisms (sectionalism) as it suggests EA initiatives should be organised into
stakeholders’ contexts and concerns in order to identify any initiatives that could
hinder implementation. Culturally, reflexive mediation of mechanisms (sectionalism)
is used to manage collaboration between the EA program and stakeholders.
6.
Communication:
a) Viewed from MA Meta-theory: This theory articulates that most people in the
program attempt to do what they think is right under certain circumstances, but
without clear communication some of their actions will do as much harm as good.
This is consistent with opportunism and the communication mechanisms proposed
by MA in that the main purpose of EA is to communicate the benefits of EA to
stakeholders. Opportunism requires a communication plan, covering stakeholders’
interests at all levels of the university’s structure and culture, ranging from an EA
initiative to governance.
b) Viewed from reflexivity theory: This theory ensures ongoing communication and
negotiation between IT and business areas about the benefits to be expected, where
the changes are expected to be made and what the expected changes may be. This
is in harmony with reflexive mediation of mechanisms (sectionalism and
diversification) as the mechanisms allow stakeholders to confirm that the program
will address their concerns.
227
7.
EA compliance:
a) Viewed from MA Meta-theory: This theory argues that EA compliance ensures
various important people with associated responsibilities are brought together, as
well as people’s actions and relationships in architectural compliance over the period
of EA implementation. It fits with opportunism and the collaboration mechanisms
proposed by MA as these mechanisms act as instruments for engaging all
stakeholders with an interest in or responsibility for EA implementation, to ensure
that the university’s interests are served and the EA objectives achieved.
b) Viewed from reflexivity theory: This theory found that stakeholders’ delegation of
responsibility for EA implementation can entrench EA within the organisation. In
terms of reflexive mediation of mechanisms (sectionalism and diversification), this
theory makes people, culturally and structurally, adopt a strategic stance towards
constraints and enablements, which can bring compliance to the implementation.
8.
EA conformance:
a) Viewed from MA Meta-theory: This theory fits with opportunism and the
collaboration mechanisms proposed by MA as conformance provides an integrated
structure for managing EA activities across interest areas by ensuring the
architecture is implemented in accordance with the specifications. In terms of MA,
stakeholders’ acceptance of the inbuilt structures and mechanisms are dependent
on EA collaboration and the opportunity to develop a peer-to-peer manner of
learning about architecture specification to further assist a peaceful transition to the
targeted state.
b) Viewed from reflexivity theory: This theory focused on managing EA activities across
interest areas where ways of doing things may have been impacted. In terms of the
reflexive mediation of mechanisms (diversification), it sought stakeholder
acceptance and endorsement for conformance as the new structure progressed to
support the program. This is consistent with reflexivity theory in identifying people’s
internal relations leading to successful EA implementation at UX.
228
9.
Shared vision:
a) Viewed from MA Meta-theory: This theory suggests that EA principles, business
linkages, baseline, and targeted architecture be identified through the shared vision
mechanism, consistent with opportunism and the communication mechanisms
proposed by MA. In terms of MA, opportunism underpinned EA initiatives in
identifying the risks associated with the vision for the architecture, and assessed the
readiness for IT-business transformation and the potential risks. It emphasised a
clear line of sight between technology decisions and business outcomes to provide
UX with a unified, targeted architecture for engaging stakeholders and delivering
stronger outcomes. Opportunism required a communication plan that covered all
the university’s cultures and met all stakeholders’ needs, ranging from an EA vision
to availability of technology infrastructure. Effective communication to the right
stakeholders at the right time was critical for successful implementation.
b) Viewed from reflexivity theory: This theory suggests the initiatives of architectural
transformation require effort to bring about structural change that will be
accomplished by a shared vision of the social relationship. Social relationship was the
basic condition of reflexivity theory for attaining knowledge, in particular about
people’s internal inclinations towards EA implementation. Once the social
relationship was understood, it became possible to define appropriate solutions and
rally stakeholders around a common vision and purpose. In terms of the reflexive
mediation of mechanisms (diversification), encapsulating an EA vision that is aligned
with social conditions is crucial for people interaction and for developing a broad
range of different targeted architectures.
10. Socio-political:
a) Viewed from MA Meta-theory: This theory suggests that targeted architecture
influences organisations and people elements (roles and responsibilities) which are
politically sensitive. It also claims EA implementation is more politically oriented than
technically or business oriented. For instance, EA can serve as a way for the
organisation to engage with business-unit projects that might otherwise proceed
without involvement in the EA program. In ensuring all political aspects, such as trust,
229
goodwill, power, and mutual interests of stakeholders are embodied in EA
implementation, in terms of MA the process of tailoring stakeholders and their
power, influence, and interests provides opportunities to remove obstacles
elsewhere in the university. This includes the creation of a shared vision,
communication amongst stakeholders, and evaluation of the EA impact.
b) Viewed from reflexivity theory: This theory explains that, without appropriate
political backing, EA implementation is bound to fail. In terms of the reflexive
mediation of mechanisms (sectionalism and diversification), sectionalism serves as a
solution for understanding the power of people’s tendencies to resist the EA
program, while diversification identifies key stakeholders, their power, influence and
interests, and defines the key business requirements to be addressed by EA. The
reflexive mediation of people-focused mechanisms enables understanding of
people’s particular courses of action (personal concerns) in relation to their social
circumstances (context of EA).
11. EA leadership:
a) Viewed from MA Meta-theory: This theory emphasises teamwork, interpersonal
skills and an ability to clearly define and communicate for strategic leadership. EA
leadership also requires extensive and substantial practical experience and applied
knowledge of the subject. Getting stakeholders to agree on an EA program might
otherwise be exponentially tougher, as the program may mean that the university’s
functional areas (business units) lose discretion over core business processes and
sometimes over the people and systems that execute them. This study uncovered (in
MA analysis) that the involvement of more people is considerably better than one or
two. Their different experiences, knowledge and perspectives led to interesting and
insightful debates. Since EA created a shared understanding of how UX would
operate, the earlier battles were forgotten and people stopped questioning the value
of EA. This occurred because of the leadership role in communicating the EA, since
MA recommends communication as a key mechanism for supporting the leadership
role. Targeted EA can be achieved by involving theory in addressing challenges of
cultural complexity. Opportunism provides communication mechanisms with
targeted information to the right stakeholders at the right time.
230
b) Viewed from reflexivity theory: This theory suggests that EA leadership is not just
about command and execution. The role of leaders is to continually empower people
with a vision and execution strategy, and to embed governance elements into EA
implementation. In terms of the reflexive mediation of mechanisms (sectionalism
and diversification), leadership is broadly shared among stakeholders by adapting to
different cultural and structural environments for social reorientation and
architectural transformation, which in turn leads to social transformation.
From the above discussion it is clear that the role of program mechanisms is to contribute to
CR theories by which social mechanisms are proposed; and to examine how different program
mechanisms (EA theories) can support different parts of: a) the social mechanisms proposed
by MA; and b) the reflexive mediation of people-focused mechanisms. Understanding the
relationship of the mechanisms in both theories is a major concern. In general, theories
provide the basis for examining the specific mechanisms to which people respond and
ultimately people’s adoption of EA, shaped by the interplay between social structures,
cultures and people’s reflexives. This perspective allows the best mechanism configurations
for enabling success of EA programs in the university sector to be revealed, and in University
X in particular.
6.4 Domain of Actual
The actual domain describes events that happen, whether we experience them or not. This
mid-dimension on the CR ontological map (see Figure 6.1) is part of the real domain. The
actual domain incorporates events and non-events which are not experienced (empirical
domain), i.e. what happens in the world is not the same as that which is observed (Danermark
et al., 2002). In other words, this domain is where key implementation mechanisms produce
events and non-events which are not directly observable within the EA events (see Figure
5.11). In identifying events and non-events that are mostly hidden in the actual domain, the
method of analysis uses abduction (see abduction stage). Table 6.5 presents the events and
non-events produced by the key implementation mechanisms in the abduction stage (section
6.3.2).
231
Table 6.5 Events and non-events in the domain of actual
Key implementation
mechanism
Event of ‘Actual Domain’
1. Diversification management:
communication events in
making decisions about
diversification of business
unit roadmap, technology
environment, and sociocultural practices.
Communication
Collaboration
1. Synergy management:
collaboration events in
coordinating the IT activities
of various business units.
2. Linkage management:
collaboration events in
tailoring stakeholders power,
influence, and interest to
provide the opportunity to
remove obstacles that are
already carried out elsewhere
in the University, thus it can
add into EA program specific
tasks in aligning the key
business requirements
Non-event of ‘Actual
Domain’
2. Strategic alliance:
enables the EA to be
communicated and
understood by all of UX
stakeholders, which may
allow them to verify that
the EA program will
address their concerns
3. Distributed leadership:
communication initiative
in engaging individuals at
all levels in leading the
transitioning the
architecture of ITBusiness transformation
(change) to keeps
everyone (the primary
agents and individual
actors) ‘on board’ and
keeps all focused on the
program.
3. Strategic alliance:
collaboration initiative to
support more detailed
definition of a
consolidated, crossbusiness unit roadmap
(the University functional
area) within the EA
program.
232
6.5 Summary
This chapter provided a thorough description of the research findings from University X’s EA
implementation and described the analysis between the abstract and the concrete from a CR
ontological viewpoint. It concludes with an overview of the three overlapping CR ontological
domains to show: a) the generative mechanisms that enabled the success of EA
implementation (in the real domain); b) the events and non-events not directly observable
within the EA events (in the actual domain); and c) the empirical facts of the UX EA events (in
the empirical domain). In terms of social theories this study explains the key implementation
mechanisms and social responses underlying successful large-scale architecture
transformation in UX. In terms of EA theories it explains the important role of program
mechanisms in supporting the CR theories whereby social mechanisms are proposed. Since
these are produced by people they can be changed by people. People’s invisible internal
relations, which made EA implementation at UX succeed, were uncovered through reflexivity
theory. To summarise, Figure 6.6 presents the research findings.
Figure 6.6: Research findings: The relation of three overlapping CR ontological domains in
University ‘X’ EA implementation
233
The following chapter presents the theoretical and practical implications of the research
findings which frame the implications of the research questions presented earlier in the
thesis.
234
CHAPTER SEVEN: DISCUSSION
7.1 Overview
This chapter presents the conclusions from the research findings discussed in the previous
chapter by revisiting the initial research questions and objectives of the research. It outlines
what was involved in theorising people’s adoption of Enterprise Architecture (EA) in relation
to the methodological framework. The theoretical implications of current EA implementation
trends and emerging social issues offer a contribution to Information Systems (IS) theory and
recommendations for EA practice. The following and final sections of this chapter discuss the
research limitations and future research directions.
7.2 Research Questions Revisited
In the following sections, the research questions are addressed in reference to the research
findings from University X’s (UX) EA implementation case analysis. Figure 5.6 shows the
related research questions and their objectives, reproduced here for ease of reference. These
research questions were addressed using Critical Realist (CR) mechanism-based explanations,
and involved theory-oriented accounts of situations in macro–micro–macro contexts
including: a) Archer’s Morphogenetic Approach (MA) schematic stage – the structure, culture
and agency (people) or SCA; and b) reflexivity theory to identify the reflexive mediation of
mechanisms constituting people’s contexts and concerns within the program.
The macroscopic context explored the interplay between structural and cultural systems and
agency, which were intended to uncover the social mechanisms under which the EA program
was implemented. Context also explored the program mechanisms (see Chapters 2 and 6)
built into the EA implementation and associated with theories gathered from the EA literature
review (program mechanisms are explained in the methodological framework implications
section). The microscopic context explains the important modes of reflexivity that
demonstrated how people responded to the embedded mechanisms of EA implementation.
People’s reflexive deliberations also formed an important mechanism for explaining how
structures and cultures constrained and enabled agents to adopt the program.
235
Figure 5.6: The schematic diagram of research questions: The interplay within and between
the cycles of social change in morphogenetic approach (adapted from Archer, 1995, pp. 193,
218, 295, and 303)
7.2.1
Sub-research question number 1: Macro – micro context of cultural
and structural system
The first sub-research question was: “What are the important situational mechanisms, that
by associated social structure and culture, causally condition individuals’ actions?” The
research objective was to “identify the situational mechanisms by which social structure
conditions individuals’ actions and how cultural environments shape their social situations.”
In terms of MA, this sub-research question is located in Time period 1 (see Figure 5.8). The
findings of this study revealed that situational mechanisms arose from the results of past
sociocultural and social interaction prior to EA implementation (the previous morphogenetic
cycle outcomes in the first-order emergent). In other words, Time period 1 was the initial
stage of EA implementation (see Figure 6.4) at UX, where the modes of conditioning at
structural and cultural levels (macro) predisposed people (micro) to using the IT environment
to promote their important projects. It was delineated that the prevailing SL was opportunism
(see Figure 6.4). The following discussion addresses the first sub-research question:
236
1.
From a cultural point of view, the analytical history shows that, in the past, an
understanding was reached with university stakeholders whereby the most suited
members, by virtue of their natural competence, capability, skills, and qualifications –
managed the IT environments for their specific business unit activities.
At the
institutional level, members of the university’s functional areas (faculties, schools, and
service centres) were shaped and guided by the results of past sociocultural interaction
in using the IT environment solely for their own purposes. Archer (1995) called these
modes of situational logic specialization (Archer, 2015). As per Figure 6.4, specialization
is viewed as the act of a business unit making the IT environment suitable for their own
purposes. This study viewed specialisation as a situational mechanism of the cultural
system because the situational logic of specialisation is a causal power of the cultural
system in the social world. Specialization was the legacy mechanism that renders the IT
environment suitable for their own interests. Culturally, the outcome of this mechanism
created a huge gap between expectations and reality (see Text Box 6.1), and led to poor
governance and decision-making caused by localised IT governance.
2.
From a structural point of view, the analytical history shows that in the past, the process
of addressing markedly different needs lead to differentiating the university’s IT
environments between one and many (different faculties, schools and service centres),
simple and complex (IT activities based on competence and capability of the university’s
functional areas), and homogeneous and heterogeneous qualifications and skills,
changed the IT environment from relatively generalised to functionalised or businessunit owned. At the institutional level, staff in the university’s functional areas had been
shaped and guided by structural competition as a result of past social interaction in the
form of divergence from the IT environment for the sake of convergence in their IT selforganising. This situational logic during the initial stages of EA implementation can be
described as differentiation (Archer, 1995), creating diversity through unity (Archer,
2013a). This study viewed differentiation as the situational mechanism within the
structural system, since the situational logic of differentiation is the generative power of
system integration (SEPs) (see Chapter 4) in shaping people’s situations within their social
systems, institutional structures, roles, and positions. Differentiation carries the
underlying mechanism that structurally influences people (agency) synchronically by
237
constraining and enabling certain agential powers. It was used as an instrument and
natural way of determining IT arrangements in the initial stages of EA implementation at
UX. It was also the legacy mechanism of the university’s functional areas, whereby IT
arrangements were made exclusively for their own purposes. Structurally, this
mechanism rendered IT at UX difficult to integrate or automate across system
boundaries, and resulted in business inefficiencies.
7.2.2 Sub-research questions number 2 and 3: Micro – micro context of
socio-cultural and social interaction
Sub-research question 2 (S-RQ2) was: “What are the consequent interaction mechanisms
triggered by EA implementation at University X?” with the objective of “describing the
interactive mechanisms linking the social situation (context) and the people’s personal
concerns (people’s thoughts and ideas that tend to consider themselves – their desires,
beliefs, values, acquaintances and interests) to influence their actions”. Sub-research
question 3 (S-RQ3 was: “How does the existing culture and structure within University X
impact the EA implementation and shape the interaction mechanisms triggered by the
implementation?” with the objective of “identifying the cultural and structural dimensions of
an action context, both how they shape and are shaped by groups of individuals”. In terms of
MA, these S-RQs are located in Time 2 to Time 3 (see Figure 5.8) and are related and
connected to each other, whereby S-RQ2 bestows meaning on S-RQ3 and vice versa. The SRQs brings together two CR premises: The Morphogenetic Approach and Reflexivity Theory.
Time 2 to Time 3 was the EA implementation period, when interaction between the social and
cultural elements shaped and were shaped by groups of individuals within the UX population.
The following discussion addresses both of the sub-research questions in terms of MA:
1.
At a socio-cultural level (S-C), the corporate agents were looking for solutions to address
the lack of architecture initiatives and governance in previous cultural systems (Time 1).
It is during this period of time (Time 2 to Time 3) that the PEPs were influenced by
corporate agents’ preponderance for looking after the interests of particular collectivities
(primary agents) and individual actors (EA end-users). Corporate agents’ actions in
obtaining appropriate political backing and focusing on the rights, roles and equitable
treatment of shareholders in the EA program are referred to as the situational logic of
238
sectionalism (Archer, 1995). In this study, underlying sectionalism was the lack of
involvement by appropriate UX stakeholders in the EA program during the initial stages
of EA implementation. Sectionalism was the interaction mechanism at the socio-cultural
level that enabled an understanding of the power of people’s tendency to resist the EA
program. Identifying stakeholders’ resistance early on led to seeking their input in
shaping the architecture, thereby ensuring their support and potentially improving the
quality of the EA program by meeting the needs of its users and stakeholders. This
mechanism to address sectionalism was built in to the EA program and sought
encouragement and university-wide participation rather than serving just local
objectives. Addressing the observed Sectionalism at the cultural level sought a culture of
understanding across the university of the purpose and benefits of the EA
Implementation. The EA “roadshow” was an important element in addressing cultural
sectionalism.
2.
At a social interaction (SI) level, the PEPs were influenced by corporate agents’
endorsement of the EA program at UX through EA events, in order for decisions on
technology investment to provide the best outcomes for the university. However, the EA
events for the university’s functional areas in particular were not as effective as was
hoped, due to a misconception about the purpose of the cross-organisational business
forums. Some primary agents were under the misapprehension that the forums were just
a potential source of funding to do what they wanted to do. Others regarded EA as a
waste of time, not beneficial for their business areas and consequently became passive
participants. To resolve the ineffectiveness of the EA forums due to a lack of engagement,
the corporate agents at UX employed a different model to gain buy-in and acceptance of
the EA program. This model engaged people in each of the university’s functional areas
individually, through a diversification mechanism that ensured alignment between local
and global objectives of the EA program. As an interaction mechanism, diversification
identified: a) who would gain and who would lose from EA implementation; b) who would
make the decisions; c) who would control resources; and d) who had influence.
Diversification provided a solution for the diversity of university business activities and
its collegial management. In terms of situational logics (Archer, 1995) at the socio-cultural
level, the corporate agents’ actions in engaging the various stakeholders’ interests in the
239
EA program is called diversification. It describes the social situation at the social
interaction level: a) between the corporate agents and the collectivities; and b) between
the corporate agents and the individuals.
The above discussion addresses the question: “How does the existing culture and structure
within University X impact the EA implementation and shape the interaction mechanisms
triggered by the implementation?” which leads to determining: “the consequent interaction
mechanisms triggered by EA implementation at University X”.
In describing the interaction mechanisms linking the social situation (context) and people’s
personal concerns the study used reflexivity theory (see Chapter 6, the retroduction stage).
By using the reflexivity investigation tool, ICONI, this theory explained how people reflexively
responded to the interaction mechanisms of EA implementation. It also reconstructed their
reflexive deliberations (their internal conversations) about adopting the EA. These reflexive
deliberations formed an important mechanism for explaining how structure and culture
constrained and enabled agents to adopt the program (reflexive mediation of mechanisms)
(see Figure 6.5). The following discussion addresses both the sub-research questions in terms
of reflexivity theory:
1.
ICONI
a) EA management: This refers to people who affected (corporate agents) and were
affected (primary agents) by EA implementation. The dominant mode of reflexivity
of the EA management was autonomous reflexive. In other words, these players had
the freedom to govern or control their own affairs with regard to EA program, which
allowed them some certainty that the program would address their concerns. As a
result, the autonomous reflexives counted on EA implementation. EA management’s
predominant concerns were communicative concerns. This means they were
ultimately more concerned about family responsibilities and family wellbeing and
therefore more likely to secure their positions and careers rather than seeking a
vocation. Communicative concern relies on the success of EA implementation to
keep their job and maintain continuity of their career. The dominant mode of
collective reflexive is collective autonomous reflexive, which means this collective
240
of people deals with the EA program in a way that is pragmatic and focused on the
wellbeing of the group.
b) EA end-users: End-users (individual actors) are those who are affected by EA
implementation. The dominant mode of reflexivity for individual actors was metareflexive. In other words, these people tended to seek vocations rather than a
career, compelled by the need to make a difference rather than simply meeting their
immediate needs. These end-users felt EA implementation was only somewhat
important as it was unlikely to impact directly on their way of doing things. This does
not mean they refused to adopt the EA program, they just accepted it. It was evident
from end-users’ ultimate concerns about EA implementation that a) those with
communicative concerns gave credit to the program; and b) those with fractured
concerns went along with the majority of their colleagues. The dominant mode of
collective reflexive was collective meta-reflexive, which means the collective of
people in this mode considered EA a relational enhancement of the way they were
doing things.
2.
Reflexive mediation of mechanisms
a) EA management: The qualitative feature of corporate and primary agents within the
social situation (context) is referred to as contextual discontinuity (Archer, 2008;
2010a), as their actions would transform the cultural and social structure. This is
predominantly a reflexive action. Upward mobility refers to corporate and primary
agents’ social stratification. It is an agential aim to improve upon social position, and
if successful, become upwardly mobile. Their relationship to structure and culture
was to adopt a strategic stance towards constraints and enablements. Despite having
the freedom to govern autonomously, their ultimate personal concerns made them
act in accordance with their moral duty rather than their own needs and desires. The
ultimate concern of corporate and primary agents was communicative. Their mode
of reflexivity saw them devote themselves strenuously to the program and
contribute most to architecture initiatives and implementation. This was the
reflexive mediation of mechanisms for EA management (corporate and primary
agents). According to Archer (2008), this reflexive mediation of mechanisms leads to
241
morphogenesis, where sectionalism and diversification are the generative
mechanisms.
b) EA end-users: End-users engaged with sectionalism and diversification mechanisms
to ensure consistent governance of their interest and concerns. The qualitative
feature of individual actors within the social situation or context is referred to as
contextual incongruity (Archer, 2008; 2010a). This social situation may have been
unsuited to the EA program, but as long as there was a continuous promise of the
program’s benefits for end-users to realise their needs, their predominant reflexive
action acknowledged faith in the EA program. As meta-reflexives they subscribe to
one of three social stratifications (see Table 5.2) depending on their social situation:
a) socially immobile; b) upwardly mobile; and c) socially volatile. However, due to the
success of EA engagement during implementation, the individual actors’ social
stratification was upward mobility. Their relationship to structure and culture was
subversive towards constraints and enablements as they tended to seek vocations
rather careers, compelled by a need to make a difference rather than simply meeting
their immediate needs and desires. However, their ultimate personal concerns led
them to give credit to the program (communicative concern) and to embrace the
program like most of their colleagues (fractured concern). Their modes of reflexivity
signalled a propensity for social re-orientation and social transformation (see Table
5.2). This was the reflexive mediation of mechanisms for EA end-users. According to
Archer (2008), this reflexive mediation of mechanisms leads to high morphogenesis
and low morphostatis, conducive for interaction mechanisms that encouraged
acceptance of EA.
The above discussion describes the interaction mechanisms that linked the social situation
(context) and people’s personal concerns and led to identifying the cultural and structural
dimension of actions which shaped and were shaped by the various groups of individuals.
242
7.2.3 Sub-research questions number 4: Micro – macro context of
elaboration
The final sub-research question was “what are the necessary conditions to encourage
individual and collective acceptance of EA practices?” Its objective was to “specify the
transformational mechanisms by which individuals, through their actions and interactions,
generate various intended and unintended social outcomes.”
A number of EA events which occurred during the past period (Time 2 to Time 3) led to Time
4 being approached in a tactically opportunistic way, conscious of policy and practice, yet
taking advantage of social events and non-events. The policies and practices of corporate
agents’ in ensuring EA compliance with stakeholders’ requirements to provide the best
outcomes for the university in the “systemic level” (Archer, 1995) was a consequence of the
prevailing situational logic of opportunism. The prevailing situational logic consequent from
the interaction at the level of ideas suggest opportunism – high systemic integration of UX in
the face of sandstone university domination with a correspondingly low social integration
where faculties, schools and centres have no interest in each other. This creates opportunism
where specialization and sectionalism at the cultural level and differentiation and
diversification at the social level (Archer, 1995, pp. 303-304) create opportunities for selfinterested newly formed corporate agents. Primary agents reduce since they must take a
stance or be left behind.
Opportunism during the period of transformation (Time 4) motivated different forms of
strategic actions by predisposing different sections of the UX population to identify their own
interests. The process of tailoring stakeholders and their power, influence, and interests,
provided an opportunity to remove obstacles that were already in place elsewhere in the
university. As opportunism was the transformation mechanism at Time 4, it can be safely
concluded that opportunism represents the generative mechanism of morphogenesis and
thus this study was able to determine “the transformational mechanisms by which
individuals, through their actions and interactions, generated various intended and
unintended social outcomes”. Opportunism required that all stakeholders with an interest in
or responsibility for EA implementation had to ensure that EA was seen as a platform for
supporting their own innovation and opportunities. As the senior EA person points out:
243
I think [culture] will be the most difficult thing to overcome and I have no
illusions about that. I think the most critical thing is for senior executives to
continue to support the governance model and hold the line because we’ll
get a lot of counter argument. And to remind people that the enterprise
architecture wasn’t developed via IT in isolation, it was developed through
discussing and having consultation right across the university about what
their current problems were, but also how did they see teaching and learning
and research and administration in the future? It’s the ability for us to
communicate and educate people as to the value of [EA] and we don’t sort
of talk about it in terms of enterprise architecture, we do talk about having
an architectural assessment against our own map and all of that sort of thing,
but we think that the organisations not mature enough to really understand
enterprise architecture. …We think that will be a gradual learning maybe
over time. We will do it at the senior level, so we will do it at the IT
governance committee level.
To be successful EA must focus on supporting the prevailing opportunism evident in the
university:
But I think the role of enterprise architecture in that is really about making
sure what you’re doing is building flexibility and agility to respond. To me,
they’re just absolutely paramount for that, so making sure that in the
technical layers of architecture, that you’re building that capability to be
flexible and to be able to respond to changes at the business process level…
For me, IT governance, or technology governance, is about making sure
we’re investing in the right things to support the business strategies and
objectives, and the enterprise architecture is the thing that tells you what
the right things are. It’s almost like the decision-making framework for
people who have got to make decisions, so I can’t turn around and say “well,
I actually like this bit of technology or this bit of software.”
Providing that UX maintains its strength and solidarity at the systemic level (facing sandstone
domination) and allow freedom to diversify at the social level opportunism should continue
244
as the main situational logic. This is good news for second tier universities. EA provides a
strong base for systemic integrity – the success of the program promises a period of
advancement as the university makes the most of the opportunities available because of its
strong innovative IT. The danger is that if the new structure is not bedded down properly and
leads to a war between the various new powers – this would be counterproductive and
destroy much of the benefit achievable from an integrated IT platform.
7.2.4 Main research question
The main research question this study sought to address was: “what are the key
implementation mechanisms and social responses triggered by EA implementation that might
constrain and enable the success of the EA program in University X and the sector in general?”
The findings reveal several different mechanisms were operating simultaneously (during Time
2 to Time 3), the relation between the generative mechanisms (during Time 2 to Time 3)
generated the transformation mechanism at Time 4, and their effects were contingent. These
contingent effects are also what generated the key mechanisms, called key implementation
mechanism in this study.
The MA analysis suggests that: a) Opportunism supported by the EA program required a
communication strategy targeting all levels of the university’s culture and structure so that
people understood what EA opportunities could be provided (meeting stakeholders’ needs,
ranging from an EA initiative to the technology governance). Effective communication to the
right stakeholders at the right time was critical. b) Opportunism presented a loose situational
logic of opportunity, which then required sociocultural practices to take advantage of it.
Rather than addressing challenges of cultural complexity, the corporate agents led the
university community’s focus to issues of sharing and sourcing business capability, and
exploring new capabilities and ideas. Developing a collaboration plan to address complex
social situations presented an opportunity for more flexible and open architecture solutions.
Therefore it can be concluded that: a) effective communication and collaboration were key
implementation mechanisms generated by the transformation mechanism (opportunism) at
Time 4; and b) the recognition of a need to support the prevailing SL of opportunism was a
fundamental key implementation mechanism to drive the architectural transformation at UX.
245
The findings also revealed that the key implementation mechanisms defined abstractions
(during Time 1 to Time 4), representing the structures and cultures within UX at the time of
EA implementation, each aimed at a particular context for stakeholders (people) and
addressing a particular set of concerns. This is called social responses in this study. The social
responses triggered by EA implementation were dependent on the interplay between social
integration (PEPs) and structural and cultural (SEPs and CEPs) integration (see Chapter 5). In
UX people’s resistance to change (social integration) was low and structural and cultural
integration was high, leading the organisation towards social transformation. This state was
appreciated through the mechanisms identified as encouraging individuals’ acceptance of EA
practices. In terms of MA, the social responses triggered by EA implementation are described
as contingent complementarity (Archer, 1995), which had a net systemic result of great
cultural variety and required that the architecture should be structured to accommodate the
business units’ cultural differences.
7.3 Research Implications and Contributions
A considerable body of research has sought to understand the social implications of EA
programs and their implementation (Boh et al., 2003; Raadt et al., 2008; Gartner, 2009;
Gravesen, 2012; Janssen, 2012; Lohe & Legner, 2013). In theorising people’s receptiveness to
Enterprise Architecture (EA) the adoption of critical realism (CR) has significant implications
with respect to the objects investigated, the progress of subsequent research and the
outcomes that can be expected as has been demonstrated. Recognising that people-focused
mechanisms explain how EA is examined, led to theories that placed mechanisms at the
centre of the research to gain an understanding of the social context of EA implementation.
This study provides a number of important insights, especially in regard to implementation
mechanisms and the social responses that drive large-scale architectural transformations. The
usage of Archer’s MA has proven to be a valuable tool for understanding the complexity of
the university social systems. The social realist tool works perhaps because the University
social environment is an open system that intimately reflects the society within which it
operates. The social and cultural environment is fluid and mirrors the morphogenetic society
of which it is a part. Thus the approach used by Archer and evidenced in her series of books
works quite well. It may not work so well in a system that is closed and unchanging.
246
The following section presents the important implications of the methodological framework,
the contribution of theories related to social and technological phenomena in the domain of
Information Systems (IS), and recommendations for EA practice to determine successful
implementation, use and effects of EA programs in the university sector.
7.3.1 Methodological framework implications
In order to appreciate the importance of CR as a philosophy in the study and identifying the
key mechanisms that contributed to the success of EA implementation at University X (UX),
this study adopted the framework proposed by Danermark et al. (2002) to guide the
methodological framework. The framework is described in five stages (see Figure 5.3, Chapter
5) and has been reproduced here for ease of reference. The following presents the important
implications of the methodological framework:
Figure 5.3: The methodological framework of the study (Adapted from Danermark et al.,
2002)
247
1.
The first stage of the study explained the often complex and composite events, situations
and phenomena by making use of a number of sources like theoretical perspectives,
existing research, observations and actors’ own accounts among others. This is called the
Description Stage. In terms of this study, the framework provides a number of important
findings:
a) Eleven possible program mechanisms were identified from existing EA research (see
Table 2.6, reproduced here for ease of reference).
Table 2.6 Potential Success Program Mechanisms Identified from EA Literature Review
Program
Mechanism
Individuals’
engagement
Stakeholders’
role
Role as a Mechanism
To ensure that all stakeholders are on board
and remain involved along the way - the
business people, the IT people on the
ground, management and the end-users
themselves.
To change stakeholders’ perceptions of the
architecture by addressing their concerns
and requirements; and by identifying the
trade-offs that will need to be made to
reconcile their potentially conflicting
concerns.
Governance
To ensure business and IT projects achieve
objectives at: a) at companywide level, b) at
business unit level, and c) at project team
level.
Linkage
To ensure that the architecture reflects and
informs the goals and priorities of all parties
through: a) architecture linkage; b) business
linkage; and c) alignment linkage.
Literature
1. Gartner (2009)
2. Bente et al. (2012)
1. Ross et al. (2006)
2. Gartner (2009)
3. Anderson and
Backhouse (2009)
4. TOGAF (2011)
5. Bente et al. (2012)
6. Janssen (2012)
1. Ross et al. (2006)
2. Anderson and
Backhouse (2009)
3. TOGAF (2011)
4. Bente et al. (2012
5. Janssen (2012)
1. Ross et al. (2006)
248
Program
Mechanism
Collaborative
Communication
Compliance
Conformance
Shared vision
Sociopolitical
Leadership
Role as a Mechanism
Literature
To ensure the EA program supports new
ways of doing business collaboratively with
partners and customers so that the result is
a “multi-entity” ecosystem that allows
interaction at more touchpoints and in more
depth.
1.
2.
3.
4.
Ross et al. (2006)
Gartner (2009)
Bente et al. (2012)
Janssen and Klievink,
(2012)
To ensure ongoing communication
and negotiation between IT and business
centres – why the changes are needed (the
motivation), the anticipated benefits, where
the changes are expected to be made and
what the expected changes may be. This
mechanism also assures sufficient
stakeholder understanding and support.
To ensure various important people and
their associated responsibilities adhere to
architectural compliance over the period of
EA implementation.
To ensure all features in the architecture are
implemented in accordance with the
specifications.
To ensure that EA principles, business
linkages, baseline, and target architecture
are identified.
To ensure all political aspects such as trust,
goodwill, power, and mutual interests of
stakeholders are embodied in the EA
implementation, from the creation of a
shared vision to communication amongst
stakeholders and impact evaluation.
To understand the importance of
communication and team building as key
critical success factors. A combination of
social skills, business focus, IT literacy, and
an ability to lead are crucial to the success of
EA programs.
1. Gartner (2009)
2. Anderson and
Backhouse (2009)
3. Gravesen (2012)
4. Bente et al. (2012)
1. TOGAF (2011)
2. Čyras and Riedl
(2012)
1. TOGAF (2011)
1. Janssen (2012)
1. Janssen (2012)
1. Ross et al. (2006)
2. Gartner (2009)
3. TOGAF (2011)
249
b) A number of possible people-focused mechanisms were identified from CR
theoretical perspectives and existing research (see Table 4.3, reproduced here for
ease of reference) to explain the contexts for success or failure.
Table 4.3 Possible social mechanisms identified in the literature
Social component
Level
Syncretism
Compromise
Unification
Containment
Correction
Pluralism
Competition
Cleavage
Polarisation
Elimination
Systematization
Integration
Reproduction
Solidarity
Protection
Specialization
Differentiation
Sectionalism
Diversification
Opportunism
c)
Situational mechanism
Situational mechanism
Interaction mechanism
Interaction mechanism
Reproduction mechanism
Situational mechanism
Situational mechanism
Interaction mechanism
Interaction mechanism
Transformation mechanism
Situational mechanism
Situational mechanism
Interaction mechanism
Interaction mechanism
Reproduction mechanism
Situational mechanism
Situational mechanism
Interaction mechanism
Interaction mechanism
Transformation mechanism
Morphogenesis
Morphostatis
Literature
Social interaction
Structural system
Generative
mechanism
Cultural systems
Situational
Logic
Socio-cultural interaction
Tendency
Power
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
Archer (1995)
Archer (1995; 1996)
Empirical facts of the UX EA events, directly or indirectly observable, were discovered
through observation (see Chapter 6, Section 6.2).
250
d) People’s own accounts of the EA implementation were collected (Primary data:
interviews and ICONI data)
This stage provided a useful representation of the empirical facts and data with regard to
EA implementation in UX.
2.
The second stage was Analytical Resolution in which the Stage 1 representation was
resolved using Archer’s (1995) morphogenetic approach (MA) as the main defining tool:
the empirical material gathered in the first stage was elaborated and analysed in terms
of three-part cycles (periods of time), each with relative autonomy and yet interacting
with the others. This process determined: a) the relationship between a time prior to EA
implementation and the time of EA implementation (the first- and second-order
emergent properties); and b) the interplay between and within social structures, culture
and agency at the time of EA implementation (second-order emergent properties). These
analyses are also called the analytical histories of emergence (Archer, 1995). In terms of
this study, this stage represented the deep dimension where important mechanisms that
enabled the success of EA implementation were found. Analytical resolution provided a
number of important findings:
a) Possible underlying causal relationships between people (agency), structure and
culture in UX were identified, thus providing an understanding of how UX
stakeholders affected and were affected by EA implementation.
b) The interplay between agency, structure and culture over time (from Time 1 to Time
4) in UX, and their important mechanisms were identified, thus providing an
explanation of how they constrained and enabled EA implementation.
c)
Different modes of conditioning and interaction in situational logics were identified,
providing an explanation of how social situations predispose people towards specific
courses of action for the promotion of their own interests.
d) Finally, the study identified the key implementation mechanisms and social
responses triggered by EA implementation (generated by the transformation
251
mechanism), which explain the important mechanisms and social responses that
propelled such large-scale architectural transformation in UX.
The above approach provided an important link between realist ontology and practical
social-IS outcomes, and supplied fundamental consistency to ontology and methodology
(see Chapter 5).
3.
In the third stage, the framework was used to interpret and redescribe the different
components or aspects according to hypothetical conceptual frameworks and theories
about structures and their relationships. This stage is referred to as abduction or
theoretical redescription, where the original ideas of the object under study are placed
in a new context of ideas. In this study, abduction was used to seek connections or
relations, not directly observable, between the EA events (Stage 1) and the EA context
identified (Stage 2), giving new meaning to already known EA events and providing a
number of important findings:
a) The study was able to identify the relationship between particular EA events and EA
contexts as: a) social situations influenced by activities associated with EA
implementation (this is a social circumstance or event that occurs whether we
experience them or not); and b) social situations influenced by the willingness of UX
members to cooperate with each other in order to realise EA implementation (this
can be regarded as social cohesion or a non-event that occurs whether we
experience it or not).
b) The study shed light on social situations as events and non-events located in the
actual domain (see Figure 5.11, Chapters 5 and 6) to detect relationships in the new
context of EA (see Table 6.1, Chapter 6) and the key implementation mechanisms
(identified at the analytical resolution stage of the MA analysis).
Abduction provided useful insights into the actual domain, which was not directly
observable within EA implementation. This stage identified three events and two nonevents (see Table 6.5, Chapter 6) generated by different mechanisms in Stage 2, thereby
producing already known EA events (the observable EA events in Stage 1).
252
4.
The retroduction stage described a mode of inference by which this study arrived at what
is characteristic and constitutive of people in adopting EA, since MA (in Stage 2) still posed
the central problem of how social structure, culture and agency were actually linked
(their social relationship: Archer, 2008). In terms of this study, the retroduction stage
allowed clarification of the basic conditions for a social relationship: people’s actions and
interactions; people-focused mechanisms; reasoning and knowledge within EA
implementation. As these basic conditions are not directly observable, the retroduction
stage went beyond the empirical (something that can be observed and experienced) by
using reflexivity theory (Archer, 2008; 2010a) to uncover people’s internal relations and
their impact on the success of EA implementation at UX. This stage provided a number of
important findings:
a) The study was able to identify how people subjectively defined their particular
courses of action (people’s personal concerns) in relation to their objective social
circumstances (context of EA) by using the reflexivity investigation tool, ICONI (see
Chapter 6).
b) The study explained how structures and cultures constrained and enabled agents to
adopt EA, known as the reflexive mediation of mechanisms in critical realist terms.
c)
The study was able to identify the collective reflexivity or social relationships in EA
implementation, thus providing an explanation of people in “special relationships”
and how those relationship made them reflexive in a social, instead of a personal
way.
Retroduction was instrumental in explaining people’s responses to the people-focused
mechanisms (mechanisms that encourage individual and collective acceptance of EA
practices), and in clarifying the basic conditions for social relationships and the identified
mechanisms.
5.
This final stage of the research methodology entailed a comparison of different theories
and abstractions and estimated the relative explanatory power of the mechanisms and
structures. The purpose of the comparison was to highlight the theory of the program
253
mechanisms (identified in Stage 1) in order to determine the conditions for what works,
for whom and in what circumstance. This stage provided a number of important findings:
a) The study was able to identify the role of program mechanisms that contribute to CR
theories by which social mechanisms are proposed.
b) The study was also able to determine how different program mechanisms support
different parts of: i) the social mechanisms proposed by MA; and ii) the reflexive
mediation of people-focused mechanisms.
This stage provided the basis for examining the specific program mechanisms by which
people responded to the EA program and ultimately their adoption of the program, as
shaped by the interplay between social structures, cultures, and people reflexives. It also
allowed the study to determine the best mechanism configurations for enabling success
of the EA program at University X.
7.3.2 Contributions to information systems theory
“The UK Academy of Information Systems (UKAIS) defines information systems as the means
by which people and organizations, utilizing technology, gather, process, store, use and
disseminate information” (Ward & Peppard, 2002, p. 3). It is in this domain that the current
study was conducted. As explained by Mingers (cited in Ward & Peppard, 2002) “IS
[Information Systems] actually is part of the much wider domain of human language and
communication, IS will remain in a state of continual development and change in response
both to technological innovation and to its mutual interaction with human society as a whole”
(p. 3).
Enterprise architecture (EA) operates broadly in the social environment of an organisation
and needs to focus on the social rather than being driven solely by the technical aspects of
information technology (IT) (Anderson & Backhouse, 2009; Bente et al., 2012; Janssen, 2012).
Thus, many social aspects need to be thoroughly analysed, understood and interpreted early
in the IT-business transformation. The social aspects of EA are just as important as
appreciating the technical aspects in aiding the implementation of EA programs, particularly
identifying the people-focused mechanisms that constrain or enable adoption. A social
254
perspective is necessary to allow for such examination, and should be considered in the early
stages of architecture visions, principles and initiatives.
Given the heavy social emphasis of this study, the use of a critical realist perspective in
examining people’s roles and the social aspects of EA implementation provided an
advantageous perspective for theorising and examining the hidden people-focused
mechanisms within the IS domain. The use of university X as a case study was well suited to a
CR approach for finding explicit causal explanations for the complex social and EA phenomena
in the IS field. The following presents the vital role of CR theories in this study and its
contribution to existing IS research:
1.
The morphogenetic approach (MA)
Previous research has demonstrated use of the morphogenetic approach for explaining
EA evolution as the interaction between the existing structural setting (existing EA) and
the action of introducing new business or IT capability into an organisation (serviceoriented architecture or SOA) (Alwadain, 2014, p. vi). Whilst Alwadain’s study is more
focused on the relationship between agency (agency seen as an ‘action’: Alwadain, 2014,
p. 116) and structure to understand the nature of change, the emphasis on action and
structure has a different central point from this thesis, which emphasises the interplay
between structure, culture and agency. The emphasis on agency (people) is important,
as Archer (1995, p. 195) suggested people are capable of resisting, repudiating,
suspending or circumventing not only the structural tendency, but also the cultural
tendency in unpredictable ways because of their creative powers as human beings. These
processes of change (or reproduction) are known as generative mechanisms (Archer,
2015).
The present research study was designed to be an in-depth exploration using a
completely morphogenetic approach, to make possible an examination of the complex
social phenomena of structure, culture and agency in EA implementation. It sees an
important role for the morphogenetic approach to identify the hidden mechanisms and
situational logics at an ontological depth that is difficult to unveil through other
alternative IS and social theories. The study makes an important link between a realist
255
ontology and practical Social theory in the IS arena, constructing a consistent foundation
between ontology and methodology and practical theory.
The findings demonstrate that MA was useful in explaining: a) how the underlying causal
relationships between people (agency), social structures, and cultures provide an
understanding of how an organisation’s stakeholders affect and are affected by EA
implementation; b) how the interplay between agency, social structures and cultures
over time (from Time 1 to Time 4) and the important mechanisms provide an explanation
of how they constrain and enable EA implementation; c) how the different modes of
conditioning and interaction in situational logics provide an explanation of how social
situations predispose people towards specific courses of action for the promotion of their
interests; d) the key implementation mechanisms and social responses triggered by EA
implementation. This study therefore makes a valuable contribution to IS literature by
demonstrating that MA increases understanding of the causal mechanisms and contexts
necessary to achieve positive outcomes in IS research, and in particular an examination
of the role of people and their social contexts in EA implementation.
2.
Reflexivity theory
This research advances existing theory of how people respond to enterprise-wide IT
change. The findings demonstrate the usefulness of reflexivity theory in explaining how
personal projects are formed and how they mediate the exercise of structural and
cultural constraints and enablements in EA implementation. The findings also show that
the reflexivity investigation tool, ICONI, is useful to identify the necessary conditions that
encourage individual and collective acceptance of EA practices. In this way, this research
represents a unique theoretical contribution to IS theory by advancing our understanding
of people’s internal conversations (reflexivity theory), which factors are associated with
the contexts and concerns of people, and how this leads to their subsequent action or
inaction in adopting IS practices.
3.
Abduction
Many IS research studies undertaken from a CR perspective, have used abduction as a
vital instrument for explaining and proposing creative use of theory in order to derive a
256
suitable explanation for something that cannot be observed directly (Smith, 2010;
Mingers, 2011; Dobson et al., 2013).
This research uses abduction for broadly identifying the events and non-events located
in the actual domain (not directly observable). The type of abduction used in this study is
creative abduction (see Chapter 5), as it applies MA to EA in a way that largely has not
been used before. The present findings (see Figure 5.11 and Table 6.1) demonstrate that
abduction can be used to identify the interconnections between the empirical domain
(observable domain) and the real (perhaps largely non-observable domain - see Figure
6.6). By focusing the lens of abduction in these two domains (the empirical and the real),
this research study provides fresh insights into IS literature from a CR perspective.
7.3.3 Recommendations for enterprise architecture practice
Implementing an EA program is a challenging task, since the program is likely to face a
complex IT environment that may be inflexible and difficult to manage. More importantly, the
program will encounter opposition from stakeholders (people) in often unknown social
circumstances (structure and culture). Many EA frameworks and methodologies to date have
proposed a detailed model of architectural work products (ranging from business to
technology architecture and from governance to change management), including stakeholder
management, to overcome the complexities and challenges of EA implementation. These
frameworks provide substantial benefits in supporting critical business applications as well as
identifying key stakeholders, their concerns and objectives. Nevertheless, the real EA
challenges are associated with people issues, as people are capable of affecting and being
affected by not only the structural tendencies but also the cultural tendencies within
organisations. EA frameworks and methodologies need to address these social and cultural
aspects and their interplay with people to properly address potential issues in the architecture
implementation.
Implementing an effective EA program means dealing with a prescribed combination of
organisation (institution), people (stakeholder), business processes (functional areas) and
technology to secure holistic architecture targets. In view of the research findings it is logical
to conclude that EA programs need to not only deal with the technical aspects of IT
257
environments and their systems, but also with people in organisations who use IT to do their
business. The social aspects identified in the case study, such as the structural and cultural
attributes people bring to an enterprise, demonstrate that a favourable social context can
potentially encourage participation in organisation-wide, rather than local objectives, and
create a culture of understanding. Key elements are collaboration programs between
stakeholders to ensure business and IT projects meet both local and company-wide
objectives, and understanding the social aspects of implementation to positively shape how
the architecture is presented and communicated to all stakeholders.
To review, this research study allowed a deep examination of the key implementation
mechanisms, social responses, and program mechanisms (theory behind the EA
implementation) that drive large-scale architectural transformation in the university sector.
The case study provided a basis for understanding the contextual factors with the most impact
on large-scale transformation efforts in the university sector, and provided an opportunity for
proposing ways that existing frameworks and methodologies can more clearly reflect the
critical role of people in the change process. This focus help us to understand the social
aspects of the enterprise and the IT-business objectives from the viewpoint of people as a key
element in EA implementation. They offer a strategic approach to evolving the IT-business
environment of the enterprise in a way that deals with the social complexity of the
environment and effectively manages the changes in the environment.
7.4 Limitations and Future Research Directions
This research deals with the emergent properties in an analysis of structure, culture and
agency (people), in an attempt to bridge the gap between the explanatory powers of practical
social IS theories and the ontological strength of a critical realist philosophy. This type of
explanation demands the researcher to formulate new ideas about the interconnections
between the EA phenomena, unveiled in the theories (abstraction) and the empirical case
study (concrete) under observation. The time limit for completion of this study means that
the investigation may not be fully realised, as aspects of the EA are still being implemented.
The case example described the challenging university environment and illustrated the
important situational logics that directed people’s actions within the complex social context
258
of a university. Although limited to one organisation, the rich data gained from this
investigation is expected to be useful for other researchers who wish to develop a better
understanding of people as a key element in EA implementation, particularly in the higher
education sector.
The concept of people as a key element in EA implementation might lead to a vocabulary
useful for examining the overlapping edges between people, processes and technology in IS
research. If people do not like the program to be provided by structuring their processes and
technologies, they may seek to reject its implementation. As argued by Archer (1995, p. 218)
the compatibility or incompatibility of a cultural system with its sociocultural interaction
creates a range of possible situational logics which create conditions for social reproduction
(morphostasis) or social transformation (morphogenesis). It acknowledges the sociocultural
consequences of interactions between the structure and the culture to provide particular
situational logics that direct, but do not determine the actions of people. Therefore, it will be
interesting for future EA studies to examine organisations where opportunism is not the
prevailing logic. As this study indicates that, if EA is presented correctly and implemented with
agility in mind, it naturally supports opportunism. In other environments it may well be less
suited.
Future EA studies can be suggested in relation to this thesis – either to examine the extent to
which situational logics define the mechanisms for change or to generalise the findings
beyond the HE sector, or even beyond the EA and IS areas. The premise that people play a
key role in successful EA implementation has the potential to change the way EA
implementation is approached in future. It is therefore important in future EA studies to
examine possible people-focused mechanisms in different organisations, as their interplay
between and within the structure and culture ultimately lead them to adopt the EA program
or not. The use of Archer’s morphogenetic approach along with reflexivity theory will further
help develop our understanding of how the key stakeholders (and their power, influence and
interests) may constrain and enable EA implementation. By including reflexivity as an
important mechanism, organisations will be in a better position to understand the role of
people and their interactions with pre-existing structures and cultures operating over
different time periods - reflexivity suggesting that “people” always have the possibility to do
259
otherwise than expected, largely dependent on their personal history and their current
personal projects and ultimate concerns.
7.5 Summary
In summary, the thesis has described how the underlying social realist philosophy and
theories helped to address the research questions. The critical realist methodology and its
most recognised methodological complement, the morphogenetic approach (MA), has added
to our understanding of EA phenomena from a social perspective. This study has developed
a connecting theory by building a theoretical explanation of the role of people as a key
element in EA implementation, notably in the university sector. In linking the critical role of
people to the relations of three overlapping critical realist ontological domains and the
important mechanisms involved in EA implementation, this middle range theory has brought
together two CR premises (the morphogenetic approach and reflexivity theory) in its analysis,
and added to the body of knowledge using an in-depth case study to uncover EA phenomena
and their social implications.
260
CHAPTER 8: CONCLUSION
Implementing the enterprise architecture (EA) program is not just architecting the technical
aspects of Information technology (IT) or redesigning the organisational business
transformation towards change. It has also to do with interpreting the organisational
structural and cultural towards change. EA practices require a widespread support from its
stakeholders and need to be actively engage in the complex social contexts in which they
exist. In an Australian university case sample, the findings show that opposition and conflict
around EA implementation were not about IT but the social complexity surrounding the
architecture that described situations boiled down to people issues. Several interviews were
held with the senior staff responsible for the program. As explained by a senior level person
responsible for the program:
… technology in itself doesn’t solve anything, … [it] is really about people.
[…], in the past [we] haven’t always recognised that, and that’s certainly
been the case I think here when I first came here.
[Implementing the EA program is] just a matter of [people] engagement.
… If I look at the executive, probably fifty-fifty there – fifty percent of them
get [EA], and also if you talk to individuals nobody would deny if you say to
them that you need technology to do what you do. Nobody will argue with
that, but then that understanding of why we’ve got to do enterprise
architecture to make sure that happens is not necessarily there.
Understanding the social context of EA can help an enterprise to understand the real
challenges associated with people issues, not only the structural constraints and enablements
but also the importance of cultural norms within the organisation.
Underpinned by a critical realist perspective, the thesis demonstrates that the MA (Archer,
1995; 1996; 2013a; 2015) is a powerful analytical tool to uncover the hidden mechanisms (the
situational logics of structures and cultures) and social responses that enable successful EA
implementation. The research examines the particular situational logics evident within the
261
University under study and how these provide opportunities and constraints to the
acceptance of EA over time.
EA is a means of enabling informed decision-making on IT-business transformation as well as
ensuring compliance with EA governance. The social complexity surrounding the architecture
is a function of the number of stakeholders involved; the variety of concerns, socio-political
dimension, diversity in their backgrounds, and work culture that brings complex behavioural
attributes into the governing of EA program and its implementation (Op ’t Land et al., 2009;
Bente et al., 2012; Janssen, 2012). Being organisation wide with a strong governance element,
the main changes to University X (UX) consequent from the new EA has significant social
implications and social dependence, as indicated by the senior EA person:
I think the main change will be about the governance and investment in
technology, so who makes the decisions? I think that will be one of the major
changes, which I think needs to be a balance of education, so that’s probably
more the people side – [it] is really educating people about the value of EA
and getting them to understand it and then communicating and keeping the
linkages between the enterprise architecture and the business of the
university. I think that’s really important, but I think the biggest one is who
makes those decisions and how the decisions are made. Because in the past,
decision-making about our investment in technology has been extremely ad
hoc and there’s been no formal process for doing it.
So I think, once again it will probably be people like heads of school, faculty
heads and directors of service centres who in the past may have been able
to make decisions based on their own specific needs in their area.
Governance is part of the implementation because if you haven’t got the
governance, how do you make sure people stay on the path?
If we don’t have governance to get to the endpoint, people will still be
making decisions in silos and you’ll never get to the endpoint, because part
of the EA is to have that roadmap to… well, you’ve got that “this is what we
want to look like and these are things we’ve got to do to get there.” If we
262
don’t totally focus on those things and if we have other people making
decisions that actually could undermine that, then we’ll just be going round
in circles.
…to me, you don’t separate enterprise architecture and governance – one
without the other is useless. So the [Enterprise Business Groups] are part of
the governance framework and that’s probably the major [collaboration]
mechanism to encourage the commitment to enterprise architecture,
supported by our planning prioritisation program to assist with all our
governance processes.
The prevailing situational logic consequent from the interaction at the level of ideas suggest
opportunism (Archer, 1995) – high systemic integration of UX in the face of sandstone
university domination with a correspondingly low social integration where faculties, schools
and centres have no interest in each other. As the senior member points out:
It’s interesting because even though I say it’s an interesting culture in the
university and it is quite different to an organisation’s, there are some
similarities. And one of those is the tension between what is seen as
corporate and what is seen as operations, and in a university, you would see
that as academia and research, and corporate is these people who set all the
rules and all of that sort of stuff. And there’s always that tension, it’s a “them
and us”. How do you take away the “them and us”? That’s why I think the
academic reorganisation [separate to EA program] was a really good move
because that decision-making is being done by the operations of the
university now. There’s some ownership and accountability for those
decisions, so the people who sit in the Operations area can’t say “’they’
made a decision that I don’t agree with, that affects my life’” sort of thing,
because the ‘they’ is them!
The MA analysis suggests that opportunism supported by the EA program required a
communication strategy targeting “all levels of the university’s culture and structure” so that
people understood what EA opportunities could be provided:
263
…we don’t communicate [EA] as talking about enterprise architecture so
much. We do talk about [EA] when we say that’s the thing that underpins
our decisions and all that sort of thing, and we talk about [EA] in plans in
simple terms of saying “we need to build a house, we need a plan. We’re
building a capability here and we need a plan to build a capability.” So we
talk about [EA] in those terms, but I think more clearly understood in terms
of governance and linking the programs that we’re doing to the achievement
of business objectives and the current problems or inadequacies.
The EA was presented and communicated as a tool for achieving organisational business
goals. The EA actually became a collaboration platform for supporting Opportunism – the
situational logic of opportunity. The EA is owned by the business and provides a platform for
collaboration via flexible and open architecture solutions:
One of the difficulties in a university is that there is not a lot of concept about
end-to-end process, so not a lot of concept about where your process goes
across different areas of the university, the siloed areas. And when we map
processes now that’s part of, I guess, our education across the university of
saying that your process actually doesn’t finish here, it continues on, and we
need to think about the impacts of what you’re doing here on here… Forget
about what systems you’ve got, what are your processes? And what data do
you need to capture for compliance purposes or whatever purpose you need
to capture that data for. Then we can look at the gap between what you’re
saying there and what we’ve got in those systems. And that’s like the
blueprint for us to implement whatever we implement in the future. It hasn’t
been driven from the top down [in the past] because there’s still not a lot of
understanding that you need to actually do that. The thinking’s there at the
executive level because they do think about “well, should that be done
across in Student Services?” That sort of high-level thinking’s there, but the
understanding that “really what we should do, we need to sit down and we
really need to map what those processes are across the university, and then
[decide] ...”
264
Reflexivity theory (Archer, 2007; 2010a) was important in attaining knowledge and
understanding about what it is about people’s internal relations that makes EA
implementation succeed. The idea is that people reflexively accept or reject constraints
depending on their personal projects, and that is a big part of social realism. The findings
demonstrate the usefulness of reflexivity theory in explaining how personal projects are
formed and how they mediate the exercise of structural and cultural constraints and
enablements in EA implementation. The senior member agreed with the potential benefits of
reflexivity:
Yes, if we can tap into [reflexivity theory], that makes [EA implementation]
a lot easier. Too much of the autonomous [reflexives] is where we’ll find the
problem.
This thesis offers organisations a means to focus on the deeper issues of EA implementation
programs by understanding the social complexity surrounding the architecture. The university
under study is evidence that recognising enterprise architecture as an organisation-wide
initiative with a strong social context is a necessary approach. As indicated by the senior EA
member at the completion of EA program:
I think that [people] will always be related to that thing we were talking
about before, about universities being a collective of individuals that don’t
necessarily all have the same focus, enterprise focus. When you’re doing
enterprise architecture it’s about the enterprise, so getting people to
understand that the thing I might want to have is not necessarily the best for
the whole university, and sometimes I might have to compromise because
it’s better for the university rather than a benefit for me as an individual. So
I think that will be our ongoing biggest challenge.
EA is intimately linked to the prevailing situational logic both in terms of the final product (the
role of governance) and in terms of the process of first implementation. It can only succeed if
the prevailing logic is understood and taken note of in the program. This thesis suggests that
for EA to be successful it must become a platform for supporting the particular situational
logic that is in place. EA cannot hope to change a situational logic since that is the
265
consequence of an analytical history of cultural and structural interaction. Therefore, for EA
to be successful it must perhaps be in line with the prevailing logic – in the case of this thesis
the importance of linking to opportunism is argued. It is argued that UX’s success of EA both
in terms of the implementation and in terms of the ongoing governance role was largely
because it supported opportunism. The situational logic, along with the MA and reflexivity
provided a powerful means for explaining the complex adoption process.
266
REFERENCES
Alwadain, A. S. (2014). A Model of Enterprise Architecture Evolution. Retrieved January 2016
from Queensland University of Technology:
http://eprints.qut.edu.au/71204/1/Ayed%20Said%20A_Alwadain_Thesis.pdf
Alwadain, A., Fielt, E., Korthaus, A., & Rosemann, M. (2013). A critical realist perspective of
enterprise architecture evolution: Preliminary findings. In Proceedings of the 24th
Australasian Conference on Information Systems, (p.
http://eprints.qut.edu.au/65202/). Melbourne, Australia.
Anderson, P., & Backhouse, G. (2009). Unleashing EA: Institutional Architectures and the
value of joined up thinking. Retrieved March 2014 from JISC Enterprise Architectures
Group Pilot – Intelligent Content Ltd: Technology & Standards Watch (TechWatch):
http://www.jisc.ac.uk/techwatch
Archer, M. (1979). Social Origins of Educational Systems. London, UK: Sage.
Archer, M. (1988). Culture and Agency: The Place of Culture in Social Theory. Cambridge, UK:
Cambridge University Press.
Archer, M. (1989). Theory, culture and post-industrial society. Sociologia: Revista di Scienza
Sociali, 1 n.s.
Archer, M. (1995). Realist social theory: the morphogenetic approach. Cambridge, UK: Press
Syndicate of the University of Cambridge.
Archer, M. (1996). Culture and agency: The place of culture in social theory, Revised Edition.
Cambridge, UK: Press Syndicate of the University of Cambridge.
Archer, M. (2003). Structure, Agency and the Internal Conversation. Cambridge, UK:
Cambridge University Press.
Archer, M. (2007). Making Our Way Through The World: Human Reflexivity and Social
Mobility. Cambridge, UK: Cambridge University Press.
Archer, M. (2008). The Internal Conversation: Mediating Between Structure and Agency [Full
Research Report: ESRC End of Award Report, RES-000-23-0349]. Swindon, UK: ESRC.
Archer, M. (2010a). Routine, Reflexivity, and Realism. Sociological Theory, 28(3), 272-303.
Archer, M. (2010b). Reflexivity. Sociopedia.isa, 1-13.
Archer, M. (2013a). Social morphogenesis. Netherlands: Springer Netherlands.
267
Archer, M. (2013b). Reflexivity (updated version). Sociopedia.isa: DOI:
10.1177/205684601373, 1-14.
Archer, M. (Ed.) (2014). Late Modernity: Trajectories towards Morphogenic Society.
Lausanne, Switzerland: Springer International Publishing Switzerland.
Archer, M. (Ed.) (2015). Generative Mechanisms Transforming the Social Order. Coventry,
UK: Springer International Publishing Switzerland.
Archer, M., Bhaskar, R., Collier, A., Lawson, T., & Norrie, A. (1998). Critical realism: Essential.
London, UK: Routledge.
Astbury, B., & Leeuw, F. L. (2010). Unpacking Black Boxes: Mechanisms and Theory Building
in Evaluation. Journal of Evaluation, 31(3), 63-381.
Bass, T., & Mabry, R. (2004). Enterprise architecture reference models: A shared vision for
Service-Oriented Architecture. In Proceedings of the IEEE – MILCOM (pp. 1-8).
Monterey, CA, USA: IEEE.
Benson, R., Bugnitz, T., & Walton, W. (2004). From Business Strategy to IT Action. Hoboken,
New Jersey, USA: John Wiley & Sons, Inc.
Bente, S., Bombosch, U., & Langade, S. (2012). Collaborative Enterprise Architecture:
Enriching EA with Lean, Agile, and Enterprise 2.0 Practices. Waltham, United States
of America: Elsevier Inc.
Best, A., Greenhalgh, T., Lewis, S., Saul, J., Carrol, S., & Bitz, J. (2012). Large-System
Transformation in Health Care: A Realist Review. The Milbank Quarterly, 90(3), 421456.
Bhaskar , R. (1989). Reclaiming Reality: A Critical Introduction to Contemporary Philosophy.
Brooklyn, New York, USA: Verso.
Bhaskar, R. (1978). A realist theory of science. Sussex: Harvester Press.
Blom, B., & Morén , S. (2011). Analysis of Generative Mechanisms. Journal of Critical
Realism, 10 (1), Equinox Publishing Ltd.
Bloomberg , J. (2014). Is Enterprise Architecture Completely Broken? Retrieved November
2016 from Forbes: https://www.forbes.com/sites/jasonbloomberg/2014/07/11/isenterprise-architecture-completely-broken/#7a4d7f2a3710
Blundel, R. (2007). Critical realism: a suitable vehicle for entrepreneurship research? In
Neergaard, Helle, & Ulhoi, Handbook of Qualitative Research Methods in
Entrepreneurship (pp. 49–74). Cheltenham: Edward Elgar.
268
Boh, W. F., Yellin, D., Dill, B., & Herbsleb, J. D. (2003). Effectively managing information
systems architecture standards: An intra-organization perspective. In Proceedings of
the International Conference on Information Systems 2003: MISQ Special Issue
Workshop: Standard Making: A Critical Research Frontier for Information Systems
(pp. 171-187). Seattle, USA: MISQ.
Brown, G. (2009). The ontological turn in education: The place of the learning environment.
Journal of Critical Realism, 5(1), 5-34.
Bunge, M. (2003). Emergence and convergence: Qualitative Novelty and the Unity of
Knowledge. Toronto, Canada: University of Toronto Press.
Bygstad, B. (2010). Generative mechanisms for innovation in information infrastructures.
Information and Organization, 20 (2010), 156-168.
Bygstad, B., & Munkvold, B. (2011). In search of mechanisms. Conducting a critical realist
data analysis. In Proceedings of the 32nd International Conference on Information
Systems. Shanghai, China: AIS Electronic Library (AISeL).
Carayon, P. (2006). Human factors of complex sociotechnical systems. Applied Ergonomics
37 (2006), 525–535.
Carlsson, S. A. (2005). Developing Information Systems Design Knowledge: A Critical Realist.
The Electronic Journal of Business Research Methodology, 3(2), 93-102.
Carlsson, S. A. (2006). Towards an Information Systems Design Research Framework: A
Critical Realist Perspective. In Proceedings of the First International Conference on
Design Science in Information Systems and Technology (DESRIST 2006) (pp. 192-212).
Claremont, CA: Claremont Graduate University.
Carlsson, S. A. (2011). Critical Realist Information Systems Research in Action. In M. Chiasson
et al. (Eds.), Researching the Future in Information Systems (pp. 269-284). Turku,
Finland: Springer.
CAUDIT. (2013). The CAUDIT Enterprise Architecture Commons for Higher Education.
Retrieved April 2014 from www.caudit.edu.au: https://www.caudit.edu.au/EAFramework
Čyras, V., & Riedl, R. (2012). Formulating the Enterprise Architecture Compliance Problem.
In Proceedings of the Tenth International Baltic Conference on Databases and
Information Systems (pp. 142-153). Vilnius, Lithuania: CEUR Workshop Proceedings
(CEUR-WS.org) .
Danermark, B., Ekström, M., Jakobsen, L., & Karlsson, J. C. (2002). Explaining Society: Critical
realism in the social sciences. London, UK: Routledge.
269
de Vaujany, F. (2008). Capturing reflexivity modes in IS: a critical realist approach.
Information and Organization, 18(1), 51-72 .
Demetriou, C. (2009). The Realist Approach to Explanatory Mechanisms in Social Science:
More than a Heuristic? Philosophy of the Social Sciences 39(3), 440-462.
DET. (2007). National Code of Practice for Registration Authorities and Providers of
Education and Training to Overseas Students. Retrieved October 2013 from
Department of Education and Training :
https://internationaleducation.gov.au/Regulatory-Information/Education-Servicesfor-Overseas-Students-ESOS-Legislative-Framework/NationalCode/Documents/National_Code_2007.rtf.
Dobson, P. J. (2001a). The philosophy of critical realism—an opportunity for information
systems research. Information Systems Frontiers, 199-210.
Dobson, P. J. (2001b). Longitudinal case research: A critical realist perspective. Systemic
practice and action research, 14(3), 283-296.
Dobson, P. J. (2003). Moving from interpretivism to critical realism in IS research : an
exploration and supporting IT outsourcing example. Retrieved June 2014 from Edith
Cowan University: http://ro.ecu.edu.au/theses/1287
Dobson, P. J. (2009). Critical Realism as an Underlying Philosophy for IS Research. In M.
Khosrow-Pour, Encyclopedia of Information Science and Technology, Second Edition
(pp. 806-810). Hershey, PA, USA: IGI Global.
Dobson, P., Jackson, P., & Gengatharen, D. (2011). Examining Rural Adoption of Broadband –
Critical Realist Perspective. In Proceedings of the 15th Pacific Asia Conference on
Information Systems. Brisbane, Australia: AIS Electronic Library (AISeL).
Dobson, P., Jackson, P., & Gengatharen, D. (2013). Explaining Broadband Adoption in Rural
Australia: Modes of Reflexivity and the Morphogenetic Approach. MIS Quarterly, 37
(3), 965-992.
Doucet, G., Gøtze, J., Saha, P., & Bernard, S. (2008). Coherency Management: Using
Enterprise Architecture for Alignment, Agility, and Assurance. Journal of Enterprise
Architecture, 4(2008), 9-20.
Eason, K. (2014). Afterword: The past, present and future of sociotechnical systems theory.
Applied Ergonomics, 45 (2014), 213-220.
Ertaul, L., & Rathod, V. (2012). The Zachman Framework, the Owner’s Perspective &
Security. In the 2012 International Conference on Security and Management
(SAM'12) (pp. 1 - 6). Las Vegas, USA:
http://www.mcs.csueastbay.edu/~lertaul/SAM9731.pdf.
270
Fleetwood, S. (2005). The ontology of organisation and management studies: A critical
realist approach. Organization, 12 (2), 197-222.
Fleetwood, S. (2013). What is (and what isn’t) critical realism? [PowerPoint Slides, presented
on CESR seminar]. Retrieved September 2015 from UWE Bristol (University of the
West of England):
http://www2.uwe.ac.uk/faculties/BBS/BUS/Research/CESR/What%20CR%20is%20an
d%20is%20not.pdf.
Fox, S. (2009). Applying critical realism to information and communication technologies: a
case study. Construction Management and Economics, 27(5), 465-472.
Franke, U., Hook, D., Konig, J., Lagerstrom, R., Narman, P., Ulberg, J., . . . Ekstedt, M. (2009).
EAF2 – A framework for categorizing EA frameworks. In Proceedings of the 10th ACIS
International Conference on Software Engineering, Artificial Intelligences,
Networking and Parallel/Distributed Computing (pp. 327-332). Washington, DC, USA
: IEEE Computer Society .
García-Ruiz, P., & Rodríguez-Lluesma, C. (2010). ‘Reflexive consumers’: a relational approach
to consumption as a social practice. In M. Archer (Ed.), Conversations About
Reflexivity (pp. 223–242). London, UK: Routledge.
Gartner. (2009). Gartner Identifies Ten Enterprise Architecture Pitfalls. Retrieved June 2013
from http://www.gartner.com/: http://www.gartner.com/newsroom/id/1159617
Gaver, S. B. (2010). Why doesn’t the federal enterprise architecture work? An examination
why the federal enterprise architecture program has not delivered the expected
results and what can be done about it. Retrieved June 2013 from The process
platform for E-Government Switzerland: http://www.echbpm.ch/sites/default/files/articles/why_doesnt_the_federal_enterprise_architectur
e_work.pdf
Gengatharen, D., Standing, C., & Knight, S. (2009). Knowledge Management in an
Organisational Climate of Uncertainty and Change: A Longitudinal Case Study of an
Australian University. In Proceedings of the 20th Australasian Conference on
Information Systems (pp. 1076-1085). Melbourne: ACIS & Monash University.
Godinez, M., Hechler, E., Koenig, K., Lockwood, S., Oberhofer, M., & Schroeck, M. (2010).
The Art of Enterprise Information Architecture. Boston, MA, USA: IBM Press (Pearson
plc).
Gravesen, J. K. (2012). Reasons for resistance to enterprise architecture and ways to
overcome it. Retrieved June 2013 from IBM Corporation: The Developer works:
http://www.ibm.com/developerworks/rational/library/enterprise-architectureresistance/enterprise-architecture-resistance-pdf.pdf
271
Gustafsson, J. (2017). Single case studies vs. multiple case studies: A comparative study
(Dissertation). Retrieved from Halmstad University:
http://urn.kb.se/resolve?urn=urn:nbn:se:hh:diva-33017
Hedström , P., & Ylikoski, P. (2010). Causal Mechanisms in the Social Sciences. Annual
Review of Sociology, 36(2010), 49-67.
Hedström, P. (2005). Dissecting the Social. On the Principles of Analytical Sociology.
Cambridge, UK: Cambridge University Press.
Hedström, P., & Bearman, P. (2009). The Oxford handbook of analytical sociology. Oxford,
UK: Oxford University Press.
Horrocks, I. (2009). Applying the morphogenetic approach: outcomes and issues from a case
study of information systems development and organisational change in British local
government. Journal of Critical, 8(1), 35-62.
Janssen, M. (2012). Sociopolitical Aspects of Interoperability and Enterprise Architecture in
E-Government. Social Science Computer Review, 30 (I), 24-36.
Janssen, M., & Klievink, B. (2012). Can enterprise architectures reduce failure in
development projects? Transforming Government: People, Process and Policy, 6(1),
27-40.
Janssen, M., & Madsen, K. (2007). Analyzing enterprise architecture in national government:
The case of Denmark and the Netherlands. In Proceedings of the 40th Hawaii
International Conference on System Sciences. Waikoloa, Hawaii: IEEE Computer
Society.
Johnston, R., & Smith , S. (2008). How critical realism clarifies validity issues in theory-testing
research: analysis and case. In D. Hart, & S. Gregor, Information Systems
Foundations: The Role of Design science (pp. 21-47). Canberra, Australia: ANU E Press
(The Australian National University).
Joint Information Systems Committee [JISC]. (2008). Enterprise Architecture an Introduction.
Briefing Paper. Retrieved March 2013 from JISC Enterprise Architectures Group Pilot
– Intelligent Content Ltd.: www.jisc.ac.uk/enterprisearchitectures
Joint Information Systems Committee [JISC]. (2009a). HE in FE: E-Assessment in Wales.
Retrieved March 2013 from JISC Enterprise Architectures Group Pilot – Intelligent
Content Ltd.: www.jisc.ac.uk/enterprisearchitectures
Joint Information Systems Committee [JISC]. (2009b). Doing enterprise architecture:
Enabling the agile institution. Retrieved from JISC Enterprise Ar March
2013chitectures Group Pilot – Intelligent Content Ltd.:
www.jisc.ac.uk/enterprisearchitectures
272
Joint Information Systems Committee [JISC]. (2010). Enterprise architecture: An
introduction, Version 2. Briefing Paper. Retrieved March 2013 from JISC Enterprise
Architectures Group Pilot – Intelligent Content Ltd.:
www.jisc.ac.uk/enterprisearchitectures
Kaidesoja, T. (2009). Bhaskar and Bunge on Social Emergence. Journal for the Theory of
Social Behaviour, 39(3), 300-322.
Kawulich, B. B. (2005). Participant Observation as a Data Collection Method. Retrieved
November 2016 from Forum: Qualitative Social Research: http://www.qualitativeresearch.net/index.php/fqs/article/view/466/996
Land, M., Proper, E., Waage, M., Cloo, J., & Steghuis, C. (2009). Enterprise Architecture:
Creating Value by Informed Governance. Berlin, Germany: Springer-Verlag Berlin
Heidelberg.
Lange, M., & Mendling, J. (2011). An Experts’ Perspective on Enterprise Architecture Goals,
Framework Adoption and Benefit Assessment. In Proceedings of the 15th IEEE
International Enterprise Distributed Object Computing Conference Workshops (pp.
304-313). Helsinki, Finland: IEEE Computer Society.
Lankhorst, M. (2009). Enterprise Architecture at Work: Modelling, Communication and
Analysis, Second Edition. Berlin, Germany: Springer-Verlag Berlin Heidelberg.
Leist, S., & Zellner, G. (2006). Evaluation of current architecture frameworks. In Proceedings
of the 2006 ACM symposium on Applied computing (pp. 1546-1553). Dijon, France:
ACM Digital Library.
Lindlof, T., & Taylor, B. (2002). Qualitative Communication Research Methods, Second
Edition. Thousand Oaks, CA.: Sage Publications.
Lohe, J., & Legener, C. (2013). Overcoming implementation challenges in enterprise
architecture management: A design theory for architecture-driven IT management
(ADRIMA). Information Systems and e-Business Management, 12(1), 101-137.
Luckett, K. (2012). Working with ‘necessary contradictions’: a social realist meta-analysis of
an academic development programme review. Higher Education Research &
Development, 31(3), 339-352.
Madsen, K. (2006). Enterprise Architecture Implementation and Management: A Case Study
on Interoperability. In Proceedings of the 39th Hawaii International Conference on
System Sciences (pp. 1-10). Kauai, HI, USA.: IEEE Computer Society.
Manzo, G. (2010). Analytical Sociology and Its Critics. European Journal of Sociology
(Archives Européennes de Sociologie), 51(1), 129-170.
273
Martin, R., Purao, S., & Robertson, E. (2009). Evolving enterprise architecture. In Proceeding
of the 15th International Conference on Concurrent Enterprising (ICE): ICEIMT
Workshop at ICE 2009 - Next Generation Enterprise Architecture (pp. 1-12). Leiden ,
the Netherlands: ICE.
Mason, K., Easton, G., & Lenney, P. (2013). Causal Social Mechanisms; from the what to the
why. Industrial Marketing Management, 42 (2013), 347-355.
Mayntz, R. (2004). Mechanisms in the Analysis of Social Macro-Phenomena. Philosophy of
the Social Sciences, 34(2), 237-259.
McCarthy, R. (2006). Toward a Unified Enterprise Architecture Framework: An Analytical
Evaluation. Issues in Information Systems, VII(2), 14-17.
Mingers, J. (1995). What is the distinctive nature and value of IS as a dicipline. Systemist,
17(1), 18-22.
Mingers, J. (2004). Real-izing information systems: critical realism as an underpinning
philosophy for information systems. Information and Organization. 14. , 87-103.
Mingers, J. (2011). Explanatory Mechanisms: The Contribution of Critical Realism and
Systems Thinking/Cybernetics. Retrieved April 2016 from Kent Business School
(Working Paper Series):
https://kar.kent.ac.uk/26306/1/Explanatory_mechanisms_sent_Web.pdf
Mingers, J., Mutch, A., & Willcocks, L. (2013). Critical Realism in Information Systems
Research. MIS Quarterly, 37(3), 795-802.
Morén, S., & Blom, B. (2003). Explaining Human Change: On Generative Mechanisms in
Social Work Practice. Journal of Critical Realism, 2(1), 37-60.
Morse, J. (Ed.) (2010). Determining Sample Size. Retrieved from Qualitative Health Research:
http://www.rcsi.ie/files/research/docs/20151204041457_MorseQual-HealthRes2000.pdf
Mueller, T., Schuldt, D., Sewald, B., Morisse, M., & Petrikina, J. (2013). Towards interorganizational Enterprise Architecture Management - Applicability of TOGAF 9.1 for
Network Organizations. In Proceedings of the 19th Americas Conference on
Information Systems (pp. 1-13). Chicago, Illinois, USA: Association for Information
Systems ( AIS ).
Mutch, A. (2007). Reflexivity and the Institutional Entrepreneur: A Historical Exploration.
Organization Studies, 28(07), 1123-1140.
Mutch, A. (2010). Technology, Organization, and Structure—A Morphogenetic Approach.
Organization Science, 21(2), 507-520.
274
Nash, R. (1999). What is real and what is realism in sociology? Journal for the Theory of
Social Behaviour, 29(4), 445-466.
Nuryatno, E. T., & Dobson, P. (2016). Examining the social aspects of Enterprise Architecture
Implementation: A Morphogenetic Approach. In Proceedings of the 26th
Australasian Conference on Information Systems 2015 (pp. 1-10). Adelaide, Australia:
University of South Australia & Cornell University Library: ISBN# 978-0-646-95337-3.
Op ’t Land, M., Proper, E., Waage, M., Cloo, J., & Steghuis, C. (2009). Enterprise Architecture:
Creating Value by Informed Governance. Utrecht, The Netherlands: Springer-Verlag
Berlin Heidelberg.
Oxford Dictionaries. (2015a). sectionalism. Retrieved August 2015 from Oxford University
Press: https://en.oxforddictionaries.com/definition/sectionalism
Oxford Dictionaries. (2015b). diversification. Retrieved August 2015 from Oxford University
Press: https://en.oxforddictionaries.com/definition/diversification
Patton, M. (1990). Qualitative evaluation and research methods. Beverly Hills, CA: Sage
Pawson, R., & Tilley, N. (1997). Realistic Evaluation. London, UK: SAGE Publications Ltd.
Pawson, R., & Tilley, N. (2004). Realist Evaluation: The Magenta Book Guidance Notes.
Strategy Unit Cabinet Office, UK. Retrieved from Community Matters:
http://www.communitymatters.com.au/RE_chapter.pdf
Pawson, R., Wong, G., & Owen, L. (2011). Known Knowns, Known Unknowns, Unknown
Unknowns: The Predicament of Evidence-Based Policy. American Journal of
Evaluation 32(4), 518-546.
Pettersen, K., McDonald, N., & Engen, O. (2009). Rethinking the role of social theory in
socio-technical analysis: A critical realist approach to aircraft maintenance. Journal
Cognition, Technology and Work, 12(3), 181-191.
Raadt, B., & Vliet, H. (2008). Designing the enterprise architecture function. In Proceedings
of the 4th International Conference on the Quality of Software-Architectures, QoSA
2008, LNCS 5281 (pp. 103-118). Karlsruhe, Germany: Springer-Verlag Berlin
Heidelberg.
Raadt, B., Schouten, S., & Vliet, H. (2008). Stakeholder perception of enterprise architecture.
In Proceedings of the European Conference on Software Architecture (ECSA 2008),
Springer LNCS 5292 (pp. 19-34). Paphos, Cyprus: Springer-Verlag Berlin Heidelberg.
Raduescu, C., & Vessey, I. (2008). Causality in Critical Realist Research: An Analysis of Three
Explanatory Frameworks. In Proceedings of the Annual Conference of International
275
Association for Critical Realism (pp. 4-41). London, UK: The International Association
for Critical Realism (IACR).
Roeleven, S. (2010). Why Two Thirds of Enterprise Architecture Projects Fail: An explanation
for the limited success of architecture projects. Retrieved from
http://www.cio.com.au/: http://www.cio.com.au/whitepaper/370709/why-twothirds-of-enterprise-architecture-projects-fail/download/
Ross, J., Weill, P., & Robertson, D. (2006). Enterprise Architecture as Strategy: Creating a
foundation for business execution. Boston, Massachusetts, United State of America:
Harvard Business School Press.
Saha, P. (2009). Advances in Government Enterprise Architecture. New York: Information
Science Reference (IGI Global).
Saha, P. (2013). A Systemic Perspective to Managing Complexity with Enterprise
Architecture. Hershey, PA: Business Science Reference (IGI Global).
Sayer, A. (1992). Method in Social Science: A Realist Approach. London, UK: Routledge.
Schekkerman, J. (2004). How to Survive in the Jungle of Enterprise Architecture Frameworks,
Second Edition. Bloomington, Indiana, USA: Trafford Publishing.
Schekkerman, J. (2009). Enterprise architecture tool selection guide version. 5. Retrieved
February 2013 from http://www.enterprise-architecture.info/:
https://www.enterprisearchitecture.info/documents/Enterprise%20Architecture%20Tool%20Selection%20G
uide%20v6.3.pdf.
Scherer, S., & Wimmer, M. (2011). Reference framework for E-participation projects. In the
Proceedings of the Third IFIP WG 8.5 international conference on Electronic
participation (pp. 145-156 ). Delft, The Netherlands: Springer-Verlag Berlin,
Heidelberg.
Searle , J. (1995). The Construction of Social Reality. New York, USA: Free Press.
Sembiring, J., Nuryatno, E. T., & Gondokaryono, Y. (2011). Analyzing the indicators and
requirements in the main components of EA methodology development using
grounded theory in qualitative methods. In Proceedings of the 2011 Conference on
Interdisciplinary Business Research: Society of Interdisciplinary Business Research (p.
http://dx.doi.org/10.2139/ssrn.1867875). Bangkok, Thailand: SSRN.
Sessions, R. (2007). A comparison of the top four enterprise architecture methodologies: 12
criteria for comparing and evaluating enterprise IT architectural methodologies.
Retrieved November 2012 from ObjectWatch, Inc.: http://msdn.microsoft.com/enus/library/bb466232.aspx
276
Smith, M. L. (2010). Testable theory development for small-N studies: Critical realism and
middle-rangetheory. International Journal of Information Technologies and Systems
Approach, 3(1), 41-56.
Spragg, N. (2015). How Does Enterprise Architecture Help [PowerPoint Slides]. Lecture. Edith
Cowan University.
The Open Group Architecture Framework [TOGAF]. (2011). TOGAF® Version 9.1: Evaluation
Copy. Retrieved December 2013 from The Open Group:
https://www2.opengroup.org/ogsys/jsp/publications/PublicationDetails.jsp?catalog
no=g116
Turban, E., Leidner, D., McLean, E., & Wetherbe, J. (2008). Information Technology for
Management: Transforming Organizations in the Digital Economy, 6th Edition.
Singapore: John Wiley & Sons. Inc.
Urbaczewski, L., & Mrdalj, S. (2006). A comparison of enterprise architecture frameworks.
Issues in Information Systems, 7(2), 18-23.
Van Den Berg, M., & Van Steenbergen, M. (2006). Building an Enterprise Architecture
Practice: Tools, Tips, Best Practices, Ready-to-Use Insights. Dordrecht, The
Netherlands: Springer.
Vandenberghe, F. (2008). Archer's Structure, Agency and the Internal Conversation [Lectures
- Recensions]. Retrieved January 2016 from Journal du MAUSS:
http://www.journaldumauss.net/?Structure-Agency-and-the-Internal
Wad, P. (2001). Critical realism and Comparative Sociology (Paper presented in the 5th
Annual IACR Conference, 17-19 August 2001, Roskilde University, Denmark).
Retrieved July 2016 from www.criticalrealism.com:
http://www.criticalrealism.com/archive/iacr_conference_2001/pwad_crcs.pdf
Wan, P. Y. (2012). Analytical Sociology: A Bungean Appreciation. Science and Education,
21(2012), 1545–1565.
Ward, J., & Peppard, J. (2002). Strategic Planning for Information System: Third Edition.
Chichester, West Sussex, England: John Wiley & Sons, Inc.
Watkins, J. (1998). Information Technology, Organizations and People: Transformations in
the UK Retail Financial Services. London: Taylor & Francis Ltd.
Wegmann, A., Kotsalainen, A., Matthey, L., Regev, G., & Giannattasio, A. (2008). Augmenting
the Zachman Enterprise Architecture Framework with a Systemic Conceptualization.
In Proceedings of the 12th IEEE International EDOC Conference. München, Germany:
IEEE.
277
Wright, D. (2013). UniAttend: A process automation case study. Retrieved August 2013 from
Leonardo Consulting:
http://hosted.verticalresponse.com/1463407/c67ea782d1/577600059/96d1dc6609/
#UniAttend
Wynn, D., & Williams, C. (2008). Critical realism-based explanatory case study research in
information systems. In Proceedings of the 29th International Conference on
Information Systems (pp. 2-19). Paris, France: Association for Information Systems
(AIS).
Wynn, D., & Williams, C. (2012). Principles for Conducting Critical Realist Case Study
Research in Information Systems. MIS Quarterly, 36(3), 787-810.
Ylimäki , T., & Halttunen, V. (2006). Method engineering in practice: A case of applying the
Zachman framework in the context of small enterprise architecture oriented
projects. Information Knowledge Systems Management 5 (2005/2006), 189–209.
Zachariadis, M., Scott, S., & Barrett, M. (2013). Methodological implications of critical
realism for mixed-methods research. MIS Quarterly. 37(3), 855-879.
Zachman , J. A. (1987). A framework for information systems architecture. IBM Systems
Journal, 26(3), 276-292.
Zachman, J. A. (2008). The Zachman Framework. Retrieved November 2016 from Zachman
International - Enterprise Architecture: https://www.zachman.com/about-thezachman-framework
Zarvi´ci, N., & Wieringa, R. (2006). An integrated EA framework for business-IT alignment. In
proceedings of the CAISE*06 Workshop on Business/IT Alignment and
Interoperability (pp. 262-270). Luxemburg, Germany: BUSITAL '06.
278
Keep reading this paper — and 50 million others — with a free Academia account
Used by leading Academics
Deniz Yonucu
Newcastle University
Eszter Banffy
Hungarian Academy of Sciences
Noel B. Salazar
KU Leuven
Jonathan DeVore
University of Cologne
