Darsoe 2001 Innovation-Making

Lotte Dars0
Innovation in the Making

1. edition 2001 Samfundslitteratur
1. edition reprint 2002 Samfundslitteratur
Samfundslitteratur 2001
Cover: Torben Lundsted
Typeset: Narayana Press
Print: Narayana Press, Gylling, Denmark
ISBN 87-593-0881-8
Published by:
Samfundslitteratur
Rosenoerns Alle 9
DK-1970 Frederiksberg C
Denmark
Telf: + 45 38 15 38 80
Fax: + 45 35 35 78 22
[email protected]
www.samfundslitteratur.dk
All rights reserved.
No parts of this book may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, recording, or by any information
storage or retrieval system, without permission in writing from the publisher.
Contents
Preface 13
Acknowledgements 17
Chapter 1: Innovation in the making: Introduction 21
Introduction to the chapter 26
Innovation 28
Industrial challenges 29
Product development cycles 30
How does innovation start? 31
Why groups? 31
Why is it interesting? 32
Summary: Why study Innovation in the Making? 33
Innovation in the making: the first research question 33
Processes of innovation 34
The working model 35
Positioning the book 37
Summary: what is 'Innovation in the Making'? 38
Objectives 38
The first objective 39
The second objective 39
The third objective 43
Summary: How is 'Innovation in the Making' examined? 43
The preject - a nascent research area 44
What this book is not about: Delimitation 44
What this book is about: Reading guide 47
Chapter 2: Industry and organization 51
Megatrends 52
Innovation policy 54
The Innovation War 55
Danish industry 56
The pharmaceutical industry. . . . . . . . . . . . . . . . . . . . . . . . . .. 56
Merger-mania 57
Patents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 58
5
To merge or not to merge - or demerge? 58
Product development cycles 60
Biotechnology - a new industry. . . . . . . . . . . . . . . . . . . . . . .. 61
Tendencies of the pharma and biotech industries ... . . . . . .. 62
Future challenges ofthe pharma and biotech industries. . . . .. 63
The company: Novo Nordisk A1S . . . . . . . . . . . . . . . . . . . . . . .. 65
Brief history of Novo Nordisk A1S 66
Strategy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 71
A 'war of position' versus a 'war of movement' . . . . . . . .. 72
Exploration versus Exploitation . . . . . . . . . . . . . . . . . . . . .. 76
Leadership and Innovation Management 79
A dual organizational structure 82
Future challenges 83
Concluding remarks 85
Chapter 3: Theoretical foundation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 89
Paradigms and models 89
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 89
Scientific framework 90
Innovation process models 93
Introduction to this section . . . . . . . . . . . . . . . . . . . . . . . .. 93
Definition of innovation 93
Innovation process models 94
Innovation stage models . . . . . . . . . . . . . . . . . . . . . . . . . .. 95
Innovation relational models 97
Emergent models of innovation processes 100
Concluding remarks _ 106
Group development 108
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 108
Group characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 109
Definitions and delimitations 111
Homogeneous versus heterogeneous groups __ ., 111
Group management and leadership. . . . . . . . . . . . . . . . .. 115
Group process models .... . . . . . . . . . . . . . . . . . . . . . . . .. 120
Concluding remarks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 124
Knowledge Creation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 126
Introduction 126
From Mode 1 towards Mode 2 Knowledge Production 126
Classification of knowledge 128
What is a problem? 141
6
What is an idea 143
Knowledge creation process models. . . . . . . . . . . . . . . . .. 151
Concluding remarks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 155
Creativity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 156
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 156
Lines of research 157
Some characteristics of creativity 157
The characteristics of the creative person , 159
Process models of Creativity 166
Creativity and group interaction 168
Crystallization 169
Innovative crystallization 172
Communication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 174
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 174
Delimitation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. 174
Definition 175
Shannon & Weaver's 'Tube' model of communication 176
Developing the model of 'Genuine Communication' . . . . .. 177
Genuine communication 178
Collective monologue 179
Context 180
Discussion 183
Listening or hearing? 186
Discussion 188
The message 188
Discussion _. . . . . . . . . . . . .. 190
Antagonistic dialogue 191
The preject-project model 196
The preject 196
Concluding summary 199
Chapter 4: Methodology 203
Outline of the chapter 204
The nature of the problem . . . . . . . . . . . . . . . . . . . . . . . . . . .. 204
The inner circle: the individuals. . . . . . . . . . . . . . . . . . . . .. 205
The middle: the group. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 206
The outer circle: the organizational context. . . . . . . . . . .. 206
Outside the circle: the environment 207
7
The case study method .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 208
Research questions and objectives 208
The propositions 210
The unit of analysis 211
The logic linking data to propositions 212
Criteria for interpreting the findings 212
Triangulation 212
Prospective versus Retrospective studies 213
Clinical research and ethnography 216
Tension between roles 218
Bias 219
Classification of Data 220
Criteria for research evaluation 223
Chapter 5: Case study overview and narrative 226
This chapter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 226
Introduction to case studies 227
The preject-project figure. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 228
Going Narrative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 229
The pilot study 230
From External to Internal 233
First cultural clash 233
Chasing the 'beginnings' . . . . . . . . . . . . . . . . . . . . . . . . . .. 234
How does a reorganization feel? 235
Base-line case study 1: Values In Action 236
Second clash: Don't ask if you don't have the answer! 239
Base-line case study 2 239
Case study 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 241
Third clash: 'Our' world versus 'your' world 243
Case study 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 243
Case study 5: Creative Problem Solving 245
Case study 6: The Mediator Group 246
Identifying Novo Nordisk cultural values 249
Analysis of Novo Nordisk culture 252
General work practice 254
Innovation Coach 255
8
Chapter 6: Communication patterns: Findings and Analysis . . . . . . . . . .. 260
This chapter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 260
Linking data with propositions 261
Establishing construct validity for the communication
framework 261
From data to findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 264
The overall pattern of communication 266
Findings and clarifying questions 267
Data from case study 1 268
Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 272
Discussion: The beginning. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 273
Beginnings in group process models. . . . . . . . . . . . . . . . .. 274
The Butterfly effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 277
Antagonistic dialogue. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 278
Chapter 7: Communication content: Findings and analysis. . . . . . . . . . .. 284
Introduction to the chapter 284
Uncertainty versus certainty 285
Establishing construct validity for the Kubus framework 287
Coding 288
Discussion of the first findings of case study 3 . . . . . . . . . . . .. 291
Data from case study 1 (base-line study) . . . . . . . . . . . . . .. 292
What is a fact? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 293
Information and exformation 294
Structure and leadership . . . . . . . . . . . . . . . . . . . . . . . . . .. 296
Investigating the Questions 297
Classification of questions . . . . . . . . . . . . . . . . . . . . . . . . .. 298
Findings in relation to questions. . . . . . . . . . . . . . . . . . . .. 302
Discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 306
Case study 3: Findings on the processes leading
to crystallization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 308
Framing through multiple perspectives (diversity) 313
Knowledge and Ignorance 315
9
Chapter 8: Conclusion and implications for future research 319
Introduction 319
In what ways can innovative processes be initiated,
supported and managed towards innovative
crystallization in heterogeneous groups? 320
Initiating innovative processes 320
Supporting and managing innovative processes. . . . . . .. 323
In what ways does communication influence knowledge
creation in heterogeneous groups? 326
Communication and Knowledge Creation 326
Synthesis of frameworks and models 328
Synthesis of models 332
Which process models contributed to 'Innovation
in the Making'? 334
Revised process model 336
Implications for future research 339
Questions and ignorance 340
Beginnings 341
The generative potential of antagonistic dialogue 342
Research as a way of life 344
Chapter 9: Recommandations for innovation in groups
and organizations 347
Introduction 347
Target group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 348
Innovative Process Model 351
The four dimensions 352
The four roles 353
Relations 354
Concepts 357
Knowledge versus Ignorance. . . . . . . . . . . . . . . . . . . . . . .. 359
Communication frameworks and meeting forms. . . . . . . . . .. 364
A typology of meeting forms . . . . . . . . . . . . . . . . . . . . . . . . .. 365
'Club' meetings 365
'Reception' type meetings. . . . . . . . . . . . . . . . . . . . . . . . .. 366
'Debate' type meetings 366
'Dialogue' type meetings 367
Innovation in Organizations. . . . . . . . . . . . . . . . . . . . . . . . . .. 368
10
Barriers for innovation in organizations 369
'Umbrella' and emergent strategies 371
Potent Reward and Incentive systems 371
Recommendations for organizations 373
A global Hot-House. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 373
Innovative partnering workshops .. . . . . . . . . . . . . . . . . .. 375
Innovation Cafes 376
A virtuous circle of innovation 376
Appendices to Innovation in the Making 379
Appendix A: CV for Lotte Dars0, Ph.D. . . . . . . . . . . . . . . . . . .. 379
Appendix B: EB Project Overview from 1996-2000 384
Appendix C: Success/Failure Visualization. . . . . . . . . . . . . . . .. 385
Appendix D: Attitude Clarification - 'Firewheel' 386
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 388
Index 404
11
Chapter 1
Innovation in the making
Introduction
John: "I hope you people are sufficiently open minded and don't con-
sider this a waste of time."
Anders: "I would like to see the group work in a very creative and open
atmosphere. I hope we can avoid getting political undertones and a
what's-in-it-for-me atmosphere."
Brian: "It's really important that we have a creative discussion, not try-
ing to just criticize when we get some fancy ideas, but try to flutter
a little bit before we get down to the roots."
Eric: "Yes, open discussions are what we need to make this a success."
Frank: "I think that we should pool our efforts and experience into a
concrete action plan."
Gary: "More should come out of this workshop than another system."
Carl: "We should use different perspectives from diverse markets,
resources and strengths. Difference is not wrong, and you're not
stupid if you come up with a different idea. We need to think
beyond the normal conventions."
Gary: "Since we have a lot of different perceptions of how we deal with
new leads, we should challenge each other's perceptions, and hope-
fully this will result in some kind of structure or overall plan that
we can all buy into."
This was part of the conversation at the beginning of a 2-day workshop with
8 persons of different backgrounds: from different nations (USA) China)
France) Denmark)) different parts of the organization (research) develop-
ment) technical service) marketing and sales) and with different functions
and tasks. A heterogeneous group of experts with the challenge oj, within two
days) coming up with a new way for the organization to collect and benefit
from good ideas from employees. The business was industrial enzymes and
the workshop had been organized in order to enhance innovation and long-
21
term business development. The setting was a conference room facing the sea
at the Hvidere mansion on a beautiful autumn morning of 1997. For more
than 50 years Hvidere was a diabetes hospital, but from the early 1990s it
has functioned as an international conference mansion for the company.
Anders: "In our part of the organization we have accepted that the
fastest of our ideas come from our customers, and we have actual-
ly assigned 3 of our field managers to use 15% of their time on new
idea development, but then the idea goes into one person's office
and he is then the killer of all ideas, because he is very very very
busy ... If the ideas were channeled into somebody with a mind
that was a little bit more creative, maybe the ideas could be bent
into something, but we are frightened that we have put too many
ideas into the system and that we haven't done our homework well
enough, and we have started to become so self critical that nothing
gets through the system."
Carl: "And when the idea gets to the market, the market person runs
around to do market research, and there it dies."
Dieter: "What is important is that if the idea has the right content,
we'll all listen and do something. It has to be the right message if
you want to create a system where this happens automatically. I
think it is impossible. We all have a lot of good ideas, but if we
should receive and deal with everything, then it will be too much."
John: "If you have an idea it might just be a tiny part of something.
Maybe another person can help if you have the right system."
Dieter: "But that depends on people. You cannot create a system - you
can have a culture, you can develop an attitude in people and make
them give this serious thought."
Anders: "But a lot of the people who send in an idea do not know that
we have had this idea before, and then the person who looks at the
idea at the other end says "Gee! We looked at this 10 years ago!""
The group had been instructed not to come up with solutions right away (as
they would usually do) but instead to use some time to identify the ('prob-
lem". They did this by talking about what was happening locally and about
how things were in their part of the organization. In accordance with the
agenda the group focused on finding the barriers for idea development local-
22
ly. The end result, i. e. the recommendations from the workshop, should meet
these challenges in order to be satisfactory.
Gary: "It's about creating an environment where it is o.k. to take risks.
In an environment with cut-backs, the problem is that we have
become averse to risk."
Dieter: "Is there any difference when we look at the American organ-
ization compared to the Danish organization? I have a feeling that
the American organization is more willing to get good ideas than in
the other part of the organization."
Frank: "It seems that Americans are much more receptive to embrac-
ing new things, like new technology, compared to European organ-
izations. They tend to sit back and wait to see somebody else do it
- they like to wait and see."
Anders: "How do you stop the bad ideas? What is the process?"
Frank: "All the easy apples on the trees have been picked, so technical-
ly you have to figure out a clever way of how to pick the high apples,
so we spend a certain amount of our R&D budget on new things,
so that the commitment is always there to do new things. Respon-
sible risk taking. At our American competitors they actually tried
out some new ideas ... made some feasibility studies. They did not
shoot them down until they had tried them. I think they spend
more of their resources trying out the ideas and less time shooting
them down."
Dieter (ironically): "And what we do is that we set down a committee
that will be discussing the ideas for three years, whereas our com-
petitors have been testing the idea for three years. The cost was the
same."
The group continued to share their knowledge in order to identify the ('prob-
lem". A lot of subjects were covered. Above the themes were culture" and the
general practice of the organization compared to how their competitors
worked. One person knew how the competitors worked as he had been work-
ing for one of them for many years.
Anders: "I still think that a lot of the things have to do with the fact
that the organization has grown too big and the networks do not ex-
23
ist that would allow people to channel new ideas into the system. I
think that if the package is right and if we get the message to the
right people who get enthusiastic, then it will work. So it is finding
these people who can do it for you, without getting it blocked from
bureaucracy."
Dieter: "It is about selling the idea."
Anders: "But you cannot sell it to the Industry Strategy Group
because they are so focused on long-term projects and big indus-
tries and only on focus areas."
Frank: "So what you are saying is that you have to keep things like a
submarine, under water and have some success and then the sub-
marine can surface?"
Dieter: "Still some people have the ability to set the whole organiza-
tion in motion around a stupid idea ... "
At the end of the first day, people were getting frustrated. The group could
still not agree on a shared problem formulation and with everything up in
the air and no common vision, the whole thing seemed rather chaotic. In
fact, there were a lot of disagreements regarding what the real problem was
- and consequently also what kind of solution would be required. During
drinks and dinner people aired their frustrations but also had a lot construct-
ive conversations making use of the occasion to talk about what was hap-
pening in the organization locally.
The next day started with a clarification of the concept of "a peer
review group".
Anders: "What you need is a center of support, not another critical
evaluation based on the fact that it does not work like that in Den-
mark."
Dieter: "This is very fluffy, we walk around like the cat around the hot
porridge, but how will it work if I sit somewhere and then I have
this wonderful idea which I think is revolutionary? Then I send it to
the peer group, and who is that and what kind of people are they,
what will they do with my suggestion?"
John: "In my opinion it should be a mix of people from all over the
organization. It is very important that they have a broad network
and that they have business knowledge."
Carl: "I disagree with you, it should be as local a group as possible ...
because you are gonna get so many ideas that if you have it all cen-
tralized without some prescreening system you could spend your
entire career chasing down silly ideas."
Anders: "But could we point out one or two people in each geograph-
ical area that are part of this prescreening, sparring partners of a
kind, so that you can get it past the managers?"
Gary: "Then you have created another gate to go through."
Frank: "Let's just imagine, just for the sake of argument, just as one
example, let us say that there is a web page, where there is a box.
You type your idea in the box, you have a box where you type your
initials, and then a number of buttons. Is this a textile opportunity,
animal feed, detergent? Those buttons are linked to mail buttons
for key people in those areas, so when you type in your idea, you
type in your initials, you hit the button, that opportunity message
goes to whoever was the support person or sounding board."
Anders: "But can you actually write in the Intra-web?"
Carl: "Yes, you can, you can put links ... "
At this point the participants were bending eagerly towards each other and
talking rapidly - everybody trying to talk at the same time. The energy was
roaring. The first solution had crystallized. The rest of the day was used to
describe this solution. The whiteboard was covered with circles, squares, words
and arrows in order to see if this solution could really overcome all the bar-
riers that had been discussed earlier.
How did this happen? W'hat were the processes that lead to the crystalliza-
tion? W'hat role did communication play? was the frustration necessary? W'hy
do some groups come up with novel results while other groups come up with
results of the ((lowest common denominator"?
In short, how does new knowledge and new solutions to problems crystallize
and in what ways can these processes be supported or even managed towards
innovative results?
That, in fact, is the content of this book.
Introduction to the chapter
The purpose of the first chapter is to introduce the research problem,
its setting and relevance. The introduction aims at answering the ques-
tions of why, what and how:
Why study 'Innovation in the Making'?
What is 'Innovation in the Making'?
How is 'Innovation in the Making' examined?
How is the book organized?
The first part of the chapter concerns the 'why' and is a figurative
journey starting with a macro-process view where the reader becomes
acquainted with some of the challenges of global knowledge society.
Gradually and logically this leads us to focus on some major indus-
trial and organizational challenges until we reach the particular organ-
ization that forms the context of the study. As we shall see, a major
challenge for pharmaceutical and biotech companies is to make the
discovery phase more effective and efficient. However, little is known
about the processes that precede the development of a new lead, and
even less is known about innovative processes in heterogeneous
groups. Thus, the answers to "why study Innovation in the Making"
can be summarized in the proposition that knowledge creation and innov-
ation in heterogeneous groups is a major competitive and an inimitable
business advantage.
The second part of the introduction concerns the 'what'. Here we
introduce two research questions and a working model, which pro-
vides a first and simple illustration of the research problem. The work-
ing model structures an introductory discussion of the main concepts
and pinpoints the focus of research in order to position the study. In
business organizations the innovation process is generally understood
as phases or stages in a linear development. Interestingly most innov-
ation process models do not include the focus of the present study: the
genesis of innovation. Strong arguments for selecting this focus are
that the genesis of innovation determines the entire innovational develop-
ment of a company (Sundbo, 1998:311).
The third part of the introduction briefly describes 'how' 'Inno-
vation in the Making' will be examined. Here we introduce three
26
objectives: 1) to gain new knowledge about innovative processes, 2)
to develop a special theory, derived from practice, and 3) to develop
a set of recommendations for the organization. We position the study
as distinct from state-of-the-art on innovation research, which con-
sists of mainly retrospective interviews or surveys. Contrary to state-
of-the-art, the present study is a prospective, processual, real-time
and real-life study of heterogeneous groups working in practice. We
briefly outline the fields involved in the theoretical foundation: social
psycholgy, cognitive psychology, organization, and business economy.
Regarding the third objective of developing a set of recommendations
for the organization, seven arguments are given for how organizations
can profit from the findings. It is pointed out that the recommenda-
tions can be generalized to other organizations and to society at
large. To sum up and clarify the focus of the study, some boundaries
and delimitations of the study are outlined towards the end of the
chapter.
Finally, a reader's guide is provided with an outline of the 9 chap-
ters and the Appendices, including (ultra) short summaries.
What you are about to read is the description and development of
a unique study on how new concepts emerge, how new ideas are born,
and how innovation processes can be developed and handled by
improving relations and communication in groups. At the turn of the
millenium project groups and networks constitute a major part of
organizations. Consequently the ways in which groups interact and
produce knowledge, services or products are becoming highly relevant
objects of investigation. In addition, the competency of exchanging
and creating new knowledge in groups has become the key competi-
tive advantage in knowledge society, because it cannot be imitated.
The book is a blend of theory and practice, of analysis and narra-
tive, based on prospective real-time case studies in a Danish pharma-
ceutical and industrial enzymes company. It concludes with a chapter
of practical recommendations for groups working in business com-
panies and organizations. The recommendations derive from the innov-
ative process model, a conceptual model, which is developed through-
out the study. In addition a set of more general recommendations and
practical suggestions are given regarding how organizations can build
a "virtuous circle of innovation".
27
Innovation
"The basic economic resource - 'the means of production' to
use the economist's term - is no longer capital, nor natural
resources (the economist's 'land'), nor 'labour'. It is and will
be knowledge... Value is now created by 'productivity' and 'in-
novation', both applications of knowledge to work."
(Drucker, 1993:7, italics original)
Peter Drucker's term 'Knowledge Society' (1993) most accurately sig-
nals the central aspects of society and industry towards the turn of the
millennium. Knowledge, and particularly new knowledge or innov-
ation, is a major criterion for the survival or growth of companies. In-
novation is defined as the creation and implementation of new know-
ledge into a product or service that yields profit. Three major types of
innovation are: incremental, radical and social innovation (Sundbo,
1998). Incremental innovations are improvements of processes,
products and methods, often found by the technicians or employees
during their daily work
l
. In this category we also find second gener-
ation products, new applications of existing products, and new mar-
kets for existing products. Radical innovation, as the term indicates,
is novel, surprising and different in approach or composition. Scien-
tific inventions belong to this category. Often radical innovation is
based on the convergence of several different kinds of knowledge. The
computer, for instance, was built on the convergence of five different
knowledges
2
However, radical innovations often include high risk and
long lead times. Finally, social innovations spring from social needs,
rather than from technology, and are related to new ways of social
interaction, behavior or function. According to Drucker, social innov-
ations may have an even higher impact than scientific or technological
innovation (1985:126).
However, the contours of a fourth type of innovation begin to
emerge. The significance of small but critical incremental changes, as
in the 'Butterfly Effect' (Gleick 1990), have been identified with
Chaos and Complexity theory. Earlier radical and incremental change
were regarded as opposites, as 'either-or'. The new paradigm of com-
plexity theory facilitates a 'both-and' perspective, which gives rise to a
28
new concept: Quantum Innovation
3
. Quantum innovation refers to
the emergence of qualitatively new system states brought about by
small incremental changes. "Seemingly insignificant changes can
unfold to create large effects" (Morgan, 1997:265). In groups who
operate in 'contexts' of freedom and flux, the process of qualifying
knowledge can unleash innovative potential, in particular if the group
is open and alert towards emergent opportunities in its information
search.
Industrial challenges
The conditions and setting of the present study is a global knowledge
society characterized by constant change, increased complexity and
high uncertainty.
In the 1990's there has been a prevalence of industrial mergers and
acquisitions in the pharmaceutical industry. Big organizations became
bigger while many middle-size companies were taken over or ceased
to exist. At the same time small specialized entrepreneurs emerged,
particularly in the areas of information technology and biotechnology.
Often larger companies bought the knowledge or products of these
small firms or alliances were made, typically between large pharma-
ceutical and small biotech companies.
Novo Nordisk A/S, the company involved in the present study, is
considered a large company in Denmark, but in an international per-
spective the company belongs to the group of middle sized pharma-
ceutical (and industrial enzymes) companies, which could be threat-
ened by hostile takeovers, a subject of frequent discussion in the Dan-
ish press
4
. Novo Nordisk has, however, a construction of ownership
that would make this rather difficult. Until recently (the beginning of
2001) Novo Nordisk has consisted of a Health Care Division, in the
pharmaceutical business, and an Enzyme Business of industrial
enzymes. In 1999 Novo Nordisk announced a demerger in order to
separate these entities (from 2001) under Novo A/S, a wholly owned
new structure. The separation took place on November 14,2000. The
reasoning behind this decision was that separate units would have bet-
ter opportunities for focusing on their special strengths and core com-
petencies} as the development of these two industries at the turn of the
millennium involves more differences than similarities. We will return
to this discussion in chapter 2. The point here is that due to this new
position} the new Novo Nordisk entities} more than before, need
strong profiles of effective and efficient innovation.
Product development cycles
A major reason that big organizations want to become bigger is a wish
to increase Research & Development. The 'Innovation War' describes
a trend of increasing the number of launches of new products every
year and at the same time decreasing the product development life
cycles (von Braun, 1997). In the pharmaceutical industry, life cycles
are presently 10-12 years
5
but a recent survey emphasizes that 'Dis-
covery processes have gone largely unaddressed'6 and need to be cut
in half, i.e. reduced from to 4-7 years to 2-4 years.
In 1995 Novo Nordisk completed a thorough revision and bench-
marking of their Research & Development processes in Enzyme Busi-
ness and in Health Care and managed to shorten the development
phase from 4-7 and 12-14 to 2-5 and 10-12 years, respectively. A stan-
dard project development cycle for the pharmaceutical industry is
illustrated below:
Timeline for Discovery and Development of a new drug
Adding the above figures} the sum is 10 years and 3 months without
counting the time of the pre-project. A pre-project concerns identify-
ing a target and generating a new lead. However} the present study
concerns processes before the pre-project - which are not even in-
cluded in the above timeline - and it is the claim of this book that this
is where the strongest potential for innovation lies. From then on and
through the whole development process} we primarily find incremen-
tal innovation.
How does innovation start?
The challenge would be to investigate how the idea or concept that
forms the basis for such a long life cycle was developed. Surely, one
would believe that a lot of focus, support and attention of the lead-
ers/managers would be dedicated to this phase, and that an abundance
of methods, frameworks and manuals would be available to facilitate
the crystallization of new knowledge. Surprisingly, this is not the case.
What happens before something turns up as a pre-project is rather
obscure} in 'real life' as well as in literature. At best it is described as a
chaotic or turbulent phase with certain individuals as central actors
who make use of informal networks, intra- and inter-organizationally.
Others describe the process as 'skunk work'?, research that is con-
ducted in the spare time or after hours because of the scientist's inter-
est or because it is 'off the agenda' by management.
Why groups?
I Preclinical I Phase 11 Proof of I
concept
36 months
Lead structure
identified
Pre-projectl Discovery
3
36 months
PLPc
Product Lead
Profile Candidate
36 months
Clinical proof
of concept
Phase 2/3
15 months
NDA- Launch
filing
I Approval
"Creative teams solve and identify problems in much the same
way whether they are developing new software, dreaming up a
new marketing strategy, seeking a scientific discovery, or con-
triving a financial ploy." (Gibbons et. aI, 1994: 124)
However, in a world of accelerating competition along with shorter
product development cycles and growing complexity, new ways are
necessary in order to survive and grow. Organizations can no longer
3
1
afford to depend on the invention and innovation efforts of single indi-
viduals, no matter how dedicated, intelligent or creative these may be.
Today the knowledge of one person is not enough. With growing com-
plexity and accelerating change, organizations need innovative teams
who are competent in knowledge creation and who can effectively and
efficiently (and in that order) generate and examine new opportunities
and new leads. Today, "indeed, organizations largely consist of per-
manent and temporary groups" (Gersick, 1988:9). However, most
studies with a group perspective are on decision making and commu-
nication, and the majority of these are conducted in laboratories
(Frey, 1996). Thus, there are good reasons for conducting a study
focusing on groups (as the level of analysis). In the words ofThomas
Scheidel (1986:125):
"Group influences on the innovation process, and any
structured communicative approaches which may facilitate
it, cry out for research attention." (emphasis added)
As a matter of fact, most studies on innovation take a management
position in order to provide managers with systems or tools for man-
aging innovation. The management perspective is evident in a major
study on innovation processes, the Minnesota Innovation Research
Program (MIRP) (Van de Ven & Angle, 1989:4):
"Although such a process theory may never reach the precision
to tell managers exactly what to do and how an innovation will
turn out, it may produce some fundamental 'law of innovating'
useful for describing and explaining a broad class of processes,
sequences, and performance conditions central to the manage-
ment of innovation".
Why is it interesting?
Stimulating innovation is a general objective of the company in order
to be competitive and survive. As stated by a corporate executive vice-
president at a Novo Nordisk international human resources meeting
3
2
in 1997: 'We need to quadruple our product innovation'. When the
proposal for 'Innovation in the Making' was presented to Novo
Nordisk management, they saw a potential gain for the organization in
starting a research project with focus on the early phases of project
development in groups of knowledge workers. The objectives were to
develop a set of recommendations for enabling and supporting in-
novative processes in heterogeneous groups. Could it be that to
achieve the desired 'hard' and competitive results, 'soft' qualities were
needed such as communication, trust and a different approach to co-
operation?
Summary: Why study Innovation
in the Making?
Because knowledge creation and innovative competencies are the
main assets for survival and competition in industries and organ-
izations in a 'global knowledge society'
Because the competency of innovative teams is a non-imitable
competitive advantage of organizations
Because research on knowledge creation and innovation in hetero-
geneous groups is nascent - and in particular communication and
interaction research from a group perspective
Because new approaches are needed to make the discovery phase
of New Product Development more effective and efficient to obtain
business advantage and growth
Innovation in the making: the first
research question
In this book we shall investigate 'Innovation in the making' by concen-
trating on what happens before a concept crystallizes, by studying the
formative processes of problem framing, by studying the communica-
tion, interaction and leadership of groups of knowledge workers, or in
33
sum: by studying the processes that lead to the crystallization of new
knowledge (innovation). How are problems found, opportunities
identified and knowledge created, which can lead to the creation of
new products or new approaches? The main question is then:
it could be because of the immense complexity involved and because
of difficulties with practical accessibility to real time studies in natural
contexts.
Research \Vorking Model
crystallization
question
Q
or
problem
people
~ ~
~
relational processes
conceptual processes
The approach to studying innovative processes is depicted in the
model below. This model is, so to speak, hypothetical - a working
model. Before starting the case studies, I had 5 years of experience
with innovative groups8 and I had developed a theoretical framework,
but I knew little about the 'real' nature of innovative processes. That
was, in fact, the knowledge I hoped to gain. This situation is "charac-
teristic of philosophical inquiry: one wishes to discuss a concept, and
hence must try to make clear what concept is being discussed, but the
purpose of the discussion is to enlighten the meaning of the concept"
(Churchman, 1971 :4). A starting point, however, is still necessary - as
phrased by Van de Ven (1988: 187): the first requirement for studying
processes of innovation is "a clear set of concepts about the object
being studied".
The working model
In what ways can innovative processes be initiated
J
sup-
ported and managed towards innovative crystallization in
heterogeneous groups?
Sundbo (1998:311): "The first phase of the innovation process,
the generation of the idea is of utmost importance since it deter-
mines the entire innovational development of the company. The
contents of the innovation is primarily created in this early
phase."
In the present study we refer to innovative processes as the generation
and development of new ideas and concepts in the formative phase of
projects. Definitions and discussions can be found in chapter 3. As the
focus is on the generation of innovative crystallization, implementa-
tion is not included in this book, except one implementation of an
innovative crystallization, which was studied for 16 months (case
study 6).
In the present study the initial phases have priority. It is the actual
genesis of innovation that is the focus. Flyvbjerg (1992: 19) has pointed
out that the decisive activities for planning projects take place before
objectives, plans and policies, i.e. in the genesis.
"At a time when so much attention is given to innovation and
entrepreneurship, it is rather pathetic that a deep understanding
of the process is lacking." (Teece, 1987:3)
Processes of innovation
If the first phase of the innovation process determines the entire innovation-
al development of the company, we would expect this area to be boom-
ing with research. How come that this is not the case? We propose that
The basic model of innovative processes is intended to provide a
simple illustration of the components of complex dynamic processes.
The goal is, ultimately, innovation. However, in this start up phase of
34
35
problem identification and concept generation, the goal is crystal-
lization of (new) knowledge, as seen in the right side of the model.
Innovative crystallization involves collective transformation of accu-
mulated and integrated ideas into a new conceptualization or proto-
type. Innovative processes are defined in this book as processes that
lead to the creation of new knowledge during problem-forming, prob-
lem-framing and crystallization of opportunities (in heterogeneous
groups). However, this definition is provisional- even hypothetical, as
the main objective of the book is to investigate and define what innov-
ative processes are. This is answered with the research questions in
chapter 8.
At the left side of the model there is a question/problem cloud,
which signifies that often a project starts with something vague, not
yet formed or identified, or with something that needs to be investi-
gated and then reformed and reframed. Below the cloud, still on the
left side of the model, is a group of people. This symbolizes a project
group of knowledge workers with different backgrounds whose task is
to create new knowledge and new leads. Heterogeneous groups re-
fer to diversity in backgrounds, perspective, culture, age, gender, and
organizational or geographical affiliation.
Two levels of analysis are proposed. As depicted in the middle of
the working model the main approach is communication, understood
both as action and as leading to action. Communication influences
two process levels: the conceptual and the relational level. The con-
ceptual processes refer to the topical and the cognitive content of
communication, whereas the relational processes concern the inter-
action between participants, the leadership process and the emotion-
al context (climate). This separation is only analytical. In real life the
processes operate at many levels simultaneously. Problem formation is
generally a rather turbulent process, characterized by iterative cycles
and back loops, rather than linearity.
Knowledge creation in groups entails communication. But how is
knowledge created in groups and in particular how is new knowledge
generated? In what way (if at all) are innovative processes linked to the
quality and type of communication? The working hypothesis for the
research and case studies is that communication is central at all levels.
Therefore we have formed a second research question:
creation in heterogeneous groups of knowledge workers?
Positioning the book
In most literature on project management the pre-phase is as non-
existent as in the pharmaceutical product development cycle displayed
earlier. Classical literature on project management simply presupposes
an idea or describes projects as phases starting with a concept and
moving to development, implementation and termination. The idea or
concept exists a priori to the project and the decision has already been
made regarding goals. In classical project management, projects were
considered tools for innovation and for testing out new strategies. As
a consequence (classical) project management was mainly concerned
with planning. As we shall see in chapter 3, classical process models
for project management, as well as for innovation and group develop-
ment, are primarily stage models with distinct steps.
In newer literature on project organization, the 'planning paradigm'
is opposed to the 'explorative paradigm', which incorporates a more
processual view (Morsing, 1997). Here projects are understood as
'temporary organizations', where individuals are temporarily enacting
a common cause. The present study is more in line with the "tempor-
ary organizing processes, i.e. the deliberate social interaction occur-
ring between people working together to accomplish a certain, inter-
subjectively determined task." (Packendorff, 1995:328).
Some attempts have been made to generate methodology in relation to
systematic innovation in the preject phase. One system, Kubus, was
developed at Copenhagen Business School (Herlau, 1995). Its main
application has been in regional development and in the area of educa-
tion, but in 1996 it was applied in Novo Nordisk research teams by the
author (Dars0, 1997a) and this led to the Ph.D. preceding this book.
In 1996 Dars0 & Herlau coined a new concept in order to discern
the pre-phase from goal-directed projects: the preject. Briefly, a pre-
ject is defined as an open (divergent) process of search for opportun-
37
ities and creation of new knowledge through group interaction and
leadership. Prejects precede projects. Prejects are characterized as
being nonlinear and process driven, whereas projects are more linear
and goal driven. This will be discussed in chapter 3.
Summary: what is 'Innovation in the Making'?
it is an analysis of 'innovative processes' or the generation and
development of new conceptualizations in the formative phases of
projects
it is research dedicated to finding answers to two research ques-
tions:
1. In what ways can innovative processes be initiated, supported
and managed towards innovative crystallization in heteroge-
neous groups?
2. In what ways does communication influence knowledge cre-
ation in heterogeneous groups of knowledge workers?
the focus is on knowledge creation and innovative crystallization
and the level of analysis is group communication
the focus is on the 'preject', a divergent process of search for oppor-
tunities under a structure of leadership
Objectives
1. To gain new knowledge about the processes that lead to
innovative crystallization through case studies in a real-
time and real-life organizational context in interaction with
heterogeneous groups working to create new knowledge
and new leads in the preject phase.
2. To develop a 'special theory'9 derived from practice for in-
itiating, supporting and managing innovative processes to
enhance innovative crystallization in heterogeneous project
groups.
3. To develop a set of recommendations for the organization: a
manual with methods, frameworks and advice for groups
working with knowledge creation and novel approaches.
The first objective
To gain new knowledge about the processes that lead to innovative
crystallization is important and the study is in many ways a pioneer-
ing adventure because of the constellation of 'innovative crystalliza-
tion', 'case studies', 'real time', 'real life', 'interaction', 'heterogeneous
groups' and the 'preject phase'. The formative phase of projects is not
well understood. As described above, it is not on the 'map' - perhaps
because it does not have an authorized 'label'. The 'preject' is an
attempt to label it and define the space. We have outlined a working
model for the study. Real-time studies on innovation processes in
organizations are rare - and even rarer in a group perspective. Thus,
answering the research questions are of major importance for groups,
organizations, industry and society.
The second objective
The second objective involves existing theory. The challenging concepts
are here: a 'special theory', 'derived from practice', 'initiating', 'sup-
porting', 'managing', 'innovative processes', 'innovative crystallization'
and 'heterogeneous groups'.
A 'special theory derived from practice' needs a few words of
explanation. A special theory has a philosophical rationale, an ideal
model based on rules, and recommendations for improving practice
(Bormann, 1996). The philosophical rationale is that innovative pro-
cesses can be initiated, supported and managed towards innovative
crystallization, and that communication is possible. The ideal model
based on rules is developed in chapter 8, and the recommendations
for improving practice can be found in chapter 9. The theoretical
foundation is comprehensive as it draws on theoretical input from
different fields. This is displayed in the figure below.
39
Theoretical contributions from the social sciences
The idea is to encompass and possibly integrate multiple theoretical
contributions that can throw light on innovative processes in preject
groups. As the field is nascent, relevant theory must be searched for in
different areas. This task has a slight resemblance to detective work, a
phrase also used byYin (1994). In this paragraph we will narrow down
the 4 major fields to the relevant sub-fields or disciplines. The real
theoretical framework will follow in chapter 3.
However, none of these theories can be applied directly to the present
study. First of all, most small group research as well as most research
on communication related to decision making in groups has been car-
ried out as laboratory or field experiments in so-called zero-history
groups. In a survey on small group research by McGrath & Altman
(Fiedler: 1967: 17) less than 5% of the 250 studies reported were nat-
ural groups in field settings. Frey (1988) reported that 64% of the
studies were of zero-history groups, 72% used students, 60% were
laboratory experiments, and 72% observed a group only once. In a
similar survey by Frey in 1994 the numbers had barely changed. State
of the art in Small Group Research is therefore as follows (Frey,
1996:39): "Most researchers thus still study a single meeting of zero-
history, college student groups in the laboratory or classroom." It goes
without saying that this is critical. It is tempting to raise the question
whether studies of this kind contribute at all to understanding natural
groups in natural contexts? The parallel to detective work does not
seem that far-fetched when the job is to find the less than 5% of stud-
ies on natural groups that could be relevant - and how many of these
groups have been studied in the preject phase?
small group research [1948] (1973) and the educational innovations of
T Groups (T = training) and 'laboratories' immediately before his
death in 1947. Small group research has made important contributions
to understanding group dynamics (Bradford, Gibb & Benne, 1964),
group climate (Gibb, 1964) and leadership style (Fiedler, 1967).
As for communication in relation to decision making in groups, this
field has grown steadily with the development of three major theoret-
ical approaches in the 1980'es: The functional, the structurational and
the symbolic convergence theory (Frey, 1996).
organization
business
economy
cognitive
psychology
social
psychology
The kaleidoscopic exercise
Social psychology
From social psychology we shall draw on two fields: Small group
research and the study of group communication and decision making.
Interestingly both fields have been inspired by the same person, a pio-
neer, not only in group psychology, but also in the psychology of lead-
ership and action research. Kurt Lewin inspired the development of
Cognitive psychology
When examining knowledge creation, we draw on theories from cog-
nitive psychology. In order to understand how ideas are generated and
problems found, formed and solved, we must gain insight into cogni-
tive processes (Simon, 1978, Hansen 1997), intelligence (including
'emotional intelligence', Goleman 1996), styles of thinking (e.g. diver-
gentJconvergent in Scheidel, 1986), the meaning of tacit knowledge
(Polanyi, 1966) and 'skilled ignorance' (Dars0, 1997c) and, of course,
knowledge creation (Nonaka & Takeuchi, 1995) and creativity
(Maslow, 1970; Edwards, 1987).
Organization
The reason for drawing in the field of organization is the 'structural'
(Giddens, 1984) influence of the organization on groups working with
knowledge creation. Processes of innovation in researcher groups do
not happen in a 'lab' - at least not in the above meaning of Small
Group Research! The organizational impact is reflected in the organ-
izational culture (Schein, 1986, 1994), in networks (Stacey, 1996), in
politics (March & Simon, 1958, Hosking & Morley, 1991) and in
strategy Gohnsen, 1994, Stacey, 1993, Mintzberg, 1989). Even though
these areas must be regarded as somewhat peripheral to the present
study, which is focused on the group as the unit of analysis, the the-
ories are necessary for describing and understanding the direct and
indirect influence of the organizational context. Also the structure of
the organization may have significance for project initiatives.
In relation to innovation, Jim March (1991) pointed out the impor-
tance of a balance between strategies of exploration and exploitation.
This parameter would influence the degree of 'freedom' of researchers
working with innovation.
Business economy
Finally, to complete the circle (or square), business economy includes
the fields of innovation & entrepreneurship, management and strategy.
There are some overlaps with organization, e.g. with strategy. Man-
agement, however, the fastest growing field in business (Micklethwait
& Wooldridge, 1997), is important in relation to: the management of
knowledge (Davenport & Prusak, 1998) and the management ofinnov-
ation projects (Van de Ven & Grazman, 1997) and project groups
(Mikkelsen & Riis, 1992; Packendorff, 1995). This perspective is dis-
cussed in chapter 2.
The third objective
The third objective is important for Novo Nordisk, but will also be
highly desirable for other organizations working with initiatives for
innovation - and indeed for society at large. Is there anywhere where
it is not desired to improve communication in heterogeneous groups
and turn problems into opportunities, leading to new products, ser-
vices or ways of adding value to the world we live in? "Research is
socially desirable precisely because it often generates such widespread
and indiscriminate benefits." (Rosenberg, 1990: 167).
As mentioned earlier, in industry it is of the utmost importance for
organizations to look for new ways to gain advantage in their business.
A more systematic and focused preject phase would serve several pur-
poses:
it would qualify the knowledge creation process by exploiting the
diverse resources of the group and of the environment
it would shorten the early phases of product development and
thereby shorten the whole process
it would make the pre-phase less turbulent
it would allow a more effective screening of new research areas ear-
ly at little cost (opposed to closing costly projects at a later stage)
it would reduce the dependency on specially talented individuals
("idea generators") as a better practice could be applied by groups
in general
the system could be applied in critical phases of product develop-
ment
it would support team work across the organization and strengthen
the working climate and the innovative culture of the organization
Summary: How is 'Innovation
in the Making' examined?
by focusing on 3 objectives in order to develop theory, practice and
concrete recommendations
through research applied in practice
43
as a prospective processual real-time study
as case studies conducted through a case study research design
combined with triangulation
The preject - a nascent research area
Summing up, this book points out a nascent area of research that has
been overlooked and neglected - or almost - nonexistent. The pre-
ject phase encompasses the entire process of problem finding, goal
seeking, knowledge creation and innovative crystallization in hetero- '
geneous groups. The preject phase is important because it is where
the seeds of innovation are sown and cultivated. The aim of this
study is to identify and examine processes that help or hinder the
seeds in sprouting and growing. It goes without saying that if the
quality of the seed of innovation is determined in the preject phase
- whether by group relations, communication, knowledge creation,
or by other factors - if the DNA is 'imprinted in the seed' by the
time of crystallization - this has important implications for future
research.
What this book is not about: Delimitation
"On those remote pages it is written that animals are divided
into (a) those that belong to the Emperor, (b) embalmed ones,
(c) those that are trained, (d) suckling pigs, (e) mermaids, (f)
fabulous ones, (g) stray dogs, (h) those that are included in this
classification, (i) those that tremble as if they were mad, (j) innu-
merable ones, (k) those drawn with a very fine camel's hair
brush, (1) others, (m) those that have just broken a flower vase,
(n) those that resemble flies from a distance."
(Borges 1966, p.108) (in Lakoff, 1987:92)
The above quote was invented by Borges, and his point (and Lakoff's)
was to show the extraordinary diversity of human categories found
44
around the world, which may look rather odd, particularly when seen
from a Western perspective. My intention with the above quote is to
point to the oddity of the headline of this paragraph and to ask the
reader not to take it literally, as this would entail a similar diversity
of things that are not included. The idea here is rather to sum up and
clarify the boundaries and delimitation of the study.
It is neither a macro-process, political or sociological study, nor is
its main level of analysis organizational.
In chapter 2 of this book, a macro-process and organizational perspec-
tive is taken, and that is in order to set the societal, business and
organizational context of the case studies. In the rest of the book, the
main level of analysis is that ofthe group. This does not mean that
the organizational perspective is left out. It means that in the case
study analysis the organization is seen from the perspective of the indi-
viduals who work with the creation of new knowledge.
It does not take the individual as its main level of analysis.
Likewise, even though at times some individuals will stand out from
the group, we do not take an individual perspective, which would
include interviews about personal motivation, background, difficul-
ties, likes and dislikes, politics and power games, etc. This book has a
business perspective and is meant to improve or transform practice
on innovation activities in a business organization and as the major-
ity of work in business organizations include groups, we have chosen
a group communication and interaction perspective. Further-
more, it is not a study of homogeneous groups, as in knowledge soci-
ety it is more often the case that groups display high diversity. Thus
we study primarily heterogeneous groups.
It neither takes a management perspective, nor a strategic, or a
power perspective.
A majority of studies of innovation take a management perspective,
obviously because management has a major influence on innovation,
particularly in relation to resource allocation and to strategy. Admit-
tedly, both the top-down management and strategy perspective are
essential for the development of innovation in organizations. However,
45
in knowledge society most managers have had to realize that in many
areas they have less knowledge than their knowledge workers. Thus,
also for managers, it is important to examine from a bottom-up
perspective, how knowledge workers create new knowledge and what
their needs and constraints in relation to management and to a busi-
ness strategy are. We thus primarily adopt a bottom-up perspective,
but do not entirely leave out a top-down perspective.
Regarding power, power has been studied as political science and
sociology from a perspective of power as conflict as well as power as
consensus (Haugaard, 1997: 136-137). In this book we do not take a
perspective of power. But because we cannot exclude power as a force
at play in organizations, we shall regard power as both conflictive and'
consensual and see power as relational and potentially construc-
tive (Foucault, 1979).
The major focus is not on innovation as a result.
It is important to point out that the primary focus and the unit of
analysis is innovative processes. Certainly, we are interested in
innovative processes because they eventually lead to innovation. Thus,
our interest in the result, which we describe as innovative crystalliza-
tion in prejects, is primarily to study the processes in real-time, which
provide the result. The challenge with working in real-time is its
'openness', that we do not know whether the process is successful un-
til it ends.
It is not a retrospective study based on interviews; nor is it a survey
based on questionnaires. It is not a laboratory or experimental
study with students (as most group studies), and it is not based
solely on observation.
The study is prospective instead of retrospective. It is a real-time,
forward-looking, open, processual, practical, 'hands-on' study of
knowledge workers. In my role as researcher, I deliberately shift
between observing, participating, and directing the group sessions in
order to gain knowledge of innovative processes, regarding how they
are best initiated, supported and managed towards innovative crystal-
lization.
In SU1n, this is a 1nicro-process social psychological study in a
group-co1n1nunication and interaction perspective. The unit of
analysis is innovative processes in the genesis of innovation
(the preject).It is a prospective, real-ti1ne case study with het-
erogeneous groups of knowledge workers in a natural, organ-
izational context interacting with the researcher in solving
practical proble1nS and in creating new knowledge.
What this book is about: Reading guide
The book is simple in its construction: Indus-
trial & Organizational
and Future Research

The general structure of the chapters is the following: Each chap-
ter begins with an introduction to the chapter, which outlines the con-
tent, and each chapter ends with concluding remarks. Chapter 3 on
theory has, because of its size, been divided into 7 subsections, each
of which begins with an introduction to the section and ends with con-
cluding remarks. Chapter 9, the distilled research findings, can be read
separately as 'Recommendations for innovation in groups and
organizations'. Since the reader can easily orientate him/herself in
each chapter by reading the introduction, the following outline of the
chapters will be brief.
47
Chapter 1: Why study 'Innovation in the Making'?
Introduction What is 'Innovation in the Making'?
How is 'Innovation in the Making' examined?
How is the book organized?
Introduction to research field, research problem and research questions, industry,
organization and context, objectives, focus, working model, main concepts, unit and
level of analysis, methodology, theory.
Chapter 2: Chapter 2 focuses on the innovational context of the study: the external
Industry and
environment (a global 'Knowledge Society') and the relevant industries
Organization
(pharmaceuticals and biotechnology). A brief history of Novo Nordisk is told.
Strategy and management of innovation are discussed, and the chapter ends by
outlining the contemporary challenges for organizations.
Chapter 3: Chapter 3 provides a state-of-the-art theoretical foundation for positioning the
Theoretical
study. At the same time it discusses and generates theory, models and theoretical
foundation
frameworks for the case study.
I. Paradigms and I. The general paradigmatic shift from Newton to Quantum is discussed and three
models categories are introduced: stage models, relational models and emergent models.
2. Innovation 2. Innovation is defined, and selected process models are discussed. The most
process models relevant models for the present study are Toyota's 'set-based concurrent
engineering', Sundbo's dual strUCture and the M1RP process model. However,
most innovation process models are organizational and are not relevant for groups
working in the preject phase.
3. Group 3. The section on groups consists of clarification of concepts, development of
development frameworks, and a review of selected process models. Apart from providing
background material on group development, the main themes include group/team,
diversity/ similarity, group leadership, and conflict.
4. Knowledge 4. The field of knowledge creation provides higWy relevant frameworks and process
creation models for understanding innovative processes. In this section we provide a meta-
framework for discussing knowing and knowledge. Several classifications of
knowledge are introduced, including discussions on 'tacit knowledge', 'intuition',
information/exformation and 'ignorance'.
5. Creativity 5. Creativity mainly contributes to the present study in relation to delimitation,
differentiation, and substantiation of concepts. The boundaries are clarified
between creativity, innovation and innovative processes. In this section we discuss
several classifications of thinking, which links to the section of knowledge creation,
and most importantly, we discuss and clarify the concept of 'innovative
crystallization' .
6. Group 6. In this section a theoretical framework for group communication is developed,
communication based on three ideal type models: genuine communication, collective monologue,
and antagonistic dialogue. Communication is used in its original meaning of
sharing. The communication models are discussed by relating them to context and
group climate, to listening or hearing, and to the message and rhetoric. The section
concludes with an overview of the communication framework.
7. The preject- The main characteristics of the preject-project model are outlined and the roots of
project model the model are explained as going back to Herlau & Tetzscner's didactic concept
from 1995. The preject-project model will form the basic framework throughout
the book for understanding and discussing innovative processes.
8. Concluding Concluding summary of the major contributions from the above fields.
summary
Chapter 4: Chapter 4 focuses on application in practice. The problem is examined and
Methodology
tentatively illustrated in a model in order to show its complexity. This justifies a
case study research design. I argue for using a prospective instead of a retrospective
research method. Clinical research and ethnography are introduced as the chosen
approaches, which are followed by a discussion on the differences and tensions
between these roles.
Chapter 5:
Case study
overview and
narrative
Chapter 6:
Communication
Patterns:
Findings and
analysis
Chapter 7:
Communication
Content:
Findings and
analysis
Chapter 8:
Conclusion and
implications for
future research
Chapter 9:
Recommendations
for innovation in
groups and
organizations
Appendices:
A: CV of Lotte
Darse
B: Enzyme
Business
Project
Overview
C: Success/
Failure
Visualization
0: Attitude
clarification -
'f>rewheel'
References
Index
The focus of chapter 5 is on praxis and starts by outlining the case-study portfolio.
The chapter is a subjective narrative of what happened and how it happened, what
problems and obstacles I met during my research. The narrative is chronological,
and the case srudies are briefly told including a workshop on Novo Nordisk's
culrural values leading to an analysis of Novo Nordisk's culture and general
practice. The narrative ends with a short description of what happened after the
research project ended: my work as innovation coach.
In chapter 6 the focus is on the relational level of analysis: communication patterns.
Construct validity is developed in order to make the communication framework
operational. The overall communication pattern of case srudy 3 is illustrated in a
graph and the data is triangulated with data from the other case studies. We diSCUSS
the influence of "beginnings" and the potential of conflict at some length and
compare that to theories outlined in chapter 3. We conclude by adding a few
elements from complexity theory and a constructive and a destructive trajectory to
the communication framework.
In chapter 7 the focus is on the conceptual level of analysis, communication
content. Construct validity is developed, and the Kubus framework IS developed
into a coding tool. An overview is provided with the main themes, "plus" words,
metaphors, and "minus" words from case study 3. Data is triangulated between the
case studies and theory in relation to two main discussions: certainty and
knowledge versus uncertainty and ignorance. The difficulty of distinguishing .
between facts and opinions is pointed out. The discussion focuses on the funcuon
of questions and the influence of framing or reframing the problem. We conclude
by developing a 'Dynamic Knowledge Map' of exploring ignorance and tacll
knowledge, combining it with rhetoric to 'qualified knowledge'.
This chapter forms the conclusion and synthesis of the book: 1) completing the
research cycle by answering the research questions, 2) making a syntheSIS of the
relational and conceprual levels of communication, and 3) integrating some of the
frameworks and models from the book and from theory. The preject-project model
is cognitively transformed into the innovative process model, and it is argued that
this model most accurately represents innovative processes in the preject space of
'Innovation in the Making.' The four main dimensions are: relations and concepts,
ignorance and knowledge.
Finally, three promising areas of future research are outlined: I) the area of
ignorance and divergent/convergent questions, 2) the influence and significance of
beginnings, 3) the generative potential of antagonistic dialogue.
This chapter displays the distilled findings and practical outcome of the book. It is
made in the form of a practical users' guide of do's and don'ts. The manual starts
by clarifying the target group.
The four dimensions of the innovative process model impose a structure on the
manual, and recommendations, frameworks and methods are given for working
with each dimension and its corresponding role in innovative groups: the
innovation gardener is responsible for relations, the innovation conceptualizer for
concepts, the innovation challenger for knowledge, and the innovation jester for
ignorance.
The second part concerns innovation in organizations. First the major barriers are
listed for innovation in organizations and secondly a 'virtuous circle of innovation'
is outlined. This is followed by some general recornnlendations and practical
suggestions for organizational transformation.
A: CV and list of publications by Lotte Darse
B: Novo ordisk Enzyme Business's process model for ew Product Development
C: SuccesslFailure Visualization is mentioned in chapters 5 and 9 and in Case
study 3 and 6. Appendix C is a template of an ideal-type SIFVisualization.
D: Attitude clarification or the 'firewheel' is mentioned in the same chapters as
appendix C. Appendix 0 is an instruction with an illustration. Both were originally
developed by Herlau (1995).
49
Notes
The close link between cooperation and innovation is pointed out by Ole
Fogh Kirkeby (1998:77).
2 The audion tube (a scientific invention); the binary theorem (a major mathe-
matical discovery); a new logic; the design concept of the punchcard; and the
concepts of program and feedback (Drucker, 1985:132).
3 The term 'quantum innovation' was coined by the author. Morgan described
it (1997, 1997:265), but did not give it a name.
4 As for instance in 'Mandag Morgen', a Danish weekly business magazine,
191098
5 EFPIA: The Pharmaceutical Industry in Figures, 1999: 12
6 'Re-inventing Drug Discovery: Issues and Actions in the Quest for Innovation ,
and Productivity', by Andersen Consulting, 1997:6
7 The term 'skunk works' dates back from the US aircraft manufacturer Lock-
heed, where it was used for the design and development of the US's first pro-
duction jet aircraft. Skunk works is now a registered service mark for Lock-
heed (Trott, 1998:210).
8 CV is included in Appendix A
9 A special theory has three main components: a philosophical rationale, an
ideal model based on rules, and recommendations for improving practice
(Bormann, 1996).
5
Chapter 2
Industry and organization
The purpose of this chapter is to examine the 'context' of the book
from an external business and organizational perspective. The focus of
the present study concerns the micro-processes of initiating, supporting
and managing innovative processes in heterogeneous groups.
Thus we will approach the unit of analysis, innovative processes, by
starting with a macro-process global perspective and point out some
general trends in society and industry that influence organizations at
the turn of the millennium, then continue with the pharmaceutical
industry and biotechnology, and, through that, meet the specific
organization, Novo Nordisk. General trends, such as globalization,
technological development, 'knowledge society', the political con-
sumer, and innovation policies, along with the overall change from
material to immaterial values, make innovation the major competitive
factor for firms and organizations who want to survive and grow. How
innovative are organizations in Europe compared to the US and
Japan? How do Danish companies fit in, and what is Danish policy in
relation to innovation?
After this we zoom in on one particular company, Novo Nordisk
A/S, the context of the book. We outline a brief history of one of the
flagships of Danish industry, a middle-sized company in an inter-
national context. The aim of the historical account is to provide the
necessary background information for understanding the present
organization as well as the situation leading up to the present study.
History also explains the special structure of this company, as it (until
November 2000
1
) consisted of two rather distinctive divisions: a
pharmaceutical and an industrial enzymes division.
In the first part of the chapter we try to maintain an external per-
spective on the organization, which is supplemented by an internal
perspective in chapter 5. It is important to know how the press and
external analysts evaluate the organization, in particular because Novo
Nordisk is known to be one of Denmark's most skilled companies
2
28 Andersen Consulting, 1997:8
29 Berlingske Tidende: 030698
35 Numbers from Forrester, in Berlingske Tidende 040299
37 Scrip Magazine, January 1999:3
39 Jan Leschly left SmithKline Beecham in april 2000 and became CEO of Care
Capital LLC, a private equity corporation investing in healthcare.
42 Novo Nordisk: Stock Exchange Announcement No. 21/1999, p. 4
43 From Novo NordiskWebhotel, 160899
44 Novo NordiskAnnual Report 1998
45 The Novo NordiskWay of Management was introduced in 1997 with 10 fun-
damental principles to be followed throughout the organization in order to
ensure productivity and alignment with Novo Nordisk values
46 Unless otherwise stated, the following is based on 'Livet pi! Novo' ("Life at
Novo") fortalt afHelge Richter-Friis, 1991
47 B0rsens Nyhedsmagasin, no. 6, 150399
48 A famous Danish architect
49 Interestingly, Enzyme Business still works in 'Business-to-Business' relations
50 CIMI (Copenhagen International Management Institute) report, Peter
Feldinger, 1997
51 Enzymer - hvor bruges de? ("Enzymes - where are they applied?") Novo
Nordisk, 1995
52 Stock Exchange Announcement No. 21/1999, page 4
53 See the organizational chart e.g. in Berlingske Tidende 220201
54 According to Mandag Morgen, No. 14, 1997, Novo Nordisk invested 15,1 %
of the turnover in 1996, whereas other pharma companies invested from 16,2
to 18,1%. In 1997 Novo Nordisk invested 16,3%, still a little behind the aver-
age of that year of 16,9%, according to Scrip Magazine, Jan. 1999:40
55 B0rsens Nyhedsmagasin, No.6, 1999:44
56 E-mail correspondence of Aug. 1999
57 B0rsens Nyhedsmagasin, No.6, 1999:44
58 BerlingskeTidende: 210201
60 Scrip Magazine, January 1999:72
88
Chapter 3
Theoretical Foundation
"In debating the epistemic status of scientific theories, likewise, it
must be made clear whether one has in mind, say, the mathematical
schema of quantum-mechanical field theory, the populational analysis
of natural selection, the microstructures and mechanisms of molecu-
lar biology, the developmental sequences of cognitive psychology, the
labour theory of economic value, the general regularities of terrestrial
meteorology, or what."1
Paradigms and models
Introduction
This chapter has two main purposes. The first is to provide a state-of-
the-art theoretical foundation, based on four different disciplines from
the social sciences: social psychology, cognitive psychology, economy,
and organization (se p. 40). The intention is to position the study in
relation to the theories and empirical findings relevant for the phe-
nomenon under study, i.e. to establish in what ways the present study
differs from state-of-the-art and in what ways the present study gen-
erates new knowledge.
The second purpose is to identify and develop theoretical frame-
works for understanding and interpreting the empirical findings, as
required by the case study method. Theoretical frameworks and
models are the researcher's 'instruments' as they provide a lens
through which complex phenomena can be understood. The disad-
vantage is that, as cognitive schemas, they may bias the interpretation
of the empirical findings. That is why this study is multi-paradigmatic
and seeks knowledge by applying different paradigms.
'Episteme', 'Techne' and 'Phronesis' were three intellectual virtues
defmed by Aristotle (Flyvbjerg, 1992:71-76). Episteme, as in epistem-
Zohar's Eight Principles of the Old and New Science, 1997 (pp 41-73)
tion and context. A new paradigm (Quantum) has gradually emerged
with the science of complexity and deterministic chaos, which
stresses non-linearity, unpredictability and disorder as the 'normal'
conditions (Gleick, 1990). In relation to the present study, we shall
look into whether this paradigm can improve our understanding of the
processes observed in prejects. The differences between the 'old' and
the 'new' paradigms are displayed below:
ology, concerns universals that are invariable and predictive, such as
Newtonian science or analytic rationality. 'Techne', as in technology,
concerns the concrete that is variable and dependent on the context.
It is about practical, means-end rationality, and concerns production
and arts. Phronesis, or prudence, concerns values, which are relevant
for praxis. It concerns conduct, which is variable, context dependent,
and related to judgment, experience, and choice.
The present study combines multiple perspectives in a real-time
prospective study in practice, a Mode 2 approach to knowledge pro-
duction (Gibbons et aI, 1994). I, the researcher, start out with theo-
retical frameworks, which are based on and generated from existing
theory, i.e. from a paradigm of 'episteme'. This perspective is preva-
lent in the present chapter. Throughout the book, however, the
'techne' perspective is evident because it is a study of problem solving
in practice, i.e. it involves a target that must be achieved, and a target
for which I am also, at least partly, responsible. As for 'phronesis',
questions of judgment and choice, based on experience and sensitiv-
ity to the context, 'phronesis' is ever present, as I work as clinician and
ethnographer in real-time processes. Thus, the present study involves
all three of Aristotle's intellectual virtues.
NEWTONIAN SCIENCE
atomism
determinate
reductive
either-or
certainty
actuality
subject-object split
vacuum
OUANTUM SCIENCE
holism
indeterminate
emergent and self-organizing
both-and
uncertainty
potentiality
participatory universe
quantum vacuum
Scientific framework
According to Kuhn (1970) a paradigm is an exemplary model, a set of
fundamental assumptions that are characteristic for a scientific field in
a specific period of time. In order to reflect the general paradigmatic
shift that is taking place in the social sciences, the scientific framework
of the present study is multidimensional. Many of the theories are
rooted in Newtonian science, some in a constructivist perspective, and
a few in a complexity paradigm. The idea is to investigate in which
ways each paradigm can contribute to our understanding of complex
innovative processes.
Newtonian or natural science is based on rationality and concerns
what can be measured, explained and predicted. The social construct-
ivist paradigm focuses on the subjective origin of making sense of the
world. It deals with interrelations between intelligent agents, organiza-
The major difference is Newtonian linearity versus the non-linearity
of complexity theory. In Newtonian science the scientist claims to be
able to predict effects, movements and reactions because nature is
seen as law-abiding, determinate and controllable, whereas, with com-
plexity theory, events or developments cannot be predicted because
nature is seen as complex, chaotic and indeterminate.
During the study of literature it became evident that a common
denominator for the theoretical fields would be needed, and the unit-
ing key turned out to be process models. Therefore process models
will form the zone of convergence between the different disciplines
and paradigms.
A process model demonstrates the relations between processes and
the elements that may influence the development. A process model is
also a cognitive tool that draws attention to some processes or rela-
tions and thereby tends to ignore others. Having a process model does
to some extent direct research, which is the case in the present study.
In social theory, models often get a 'quasi-ontological' quality as a
9
1
model can create a new way of perceiving the object/process under
study based on new concepts (Yin, 1994:234). This aspect is also
important in this book. One objective is to perceive innovative pro-
cesses in new ways at the conclusion of the text.
In the present study we have developed a framework of three major
categories of process models: stage-models, relational models and
emergent models. Not all process models belong to one category only,
there are, of course, overlaps and approximations.
Stage models are linear and sequential - and thus predictable. In
this perspective innovation is seen as 'an orderly, logical process'
that can be broken down into its component parts (Saren,
1984:21).
Relational models focus on the interfaces between stages or on the
links between people, and between people and context, culture,
organization, and environment. The model is holistic in trying to
provide a general picture of processes of interaction and in trying
to show interrelations, which mutually affect each other.
In emergent process models uncertainty, change, and flux are
central. And so is the potentiality of becoming. Chaos deals with
irregularities, discontinuity, unpredictability and complexity.
Models of emergence place more importance on the context, as the
context plays a major part in changing patterns ('attractor') of
behavior (Gleick, 1990).
In the following we shall review some of the major process models for
innovation/new product development, group development, knowledge
creation, creativity, and communication. Each section will be intro-
duced separately. The chapter ends with a summary in which the pre-
ject-project model is positioned in relation to the reviewed process
models (state-of-the-art). In this chapter the focus is on 'episteme'.
'Techne' is the main perspective in chapter 4 in relation to methodol-
ogy, and the 'phronesis' aspect is involved in chapter 5, the narrative
account of the case studies.
The represented theories and models are not meant to be an extensive
or full account of all process models of the above disciplines, but
rather a way to discuss some similarities and differences, strengths and
weaknesses of the most relevant approaches.
Innovation process models
Introduction to this section
It is natural to start out with innovation process models in a project on
innovative processes, but as we shall see, most of the models presented
in this section are organizational models and do not add much value in
relation to group interaction. Used as descriptive frameworks for
organizational innovation some of these models are, however, enlight-
ening, because a framework provides an opportunity for discussing the
taken-for-granted organizational project models of the specific com-
pany - which can result in some concrete suggestions for improvement.
The roots of these models are mainly economical, as the concept of
'innovation' was coined by Joseph Schumpeter (1934), an Austrian
economist. We take our point of departure in reviews of some general
innovation process models and add other types of models, which are
particularly relevant for the present study: TheTakeuchi & Nonaka 'Re-
lay Race', 'Rugby' and 'Sashimi' models, the Toyota model of ' set-based
concurrent engineering', the 'edge of chaos' model by Stacey, Sundbo's
combined dual organizational structure, and the MIRP meta-process
model. We also discuss the paradigmatic change that is taking place in
project management from stage-models (the project as a tool) towards
relational models (the project as a temporary organization). We con-
clude the section with a discussion and concluding remarks.
Definition of innovation
Before describing process models of innovation, a definition is need-
ed. We use Joseph Schumpeter's original definition as described by
Sundbo (1998: 13):
innovation is an effort of one or ''lore individuals to
create econo''lic profit through a qualitative change.
93
Innovation can be a new product, a new production process, a new
form of organization or management, and a new form of marketing or
general market behavior. Drucker too builds on the theoretical foun-
dation of Schumpeter, but Drucker has added the systematic and
rational aspect which is attractive, particularly from a management
point of view. And perhaps innovation is rational, but does that go for
innovation processes too? We provisionally defined innovative pro-
cesses as processes that lead to the crystallization of new knowledge
through problem forming and problem framing. We do not repeat the
definitions of incremental, radical, social, or quantum innovation, but
refer the reader to chapter 1. Other classifications of innovation are
available, e.g. product, technological, or market innovation, but these'
are not relevant for our study of innovative processes and will not be
discussed.
Innovation process models
"But a good process model of innovation development does
more than simply define its component events; it strings them
together in a particular temporal order and sequence to explain
how and why innovations unfold over time."
(Van deVen & Poole 1995:163)
The following is built on a review on process models by Saren (1984)
and by Trott
2
(1998). Saren (1984:11-12) defines innovation as differ-
ent from invention and diffusion. "Innovation is the process by which
an invention is first transformed into a new commercial product, pro-
cess or service." If we follow Saren's definition, "innovation in the
making" would be a project in the field of invention and only case
study 6 would belong to the field of innovation. The process models
that Saren lists, however, do not have the exact same boundaries, and
he quotes a much more encompassing definition by Rutten (1959) of
innovation covering "the entire process by which 'new things' emerge"
(Saren, 1984:12). Paul Trott (1998:12) also has a broad definition of
innovation (from a management perspective): "Innovation is the man-
agement of all the activities involved in the process of idea generation,
94
technology development, manufacturing and marketing of a new (or
improved) product or manufacturing process or equipment."
Saren (1984: 11) lists 5 types:
Departmental-stage models
Activity-stage models
Decision-stage models
Conversion process models
Response models
Trott adds two more (1998:126):
Cross-functional models (teams)
Network models
Innovation stage models
The three first and the fifth of Saren's categories are stage models.
They have different purposes and different focus, but are all based on
stages of development.
The departmental-stage model is merely descriptive and based
on concrete organizational structures. Trott calls this type of model for
an 'over-the-wall' model, which points to the problematic interface
between departments. This model is becoming outdated as too much
information and knowledge is lost every time a project is thrown 'over
the wall'.
The activity-stage model focuses on the (separate) activities relat-
ed to the innovation process, e.g. idea generation, screening, commer-
cial evaluation, technical development, testing, and commercialization.
While the departmental-stage model merely describes where the new
product is being developed, the activity-stage model focuses on what is
taking place, which activities and tasks are being performed in order for
an idea to be transformed into a product. Advanced activity-stage mod-
els furthermore describe influences from the external environment.
The decision-stage model takes a different perspective by focus-
ing on decisions related to innovation processes. In some models deci-
sions are listed sequentially, this involving decision criteria for each
95
stage of GOINO GO. This way some projects are gradually eliminated
while some survive. In other models decisions are listed according to
key decisions and action points regarding e.g. personnel, marketing,
production process, etc. While the former concerns decision on activ-
ities, the latter concerns decisions related to departments.
The Response model is a stage model developed by Becker and
Whisler in 1967, which focuses on the response of the organization, as
"innovation is represented as the firm's 'response' to some external or
internal stimulus" (Saren, 1984:23). The stages of the model are: Stim-
ulus (to create a new idea), Conception (of the idea), Proposal (of a
project), Adoption (or rejection). What is interesting in relation to this'
book is the focus, the scope and the significance given to the initial
activities as key determinants for the response - and thus for determin-
ing the start of the innovation process. This over-emphasis is, of course,
also its weakness, as the rest of the innovation process is not included.
The innovation stage models described above all have in common that
they are viewed from an organizational perspective. They have mainly
an internal focus, and are concerned with innovation processes inside
an organization. The influence from external sources are mostly
ignored (except in the response model). They also tend to have a man-
agerial or top-down organizational perspective. As this is also the case
with traditional project management, we shall discuss the shift of para-
digm that is taking place in project management before introducing
process models of the relational category.
Project management
Projects started as unique enterprises for testing out new strategies
Gohnsen, 1994), but became established as a more permanent way of
organizing with the matrix organization. Project management thus
started as a functionalist or engineering matter, primarily related to
planning and resource allocation. Thus, the first models were stage-
models. The Project Management Institute, describes four basic
phases of project work (Lundin & Soderholm, 1995): concept, devel-
opment, implementation, and termination. The phases are based on
function, which makes it an activity-stage model. An important
aspect, however, is that certain key variables have to be managed dur-
ing specific periods of the project. Thus, the time aspect is of major
importance in project management, both in classical project manage-
ment and also in the new paradigm of the 'temporary organization',
but in the new paradigm, action (instead of decision making) is one of
the main parameters. Below, the change of paradigm that is taking
place in project management is displayed.
OLD AND NEW PARADIGMS FOR PROJECT MANAGEMENT
~ o
the project as a
research focus
the project as a tool
temporary organization
development plan expectations
t
--r
i
----
implementation control action
t
-- - - - - - - ~ - - - - - - - - - - - - - - - -
termination evaluation learning
Packendorff, 1995:328
The major difference between the two paradigms is the shift from a
functional, mechanistic, elegant, not human design - a prototype of
the 'engineering' culture (Schein 1998) - to a design involving human
agents who act according to expectations and learn (Packendorff,
1995). The shift can also be interpreted as a shift from a product focus
towards a more processual focus, acknowledging that human beings
influence processes and products and that most errors and failures can
be traced back to lack of information, miscommunication, conflict, or
inept management.
Innovation relational models
According to the conversion-process model, innovation is per-
ceived as a process of conversion of input through a 'black-box' to out-
put. This approach avoids breaking down the process into parts or
97
stages and can give some overall insights into what factors influence
innovation, its direction and costs. The black-box is, however,
unknown.
Cross-functional and network models are team-based models, which
means that we move to a different level of analysis, in fact, to the
level of analysis of the present study: the group level. As the network
model, the way it is described by Trott, is based on process models
that have been categorized in the present context as knowledge cre-
ation models (e.g. Nonaka & Takeuchi, 1995), we will leave that for
later.
The idea of cross-functional teams, representing different
departments of the organization during large parts of the project and
particularly through the interfaces, has been implemented in many
organizations. An example of this approach is 'concurrent en-
gineering', which involves all functions from the start and throughout
the project (Trott, 1998:128).
'Relay race', 'Sashimi' and 'Rugby'
The difference between stage models and relational models is articu-
lated by Takeuchi & Nonaki in an article from 1986, where they equal
the traditional sequential model of product development to a 'relay
race', "with one group of functional specialists passing the baton to
the next group" (Takeuchi & Nonaka, 1986: 137). An alternative
(relational) model resembles a 'rugby' approach, "where a team tries
to go the distance as a unit, passing the ball back and forth" (ibid.). In
the cases studied by Takeuchi & Nonaka, a mix of the two approaches
was found: the 'sashimi' system. This originated from a stage model
(relay race), but was changed to overlap like 'sashimi' (slices of raw
fish) on a plate, this way strengthening the interaction between the
project members internally but also with suppliers externally. In
comparison the Honda 'rugby team' consisted of one big team from
start to end. The advantage of the overlap approach is that it "enhances
shared responsibility and cooperation, stimulates involvement and
commitment, sharpens a problem-solving focus, encourages initiative
taking, develops diversified skills, and heightens sensitivity toward
market conditions." (Takeuchi & Nonaka, 1986: 141) The difficulty of
the approach was that it tended to create more tension and conflict in
the group.
Set-based concurrent engineering
Another 'concurrent engineering' model is Toyota's 'set-based concur-
rent engineering' as opposed to a 'point-to-point' design (stage-
model). The point-to-point design, which is common practice, is to
make modifications or improvements in series on one alternative until
'the' solution has been obtained. Toyota has a different approach of
having many alternatives, and as Toyota has been very successful and
fast in its car design, a group of researchers set out to investigate why
this was so. The following paragraph (including all citations) is built
on an article by Ward, et al. from 1995. They found the paradoxical
situation that in spite of their effectiveness, many steps of Toyota's
developmental process seemed highly inefficient. One of the secrets
appears to be that Toyota keeps the design process divergent for as long
as possible: "The second paradox, in brief, delaying decisions, commu-
nicating 'ambiguously', and pursuing excessive numbers of prototypes,
enables Toyota to design better cars faster and cheaper." (1995:44)
A few of the surprising differences between Toyota and US automo-
bile companies were (1995:47-49):
Toyota explores a much larger number of concepts in 1/4 or 1/5
clay models. As expressed by a Toyota general manager of styling,
they "prefer lots of torpedoes to a single sniper bullet."
They delay freezing specifications for body shape. The remark of
the general manager of body engineering was: "The manager's job
is to prevent people from making decisions too quickly."
Giving approximate targets in order to give the suppliers the oppor-
tunity of exploring various alternatives.
"The large number of prototypes and alternatives is not a con-
sequence of rapid changes in the design concept, but an effort
to explore broad regions of the design space simultaneously."
Of major interest for the present study is the deliberate exploration of
multiple opportunities and alternatives in an extended divergent
99
phase and the deliberate postponed decisions. The authors conclude
(1995:58): "In the conventional, point-to-point search, every change
that part of an organization makes may invalidate all previous decisions ...
Conversely, in the set-based approach, all communication describes
the whole set of possible solutions. As the set narrows, the earlier com-
munications remain valid but are supplemented with further, more
precise information."
We shall let the authors conclude with another promising perspective,
which in many ways describe a fully-fledged preject:
"Set-based concurrent engineering, conversely, explores many
concepts in depth and can potentially fmd better solutions based
on radically new concepts. It also allows a company to pursue
radical improvements with a fair degree of safety: if one idea
does not work out, another is likely to". (1995:59)
Emergent models of innovation processes
In the revised reviews, there were no innovation process models of
emergence, except perhaps Nonaka & Takeuchi's knowledge creation
model (1995), mentioned briefly above. The most salient example of
an emergent process model is that of Ralph Stacey (1996), who draws
on the framework of complexity and deterministic chaos. Stacey
describes the processes of creativity and innovation as being 'on the
edge of chaos'. Stacey claims that a space for novelty exists in the ten-
sion between the extremes of stability (ossification) and instability
(disintegration). In a state of 'bounded instability' innovation can
emerge in groups of people. Stacey (1996: 13) writes:
" ... these are the very conditions required for creativity, an excit-
ing journey into open-ended evolutionary space with no fixed,
predetermined destination. The whole universe, it seems, is law-
ful and yet it has freedom of choice. The price for this freedom
is an inability to know the final destination or to be in control of
the journey."
100
Stacey's strong points of are the underscoring of innovation as emer-
gent, as a potential ("archetype") that can be actualized given the right
conditions. A weak point is Stacey's rather vague definitions of creativ-
ity and innovation. According to Stacey, both creativity and innovation
can be potential and actual:
"However, we only apply the labels creative and innovative if the
resulting behavior actually turns out to improve fitness. Creativ-
ity, then is a change in the pattern of symbols in the mind that
eventually turns out to produce behavior that increases fitness,
and innovation is that behavior. Changes in the recessive symbol
system are potential creativity. If these changes replace existing
dominant symbol systems, then performance ofthe current prim-
ary task changes; this is potential innovation. If that innovation
actually improves fitness, we can say that innovation and creativ-
ity have actually taken place." (Stacey, 1996:131-132)
The chaos framework of emergence adds meaning and value to a pro-
cess perspective, but as it appears from the above quotation, Stacey's
definitions of creativity and innovation are not very clear.
Ion Sundbo found empirical evidence for a dual organizational struc-
ture of innovation processes in large service companies. In some
instances the innovation process followed a linear intended trajectory,
close to a stage model (mainly in large top-strategic organizations),
and in other instances it was more complex, non-linear and emer-
gent (mainly in loosely coupled interactive structures) (Sundbo,
1998:370). Furthermore, Sundbo found that the early phases of idea
generation and acceptance would often consist of a collective bottom-
up process starting in the loosely coupled structure. Thus Sundbo
describes a combined relational/emergent bottom-up innovation pro-
cess and a top-down sequential stage model.
The Minnesota Innovation Research Program (MIRP)
The MIRP study, a longitudinal study, examining 14 different techno-
logical, product, process and administrative innovations in both indus-
101
7. Setbacks
Angle, H.L. & Van de Ven, A.H. (1989:666): Key Components of the Innovation Jonrney
A
13. Adoption
...
,
,
,
,
,
,
,
_/
~ 8.
Criteria ~ 14. Termination
shift ",
o \ ~ 5 . Attribution
,
,
't'
6. People
transitions
10. Relationships 11. Infrastructure
with others development
Organization Direction Over Time ~
1. Gestation 2. Shock

10
3
tive and negative. In that sense shock merely indicates a major change
or critical event. In the MIRP this could be a change in leadership, a
product failure, a new disease or even an approach from another com-
pany offering a joint venture. 3. The third period concerns the pro-
posal submitted to resource controllers in order to launch the devel-
opment of the innovation. Here Poole & Van de Ven state (1989:657):
"In the initial stages, when the need for innovation is felt, the innova-
tion concept is emerging, and only a small group supports the innova-
tion, a 'marketplace of ideas' prevails. Events on the idea and people
tracks are the driving force and the innovation path is complex and
disorderly."
Schroeder et aI's 'fireworks' model was developed by Angle & Van
de Ven into the model displayed below.
The development period, as can be seen, consists of the steps 4-11.
Implementation and termination concern steps 12 to15.
try and in the public sector, ran from 1984 to 1990. In that study,
Schroeder, Van de Ven, Scudder & Polley (1989) did a review of pro-
cess models, comparing literature from group development models,
decision process models, organizational planning models, organiza-
tional change and development models, and innovation process
models. None of the reviewed models were found adequate for
describing the developmental patterns of innovation processes in
organizations. The research teams developed a common framework
and a process model, the 'fireworks model', which was grounded on
data and empirical studies. This model was quite different from the
theories described in literature. Schroeder et al. (1989: 132) describe
innovation as "a rather fluid process where an idea seems to start off
with a shock, then proliferates, is subject to setbacks and surprises,
and then links with the old organization along the way."
The innovation process was indeed found to be complex and vari-
able: "We conclude that a much more complicated multiple progres-
sion process of divergence and parallel and convergent streams of
activities occurs in the development of innovations." (Schroeder et aI,
1989: 133)
One of the important aspects of the MIRP is its comprehensiveness
and the researchers' attempt to form a metatheory on innovation pro-
cesses. In outlining their metatheory, Poole & Van de Ven combined
the theories into global and local interacting theories, and furthermore
divided the innovation process into three major periods: Initiation,
development, and termination.
Initiation
Angle and Van de Ven (1989:665) describe the initiation period as
consisting of three parts: Gestation, shocks and plans. 1. Gestation is
described as the period (often lasting three or more years) of 'setting
the stage' for the innovation. Usually this consists of many events
occurring that eventually lead to an innovation being initiated. Ac-
cording to Van de Ven, Angle & Poole (1989:9), this part of the
research was retrospective, i.e. it derived from 'published reports and
other documents, interviews, and questionnaires'. 2. The 'shock' that
seemed to trigger the initiation of an innovation could be both posi-
102
Development
4. The proliferation stage is described as a divergent process of mul-
tiple, parallel and interdependent paths of activities. 5. The fluid par-
ticipation concerns the shifting of individuals who participate only
part-time and enter and leave the project continuously. 6. People tran-
sitions refer to the emotional content or climate of the project group
and to the individual emotional changes from euphoria to frustration
to closure. 7. Setbacks often happen during innovation because of
encounters with unexpected events or problems. 8. Criteria of success
or failure change over time as the environment change in favour or
disfavour of the project.
3
9. Top management influences the project
through four different roles: sponsor, critic, mentor, and institutional
leader. 10. The relationship with others, internal or external units,
influence the project in various ways. 11. Infrastructure development
in relation to the external environment such as an industry or a com-
munity can be influenced by management in favour (or disfavour) of
the innovation being developed.
Termination
The termination period deals with 12. Linking the old with the new,
which can be done through gradual integration or through a more rad-
ical transformation or even replacing the old with the new. 13. The
adoption of the innovation concerns the way that the innovation is
implemented and accepted or rejected. 14. Termination is due to the
innovation having been successfully implemented or to that resources
run out. 15. Attributions about the innovation refers to the fact that
managers make (subjective) attributions about the success or failure
of the innovation, which influence the next innovation project. The
difficulty regarding how individuals make attributions is that we tend
to blame others' failure on them and our own failure on the situation.
The conclusion of one study was (Angle & Van de Ven, 1989:690):
"The evidence indicates that attributing failure to mismanagement
was incorrect, and resulted in making managers the scapegoats for
events beyond their reasonable control."
14
Discussion
The reason for devoting so much attention to the MIRP is that it is
the largest, broadest, longitudinal real-time case study of innovation
processes. As most studies on innovation, it takes the perspective of
management. The present book differs in that perspective and also in
scope, as here the primary perspective is that of group interaction
whereas the management perspective is secondary. Another difference
is the focus on micro-processes in the 'embryonic' innovation pro-
cesses. In the framework of the MIRP 'fireworks' model, the present
(main) study would start after step 3 (the decision to screen oppor-
tunities for starting an innovation) and mainly concentrate on step 4
of proliferation, and on steps 5 and 6. Interestingly, Angle and Van de
Ven describe this part of the process as not very well understood
(1989:672): "As a consequence, after a short initial period of simple
unitary activities, the management of innovation soon lapses into an
effort to direct controlled chaos. This mushrooming of activities over
time appears to be a pervasive but little understood characteristic of
the developmental process." From this we see why the present study
is important and relevant, and distinct from the MIRP. The present
study intends to enlighten this early part of the developmental pro-
cess, the preject. Furthermore, the MIRP does not seem to capture
the crystallization of innovations.
Peter Drucker, the grand old man of innovation, is not included in the
above process models as his theory did not fit any of the categories.
Drucker is primarily concerned with the practical aspects of innovation,
the 'techne', which he presented in his seminal work on 'Innovation and
Entrepreneurship'in 1985. Drucker defines innovation (1985:49):
"Systelnatic innovation therefore consists in the purpose-
ful and organized search for changes, and in the systeln-
atic analysis of the opportunities such changes lnight
offer for econolnic or social innovation."
In his book Drucker gave formulas on how to exploit the seven
sources of innovation (1985). Drucker's contribution is a systematic
approach based on 20 years of practice and the idea that innovation
15
can be taught and learned. Drucker, however, describes knowledge-
based innovation (the seventh source) as highly risky: "Knowledge-
based innovation differs from all other innovations in its basic charac-
teristics: time span, casualty rate, predictability, and in the challenges
it poses to the entrepreneur. And like most 'super-stars', knowledge-
based innovation is temperamental, capricious, and hard to manage"
(Drucker, 1985:126). The risks are highest in the areas of science and
technology, whereas the risk is a lot lower when the knowledge base is
different, as e.g. in social innovations. Finally, according to Drucker,
bright ideas are the riskiest and least successful source of innovative
opportunities with an enormous casualty rate.
Concluding remarks
The innovation process models revised in this section primarily form
a descriptive framework for the overall innovation process in organiza-
tions. The revised models serve as background and furthermore reflect
the way organizations choose to illustrate New Product Development.
The Novo Nordisk pharmaceutical development cycle is a depart-
mental stage model with a sharp division between Discovery and De-
velopment, and Novo Nordisk Health Care works this way with the
problems it entails. The culture and general practice of Discovery and
Development are different and much knowledge is lost when projects
are thrown 'over the wall'. Still, the real life of projects is more dynam-
ic than the model alludes, and procedures have been made to support
the transfer, e.g. with project managers participating in a more over-
lapping 'Sashimi' style. The pharmaceutical business is restricted by
rigid laws and regulations, and the development phases, in particular,
have to follow the steps and procedures of quality and regulations. The
division between Discovery and Development, however, is optional,
and other pharmaceutical companies have chosen more holistic ways
of organizing New Product Development.
The project structure of Novo Nordisk Enzyme Business is some-
what different. There used to be a separation between Discovery and
Development - until the summer of 2000, when it was decided to
make an organizational change and merge them. Enzyme Business is
106
bound by different regulations that are not as rigid in relation to
sequential steps as is the case in the pharmaceutical business, and En-
zyme Business has a much shorter development cycle of 2-5 years. In
Enzyme Business's 'new' project structure (from 1996), a 'concurrent
engineering' style has been chosen. This means that the same project
director will supervise a project all the way through discovery, devel-
opment, production, launch, and product maintenance (see Appendix
B), and that the project groups will have at least one person from each
function participating in the work throughout the project. The other
participants shift with the demands of the project, e.g. with most par-
ticipants from the discovery area when the project is in the discovery
phase. Furthermore the Enzyme Business structure includes decision-
stages involving decision criteria for each stage as well as industry
strategy groups that provide Godfathers/-mothers for each project.
Appendix B exhibits an overview of the Enzyme Business's project
structure before the merge in 2000.
The purpose here was to demonstrate how the models outlined in
this section provide a framework for discussing the organization ofinno-
vation and new product development. With this said, most of the revised
innovation process models are organizational and do not serve the pur-
pose of understanding innovative processes in their early stages. The
most relevant models were the Toyota 'set-based concurrent engi-
neering', Sundbo's dual organizational structure and the Minnesota In-
novation Research Program (MIRP) process model. The first is relevant
because of its focus on divergence and postponed decision making -
and the surprisingly positive results in spite of the apparently inefficient
development process. The second is relevant because its dual structure
reflects how organizations consist oftop-down strategies, objectives and
management interacting with bottom-up entrepreneurial enterprises of
emergent innovative activity. In particular, the latter is relevant, as
according to Sundbo, this is where idea generation and innovative pro-
cesses emerge. Finally, the MIRP process model stands out as the most
holistic and comprehensive process model of all the revised models.
MIRP is interesting because it consists of 14 real-time, longitudinal,
innovation process case studies. More than the others the MIRP pro-
cess model serves the purpose of providing an overall framework for
positioning and discussing the present study. The MIRP, however, does
10
7
not provide a framework for understanding the preject and the process-
es of interaction in heterogeneous groups that lead to innovative crys-
tallization. Thus, we shall examine the development of groups.
Group development
Introduction
The main purpose of this section is to clarify and discuss the concept
'heterogeneous group' (of knowledge workers), and next to that, two
major challenges for groups: leadership and conflict. This section on
group development is important because the present study, contrary
to the majority of studies on innovation, takes a group level of analysis.
As mentioned earlier, most studies on groups have taken place with
'zero-history groups of students meeting once in a lab-type setting'
(Frey, 1996:39). Less than 5% of the groups have been studied in real-
time natural settings - and even less in organizations. Apparently very
few groups (if any) have been studied in real-time in relation to 'innov-
ation in the making' in organizations.
We start with a clarification of the difference between a group and a
team. These terms are often used intermittently but in this book we
argue that there is a difference. Next we discuss homogeneous versus
heterogeneous groups, the study of which is becoming more urgent as
globalization increases. A matrix of perceived similarity/diversity is
introduced which forms four different types of interaction. After this
we turn to group leadership, and discuss Herlau's model of group
work structure and Stacey's 'edge of chaos' model. We end the chap-
ter with a brief description of some of the main process models of
group development and include models that focus on conflict. The
study of conflict is relevant for the present study, as conflict is inher-
ent in diversity. Thus how conflict develops, how it is triggered, and
how the innovative potential of conflicts can become constructive are
some of the questions we want to examine. This will be continued un-
der the section on communication, where it is pointed out that one of
108
the communication models, the 'antagonistic dialogue', potentially
holds a constructive as well as a destructive trajectory. This discussion
is closely related to both of the research questions and will be con-
tinued throughout the book.
Group characteristics
As the concepts of 'group' and 'team' are often used synonymously in
literature and in practice, we need to define them. One feature is
included in most definitions, but with different meanings: interde-
pendence. Kurt Lewin [1948](1973), among others, argued that
interdependence was an important characteristic of groups.
Interdependency can be understood as:
psychological relations (Lenneer-Axelson & Thylefors, 1983)
sharing a common goal (Lenneer-Axelson & Thylefors, 1983;
Sjolund, 1965, Fiedler, 1967)
purpose
4
(Herlau & Darso, 1994; Katzenbach & Smith, 1993;
Buchholz & Roth, 1987)
In the bona fide group perspective interdependence is seen as an
inter-group phenomenon related to permeable boundaries. Ac-
cording to this approach, a group is defined as "patterns of repetit-
ive behaviors that become interdependent and interstructured"
(Putnam & Stohl, 1996: 175). The advantage of this perspective is
the permeability of the boundaries between group and environ-
ment and between groups and other individuals/groups. Putnam &
Stohl criticize the traditional intra-group perspective of most the-
ories for seeing the group as a 'container' with rigid boundaries.
Finally, Edgar Schein (1994:23), known from his clinical work with
organizational culture, points out history as the distinguishing fea-
ture between a group and a gathering of people. According to
Schein a crowd of people turns into a group when there has been
sufficient shared history to create some form of culture.
10
9
Groups and teams
Thus, groups and teams can be seen as stages of a development.
Buchholz & Roth (1987: 15) outline three steps ofteam development:
From a collection of individuals, through 'group' and finally becom-
ing a 'team'. A collection of individuals is, as the term connotes, indi-
vidual-centered and have individual goals. Groups have established
norms for working together and have clarified their purpose. Groups
are leader-centered, as the leader directs, assigns tasks and controls
performance. Teams, on the other hand, are purpose-centered and
share responsibilities and rewards. Mutual accountability is con-
sidered a decisive feature for being classified as a team. .
The most developed definition of groups and teams is seen in the
'team performance curve' by Katzenbach & Smith (1993:91-92)
where distinctions are made between five modes of cooperation.
Katzenbach & Smith mention two conditions for creating a team: a
demanding performance challenge and a shared purpose, similar to
Hamel & Prahalad's concept of 'stretch' (1994) and Collins & Porras'
BHAG's (Big Hairy Audacious Goals, 1994). The five types of coop-
eration are:
1. Working group: characterized by having 'no significant incremen-
tal performance need'. A working group has no real common pur-
pose except sharing useful information for each individual to per-
form better separately.
2. Pseudo-team: not focused on collective performance and using
more energy for group maintenance than on performance itself,
which often results in the 'lowest common denominator'.
3. Potential team: trying to improve performance, but still lacking
clarity about purpose, goals, working approach, etc. Also still not
sharing accountability.
4. Real team: people with complementary skills, clear purpose, goals,
working approach and with mutual accountability.
5. High-performance team: a team of people who are not only mutu-
ally accountable in relation to working skills but also deeply com-
mitted to supporting each other's personal development and
growth.
110
Finally Herlau & Dars0 (1994), in relation to the didactic Kubus
system, describe innovative teams normatively with the following
eight characteristics: 1) a shared sense of purpose, 2) open communi-
cation (in a supportive team-culture), 3) trust and mutual respect
(with constructive feedback), 4) shared leadership, 5) effective working
procedures, 6) building on differences (complementary skills), 7) flex-
ibility and adaptability, 8) continuous learning (reviewing the process).
Definitions and delimitations
We are now able to outline the following definitions. A group is a col-
lection of individuals with a common purpose or goal. The individu-
als of a group are, however, accountable only for their own tasks or
contributions. The group may have a formal leader, no leader or
shared leadership. A
(more than a group) shares information, knowledge, ignorance, lead-
ership, energy and accountability. Interdependence is stronger in a
team, and the 8 characteristics mentioned above generally apply.
There are multiple ways of classifying groups. Fred Fiedler (1967)
distinguished between social groups, therapy groups and task groups.
The main focus in this book is on 'task' groups, which will be used
interchangeably with 'project' groups.
Homogeneous versus heterogeneous groups
"Because businesses are rapidly restructuring around work
teams (versus individual contributors), understanding the
dynamics of diversity within work teams is especially import-
ant." (Sessa & Jackson, 1995: 134)
'Homo' means same, 'hetero' means other. Accordingly, homogeneous
groups are of the same kind (i.e. similar), and heterogeneous groups
are of other kinds (i.e. diverse). In this book we shall use the term
homogeneity interchangeably with similarity, and heterogeneity with
diversity. In a framework by Sessa & Jackson, diversity is conceived
111
similarity
Matrix of Diversity and Similarity
abling for novelty. This may depend, however, on whether the divers-
ity is perceived or exposed in a particular group. If we combine the
author's classification of diversity in groups5 (1. not perceived, 2. per-
ceived but ignored, 3. exposed leading to destructive conflict, 4.
exposed leading to creative conflict) with 'perceived similarity', a cen-
tral concept from Harry Triandis' theoretical framework for diversity
(Triandis, 1995: 17), we obtain the following matrix for describing dif-
ferent types of group interaction.
According to a review by Triandis from 1977, various studies have
shown that we are attracted to those we see as similar to ourselves
(Triandis, 1995:21). Perceived similarity tends to make interaction
smooth and easy, as indicated in the top right box (B). When neither
similarity nor diversity is perceived, this indicates relations that are
temporary and superficial (A). In a 4-hour course with 50 partici-
pants, one would probably focus on the content of the course and not
care much whether the other participants were similar or diverse.
When diversity is perceived, and there seems to be no similarity, most
people get anxious and on guard (C). Culture shock is related to a loss
perceived
Matrix of group interaction
not perceived
superficial smooth
relations and in-group
interaction interaction
A B
C D
on guard dynamic
culture shock tension
conflict on the edge
of chaos
not perceived
perceived
diversity
One of the consequences of globalization is a greater awareness of
diversity, which has effects at all levels of society. In macrolevel studies
of diversity there are two major poles: the 'melting pot' ideology (e.g.
Japan), which tries to preserve an original culture by absorbing differ-
ences through assimilation; and the 'multicultural' ideology (e.g.
Canada), which tries to preserve differences through integration into
a dynamic whole (Triandis, 1995: 14). The first aims at homogeneity,
convergence and consensus, whereas the latter involves heterogeneity,
divergence and multiple voices.
In relation to organizations, Sessa & Jackson (1995: 140) observe:
"Although research and theory based in the horizontal approach
suggest that diversity has a positive impact on performance, diversity
is hypothesized to have the opposite effect on cohesion."
Thus, diversity in groups can be both potentially enabling and dis-
Heraclitus said: "Nature desires eagerly opposites and, out of
them, it completes its harmony, not out of similars."
(from Bennis, Parikh & Lessem, 1994:137)
Diversity - Similarity
both as 'horizontal' referring to characteristics such as race, gender,
age, profession, etc., and as 'vertical' referring to status and power
relations. The first perspective, often assumed by psychologists, is that
'all differences are created equal', whereas the second, 'differences
among team members create a rank ordering', is mostly assumed by
sociologists (Sessa & Jackson, 1995: 140-143). In this book primarily
a 'horizontal' perspective is assumed, focusing on diversity in relation
to professional, cultural, age (both chronological age and length of
employment in the organization), and organizational affiliation. The
'vertical' perspective is not adopted in this book, even though power
relations may have had some influence in the studied groups. Status
could have been considered if the case study groups had been com-
posed of scientists and e.g. technicians or blue-collar workers. As it
was, the groups consisted of mainly white collar knowledge workers.
What is important is the concept of diversity, which we want to exam-
ine a little closer in the following.
112 113
Herlau developed a model of group leadership in order to discuss the
boundaries of Stacey's (1996) model of 'stability - edge of chaos -
instability'. The following is taken directly from Herlau (1997:9):
1 "The project work system is characterized by: The uncer-
tainty is calculated and it is partly transparent. Knowledge
is generally explicit. It is goal oriented and the leadership is
personified - in a project leader. The project has resources
and a time limit.
2 The loosely bonded phase is characterized by: a high level of
uncertainty, turbulence and chaos. Opaque. Knowledge is
prepared explicitly. There are many goals. The leadership is
not clearly defined, everybody is a leader. Neither time nor
resources have been allocated. The process is self-organiz-
ing, which emerges as a result of uncertainty, i.e. chaos.
3 The preject phase is characterized by: a high level of uncer-
tainty, turbulence, chaos, transparency. Knowledge is pre-
pared explicitly. Activity is partly goal oriented (theme-
goal). The leadership is well-defined. Time and resources
have been allocated. The process is moving towards a
leadership- and knowledge-potential system that can be
converted into a project work system."
In the following Herlau's model for group interaction will structure
the discussion. Ideal types allow a stronger distinction between cat-
egories, which further the understanding of guiding principles but
which is, of course, also farther from reality. In real life situations, e.g.
meetings, all three forms may be manifested, interchangeably, without
the participants noticing this.
Type one: the project system
Traditional project management works by the planning-evaluation-
control paradigm, where projects are seen as 'tools' for improvement
and innovation. Just as Taylor divided the work processes into frag-
ments, the planning paradigm of project management has a work
breakdown structure, identifying tasks and resources and arranging
116
these in optimal sequences as in 'Gantt'9 or 'Pert' charts. "'Best
practice' advocates that project managers should adopt the hierarchy
strategy for coping with accountability. Authority resides in the nego-
tiated design and plans." (Kreiner, 1995) ,
If the project manager is completely on top of the task, this may be
the fastest way to achieve the objectives, simply by delegating the tasks
to the participants, and controling that the tasks are carried out on
time. This works best in relation to routine tasks, under circumstances
of clarity and certainty. In such cases group interaction has a quality
similar to the 'smooth in-group interaction' category (B) in the matrix
of diversity and similarity. In a group with an autocratic leader, the cli-
mate will most likely be persuasive with relations of distrust, a good
deal of argumentation and many hidden agendas. The communication
will be mainly defensive. Whether conflict will actually come up, will
depend on what the project manager tolerates. In an autocratic set-
ting, the group will meet on only rare occasions to coordinate their
tasks and not in order to discuss. Whether the project is successful or
not, the reward/blame will go to the project manager.
Groups of knowledge workers working with innovation and new
leads are, however, mostly in non-routine tasks, under conditions of
uncertainty and ambiguity. Under such conditions the autocratic lead-
er-followership can have a different purpose and take the form of a
'specialization strategy' (Stacey, 1996). In the specialization strate-
gy the leader takes on all the anxieties and behaves as a shield for the
group members, at the same time trying to influence them to follow
his ideas and get inspired by the opportunities. This takes a strong and
creative leader, as the leader alone has to be creative and come up with
all the good ideas and new opportunities, and it also takes a well-
functioning leader-followership, but as Stacey points out, this has
proved to be a successful strategy for many organizations (1996: 156).
The problem with project management of the 'ideal type' autocratic
style is the utter dependence on the individual manager and the personal
qualities of that manager. Most managers have had courses on project
management where they have learned, primarily, how to plan and con-
trol tasks and resources. The human side is often left to skills the man-
ager mayor may not possess from experience and personal interest.
117
We have discussed, however, a management prototype. In real life,
project management is dependent on the situation, the participants,
the task, the community of practice, the organizational culture, the
time limit, the attention from managers, power structure and politics,
changes from the external environment - and of course the personal-
ity and style of the individual project manager. Next we shall examine
situations without a leader or manager: the loosely bonded structure
or informal networks.
Type two: Networks
Often groups get together to discuss a subject of common interest
without an appointed leader. Here leadership will shift according to
who takes initiatives. Leadership by profession can work well, if every-
body feels at ease with this. More often than not, however, the partici-
pants experience uncertainty, ambiguity and anxiety, which may result
in superficial relations and interaction (category A in the matrix of
diversity and similarity), as nobody dares to ask questions that could
reveal their ignorance or confusion. Thus, on the surface this type of
interaction would seem democratic, but also prone to manipulation by
strong individuals. According to Ralph Stacey (1996), creativity and
innovation emerge in the zone between stability and instability. The
key control parameter of human systems is anxiety and being able to
'hold' or sustain it, and normally it is the task of the leader to manage
this process: "The paradoxical space for novelty must be occupied for
creativity to occur either by the leader alone or by all the members,
and this remains true no matter how powerful the leader" (Stacey,
1996:159).
Above we described the specialization strategy. In the participa-
tion strategy, as the word indicates, all group members occupy the
space for novelty and take shifts at being 'leaders' when they contrib-
ute with their knowledge or ideas. This way everybody contributes to
the creative process, and at the same time everybody 'holds' the anx-
iety of being at the edge of chaos. The participation strategy is more
probable when the participants know each other. If the group is new,
the network type of interaction most often stays superficial.
118
Type three: Poled leadership
The concept of poled leadership took its beginning in 1989 with the
Kubus framework (Herlau, 1995) and is based on functions by turns.
The idea of poled leadership derives from the Greek word 'polis'
(Argyris & Schon, 1996:9), and signifies an open discussion followed
by a collective decision and action. Thus, polis is a way of organizing,
and corresponds to the original sense of the term democracy.
Poled leadership is displayed in a tension field between two dialec-
tic, yet complementary leadership functions in a group. One role is
task-oriented and the other is relation-oriented. The task-oriented
leader pulls in the direction of outcomes, decisions, and results,
whereas the relation-oriented leader is responsible for the process, the
group climate, the creativity and the social interaction. The group
works in the tension between these two poles. What differentiates this
system from the approaches mentioned above is that the functions are
regularly shifted between the participants through taking turns. This
has the advantage that the functions can be gradually formed accord-
ing to the special human composition of the groups and their wishes
concerning leadership. Some groups are very democratic and thrive
better with consensus decisions, whereas others prefer a more author-
itative style.
The idea of leadership taking place between two poles may create
a 'dynamic tension on the edge of chaos' (category D in the matrix)
and is also found in 'dialectical leadership' (Van de Ven & Grazman,
1997:298) and in 'contention management' (Snyder & Clontz,
.'
1997:68).
Leadability
A major advantage of poled leadership is the shared accountability,
which is formed along with the shift of roles. This is what is meant by
individual or group 'leadability'. The point is that an individual
becomes a lot more leadable (able to be lead, i.e. a good follower),
after having tried to be group leader and having experienced the hard-
ships of leadership - especially if people do not cooperate. 'It is diffi-
119
cult to appreciate the pressures on leaders until you have walked in
their shoes'. Even though the responsibility of leadership is assigned to
the (shifting) leaders, it is important that every group member sup-
ports, participates in and feels responsible for a well functioning lead-
ership. During the process groups really find out that leadership is a
serV2ce.
In the end, it is not a question of having two leaders in a group, but
more a question of building an appropriate leadership process. By
splitting the leadership into two complementary roles, leadership
becomes 'transparent', and genuine learning can take place. Through
that people are encouraged to develop new skills and competencies.
Thus, the system can either be used for action learning/leadership
training or for team/culture building. In both cases, the spin off that is
hoped for, is that afterwards the competent leader can perform both
roles as one person according to the needs of the situation.
Group process models:
Stage models
A good deal of research on groups has been dedicated to the study of
their developmental stages in order to look for a general pattern. In
1977 Tuckman & Jensen made a review on twenty years' literature on
group development and concluded that all groups (therapy, natural,
laboratory, self-study and task groups) go through basically five stages:
Forming, storming, norming, performing and adjourning (Napier &
Gershenfeld, 1989). This is a unitary sequence model, which tries
to explain group behavior through a sequence of predictable events.
The unitary sequence model dates back to Dewey's reflective
thinking modepo of 1910, which inspired the classical phase-model
(l.orientation, 2. evaluation, 3.control) of Bales & Strodtbeck from
1951. The advantage of unitary sequence models is, that if group
behavior could be explained according to a predictable pattern of
behavior, this would make facilitation a lot easier, as each phase could
then ideally be handled according to a normative 'recipe'.
As the preject has been defined as an nonlinear and rather turbu-
120
lent space, it would be counterintuitive to apply a unitary sequence
model approach. "Traditional phasic definitions - which emphasize
isolating a simple set of periods of unified activity - are too general
and too vague to encompass the diversity of group activities. Studies
of group development show a far greater variety of activities than can
be covered by any simple set of phases;" (Poole, 1983b:326). Poole
has criticized classical stage models of group development for being
too general and vague, and too rigid for describing and understanding
the dynamic processes of group development. Based on his empirical
studies, Poole developed a 'multiple-sequence modeI'll, which will be
discussed under relational models.
Group models of conflict
We include conflict models here as most of these started as stage
models. Aubrey Fisher (1970) did a study on group interaction in 10
groups by examining three categories. The third category is relevant
for the present study and had the following dimensions: favorable,
unfavorable, and ambiguous attitudes. Fisher found four distinct
phases of decision making: 1) orientation, 2) conflict, 3) emergence,
4) reinforcement. The orientation phase had more ambiguous and
favorable than unfavorable verbal cues, and here ambiguity was
described as 'tentatively favorable'. In the conflict phase a polariza-
tion of attitudes, disagreements and ideational conflicts were seen in
relation to the decision proposals that emerged. Thus, this phase had
more unfavorable units than favorable and ambiguous attitudes. Fa-
vorable attitudes, however, increased, compared to the orientation
phase, probably due to the polarization. Conflict more or less dissi-
pated during the phase of emergence of the decision, where ambi-
guity seemed to be the prevalent pattern of interaction. In this phase,
however, ambiguity had a different function (of modified dissent),
mediating the change of unfavorable towards favorable attitudes
towards the emerging decision (Fisher, 1971 :63). Fisher noted
(ibid.:64): "But since the dissipation of conflict is gradual and marked
by ambiguity, the question concerning the point in time at which deci-
sions are made must remain unanswered." Thus, the third phase was
labeled 'emergence'. Finally the fourth phase of reinforcement was
121
characterized by more favorable attitudes of interpretation towards
consensus and unity.
Linda Putnam (1986:181-182) distinguishes between three types
of conflict: substantive, affective, and procedural. Substantive conflict
is related to the ideas or the content of the task (corresponding to the
conceptual level of analysis in this book), affective conflict is related to
relations or personality clashes (corresponding to the relational level
of analysis), and procedural conflict concerns meeting procedures,
etc. (i.e. structure or methods in this book).
An interesting perspective is that of deviance. Of course, the label
'deviance' is in itself problematic as it implies conformity or normal-
ity. In a group perspective, however, 'deviant' behavior involves differ-
ent or diverse perspectives, and is therefore relevant to the present
study. Putnam (1986: 188) defines deviance as 'behavior that departs
from the social and task norms of the group' and lists three types of
deviance: role, opinion, and innovative deviance. Here we are mainly
interested in the last two types. Deviance of opinion indicates a differ-
ence or opposition to the overall opinion of a group. When this devi-
ance of opinion manages to break the frames or expand the boundar-
ies of the group's thinking, it is called innovative deviance.
In 1974, Valentine & Fisher made a pilot study (of 6 zero-history
groups) on Verbal Innovative Deviance, building on Fisher's earlier
study and his 4-phase theory. Verbal Innovative Deviance consists of
contradictions, rejections, strong assertions, amplifying disagree-
ments, pointing out new directions, and advocating opposing opin-
ions. Two findings are interesting: Verbal Innovative Deviance was
found to constitute one fourth of all group interactions (i.e. of 6 zero-
history groups). "While VID appears to be quite acceptable during the
Conflict phase and, to some extent during the Emergence phase, such
deviant behavior is probably detrimental to the group process during
the formative stages of Orientation and the final phase of Reinforce-
ment as the group nears consensus." (Valentine & Fisher, 1974:420)
and (ibid.) "Deviance appears to function differently in each succes-
sive phase or group interaction."
Tjosvold (1982) found that confrontation through cooperative
controversy, rather than 'smoothing' (avoiding controversy) or 'forc-
ing' (competitive controversy), was most effective in group decision
122
making and most effective for conversations between managers and
subordinates (Tjosvold, Wedley & Field, 1986).
Finally, Scheidel (1986) proposes that it is the shift from divergent
to convergent thinking that provokes conflict, and that the core prob-
lem concerns judgment and evaluation as this causes defensiveness,
e.g. as described by Gibb (1961).
Putnam concludes (1986: 195): "This review suggests that ineffec-
tive conflict management stems from listening evaluatively and defen-
sively, clinging tenaciously to positions, behaving inappropriately at
particular stages of conflict development, failing to adapt to a partic-
ular type of conflict, relying on "trained incapacity" to determine work
habits, and developing conflict spirals from repetitive power plays. Ef-
fective conflict management, in contrast, entails both procedural and
content matters."
The issue of conflict is relevant when groups meet to discuss urgent
and important matters related to innovation, particularly in heteroge-
neous groups with diverse perspectives. Conflict is usually seen as
destructive, but as we shall see in the following chapter, the conflictive
element does entail a potential for 'creative abrasion' (Leonard-Bar-
ton, 1995), 'constructive controversy' (Tjosvold, et aI., 1986) or a 'con-
structive trajectory' towards innovation (Dars0, 1998).
Relational group models: Continuous models
As alternatives to stage models we find continuous models or Multiple
Sequence models, as the 'three path' model proposed by Poole (1983b).
The three paths focus on: a) task-process activities, b) relations, and c)
a topical focus. To these paths Poole added three types of'breakpoints>l2:
normal breakpoints (topic shifts, natural breaks, planning, etc.), delays
(back-looping or 'comprehension cycles')3) and disruptions (conflict or
failure) (Poole, 1983b:330). The concept of breakpoints facilitates ob-
servation of changes or events that influence group interaction.
Finally, Poole (1989) found three types of activity cycles in his
work: a) a unitary sequence (a linear path), b) a complex cycle (work-
ing in circles, back-looping) and c) solution cycles (focusing on solu-
tions without examining the problem). Poole argues that some courses
12
3
of events are best described in a linear pattern, whereas others need to
be understood as continuous models or cycles.
Poole's framework is interesting as his path approach as well as the
different types of activity cycles are relevant and incorporate multiple
aspects of complex patterns found in the present study. From the
above description it appears how Poole's work has influenced the
development of the MIRP innovation model described earlier.
Emergent group models: Critical event models
A third way of studying group development is through critical event
models, represented by Gersick (1988; 1989). Gersick started in 1988
by observing 8 groups in real time and natural contexts. Through
grounded theory she unexpectedly found that the activities of the
groups did not follow the classical stage models. Instead they pro-
gressed in two rather steady patterns of 'inertia', interrupted by a sud-
den 'revolutionary period', which took place in the middle of the work
period of the project. The revolutionary period was triggered by an
awareness of time and deadlines. In 1989 Gersick repeated her study,
this time with students in a laboratory setting and of a much shorter
duration. Gersick's studies are very well documented. Gersick found
that time turned out to be more decisive than any specific develop-
mental phases. Is it surprising that groups are aware that 'about half
of the time has passed, let's move on'? In relation to 'killing' the tradi-
tional stage model, it is interesting.
Even more interesting was that Gersick found that the working pat-
tern of groups formed within the first minutes of the first meeting and
continued up till the revolutionary period. After that it would change
into a different pattern, which was kept for the rest of the time, or it
would go back to the earlier pattern and continue. We shall return to
the significance of 'beginnings' in chapter 6.
Concluding remarks
Groups form a major part of organizations and society. Therefore it
is important to find out how groups perform and what the challeng-
12
4
es are in relation to group work. It is, however, difficult to study
groups in real-time and real-life - particularly in relation to the early
phases of innovation. In fact, very few studies of this kind have been
made.
With increasing globalization comes an increasing diversity of
society and consequently of working groups in organizations. As we
have tried to demonstrate in this section, the main point, however, is
not diversity per se, but rather the degree of perceived diversity (and
similarity) as shown in the matrix on page 113. Therefore more em-
phasis should be made on getting groups to work with attitudes, com-
munication and relation building.
Another feature of diversity is that it often provokes conflict. In this
section we have provided some 'raw material' for a discussion about
the potential of conflict, by revising process models related to conflict
and 'innovative deviance'. Scheidel's proposition that conflict is bound
to happen when changing thinking mode from divergent to conver-
gent thinking is interesting and will be taken up later.
A challenge in groups is also the question of leadership. This is of
particular interest in preject groups, because the working situation
without a clear goal differs considerably from normal project work.
The image of leadership is undergoing change. Old management
models are not sufficient or even relevant for heterogeneous groups
who work with knowledge creation and novelty. Margaret Wheatley
mentions some of the new metaphors that describe leaders: gardeners,
midwives, stewards, servants, missionaries, facilitators, and conveners
(1999: 165). In this book we have suggested participative or poled
leadership for the early phases of innovation, and it was argued that in
preject groups leadership is a service.
The main purpose of this section was to revise literature on group
development and group interaction as this book operates from a group
perspective. In sum, this section has provided clarification of concepts,
new frameworks, raw material for later discussions and a general back-
ground for understanding group development.
12
5
Knowledge Creation
"We hold that knowledge is a tool of knowing, that knowing is
an aspect of our interaction with the social and physical world,
and that the interplay of knowledge and knowing can generate
new knowledge and new ways of knowing."
(Cook & Brown, 1997:6)
"Whether it is held by a bat or a biologist, knowledge minimizes
an organism's consumption of energy, space, and time for a
given amount of effort." (Boisot, 1998: 11)
Introduction
In this section we combine social psychology with theories from cog-
nitive psychology. The purpose is to clarify, classify and discuss know-
ledge and knowing in theories that relate to or derive from practice.
The American pragmatist, John Dewey's book 'How we think' from
1910 has provided many insights and contributions, which we build
on in this section. His seminal stage-model of reflective thinking has
inspired much later work, practical as well as theoretical. Dewey has
inspired group stage-models (Bales & Strodtbeck, 1951) and action
science (Argyris & Schon, 1996) as well as action learning (Revans,
1991). In this section, in particular Dewey's perspective on the nature
and function of 'problems' and 'ideas' is illuminating. A much later,
but also important, influence on the discussion of theory and research
on knowledge is the Mode 2 approach by Gibbons et. al. (1994), but,
in fact, various valuable frameworks and models are presented in the
following section.
From Mode 1 towards Mode 2 Knowledge Production
The general paradigmatic shift that is taking place at the turn of the
millennium is also reflected in research and 'knowledge production'.
Gibbons et a1. (1994) describe this shift as going from Mode 1 to
126
Mode 2. Gibbons et al. write (1 994:vii) : "A new mode of knowledge
production affects not only what knowledge is produced but also how
it is produced; the context in which it is pursued, the way it is organ-
ised, the reward systems it utilises and the mechanisms that control
the quality of that which is produced."
What distinguishes Mode 2 from Mode 1 is:
Mode 2 is focused on application in practice - rather than within a
particular discipline
Mode 2 is centered on the usefulness for the involved parties and
for society in general - Mode 1 produces knowledge in the absence
of interested parties
Mode 2 is a collective phenomenon with a wider set of criteria -
Mode 1 is an individual matter with criteria of one particular dis-
cipline
Mode 2 is transdisciplinary and heterogeneous - Mode 1 is homo-
genous and disciplinary within one specific scientific community
According to Gibbons et aI, transdisciplinarity has four characteris-
tics:
It involves creativity and frameworks that evolve in the process
It makes a contribution to knowledge, which is developed in the
special context of application and people
The outcome is primarily processual, absorbed by the people
involved and transformed into new knowledge in new applications
and situations
It is dynamic and is not to be evaluated by traditional academia but
rather to be communicated and become parts of new configur-
ations
The present study has many of the characteristics of Mode 2. It con-
cerns problem solving in practice and in real-time, in heterogeneous
groups, creating new knowledge. It is useful for the parties involved -
and for society in general. It is transdisciplinary in the sense that it
combines different disciplines and different fields. There are, of
course, features of Mode 1 in the sense that the research has been
reported and evaluated as a Ph.D. - but then again the committee was
international and included both academia and industry.
12
7
Thus, what we are aiming at discussing in this book is knowledge cre-
ation in practice, the process of knowing and of generating new know-
ledge in interaction with others. We are studying the processes that
lead to innovative crystallization, which refers to the conceptual out-
come of the 'generative dance between knowledge and knowing'
(Cook & Brown, 1997).
At the turn of the millennium, in our global knowledge society,
knowledge is high on the agenda or discourse of media, management
and academia. From the 'learning organization' we have moved to a
new fad of 'knowledge management' with new 'witch doctors'. Lately,
much literature has been published on this matter, as e.g. Nonak;a &
Takeuchi (1995), Krogh, Roos and Kleine (1998) and Davenport &
Prusak (1998), just to mention a few of the more recent ones.
Classification of knowledge
There are various classifications and categorizations of knowledge.
Venzin, von Krogh & Roos (1998) take a strategic management per-
spective in examining the roots of three major epistemologies. Nona-
ka & Takeuchi discuss organizational knowledge creation (1995). Her-
bert Simon applies a cognitivist perspective on individual cognition
(1986). Jerome Bruner examines cognition and linguistics from a
social constructivist view (1986). Max Boisot (1995a) outlines a polit-
ical economy of information and examines the production and
exchange of knowledge.
PPEP Process Model
In this book we apply the framework ofJohn Heron (1981) as it facil-
itates an advanced differentiation of knowledge. According to Heron
there are three kinds of knowledge, which may (or may not) be active
at the same time:
Propositional knowledge
Practical knowledge
Experiential knowledge
128
Propositional knowledge corresponds to Latour's term of 'ready
made science' (1987). This means that it has been established as a
fact, a product, an entity, or a statement about the world. It is not a
concrete or direct part of the world, but it contains information about
the world, as for instance, the law of gravity. We cannot see the law of
gravity, but we can see that it works. This is explicit knowledge (Nona-
ka & Takeuchi, 1995), declarative knowledge (Hansen, 1997), or
knowledge as 'possession' (Cook & Brown, 1997). It corresponds to
Aristotle's concept of 'episteme'.
Practical knowledge is about knowing how to do something. Pract-
ical knowledge is seen in craft and in all skills, and consequently pre-
supposes some training and experience. This involves 'tacit' knowl-
edge (Polanyi, 1966), procedural knowledge (Hansen, 1997) or ,-
knowing' (Cook & Brown, 1997), and corresponds to Aristotle's con-
cept of 'techne'.
Experiential knowledge signifies a direct experience, knowing some-
thing or somebody from a face-to-face encounter and interaction.
Reading about a place or seeing a video of some people is never the
same as a direct encounter. Being able to express knowledge as expe-
riential is not a direct part of the Nonaka & Takeuchi framework, yet
it is a distinct way of knowing, different from that of know-how or
skills. The knowledge that one has from having been present and hav-
ing been part of the 'history', first-hand experience, is highly relevant.
It is one of the prime advantages of real-time case studies. This know-
ledge, in fact, constitutes the 'context' (Hall, 1983), which we shall get
back to shortly. Cook & Brown (1997:9) make a point: "we see the
interplay of knowledge and knowing as a potentially generative phe-
nomenon - that is, for human groups, the source of new knowledge
and knowing lies in the use of knowledge as a tool of knowing within
situated interaction with the social and physical world. It is this that
we call the generative dance." Experience is contextual. Experiential
knowledge leaves open the possibility of including the ethical or value
aspect of Aristotle's concept 'phronesis', even though the concept is
not quite the same.
12
9
Heron coined an additional term: presentational knowledge, which
is experiential knowledge at a deeper level and concerns the direct
experience of a non-linguistic, spatio-temporal quality of presence
(1981 :27-29). As this thinking is holistic (characteristic of the right
hemisphere of the brain), it can best be expressed in drawing, fantasy,
story-telling or movement (Reason, 1994:44-46). The process of
knowledge creation in the paradigm of cooperative inquiry is displayed
above.
As seen above, the PPEP process model shows a cyclic process of
knowledge and reflection that can start anywhere. Most often it starts
with some proposition or problem that a group wants to examine. This
kind of knowledge creation takes what Reason calls 'critical
subjectivity', which refers to the quality of attention (Reason,
1995:46): "It develops through the cyclical process of cooperative
inquiry, in the iteration between experiential knowing through direct
encounter; presentational knowing expressed in patterns of imagery;
start here
)
Seed idea
(Phase 1)
First
(phase 4)
Provisional
presentational , propositional
portrayal of data knowledge
Second Refined
" propositional presentational
portrayal of data knowledge
~
Etc. ,I.'
Second Second
experiential action-plan
,I.'
grounding and practice
First First action-
experiential
oJ
plan and
grounding
...... practice
(Phase 3) (Phase 2)
,_ p:.r':P!'!..a! ~ n . ? _
, conceptual filters :
I I
I ,
I ,
,
I
I
Boisot,1998:12
event
(data source)
~
~
~
~
L- --' data
Data, Information and Knowledge
propositional knowing expressed in concepts and theories; and pract-
ical knowing expressed in the skills of living (Heron, 1992)."
Individual knowledge acquisition
To some extent this section builds on an earlier paper (Dars0, 1997c),
describing the processes of individual knowledge acquisition. Here we
briefly discuss the four phases: perception, cultivation, integration and
recuperation.
Individual knowledge is cultivated, integrated information.
Perception is a screening stage, filtering what is taken in. Basically, you
see what you 'want' to see, you hear what you 'want' to hear, etc.,
understood as a usually non-conscious process, based on individual
inclination, experience and cognitive structures developed during the
socialization process. These are referred to as: schemas (Stacey, 1996),
schematas (Nonaka & Takeuchi, 1995), or mental models (Senge et aI,
1994), and described as processes of sense making by Karl Weick
(1995). Below is a simple illustration by Boisot, which captures the
difference between data, information and knowledge.
Propositional
Practical
Presentational
Experiental
Reason, 1995:45 (After Heron, 1992)
13 13
1
According to Boisot (1995a:22): "Data in its most basic formulation
is a discernible difference in the energy states of phenomena as they
occur and propagate in space-time, whether as matter or electromag-
netically." Perceptual and conceptual filters let some data through.
This is turned into information and is processed to become individual
knowledge. An interesting question here is whether the perceptual and
conceptual filters are passive like screens, or whether they are active,
i.e. can be controlled or adapted by an individual?
The processing of information is what is meant by 'cultivation'
(Dars0, 1997c). This is an active process of thinking that relates infor-
mation to prior knowledge (any kind of the PPEP), and either leads to
discarding the information - or to a process of integration through
association or mental categorization (i.e. storing
14
it).
Cultivation is a process of the short-term memory, our mental work
space. Recuperation is an activation of long-term memory, which then
influences what data are selected. Long-term memory is spread all
over the brain. The American psychologist Karl Pribram claims that
the brain functions as a hologram, which means that the whole is
located in the parts, and that removing parts will make long-term
memory only more vague (Hansen, 1997: 179). The process of indi-
vidual knowledge creation is illustrated below.
Investigations show that short-term memory has a capacity to hold 7
(plus/minus 2) elements in our consciousness at the same time
(N0rretranders, 1991: 170). In our mental work space we can digest
the information we have taken in, compare it or integrate it with our
prior knowledge (through access to our long-term memory), organize
or recombine it - even think new thoughts. In order to be able to hold
more than seven elements in consciousness, the brain has developed
the process called 'chunking'. This means that similar concepts are
linked into chunks, which form greater wholes of mental categories or
overall symbols (Boisot, 1995a:48).
Tacit knowledge
Human consciousness has limited capacity. Psychologists, however,
have found that at a different, subconscious level a lot more informa-
tion is taken in. This phenomenon was studied by Lazarus &
McCleary in 1949 as 'subception', and by G.S. Klein as 'subliminal
activation' (Polanyi, 1966:7;95). N0rretranders argues (1991:165):
"Perception opens for millions of bits per second; consciousness
only a few dozens. The stream of information, measured as bits
per second, is understood as bandwidth. The bandwidth of
consciousness is much lower than the bandwidth of perception"
(my translation).
action
recuperation
inte2ration
cultivation
framine:
data selection
direct data experience
There seem to be diverging opinions or a lack of clarity as to what
exactly Polanyi's concept of tacit knowledge encompasses. Polanyi
states (1966:7): "These two aspects of knowing have a similar struc-
ture and neither is ever present without the other... I shall always
speak of "knowing," therefore, to cover both practical and theoreti-
cal knowledge." Nonaka & Takeuchi, who build their process model
for knowledge creation on the iteration between tacit and explicit
knowledge, follow Polanyi when they argue that tacit knowledge
includes both cognitive and technical elements, the cognitive part
being the 'mental models' through which we perceive and define the
world:
The Ladder of Knowledge
13
2
133
"For example, knowledge of experience tends to be tacit, phys-
ical, and subjective, while knowledge of rationality tends to be
explicit, metaphysical, and objective. Tacit knowledge is created
"here and now" in a specific, practical context and entails what
Bateson (1973) referred to as "analog" quality. Sharing tacit
knowledge between individuals through communication is an
analog process that requires a kind of "simultaneous processing"
of the complexities of issues shared by the individuals."
(Nonaka & Takeuchi, 1995:60).
Cook & Brown (1997: 14) make a point of stating that tacit knowleqge
cannot be converted into explicit knowledge, thereby implicitly criti-
cizing Nonaka & Takeuchi. Using Polanyi's example of bicycling, they
claim that explicit knowledge is "generated in the context of riding
with the aid of what we knew tacitly."
Whether tacit knowledge is converted into explicit knowledge (Nona-
ka & Takeuchi, 1995) or explicit knowledge is generated through tacit
knowledge (Cook & Brown, 1997) is, in fact, merely a matter of
words, i.e. an academic dispute. The main question is whether tacit
knowledge must remain tacit or whether by some activity or method
- be it through conversion, generation or translation - that tacit
knowledge can be rendered useful in communication and interaction
with others for generating new knowledge. According to Polanyi, this
is possible 'provided we are given adequate means for expressing
ourselves' (Polanyi, 1966:5) - i.e. presentational knowledge in the
PPEP framework. According to Jacobsen (1971), psychic material
with 'tacit knowledge' quality can be expressed. This is based on Ku-
bie who in 1958 argued that primary processes related to creativity
could be brought from the 'preconscious' into consciousness. Primary
thought processes have a dream-like quality and will be discussed in
the next section on creativity.
Thus, when Polanyi claims that we know more than we can tell
(1966:4), the explanation may be that the body takes in much more
information than the individual is conscious of. People with developed
134
intuition may simply know how to access this bodily information to a
greater extent than others. As intuition is another concept that lacks
clarity and that is often used intermittently with tacit knowledge, let
us examine it a little closer.
Intuition
In 'Mind over Machine' the brothers Dreyfus (1986) have described
their studies of human learning, e.g. language acquisition, the game of
chess, different kinds of human skills. They found that human learn-
ing goes through five stages and that there is a qualitative leap from
the third step of being competent to the fourth of being skilled. They
claim that this qualitative leap is caused by going from rule-based to
context-based activity, and that this involves experience and intuition.
An expert is recognized by his/her fluent performance and immediate
action, based on skills and intuition. Many skills have the quality of
'flow' (Csikszentmihalyi, 1990), where the person is one with the
activity and looses the sense of time. The Dreyfus brothers called this
quality 'arationality' (1986:36). In the illustration below the sixth step
has been added by Bent Flyvbjerg (1990) in order to describe another
qualitative leap from expert to renewer. This concerns the creation of
new knowledge or rewiring the brain. "Rewiring the brain, challenging
all those initial habits and assumptions in the face of new experience,
requires far more energy than the original wiring" (Zohar, 1997:28).
The qualitative leap from expert to renewer concerns the difficulty of
breaking habits, whether they are mental or physical. Biological organ-
isms have an in-built mechanism of following the mental paths that
have already been trodden, and it takes a lot of energy and willpower
to break away from expertise or excellence.
135
Six steps in human learning:an illustration of the Dreyfus &

Dreyfus model, 1986. The sixth step added by Flyvbjerg, 1990
Tacit knowledge involves much more than skills of e.g. bicycling, or
'practical knowledge' (PPEP framework). One of Polanyi's own ex-
amples concerned the recognition
15
of faces, which could not be
described (through explicit knowledge), but could easily be recog-
nized. This, however, does not concern 'practical' knowledge or skills,
but rather 'experiential' knowledge. By applying the PPEP framework
we can see how Polanyi's original concept of tacit knowledge encom-
passes both 'practical' as well as 'experiential' knowledge and could
possibly be articulated through 'presentational' knowledge.
Do we have two 'bodies' of knowledge? In 1981 the Danish psychol-
ogist Olav Storm formed a theory about 'the two bodies', the cogni-
tive-voluntary body (carried by the ego) and the emotive-vegetative
body (not controllable by consciousness). The first is our controllable
body, which we can make walk, do gymnastics and bicycle. The sec-
ond can be made partly conscious, like we can control our breathing
and swallowing, but we cannot stop breathing entirely through our
will (because then we will faint and start breathing again) and we can-
not control our feelings or reactions when e.g. getting scared. Accord-
ing to Storm, breathing and sexuality bridge the two bodies, which is
why most mental and spiritual techniques build on breathing exercis-
es (N0rretranders, 1991 :400).
The concept of two 'bodies' adds meaning to Polanyi's description
on how we must keep our attention elsewhere while producing this
tacit knowing, just like we, at night, need to focus our eyes on some-
thing next to what we really want to look at in order to see it. Like-
wise, in studies on creativity many scientists have reported that illu-
mination (see later paragraph on creativity) always happens when they
are attending to something else, preferably a routine job (Edwards,
1987). Many scientists rely on this way of thinking. It is as if the brain
works on its own - much better than when you focus your attention
directly on the problem. Maybe this is what Scharmer (1998:3) relates
to as the 'sources for producing things' or 'self-presencing knowledge'.
Other theories that confirm the body perceiving independently of
our consciousness is the findings from 'subliminal' research (N0rre-
tranders, 1991 :200). These findings were commercialized in the 1950s
as subliminal advertizing, showing ultra-short images during a TV
show. The consumer would not be conscious about a subliminal com-
mercial but it would still influence his behavior. This was, of course,
forbidden by law. The point here is that the findings proved that the
body takes in and reacts to much more than we are conscious of.
In sum, tacit knowledge is primarily bodily knowledge, but it involves
more than skills. The body perceives and absorbs signals that the 'I' is
normally not conscious of, because the'!' is attending to something
else, as Polanyi has described it. Tacit knowledge is produced while the
mind attends to something else and may surface during cyclic think-
ing processes. Thus the body learns and has knowledge, which is inde-
pendent of our rational intellect. Intuition involves access to the tacit
bodily knowledge
16
, often described as 'hunches' or 'gut feeling'. The
Dreyfus brothers (1986) studied how intuition developed from
hunches to virtuosity. Intuition concerns the immediate and simultan-
eous thinking, knowing and acting according to the specific time, situ-
ation and people, i.e. to the context. Related concepts are 'flow' (Csik-
szentihalyi, 1990), 'self-presencing awareness' (Scharmer, 1998) and
'abduction' (Kirkeby, 1994).
137
The knowledge map is created by crossing the axes between informa-
tion and exformation. The axes
17
, which go from plus to minus, are
thus: information (vertically) and exformation (horizontally). When
these are combined, we get four abstract categories.
In the north-east corner we find ignorance, the combination of no
information with no exformation. There are, however, varying degrees
of ignorance. In the top corner ignorance is 'squared', i.e. ignorance of
being ignorant, or 'cognitive blindness'. After that we could speak of ,-
conscious' ignorance, a person knowing that he/she is ignorant. The
first step is to know that you do not know, the second is to have iden-
tified a gap regarding what you do not know. This was expressed by
Reg Revans (1991:6): "Action Learning suggests that, only if a man,
particularly the expert, can be persuaded to draw a map of his own
ignorance, is he likely to develop his full potential."
Information and Exformation
Also the concept of 'exformation' by Tor N0rretranders (1991) is
somehow related to tacit knowledge. When producing information,
some is selected and a lot is discarded. The latter is the exformation.
Information is the result, it is concrete and can be measured. Exfor-
mation is what was discarded, the mental work that was done in or-
der to produce the information. The purpose of information, of the
explicit content, is to indicate what is behind, what is implicit in the
content. "Exformation is the history of information, information is
the product of the history. Both are meaningless without the othey -
information without exformation is empty talk; exformation without
information is not exformation, but rather discarded information."
(N0rretranders, 1991:131)
Of specific relevance to the present study are the processes of
knowledge creation, i.e. the history of information, both because it
indicates the 'depth' or the 'weight' of the information, and because
the history is contextual and practical, and includes direct 'experi-
ence'. It is in this meaning that exformation is applied in the follow-
ing. Thus, exformation without information is here understood as
history, presence, experience, and depth. Exformation may be con-
scious, but hard to articulate, it may be non-conscious as intuition,
it may be experience, knowledge or skills, which could possibly be
articulated, i.e. combined with information. Whereas N0rretranders'
point of departure is information, where exformation is seen as dis-
carded information and thus as part of the past, I have 'adapted' the
concept of exformation to form a forward-pointing potential for build-
ing 'qualified knowledge'. This is illustrated in the knowledge matrix
below.
information
+
tacit
ignorance
knowledge
qualified
rhetoric
knowledge
exformation
Dar,o, 1997c: Knowledge Map
West of 'Ignorance' we find tacit knowledge, the combination of no
information with exformation18. Tacit knowledge consists of individu-
al memory, experience, ideas, intuition, hunches, feelings, values,
which are hard to articulate in language. There are two levels of tacit
knowledge, non-conscious and conscious, tacit knowledge. The social-
ization process is an example of a non-conscious process. Also our
cognitive structures, our very personal way of perceiving and framing
139
the world, our schemata, are normally non-conscious. These, howev-
er, are sometimes revealed or exposed because of voluntary or invol-
untary clashes with reality. Our conscious tacit knowledge
19
can be
technical or practical as in skills or know-how or they can be gut-feel-
ings or hunches connected to action and experience. An important
part of knowledge creation is trying to surface tacit knowledge and
communicating it to others (representational knowledge in the PPEP
framework). Nonaka & Takeuchi (1995) argue that the core of know-
ledge creation concerns this conversion of tacit knowledge into explic-
it knowledge.
.
Moving south-east in the 'landscape' we find rhetoric, information
with no exformation, one of the great challenges of the Internet and
global Knowledge Society. How can we validate information? At the
'bottom' of the box of rhetoric is the area of hollow presentations,
the strategy of eloquent 'shooting from the hip'. Some people are ex-
tremely skilled in persuasive talk and get high up in the hierarchies
of organizations. Roger Bernstein, the manager of Random House,
formulated this dreaded person as the 'articulate incompetent'
(Larsen, 1995). The next step is rhetoric with a little exformation.
Entertainment and politics are examples that could very well illus-
trate this area. Certainly also the status of the person may influence
how information with little exformation or depth is interpreted by
others.
Finally qualified knowledge combines information and exformation.
Qualified knowledge is explicit knowledge based on qualified exfor-
mation. Qualified knowledge is the outcome of a process of exploring
the other areas of the knowledge map. This process may start by try-
ing to make tacit knowledge explicit, what Nonaka & Takeuchi
describes as 'knowledge conversion', or by uncovering and clarifying
important areas of ignorance, as described by Revans. Ideally a
dynamic process emerges between the two. Qualified knowledge is
created when combining this internal process with an external scan-
ning for information in a searching spiralling process and by adding
rhetoric throughout the process, and particularly in the end for the
presentation (or report).
What is a problem?
Plato's paradox (Polanyi, 1966:22): If you know what you are looking
for, then there is really no problem, but if you do not know what you
are looking for, then how would you be able to identify it?
Plato's explanation of this paradox was that discovery is simply
remembering past lives. Polanyi, however, uses this discussion to un-
derscore his claim that "we can know more than we can tell" (Polanyi,
1966:4) and that knowing that a problem is a 'good' or an 'original'
problem must be a function of our "tacit foreknowledge of yet undis-
covered things" (Polanyi, 1966:23). Polanyi's main conclusions are
that tacit knowledge can account for:
a valid knowledge of a problem
the scientist's capacity to pursue it, guided by his sense of approa-
ching its solution
a valid anticipation of the yet indeterminate implications of the dis-
covery arrived at in the end
John Dewey describes four types of problems [1910](1997:72 -74):
1. "the difficulty resides in the conflict between conditions at hand
and a desired and intended result, between an end and the means
for reaching it."
2. "the difficulty experienced is the incompatibility of a suggested and
(temporarily) accepted belief (that the pole is a flagpole), with cer-
tain other facts" (last parenthesis added)
3. "an observer trained to the idea of natural laws or uniformities
finds something odd or exceptional in the behavior of the bubbles.
The problem is to reduce the apparent anomalies to instances of
well-established laws."
4. "In cases of striking novelty or unusual perplexity, the difficulty,
however, is likely to present itself at first as a shock, as emotional
disturbance, as a more or less vague feeling of the unexpected, of
something queer, strange, funny, or disconcerting."
The first two types of problems are similar to the common sense appli-
cation of problems, which usually have negative associations and are
regarded as something to 'get rid of'. In this context problems are
preferably avoided.
The two latter types of problems are disturbing in a different way
as they attract the curiosity of a 'trained observer'. To a researcher, a
problem can be challenging - even 'beautiful'- yet still something that
is meant to be solved. Thus, to a researcher a problem is more like an
opportunity. This leads to two further distinction of problems by Bales
& Strodtbeck (1951) and Reg Revans (1991).
Bales & Strodtbeck distinguished between 'truncated' and 'full-fledg-
ed' problems (1951 :487). Truncated problems are e.g. 'open and ~ h u t '
cases, i.e. closed and with little room for divergence and diverse per-
spectives. Full-fledged problems have some degree of ignorance and
uncertainty, involve different values and interests, and must be solved.
Revans' (1991) distinction is between problems and puzzles. Ac-
cording to Revans (1991: 11), a puzzle is "an embarrassment to which
a solution already exists, although it may be hard to find even for the
most accomplished of experts." Many technical or production 'pro-
blems' are really puzzles, as they can be solved by finding the right
solution. In contrast, a problem "has no existing solution, and even
after it has been long and deliberately treated by different persons, all
skilled and reasonable, it may still suggest to each of them some dif-
ferent course of subsequent action. This will vary from one to another,
in accordance with the differences between their past experiences,
their current values and their future hopes."
In this book we take Revans' definition and view problems as
open-ended and as having potentially many different solu-
tions. This definition goes well with scientific search, which is often
an open-ended process of search and discovery. The heart of the mat-
ter concerns the journey from an unarticulated or yet unknown prob-
lem towards formulating the problem or knowing what the problem is.
This is problem-finding as opposed to problem-solving.
"The formulation of a problem is far more often essential than
its solution, which may be merely a matter of mathematical or
experimental skill. To raise new questions, new possibilities, to
regard old problems from a new angle requires creative imagin-
ation and marks real advance in science," as Einstein and Infeld
have expressed it in 'The Evolution of Physics' of 1938
(Maslow, 1970:18).
In 'How We Think', Dewey focused on thinking as reflection [1910]
(1997:57):
"Reflection is turning a topic over in various aspects and in var-
ious lights so that nothing significant about it shall be over-
looked - almost as one might turn a stone over to see what its
hidden side is like or what is covered by it."
Dewey outlined 5 steps of reflection (1910:72):
i) a felt difficulty
ii) its location and definition
iii) suggestion of possible solution
iv) development by reasoning of the bearings of the suggestions
v) further observation and experiment leading to its acceptance
or rejection; that is, the conclusion of belief or disbelief
What is an idea?
"An active consciousness holds all ideas lightly"
Gudy Marshall at a Conference in Bath, 1995, in Torbert,
1997a:14)
Ideas are suggestions that can be more or less articulated or advanced.
Some are diffuse and vague, others are specific and ready to implement.
In Dewey's framework an idea is primarily a factor in judgment
(1997:108):
"In this process of being only conditionally accepted, accepted
only for examination, meanings become ideas. That is to say) an idea
is a meaning that is tentatively entertained,iormed) and used with ref-
erence to its fitness to decide a perplexing situation, - a meaning used
as a tool ofjudgment ...
143
.. But if we treat what is suggested as only a suggestion, a sup-
position, a possibility, it becomes an idea, having the following
traits: a) As merely a suggestion, it is a conjecture, a guess,
which in cases of greater dignity we call a hypothesis or a theo-
ry. That is to say, it is a possible but as yet doubtful mode of
interpretation. b) Even though doubtful, it has an office to per-
form; namely, that of directing inquiry and examination ... Tak-
en merely as a doubt, an idea would paralyze inquiry. Taken
merely as a certainty, it would arrest inquiry. Taken as a doubt-
ful possibility, it affords a standpoint, a platform, a method of
inquiry. Ideas are not then genuine ideas unless they are tools in.
a reflective examination which tends to solve a problem."
Thus, according to Dewey ideas are 'working hypotheses' that
help in directing inquiry and examination. This definition, in
fact, covers our common sense understanding of an idea that could
be for a new product or application, as well as suggestions for some
activity. It also makes it possible for us to distinguish between a
problem and an idea, even when they are both vague, as the former
involves a perplexity and the latter involves a suggestion. Both direct
thinking.
"The problem fixes the end of thought and the end controls the
process of thinking." (Dewey, 1997: 11)
"Logical ideas are like keys which are shaping with reference to
opening a lock." (1997:109)
"I like to stress that ideas are the raw material of solutions,
potential stimuli for the high-quality ultimate solution," says
VanGundy (in Grossman, 1997:10).
The shaping and fixing quality of ideas is seen in the framing or
reframing of problems. The way that a problem is defined more or less
determines the solution (Schon, 1986).
144
Problem framing
"Reality has a tendency to reveal itself in accordance with the
perspectives through which it is engaged." (Morgan, 1997:350)
Goffman (1974) uses framing to explain the 'organization of
experience'. This incorporates the whole situation and all the cues that
an individual uses in order to make sense out of 'what is going on
here?' Goffman's concept thus concerns relational as well as concep-
tual matters. In the present chapter our scope is narrower as we want
to examine the activity of conceptual framing.
An area where framing and reframing problems have been of ma-
jor concern is in the psychology of brief therapy. Watzlawick et al
describe reframing (1974:95): "What turns out to be changed as a
result of reframing is the meaning attributed to the situation, and
therefore its consequences, but not its concrete facts." This is relevant
in the present book as reframing concerns the changing of perspec-
tives - not things. In this sense reframing has to do with the concep-
tual framework, i.e. the views, the expectations, the reasons and the
premises (Watzlawick et aI, 1974: 104). De Shazer, also into brief ther-
apy, defines frames as 'rules' by which we construct our reality, differ-
ent rules might apply in different situations. And it is the transform-
ation of these rules that are necessary for creating change. De Shazer
concludes (1988: 118): "The best way to design a failure is to estab-
lish a poor definition of the complaint."
Thus problem framing happens early in perception, as seen in the 'lad-
der of Knowledge', described earlier in this chapter. Donald Schon
(1986) describes, how we, intentionally or unintentionally, make use
of 'generative metaphors', which derive from our cognitive schemas.
In his example from the area of social policy, he points out how see-
ing an urban housing situation as a 'slum area' and as 'possessed of a
congenital disease' indicates a different solution than describing the
same situation as a 'locus for social relationships' and a 'natural com-
munity and its dislocation'. "Each story constructs its view of social
reality through a complementary process of naming and framing"
145
(Schon, 1986:264). The point is that in the description of a story or a
problem there is already an inherent subjective evaluation. And it is
"this sense of obviousness of what is wrong and what needs fixing
(that) is the hallmark of generative metaphor in the field of social pol-
icy" (parenthesis added, Schon, 1986:266). The process of problem-
setting matters profoundly because it reflects subjective values and
purposes, and because it (implicitly) prescribes the diagnosis and sets
out the direction for seeking solutions.
often used for benchmarking. Of course, it is important to develop the
best routines and to keep improving them. Sometimes organizations
and groups fall into the trap of doing things right (routines) instead of
doing the right things (Larsen, 1992). The drawback of routines is the
path-dependency and the in-built conservatism of being habit-bound.
Therefore routines need to be revised and scrutinized regularly. Organ-
izations, in fact, need to build in 'organized abandonment' of all rou-
tines (Drucker, 1993).
Problem-finding and problem-solving
"Successful leaders are "committed to problem-finding, not just
problem-solving. They embrace error, even failure, because they
know it will teach them more than success."
(Warren Bennis, 1992) (in Snyder & Clontz, 1997:71)
Routines: When combining a known problem with a known solution,
we are in the field of routines or procedures. There is no need for
'novelty' when 'continuity' is working well. Routines are time saving
and efficient, and are part of the structural capital of the organization
(Sveiby, 1991). Routines can be shared as 'best practices' and are
routines problem-
SOlving
exploitation explorative
search
Problem solving: When a solution to a known problem (in fact: a puz-
zle) has to be found, we can speculate, seek information, make trial-
and-error experiments, or design scientific experiments. Problem solv-
ing involves investigation and focused search for information. Some
scientific discoveries belong to this area. Thomas Edison knew the
problem very well, but it took more than 500 trials and a lot of persist-
ence to find the solution
20
Two of Drucker's seven sources of innov-
ation (1985) concern problem solving: investigating the problem of
an 'incongruity' or mismatch, and trying to find the 'missing link' in
relation to a 'process need'.
Problem solving involves real practical problems, which Dewey
describes as forked-road situations. By that he indicates that there is
often an urgency for action.
Exploitation: In knowledge society one way of applying knowledge is
to exploit existing knowledge, products or compounds for new appli-
cations or new markets. One area in which a known solution can be
exploited to find answers to unknown problems is in technology-based
search, where technology is exploited in order to find new applica-
tions. The company has the 'solution' but lacks the 'problem'. Simi-
larly, industries such as chemical or pharmaceutical companies may
be in the possession of various compounds that can be screened, put
through experiments and adapted or changed. The point is to exploit
existing assets in every possible way.
(Dewey, 1997: 11): "Thinking begins in what may fairly enough
be called a forked-road situation, a situation which is ambiguous,
which presents a dilemma, which proposes alternatives ...
Problem Map
unknown
solution
known
known
unknown
problem
Problem-finding versus problem-
solving implies two vastly different
perspectives, but involves the same
two elements: a problem (known or
unknown) and a solution (known
or unknown). Whether these ele-
ments are known or must be searc-
hed for determines the kind of
approach that is appropriate for
project work. The matrix offers
four combinations ofproblems and
solutions, resulting in four different
approaches: routines, exploitation,
problem-solving, and explorative
search.
147
The dilemma of project planning, Mikkelsen & Riis, 1992
Problems of the third kind
The figure illustrates how important decisions, taken early in a project,
as for instance the choice of a certain technology, is decisive for the scope
and path of the project. Other technologies were from the beginning
counted out, even though they might actually have proved to be better.
.......
'.
..........................
available information and knowledge the impact of decisions
. .
.....
".
".
.....
Thus, the problem formation process is usually skipped, as the prob-
lem is given. This means that quite often, by the termination of a pro-
ject, project groups realize that they have been working with the wrong
problem. And the irony is that now that they know what the problem
really is, the project is over. This type of error has been called 'error of
the third kind' or the 'fallacy of misplaced precision' (Mitroff &
Featheringham, 1974).
Reg Revans (1991 :6) describes this situation: "there is nothing so
terrible in all human experience as a bad plan efficiently carried out,
when immense technical resources are concentrated in solving the
wrong problems. Hell has no senate more formidable than a conspir-
acy of short-sighted leaders and quick-witted experts."
What is described here could very well be caused by a well-known
dilemma of projects: that important decisions are often made early in
a project at a time when there is little information and knowledge. This
is illustrated in the model below.
Demand for the solution of a perplexity is the steadying and
guiding factor in the entire process of reflection ... But a ques-
tion to be answered, an ambiguity to be resolved, sets up an end
and holds the current of ideas to a definite channel. Every sug-
gested conclusion is tested by its reference to this regulating
end, by its pertinence to the problem at hand. This need of
straigthening out a perplexity also controls the kind of inquiry
undertaken."
Explorative search: An unknown problem combined with an un-
known solution is difficult to tackle, and some would be inclined. to
say: why bother, if there isn't even a problem? In a competitive
business world, however, where innovation is essential, new areas
must be explored and new problems or needs must be identified
(and preferably before the competitors). As Simon stated (1986: 13):
"By its very nature, scientific discovery derives from exploring pre-
viously unexplored lands. If it were already known which path to
take, there would be no major discovery - and the path would most
likely have previously been explored by others." Exploration can be
a time-consuming process. Even Peter Drucker (1985) warns entre-
preneurs of the hardships of the seventh source of innovation: new
knowledge.
Explorative search often starts with searching for an original prob-
lem, by posing a 'burning' question, by wondering about something,
or by having a vague idea. It implies a general scanning for informa-
tion and attempts to build a general database for decision making.
Interestingly, the stage of finding or forming the problem is often
skipped. In scientific literature as well as in everyday life, individuals
and groups prefer 'solution cycles' (Poole & Roth, 1989a). Thus the
main interest lies in problem-solving. Donald Schon criticizes this
prevailing attitude, which directs our attention to search for solu-
tions, even before the problem has been identified. The problem is
often taken more or less for granted. Schon writes (1986:261):
"There are great difficulties with the problem-solving perspective ...
Problems are not given. They are constructed by human beings in
their attempts to make sense of complex and troubling situations."
149
Postponing decisions is in accordance with Dewey's reflective think-
mg.
"Set-based concurrent engineering bases the most critical, early deci-
sions on data. The earliest decisions about designs have the larg-.
est impact on the ultimate quality and cost, but these decisions
are made with the least data." (Ward, et. aI, 1995:59)
In this book the applied methods aim at lifting the curve of knowl-
edge, before making important choices (see figure below). In fact, one
proposition of this book is that by building enough data, information
and knowledge concerning the problem area, the problem formula-
tions become more qualified, more potential is captured, better oppor-
tunities emerge and decisions (whether to 'stop or go') can be made
on a more solid basis. This is demonstrated in the Toyota approach,
discussed earlier:
tion. To turn the thing over in mind, to reflect, means to hunt
for additional evidence, for new data, that will develop the sug-
gestion, and will either, as we say, bear it out or else make obvi-
ous its absurdity and irrelevance. Given a genuine difficulty
and a reasonable amount of analogous experience to draw
upon, the difference, par excellence, between good and bad
thinking is found at this point. The easiest way is to accept any
suggestion that seems plausible and thereby bring to an end
the condition of mental uneasiness. Reflective thinking is
always more or less troublesome because it involves overcom-
ing the inertia that inclines one to accept suggestions at their
face value; it involves willingness to endure a condition of men-
tal unrest and disturbance. Reflective thinking, in short,
means judgment suspended during further inquiry; and
suspense is likely to be somewhat painful. As we shall see lat-
er, the most important factor in the training of good mental
habits consists in acquiring the attitude of suspended conclu-
sion, and in mastering the various methods of searching for
new materials to corroborate or to refute the first suggestions
that occur. To maintain the state of doubt and to carryon
systematic and protracted inquiry - these are the essen-
tials of thinking." (emphasis added)
Knowledge creation process models
We have already examined a few process models of knowledge cre-
ation. We started with Heron's PPEP model of Cooperative Inquiry, a
cyclical model, which has formed the framework for our discussion.
We have described two stage models: the Dreyfus & Dreyfus learning
model and Dewey's reflective thinking model. We want to conclude
this section on knowledge creation with Nonaka & Takeuchi's two
models (1995): the 'knowledge spiral' of four modes of knowledge
conversion, and the 'spiral of organizational knowledge creation'.
.... -
._--_.
available information and knowledge the impact of decisions
. ~ - r ...
.., "." .
.' ' .. /."
./ :'" I "
...... ,/ "'Y>---' '\
/ .... "" ,
"". ,
"". .
,;,; '. '.
-,," ",
- _ . : : : : ~ - - ...
Ideal Project flow (inspired by Mikkelsen & Riis, 1992)
(Dewey, 1997:13): "If the suggestion that occurs is at once
accepted, we have uncritical thinking, the minimum of reflec-
"The cornerstone of our epistemology is the distinction between
tacit and explicit knowledge ... the key to knowledge creation
Nonaka & Takenchi 1995:72. Contents of knowledge ereated by the four modes.
Tacit knowledge Explicit knowledge
To
Even though the four modes of knowledge conversion are experienced
by the individual, the conversion between tacit knowledge and explicit
knowledge is also social, according to Nonaka & Takeuchi (1995:61).
There appears to be some inconsistency in the theory regarding how
individual knowledge is turned into organizational knowledge, which
has also been pointed out by Griffin, Shaw & Stacey (1997).
According to Nonaka & Takeuchi, the social process of knowledge
creation is triggered by different mechanisms. In the socialization
mode it is triggered by building a 'field' of interaction for sharing
mental models and experience. Tacit knowledge is externalized
through dialogue or collective reflection, using metaphors or anal-
ogies to articulate tacit knowledge. Combination of explicit knowledge
is facilitated by networking and thereby creating a new product, ser-
vice or system. Finally, 'learning by doing' is what triggers internal-
ization.
Eventually a spiral emerges and knowledge is elevated or 'amp-
lified' from individual, through groups, to organizational levels. At the
organizational level a fifth phase is added: cross-leveling knowledge.
This means that the knowledge generated is extended and spreads
internally and (sometimes) externally and thereby gives rise to a new
knowledge spiral (Nonaka & Takeuchi, 1995:88).
(Socialization) (Externalization)
Sympathized Conceptual
Knowledge Knowledge
(Internalization) (Combination)
Operational Systemic
Knowledge Knowledge
From
Tacit
knowledge
Explicit
knowledge
lies in the mobilization and conversion of tacit knowledge."
(Nonaka & Takeuchi, 1995:56)
The four modes of individual knowledge creation, described below,
are the 'engine of the entire knowledge-creation process' (Nonaka &
Takeuchi, 1995:57). It will be complemented by the four characteris-
tics of 'Ba', which have been added to the original knowledge spiral by
Nonaka & Konno (1998:46). 'Ba' stands for a shared space that can
take several forms:
1 Socialization (from tacit to tacit): Sharing experiences and build-
ing shared 'mental models' through observation and direct e x p ~ r i
ence (e.g. like setting up a 'brainstorming camp') helps to create
'sympathized' knowledge. This is 'Originating Ba'. In the PPEP
framework this would correspond to 'experiential' knowledge.
2 Externalization (from tacit to explicit): Articulating tacit know-
ledge can be encouraged through metaphors, analogies, images and
models. This process is typically seen in concept creation through
collective reflection - thus 'conceptual' knowledge. This is 'Inter-
acting Ba', and would correspond to 'presentational' knowledge in
the PPEP framework.
3 Combination (from explicit to explicit): Concepts are linked to
explicit knowledge and sorted, systematized, and recombined into
more complete documentation or 'systemic' knowledge (e.g. proto-
types). Nonaka & Konno label this 'Cyber Ba'. It would correspond
to 'propositional' knowledge in the PPEP framework.
4 Internalization (from explicit to tacit): Through documents, man-
uals or stories explicit knowledge is internalized, absorbed or inte-
grated into each person's individual knowledge and experience. It
is closely related to 'learning by doing' and is called 'operational'
knowledge, or 'Exercising Ba'. In the PPEP framework this would
entail a completed learning cycle of combined 'practical' and 'ex-
periential' knowledge.
153
Nonaka & Takeuchi's process model at first sight appears holistic and
relational as it is cyclic and described as a 'spiral' of knowledge creation.
In some ways, however, it also has similarities to a stage model, as it is
described as a 5 phase model of 1. Sharing tacit knowledge, 2. Creating
concepts, 3. Justifying concepts, 4. Building an archetype, and 5. Cross-
leveling knowledge (1995:84). At the same time it is pointed out that the
involved teams are cross-organizational, working more like 'rugby teams'
than as the participants of a 'relay race' (Takeuchi & Nonaka, 1984).
Nonaka & Takeuchi's theory of 'Organizational Knowledge Creation'
(1995) has been trend setting in the new field of 'Knowledge' 5=re-
ation and Knowledge Management. At the same time it revitalized
Michael Polanyi's concept of 'tacit knowing' (1966) and started a
debate. Whether we call it tacit knowledge, bodily knowledge, implicit
knowledge or 'Ba' is not the point. The point is that human beings
'feel'/'sense' something that is hard to articulate in language, but
which can be transformed into images, drawings or metaphors and
from there into new knowledge. What Nonaka & Takeuchi have, to
some extent, underestimated is the significance of communication and
interaction. Their perspective is partly seen from the level of the indi-
vidual, partly organizational. Theirs is not a group perspective, as in
the present case studies. Still, as we shall see later, their model, at least
partly, does serve as a constructive framework for understanding com-
munication in heterogeneous groups.
By others Nonaka & Takeuchi have been criticized for a discrepancy
between a claimed social process but actually maintaining an individ-
ual perspective throughout their book. Likewise Nonaka & Takeuchi
are criticized for promoting a linear process of knowledge develop-
ment, seeing knowledge as 'possessed' by individuals and shared
between individuals instead of seeing knowledge as participative and
as emerging out of edge-of-chaos situations (Griffin, Shaw & Stacey,
1997: 11): "What Nonaka and Takeuchi end up with, then, is a process
for knowledge creation which can be managed and controlled. We
would maintain that management has important and legitimate con-
trol functions, but that knowledge creation is a process which cannot
be controlled or managed from outside of that process ... "
154
This, indeed, leads to a good question: can knowledge creation be
managed from outside the process - and can it be managed from
within?
Concluding remarks
In Knowledge Society knowledge creation is the heart of the matter.
Information floats freely and can connect people everywhere - at least
those who have access to the new global infrastructure: the Internet.
But if information is to be turned into knowledge it needs to be pro-
cessed by individuals - or even better - by interacting individuals who
apply multiple perspectives (as in heterogeneous groups) in order to
create qualified knowledge.
As we have seen in this section, knowledge can be differentiated in a
variety of ways and understood from a variety of theories and frame-
works. As a meta-framework we selected John Heron's PPEP (Propo-
sitional, Practical, Experiential, Presentational) model, developed in
cooperation with Peter Reason, because we will need a differentiated
framework for discussing different kinds of knowledge and knowing
throughout the book. We introduced a Knowledge Map as we found
that one particular aspect of knowledge was missing in literature:
ignorance. Other aspects of knowledge that are attracting interest in a
Knowledge Society are intuition and bodily knowledge. As long as
people have existed on this earth intuition has been part of decision
making. Managers admit that even in their rational world they still rely
on intuition when in doubt. More studies are needed on intuition and
bodily knowledge.
One of the challenges of this book is to find answers to Plato's para-
dox: How can you identify a problem, when you do not know what it
is? What is a problem - and how are problems best formed, framed,
identified, and constructed? Interestingly, John Dewey's contributions
from 1910 were highly relevant and valuable in defining 'idea' and
'problem', which are central concepts for the present study.
155
In sum, the field of knowledge creation is rich in providing frame-
works, classifications, and process models that are highly relevant for
the study of innovative processes. When reviewing literature on
knowledge, however, it appears that the group perspective is miss-
ing, 'ignorance' is ignored, and 'communication' is 'commonsensical'
and not in focus, i.e. not built on a profound psychological founda-
tion.
Creativity
"In our willingness to step into the unknown, the field of all
possibilities, we surrender ourselves to the creative mind that
orchestrates the dance of the universe."
(Deepak Chopra 1994: 81)
Introduction
The purpose of this section is primarily to clarify and discuss the
'chameleonic' concept of creativity, and secondly to compare and
distinguish creativity from innovation. The concepts of creativity
and innovation are often intertwined because of their close relation-
ship. As we shall see, however, there are important differences. In
relation to the research questions of this book, we would like to
know: What is the role of creativity in enhancing innovative process-
es? Is creativity a necessary condition for innovative processes to
come about?
This discussion entails a clarification of different types of thinking
and different types of creativity. Another important aim of this section
is to clarify and discuss 'innovative crystallization', a concept coined in
the present study in order to describe the emergence of novelty. In
accordance with the previous sections, a few process models of cre-
ativity are introduced.
Lines of research
Creativity is hard to define and most researchers refrain from exact
definitions and try instead to pin down some of its characteristics. Re-
search on creativity has been inspired by Osborne, who invented the ,-
brainstorming' technique, of which the most original feature was the
rule of non-judgement. Also Abraham Maslow's studies of self-actual-
izing individuals have influenced research (1970). Maslow proposed a
humanistic psychology studying creativity in healthy, middle-aged
people as opposed to most psychological studies focusing on abnor-
malities and deviance. The core of creativity, in my opinion, lies in per-
ception and in the ability to make an original change of perspective.
David Bohm writes (1998:4): "But real perception that is capable of
seeing something new and unfamiliar requires that one be attentive,
alert, aware, and sensitive." It takes curiosity and alertness to perform
a change of perspective and from that see what could be possible.
Bohm finds that the essence is to see new sets of relevant differences
and similarities, which can give rise to new order, and to new hierar-
chies of order.
There are two main schools of creativity: Creative Problem Solving
(CPS), also called the Buffalo school, that has a strong orientation to-
wards tools; and Synectic, a theory developed by Gordon and Prince
21
,
that focuses on unconscious processes. These are brought forward
through fantasy, analogy, metaphor, and intuition or through a climate
of acceptance. This line of thought was discussed in the prior section
on tacit knowledge and intuition. We shall outline the CPS process
model later. First we want to focus on five characteristics of creativity.
Some characteristics of creativity
Knowledge: It has been stated time and again that knowledge is a
prerequisite for creativity. This point was well phrased by Herbert
Simon (1986: 11): "It is the surprise, the departure from the
expected, that creates the fruitful accident; and there are no sur-
prises without expectations, nor expectations without knowledge".
157
Cognitive processes: In general, creativity is associated with imag-
ination, perception and association, i.e. primary processes, holistic
thinking and divergent thinking. Creativity is about finding new
solutions to old problems, combining things in new ways or seeing
things in a different perspective.
Emotions: Creativity is related to emotions, expressed in art such
as poetry, music, paintings, dance, etc. In addition, creativity, when
expressed in art, incorporates an aesthetic feature.
Activity: The root of creativity is creation. The activity can be phys-
ical or mental, but it is hard work.
Novelty: Creating is often a combination of known e l e m e n ~ s to
form something new, ranging from 'a little different' to 'radically
new'. Chikszentmihalyi (1996:28) defines creativity as "any act,
idea or product that changes an existing domain, or that transforms
an existing domain into a new one."
When the 'new' product can form the basis for economical growth, we
talk about innovation. Innovation is often the result of a creative pro-
cess. There are many similarities between creativity and innovation.
They both deal with novelty and activity, and both take hard work,
knowledge and skills (mental or physical). Let us look at some differ-
ences between creativity and innovation.
An important difference between creativity and innovation stands out
when looking at the key elements of creativity, drawn out of 22 con-
tributions to the field by Welsch (1980, in Isaksen, 1988):
"Creativity is the process of generating unique products by
transformation of existing products. These products, tangible
and intangible, must be unique only to the creator, and must
meet the criteria of purpose and value established by the cre-
ator."
In contrast, when aiming at innovation, the products must be unique
to the recipients, and must meet the criteria of purpose and value of
the recipients, i.e. the company board or evaluation committee and,
later, the customers or clients.
Summing up:
Creativity is a process - innovation is a result
Creativity is not primarily aimed at economic gain - innovation is
Creativity is evaluated by the creator - innovation is evaluated by
the recipients
"Creativity is getting the idea, and innovation is making it hap-
pen" (Gamache, 1988).
We shall now distinguish creative processes from innovative processes,
which are the focus of this study. The main difference lies in the qual-
ity, the strength and the active use of emotions.
In creative sessions emotions are at play. The persons involved
express their emotions, e.g. people laugh a lot. "If you're laughing,
you're more likely to break all that education and come up with a
wicked-good idea" (Doug Hall, in Grossman, 1997:4). Creativity
allows people to become playful and foolish, which, in fact, is recom-
mended for organizational renewal (March, 1979). By some ther-
apists, creativity is even regarded as crucial for maintaining mental
sanity. Creative activities have been described as the experience of
'flow' in a theory of optimal experience (Chikszentmihalyi, 1990).
Innovative processes, on the other hand, are conceptual and cog-
nitive processes of forming and framing a problem. The topic is inves-
tigated rationally, based on internal and external knowledge. Both
divergent and convergent thinking is applied, and emotions are most-
ly ignored. The search is focused or at least has some direction.
The characteristics of the creative person
It is beyond the scope of the present study to examine individuals. It
is a fact, however, that some individuals appear to have more creative
talent than others. Buzan & Keene (1994) made a thorough study of
great geniuses and found 20 capabilities that are characteristic for tal-
ented individuals. They advocate that these skills can be learned,
which is in line with other approaches. Critical voices maintain that
schooling and Western dedication to rationality literally kill the seeds
159
of creativity that all children possess and express in play (Spang Olsen,
1998). "As our education increases, our imagination decreases" says
Doug Hall, US innovation guru (Grossman, 1997:4).
As the context of the case studies is one of science, we shall con-
clude this paragraph with Simon's findings on the characteristics of
creative scientists (1986: 18):
"From our review of scientific discovery, we have seen that at
least three stigmata seem to characterize scientists who are
unusually creative: first, sensitivity to 'accidents' and readiness
to respond to them, even abandoning an ongoing program (a.s
the Curies did in their search for radium); second, care and
thoughtfulness in defining and selecting research goals and
research problems; third a propensity for risk taking."
Divergent and convergent thinking
How are innovative processes initiated? How does a search process
start?
The pragmatist philosopher, John Dewey described the search pro-
cess in his 'reflective thinking model' from 1910 as consisting of five
steps (described earlier). Dewey regarded curiosity, playfulness, and
suspended judgment as important ingredients in problem solving
activities and used ideas as working hypotheses. Dewey's model and
method from 1910 have inspired and influenced many researchers
who were interested in creativity, thinking, communication and prob-
lem solving.
Later two major thought patterns, divergent and convergent think-
ing, were identified and discussed i.e. at a conference in 1963, where
Guilford, Getzels and Jackson, andThomas Kuhn, among others, par-
ticipated.
Divergent thinking is a searching process of generating ideas, look-
ing for information, scanning the environment, inquiring into matters.
It is a process of searching, exploring, expanding, developing and
unfolding. Most importantly, it involves the suspension of judgment.
Convergent thinking is a process of narrowing down and focusing.
It involves comparing, classifying, examining, analyzing, selecting,
160
eliminating and synthesizing. The goal is judgment, evaluation, choice
and decision making. The generally accepted view is that both types
of processes are essential for creativity, but that especially the shift to
the convergent phase may create some tension (Scheidel, 1986: 119,
128-129).
The Nobel-prize winner, Dr. Roger Sperry, a pioneer in brain
research, was the first to describe two fundamentally different ways of
thinking: the 'logical' left-hemisphere (linear, verbal, sequential,
syntactical, digital) versus the 'holistic' right-hemisphere (global,
nonverbal, simultaneous, perceptual, spatial) thinking processes
(Buzan, 1993). Later research modified the actual location of these
processes, but not the distinction. It is therefore more adequate to
regard these as different styles of thinking.
With the rapid development of technology, research on cognition has
accelerated. The concept of intelligence has been expanded to encom-
pass 'emotional intelligence' (Goleman, 1996) or, in fact, multiple
intelligences (Buzan & Keene, 1994). In her book on Quantum think-
ing (1997), Danah Zohar describes recent research, which has iden-
tified three neural brain patterns:
Serial thinking
Associative thinking
Quantum thinking
Serial thinking is the brain's intellect. It is rational, logical and rule-
bound. In the brain, neurons are connected in neural tracts, in
series of one-on-one neurons. This thinking is linear and determin-
istic. Serial thinking is fast, accurate, precise and reliable. The dis-
advantage is that it is inflexible.
Associative thinking is the brain's heart. It is emotional, social and
habit-bound. In the brain this thinking is linked to neural networks
that can wire and rewire themselves according to experience. This
thinking is linked to practice and develops through trial-and-error
learning, bodily skills (such as bicycling), tacit learning and pattern
recognizing. The disadvantage is that it is slow, inaccurate and
habit-bound.
Quantum thinking is the brain's spirit. It has to do with deep values,
visions and a quest for meaning. Quantum thinking arises from a
field across the brain of synchronized oscillations of neurons. The
pattern has been identified in brain research, but cannot yet be
explained. Quantum thinking is creative, intuitive, insightful, rule
breaking and rule making. It is holistic and unifies and integrates
serial and associative thinking.
Primary and secondary thought processes
A weakness of the 'hemispheric' division of thought processes i ~ that
emotions are generally attributed to the right hemisphere. In compar-
ison, the concepts of primary- and secondary thought processes, stem-
ming from Freud Qacobsen, 1971), encompass distinct thought pro-
cesses as well as different ways of organizing emotions.
Primary thought processes are characterized as simultaneous,
spontaneous and with changing images, no borders, no sense of time,
no language, no precision and no nuances. They have a dream-like
quality and can hold ambiguity, as opposite feelings can exist at the
same time without conflict and as feelings have changing bindings to
objects. Needs must be gratified immediately. Primary processes work
according to some general principles, which are easily recognizable
from dreams Qacobsen, 1971:17-23):
1) Resemblance between people, animals or objects means that they
are represented as identical in dreams. This means that if an object
resembles something else (even as a certain 'feeling quality'), they
appear identical, e.g. when a bird is suddenly a dog and later turns
into something else.
2) Pars pro toto =the part is the same as the whole. A person can, for
example, be represented in a dream by a small part, e.g. an earring
or a special walking style.
3) Condensation. A number of different objects may have a common
part, which then can express all the objects.
Secondary thought processes are logical and closely related to lan-
guage. Ambiguity is difficult to hold and creates conflict. A sense of time
is prevalent and gratification of needs can be postponed Qacobsen,
1971). Feelings have stable bindings, which is essential for the develop-
ment oflanguage.The same stable bindings offeelings enable 'chunking',
the linking of related concepts/feelings/experiences that expand the hu-
man potential for thinking and memorizing (Hatch, 1997: 11).
Primary and secondary processes are usually not sharply divided in
normal individuals. Particularly secondary processes are not found in
their 'ideal form'.
Discovery
The process of discovery is often described as a 'heureka' experience
where the solution to a problem suddenly 'pops' up in the brain. In
order to explain how this can happen without conscious secondary pro-
cess thinking, we need, however, to look deeper into the area of the pri-
mary processes. In 1958 Kubie, who studied neurotic behavior and
creativity, found it necessary to categorize primary processes into the
unconscious, which consists of repressed ('sick') psychic material, and
the pre-conscious which consists of psychic material, thoughts, feel-
ings and memory that can be brought into consciousness Qacobsen,
1971). Material from this area will, however, often be difficult to express
in language (tacit knowledge, Polanyi, 1966). In this way primary pro-
cess thinking is used for thinking many thoughts simultaneously and for
swift movements between thoughts and feelings (e.g. according to the
laws of primary processes, e.g. part pro toto). This is followed by de-
cision or selection of one among many - the characteristic precision of
secondary processes. All this takes place in the pre-consciousness.
Jacobsen points out that the most difficult achievement is to think
up something that has no language, which means that it cannot even
be translated into secondary process thinking. Here images of primary
process thinking must be pulled into secondary process thinking,
words are borrowed from other areas until the idea or concept finally
gets its own vocabulary. This is, according to Jacobsen, one of the
highest forms of creativity. In fact, this is the key to Synectics. This
may also happen with experiential knowledge and feelings, which can
be extremely difficult to describe. Often this must be described in
metaphorical language or in art. Ultimately art and science seem to
derive from the same source (Maslow, 1970).
Abraham Maslow has taken this a bit further and distinguishes between:
Simon said (1986:4): "The processes required for creative acts are the
same as those required for all intelligent acts." But are the processes
the same or is there a difference between 'divergent thinking', 'primary
process' and 'holistic thinking'?
The model below was constructed as an attempt to answer that
question.
1. Primary creativity: a creativity that mainly uses primary processes,
as e.g. in associations (brain-storming), imagination, fantasies, etc.
This creativity involves divergent thinking in the sense that it means
opening up for possibilities, asking lots of questions and taking in
lots of information.
2. Secondary creativity: a creativity based mainly on secondary pro-
cess thinking, as e.g. in production of bridges, houses, new cars,
which is often an improvement of already existing ideas. This is
mainly convergent thinking aiming at narrowing down possibilities
and making decisions or obtaining solutions to problems.
3. Integrated creativity: a creativity that equally applies both types of
thinking, intermittently and sequentially. According to Maslow, it is
this kind of creativity that yields the great scientific discoveries and
works of art.
Creativity: Combining primary processes with divergent thinking is
the prototype of creativity. It stretches from crazy fantasies (on the
borderline of psychosis) to playfulness, idea generation, associ-
ations, images. Many of the activities of composers, poets, writers,
choreographers and painters, among others, start in this form.
Creative problem solving: Here we find primary processes combined
with convergent thinking. Most Creative Problem Solving tech-
niques work with associations or analogues (Harriman, 1988) char-
acteristic of primary process thinking, but the fact that the process
starts out with an identified problem makes it convergent already. It
should be noted that the concept of creative problem solving and the
school, Creative Problem Solving, need not be identical. The latter
is a school with its own techniques. It involves various sets of diver-
gent/convergent steps (Isaksen, 1988), as we shall see later. Other
techniques are straightforward problem solving techniques involv-
ing creativity. The main difference is whether the problem has
already been identified or whether this is a part of the exercise
Focused search: We now combine secondary process thinking with
divergent thinking. In most research search is focused. It is diver-
gent in the sense that it opens up for new information and knowl-
edge, but it is still focused on or around a topic.
Analytical problem solving: The final space combines secondary pro-
cess and convergent thinking. This thinking is analytical, critical,
goal-oriented, and strategic. This is the 'prototype' of science.
After this discussion it becomes possible to distinguish creativity from
scientific search. Creativity involves primary process thinking, which
can be combined with divergent as well as convergent thinking (as in
the bottom two quadrants).
Scientific search involves secondary process thinking, combined
with divergent and convergent thinking (as in the top two quadrants).
Integrated creativity combines all 4 quadrants as indicated by the
dotted line, and holistic/sequential thinking can, in principle, be
applied in any quadrant. Even though this is taking the concept
beyond Maslow, we shall use his (1970: 159) descriptions of self-
actualizing human beings. Maslow noticed how ignition, inspiration
and illumination would coexist with hard work, long training, harsh
convergent
Integrated creativity
divergent
/
/
,
/
,
/
,
,
/
/

fOCI15'e'd

/
/ solving',
/
,
/
,
,
/
,
,
/
/
,
/ ,
, /
,
/
,
, creatjve
creaqvity prlem
,
,
,96lving ,
,
,
/
/
,
,
/
primary
process
secondary
process
criticism and perfection. He also noted that the spontaneous was
followed by deliberation; total acceptance was followed by critique;
intuition was followed by logic; fantasy was followed by reality testing.
And actually, these descriptions of integrated creativity agree well with
the implications of the depicted model.
Process models of Creativity
The first to describe different stages of creativity was the German
physiologist and physicist, Helmholz, who, at the turn of the 19
th
.cen-
tury, described three stages of creativity (saturation, incubation and
illumination). A fourth stage (verification) was added in 1908 by the
French mathematician, Poincare. The last phase (which was actually
the first stage of the creative process) was added by the American psy-
chologist Getzels in the 1960s: the problem-finding stage or the first
insight (Edwards, 1987).
The creative process consequently consists of the following five
phases:
1) First insight, 2) Saturation, 3) Incubation, 4) Illumination and
5) Verification.
Betty Edwards (1987) has linked holistic and sequential thinking to
these phases:
1 The first insight often derives from curiosity or from a question.
Somebody starts wondering about something. This mainly consists
of holistic thinking.
2 The stage of saturation involves logical thinking as this concerns
obtaining information and data about the problem or question.
3 Incubation is described as a period where the information is
reflected upon and cultivated, and then thrust away from con-
sciousness. This is mainly a process of holistic thinking.
4 Illumination usually happens after a period of relaxation or of do-
ing cyclical work (e.g. weeding the garden, taking a shower). It is
described as a lightning 'AHA' (holistic thinking).
5 Then follows verification where the idea or the solution is 'ra-
tionalized' through processes of logic.
166
The Creative Problem Solving Model (CPS)
The CPS model consists of 6 stages, of which each has a divergent and
a convergent phase. "The current model is built on the belief that
effective problem solving relies upon both creative and critical think-
ing." (Isaksen, 1986: 154) The stages are:
Mess Finding
Data Finding
Problem Finding
Idea Finding
Solution Finding
Acceptance Finding
Isaksen describes creative thinking as divergent: It aims at making new
connections in order to generate many, different and unusual possibil-
ities. Critical thinking is convergent and analytical and concerns com-
paring, improving, selecting, judging, and deciding.
The CPS model is a stage process model, but Isaksen (1998:151)
notes that real creative problem solving is a 'messy' process, which will
depend on the specific task and situation, and that some of the stages
will be accentuated and others left out. The CPS model includes 'Ac-
ceptance Finding', which concerns the implementation of the solu-
tion. This is unusual, but constructive. Often the creative process stops
when a thousand ideas have been generated - and then what? It also
happens that ideas are screened and one solution selected, but the
adoption of the solution by the organization is often ignored - detri-
mental to the result, i.e. losing the (opportunity for) innovation.
The '4 Steps of Problem Solving' process model
Herluf Trolle, who has worked practically with creativity for many
years, has integrated the main ideas of Synectics and CPS into a pro-
cess model, 'Steps of Problem Solving', including the following four
steps (1988):
Problem Formulation
Problem Reformulation
Idea Development
Idea Improvement
Trolle explains the process of problem formulation as a process that
can be focused on general or more specific levels. With 'why' ques-
tions a 'problem' can be taken to a more general level of thinking
(involving goals and motives), whereas 'how' questions pull the prob-
lem into a more concrete mode (involving means). Trolle proposes a
goal/means problem hierarchy (1988: 10) with a general problem at
the top being divided into different formulations of the problem,
indicating totally different means. Thus, Trolle has found a way of
operationalizing Schon's (1986:264) thoughts about 'naming and
reframing'. Furthermore, Trolle's suggestions of reformulating the
problem incorporate the perspective of diversity in a
way, which is relevant for the present book. Trolle's model is relevant,
operational and systematic.
Creativity and group interaction
How does creativity affect group interaction and how does group
interaction affect innovative processes? A quote by Edward de Bono
from Metcalf and Felible answers (1992, in Lumsden & Lumsden,
1997:201):" ... But it is the willingness to play with ideas, to risk fool-
ishness without fear, that are the hallmarks of the creative thinkers.
And it is the creativity which springs from humor that increases our
effectiveness."
It seems that one of the major impacts that creative sessions have,
apart from generating ideas, has to do with the atmosphere. The
demands of postponing judgment, maintaining a positive attitude and
a general looseness provide a supportive comfortable climate. It is
healthy to laugh together and laughter can relieve a lot of tension in a
task group.
Regarding how group interaction can affect innovative processes,
we found a suggestion by Leonard & Sensiper (1998): "When a group
of diverse individuals addresses a common challenge, each skilled per-
son frames both the problem and its solution by applying mental sche-
mata and patterns he or she understands best. The result is a cacoph-
ony of perspectives. In a well-managed development process, these var-
168
ying perspectives foster creative abrasion, intellectual conflict between
diverse viewpoints producing energy that is channeled into new ideas
and products."
Creativity and creative problem solving techniques can be helpful
in generating new ideas and angles to solve problems as well as for
building up a constructive group climate. The need for integrative
creativity is related to the task, the participants, the context, and in
particular to the type of problem. Creativity may give the spark or light
the fire, may 'kick-start' a stranded project, but it takes more than
creativity to make innovation happen.
Let us conclude the paragraph with a quote that underscores this line
of thought:
"The quote 'Genius is one part inspiration and ninety-nine
parts perspiration' has been attributed to at least a half-dozen
people. I would like to paraphrase it and say that "Innovation is
one part creativity and ninety-nine parts productivity"
(Gamache, 1988)
Crystallization
"We could review all the OpInIOnS offered to explain why an
open controversy closes, but we will always stumble on a new
controversy dealing with how and why it closed. We will have to
learn to live with two contradictory voices talking at once, one
about science in the making, the other about ready made science
... The left side considers that facts and machines are well deter-
mined enough. The right side considers that facts and machines
in the making are always under-determined
22
."
(emphasis original) (Bruno Latour, 1987: 13)
The overarching goal of problem search, problem framing and know-
ledge creation is crystallization. The ideal type of crystallization is the
moment when everything falls into place, when opportunities emerge,
when the strategy or the solution becomes clear, when the problem
can finally be formulated and when the group breathes a sigh of relief
or shouts 'AHA!' Crystallization resembles 'illumination', the fourth
phase of the creative process, described earlier.
Emergence
Emergence is the unfolding of potential. "When simple rules are
allowed to operate for enough time or on enough elements, new qual-
ities emerge, break out, surface, appear" (N0rretranders, 1991 :445,
my translation). Emergence is not seen when studying a small ar:nount
of elements; emergence appears only when these characters have
accumulated to a degree that it has collective effects, group attributes.
Thus emergence cannot be studied by studying particular elements.
This is exemplified by measuring temperature in a liquid. Temperature
cannot be measured by measuring particular molecules; temperature
is a collective thing, expressed by the dispersion of speed among many
molecules. "In emergence, global patterns cannot be predicted from
the local rules of behavior that produce them. To put it another way,
global patterns cannot be reduced to individual behavior." (Stacey,
1996: 287).
One of the challenges of this study is that emergence cannot be
foreseen, calculated beforehand or deducted from studying the par-
ticular elements - and that it must necessarily be studied in its prop-
er context and in real-time. "Tight control is achieved at the expense
of lost potential. Fullest potential is achieved by letting the system
unfold, emerge, as it will. No amount of controlled intervention can
foresee and realize emergent possibilities." (Zohar, 1997:53)
Archetype
A way of understanding emergence is through 'archetypes'. Stacey
defines an archetype as (1996:283):
"A potential behavior that preexists experience and awaits spe-
cific experience to be actualized or realized. Although the arc-
hetype exists in a recognizable general form, its specific actu-
17
alization is always unique and depends upon the specific expe-
rience. An archetype is therefore a similar concept to an imma-
nent, implicate, or enfolded order as used by Bohm (1980)
and by Aristotle. It is also similar to Plato's concept of ideal
form."
The key elements are the disposition or potential in relation to the
actual experience. "When I use the terms archetype, potential, or dis-
position, I am seeking to describe a possibility that is enfolded in a set
of rules of interaction." (Stacey, 1996:55)
What this entails is that by changing the rules of interaction, new
archetypes may be realized.
In Stacey's framework crystallization would be the actualization of
archetypes. In a group this happens through building up the reces-
sive schemas of the group. Groups, like individuals - and organiza-
tions, have dominant and recessive schemas. We know from the
above that cognitive schemas are filters for our perceptions, interpre-
tations and actions. Dominant schemas of a group are related to pro-
cedures, norms, functions, responsibilities, and formal organization.
Recessive schemas concern much of what goes on at another level of
interaction, the 'shadow' system
23
. This is related to emotional rela-
tionships, political maneuvering, fantasizing and other non-verbal
behavior. Crystallization can then be explained as a change of the
dominant schemas of a group, a change that emerges from the shad-
ow system as a process of interaction in the recessive schemas. Or it
can be explained as something emerging at the right moment, 'kairos'
(Kirkeby, 1998:77).
In real-time situations, however, crystallization is usually not that
simple, even though it may seem simple in hindsight. In groups working
with the creation of new knowledge or new leads, the question of
when 'enough is enough' (knowledge) is often debated. Sometimes a
deadline is the decisive factor for a group having to produce some-
thing. Then they must use the collected data and knowledge to make
a proposal, which is the best they can provide under the given condi-
tions. This is not crystallization - but closure.
17
1
Innovative crystallization
We shall conclude by setting up some criteria for recognizing 'innov-
ative crystallization' and then attempt a definition. It is not unusual for
authors to describe that something has 'crystallized'. E.g. Nonaka &
Takeuchi write (1995:86). "The shared tacit mental model is verbal-
ized into words and phrases, and finally crystallized into explicit con-
cepts." I have not, however, found the concept of 'Innovative Crystal-
lization' in others' writings. I shall argue that such a concept is needed,
because it can be distinguished:
from a decision - even though a decision is implicit in it
from a solution - even though it is sometimes a solution to a com-
plex full-fledged problem
from an idea - even if the end result may resemble a perfect and
finished idea
from an innovation - even though it definitely involves novelty
and from a conceptualization - even though it involves cognitive
elements.
Characteristics of innovative crystallization
Most of all, it resembles 'illumination', the fourth phase of creativity,
but whereas illumination is individual, innovative crystallization emerges
out of collective interaction.
It is a stepwise process of accumulation and integration of informa-
tion and knowledge.
It incorporates many ideas and multiple perspectives.
It involves a transformation that produces novelty.
It synthesizes tacit knowledge with the accumulated and integrated
knowledge material into an overall model, a substantive form, a
prototype, or a new conceptualization.
It is simple and complex.
It yields enthusiasm and creates immediate commitment from the
persons involved.
It needs to be examined, adapted and tested before it is ultimately
accepted.
17
2
In sum, innovative crystallization is defined as the outcome of
a process involving collective transformation of accumulated
and integrated ideas into a new conceptualization or proto-
type.
Concluding remarks
The field of creativity is large, and what we have included here is far
from exhaustive. We have mainly tried to uncover the boundaries
between creativity and innovation. Creativity contributes to the study
of innovative processes primarily as a source for delimitation. We clar-
ified the boundaries between creativity and scientific search through a
discussion of cognitive psychology's different thinking styles, and we
argue that creativity is not a necessary ingredient for innovative crys-
tallization to occur. Creative methods can, however, add a different
quality to the conceptualization process, and will often make the con-
tribution more original.
Innovative crystallization is a focal concept of this study. When coin-
ing a new concept it is often easier to describe it through what it is not
than to come up with a straightforward definition. We found a resem-
blance between crystallization and 'illumination', the only difference
being that illumination is an individual process, whereas crystalliza-
tion is a collective process. Crystallizations emerge out of group inter-
action and sharing of knowledge.
Creativity can, apart from supporting idea generation, help create a
pleasant group climate where people feel confident about one another.
And, as we shall see in the next section on communication, that is
important.
To sum up, creativity has contributed important material for delim-
itation, differentiation, and substantiation of concepts and prior dis-
cussions.
173
Communication
"There is only one language, the language of the heart" (Sai Baba
24
)
Introduction
The purpose of this section is to construct a theoretical communica-
tion framework, which can be applied in the case studies. Related to
the second objective of the book, the development of a 'special theory'
of communication, we will deal with the two first component.s: the
philosophical rationale and an ideal model based on rules. The third
component, recommendations for improving practice, will follow in
chapter 9, after the case study narrative (chapter 5) and the data an-
alysis (chapters 6 and 7). In this section we will look at two opposite
philosophical rationales and develop a theoretical framework for com-
munication, which consists of three ideal-type models.
We start with a delimitation and definition of the concept of com-
munication and a discussion of the traditional 'tube' model of commu-
nication by Shannon & Weaver (1948), which forms the background
for the majority of models on information and communication. Then
we describe the development of two ideal type communication models,
the 'Genuine Communication' model, inspired by Bormann (1996)
and the 'Collective Monologue', inspired by Hewes (1996). We attempt
a dialectical discussion of these by examining three dimensions: the
context, 'listening' or 'hearing', and the message.
After introducing and discussing the third ideal-type model, the
Antagonistic Dialogue, we summarize the chapter with a presentation
of the communication framework, consisting of the three models.
Delimitation
"Spoken discourse among persons interacting on matters of
mutual concern and coming to agreement on an appropriate
course for common action is the social atom - the building block
ofhuman society and human social activity." (Scheidel, 1986: 113)
174
Communication is basic and penetrates human life at all levels.
Even animals communicate. In fact, everything is communication.
This idea derives from Paul Watzlawick who said, "One cannot not
communicate" (Lieth, Kuschel & Petersen, 1991 :7). Thus, we need
to delimit communication to encompass only interpersonal commu-
nication in relation to knowledge creation in groups. We could, of
course, make the concept still more operational by narrowing the
scope to cover only communication that is intended, perceived and
reacted upon, as proposed by D.M. McKay (in Dars0, 1992). This
is tempting, but would be to 'throw out the baby with the bath
water', as the focus of this book is on complex processes. Reducing
the complexity too much would give a false appearance of simplic-
ity. Summing up:
In the present study communication concerns interpersonal
communication in groups and includes conscious or non-con-
scious, intended or not intended, verbal or non-verbal sharing
and interchanging of words, meanings, sentiments and mes-
sages.
Definition
The concept of communication comes from Latin and means 'to share
with'. As Hatch (1997) pointed out in relation to culture, sharing has,
however, two contrary meanings: One emphasizing similarity and the
other emphasizing separateness. An example of the first is when shar-
ing an experience and thereby creating 'sameness' by having that in
common. An example of the second is sharing a cake. In this case
sharing means dividing the cake into individual pieces and distribut-
ing them among the group. Hatch incorporates the double meaning of
sharing this way (Hatch, 1997:206):
"Sharing (culture) means that each member participates in and
contributes to the broad patterns (of culture), but the contribu-
tions and experiences of individual members (of the culture) are
not identical." (Parentheses added)
175
We will adopt Hatch's point, but add one more ingredient. As we will
develop that ingredient into more depth shortly, we will only intro-
duce it here. Communication requires listening, and there is a differ-
ence between listening, which is active, and hearing, which is passive.
Listening is essential for communication. Thus:
Q encoding
/\ )
noise
~
message
commnnication channel
decoding Q
)/\
Communication consists ofsharing between members in such
a way that each member listens, participates in and contrib-
utes to sharing, but the contributions and experiences of indi-
vidual members are not identical
Shannon &Weaver's 'Tube' model of communication
Unintentionally it seems, Shannon & Weaver's model has become the
background as well as the battleground for studies of information and
communication. Earlier in this chapter we used a figure by Max Boisot
to illustrate the difference between data, information and knowledge.
Information is, to use Bateson's famous quote, a 'difference that makes
a difference', meaning that what appears as information is data that
was different enough to be "let through" the individual's perceptional
and conceptional filters.
The original meaning of information concerned something that
was 'indeterminate', contrary to how it is used today. According to
Nonaka & Takeuchi (1995:90), Shannon used only the 'syntactic'
dimension of information and saw it as the 'difficulty in transmitting
the sequence produced by some information source'. As engineers in
the Bell Labs, Shannon & Weaver were mainly interested in noise
reduction and in being able to transfer signals (messages). They were
not interested in the semantic aspect, the meaning.
The model is illustrated below: a sender encoding a message, send-
ing it through a communication channel, to be decoded by the re-
ceiver. The 'tube' metaphor has influenced models of communication
since it came out. Certainly, it has been modified, changed, and added
to, but for many decades this has been the main model.
17
6
An illustration of Shannon & Weaver's communication model
The 'tube' metaphor is criticized by Maturana & Varela (1998: 196):
"From the perspective of an observer, there is always ambiguity in a
communicative interaction. The phenomenon of communication
depends not on what is transmitted, but on what happens to. the per-
son who receives it. And this is a very different matter from 'transmit-
ting information'." Thus, what Maturana & Varela object to is Shan-
non & Weaver's focus on the congruity of the message (on the
mechanistic noise reduction), rather than focusing on the meaning
(semantics) of the message (and its effect). The question is, however,
whether Shannon & Weaver should be blamed? Mter all, they were ful-
filling a prescribed purpose of an engineering job. Should the critique
not be addressed to the people who adopted the model?
Developing the model of 'Genuine Communication'
My own first model of 'genuine communication' was developed on the
conceptual foundation of Shannon & Weaver's model (Dars0, 1995).
It has also been inspired by other contributions, e.g. the model of
intercultural communication by Samovar & Porter (1985). After some
reflection, however, I saw how using Shannon & Weaver's communi-
cation model as a fixed framework had, in fact, limited my thinking.
Actually, I had a quite different model in mind. I found that the model
of 'genuine communication', the essence of which is to build a
'common ground', was much more in line with the symbolic conver-
gence theory, developed by Ernest Bormann and others during the
1970s and 1980s. Symbolic convergence focuses on the formation of
a group consciousness through shared concepts, ideas and motives.
177
We will adopt Hatch's point, but add one more ingredient. As we will
develop that ingredient into more depth shortly, we will only intro-
duce it here. Communication requires listening, and there is a differ-
ence between listening, which is active, and hearing, which is passive.
Listening is essential for communication. Thus:
Q encoding
/\ )
noise
~
message
communication channel
decoding Q
)/\
Communication consists ofsharing between members in such
a way that each member listens, participates in and contrib-
utes to sharing, but the contributions and experiences of indi-
vidual members are not identical
Shannon &Weaver's 'Tube' model of communication
Unintentionally it seems, Shannon & Weaver's model has become the
background as well as the battleground for studies of information and
communication. Earlier in this chapter we used a figure by Max Boisot
to illustrate the difference between data, information and knowledge.
Information is, to use Bateson's famous quote, a 'difference that makes
a difference', meaning that what appears as information is data that
was different enough to be "let through" the individual's perceptional
and conceptional filters.
The original meaning of information concerned something that
was 'indeterminate', contrary to how it is used today. According to
Nonaka & Takeuchi (1995:90), Shannon used only the 'syntactic'
dimension of information and saw it as the 'difficulty in transmitting
the sequence produced by some information source'. As engineers in
the Bell Labs, Shannon & Weaver were mainly interested in noise
reduction and in being able to transfer signals (messages). They were
not interested in the semantic aspect, the meaning.
The model is illustrated below: a sender encoding a message, send-
ing it through a communication channel, to be decoded by the re-
ceiver. The 'tube' metaphor has influenced models of communication
since it came out. Certainly, it has been modified, changed, and added
to, but for many decades this has been the main model.
An illustration of Shannon & Weaver's communication model
The 'tube' metaphor is criticized by Maturana & Varela (1998: 196):
"From the perspective of an observer, there is always ambiguity in a
communicative interaction. The phenomenon of communication
depends not on what is transmitted, but on what happens to the per-
son who receives it. And this is a very different matter from 'transmit-
ting information'." Thus, what Maturana & Varela object to is Shan-
non & Weaver's focus on the congruity of the message (on the
mechanistic noise reduction), rather than focusing on the meaning
(semantics) of the message (and its effect). The question is, however,
whether Shannon & Weaver should be blamed? Mter all, they were ful-
filling a prescribed purpose of an engineering job. Should the critique
not be addressed to the people who adopted the model?
Developing the model of 'Genuine Communication'
My own first model of 'genuine communication' was developed on the
conceptual foundation of Shannon & Weaver's model (Dars0, 1995).
It has also been inspired by other contributions, e.g. the model of
intercultural communication by Samovar & Porter (1985). After some
reflection, however, I saw how using Shannon & Weaver's communi-
cation model as a fixed framework had, in fact, limited my thinking.
Actually, I had a quite different model in mind. I found that the model
of 'genuine communication', the essence of which is to build a
'common ground', was much more in line with the symbolic conver-
gence theory, developed by Ernest Bormann and others during the
1970s and 1980s. Symbolic convergence focuses on the formation of
a group consciousness through shared concepts, ideas and motives.
177
The theory describes how the sharing of fantasies or visions in relation
to the task influences communication and decision making. In the
words of Bormann (1996:89): "Symbolic convergence creates a sym-
bolic climate and culture that allow people to achieve empathetic
communion as well as a 'meeting of the minds'." In this respect, it
came very close to the essence of the model of genuine communica-
tion, which is sharing. Also Nonaka & Konno's (1998:40) concept of
'Ba' concerns sharing: "Ba can be thought of as a shared space for
emerging relationships."
In the following I will outline two contrasting communication tpodels:
the model of 'Genuine Communication' and the model of 'Collective
Monologue' .
"We cannot leave language, for then we cannot talk to one
another. But neither can we say what we would like to say, for
we have only language to communicate with."25
(N0rretranders, 1991 :380)
To the extent, however, that an idea can be rephrased by another per-
son through his own words or by model or metaphor to the satisfac-
tion of the originator, this will pertain to the concept of 'genuine
communication'. The model distinguishes itself through two import-
ant features: 1. the building of 'common' ground, and 2. the forming
of a relational bond, which in this model is considered a precondition
for the development of real sharing. Shotter provides the following
'ideal type' image of sharing (1993:146):
Genuine Communication
(inspired by Bormann, 1996)
Collective Monologue
(inspired by Hewes, 1996)
"But we can, I think, at least say this: first, that there are certain
special moments which one shares with others - I shall call them
moments of common reference - in which, as two people (two
beings) regard one another and their common situation, they
know from each others' 'attunements', as I shall call them, that
they are each sensing it in the same way ... The second thing is
that in such moments, one's expressions can work to give a
shared significance to such shared circumstances ... "
Collective monologue
Genuine communication
The model of genuine communication is based on the philosophical
rationale that communication is possible. Genuine communication is
the result of a successful communication process where, ideally, 'true
sharing' has taken place. Perfect congruence of understanding is, of
course, an abstraction, and it is easily argued that this is not possible,
as you will only be able to understand the life of a person if you have
lived it. Another limitation is the inadequacy of language.
The second model of communication is somewhat provocative. The
underlying rationale is here that real communication is neither pos-
sible nor intended. The 'socio-egocentric model' by Dean E. Hewes
(1996) claims that what is going on in decision making groups is not
communication at all, but rather 'collective monologues', character-
ized by turn-taking and 'vacuous acknowledgments'. The model is
inspired by Jean Piaget's empirical studies of children's 'egocentric
speech', a pattern that exists among young children. At first glance the
children seem to be engaged in some sort of dialogue. On closer study,
however, each child concentrates on her own conversation and stays
with that only. Piaget's explanation is that it is difficult for children at
179
that age to contain both their own thoughts and those of the other
child. Hewes draws a parallel to conversations or meetings of adults,
where it may seem initially that the participants are communicating,
as they are dealing with the same subject and taking turns in speaking
about it. When examined more closely, however, people are actually
talking to themselves, following their own trains of thought, and at the
same time pretending to communicate by utilizing 'vacuous
acknowledgments'. These are polite statements that ostensibly link the
messages, but which, when analyzed in more depth, have no semantic
links at all. Hewes' model is original and critical towards much
research on group communication.
Context
"I have called the kind of knowledge required to be able to talk
and to understand in this self-specifying way, knowing of the
third kind: it is a knowing from within a discursively constructed
situation; that is, from within an event. As such, it is a form of
knowledge whose nature cannot be described theoretically, in
ways amenable to evidential support. Even to try to do so would
be paradoxical: for we want an account of it in practice, a con-
texted understanding of it from within the context of its use, ... "
(Shotter, 1993: 113)
Context encompasses the physical, social and psychological environ-
ment of a situated event 'alive in its present', of 'events in their going
on now', (Shotter, 1993: 110-111). Context is the 'sister' of praxis. In
his doctoral thesis, Flyvbjerg deconstructs the concept of theory in the
social sciences, and replaces it with context and practical rationality
(1992: 159).
Communication is a complex whole that is dependent on both
rationality and intuition, culture and context and synchronicity
between the people involved. A useful model for (intercultural) com-
munication involving diversity is the model of High-Context and Low-
Context Communication by Edward T. Hall (1974). The model illus-
trates how meaning is generated in conversations of different cultures.
180
In a High-Context culture more meaning is derived from the context
than from the message itself. This means that, for instance, attention
must be directed towards the situation, to prior history, to gestures
and tone of voice, pauses and subtleties. Sensitivity to the occasion
and the situation is more necessary than listening to what is actually
said. (Classical) Japan is supposedly a good example of a High-Con-
text culture. Low-Context cultures, on the other hand, attach mean-
ing to the message itself, having context playa minor part. This form
of communication is very direct, aiming at efficiency. Europe and
North America are typical Low-Context cultures. Many failures of
communication and cooperation between High-Context and Low-
Context cultures have been reported. Hall's model is relevant for the
present study, partly because of the focus on differences, partly
because 'context' is relational, and 'information' conceptual, like our
levels of analysis. Furthermore, examining how meaning is generated
is central for communication and knowledge creation.
High-Context
Fig. 3.6.4 Hall's model of High-Context
and Low-Context Communication
Low-Context
Group Climate
An important part of the context is the psychological environment or
the group climate. Among the first to point out the significance of
group climate was Jack R. Gibb (1964). His theory is based on
research in various T-groups (Training groups) as well as on a large
number offield studies in industries and organizations. Gibb clustered
four general dimensions of social interaction: Acceptance, Data flow,
Goal formation, and Social control.
GROUP CLIMATES
Acceptance concerns the formation of trust and acceptance of self
and others and of growing confidence. It is linked to group member-
ship. Data-flow concerns all data, feelings as well as perceptions and
attitudes, which finally produce decision making or choice. Goal-for-
mation relates to the integration of the group members' motivations
regarding producing something, whether it is learning, growing or
performing a task. The control dimension has to do with regulations
of group behavior, i.e. to group organization. Gibb proposed two
ideal-type models of group member behavior: a) persuasive behavior,
and b) participative behavior. Below is a table contrasting these two
behaviors:
PERSUASIVE BEHAVIOR
1. fear
distrust
2. strategy
facade
3. manipulation
persuasion
4. control
bargaining
Gibb (1964:294)
PARTICIPATIVE BEllAVIOR
1. confidence
trust
2. openness
spontaneity
3. Self-assessment
problem solving
4. permissiveness
interdependence
data, false assumptions, inadequate theory. Regarding the goal 3),
reactions would arise like resistance, low commitment, manipula-
tion, competition, a need for structure, which would finally lead to
4) chaos, disorganization, dependency, hostility (often latent), pow-
er struggles.
On the other hand, if acceptance was developed in a group, then
the participative track would be highly possible with 1) trust and
acceptance, diversity and nonconformity, 2) clarity, problem-solving
behavior, open expression of feeling and conflict, 3) work orientation,
creativity, reduction of competitive behavior and conflict, and 4) inter-
dependence, allocation of work by consensus or ability, informality
and spontaneity.
Gibb's conclusion was (1964:298): "The relevant dimensions of
growth of groups seem to be in the direction of supportive climate,
reality communication and feedback, maximal goal integration, and
functional interdependence in action and structure." Gibb's findings
have later been complemented by Ekvall (1991), who found that an
innovative climate was characterized by dimensions of perceived high
challenge, freedom, idea-support, trust, dynamism, playfulness, de-
bates and risk taking.
Discussion
As this discussion is based on the analytical abstraction of ideal type
models, we will draw up some sharp lines, but remind the reader that
it is not our intention to simplify real life communication. 'Ideal types'
are extremes that enhance understanding.
According to Gibb, the four dimensions are hierarchically contin-
gent upon each other, as each sets boundaries for dimensions lower
in the hierarchy, the decisive entry point for behavior in a group be-
ing acceptance, the first dimension. If that fails to develop, most
groups would get trapped in the persuasive mode, like a vicious cir-
cle of: 1) distrust, fear, resistance, cynicism, 'polite' behavior. This
would lead to 2) ambiguity, tricks, facade building, distortion of
In 'genuine communication' context and climate playa major part.
Gibb found that acceptance and trust were the building blocks for
group processes, communication, decision making and outcomes.
Stephen Covey takes this a bit further in a matrix with axes of trust
and cooperation:
High
TRUST
LEVELS OF COMMUNICATION
Syner . 'c (WinlWin)
Respe ul (Compromise)
where context matters more than information. The exact combination
of context and information will, of course, depend on the purpose of
the group. In Western business contexts, which tend to pull towards
Low-Context, i.e. more weight on information, it would be unusual to
find groups who would invest (even) equal attention in context and
information. This seems, however, to be the strength of many Eastern
companies (Nonaka & Takeuchi, 1995). In relation to the matrix of
diversity and similarity (see page 113), sharing builds on perceived
similarity and tends to produce 'smooth in-group interaction'.
The lowest level of communication, the Defensive, is characterized
by defensiveness, rules and regulations. The outcome is often
Win/Lose or LoselWin. This is similar to Gibb's 'persuasive' behavior
and further substantiated in Gibb's article on 'Defensive Communi-
cation' (1961). The middle level with medium trust and medium
cooperation is called Respectful communication. Here people are
polite, but avoid conflicts. Often the result is a compromise. Accord-
ing to Covey, however, a compromise means that 1 + 1 = 11/2. It can
still be a lowWinIWin solution, but the creative potential is not used.
Finally, the highest level is Synergistic communication, which is a
totalWinlWin solution, where "1 + 1 may equal 8, 16 or even 1,600."
(Covey, 1992:271). This form of communication is highly creative,
enjoyable and benefits from valuing differences between people and
their paradigms. In comparison, Gibb's 'participative' climate is simi-
lar to Respectful communication and has within it the potential for
Synergistic communication.
Low
De sive (Win/Lose or LoselWin)
Low
COOPERATION
Stephen R. Covey, 1992:270
High
As for the other extreme, the 'collective monologue', context is less
important. This type of communication is superficial. People are, so to
speak, 'skating on ice', greeting the people they meet in a friendly way,
shouting 'Hi, how are you', and are gone before they can hear the
answer. In the matrix of diversity and similarity, the interaction is
based on not perceived similarity or diversity (as perception is totally
self-centered), which we have called superficial relations and interac-
tion. Participants of the 'collective monologue' are mostly interested
in themselves. In an interview Hall explained the difference of High-
Context and Low-Context cultures this way (Bluedorn & Hall, 1997):
"I like to say that more information is transmitted in mono-
chronic
26
/low context cultures than in polychronic/high context
cultures, but more information is shared in the polychronic than
in the monochronic cultures. As an example of context's import-
ance, think of the situation when somebody describes an event
that had everyone in stitches when it happened, but that doesn't
even produce a smile when it is described to people who weren't
there. The teller usually shrugs and says, "I guess you just had to
be there." "Being there" is a reference to context and illustrates
the point that not all of the meaning is contained in the message
(the story)."
In fact, a climate of acceptance and confidence is the foundation for
'genuine communication'. In relation to Hall's figure of High-Context
and Low-Context communication, ultimate sharing is High Context,
In Hall's terminology the 'collective monologue' is Low Context. Still
one dimension is, however, missing: that of attention and listening.
Listening or hearing?
In the traditional 'tube' model of communication, communication
would always take place between a 'sender' and a 'receiver'. The
sender would have to have skills in encoding the message in order for
the receiver to be able to decode, i.e. understand it. The former will be
discussed shortly under the headline of 'the message'. The latter is the
subject of this paragraph, where we shall try to illustrate the signifi-
cance of listening skills for communication. Stephen Covey
(1992:237) has made the following important point:
"If I were to summarize in one sentence the single most import-
ant principle I have learned in the field of interpersonal
relations, it would be this: Seek first to understand, then to be
understood. This principle is the key to effective interpersonal
communication."
According to Covey, few people have had any training in listening,
whereas most people spend years on learning how to read and write
and how to speak. The greatest problem is that people do not listen to
understand, most people listen to reply and are either speaking or pre-
paring to speak. People often give solutions before they know the
problem. They 'prescribe' before they 'diagnose'.
Covey lists four ways of 'listening' that people normally practice.
The fifth way, empathetic listening, is rarely taken into use.
1. ignoring the other, simply not listening at all
2. pretending, saying sounds of confirmation, but not listening
3. selective listening, putting our attention to certain parts
4. attentive listening, paying attention to the words, as in 'active
listening', where you repeat the words of the person in order to
check your understanding
5. empathetic listening is seeking an understanding of the person
that goes beyond words, i.e. an understanding based on empathy,
where you both emotionally and intellectually seek to understand
the other.
186
The French doctor Alfred Tomatis, who has conducted research on
the functions of the ear for 50 years, found that auditive perception
takes place through the ears, the skin and the skeleton
27
. Interesting-
ly, he found two distinct functions: one active process of focusing, fil-
tering, remembering, and reacting - the activity of 'listening', and one
passive process of letting auditive impressions pass, i.e. 'hearing'.
Tomatis has proved his 'listening-therapeutic' theories by teaching
deaf people to listen, and even to play music in symphony orchestras
28
by feeling the vibrations of music through their skin and skeleton. The
point is that they can learn to listen, even if they will never be able to
hear.
Covey's taxonomy is built on an increasing degree of listening, start-
ing with 'hearing' (the first three steps) and gradually turning into
'listening'. Before discussing this in relation to our communication
models, we will add one more dimension.
Emotional Intelligence
The concept of emotional intelligence derives from Howard Gardner's
concept of multiple intelligences. One of Howard Gardner's intelli-
gences is 'interpersonal intelligence', which involves the ability to work
cooperatively in a group and to communicate with others. Emotional
intelligence includes five parameters
29
(Goleman, 1996:43), of which
two are related to communication: Recognizing emotions in oth-
ers (the essence of this is empathy, the ability to feel what other peo-
ple feel) and handling relationships, a cornerstone of leadership
and interpersonal effectiveness. It has been found in several investiga-
tions that individuals with a high emotional intelligence often do bet-
ter than their more intelligent (higher IQ30) peers. In fact Goleman
states (1996:36): "Emotional aptitude is a meta-ability, determining
how well we can use whatever other skills we have, including raw
intellect" .
Discussion
Tomatis's findings regarding the difference between 'listening' and
'hearing' highlight Covey's taxonomy. Genuine communication is
related to listening and to Coveys 'attentive' and 'empathetic' listen-
ing. Furthermore, genuine communication can be enhanced through
interpersonal and emotional intelligence. Collective monologue, on
the other hand, is related to 'hearing' and to the first three of Covey's
listening modes: ignoring, pretending and selective listening. In meet-
ings characterized by 'collective monologue', most participants are so
eager to prepare their own contribution that they only ' l i s t ~ n ' very
superficially in order to catch words that will allow them to enter into
the conversation.
The message
Interpersonal communication is a two-way dynamic process, taking
place at multiple levels. Above we discussed how communication is
received, actively or passively. Here we want to examine how commu-
nication is prepared and expressed. Encoding concerns the ability to
express thoughts or ideas clearly. Rhetoric concerns presenting it con-
vincingly to others. If we start with the communicator's intention; this
could be to be understood - or the opposite. In the first case the object-
ive could be anything from creating interest, getting information
across, inducing commitment - to producing change or action (Dars0,
1995). When the intention is to be understood, the sender has the
greatest chance of 'getting across' by understanding the recipient and
his or her context/situation. One would have to present things different-
ly to a colleague with a different background than to a colleague with a
similar background. Likewise, the selection of content should be differ-
entiated, as different data or different arguments would be convincing
to different audiences. What this boils down to is, in fact, emotional
intelligence: the interpersonal skill of empathy. S0ren Kierkegaard
expressed it this way: "That if, indeed, one shall succeed in guiding a
human being to a certain place, one must first and foremost take care
to find him where he is, and start there." (M0nsted, 1991)
188
In some cases, the intentions of the communicator is not to be under-
stood. Ambiguous messages are also part of conversations, but they
often go undetected. This ambiguity may be conscious as when man-
agers leave certain decisions open for interpretation in order to be
able to adapt or change the decision later. Or ambiguity may be non-
conscious. Many knowledge workers and 'experts' are masters at pro-
ducing eloquent messages that sound convincing, but which, upon
further analysis, turn out to be personal opinions (but sound like
facts), 'cover-ups' for not-knowing or not wanting to tell - or pure
'platitudes' .
Rhetoric
The Greeks described three forms of appealing to others: ethos,
pathos and logos. Ethos is related to ethics and expresses the person-
ality or 'image' of the speaker, i.e. the authenticity, the integrity, the
trustworthiness, the history. Ethos is built slowly and is easily worn
out. Pathos appeals to feelings, sentiments and atmosphere. It
appeals to spontaneous reactions. A good example of pathos is Martin
Luther King's famous speech of 1963 'I have a dream'. Finally, logos,
related to logic, appeals to the intellect of the recipient. This type of
presentation is rational and objective, based on data and facts.
In communication studies two different approaches are: the self-
other, rhetorical-responsive, dimension of interaction as opposed to
the person-world, referential-representational, dimensions of interac-
tion. In social constructivism the former is considered primary and
the latter secondary: "The account of language offered is a communi-
cational, conversational, or dialogical account, in which people's
responsive understanding of each other is primary." (Shotter, 1993:8)
Shotter points out the formative power of words and language, where
words can materially 'move' others and thus transform people and
organizations.
Certainly, this 'moving' power is most often attached to leaders. A
rhetorical leadership style of managers can persuade followers to
identify with their vision. McCloskey (1998
31
) considers 'sweet talk' a
major instrument of management, the rationale being that if you can
convince people of something, they will do it. Shotter argues that good
managers not only 'make history' but also should be seen as 'authors'.
In that sense, language is power, not only of leaders, but also in every-
day conversations (Shotter, 1993:20):
"By contrast, the studies in this book, display an interest in the
contested activity of words in their speaking; that is, in the practi-
calities of their use as means or as 'tools' in effecting everyday
communicative processes, and in particular, in their formative or
'shaping' function, and the 'resistances' they meet, in such pro-
cesses. Thus the stance I take in all the following chapters, is that
in an everyday process involving a myriad of
responsive, practical, unselfconscious, but contested interac-
tions, we unknowingly 'shape' or 'construct' between ourselves,
as already mentioned, not only a sense of our own identities, but
also a sense of our own 'social worlds'."
Discussion
There is a profound difference between a person who communicates
to be understood and a person who communicates to hear himself
speak. At a surface level, they both seem to convey a message, and, in
fact, both may influence others. In genuine communication, a deeper
understanding is generated when each person carefully 'frames' their
utterances and each person applies listening skills. The formative
power of words and language can be supportive in creating 'common
ground'. In collective monologue, the focus is more on the self than
on others. As this 'communication' is superficial, utterances become
rhetorical in a correspondingly superficial way, using 'buzz words' or
provocation in order to catch others' attention. Messages are thus a
matter of self-monitoring.
In the beginning we proposed to discuss three models of communica-
tion. Until now we have discussed two with opposing rationales. The
final model of communication, the antagonistic dialogue, involves ten-
sions between opposites.
Antagonistic dialogue
Between the extremes of two ideal type models of non-communica-
tion (collective monologue) and genuine communication we find the
third: antagonistic dialogue. An antagonist is a biological term used to
describe something with the opposite effect (of an agonist). The word
dialogue comes from the Greek word 'dialogos', 'dia' meaning
'through' and '10gos'32 meaning 'the word'. Antagonistic dialogue is
thus a dialogue between opposing opinions or between diverting per-
spectives. The opposition can take two forms, which are based on dis-
tinct logics: 'rock' logic or 'water' logic (De Bono, 1991 :8). A rock is
hard and unchanging, and if you add one rock to another there will be
two rocks. Water flows and takes the form of the container in which it
flows, and if you add water to water, there will not be two waters but
rather more water.
Schein gives a vivid example of a 'traditional academic' argumen-
tation from a seminar on clinical method (1987: 14): "We found our-
selves 'interrogating' the presenter, competing with each other for air-
time, arguing with each other about the validity of ideas presented,
and generally behaving in a manner that I labeled as 'aggressive',
though some members aggressively denied that there was any aggres-
sion present." We could call this antagonistic friction combined with
'rock' logic.
The potential of antagonistic friction in organizations is described
by Brown & Duguid (1991 :54): "Out of this friction of competing
ideas can come the sort of improvisational sparks necessary for ignit-
ing organizational innovation." The friction of antagonistic forces can
be softer, as seen in the following description by David Bohm
(1996:6): "The picture or image that this derivation suggests is of a
stream of meaning flowing among and through us and between us. This
will make possible a flow of meaning in the whole group, out of which
may emerge some new understanding. It's something new, which may
not have been in the starting point at all. It's something creative. And
this shared meaning is the 'glue' or 'cement' that holds people and
societies together." Bohm's description (and his choice of words) is a
good example of what we could call 'water' logic.
Antagonistic dialogue can be more or less constructive, spanning
from battle and destructive conflicts to cross-fertilizing confrontations
leading to innovative knowledge creation. In this respect antagonistic
dialogue displays a tension similar to 'edge of chaos' situations
(Stacey, 1996). Originally dialogue was a 'nonconfrontational' way of
solving problems in ancient Greece and only later did it turn into
debate and voting, as in our modern democracy. In the words ofVan
Maanen (1995:140) who contrasts 'debate' with 'conversation': "The
object of debate is of course to overwhelm or obliterate one's oppon-
ent: to prune, pare and discard. The object of conversation is to keep
it going: to plant, nurture and cultivate."
Several researchers have pointed out the latent creative potntial of
opposition. For example, Eisenhardt states: "Creative insight often
arises from the juxtaposition of contradictory or paradoxical evi-
dence" (1995:85). The advantage of the model of antagonistic dia-
logue is its potential for describing conflict in a prospective context
with a yet indeterminate outcome.
The Harmony-Conflict Span
Harmony is linked to stability. According to the Random House dic-
tionary, harmony is an "agreement; accord; a consistent, orderly, or
pleasing arrangement of parts" (Snyder & Clontz, 1997:63). In a group
context, harmony concerns group cohesiveness and consensus. Cohe-
siveness can be described as a 'we-stick-together' feeling. If all partici-
ipants agree on a decision or a goal, group cohesiveness grows. Con-
sensus can, however, be superficial, caused by fear of conflict, fear of
loosing face, fear of a superior (present), or by simple miscommunica-
tion, etc. An apparent consensus not based on (genuine) communica-
tion can be explosive.
Thus, the function of cohesion and consensus is that it provides
stability and 'glues' the group together. When a group has cohesion,
anxiety is reduced and the participants can relax, as cohesion, once
established, demands little energy. The danger is, of course, 'group-
think' which Janis observed empirically and described (Janis,
1989:276): "excessive risk-taking based on a shared illusion of invul-
nerability, stereotypes of the enemy, collective reliance on ideological
rationalizations that supported the belligerent escalation to which the
group became committed, and mind-guarding to exclude the dissident
views of experts who questioned the group's unwarranted assump-
tions". Thus the dangers of cohesiveness is a very strong internal focus
that resists opposing data or information from external sources.
If harmony is linked to stability, conflict is linked to instability. "Con-
flict exists, presumably, because of the simultaneous presence of at
least two mutually incompatible response tendencies" (Festinger,
1969:3). From a perspective of harmony, conflict is negative, as it is
disturbing and disruptive. If, however, we choose to regard conflict as
'unresolved difference', new perspectives open up towards resolving
these. Bennis, Parikh & Lessem, (1994: 153) describe three ways:
domination, compromise, and integration.
A. Domination is the warfare method, where the strongest or most
powerful will win, a win-lose situation. In relation to the discussion
of this chapter, domination is related to a defensive group climate,
to defensive communication and to the destructive trajectory of
antagonistic dialogue.
B. Compromise involves bargaining, but it is in many respects similar
to domination, as the strongest or most powerful will usually get
the best compromise. In a compromise, however, differences are
smoothed out and covered up, and therefore, in reality, a comprom-
ise solves nothing. Usually a new conflict will arise, e.g. regarding
the interpretation of the compromise. In our terms, compromise
could be a product of collective monologue, as this is very superfi-
cial, or to use Covey's framework, a product of respectful commu-
nication. Compromise has no winner, it is a lose-lose situation.
Certainly, there are times and situations where this is preferable to
a disruptive conflict - like war, for instance.
C. Finally, integration means finding a third and new alternative that
can satisfy the highest prioritized wishes of both parties. This solu-
tion, however, demands openness and transparency on the part of
the participants, 'no cards up the sleeve', as this approach does
need the efforts of both parties in trying to find a solution to their
(now) joint problem. If we try to draw some parallels to the frame-
work of communication, integration seems close to the concept of
genuine communication of sharing in an open way.
193
But something is missing in the above methods for resolving conflicts.
The concept of integration implies inclusion of what is known, but
there is no indication of novelty, as implied in the constructive trajec-
tory of antagonistic dialogue. Conflicts can be caused by both concep-
tual and relational matters, and are often a strong mix of both. Ac-
cording to Schneider, Garcia-Prieto & Bellard (1998), Jehn found that
people in work teams perceived relationship conflicts as an impedi-
ment for group effectiveness, whereas task related conflicts, in fact,
improved performance. Thus, in innovative groups, where the cause of
conflict is often related to the topic, at least at the surface level, con-
flict or antagonistic dialogue could be enhancing. Putnam (1986: 190)
has proposed, based on literature, that there is a high frequency of
controversy in the middle stage of group decision making 'when prob-
lems are defined and solutions considered'.
Thus, the essence of 'antagonistic dialogue' is much like the Chinese
character for 'crisis', which is composed of two characters: one for
danger and one for possibility. The challenge in the following chapters
is to examine processes that initiate, support or trigger a destructive
versus a constructive outcome - innovative crystallization. With this
we have concluded the construction of an analytical framework for
communication, displayed in the model below:
Concluding remarks
Communication is an essential part of knowledge creation - both
regarding what knowledge is shared and how this knowledge is shared.
It matters profoundly whether people are more focused on promoting
themselves (collective monologue) or they listen attentively and try to
build on other people's ideas (genuine communication).
The purpose of this section has been to construct a theoretical and
analytical framework for communication in groups. The 'genuine
communication' model is based on two core features: building 'com-
mon ground'33 and forming a relational bond between people. In
contrast the 'collective monologue' is not real communication (as we
have defined it). People talk to each other - but actually, when you
analyze it, they talk to themselves. The main difference is the attitude
of the participants of being either active (listening) or passive (hear-
ing). Genuine communication requires listening. We also saw in this
section that the type of conversation that will take place in a group to
a large extent depends on the context and in particular on the group
climate.
The third type of communication introduced is the 'antagonistic
dialogue', built on the communication of opposing views. It will be
interesting to investigate this opposition to find out which processes
lead to constructive results and which lead to destructive ones.
collective
monologue
194
BASIC MODELS OF COMMUNICATION
PROCESSES IN HETEROGENOUS GROUPS
Darsa, 1997c: Communication framework
genuine
communication
The communication framework consists of these three ideal-type
models. By using ideal-type models of communication we want to be
able to identify communication patterns in natural groups of know-
ledge workers working on innovation. In chapter 6 we will operation-
alize the communication framework in order to establish construct
validity and then, through the analysis that will follow, we will identify
the communication patterns that turn out to enable innovative crys-
tallization.
195
The preject-project model
The preject
ject is characterized by a structure that differs from the structure of a
goal. This structure is leadership -leadership understood as functions,
roles, processes of communication and interaction.
A project is per definition (Herlau & Tetzschner, 1995:87) a goal
directed, time limited, unique effort, which uses the available human,
material and economical resources from across an organization. The
word project derives from the past participle of the Latin verb
'projicere' (to throw), and refers to an activity that has been 'thrown',
i.e. it concerns planning to obtain a goal. In an attempt to distinguish
between goal-driven and goal-searching processes, and in order to com-
municate this distinction, the term 'preject' was coined
34
(Dars0 &
Herlau, 1996). 'Preject' refers to the process 'before' a goal has been
set. A preject thus precedes a project. The former describes the pro-
cess before ideas have been generated, decisions made and goals
formed, whereas the latter indicates activities carried out to achieve an
existing goal.
D:1fSO & Herlau, 1996
PREJECT-PROJECT DIAMOND
Phase 4 Phase 3 Phase 2 Phase 1
Herlau & Tetzschner's model is a phase model, outlining four phases
(1995:223):
1. An intensive course of training tools and developing a common cul-
ture.
2. The inventive phase consisting of project work with stochastic lead-
ership.
3. Linking the project to 'reality' through goal-directed project man-
agement.
4. Implementation.
Thus, the preject is a phase of problem identification, problem fram-
ing and reframing, whereas the project phase concerns problem solu-
tion. The preject-project process model is primarily a cognitive model,
which can be applied in different ways. It was inspired by and derived
from a larger model, the didactic concept by Herlau & Tetzschner,
used for training interdisciplinary student groups (LFPP5).
goal
project phase

time
--
process.
execution
-
preject phase
start
As seen in the Preject-Project
diamond, the preject is charac-
terized as nonlinear, divergent
and process driven, whereas the
project is characterized as linear,
convergent and goal driven
(Herlau, 1995). The significant
difference is the lack of goal in
the preject phase, which means
that the normal structure ofpro-
jects is missing, and consequent-
ly gives it an apparent turbu-
lence. In projects the goal dic-
tates planning and tasks, which
makes the process more predict-
able and linear.
In order to distinguish between new knowledge emerging from net-
working (Kreiner & Schultz, 1993) and new knowledge being created
in project groups, it is necessary to add one more ingredient: the pre-
Herlau & Tetzschner, 1995:223: The Didactic concept
(drawn (and translated) after the original by the author)
197
The didactic model is understood as a sequence, starting out, as illus-
trated, at the left side of the model by the convergent funnel. This sym-
bolizes the development and training of a shared framework among
the participants before starting on the inventive (preject) phase 2. In
the inventive phase the group applies the Kubus tools in order to solve
or identify a problem. Here they USe internal as well as external
resources for building a project that can be used by the target group.
In phase 3, the project is linked to the target group and in phase 4, the
project is implemented, and feedback is fed into the funnel.
The above sequential phase model was part of my cognitive frame-
work
36
The preject-project diamond was developed later in a-different
. 37 ki d
project as a wor ng an instruction model. In this way, the preject-
project model became the point of departure and has, together with
the working model from chapter 1 formed my cognitive map through
all the case studies.
The main advantage of the preject-project model is that it is holistic
and simple. Nonaka &Takeuchi's knowledge spiral model could also
have been applied, but its vagueness regarding individual, group, and
organization makes it difficult to apply as the main framework. Thus,
it will serve better as comparison. Another advantage of the preject-
project model is that it is familiar and well tested. It provides a frame-
work for discussing processes that are difficult to understand and
articulate, and it enables a conversation about divergent versus con-
vergent processes, and likewise, it facilitates a conversation about
vague and fluffy goals that emerge when enough knowledge has been
created in a group. To use the words of Polanyi (1966:7): "We can,
accordingly, interpret the use of tools, of probes, and of pointers as
further instances of the art of knowing, and may add to our list the
denotative use of language, as a kind of verbal pointing." Here, we
extend Polanyi's idea to call the model a processual 'pointer'. We will
continue with this model as a processual pointer throughout the book.
We will struggle with it in chapter 5 in relation to illustrating the case
studies, and we will transform it in the synthesis of chapter 8. In chap-
ter 9 it appears as the main structure for understanding and discuss-
ing innovative processes. Thus, the conjectures of the preject-project
model survive, but a new perception transforms its meaning.
Concluding summary
The purpose of this summary is to conclude what the major contribu-
tions have been from the different fields we have reviewed here. In-
deed, they have all contributed to form a multifaceted and compre-
hensive synthesis - but in different ways.
The innovation process models were organizational models that
provide the context and the setting for the case studies. The frame-
works serve a discussion of how new product development is organ-
ized, which we did for both the pharmaceutical and the industrial
enzyme units of Novo N ordisk. This way we discussed different types
of systems in which innovative groups have to form and frame their
work. Apart from providing the necessary organizational framework,
three process models were identified as relevant: the set-based Toyota
model of a prolonged divergent phase with postponed decision mak-
ing, Sundbo's dual organizational structure (1998), and the MIRP
meta-process model with its holistic and processual perspective.
The group process models contributed with background materi-
al, clarification of concepts, new frameworks and research on group
conflict and other process models. Reviewing research on group pro-
cess models was furthermore necessary because we apply a group per-
spective in this book. We find the group perspective a major challenge
in global knowledge society, as now and in the near future organiza-
tions and society at large are becoming, and will become increasingly
dependent on the quality of group communication and knowledge
creation. As Gibbons et al. have pointed out (1994: 120) "Created
comparative advantage results not only from resources but from the
creative combination of resources and resourcefulness." And these are
increasingly human. It is therefore disturbing that only little research
has been and is done with real-time real-life groups working with
innovation in the early phases.
From group development we sought the necessary clarifications of
concepts and frameworks, and developed a framework for perceived
diversity and similarity, which will be used throughout the book. As
prejects do not have specific goals, a different structure is applied:
leadership, and here we used Herlau's framework for discussing lead-
ership in groups. Finally, we found relevant theory on conflict in
199
groups, which will be used later during the analysis. Interesting pro-
cess models were Gersick's punctuated equilibrium model and Poole's
multiple sequence model.
The field of knowledge creation contributed to the overall frame-
work considerably, both in relation to methodology (Mode 2
approach), and in relation to classifications, frameworks and process
models. This section entailed important discussions on different types
of knowledge and knowing and on forming, framing and solving
'problems'. In this field some highly relevant process models were
identified, in particular Nonaka & Takeuchi's knowledge spiral (1995)
and Dars0's Knowledge Map (1997c).
Creativity was important for discussing delimitations and bound-
aries, in particular between creative and innovative processes. Through
the discussions related to creativity we also managed to differentiate
thinking processes and to substantiate earlier discussions. In particu-
lar, the concept of innovative crystallization was important to clarify.
Process models of creativity contributed to the overall knowledge but
were not directly applicable.
Communication is central to the present study, and from the theo-
retical contributions we developed a communication framework of
three ideal type models: genuine communication, collective mono-
logue, and antagonistic dialogue. This framework will be operational-
ized later for construct validity and will form the foundation for pat-
tern matching with the data from the case studies.
Finally, we outlined the major characteristics of the preject-pro-
ject model and provided the background for the development of the
model. We concluded by emphasizing that the model is primarily a
processual 'pointer' and will form the basic framework for under-
standing and discussing innovative processes throughout the book.
In chapter 3 we have focused on 'episteme'. In the next chapter we
shall focus on 'techne' and examine how we can generate knowledge
in practice for answering the research questions.
200
Notes
"Philosophies of the Branches of Knowledge: Philosophy of science: PHILO-
SOPHICAL STATUS OF SCIENTIFIC THEORY: Philosophical analysis
and scientific practice." Britannica Online (Nov. 98)
2 In the article Saren (1984) lists only 5 categories. Still, Trott writes
(1998:126): "Among the burgeoning management literature on the subject it
is possible to classify the numerous models into seven distinct categories
(Saren, 1984):" Thus it seems that Trott has added two 'extras' on his own
account: cross-functional models and network models.
3 This was also found by Christensen & Kreiner, 1994
4 According to Buchholz & Roth (1987:57) the difference between a purpose
and a goal is that "a purpose is an ongoing, general direction, whereas a goal
has a beginning, middle, and end. It is a specific target. A goal that fits under
the purpose is 'on purpose'. A purpose is like an umbrella under which fit the
specific goals the team chooses."
5 The following is based on a paper presented at the EGOS conference of 1998:
Lotte Dars0: "The Butterfly Effect. The Difference that makes a Difference
for the Emergence of Innovation in Researcher Teams"
6 Myers-Briggs Type Indicator, a 'test' developed to understand and value the
differences of individual preferences. It was based on the work of lung, and
developed by Isabel Myers and Katherine Briggs.
7 S = Sensation, N= Intuition (the I is used in the E-I code: Extrovert versus
Introvert)
8 See Copenhagen Institute for Futures Studies, 1999, 3: 'Medarbejderdiver-
sitet som konkurrenceparameter' (Employee diversity as a parameter of com-
petition, my translation)
9 The originator Henry Gantt was a 'disciple' of Taylor and developed the
Gantt-chart in 1910 (Packendorff, 1995:321)
10 See also under Knowledge Creation models
11 Poole, 1981, 1983a, 1983b, Poole & Roth, 1989a, 1989b
12 A breakpoint is a key situation, where the group can either change direction
or can choose to continue in the same direction (Poole, 1983b: 330)
13 Mintzberg et a!., 1976:265
14 In my paper of 1997c, individual knowledge acquisition was described as per-
ception, cultivation, storing, and recuperation. Storing has now been replaced
with the more dynamic concept of integration.
15 According to Zohar (1997:36) 80% of a given pattern can be removed, and
the brain will still be able to recognize it.
16 In 2000 a book about this subject came out: 'Ubevidst Intelligens. Du ved
mere end du tror' ('Non-conscious intelligence. You know more than you
think', my translation) by Ole Vedfelt, Gyldendal
201
17 The axes have been reversed in relation to the figure of Dars0, 1997c: 11, but
the meaning is the same.
18 I recognize that I am probably digressing from N0rretranders' definition here
19 The question is whether 'tacit knowledge' is tacit per definition or whether it
can actually surface
20 And then in the end he bought the solution from Swan (Drucker, 1985)
21 e.g. discussed in Scheidel, 1986
22 The Duhem-Quine principle asserts that one single factor is not enough to
explain the closure of a controversy or the certainty acquired by scientists.
23 The shadow system "comprises all social and political interactions that are
outside the rules strictly prescribed by the legitimate system. It is the arena in
which members of an organization pursue their own gain but also the arena
in which they play, create, and prepare innovations." (Stacey, 1996:290)
24 Overhead quote from a lecture by Dr. Chibber, at the Institute of Manage-
ment, Politics and Philosophy, Copenhagen Business School, Sept. 1996
25 This has been translated by the author from Danish: "Vi kan ikke forlade
sproget, for sa kan vi ikke tale sammen. Men vi kan heller ikke sige det vi
gerne viI, for vi har kun sproget at meddele os igennem."
26 in monochronic cultures people prefer to be engaged in one thing at a time,
in polychronic cultures people engage in several activities and events at the
same time
27 Berlingske Univers: Det bevidste 0re (the conscious ear), 180599
28 The Scottish percussionist Evelyn Glennie, ibid.
29 1. Knowing one's emotions, 2. Managing emotions. 3. Motivating oneself. 4.
Recognizing emotions in others. 5. Handling relationships
30 IQ= Intelligence Quotient, the traditional way of measuring intelligence
31 Lecture at Miniconference at The Institute for Management, Politics &
Philosophy, CBS, 1998
32 Zohar makes the point that there is a more original Greek meaning of 'logos'
meaning 'relationship'. Thus 'dialogos' means 'through relationship' (Zohar,
1997:136)
33 Interestingly, the first feature was developed before Nonaka & Konno (1998)
described the Japanese concept of 'Ba', but the content and meaning is per-
fectly similar.
34 Schein notes that a new concept is valuable when it explains events that are
otherwise difficult to grasp (Schein, 1986: 12)
35 Lereanstalternes F::elles Projektledelse og Innovationskursus (Interdisciplin-
ary project management and innovation course)
36 I have instructed students, teachers and business people in the Kubus concept
for several years, see CV in Appendix
37 The Kubus pilot study under the Ministry of Education (Dars0 & Herlau,
1996)
202
Chapter 4
Methodology
"The proper place for the scientist - once in a while at least - is
in the midst of the unknown, the chaotic, the dimly seen, the
unmanageable, the not-yet-well-phrased. This is where a prob-
lem oriented science would have him be as often as necessary"
(Maslow, 1970: 17)
The research problem concerns how heterogeneous groups can foster
innovative crystallization in order to produce wealth - a complex
problem - as we will see in the following. The methodology concerns
a Mode 2 approach to knowledge production as the research is
focused on usefulness and application in practice. According to Flyv-
bjerg (1992: 45), methodology concerns concrete practical rationality
or 'techne', and that is the main perspective of this chapter.
In accordance with the Mode 2 approach of Gibbons et al.
(1994:44), we will apply "a pluralism of approaches which combine
data, methods and techniques to meet the requirements of specific
contexts." Thus the chosen method is a series of real-time case-stud-
ies, combining mainly two approaches: clinical and ethnographic
research.
What makes this study stand out is, however, that it is, indeed,
'Innovation in the Making'. Unique characteristics of the study are:
its 'real-timeness', its 'thrownness'l into innovative processes, the
researcher's interaction with the 'researchees'2, and its prospective
approach of looking forward instead of looking in retrospect. Retro-
spective studies are state-of-the-art in studies on innovation and
knowledge creation. We try to avoid "the' ex post facto fact' fallacy: the
fallacious retrospective claim that, for present events to be as they
are, their causes must have been of a certain kind." (Shotter,
1993:25)
20
3
Outline of the chapter
As the objective is to gain knowledge about Innovation in the Making,
we need to ask: what is the nature of the problem? and secondly:
through which methodology can this kind of problem be studied? We
will examine the nature ofthe problem by outlining processes and elem-
ents that may influence the research problem area. The complexity
of the problem justifies the case study method, and we introduce the
major features of the case study method as well as the research design
and the propositions. The most important part of this chapter is, how-
ever, the deliberation on prospective versus retrospective research.
Here we draw on the concept of sense making by Karl Weick (1995),
but also include arguments from John Shatter (1993) and Bruno La-
tour (1987). The discussion focuses mainly on the difference between
situations in their making that are open, and situations when they are
'made', i.e. closed.
Also the researcher's role of clinician and ethnographer is discussed
and some differences and tensions between the two roles are pointed
out. We conclude by contrasting the criteria for evaluating research of
the positivistic (functional, modernist) and the naturalistic research
paradigms: internal validity versus credibility, and reliability versus
dependability. As the study applies multiple perspectives, we argue for
applying both.
The nature of the problem
In order to demonstrate the complexity of the problem we are examin-
ing in this study, we will try to highlight some of the aspects and
dynamics that could influence the process and the outcome from a
group of knowledge workers dealing with knowledge creation and
problem solving. Innovative processes are multifaceted, intangible,
dynamic and complex. We want to study the processes that lead to and
enhance crystallization of new concepts or new knowledge in hetero-
geneous groups. An attempt to illustrate the multifaceted problem is
displayed below. The model has certain limitations. Its point of depart-
24
ure is psychological, based on (different) levels of interaction, starting
with the single individual, then group, organization and society. The
model is meant to be holistic in showing the different life-worlds as
'layers' of the same world, indicated in the model by the dotted boun-
daries. The model may, however, give the impression that individuals
are not in direct interaction with the organization or with the environ-
ment. This interpretation is, of course, not intended. Another short-
coming of the model is that it is not all-inclusive.
globality NN image
global trends - .flI'"aanizrrtiunuHevel-- - - - __ _ stakebolders
//// Industri Strategy Group ----_ customers
,
" '"
uncertainty ,,/' timing ..... -- --..... networkS-, .
// ...... /.......... using diversity ..................... rhetoric ',competitors
I/opportunities/.......... ideas _--- _ info by networks """''-, . "
I
I ideas / tti -irnlivldual revet-_ ' anchonng "
/ II oppor um ....... """ ' ...... emergent strategy'\ \
cbange I I / competition ' , t t ,
I visions I problem / perceptiono " rbetoric ,s ra egy ,
I ,I forming I . t 'ti' 0 0 personal gain' 'I . . I
I I In ill on \ . . . deCISIon I
I needs I I 0 I deCISIOn making I . I
I I org. culture \ ignorance liVid"1 knowledge I I making I
\ problems \ , I
\ ' harmony _, inner drive. tasks / goals I evaluatiOl"\'
\ \" / /
, org. culture \ conflict "commitment time pressure." methods / /
\ ',climate .............. -..__ values __-// tasks III ppe /
\ " ------ ,/ /
' ........... communication . time/ressource ...-// resources II
" .............. _..... leadershLp .... /...... II
, . k -- I -// /
. ..,tnternal networ s - - v.! - - - reward systems / I d
ecologIcal '" matrix /// aws
concerns .......................... management .......... ,....,.... regulations
............... --_..........
___
Research problem area
The inner circle: the individuals
Knowledge workers are intelligent in diverse ways, and some are high-
ly creative. Different education brings about different subcultures, e.g.
engineering cultures are prone to problem-solving and technical de-
signs (Schein, 1998), whereas academic cultures are known for strong
analytical skills. Still there is a large spectrum of personalities, as most
scientists have their highly individual style. Diversity is abundant re-
25
garding individual commitment, individual perception of opportu-
nities, personal goals, immediate personal situation (both family-wise
and job-wise), etc. The influence of specific individuals on 'innovation
in the making' can vary from little to very high. We could have taken a
perspective focusing on the individual personality or we could have
examined the composition of a good team. Certainly each group is
unique as it consists of unique individuals. Therefore individuals do
matter. In the present study on innovative processes, however, we will
look at the group as a whole, as having a group identity ('syntality'),
while still acknowledging the fact that a group consists of individual
agents, and that each individual could make a difference for the final
result.
The middle: the group
At the group level many aspects may influence the final result. Re-
garding input, apart from the professional knowledge and skills each
individual brings to the group, the purpose, goal or task of the preject
group, the available ideas or opportunities, the knowledge brought in
through literature and networking as well as the resource allocation _
all these inputs will to some degree influence the result. As pointed out
earlier, the forming and framing of the problem, the type of commu-
nication, the climate, the harmony-conflict potential, the leadership,
the decision making, the type of goals and the strategy of the group will
also influence the result. In addition, the methods and approaches the
group applies to accomplishing the result, as well as the organization-
al culture, the diversity of professional cultures, national culture and
global culture, influence the result and thus add to the complexity of
the interaction pattern.
The outer circle: the organizational context
But this is not all. As seen from the illustration, the group (and the
individuals) cannot be separated from the context. The working group
is influenced by the overall strategy and by the visions and decisions
206
of top management. Different styles and roles of leadership, as well as
the direct influence from line management (e.g. in relation to resource
allocation) can encourage or discourage groups working with innova-
tion. Politics and rhetoric play a part in most organizations through
the internal networks. The outcome from the group must fit into the
larger context, as for instance into the project portfolio, the urgent
needs and the contemporary priorities. The perception of the evalu-
tion board regarding quality and viability, combined with the group's
rhetoric in presenting the proposal also play a part. Finally, timing is
of major importance, and luck is a component that should not be
underestimated.
Outside the circle: the environment
Groups working with innovation depend on the relation to the cus-
tomers and the development in the local and global market. The activ-
ities of competitors must be taken into consideration (as for instance,
whether the competitors plan to launch a similar product and thereby
change the 'fitness landscape') as well as the relation between the
organization and its stakeholders. Finally, in this study the concurrent
trends of the pharmaceutical industry, biotechnology, and industrial
enzymes are important. Globalization and change are challenges of
major influence along with the development in information technolgy
and communication.
All the above may influence directly or indirectly, to a greater or less-
er extent, 'Innovation in the Making'. Indeed, the problem is complex.
It involves many sub-units, relations, dimensions, dynamics, and pro-
cesses that are ongoing, interrelated and that mutually shape each
other. In the words of Bales & Strodtbeck (1951), it is a full-fledged
problem - and with plenty of ambiguity, uncertainty and flux.
How can we best approach and investigate the problem of 'Innovation in the
Making'?
20
7
The case study method
The criteria for selecting the case study approach are that the research
problem involves complexity, real-time events and processes that are
not easily distinguishable from their context and that must conse-
quently be studied in real-life (Yin, 1994). The case study method
therefore suits the purpose. The case study has a strong design and at
the same time can include qualitative as well as quantitative data. Case
studies cover contextual conditions and can be applied to explain
(causal links), describe, illustrate, and explore these. The present study
explores, describes, and illustrates innovative processes in their mak-
ing and also attempts to explain or draw inferences in order to answer
the research questions.
Yin lists five components that are important for the case study design
(1994:20):
1. The questions
2. The propositions
3. The unit of analysis
4. The logic linking data to propositions
5. Criteria for interpreting the findings
Research questions and objectives
According to King, Keohane &Verba (1994:18) a good research ques-
tion satisfies two criteria: "it should deal with a significant real-world
topic and be designed to contribute, directly or indirectly, to a specif-
ic scholarly literature."
The research questions:
In what ways can innovative processes be initiated, sup-
ported and managed towards innovative crystallization in
heterogeneous groups?
creation in heterogeneous groups of knowledge workers?
208
The necessity for research on heterogeneous groups working with the
genesis of innovation and new knowledge has been pointed out by
researchers of various fields. For example, Gibbons et al. emphasize
the comparative advantage of team innovation and thus a need for
research in this area (1994:120):
"The ability to innovate continuously is crucial to long-term
performance. It is the source of creative comparative advan-
tage which drives forward changes in terms of products on the
first level. On the first level competition is concerned with tech-
nology, on the second level with knowledge and skills.
Created comparative advantage results not only from resources
but from the creative combination of resources and resourceful-
ness. The novel element is that the relevant resources are
increasingly human ones and widely distributed. The trend
towards alliances is a natural outcome of the need to access
these human resources. Resourcefulness consists in the ability
to configure these resources and the source of the value-
added lies in the precise form of the collaborative groups
and in the skills of their members."
(emphasis added)
In order to answer the research questions, the present study proposes
three objectives. In relation to King, Keohane & Verba's criteria
(above), the first objective concerns a 'real-world topic', the second
concerns the contribution to 'scholarly literature', and the third adds
the important aspect of practical usefulness.
Objectives:
1. To gain (new) knowledge about the processes that lead to
innovative crystallization through case studies in a real-
time and real-life organizational context in interaction with
heterogeneous groups working to create new knowledge
and new leads in the preject phase.
2. To develop a 'special theory'3 derived from practice for
initiating, supporting and managing innovative processes
20
9
to enhance innovative crystallization in heterogeneous pro-
ject groups.
3. To develop a set of recommendations for the organization: a
manual with methods, frameworks and advice for groups
working with knowledge creation and novel approaches.
The propositions
Propositions point out the direction of the case study as to how the
research question may be examined and what should be investigated.
Of importance when discussing propositions, is the question of falsifi-
cation. King, Keohane & Verba (1994: 19) recommend that the
researcher select theories or propositions that can be falsified by using
Karl Popper's principle of falsification (1968). He argues that a
researcher must primarily be able to answer one question: "What evi-
dence would convince us that we are wrong?"
Because of the complexity of the research problem it seems adequate
to apply Popper's approach. Considering the many possible influences,
how could we know that the processes we focus on are not influenced
by other elements than the ones we examine. Thus, we shall adopt
Popper's criteria of falsification in order to see if we can reduce the
complexity by eliminating some elements. We will start with the prop-
ositions and then examine the question of falsification.
1. When diversity is matched with genuine communication, innov-
ative crystallization is enhanced
2. Innovative crystallization is triggered by certainty and know-
ledge, rather than by uncertainty and ignorance
Re. 1. Earlier we defined the concept of diversity and constructed a
matrix of perceived diversity and similarity. We have also defined and
discussed the concept of genuine communication, but we have yet to
make the communication framework operational. This is done in
chapter 6. Regarding innovative crystallization, we managed to devel-
op criteria for recognizing the phenomenon. Then how could the first
proposition be falsified? This would be possible if communication pat-
terns, other than 'genuine communication' (e.g. collective mono-
210
logues), were seen to enhance innovative crystallization in a hetero-
geneous group. We would not regard it as falsified if we found a pat-
tern of genuine communication in a group, which did not coincide
with crystallization. We would, however, look further into the patterns
of communication if genuine communication coincided with innova-
tive crystallization.
Re. 2. The second proposition relates to the content of communica-
tion, to the conceptual level. The Kubus model (Herlau, 1995) will be
used as an analytical tool to classify the topical conversation, as con-
versations can be coded by applying the Kubus framework into three
arenas of uncertainty and three arenas of certainty. This way the con-
tent of the conversation can be coded and quantified (see chapter 7).
Thus, if crystallization emerged when more time was spent in areas of
uncertainty and ignorance than in arenas of certainty and knowledge,
the proposition would have been falsified.
The unit of analysis
The unit of analysis is the innovative process, i.e. a process that leads
to 'innovative crystallization'. In chapter 3 we made a tentative defin-
ition of innovative processes as the formative processes of problem
forming, problem framing and crystallization of (new) knowledge. We
want to gain a better understanding of innovative processes in order
to understand how innovative processes are initiated, supported and
managed in heterogeneous groups of knowledge workers, and we
decided to examine patterns of communication on two levels of analy-
sis, the relational and the conceptual level.
By studying communication patterns, could we identify elements,
processes, or patterns that encourage or discourage innovative activ-
ities?
211
The logic linking data to propositions
The empirical data will be linked to the propositions through 'analyt-
ical generalization'. Analytical generalization is based on the develop-
ment of theory prior to data generation. "This role of theory devel-
opment, prior to the conduct of any data collection is one point of
difference between case studies and related methods such as ethnog-
raphy" (Yin, 1994:27). The communication framework of ideal-type
models will primarily serve as the basis for comparison with the data
obtained on the two levels mentioned above. But also the relevant
theories and process models discussed in chapter 3 will support ana-
lytical generalization.
Criteria for interpreting the findings
As the data will derive from different sources, methodological triangu-
lation will ensure construct validity. The findings will be interpreted
through pattern-matching in relation to the frameworks developed in
chapter 3. The intention is to find a match with one of the outlined
(rival) process models, or regarding communication, with one of the
ideal-type communication models. Alternatively new theoretical mod-
els will be developed to match the findings. "The method of general-
ization is 'analytical generalization', in which a previously developed
theory is used as a template with which to compare the empirical
results of the case study." (Yin, 1994:31) Thus a synthesis is reached
through combining the findings, possibly in a causal or some inferen-
tial relationship, in relation to understanding innovative processes and
answering the research questions. Finally recommendations will be
made based on the findings.
Triangulation
"Consider the difficulty of establishing the occurrence of an
event. You would be more confident in saying that the event
actually had occurred if your study showed that information
212
from interviews, documents, and your own observations all
pointed in the same direction. With such converging evidence,
you might even feel very confident about your conclusion that
the event had occurred. This type of triangulation is the most
desired pattern for dealing with case study data, and you should
always seek to attain such an outcome." (Yin, 1993:69)
The concept of triangulation originates from the field of cartography
and navigation, where it is based on the rationale that it takes meas-
urement from three different points to locate a new point or to take
bearing. The parallel in the social sciences is primarily a combination
of methodologies in the study of the same phenomenon.
Triangulation means using different or multiple sources of data,
methods, investigators, or theory. The present study, aiming at a com-
prehensive or multi-perspective approach, applies the triangulated
approach in three ways. Firstly, it serves as a method for generating
data (primary and secondary data) from a variety of sources to con-
struct valid knowledge about innovative processes. Secondly, the
methods are triangulated by applying the case study method, clinical
research and ethnography. Finally, analytical generalization is matched
with Popperian falsification. Evidently, triangulation most adequately
reflects the complexity of the problem and will help to point out con-
verging evidence and thus to answer the research questions.
Prospective versus Retrospective studies
"Most studies of innovation or change to date have been retro-
spective case histories conducted after the outcomes of change
were known. However, it is widely recognized that prior know-
ledge of the success or failure of an innovation invariably biases
a study's findings. Historical analysis is necessary for examining
many questions and concerted efforts can be undertaken to
minimize bias, but it is generally better, if possible, to initiate
historical study before the outcomes of a strategic change pro-
cess become known" (Van de Ven and Poole, 1989:35).
21
3
What makes the case studies of 'Innovation in the Making' stand out
methodologically is the real-time, processual, prospective approach. As
stated above, most studies on innovation have been retrospective, i.e.
interviews based on success stories. That includes one of the largest
studies on knowledge creation by Nonaka & Takeuchi, which is based
on 130 interviews (1995:17).
Two main points will be made here. The first concerns 'hindsight bias'
(Weick, 1995). The second examines the openness versus the'closed-
ness of situations. Together these arguments justify a real-time pros-
pective study of innovative processes.
Regarding hindsight bias, Weick (1995:30) reports experiments with
student groups who were, after a task, randomly given feedback on their
performance (high or low) and then given questionnaires regarding
motivation, communication, group cohesiveness, openness to change,
etc. The ratings of 'high-performance' groups were significantly more
positive than the 'low-performers'. This is one of many examples from
empirical research of hindsight bias. Weick summarizes this (1995:28):
"The nature of these determinant histories is reconstructed differently,
depending on whether the outcomes are seen as good or bad."
Many other examples are reported in literature of sense-making
processes (Weick, 1995:18): "people concerned with identity in the
context of others engage ongoing events from which they extract cues and
make plausible sense retrospectively, all the while enacting more or less
order into those ongoing events" (Italics added).
The second argument, regarding openness versus dosedness, will take
its point of departure in John Shotter's explanation of the 'ex post facto
fact' fallacy. The following sequence is quoted at full length (Shatter,
1993:85):
1. "Firstly, a situation is described which, although we do not
realize it at the time, is open to a number of possible inter-
pretations.
2. We are, however, then tempted to accept one of these
descriptive statements as true.
21
4
3. The statement then 'affords' or 'permits' the making of fur-
ther statements, now of a better articulated nature, till a
systematic account has been formulated.
4. The initial interpretation (already accepted as true, of cour-
se) now comes to be perceived, retrospectively, as owing its
now quite definite character to its place within the now
well-specified framework produced by the later statements.
In other words, the original situation has now been 'given' or
'lent' a determinate character, within the terms of the system,
which it did not, in its original openness, actually possess. This,
I think, is a fallacy which operates on a grand scale in the social
sciences, where we always attempt to make sense of social and
psychological phenomena within well-defined systems of terms
- that is, systematic discourses."
The above account adds valuable insight to the concept of 'process', in
how open situations become closed events, and how sense is made after-
wards through 'narrative smoothing'4. Shotter continues (1993: 128):
"Our attention is diverted, because, in a hermeneutical construction, all
the fragments which have occurred are decontextualized, and made into
a orderly or systematic whole - often with, as Freud put it, the insertion of
the 'missing points' which must have 'originally' been there if things are
to be orderly."
The real-time case study of 'Innovation in the Making' is an attempt to
study processes in their openness, in their making. Processes are gener-
ally fragmented and incoherent, and may not make sense at the moment.
Afterwards, however, when events have been 'constructed' or 'labeled',
they make sense, but this is a retrospective construction (Weick, 1995).
If sense-making or retrospective 'construction' of events and facts
guides experience and recollection, then there is a significant differ-
ence between things in their making and things made. The point here
is not to judge or evaluate which is better or more 'scientific'. Weick
distinguishes between 'pure duration' and 'discrete segments' of time.
According to Brown & Duguid (1991:41), Bourdieu distinguishes
between a 'modus operandi' (work in progress, i.e. open) and an 'opus
21
5
operatum' (the finished task). And Latour, who did an ethnographic
study on scientists in their labs, emphasize a similar difference in the
construction of scientific facts:
"Uncertainty, people at work, decisions, competition, contro-
versies are what one gets when making a flashback from certain,
cold, unproblematic black boxes of their recent past. If you take
two pictures, one of the black boxes and the other of open con-
troversies, they are utterly different. They are as different "as the
two sides, one lively, the other severe, of a two-faced Janus. 'Sci-
ence in the making' on the right side, 'all made science' or 'ready
made science' on the other;" (Latour, 1987:4)
The sum of the argument is the following. In order to produce new
knowledge about innovative processes, we need to study processes in
their making, i.e. in their openness. Thus the researcher needs to be
present, to experience (experiential knowledge) the potential of the
situations. I, the researcher, participate and can follow how knowl-
edge is created. Shotter (1983: 156) claims that action creates condi-
tions for further action, i.e. action sets a direction and a path. Only
in the midst of the process in its making is it possible to see how
innovative crystallization comes about and perhaps what makes it
emerge.
According to Shotter, that which is well-organized and rational actu-
ally prevents us from seeing that our circumstances surround us with
possibilities for constructing a new and different future. We cannot
help constructing a 'smooth' narrative afterwards, but we can comple-
ment the retrospective with the projective narrative through experi-
encing and participating in the process.
Clinical research and ethnography
The approach of the present study is clinical and ethnographic. Two
major characteristics of the clinical role are the qualitative aspect of
216
the relation to the client, and the 'helping' aspect of trying to improve
relations or solve practical problems. The clinical process model is pd-
marily built on the relations between the researcher and the client. For
the clinician, it is always the needs of the client that come first - and
this to the extent that research is secondary and must be abandoned
if it suits the client better. Regarding applicability, it is the function or
the practical aspect of the methods that are central. Thus methodol-
ogy depends on the nature of the problem.
In ethnography the aim is to understand and describe a group of
people and try to reveal some underlying structure - but without dis-
turbing or changing the organization (Schein, 1987:30). The primary
goal is thus to obtain scientific knowledge by understanding a group
or a culture from within and by uncovering deep structures. This has
been elegantly done by David Guss who lived for many years with the
Yekuana tribe in the jungle of Venezuela. He found the underlying
structure to be 'To Weave and Sing'S, which penetrated the Yekuana
world from constructing houses to communication about all daily
activities (Guss, 1989).
In his book on The Clinical Perspective in Fieldwork from 1987,
Edgar Schein contrasts the clinical with the ethnographic view. Schein
concludes:
" Both perspectives are ultimately necessary to understand fully
what goes on in groups, communities, and organizations, but
only by understanding the differences in the perspectives can we
grasp fully what is meant by saying that we need both perspec-
tives." (Schein, 1987: 14)
According to Schein, some differences between the roles of the clin-
ician and the ethnographer should be noted. The clinician must be
oriented towards the needs of the client, as the clinician has been
asked to come into the organization and is paid for it. The ethnogra-
pher, on the other hand, gets a more free-riding situation, as s/he has
chosen the organization and has created his/her own entry and condi-
tions for being there. The clinician is there to do the job of helping the
client to solve a problem. Therefore the clinician cannot leave until the
job is finished (or s/he is kicked out).
21
7
The ethnographer is there to learn and is free to leave when s/he
has learnt enough. The clinician has a limited amount of time, while
in principle the ethnographer has the amount of time it takes.
Another difference is that the clinician must concentrate on the
problem at hand and therefore goes into depth, whereas the ethno-
grapher tries to cover a wider field, and mayor may not get into the
deeper layers of the group or organization. Schein illustrates this by
pointing out that the clinician tends to be invited 'back stage', i.e.
to see what goes on behind the curtain from the management per-
spective, whereas the ethnographer is only allowed to see the front
stage from the lower levels of the organization. Schein concludes
(1987:59):
"For the clinician to learn to translate insights obtained in the
field into scientifically valid useful information is a secondary
skill, just as for the ethnographer to learn to help his or her sub-
jects is a secondary skill. It is this secondary learning for each
type that is the real challenge because in organizational work the
two roles so often come to blend into each other."
Evidently, the complementarity of the roles of clinician and ethnog-
rapher is a methodological strength - but also a challenge. We will
examine the tension between roles in the following.
Tension between roles
Schein (1987:29): "The implication of this situational reality is that
the person who works with living human systems must be able to
function in both the clinical and ethnographic role, and, furthermore,
must be highly aware of when he or she is in which role so that nei-
ther relationship is fundamentally compromised."
One of the difficulties of performing several roles, maybe even dur-
ing the same session, is the tension between predicting the outcomes
and in letting outcomes emerge. This is, in fact, a tension between two
opposing paradigms, Newtonian versus Quantum Science.
218
We said earlier that the case study method is anchored in the rational
(or Newtonian) paradigm. A major difference between the case study
method and ethnography is thus that the researcher has developed a
theoretical framework as part of the case study design in the former,
whereas the ethnographer uses an inductive method of letting people
and events 'speak for themselves'. The ethnographer is thus closer to
a 'grounded theory' approach, as described by Glaser & Strauss in
1967. Having a theoretical framework prior to fieldwork does not,
however, necessarily hinder people and events in speaking for them-
selves or in the emergence of unexpected findings. As described ear-
lier, a theoretical framework certainly influences the researcher's per-
ception and interpretation of these (in particular the clinical field
worker who tries to predict the outcome). But when these interpre-
tations are contrasted or challenged by other sources of data, as in
methodological triangulation, this influence will be modified or cor-
rected.
Bias
We have mentioned earlier that cognitive schemas or the underlying
assumptions of the researcher are strong filters for what is seen and
for what data are generated. There are, however, ways to meet that
concern. The researcher can make her general presuppositions re-
garding philosophical paradigm, values and beliefs explicit and try to
hold ideas and hypotheses 'lightly' in order to let the data direct the
findings instead of framing the findings beforehand. Another way to
meet the problem is through methodological triangulation, discussed
earlier.
The demand for intersubjectivity is becoming more in demand by
researchers, mainly because of the ethical dimension. In the naturalis-
tic
6
approach 'member checking' is one way of establishing credibility
as the members or subjects of the study are asked if they recognize the
description or whether they agree with the findings or the conclusion
(Erdunder, 1993:142). This is, certainly, not always unproblematic,
21
9
depending on the subject being studied and on the controversy of the
findings.
Other factors are the conditions of the research itself and the condi-
tions of the 'subjects' (Argyris, 1992). The major importance for us is
the effect and degree of changed behavior, which the researcher's
presence calls forth in the subjects. This includes a varying degree of
hostility (covert or overt) and deception, as for instance trying to
deliberately misinform the researcher or pretending to be cooperative
and "giving the researcher what he wants in such a way that the
researcher does not realize that the subject is doing this" (Argyris,
1992:426).
Another factor is the degree to which the researcher is perceived as
representing management. The clinician is normally engaged by man-
agement, which could easily set off some of the above behaviors. For
the ethnographer it is more a question of being a stranger and conse-
quently of developing relations of trust.
Classification of Data
We usually talk about data in the plural and rarely employ the sin-
gular term of datum, which signifies a basic element of information.
In the present study a distinction between primary and secondary
data makes sense, as it is a real-time study in a real-life situation.
Only in this type of study is primary data generated. Most case
studies are built on secondary data, based on qualitative interviews
about past events. Primary data relates to direct 'experiential',
'practical' and 'presentational' knowledge (Heron, 1981). Secondary
data concerns what happened after the act/ interaction, such as
written reports, empirical data, analytical methods and theoretical
models. Thus secondary data mainly concerns propositional knowl-
edge (and experienced knowledge). Primary data is contextual,
open, and concerns things in their making, whereas secondary data
is decontextual, and concerns closed 'black-boxes' of facts or infor-
mation.
220
It is evident, that the immediate quality of primary data cannot
be kept in any way. Some of the qualities can be revived with person-
al notes, symbols and mind maps. Inter-subjective interaction can be
taped, but even a videotape does not have the authenticity or imme-
diacy that existed at the meeting. Audiotapes and videotapes are,
however, still the most authentic information we can produce from a
real-life situation. But certainly it makes a qualitative difference
whether the person who is to analyze the data has been present or
not.
Regarding secondary data, which most empirical studies must
rely on, a few comments are necessary. Relying on first-person sub-
jective interpretation only, can be problematic. The skill needed here
is for the researcher to develop a kind of 'transparent subjectivity' of
displaying at least some of the steps from raw data to assumptions
and conclusions. This could be, for instance, by making use of the
'Ladder of Inference'.
7
Research with a high degree of reliability
should include a description of some of the steps, aiming at making
the researcher's inferences transparent to others. No doubt this will
also add reflective quality to the researcher's own dialogue with the
text.
Summing up, in relation to science and research the essential point
is that 'voices' of other subjects than the researcher are also heard. In
the present study this is attempted in the triangulated approach where
the aim is to reach convergence of primary and secondary data.
221
Table 4.1: Data type and researcher role in case studies
Descrip- Pilot (0) Base-line Base-line Case 3 Case 4 Case 5 Case 6
tion 1 2
PRIMARY
DATA
Experi- clinician participant participant clinician clinician clinician clinician
encing role and observer observer role and role and role and role and
things in participant participant participant participant participant
their observer observer observer observer observer
making
'High-
context'
interaction
data Notes Notes notes notes notes notes notes
generated: mind maps mind maps mind maps mind maps mind maps mind maps mind maps
notes overheads audiotapes audiotapes audiotapes
mind maps flip-over overheads
audiotapes papers,
flip-over evaluation
papers
evaluations
SECON-
DARY
DATA
Minutes Minutes Minutes Minutes Minutes Minutes Minutes Minutes
e-mails e-mails e-mails e-mails e-mails e-mails e-mails e-mails
question- questions questJons questions questions evaluation questions
naires evaluations evaluations (Success/ (Success/ (success/
evaluations Failure) Failure) Failure)
evaluation evaluation evaluation
informal with many with many With many with with with with all
COllver- scientists & partici- scientists & chairman chairman chairman partici-
sations director pants director and many and many and many pants also
project partici- partici- partici- between
leader pants pants pants meetings
steering gr.
qualitative with with CVP with with with with CP, three
interviews director & project director & chairman chairman chairman CVP's
leader managers director etc
internal teports instruction memos e-mails, e-mails e-mails e-mails,
Org. data: Novo memos e-mails present- present- present- present-
reports Nordisk e-mails ations, ations ations ations,
strategic vision repor- Intra-Web transcribed transcribed Intra-Web
goals memos tings Data-base, tapes from tapes from Data-base,
memos e-mails fInal report article in other prs. other prs. memos
Intra-web Dialogue interviews interviews reports
e-mails strategy
Intra-web newsletter
info articles in
press Dialogue
releases
Dialogue
(int. news-
paper)
external articles articles in Internet-
data: from news- newspaper page: could
media and papers CEO in enzymes
press TV be the key?
222
Criteria for research evaluation
"The unity of all science consists alone in its method, not in its
material" (King, Keohane & Verba, 1994:9)
Regardless of the purpose of research, methodological soundness is
the focal point. In fact, methodological soundness or 'trustworthiness'
seems to be a uniting value for all paradigms. As both a clinical and an
ethnographic approach are applied in the present study, we have tried
to encompass criteria of validity from conventional (positivistic)
research as well as from naturalistic
8
research. But here we will only
mention two pairs or contrasts (Erdunder, 1993: 133):
internal validity versus credibility
reliability versus dependability
According to Yin, internal validity is central for data analysis in case
studies that are explanatory or causal, but not for descriptive or
exploratory studies (Yin, 1994:33). Dealing with a complex problem,
some inferences and explanations will be attempted, but a causal rela-
tionship will hardly be the outcome.
Credibility involves some of the following elements: prolonged
engagement, persistent observation, triangulation, member checking,
and a reflexive journal. Prolonged engagement leads to a deeper
understanding of an organization, e.g. regarding the difference
between being an insider and an outsider
9
The risk is, of course, 'go-
ing native' whereby a researcher becomes so deeply involved that the
research perspective is lost (Erdunder, 1993: 135). Persistent observa-
tion provides depth and involves a thorough examination of different
hypotheses, avoiding early closure, etc. Member checking concerns
interaction and dialogue with the participants about key issues and
also letting them read through the report and comment on it or
change it before it is published. A reflexive journal or a case study pro-
tocol is highly recommended, as this helps to provide reliability. Re-
liability concerns demonstrating that by following the same procedure
as the researcher, this will lead to the same results. Likewise, in natu-
ralistic studies dependability is related to the reflexive journal as well
223
as to an 'audit', based on the journal, data from all the different stag- .
es of data processing, and notes. This corresponds toYin's recommen-
dations of setting up a case study data-base.
This goes to show that, in many ways, what was earlier called the
'soft' sciences are becoming 'hard' in their strict demands for credibil-
ity and dependability, particularly in relation to data processing. In the
words ofYin (1994: 16): "Case study research is remarkably hard, even
though case studies have traditionally been considered to be "soft"
research. Paradoxically, the "softer" a research strategy, the harder it
is to do."
Concluding remarks
Methodology is the focal point of research. The type of research
undertaken in the present study is different from other studies on
innovation, and in this chapter I have tried to explain in what ways. In-
stead of gathering qualitative interviews after innovation has taken
place, I became part of the innovation groups to see what happened
during the innovation. That is why the book is called 'Innovation in the
Making'.
In this chapter I have argued for the advantages of a prospective
method compared to the state-of-the-art, retrospective methods. In
particular I discussed the influence of hindsight bias after things have
developed (or not developed), and pointed out the difference between
the openness of being in a process versus the closedness (black box)
of 'things already made'. This feature makes the present study stand
out.
The selection of methods depends on the problem to be investigat-
ed. The research problem was illustrated through a map of the mul-
tiple facets that could potentially influence the emergence of innova-
tion in the preject phase. This was done to illustrate the complexity
involved and to show the multiplicity and interrelatedness of process-
es and sub-units, as well as the permeability of boundaries between
groups and context. The complexity of the problem justified choos-
224
ing the case study method. To ensure trustworthyness this was com-
bined with triangulation of data generation, research methods and
analysis.
Regarding my own role as researcher, I chose two complementary
approaches: clinical research and ethnography. Eric Jantsch has argued
that different approaches elucidate different aspects of the world:
"Scientific inquiry does not produce 'objective truth' - it pro-
duces a human design which orders and formalizes certain
aspects of reality in a communicable way."
Notes
1 I.e. ongoing experience, see Weick (1995:44)
2 This word is invented (as interviewee, employee) to avoid 'objectification' of
the individuals
3 A special theory has three main components: a philosophical rationale, an ide-
al model based on rules, and recommendations for improving practice (Bor-
mann, 1996).
4 'Narrative smoothing' was coined by Donald Spence (in Shotter, 1993: 127)
5 The name of the book is 'To Weave and Sing'
6 There seems to be some confusion around the term 'naturalistic'. As "the pur-
pose of naturalistic inquiry is to understand the constructions of the respon-
dents on their own terms" (Erdunder, 1993:132) this approach resembles
ethnography, but with an aim at building empowerment and authenticity into
the research, similar to the ethics of cooperative inquiry. Richard Norman,
however, applies the term for the natural sciences, i.e. positivism (Norman,
1979:188).
7 The Ladder ofInference consists of 7 steps (Senge, et a!., 1994): Observable
data, selection of data, adding meanings, making assumptions, drawing con-
clusions, adopting beliefs, and taking action (see also chapter 7)
8 The naturalistic paradigm "deals in the constructions created by the stake-
holders in the context being studied" (Erdunder, 1993: 132)
9 The researcher's own experience regarding this is described in chapter 5 in
the case narrative
225
Chapter 5
Case study overview
and narrative
"In all of these situations, the distinctive need for case studies
arises out of the desire to understand complex social phenom-
ena. In brief, the case study allows an investigation to retain the
holistic and meaningful characteristics of real-life events ... "
(Yin, 1994:3).
This chapter
ture workshop and the base-line studies. The chapter ends with a
'post-study' narrative of my work as innovation coach, the implemen-
tation of the results.
Introduction to case studies
The following case study overview provides a simple illustration of the
case studies that were conducted to study 'Innovation in the Making'.
The circles illustrate the number of groups involved in each study.
The pilot study does not 'count' as a case study, but was conducted
in order to assist "an investigator to develop relevant lines of questions
- possibly even providing some conceptual clarification for the
research design" (Yin, 1994:74). Therefore the pilot study gets the
number O. The pilot study was conducted when I was an external con-
sultant, whereas the case studies were all conducted from within the
organization.
Case study overview
In this chapter we focus on praxis, the case studies. The guiding prin-
ciple here is Aristotle's third intellectual virtue of'phronesis', i.e. pru-
dence. Prudence concerns values and judgment and depends on the
situation and context.
The purpose of this chapter is fourfold: 1) To provide the reader
with a general understanding of the case study portfolio, 2) to add
'life' to the case studies through a subjective narrative account, 3) to
describe the focus of the preject more accurately, and 4) to arrive at
a description of Novo Nordisk culture and 'general practice', which
will (later) serve as background knowledge for the findings and data
analysis.
The case study portfolio is illustrated in a case study overview,
which is accompanied by a first-person subjective narrative of how the
case studies came about and developed. The narrative includes
excerpts from the research diary as well as some 'cultural clashes' that
happened on the way. To increase the understanding of the preject-
project figural space, the focus and process area of each case study is
illustrated with arrows. An analysis of Novo Nordisk culture and 'gen-
eral practice' of knowledge creation follows, based on data from a cul-
226
Pilot study (0)
Base-line case study (1)
Base-line case study (2)
Case study 3
Case study 4
Case study 5
Case study 6
oe5bo
00
o
o
o
o
Role of researcher
participant-observer
and intervention
and intervention
and intervention
and intervention
and intervention
227
Simple preject-project
Going Narrative
PROJECT PREJECT -
For each case study the illustration
will display whether the case study
operates in the divergent or in the
convergent space - or both. It will
further display 'where' in the figural
space the beginning takes place, i.e.
if it starts in the divergent or con-
vergent space, and which course
and direction is taken by the group.
Finally it will appear that the total-
ity of cases cover all of the figural
space and that one case in particu-
lar covers most of the space. To
make the illustration simpler, most
of the explanatory terms have been
erased from the original model.
In the case studies, all persons,
male or female, are described in the male form to ensure anonymity.
The 'time of writing' referred to during the analysis was Spring 1998
in the excerpts from the diary, and February 1999 in case study 6.
Two base-line studies were conducted with me (the researcher) as par-
ticipant-observer before intervention was attempted. This was done
because a real-time, real-life study with a researcher doing clinical and
ethnographic research poses a few methodological challenges. Apart
from the tension between the two roles (described in chapter 4) and
the difficulty of discerning these in practice, there is the 'problematic'
(from a research perspective) irreversibility of human interaction.
Meetings cannot be rewound and started again in order to tryout a
different approach. Intervention cannot be 'undone' or 'redone'. So
the hypothetical question of what would have happened if ... remains
unanswered.
A related challenge is to establish a 'before' the intervention in
studies of process. Process is continuous and ongoing, there is no 'be-
fore' and no 'after' but only 'in'. The base-line studies were conduct-
ed to establish a 'before' or a description of Novo Nordisk's 'general
practice'. In particular the first base-line study was informative and
generative. The central case study was no. 3. In fact, case study 3 has
qualities that make it 'revelatory'. According to Yin (1994:40): "A
third rationale for a single case study is the revelatory case. This situ-
ation exists when an investigator has an opportunity to observe and
analyze a phenomenon previously inaccessible to scientific investiga-
tion ... " This is the case here: The phenomenon under study is, under
normal circumstances, highly inaccessible. Still, the research design is
not a single case study, but a multiple case study including a revela-
tory case.
The preject-project figure
"We cannot avoid using narratives, metaphors or theories, but
what we can avoid is becoming entrapped within their confines
by claiming anyone of them to be the single correct narrative,
metaphor or theory. They are instruments, not depictions."
(Shotter, 1993: 132)
The idea of using the preject-project figure is to illustrate the 'space'
occupied by each case study. This space is abstract and figurative,
needless to say, but it will serve the purpose of a 'processual pointer'
bridging 'immaterial' processes in their making with a 'cognitive
space'.
The reader should now prepare for a different style of writing, as we
turn from a professional-scientific neutral text into narrative. From
describing and discussing in the grammatically 'passive' mode or us-
ing the form of 'we', the text turns into a story from real life with an
individual 'I' who experiences and reflects.
228
229
Three features of narrative deserve to be mentioned: 1) narratives
describe intentions of a subject 2) they are indeterminate or ambigu-
ous and 3) they involve the reader's imagination in a particular way:
"The reader receives it by composing it" (in Bruner, 1986:24).
Regarding the intentions, the following narrative is topic-centered
around the theme of 'chasing the beginnings'. The reason for chasing
beginnings rests on the proposition that beginnings are important for
group climate and communication - and thereby also for innovative
processes. As for indeterminacy, this quality has been difficult to
maintain - after things have developed. I have tried to stick to the
quality of indeterminacy and openness by not changing what was
already written. Instead I have made efforts to add retrospective
reflections afterwards as footnotes. Nevertheless, the narrative does to
some extent suffer from 'hindsight bias' and 'narrative smoothing'
(Shotter, 1993).
My first official presentation took place at a quarterly meeting of
about 80 persons at a Health Care Discovery department in October
1995. I remember how difficult it was to judge the reaction to my
presentation, but it seemed that at least some people were interested.
I was told later, that it had been well received. Shortly afterwards I had
meetings with the director and with his manager group, and it was
decided that I should conduct a two-day workshop for a group of 12
people, introducing them to frameworks
4
and methods for working
with the preject phase of innovation. This took place in December
1995 and the outcome of the workshop was very satisfactory. The
group came up with four new areas they wanted to examine by using
the new framework (Kubus). I enjoyed working with the groups. They
were professional, knowledgeable and open. Evidently, it was a way of
working that differed from what they were used to, e.g. they were not
used to any procedures of sharing knowledge.
The pilot study
Czarniawska-}oerges
1
writes about the narrative form: "In order
to understand our own lives we put them in narrative form and
we do the same when we try to understand the lives of other
people. Thus, every action acquires meaning by acquiring a
place in the narrative of lives."
My first contact with Novo Nordisk was due to an interview to a Dan-
ish weekly business news1etter.
2
Someone from one of Novo Nordisk
Health Care Human Resources departments was interested in what
we were doing regarding systematic development of innovation in the
Center for Innovation and Entrepreneurship, Copenhagen Business
School, and we had a couple of meetings. As I was, at that time, look-
ing for an organization that would be interested in sponsoring an In-
dustrial Ph.D., I was told to go ahead with this opportunity by my col-
leagues. I hoped that if! did a good job, Novo Nordisk would want an
Industrial Ph.D. 3 in that area.
23
At the first status meeting, after six
weeks, all groups had found and
developed substantial knowledge
in the new areas, and two of the
groups had come up with new
concepts and ways to test these
that made them potential pre-pro-
jects. I could, of course, not evalu-
ate the quality or novelty, but two
prejects were presented at the fol-
lowing Health Care Discovery
Management meeting. One was
accepted as a 'formal' pre-project,
which meant allocation ofresourc-
es.The other was recommended to
continue working with the ideas,
to test them, and then to come
back. As illustrated, the groups
were preject groups searching for
new knowledge, i.e. in the diver-
gent phase.
PREJECT - PROJECT
V
I
<>I
I
Covered 'space' of Pilot study (0)
23
1
In retrospect I learnt that the timing of my entrance had been perfect.
The director of the function had a fairly new organization due to a
recent re-organization, and people were spread geographically in
many areas) which meant that people did not know each other well.
Thus, sharing knowledge in groups was a constructive way of
strengthening cooperation. Furthermore) the Corporate Vice Presi-
dent of Health Care Discovery had asked all directors to come up with
innovative pre-projects as the project portfolio was 'thin' in the enter-
ing stages with few pre-projects.
By the second status meeting, however, a lot of the previous condi-
tions had changed. By now, management had enough pre-projects and
did not have the same vivid interest anymore. Some central scientists
left the function (and the preject groups)) and ongoing projects called
for all resources (and more). The pilot project ended with a highly
favorable
5
evaluation in November 1996, but also pointed out some
drawbacks
6
to the framework (Dars0, 1997a). By then I had already
started my Industrial Ph.D. in the organization.
The pilot study differed in many ways from the present study. The pri-
mary task of the pilot study was classical clinical work done by an
external consultant, trying to help solve a problem of feeding new pre-
projects into the project portfolio. The secondary task was for the
researcher to test some methods and frameworks, and find out if and
how they fit the organization. This was done by conducting an evalu-
ation with questionnaires, followed by a presentation of the findings to
the participants and a discussion about these (Dars0, 1997a). Thus
the pilot was a way to test out a tentative case study design, as recom-
mended byYin (1994:52).
The pilot study was based on the Kubus framework, which was devel-
oped from empirical studies of conversations in groups (Herlau, 1995)
into a normative prospective method. In the present study some of the
Kubus methods? have been applied, since I had practical experience
with these, but not the main Kubus model (except once in the medi-
ator group, case study 6). The advantage of that decision was that the
model could be used deductively for coding the topical content of the
conversations, as discussed in chapter 7.
23
2
From External to Internal (Excerpts from diary 1998)
"I started working in Novo Nordisk Health Care Discovery & De-
velopment Human Resources unit on September 1, 1996. The first
thing that struck me was the difference from being a visitor (a con-
sultant) to becoming a 'Novo Nordisk person'. Suddenly all the
things I had not previously been able to get hold of were easily avail-
able. A strange mix of confidential and trivial information kept com-
ing in, both through my e-mail and through conversations and meet-
ings. I was overwhelmed with the loads of information that kept
flowing in.
I remember how I kept saying 'you in Novo Nordisk' in the begin-
ning when talking to people in the organization) and I remember how
they looked a little surprised, like saying 'are you not part of Novo
Nordisk?' So then I decided, after a month or so, to say 'we in Novo
Nordisk'. This was, however, noticed by people outside the organiza-
tion and interpreted as if I had been absorbed and had quickly tak-
en over the Novo Nordisk culture."
During the four years I worked in the organization I tried to main-
tain a critical perspective. I managed to understand the culture a lot
better, bur cultural clashes still happened.
First cultural clash
During the pilot project, some time after the first pre-project had
been accepted, I asked the director what would now happen with the
pre-project? He said that it would have resources allocated, but that)
in fact, he wondered why he had not heard any news about it.
Through a phone call he found out that the pre-project was 'stuck'
in the administrative system) because there was no appointed project
manager. The system had no 'category' for shifting leadership (the
Kubus method). In order for the pre-project to get resources, the
director finally had to point out an official project leader (in agree-
ment with the group). In my reflective diary I described this event as
"a clash between individual versus collective thinking". "They talk
'teams' (espoused theory) and act 'individuals' (theory-in-use
8
)."
233
This is a good example of how systems influence what is possible
and how systems can block novel initiatives (systemic thinking, as
Senge, et al, 1994).
Chasing the 'beginnings'
The first thing I was told was to network. "You can't survive without
it," my advisor said. I had no idea then how right he was. He started
by introducing me and my project via the internal e-mail system, but
the news had few reactions. This could be due to many things. People
were busy (as usual) and suffering from information overload. It was
in the middle of the MAX process, where 20% of the Staff and Ser-
vice functions would be cut down. The project was maybe seen as aca-
demic (i.e. too theoretical) coming from a Business School or it was
classified as a HR project (i.e. process oriented).
I was introduced to a lot of people, among others to a director of
Discovery, for whom I carried out a climate survey in his department
(40 people) in order to get to know the scientists. It seemed that
working with people and earning their trust and respect was the only
way to get what I wanted: starting preject groups in order to study
innovative processes and innovative crystallization. Thus, when my
advisor spotted tasks that could generate opportunities for my pro-
ject, he asked me if I wanted to take them. In this way I soon became
involved with three parties: a director, a manager and a group of
managers. Only one case study (base-line study 2) came out of these
contacts. I also generated data with the manager's group, primarily
on processes of cooperation and strategy, but the objective was not
innovation. It is therefore not included. The group of managers lost
interest in the project on organizational innovation before the prob-
lem was identified. I think it was due to prioritization and time pres-
sure.
During the first six months I socialized with a lot of Novo
Nordisk people, and I introduced my project to many people and
groups. I always explained very carefully how important it was for
my research to join a pre-project group right from the beginning. It
often happened that I heard of a group that had just started, but it
234
seemed impossible to catch projects right at the starting point. It
became more and more mysterious how projects actually started. I
did some interviews with directors and project leaders, and they
explained how some people were very good at getting ideas by doing
'skunk-work', and that all pre-projects were started by these 5-10 key
persons. I explained that there was a different way of doing this, that
these competencies could be developed in groups, that this had been
done both inside Novo Nordisk and outside Novo Nordisk with
good results, but they were not convinced, and it seemed as if they
were not even interested in finding out if this could, indeed, be a bet-
ter method and add quality to the early pre-projects. Thus, the prev-
alent view was that bottom-up pre-projects were started through
informal networks and skunk-work - and by individuals. Pre-projects
would develop and grow and at some point they would be presented
for management. This meant that some pre-projects took ages to
develop, depending on the time available and the drive and networks
of the individual researchers. Others came about quite fast, as an
individual or two would be working day and night on something that
looked promising.
How does a reorganization feel?
(Excerpts from research diary, 1998)
"You hear stories, during lunch for instance, or at the Friday morning
coffee sessions. One story I heard a couple of times was that organiza-
tional change would usually take place right before Christmas. Appar-
ently this had happened a few times. The large reorganization in 1994
was said to have happened like this. 'You are fired, but we'll of course
try to replace you, Merry Christmas'.
This brings about a certain cynicism. It took time before I under-
stood. I think I really understood when a reorganization hit our own
unit of Discovery & Development Human Resources. For some time
I did not know where I would be in the future, as my advisor was
negotiating his new position. It provoked some anxiety about my pro-
ject and future. But when that was over with, it meant that we were
now 25 people instead of 6 in the Health Care Human Resources
235
department under a different director. After having been here for 20
months, I have realized that reorganizations happen all the time
9
,
small and large. Actually every time a new person changes job and is
put in charge of a different unit, a reorganization is expected. I found
out that one person uses 1/3 working time for changing the organiza-
tion charts. I wonder if reorganizations are a way of keeping the organ-
ization on the move away from stability, from equilibrium and whether
it is done consciously?"
Base-line case study 1: Values In Action
The first time I heard of a starting project before it actually started was
the "Values In Action" project. I heard of it the day before it started,
through two channels. One was a professor at my institute at the Busi-
ness School, who told me that he was going to do a presentation in
Novo Nordisk about a project that was just starting. The other was a
director at Novo Nordisk that I knew well, who was going to partici-
pate in the project. I contacted the person in charge, and more or less
told him that he had to invite me to participate, which he then did. So
I literally threw myself into the project, which became base-line study
1. I hoped this way to find out how project groups would normally
tackle complex problems and to be a participant-observer in case
some kind of crystallization happened. This study has been described
in a paper written for the Conference of Uncertainty, Knowledge &
Skill, in Limburg, Belgium, 1997, and the following will, to some
extent, consist of excerpts from that paper. First a short description of
the project (Dars0, 1997c:6):
"In May of 1997, a large trans-organizational project was
launched by corporate management. The purpose was to ensure
that company values would be embedded in the daily handling
of issues related to social, ethical and environmental matters and
the objective was to form a proactive strategy and action plan.
The scope of the project was global, as it involved the whole
organisation, including the subsidiaries in various countries.
That made the project not only cross-organisational but also
cross-cultural. The project dealt with knowledge creation under
uncertainty, complexity and change.
The project involved 30 - 35 persons from all over the organisa-
tion. The decision was thus top-down, but the work consisted of
a bottom-up co-operation. A project organisation was estab-
lished, consisting of a project leader, a steering group and 5
working groups. The time frame was set to four months, and the
project was divided into three phases, concluding with recom-
mendations and action plans to be presented for corporate man-
agement."
"As the main area of interest in this study (the Ph.D.) was the
point of departure and management of the start-up process, I
attended the first meetings of all the working groups as well as
those of the steering group. All in all, I have attended 15 meet-
ings and 2 all-day workshops."(Dars0, 1997c:7)
I was happy that I had managed to get involved in the Values-In-
Action project because I learned a lot about the Novo Nordisk way of
project work - and, through that, about Novo Nordisk culture. (This
will be discussed at the end of the chapter.)
"All in all, in the present case study, as in most project work, the
divergent phase was mostly ignored (therefore: the dotted line),
and 'the usual way' of doing project work was applied. It can be
seen in the data from the meetings that these have to a very large
extent been in the area of uncertainty, e.g. lots of questions and
clarifications. This is symbolised by the loops back into the
divergent phase. However, this was to some extent relieved by
the project leader as guidance was provided in the templates
given to the groups to be filled in." (Dars0, 1997c: 9)
237
Illustration of base-line 1
I had many deliberations re-
garding how to illustrate the
first base-line study. How
could I demonstrate that the
majority of the groups skip-
ped the preject phase? And
howcould I plot in a time line?
I finally decided to illustrate
it by starting a time line in
the middle of the figure, indi-
cating that the groups started
as if it was a project, but then
evidently had to back-loop
into the preject phase in or-
der to clarify things.
In the writing process I
made several models.
divergent prejeet phase
start
... I
time
convergent project phase
target
)
Second clash: Don't ask if you don't have the answer!
At a meeting in the steering group (of the Values in Action project) I
had pointed out that the term 'ethics' was problematic, as people
understood it in multiple ways (without realizing it). Some thought it
might be a problem, others didn't. We rushed on to keep within the
time and the agenda. At a meeting afterwards with one from the
steering group I raised the issue again. He asked if I could provide a
clarification. (I had just had a Ph.D. course in ethics and had tried
to make a simple table for clarification, but had given it up as I was
very short of time and as I generally took on only tasks that would
generate data for my Ph.D.). I explained that I unfortunately did not
have the time. He got furious and shouted: Don't ever bring some-
thing up in a group if you cannot provide the answer! I was literally
shocked. It was the only time anybody shouted at me in Novo
Nordisk. But if you come to think about it, it is a frightening remark.
It demonstrates a clinging to certainty, and no curiosity or interest in
uncertainties. I wondered if that was a personal attitude or if it was
part of the culture?
JlIustration of base-line case study 1
A different version is fig. 5.6,
indicating that the scope of
most projects was limited.
The following excerpt is from
the conclusion: "Certainty or
uncertainty, the ways of pro-
ject management have not
changed. We continue with
old routines without realising
that the world around us has
changed and that conse-
quently new methods are
needed."
start
divergent prejcet phase
time
COQ\'ergent project phase
Base-line case study 2
As explained above I began to cooperate with a director in Discovery
through a climate surveylO. Later I met with him and his management
group, presented my project and my ideas and started to follow the
meetings of two project groups. I wanted to find out what went on in
these meetings, what kind of interaction took place and how the com-
munication functioned in such a group. I was also interested in the
project management style.
239
I invested a lot of time and energy in the people from that particular
department to establish a relationship of trust and to step in and start
a new group (a case study). I even carried out one more climate sur-
vey and was negotiating with the director about a new approach to
team-work aiming at getting a better use of the expertise and know-
ledge in the unit. From one day to the next, however, the situation
changed when, unfortunately, due to the second re-organization of
Health Care Discovery in December 1997, this function was split up
and the director went back to work with his area of expertise. I believe
that it was hard on the director and on the unit, which was function-
ing well. For me it meant a lot of work, a lost opportunity, but also
some gained experience and data.
My general impression from the
meetings that I attended was
that project meetings were rath-
er formal and goal directed,
scheduled beforehand with peo-
ple presenting their findings.
This is illustrated by the arrows
in the project area. There was
not much spontaneity, but
sometimes a great deal oflaugh-
ter. The meetings included both
scientists and technicians. In
one of the groups there were
two people who posed some
sharp questions that led to an
interesting discussion, but these
traces were not followed. Maybe
it was outside scope, maybe it
was not interesting enough. My
knowledge of Diabetes was not
sufficient to judge this.
PREJECT - PROJECT
Covered 'space' of base-line case study 2
Case study 3
The situation would probably have been unbearable if I had not,
already then, been involved in some enterprises in Enzyme Business.
During the summer of 1997 I was contacted by two people (separate-
ly) regarding a workshop that was to take place in October 1997. En-
zyme Business management had selected some business processes for
investigation, and one of these was the 'From Idea to Sales' process.
One of the persons who contacted me was responsible for the first part
of that process and had planned a two-day session regarding 'Idea
Generation and Evaluation'. People were joining the workshop from
USA, China, Singapore, Paris, so I was asked if I could help him plan
the session. I explained to him about some of the 'traps' that I saw in
such a session, the trap of looking for solutions before the problem
had been identified, the trap of believing to start from common
ground, when the ground had not been covered, and the trap of forc-
ing closure before enough data and knowledge had been generated.
He could easily see my points and seemed content when I offered to
plan and chair the workshop - on the conditions that I could tape all
the conversations, as it would be part of my Ph.D. study.
During the fall of 1997 I was contacted by many different people
from Enzyme Business in relation to other innovative activities and I
had several meetings regarding how to encourage innovation in En-
zyme Business Research and Enzyme Development & Application.
The workshop started on a sunny October morning in Hvid0re in
a room facing the seaII. People were generally kind and open towards
me, the program, and even to the video camera on the table. One per-
son from a high managerial level, however, was not content. He want-
ed to do 'business as usual', because it worked well, and he could see
no use for all this sophistication. Luckily I have had experience with
all sorts of people, and I had decided that no matter what happened
(almost!) it would always be interesting from a research point of view.
On the other hand, I have also seen sessions being ruined by one very
negative person, so I was on guard. This person did not change his
mind (about the usual way being the best way) throughout the work-
shop, but despite that he thought it was successful.
I had told the participants from the start that this was more or less
what would happen. I had shown them the preject-project model and
explained that after the first day we would be close to the middle of
the window, either on the left side or in the middle or, if we worked
really well, through the window. By the end of the first day, the group
was (figuratively) still on the left side of the window. People did not
feel good about this confusion, which they were evidently not used to,
but at the time I did not think it could be any different. The advan-
tage, I thought, was that confusion urges the brain to keep on think-
ing, even if we are unaware of it.
The next morning, however, about 35 minutes into the session,
crystallization happened. I recall the great thrill when everybody was
bent towards each other, talking eagerly about the new concept. The
blackboard was used, concepts were outlined. It was as if everything
I had asked everybody beforehand
to prepare a Success/FailureVisu-
alization12 for the session and
most people had done so. Thus the
opening session was positive and
an open and participative climate
developed. In the afternoon the
main task was to identify barriers
against what the group was about
to invent. The beginning went
fine, but towards the end of the
day, some confusion roamed. A lot
of barriers, ideas, stories, exam-
ples, thoughts had been aired and
were still in the air (actually on
flip-over papers and white boards)
when we closed the session.
On the illustration the arrows
are meant to demonstrate that
both the preject and (part of)the
project space were covered.
PREJECT - PROJECT
I
Q
<>
Q
o
I
Covered 'space' of case study 3
that had been in the air from the day before fell down again right into
its place. All the data and knowledge was used, and everybody contrib-
uted. This took all day, but it was a happy ending, and the evaluation
of the two days and the results were positive.
Third clash: 'Our' world versus 'your' world
I met a particular attitude in the organization several times during my
work. In my research diary I wrote: "The typical Novo Nordisk per-
son is extremely kind, but also very skeptical. I do not know whether
this is a general cultural value, or whether it is because I am somehow
perceived as representing Human Resources, or as coming from
Copenhagen Business School. I have heard that Human Resources is
a waste of time (particularly when I arrived in 1996). I have also been
accused many times of being too theoretical or too abstract. I try to
speak the same language, but it is perceived as being different."
A similar clash happened in the mediator group, when one of the
mediators and I had been assigned the task of producing a draft about
the mediator work for the internal newspaper 'Dialogue'. We mailed
this to the group and immediately received an answer from another
mediator that this was not meant to be an academic exercise about
process, but concerned a goal-oriented project about developing the
business. The issue was debated at the next mediator meeting, where
clarifications, strong opinions and disagreements were aired. When the
article came out in the internal paper (Dialogue) the description of the
work process was more or less left out with focus on the results.
Case study 4
The 'flow maker'13 of another business process project contacted me
because he had heard about the results of the above workshop. He had
experienced a rather frustrating and unsuccessful workshop when try-
ing to create a strategy for one of the five business processes that was
being dealt with in Enzyme Business during 1997. We had a couple of
243
This was clearly expressed by a participant in the written evaluation:
244 245
PREJECT - PROJECT
Case study 5: Creative Problem Solving
Aproject group was stuck with a com-
plex problem, the causes of which
they could not identify as the process
involved many elements that could
mutually affect each other. As I had
posed the question many times to
myself regarding what role creativity
could play in innovation I agreed to
work with this group. I undertook the
task and designed a workshop, hoping
that something useful would come
out of it, both regarding some practi-
cal results for the group and regarding
some data for my Ph.D.The process is
illustrated by arrows in the preject
space, as we tried to frame the prob-
lem in multiple ways, and then gener-
ated a lot of ideas. f
This case study was conducted to find out what could be accom-
plished through creative problem-solving techniques.
Two things stand out (in my memory) from that seminar: The 'fire-
wheel' exercise
l4
and some political tensions between two units. I will
dedicate a few explanatory words to the former. The idea was to define
and clarify the central concept by trying to identify a common core and
allowing different perspectives to connect to the core. This exercise
turned out in a very positive way and became common ground for the
rest of the discussion. The perspectives were found to fit as strategic
elements, and these were minutely examined for 'flow-stoppers' (bar-
riers) and 'flow-enhancers' before creating an action plan and assign-
ing action owners. Several gaps were identified in relation to activities
that had no action owner at the time. The results of the workshop were
considered satisfactory by the participants, and later by management,
and actions were assigned in order to fill the identified gaps.
PREJECT - PROJECT
"Given that none of us really felt strongly about the success of
the seminar (except that this should be the last one), I think the
attitude was positive, cooperative and dedicated."
meetings examining what went wrong in the previous workshop. He
described the unsuccessful half-day workshop as one big brainstorm-
ing without anybody being able to agree on anything. He and another
person were trying to investigate what actually happened by interview-
ing some of the people involved. I suggested that the problem could
be that the participants had completely different perceptions regard-
ing the central concept. This brought about an 'AHA' experience,
because the difference in perspectives was clearly seen in the inter-
views. I also suggested that the amount of time (1/2 day) and people
(approx. 20) had not favored a good result.
It did not take much to talk me
into planning and chairing the
next workshop. Here I had a
chance to try to 'redo' a beginning
and find out if that would make a
difference. This is illustrated by
the arrows to the left. Once the
different perspectives of the con-
cept had been clarified, however,
the rest of the workshop was spent
in a more focused project space,
indicated by the arrows there.
It was explained to the people
invited to the workshop that we
wanted to do the workshop all
over again, trying out a different
methodology and approach. Still,
the attitude of the participants
was not very positive at the out-
set, but it changed for the better.
Unfortunately my tape recorder did not tape the conversation becaus@
of some technical difficulty. I would afterwards have liked to be able
to listen to the first session, where we used some time to discuss the
problem from different angles trying not to close it or frame it too rig-
orously. I do, of course, have my own notes and the group's summary
from the meeting.
After having worked out an open formulation of the problem, the
participants were divided into three groups and a creative technique
applied
I5
. This way many good ideas, tests and solutions were found
16
,
classified and evaluated, which the project group could bring back and
apply to solve the problem.
Case study 6: The Mediator Group
The last case study is an embedded longitudinal case study, consisting
of monthly meetings during more than a year. Case study 6 had a
wider scope than the focus of the research and concerned the devel-
opment and implementation of one particular innovative crystalliza-
tion. Here the MIRP process model was applied and used as a guid-
ing structure for the analysis. It was found to be constructive for
understanding and discussing the implementation of innovation and
afforded a holistic process perspective.
This case study is interesting in many ways. Firstly, it concerns the
implementation of the 'crystallizations' from case study 3. Secondly, it
includes several small sessions of trying to develop and evaluate the
more 'wild' or 'off' ideas towards innovative crystallization. Thirdly, it
displays culture, management and political issues when the group
meets the constraints of introducing a new organizational approach.
The mediator group, a name that was developed after the work-
shop of case study 3, met for the first time in April 1998. Since then
I have attended approx. 10 meetings, which have been taped. The
original idea of the mediator group concerned connecting people with
good ideas to people with knowledge about the area. In order to
ensure interest and commitment from the people appointed to the
mediator task, people had to send in applications (instead of being
appointed). Out of 28 applications, seven mediators were selected,
who could cover all the different lines of business and who had net-
works throughout the organization. They were given 25 % of their
time for this work, or rather they were supposed to be given 25%. In
reality only one or two people were allocated the resources. The input
to the mediators came primarily from the new Opportunity web page
that had crystallized at the workshop (case study 3). This 'New Op-
portunity Page' was launched in April 1997. The challenge was how
to evaluate, develop and apply the ideas and opportunities that came
streaming in.
Another reason for following the implementation of the mediator
concept was that it was new and had not been tried before. And
because I had been involved in the making, I 'caught' the beginning.
At the first meeting I introduced myself, presented some of the ideas
of my Ph.D. and stressed the importance of creating common ground
as a point of departure.
I suggested that everybody should make a Success/Failure Visual-
ization in order to create some of that common grounding. I urged
them to discuss individual ambitions and desires, expectations
regarding how the group should work and cooperate, and visions
about what kind of results or successes the group should aim at.
Most of the participants filled in the S/F template, but there was not
much discussion. Some participants would rather 'get started'. Since
I had decided to be mainly participant-observer and step in only if I
found that frameworks or methods were needed, I left it to the group
to decide how to get on with the task. They decided to start work-
ing and then take up principles and process along the way. This
meant, in my interpretation, to do things the usual way: by Trial-
and-Error.
247
At the time of writing
l
? it is hard to tell whether the mediator group
is a success. Their work is not very visible, even though the Oppor-
tunity Web Page has been described in the internal monthly paper
twice, and wine have been given for the three best ideas so far. To me
it looks like they have become trapped in administration, as the ideas
kept coming in. The group is not given enough resources to put into
innovation activities and have difficulties getting hold of the necessary
knowledge about market potential.
It has been interesting to follow the development from having no
system or model for evaluating incoming ideas till the opportunity
'score card' crystallized. It had to be adjusted a little when put to use,
but has, since then, worked well. After the score card was developed it
was 'tested' on some 'new lead' projects from 1995. The score card
was applied to check out which ideas would have been started if the
score card had existed then, and afterwards this was compared to
which 'new lead' projects had actually made it. Only one project that
would have been stopped by the score card had made it, and this pro-
ject was special in relation to a network connection, so in a way it did
not really count. More importantly, some new lead projects from 1995
I have tried to illustrate the process
of the mediator formation and pro-
gression. There is a small arrow (to
the left) indicating some work with
creating common ground, but then
moving over to the project side. The
backlooping circle illustrates that at
times the process was stopped and
process matters were taken up when
necessary. The small punctured
cubes indicate new prejects in the
making and the crystallization was
the New Opportunity Score Card.
The long arrows to the right are
meant to illustrate that some ideas
could be implemented right away.
PREJECT - PROJECT
<>
to Ii I ~
....... : ~
< > .
had occupied resources for a considerable amount of time before they
were closed, and these would not have been started, if the score card
had been in use.
Another interesting session was when an idea came in through the
Web Page and a problem was formulated. In order to evaluate the idea
a specialist was invited to do a presentation on the subject. Afterwards
the problem was reformulated with a completely different focus. This
was an example of framing and reframing the problem.
All in all, case study 6 has a richness and abundance of data, of
which only a small part has been used. I think that most researchers
are afraid of not having enough data, but I also wonder whether it is
not most often the other way around.
Identifying Novo Nordisk cultural values
The last account from the study is about a session of approx. three
hours, where I was involved as a 'catalyst' for a group process. It con-
cerned the crystallizing of Novo Nordisk cultural values in a so-
called Better Practice Group of directors from across the organiza-
tion. The group had been working for some months on values and
value based leadership and was at a point where they needed a per-
son and a method for this process of (forced) crystallization. My gain
was that the process generated valuable data that could be used in
my Ph.D.
I started with an introduction of Schein's model of organizational
culture
l8
(artifacts, values, and assumptions), McGregor's theory X
and Y regarding human nature (control versus trust), and three types
of basic orientation (being, developing, doing). After this we did a
'tracking' exercise, first looking 10 years back and then looking 10
years ahead, in order to find some general trends and some specific
events, which had influenced Novo Nordisk (see table below). The
idea was, through this exercise, to anticipate in which ways the gener-
al trends of the present could influence the future values of the com-
pany. Looking back, it was agreed that the most significant and
influential event for the company was the American Food & Drug Ad-
249
I have tried to illustrate the process
of the mediator formation and pro-
gression. There is a small arrow (to
the left) indicating some work with
creating common ground, but then
moving over to the project side. The
backlooping circle illustrates that at
times the process was stopped and
process matters were taken up when
necessary. The small punctured
cubes indicate new prejects in the
making and the crystallization was
the New Opportunity Score Card.
The long arrows to the right are
meant to illustrate that some ideas
could be implemented right away.
PREJECT - PROJECT
had occupied resources for a considerable amount of time before they
were closed, and these would not have been started, if the score card
had been in use.
Another interesting session was when an idea came in through the
Web Page and a problem was formulated. In order to evaluate the idea
a specialist was invited to do a presentation on the subject. Afterwards
the problem was reformulated with a completely different focus. This
was an example of framing and reframing the problem.
All in all, case study 6 has a richness and abundance of data, of
which only a small part has been used. I think that most researchers
are afraid of not having enough data, but I also wonder whether it is
not most often the other way around.
Identifying Novo Nordisk cultural values
At the time of writing
l
? it is hard to tell whether the mediator group
is a success. Their work is not very visible, even though the Oppor-
tunity Web Page has been described in the internal monthly paper
twice, and wine have been given for the three best ideas so far. To me
it looks like they have become trapped in administration, as the ideas
kept coming in. The group is not given enough resources to put into
innovation activities and have difficulties getting hold of the necessary
knowledge about market potential.
It has been interesting to follow the development from having no
system or model for evaluating incoming ideas till the opportunity
'score card' crystallized. It had to be adjusted a little when put to use,
but has, since then, worked well. After the score card was developed it
was 'tested' on some 'new lead' projects from 1995. The score card
was applied to check out which ideas would have been started if the
score card had existed then, and afterwards this was compared to
which 'new lead' projects had actually made it. Only one project that
would have been stopped by the score card had made it, and this pro-
ject was special in relation to a network connection, so in a way it did
not really count. More importantly, some new lead projects from 1995
The last account from the study is about a session of approx. three
hours, where I was involved as a 'catalyst' for a group process. It con-
cerned the crystallizing of Novo Nordisk cultural values in a so-
called Better Practice Group of directors from across the organiza-
tion. The group had been working for some months on values and
value based leadership and was at a point where they needed a per-
son and a method for this process of (forced) crystallization. My gain
was that the process generated valuable data that could be used in
my Ph.D.
I started with an introduction of Schein's model of organizational
culture
l8
(artifacts, values, and assumptions), McGregor's theory X
and Y regarding human nature (control versus trust), and three types
of basic orientation (being, developing, doing). After this we did a
'tracking' exercise, first looking 10 years back and then looking 10
years ahead, in order to find some general trends and some specific
events, which had influenced Novo Nordisk (see table below). The
idea was, through this exercise, to anticipate in which ways the gener-
al trends of the present could influence the future values of the com-
pany. Looking back, it was agreed that the most significant and
influential event for the company was the American Food & Drug Ad-
249
ministration inspection of 1994 of the insulin production. The Amer- .
ican Food & Drug Administration found an unacceptable amount of
errors, which if not corrected immediately would mean that the US
market could be lost. The company was in a state of chock and an
army of consultants was called in to assist. This crisis was found to
have major impact on the change in values. The table below was made
by the Novo Nordisk managers.
Values before FDA in 1994 Values after FDA/Current values
Values before FDA in 1994 Values after FDA/Current values
anti-bureaucracy accept of bureaucracy (tapping energy, over-systematized)
each person knows how to behave control/motivation is necessary (theory X)
(theoryY)
acting (free, acting person) systems make it more difficult to act. Partial paralysis/ lack of
decisions. Can we act, or do we not dare to? Empowerment is
not only delegation. Management must set frames and add
clarity for the empowered group
improvise/be creative/ have courage internal stakeholders are sworn in. CMA (cover my ass)
attitude. Less risk willingness
flexibility: the orientation of the flexibility demanded by the organization (external to the
individual individual) a readiness for change - internally or externally
forced?
ambitious, professional ambitious, business-oriented and professional. Novo Nordisk
wants to be best, but are we, do we really try?
respect for the individual the individual can be replaced (succession planning)
'family' (connectedness) - both lack of ceremonies
positive and negative value
automatic salary increase, based on performance evaluation, job classes, bonus
education and seniority
trust in people - not systems more trust in systems (theory X)
no measuring/directing directing/measuring everything incl. people; links to salary
system
freedom encapsulated in rules, freedom within boundaries
conflict avoidance (links to respect for continued conflict avoidance
the individual)
unorganized anarchy necessary managing. Organized anarchy (people do not do
quite as they like). Future organization: Systematized
nerworks?
25
The crisis forced Novo Nordisk to change values from the values of a
'professional' organization ('we know best', 'we do not need any
procedures', 'we can improvise and make anything happen') towards
the values of a 'machine' organization, building on systems, proced-
ures, rules, control, and measurement (Mintzberg, 1989). At the same
time there was a shift from values of 'theoryY' to values of 'theory X',
from total freedom and respect for the individual to freedom within
boundaries and seeing the individual as an economical resource or
even expense that could be controlled and measured.
Regarding the question of which trends the group foresaw as going to
influence the development of future values, the following were men-
tioned: acquisitions/mergers, satellites, partnerships, globalisation/
Internet/changed doctor-patient relation, experience/perceived value
(the Diabetes Care concept), decreasing prices on drugs (local author-
ities), individualization.
Regarding the desired future for Novo Nordisk, the group built its vision
on the following trends: Satellites (partnerships, niches, specialist
areas), environments that enhance cultural diversity and people who
can manage it, innovative products, encouraging difference, individu-
alization. Individualization was not understood as putting the individ-
ual above the group, but rather as individual development as part of
the whole.
Out of all the listed values the group selected the values that they felt
were essential for the organization to remain a market leader and a
challenging place to work. These were:
1) ambitions
2) flexibility/readiness to change
3) cross-disciplinary cooperation with respect, trust and skilled com-
munication
4) freedom, safety, power to act, risk willingness, creativity, and Im-
provisation.
Of these values, the first two have influenced the organization for
many years and still do. Flexibility had, however, changed from being
25
1
an internal capability towards being something that was demanded
externally. The last category of values (4) had been part of the prior
organization, but had been lost in the recent machine bureaucracy.
These values were needed and should be reinvented in order for the
organization to have creative power for action and for learning. The
third category was partly new and was added because cross-discipli-
nary cooperation would be essential in the future for managing cul-
tural diversity, creating innovative products and working in satellites
and with partnerships.
Analysis of Novo Nordisk culture
Summarizing the data from the two base-line studies and the culture
workshop Novo Nordisk's culture can best be understood as a 'pro-
fessional' configuration that has been forced into a 'machine' form by
necessity (the American Food & Drug Administration crisis). Accord-
ing to Mintzberg (1989) a professional organization is characterized
by autonomy, proficiency and ambition along with arrogance and an
anarchistic attitude. The 'machine' configuration or bureaucracy is,
however, prevalent in all the systems, evaluation boards, Key Perform-
ance Indicators, measurements, Standard Operating Procedures, con-
trols, etc. The result is an aversion to or fatigue in the face of new
systems, no matter whether these are good or bad.
According to Mintzberg, the machine form is most apt for a stable
environment, which is hardly the case in the pharmaceutical industry
or in industrial enzymes. The environment is more one of uncertain-
ty, ambiguity and flux. Together with all the internal changes of the
last few years resulting in a large amount of reorganizations and in
some necessary downsizing, this has made many employees insecure
and has made them cling to old routines and not want change or
innovation. People prefer to work with what is known and certain
instead of inquiring into new areas and experimenting with risk and
uncertainty. This attitude was seen in both base-line studies, but most
apparent in the Values in Action project, which concerned the creation
of new knowledge. Putnam (1986: 181) mentions the concept 'trained
incapacity', which concerns the well-known psychological feature of
falling back on past work habits when experiencing conflict or when
exposed to a high degree of uncertainty and ambiguity. A quote from
case study 3 compares 'before', where you could make mistakes if you
could explain how and why, with 'now':
"N0, no, no we didn't need to go and argue for it, as long as your
argument holds, but that means that you can go out and make a
mistake and say I did it because of that and that and that, and the
arguments were logical and you showed that you had actually
thought it through ... then you don't get blamed for making a
mistake. Now most people tend to argue before they even do it in
order to get permission. People are more scared, more insecure"
As in most middle-sized and large organizations, this means that many
forces are at play and many tensions exist. Forces of cooperation ver-
sus forces of competition, individualistic forces versus collective
endeavors. These forces become apparent in the organization through
networks. Novo Nordisk is a network-based organization ("you can't
survive without them"). The legitimate networks consist of project
groups, boards and committees, and task forces who have mandates to
carry out. But underneath is a large informal 'shadow network'
(Stacey, 1996) of cooperation and competition, of alliances and pol-
itics. At times novel ideas or approaches emerge from the shadow net-
works and become part of the legitimate networks, as described by
Stacey. At times, people are maneuvered in or out of positions through
alliances and politics. After the demerger was announced (in 1999),
negotiations and alliances were formed in the 'shadow networks', even
though the official message was not to negotiate and that everyone
would have jobs after the demerger. This is not said to criticize Novo
Nordisk. Politics appears to be (an inevitable) part of large organiza-
tions.
Thus, there is no uniform picture of the Novo Nordisk culture, but
a dynamic mosaic of interacting forces. The main impression is that
the company is more in the mode of 'doing things right' (efficiency) -
than doing the 'right things' (effectiveness). Often when people sum
up their percentages of allocation on different projects, the total adds
253
,ill a hundred percent, which means that they will need
,a time if they are to fulfil their tasks. This does not leave
,ce for creativity, experimentation and innovation. 19
.l'eneral work practice
Regarding a general work practice there are some differences regard-
ing professional subcultures that can most easily be seen as a differ-
ence between Health Care and Enzyme Business. The organization is
a dynamic mosaic of subcultures: The academic subculture is charac-
terized by 'competitive debates' and competitive attitudes of hoarding
knowledge or ideas in order to get credit for it (publish or take pat-
ents). The Engineering culture is very result oriented and technology
fixated, and prefers 'solution cycles' (i.e. finding solutions before ana-
lyzing the problem), sometimes resulting inType III errors: solving the
wrong problem (Mitroff & Featheringham, 1974).
I have asked many people if there is a difference between Health
Care and Enzyme Business, and all have confirmed that this is so. In
case study 3, the following was said: "There is a big difference between
Health Care and Enzymes, because in enzymes it is the results that
count no matter how you got them. In Health Care you have to follow
the rules and do such and such, so it matters how you get the results."
This difference could be related to the difference in product develop-
ment time. Enzyme Business has a development cycle of 3-5 years,
whereas pharmaceutical cycles take 10-12 years. People working in
Enzyme Business are generally said to be more informal, more open,
more market oriented (though far from enough), whereas people
working in Health Care are generally more reserved, formal and not
market oriented. There are more engineers in Enzyme Business and
more academics (e.g. medical doctors) in Health Care.
One of the participants from case study 3 expressed the difference
this way:
"At the University you learn how to go by the rules, this is a typ-
ical example of a Ph.D. thing ... but this is not the way of real
work, you go for the solution and afterwards you define things."
254
The Ph.D. thing was, of course, aimed at me, but the solution cycle
was probably a rather accurate way of describing the general practice
of Enzyme Business in particular. But if judging from the base-line
studies, which involved people from all over the organization, the
description illustrated the work practice of the Novo Nordisk organ-
ization in general. Fear of failure, lack of experimentation and bureau-
cracy are stifling for innovative activities, aspects which have been
described in the case studies, in informal discussions and in 10 inter-
nal reports
20
on innovation. With the demerger, however, both Health
Care and Enzyme Business have announced that more focus and
effort will be aimed at innovation. The findings and recommendations
in this book point out new ways and directions.
Innovation Coach
From October 1999 to December 2000 I worked as innovation coach
in Enzyme Business. Since I started many people have asked me what
it means to be an innovation coach. Coaching is a new supportive
leadership style. A coach asks questions that make people reflect on
their work, on their attitudes and on the interaction that goes on inthe
organization. As innovation coach, my job was basically to initiate and
support innovation - and in particular projects related to new business
development.
At first I became part of an international group working with the
New Idea Database and the mediator set-up that had been moved
from Research & Development to Enzyme Business Operations. At
that time most ideas came from Research & Development in Head-
quarters, Denmark. Together we developed a pre-screening facility for
the ideas that came into the idea database: a 'Quick and Creative Op-
portunity Assessment' . The purpose was to make a quick and intensive
evaluation of the ideas and find the key issues that had to be investi-
gated before allocating resources for further examination. We invited
scientists from all over the organization to take part in these sessions,
and worked in parallel in mixed teams for two days in a row. At the end
of the second day we all met with a team of managers to share our
255
results and to give recommendations for each idea - whether it should
be trashed or continued. These sessions were good in several respects.
A lot of work was achieved in a short time. The participants were intro-
duced to new methods and tools for idea generation and idea evalua-
tion. The energy was high because we worked in parallel, and good
networks were created between marketing people and researchers.
Enzyme Business has subsidiaries on all the major continents, and it
became evident that, if new business were to develop according to the
strategies and goals, a better understanding of the necessity for new
product development and stronger links to the subsidiaries were
needed. Consequently in 2000 we arranged creativity and innovation
sessions in North and South America, Europe, and in the Far East
Gapan, China and Malaysia). Before travelling, however, I developed
a 'Creativity and Innovation Toolbox' to bring as a hand-out. This was
a small manual with methods, frameworks and tools based on the
findings from my research (chapter 9).
Each session was developed according to the specific needs of the
regions, and in each country we worked on their own concrete local
ideas. It was illuminating to work with people in their proper sur-
roundings and to learn about their problems and desires. I learnt a lot
about how people in the subsidiaries saw the Danish 'Headquarters',
and what made them want to work for Novo Nordisk. The workshop
participants also learnt a lot and were satisfied with the toolbox and
with the concrete results they had achieved, which they continued to
work on. There were, of course, some interesting cultural differences,
e.g. in relation to how free and wild people would get during idea gen-
eration, but all the methods worked and people generally found the
sessions worthwhile and constructive.
As innovation coach I also planned and conducted the 'Innovation
Cafes' with a partner I had met earlier from Research & Development.
The idea emerged from my literature studies on creativity and innova-
tion, where I had seen examples of innovation happening as a result of
crossing professions, technologies and sectors that would normally
not be linked in any way. I thought that organizations needed provo-
cation from different and preferably totally 'irrelevant' industries
because this could spark or inspire novel combinations.
25
6
We presented the idea for top management who agreed that it
would be worth trying and each manager agreed to host an Innovation
Cafe. The first Cafe was with a very experienced producer of children's
television. His story was a tale about how he had had to innovate in
order to produce an entire month of children's television - on a low
budget and with a mixed group of staff that was too small. He illustrat-
ed his story with vivid video clips from the takes. Afterwards we had
Cafe discussions sitting around tables using some creativity methods
for idea generation and development. A lot of interesting conversations
took place, ideas were generated, and new relations were formed. In
my opinion it was an important innovation culture-building event.
We continued with three other Innovation Cafes. The next was on
the theme 'a swinging organization' and here we had a 9-person swing
band entertain us about management, teamwork and cross-organiza-
tional cooperation. All of this was illustrated by the musicians and it
was a hilarious and energetic session, where everybody was snapping
their fingers to the music.
The third was on storytelling. Here we were told an Irish legend,
which we interpreted and related to our own organization. We got into
some very deep conversations about innovation and what values are
needed for an organization to be innovative.
The fourth was on entrepreneurship. Here we had invited the CEO
from a very successful entrepreneurial company. We discussed the
qualities of entrepreneurship and the special talents, energy and per-
severance it takes to work with high stakes and high risk.
Each Innovation Cafe was special and distinct from the others.
They all, however, contributed to the creation of an innovation culture
and stimulated a continuous focus on innovation and entrepreneur-
ship. The evaluation from the participants after four Cafes was that In-
novation Cafes were worthwhile and should be continued.
On November 14, 2000, the planned demerger took place. This
was the start of a new Novo Nordisk NS and Novozymes NS. I
enjoyed being part of the celebrations and felt that the circle had been
concluded. It was the right time for me to pass on to new challenges
in a new job.
257
Concluding remarks
This chapter was based on experience from praxis with stories of what
happened and how it felt. It included some of the difficulties, problems,
struggles, deliberations and cultural clashes that are inevitable in real-
life situations where things do not always proceed as planned. We pro-
vided the flesh and bone of the case studies and brought them to life.
A workshop on cultural values provided valuable insight into the cul-
ture and the general work practice of the organization. The analysis
did not result in a homogeneous picture, but rather in a heterogeneous
multifaceted picture of forces pulling in different directions. Stacey's
(1996) concepts of 'legitimate' and 'shadow' networks provided a re-
alistic picture of Novo Nordisk, including both mechanistic and an-
archistic features. The general work practice involved 'academic' as
well as 'engineering' subcultures, and demonstrated a very result-
oriented approach to problems, which is alarming if it results in solv-
ing the wrong problem.
I will conclude the chapter and narrative with a brief reflection on
fieldwork. Before I started on my fieldwork, one of my professors said:
"Remember that you are delivered to the field on its terms".21 It made
sense to me then, but in retrospect this sentence reflects the essence
of my fieldwork.
Mter this narrative we move on to analysis. In the next chapters (6 and
7) we will develop construct validity and start pattern matching and
analysis
Notes
1 Quotation from an overhead from a Ph.D. course at the Institute of Manage-
ment, Politics and Philospphy, Copenhagen Business School, in Oct. 1997
2 Ugebrevet Mandag Morgen, 20.02.95: 'Globale koncerner vii bruge dansk
iva:rksa:ttermodel' (Global companies want to apply Danish Entrepreneurial
model)
3 An Industrial Ph.D. is a co-operation between an organization and a univer-
sitylbusiness school, partly sponsored by ATV (the Academy of the Technichal
Sciences)
4 Kubus, as described in Herlau, 1995, see also in chapter 7: Developing con-
struct validity
5 E.g. from written evaluation: "It focuses and speeds up the difficult job of
evaluating new research efforts and it facilitates greatly the interactive-infor-
mation gathering process needed"
6 E.g. from written evaluation: "It may seem somewhat rigid"
7 E.g. the Success/Failure Visualization (see Appendix C) and the Barrier exer-
cise (see chapter 9)
8 Argyris & Schon, 1996
9 From 010998 to 011099 776 organizational changes were registered in Novo
Nordisk (i.e. change of structure, or name of unit/area, or management
change)
10 From 1996 climate surveys were obligatory.
11 Hvid0fe used to be a diabetes hospital, but is presently used as an internation-
al course mansion and small hotel, owned by the company. The reader may
recognize the setting from the beginning of this book. The story from chapter
1 builds on data from case study 3.
12 See Appendix C
13 People responsible for the five business process projects started by Ensyme
Business management were called flow makers, probably in order to indicate
the cross-organizational aspect of these processes.
14 See Appendix D
15 A technique called 'Analogue with Nature', which I had picked up at a course
in creativity in Brussels, 1996
16 Of the many ideas 16 were considered feasible and original
17 This was written in February 1999. It is a good example of things in a state
of openness. Six months later things looked different - and better. The medi-
ator group was merged with the New Business Unit from Enzyme Business
Marketing and Sales, and thereby got the market information and the
resources they needed.
18 Schein (1994:121-138)
19 Data from 10 CIMI (Copenhagen International Management Institute)
reports on Entrepreneurial behavior in Novo Nordisk, 1997
20 Data from 10 CIMI (Copenhagen International Management Institute)
reports on Entrepreneurial behavior in Novo Nordisk, 1997
21 Erik Maaloe said this on a Ph.D. course in Oct. 1997: "Man er udleveret til
feltet pa dets betingelser"
259

Darsoe 2001 Innovation-Making

Uploaded by

Copyright:

Available Formats

Darsoe 2001 Innovation-Making

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Darsoe 2001 Innovation-Making

Uploaded by

Copyright:

Available Formats

Lotte Dars0

Innovation in the Making

Six steps in human learning:an illustration of the Dreyfus &

You might also like