Int. J. Business Information Systems, Vol. 1, Nos. 1/2, 2005 83

Copyright © 2005 Inderscience Enterprises Ltd.

Systemic knowledge and the V-model

Jim Sheffield Department of Information Systems and Operations Management University of Auckland, Private Bag, 92019 Auckland, New Zealand Fax: (64–9) 3737 430 E-mail: j.sheffield@auckland.ac.nz

Abstract: A simple but surprisingly useful system of inquiry is outlined and applied to elicit and validate systemic knowledge. Systemic knowledge is the holistic understanding of interpersonal expectations or norms, the technical system, and the relationships between the two. General systems concepts such as hierarchy and intentionality are employed to generate a new V-model that incorporates the familiar concepts of ‘top down’ design and ‘bottom up’ implementation. The V-model validates systemic knowledge by testing the strength of a chain of evidence that emerges from the application of design principles to a practical knowledge management problem.

Keywords: action science; inquiring systems; knowledge management; multiple perspectives; organisational learning; problem structuring; sensemaking; systemic knowledge; V-model.

Reference to this paper should be made as follows: Sheffield, J. (2005) ‘Systemic knowledge and the V-model’, Int. J. Business Information Systems, Vol. 1, Nos. 1/2, pp.83–101.

Biographical notes: Dr. Jim Sheffield graduated with a PhD from the University of Arizona in 1990. His doctoral work in group support systems led to the creation of a Decision Support Centre at the University of Auckland and participation in major policy development exercises. Action research initiatives include the development of economic strategy, science policy, and comprehensive regional planning. He has published in Group Decision and Negotiation and Journal of MIS. His current interests include The Habermasian Inquiring System, knowledge management, collaborative learning, and research methods.

1 Introduction

Business information systems employ technology in a setting that is problematic. Factors like leadership style, culture, organisational structure, and power relationships are typically beyond the control of the information systems group yet may strongly influence success. Methods of inquiry are therefore required to understand or ‘structure’ systems issues, decision situations or problems, rather than to ‘solve’ them.

Systems professionals rely heavily on ‘models’ to convey notions of systems structure and function. There are a profusion of models, modelling techniques, and approaches to modelling, each serving a particular purpose within a particular area of application. Typically these models provide rigorous guidelines for the more detailed and technical aspects of systems development and implementation. It is these same

84 J. Sheffield

characteristics that make them less useful when individuals from different functional specialties (e.g., senior management, users, and IT personnel) must develop a mutual understanding of the broader aspects of a system.

Systemic knowledge is the holistic understanding of interpersonal expectations or norms, the technical system, and the relationships between the two. Models for eliciting and validating systemic knowledge have many applications. Information Systems applications include joint application development, project management, quality assurance, requirements gathering, reviews, systems analysis, and design. Other applications include knowledge management, organisational learning, programme review and evaluation, research processes and structures, and strategic planning.

A user-friendly model that offers robust guidance in a multitude of situations would appear to be most useful. Such a tool would need to incorporate universal concepts from general systems theory such as hierarchy and intentionality in a way that supports the development of business systems and technical IT issues. The model should be equally acceptable to systems personnel and those from other business disciplines, such as management and marketing. Yet current systems approaches such as Visio and the Universal Modelling Language (UML) are relatively narrow and technical – they do not directly support broad discussions of business and organisational issues.

A truly user-friendly model should be universal. It should capture all elements known to be crucial to success, whether or not these relate to personal, organisational, or technical aspects. Many authors argue that these issues are so intertwined that systems failure is assured unless personal and interpersonal issues are developed in conjunction with technical issues (Checkland and Holwell, 1998; Fahey and Prusak, 1998; Ulfelder, 2001). Stakeholders expectations emerge from dialogues by humans who collectively possess a plurality of perspectives and an awareness of the richness of possibilities. A model that is useful in a wide variety of situations must support both the interpersonal organising process and the development of a technical artefact.

The V-model assists individuals or groups in learning about and remembering system components and the whole system. It consists of a small number of systems concepts (hierarchy, intentionality, chain of evidence, equivocality, and uncertainty) made memorable by organising them around a V (for Validity) graphic. The vertical axis of the V measures hierarchy and the horizontal axis measures intentionality. The process steps (which may relate to people, process, product, and technology) are placed on the V (i.e., both vertically and horizontally) so that key dependencies among the components and the system as a whole can be appreciated.

An action science approach is adopted. In Section 2 selected literature on systemic knowledge is reviewed, and the theory is combined with practice to form an initial model. The concepts of hierarchy and intentionality are described in Section 3 and Section 4, respectively. In Section 5, these concepts are combined with additional concepts to complete the derivation of the V-model. The conclusion in Section 6 reveals that the paper itself constitutes a system, the structure of which may be investigated via the V-model.

Systemic knowledge and the V-model 85

2 Systemic knowledge

The author adopts an action science approach. As a reflective practitioner (Argyris, 1999) he seeks to describe a process for eliciting and validating systemic knowledge in a way that directly supports practice. The V-model is a theory grounded in extensive practice in various problem domains. This paper has been developed over multiple cycles of literature review, practice, and theory building. In Section 2, these elements are reviewed to create a foundation for the subsequent development of the V-model.

2.1 Literature review

The literature review is conducted to identify the work of other researchers and build on it in a cumulative tradition. The review starts with dictionary definitions of ‘system’ then amplifies selected themes via a review of the academic literature.

2.1.1 Dictionary definitions

The word ‘system’ has many applications. A search in standard dictionaries reveals two clusters of definitions that indicate the application of general systems concepts such as those used in the V-model.

The first cluster of definitions emphasises less generic concepts. One element is an organised society or social situation regarded as stultifying. This definition suggests that the system may be too structured, perhaps because it has lost legitimacy and produces outcomes deemed inappropriate, and therefore in need of radical revision (Burrell and Morgan, 1979). This suggests that when qualitatively different elements of systemic knowledge are arranged in a hierarchy, personal choice, like freedom in a democracy, is paramount. Successful systems need the informed acceptance of stakeholders.

The second cluster of definitions emphasises more generic concepts. Typically a system was defined as ‘an organised integrated whole made up of diverse but interrelated and interdependent parts.’ Elements in this cluster include concepts at two levels:

1 Upper level: interpersonal expectations or norms

Method, orderliness, plan, procedure, proceeding, process, fashion, manner, mode, modus, technique, way, wise. These concepts are more inclusive and less structured (Gorry et al., 1971; 1989) than…

2 Lower level: technical system

System object or artefact (complex, whole, aggregation, array; mesh, network; arrangement, scheme, setup; order, pattern, entity, integral, integrate, sum, totality). These concepts are less inclusive and more structured.

2.1.2 A review of the academic literature

Systemic knowledge has been defined as the holistic understanding of the interpersonal expectations or norms, the technical system, and the relationships between the two. Action science researchers with this focus have the goal of facilitating learning and improving practice (Morton et al., 2003). They develop holistic models containing qualitatively different elements for the express purpose of assisting actors to learn about

86 J. Sheffield

the system(s) created by their (inter)actions. Systemic knowledge is complex because understanding is diffuse and incomplete, and there is conflict over the allocation of scarce resources. Understanding one’s own mental processes and articulating personal understanding (mental models) are important aspects of systemic knowledge. Managing complexity takes the form of appreciating multiple perspectives and mutual influence.

Research in this tradition includes:

• action science (Argyris, 1999)

• general systems theory (Jackson, 2000)

• knowledge management (Davenport and Prusak, 1998; Garvin, 1998; 2000; Igel and Numprasertchai, 2004; Johannessen et al., 2005; Nonaka and Takeuchi, 1995)

• multiple perspectives (Habermas, 1984; 1987; Mingers, 2001a–b; Sheffield, 2004; 2005)

• organisational learning (Garvin, 1998; 2000; Senge, 1990; 1994)

• problem structuring and operations research in the UK tradition (Ackermann and Eden, 2001; Checkland and Holwell, 1998; DeBono, 1999; Jackson, 1982; Rosenhead and Mingers, 2001)

• sensemaking (Weick, 1979; 1995; 2001).

In each of these traditions the concept of hierarchy plays an important role. Each of these traditions employs paired concepts organised in a two-level hierarchy:

1 Upper level: More inclusive (i.e., more subjective, interpersonal, and humanistic, as in the arts). Less structured.

2 Lower level: Less inclusive (i.e., more objective and technical, as in science and technology, including information technology). More structured.

Table 1 summarises the relevant literature by providing one or more examples of hierarchically paired terms in each of the research traditions identified above. A representative and recent reference is provided for each.

Systemic knowledge and the V-model 87

Table 1 Systemic knowledge and hierarchically paired terms in relevant research traditions

Sens

emak

ing

Equ

ivoc

alit

y

Lac

k of

per

sona

l and

in

terp

erso

nal k

now

ledg

e of

the

‘rig

ht p

robl

em’

Unc

erta

inty

Lac

k of

tech

nica

l kn

owle

dge

and

othe

r re

sour

ces

to im

plem

ent

the

‘rig

ht s

olut

ion’

Wei

ck (

2001

)

Pro

blem

st

ruct

urin

g

1. U

nder

stan

ding

the

prob

lem

in c

onte

xt

2. S

oft s

yste

ms

3. S

oft O

R

1. D

evel

opin

g a

spe

cifi

c so

lutio

n

2. H

ard

syst

ems

3. H

ard

OR

Ros

enhe

ad a

nd

Min

gers

(20

01)

Org

anis

atio

nal

lear

ning

Acq

uiri

ng

know

ledg

e

App

lyin

g kn

owle

dge

Gar

vin

(200

0)

Mul

tipl

e

pers

pect

ives

Lif

e w

orld

Rel

ativ

ely

mor

e w

eigh

t atta

ched

to

pers

onal

, tha

n in

terp

erso

nal o

r te

chni

cal i

ssue

s

Syst

em w

orld

Rel

ativ

ely

mor

e w

eigh

t atta

ched

to

tech

nica

l, th

an

inte

rper

sona

l or

pers

onal

issu

es

Hab

erm

as (

1987

)

Kno

wle

dge

man

agem

ent

1. H

olis

tic

k

now

ledg

e

2. T

acit

kno

wle

dge

3. K

now

how

1. S

peci

alis

t

kno

wle

dge

2. E

xplic

it

k

now

ledg

e

3. K

now

wha

t

Joha

nnes

sen

et

al.

(200

5)

Gen

eral

sys

tem

s th

eory

1. O

rgan

ic

2. S

ynth

etic

3. U

nity

4. S

yste

mic

5. H

olis

tic

6. G

ener

al

1. M

echa

nist

ic

2. A

naly

tic

3. R

educ

tion

ism

4. ‘

Div

ide

and

con

quer

’

5. D

etai

led

6. S

peci

fic

Jack

son

(200

0)

Act

ion

scie

nce

Dou

ble-

loop

le

arni

ng

Sing

le-l

oop

lear

ning

Arg

yris

(19

99)

Res

earc

htr

adit

ion

Lev

el in

hie

rarc

hy

Upp

er le

vel

M

ore

incl

usiv

e

(i

.e.,

mor

e su

bjec

tive,

inte

rper

sona

l and

hum

anis

tic, a

s in

the

arts

). L

ess

stru

ctur

ed.

Low

er le

vel

L

ess

incl

usiv

e (i

.e.,

mor

e ob

ject

ive

and

tech

nica

l, as

in s

cien

ce

an

d te

chno

logy

,

incl

udin

g IT

). M

ore

stru

ctur

ed.

Rep

rese

ntat

ive

refe

renc

e

88 J. Sheffield

2.2 Combining theory and practice in an initial model

In the remainder of Section 2, theory and practice are combined into an initial model (Figures 1 and 2).

Figure 1 Eliciting systemic knowledge as a journey of discovery

Figure 2 Validating systematic knowledge from multiple perspectives

Voyage of discovery: six questionsQ1.Why embark on this journey? Who is the idea champion? Who is the sponsor? Who are the other stakeholders? What are each individual’s expectations?

Q2. What sort of journey and destination are envisaged? What type of vessel will support our aspirations?

Q3. How do we plan to get there?How sound is our vessel?How will we navigate?How can one know?

Q4. How is it going? Are we there yet? How sound are our sailing skills? Where are we right now? How can one know?

Q5. What can we say about the destination we reached? In what worlds do we find ourselves?

Q6. Why do I feel the journey was worthwhile? What if they say “So what”?Why is there a good story to tell? Why does each stakeholder care ?

Systemic knowledge and the V-model 89

The initial versions of the V-model emerged from multi-year action research in economic development with Professor Michael Porter of Harvard University. The project was titled Upgrading the Competitive Advantage of New Zealand. It involved the extensive use of Group Support Systems (Sheffield and Gallupe, 1994; 1995). Because the task was not well understood, and each participant had only a piece of the mystery, a model to elicit and validate systemic knowledge was required. The model served as a roadmap for the journey towards actionable objectives and implemented solutions.

In facilitating industry-based strategic planning groups the author quickly discovered that a focus on a priori or external structures in the form of relevant content-oriented theory (Porter, 1990) was not sufficient. This is not a reflection on the theory, which had been synthesised from case studies in a large number of industries and was highly relevant to the topic of industry-based competitiveness. Yet each industry is unique in many respects. Stakeholders need a model of systemic knowledge to guide the elicitation and validation of interpersonal expectations or norms, the technical system, and the relationships between the two within their own industry. This requirement is captured in the initial organising model described below.

To avoid disenfranchising the views of the stakeholders, the process of reducing equivocality about intentions (Weick, 2001) and surfacing tacit knowledge (Nonaka and Takeuchi, 1995) proceeds from the holistic (Garvin, 1998,p.55; Johanessen et al., 2005) and general (upper level) to the specific (lower level) (Jackson, 2000). Unfortunately communication processes may stall because the initial action plan is not resilient enough to withstand industry politics. The interaction may become a talkfest devoid of feasible project plans. To avoid this eventuality uncertainty must be reduced (Weick, 2001) about explicit candidate action plans by combining their elements (Nonaka and Takeuchi, 1995) into a more integrated and robust project. This process proceeds from the specific (lower level) to the general (upper level) (Garvin, 1998,p.55; Jackson, 2000).

Communication among industry representatives promotes interorganisational learning (Garvin, 2000), readiness for mutual problem solving (trust), and a more nuanced or double-loop feel for when to compete and when to cooperate (Argyris, 1999). The process of learning the soft or humanistic system (Checkland and Holwell, 1998) in which each participant is embedded takes the form of a journey of discovery (Ackermann and Eden, 2001) with six process steps (Garvin 1998,p.55) arranged in a three-level hierarchy (Figure 1). Multiple perspectives on knowledge, each with qualitatively different phenomena and criteria for success, are connected to this three-level hierarchy. The freedom associated with individual choice is paramount and is placed at the uppermost level. The success of the journey is validated by personal commitment to an interpersonal agreement for technical excellence (Figure 2). In summary, theory and practice have been combined into an initial model that has a distinctive V-shape. The initial model (Figures 1 and 2) has subsequently been validated via application in the traditions identified above.

3 Hierarchy

The vertical axis of the V-model indicates levels in a hierarchy. As indicated in Section 2, upper levels are more inclusive and less structured; lower levels are less inclusive and more structured. The upper two points of the V represent phenomena that include ‘personal visions’ of the ‘big picture’; a holistic or ‘helicopter view’ of the business

90 J. Sheffield

application system; and a high level of abstraction in dealing with system components. The lowest point of the V represents more structured/less inclusive phenomena (i.e., the details or ‘nuts and bolts’).

The theory developed in Section 2.2 produced a hierarchy of qualitatively dissimilar phenomena that bears some resemblance to Maslow’s Hierarchy of Needs. With the Section 2.2 hierarchy in mind, it may be worthwhile to consider a concept normally attributed to hierarchies, that of abstraction. A review of dictionaries identified that the adjective ‘abstract’:

• Denotes (depending on the context one or more of the following): disassociated from any specific instance, insufficiently factual, expressing a quality apart from an object, having conceptual rather than concrete existence, hypothetical, ideal, theoretical, transcendent, transcendental, academic, impractical, utopian, visionary; speculative, undemonstrable, conceptual, notional.

• Contrasts with corporeal, material, objective, phenomenal, physical, actual, factual, real.

• Is the opposite of concrete. To instantiate is to ‘represent (an abstraction) by a concrete instance’.

Many authors adopt a three-level hierarchy. Authors in the area of planning and control systems (Anthony, 1965; Anthony and Govindarajan, 2004) describe three levels:

1 strategic planning

2 management control and tactical planning

3 operational planning and control.

Note that elements of a ‘big picture’ or ‘total system’ view exist to a considerable degree in the hearts and minds of the individual humans who construct (effect) or are affected by the system. Each person has a different perspective and therefore sees a different vision. Each sees system possibilities through the lens of a particular set of interests and values. While each individual may have strong feelings about the system these feelings have to be articulated before the structure of the whole system may be determined. The big picture view is ‘loose’ or ‘fuzzy’ in that it does not specify the common framework that emerges when these perspectives ‘come together.’ In summary, the big picture view from the top of the V is unstructured.

Only after individual views are articulated, and individuals are willing and able to accommodate at least some of the perspectives of others, may consensus on a common vision emerge. Common objectives are articulated and appreciated within a framework of communication and mutual understanding (Checkland and Holwell, 1998). The acronym SMART (Specific, Measurable, Actionable, Relevant and Timed) captures the characteristics of objectives that facilitate implementation. Yet ongoing environmental influences, changing priorities, and implementation issues cause objectives to be reviewed and changed. Too much analysis in such situations will lead to paralysis. Consequently the mid-range systems view from the middle of the V is semistructured.

Implementation requires all of the conceptual ‘nuts’ and ‘bolts’ to be visualised in great detail, then realised and connected into a fine-grained and stable pattern. The view from the bottom of the V is structured.

Two worked examples and two exercises on hierarchy and the V-model follow (see Appendix).

Systemic knowledge and the V-model 91

4 Intentionality

The horizontal axis measures the links among intentions and outcomes. The left half of the V-model contains process steps associated with intentions (e.g., identifying user requirements, analysis, and design). The right half of the V-model contains process steps associated with outcomes (e.g., tangible outcomes or deliverables). Lines and/or arrows are drawn between steps to identify how the process steps depend on each other. There are three common types of dependencies:

1 Time order of events or start-stop precedence of qualitatively different activities. These are shown as directed lines down the left side and up the right side of the V-model. The details of the time structure of the system depends on the way the overall project is framed and the more strategic aspects of the development process.

2 Logical cause and effect between the planning and the implementation steps for the same type of activity (shown as a horizontal line at the same level in the hierarchy).

3 Iteration. In project management approaches such as prototyping and continuous improvement intentions and outcomes are linked in a cycle of repeated activities.

The application of the V-model concept of intentionality to quality assurance of corporate information systems is illustrated by the horizontal lines in Figure 3. Note that these horizontal lines occur at three different levels in the hierarchy (business modelling, system modelling, and module specification).

Figure 3 Application of the V-model to corporate IT

92 J. Sheffield

Note also that the quality of the left-to-right linkages varies as indicated in Figure 2:

• Closing the gap between vision and implementation review is relatively unstructured and relies heavily on personal commitment of those involved. This commitment cannot be ‘outsourced’.

• Closing the gap between business analysis and user acceptance of the implemented system is semi-structured and relies heavily on an interpersonal agreement to provide the services in the manner ‘specified’ (by the agreed contract).

• Closing the gap between module specifications and completed system components relies heavily on technical excellence. This knowledge can be outsourced.

4.1 Linking hierarchy and intentionality into a chain of evidence

Figure 4 illustrates an incomplete V-model. Only the concepts developed so far are employed to link hierarchy and intentionality into a chain of evidence. The bold horizontal and diagonal lines represent the chain of evidence that link the six types of activities identified in the V-model.

Figure 4 An incomplete V-model

The chain of evidence is the logic underlying the arrangement of the steps in the process. Each link of the chain provides a stepping-stone to move from intention to outcome in a sensible or coherent manner. Systems may be viewed from three complementary perspectives:

6. THE PAYOFFWhy did pursuingthis idea add value?

4. THE PLAN IN ACTION

How strong is the evidence that we have met each

sub-objective?

3. THE ACTION PLAN

How will we meet each sub-objective?

2. THE OBJECTIVEWhat is our (my? your?)

objectives and sub-objectives?

1. THE IDEAWhy will pursuingthis idea add value?

INTENTION OUTCOME

5. THE RESULTSWhat is the evidencethat we have met our

(my? your?) objectives?

WHAT?

WHY?

HOW?

5

6

3

2

4

1

Systemic knowledge and the V-model 93

1 Planning perspective, as in an investigation of prospects and the development of initiatives (looking right or forwards in time from the left of the V).

2 Review perspective, as in a retrospective account of the success of an initiative or a newly implemented system (looking left or backwards in time from the right of the V).

3 Audit perspective, as in a critical appreciation of the quality of procedures linking both planning and review (looking both forwards and backwards).

The V-model supports planning, review and both of these (audit). It recognises that unstructured and structured phenomena are very different and that models of cause and effect must start with the intentions of individuals and the way they make sense of their world (Weick, 1979; 1995; 2001). It is intended as a conceptual framework or scaffold that links theory and action (Argyris, 1999; Senge, 1994) in a pragmatic and robust manner. It should not to be applied in a mechanistic manner that is insensitive to the situation at hand.

For example, the movement from the ‘big picture’ to the ‘nuts and bolts’ is a continuous series of small steps. These small steps may, for convenience, be packaged into three levels. On occasion two or four levels may be preferable. There is nothing magical about how many steps are shown on the V, nor where one step ends and the next starts.

Table 2 Systemic knowledge and some terms for three-level hierarchies in selected domains

Business function

Level in hierarchy

General systems

Developing IT systems

intentions

Developing IT systems outcomes

Developing research intentions

Developing research outcomes

Planning and control

Upper level Approach Idea (IT system concept)

Payoff (IT system review)

Idea (introduction)

Payoff (discussion and conclusion)

Strategic

Middle level Framing Objectives of IT system (business requirements)

Results of IT system (business application)

Objectives of research (literature review)

Results of research (analysis)

Tactical

Lower level Decomposition IT system design

IT software development and testing

Action plan (methodology)

Plan in action (data gathered)

Operational

The V-model is a generic metamodel that imposes a minimum number of distinctions or meta-principles. It champions pragmatism (Menand, 1997; 2001) and breadth of application across problem situations rather than prescriptive accuracy within any one domain. It does not deal directly with the iterative and incremental approaches associated with improving process quality, nor the richly interconnected and adaptive aspects of complex systems (Cilliers, 1998), nor the quantitative formulation of the cyclical aspects of dynamic systems (Maani and Cavana, 2000). Rather it provides an organising model for systemic knowledge (Johanessen et al., 2005) sensitive to personal, interpersonal, and technical concerns (Habermas, 1984; 1987; Ropohl, 1999). The V-model guides progress by eliciting and validating the assumptions underlying the need for a more specialised approach (Table 2).

94 J. Sheffield

The purpose of the V-model can now be summarised as follows – To validate systemic knowledge by testing the coherence among intentions and outcomes at three levels of inclusiveness.

5 Result: the complete V-model

Previous sections have linked hierarchy (Section 3) and intentionality (Section 4) into a functional but incomplete V-model (Figure 4). However two additional concepts from the sensemaking literature – equivocality and uncertainty – may enrich understanding. The complete V-model is shown in Figure 5. As illustrated by the boxed text, the new concepts serve to differentiate the top half of the V from the bottom half. It is desirable to make this distinction as the activities and knowledge requirements for each half are very different. Inability to reconcile the more humanistic and managerial concepts at the top of the V with the more technical concepts at the bottom of the V is a major cause of system failure (Ulfelder, 2001; Fahey and Prusak, 1998).

Figure 5 The complete V-model

5.1 The top half of the ‘V’ reduces equivocality about tacit knowledge

The top half of the V describes aspects directly related to the fact that there are multiple stakeholders and multiple viewpoints. The knowledge and interests of stakeholders form the ingredients and recipes from which the system must emerge (Habermas, 1972). There is almost always more than one cook in the kitchen and more than one customer at the table. Each stakeholder experiences different perspectives and possibilities. Initially the quality of mutual understanding may be poor. The ‘system’ is merely the site of efforts to

6. THE PAYOFFWhy did pursuing

this idea add value?

2. THE OBJECTIVEWhat is our (my? your?) objectives

and sub-objectives?

1

3 4

2 5

6

5. THE RESULTSWhat is the evidencethat we have met our(my? your?) objectives?

4. THE PLAN IN ACTION

How strong is the evidence that we have met each

sub-objective?

1. THE IDEAWhy will pursuing this

idea add value?

3. THE ACTION PLAN

How will we meet each sub-objective?

OUTCOMESINTENTIONS

UNSTRUCTURED

SEMI-STRUCTURED

STRUCTURED

Reduce equivocality about themultiple ways in which the idea may

be approached, stabilised and framed

Reduce equivocality about contextin which the results are meaningful

Reduce uncertainty aboutthe multiple ways in which

the objectives can be instantiated

Reduce uncertainty aboutthe multiple ways in which the actions can be evaluated

Systemic knowledge and the V-model 95

stabilise personal expectations via exploration of mutual interdependencies (Weick, 1979,p.3). Each must interact with the others to know that they do not know each other’s perspectives or the system as a whole. Each is likely to be aware of the need to manage the conflict that results when multiple values and multiple objectives are opened for discussion. The discussion process must deal directly with issues of governance, the exercise of power, and the resolution of conflict. For each participant the world of desires and meaning has yet to be reconciled with the world of consensus for systematic action. No ‘right’ solution can be developed and validated because there is no mutual understanding of the ‘right’ problem (Table 1) (Rosenhead and Mingers, 2001).

The term equivocality (Daft and Lengel, 1986) is used to describe the ‘looseness’ or ‘fuzziness’ or ‘conceptual uncertainty’ or unstructured aspects of a system. Understanding of system elements and interdependencies (e.g., among people, technology, processes and outcomes) are equivocal because there are multiple perspectives on the way that they can, and should, fit together. The degree of fit is always less than perfect and always threatened by change. The communication to realign possibilities, to accommodate different perspectives, and to give birth to a common framework ‘reduces equivocality’. The need to reduce equivocality may be signalled by the presence of persons with strong but conflicting opinions, but who must work together. Without interpersonal learning and mutual accommodation, the problem in context is undefined. The system designer cannot claim privileged access to the ‘right’ problem. Claims that any specific technology is the solution are premature.

As illustrated in the boxed text in the top half of Figure 5:

• Movement from Step 1 (the idea) to Step 2 (the objective) requires those who have a stake in the system to reduce equivocality about the multiple ways in which the idea may be approached, stabilised, and framed.

• Movement from Step 5 (the results) to Step 6 (the payoff) requires those who have a stake in the system to reduce equivocality about the context in which the results are meaningful.

5.2 The bottom half of the ‘V’ reduces uncertainty about explicit knowledge

The bottom half of the V describes actions to implement a preexisting and explicit interpersonal agreement among those who have a stake in the system and the power to make things happen. The assumption is that the stakeholder’s desires and meanings have been appropriately reconciled with the world of consensus and instrumental action.

The term uncertainty is used to denote procedural or instrumental confusion, or lack of understanding about how to perform within a known frame of reference. Uncertainty relates to facts and ‘factual uncertainty’, and the technical skill required to harness them to develop the ‘right’ solution to a preexisting objective (Table 1). A team implementing a system specification (i.e., a statement of requirements that is ‘smart’ and otherwise unproblematic) ‘reduces uncertainty’.

As illustrated in the boxed text in the bottom half of Figure 5:

• Movement from Step 2 (the objective) to Step 3 (the action plan) requires those who have a stake in the system to reduce uncertainty about the multiple ways in which the objectives can be instantiated.

96 J. Sheffield

• Movement from Step 4 (the plan in action) to Step 5 (the results) requires those who have a stake in the system to reduce uncertainty about the multiple ways in which the actions can be evaluated.

Note that structure is not an end in itself. In Section 2, one definition of a system is ‘an organised society or interpersonal situation regarded as stultifying.’ Some stakeholders are likely to view the system as oppressive – something to be resisted. These stakeholders may seek freedom via radical change. They wish to destroy certain aspects of the current interpersonal and technical arrangements so as to make way for a new order (Burrell and Morgan, 1979).

6 Conclusion: summary of design principles

This paper has added considerable detail to the concepts sketched in the initial model. (Figures 1 and 2) The paper concludes with a summary of the previous sections. In keeping with the theme of action science, and the idea that the value of theory must be demonstrated in practice (Menand, 1997; 2001), this task will be accomplished with the aid of a V-model shown in Figure 6.

Figure 6 Applying the V-model to this paper

Section 1 (shown in the top left of Figure 6) articulates the idea motivating the V-model. This section describes the need for systems theories, concepts, and tools that aid in an understanding of intertwined technical, interpersonal, and personal aspects.

1

3 4

2 5

6

WHY?

WHAT?

HOW?

6. THE PAYOFFWhy did pursuing this idea add value?Conclusion: Summary of design principles

4. THE PLAN IN ACTIONHow strong is the evidence that we have met each sub-objectives?

Intentionality

1. THE IDEAWhy will pursuingthis idea add value?Introduction

2. THE OBJECTIVEWhat is our (my? your?) objectives and sub-objectives?

Systemic Knowledge

3. THE ACTION PLANHow will we meet each sub-objective?

Hierarchy

5. THE RESULTS What is the evidence that we have met our (my? your?) objective?

Result: The complete V-Model

Developing intentions Developing outcomes

Systemic knowledge and the V-model 97

Section 2 provides a foundation for understanding systemic knowledge. Following the action science approach, existing theory and personal practice are combined to produce an initial model. Eliciting systemic knowledge (including that, which emerges from one’s own actions) is likened to a journey of discovery. The success of the journey is validated by personal commitment to an interpersonal agreement for technical excellence. Theory and practice are combined to develop an initial model that has a distinctive V-shape (Figures 1 and 2).

Section 3 investigates the vertical axis of the V-model (hierarchy). Higher, more inclusive and abstract levels in the hierarchy are associated with low levels of structure, and vice versa. Three levels are described. The highest (and least constrained, analysable, observable or structured) level is linked to the possibilities and perspectives implicit in the feelings, thoughts and discretionary actions of individuals. A moderate level is linked to articulated interpersonal agreements for explicit systems objectives. The lowest (and most constrained, analysable, observable, or structured) level is linked to the technical ‘nuts and bolts’ of detailed designs and technical solutions.

Section 4 describes the horizontal axis of the V-model (intentionality). It illustrates the need for similar types of activities to be represented at the same level on the left (intentions) and the right (outcomes) side of the V-model. In the V-model horizontal and diagonal lines represent the chain of evidence linking six types of activities.

Section 5 illustrates that a system (Section 2) described in terms of hierarchy (Section 3) and intentionality (Section 4) has a distinctive V shape (Figures 3, 4, and 5). The V-model elicits and validates systemic knowledge by developing and testing the coherence among intentions and outcomes at three levels of inclusiveness. Activities at the top half of the V ‘reduce equivocality’ about the nature of the ‘right’ problem by instantiating and linking the upper and middle levels in the hierarchy. Activities in the bottom half ‘reduce uncertainty’ about the nature of the ‘right’ solution by instantiating and linking the middle and lower levels. The completed V-model is illustrated in Figure 5.

The six steps are organised into pairs, with one pair at each of the three levels of a hierarchy (Figure 4). Steps 1 and 6 describe why the V-model idea or concept will add (and is adding) value. Steps 2 and 5 describe what is involved in a system of inquiry that elicits and validates systemic knowledge, particularly those systems constructed using the concepts in Steps 3 and 4. By explaining the concepts upon which the system is built, Steps 3 and 4 effectively describe how the V is constructed.

The six steps are also organised by intentionality. The pairs of steps at each level clearly deal with the same type of phenomena. Step 1 ‘logically’ comes before Step 6, while Step 2 ‘logically’ comes before Step 5. The order in which the vertical and horizontal vertical axes are introduced is immaterial. Providing that the concepts related to diagonals are treated at the end, Steps 3 and 4 may be interchanged, a fact that suggests that they are indeed independent constructs. In addition coherence demands that Steps 1, 2, (3 or 4), 5, and 6 are described in that order. That is, these steps are conceived as belonging to a time order of events or start-stop precedence sequence. Only Steps 3 and 4 can be reversed without damaging understanding.

The six steps form a chain of evidence that must emerge from the application of design principles to a practical knowledge management problem. The meta-principles or design guidelines incorporated in the V-model are summarised in Figure 7.

98 J. Sheffield

Figure 7 Meta-principles or design guidelines incorporated in the V-model

Thus, the V-model presented in Figure 6 of Section 6 encapsulates the system that is the paper entitled Systemic Knowledge and the V-model (and in which Section 6 is a part).

The contribution of the research is the generation from general systems concepts of a deceptively simple but robust model that is demonstrated via recursion. The limitations are a lack of an epistemological basis and lack of application examples. These issues are addressed in Sheffield (2004; 2005).

References

Ackermann, F. and Eden, C. (2001) ‘SODA – journey making and mapping in practice’, in J. Rosenhead and J. Mingers (Eds.) Rational Analysis for a Problematic World Revisited: Problem Structuring Methods for Complexity, Uncertainty and Conflict, Chichester, UK: Wiley.

Anthony, R.N. (1965) Planning and Control Systems: A Framework for Analysis, Boston, USA: Harvard University Press.

Anthony, R.N. and Govindarajan, V. (2004) Management Control Systems, Boston, USA: McGrawHill/Irwin.

Argyris, C. (1999) On Organizational Learning, Oxford, UK: Blackwell.

Burrell, G. and Morgan, G. (1979) Sociological Paradigms and Organizational Analysis: Elements of the Sociology of Corporate Life, London, UK: Heinemann.

Checkland, P. and Holwell, S. (1998) Information, Systems and Information Systems, Chichester, UK: Wiley.

Cilliers, P. (1998) Complexity and Postmodernism, London, UK: Routledge.

Daft, R. and Lengel, R. (1986) ‘Organizational information requirements, media richness and structural design’, Management Science, Linthicum, USA, May, Vol. 32, No. 5, pp.554–571.

Davenport, T.H. and Prusak, L. (1998) Working Knowledge: How Organizations Manage What they Know, Boston, USA: Harvard Business School Press.

DeBono, E. (1999) Simplicity, London, UK: Penguin Books.

“Employ a cons is tent framework or sys tem of inquiry that resembles a V. The left half of the V contains process steps associated with intentions (e.g., 1. developing ideas , 2. objectives and 3. activities ). The right half of the V contains process s teps associated with developing outcomes (e.g., doing 4. thoughtful activities to 5. achieve results that 6. payoff). Steps one to three success ively refine and narrow intentions . Steps four through s ix success ively aggregate and expand outcomes . Validate sys temic knowledge via testing the coherence among intentions and outcomes at three levels of inclus iveness .”

INTENTIONS OUT COMES

1

2

3 4

5

6

Systemic knowledge and the V-model 99

Fahey, L. and Prusak, L. (1998) ‘The eleven deadliest sins in knowledge management’, California Management Review, Berkeley, USA, Spring, Vol. 40, No. 3, pp.265–276.

Garvin, D. (1998) ‘Building a learning organization’, Harvard Business Review on Knowledge Management, Boston, USA: Harvard Business School Press.

Garvin, D. (2000) Learning in Action – A Guide to Putting the Learning Organization to Work, Boston, USA: Harvard Business School Press.

Gorry, G.A., Scott-Morton, M.S. (1971) ‘A framework for management information systems’, Sloan Management Review, Cambridge, USA, Fall, Vol. 13, No. 1, pp.55–70.

Gorry, G.A., Scott-Morton, M.S. (1989) ‘A framework for management information systems’, Sloan Management Review, Cambridge, USA, Spring, Vol. 30, No. 3, pp.49–61.

Habermas, J. (1972) Knowledge and Human Interests, translated by J.J. Shapiro, London, UK: Heinemann Educational.

Habermas, J. (1984) The Theory of Communicative Action Volume 1: Reason and the Rationalization of Society, translated by T. McCarthy, Boston, USA: Beacon Press.

Habermas, J. (1987) The Theory of Communicative Action Volume 2: Lifeworld and System, translated by T. McCarthy, Boston, USA: Beacon Press.

Igel, B. and Numprasertchai, S. (2004) ‘Managing knowledge for innovation: multiple case studies of science-based R&D in Thailand’, Innovation: Management, Policy and Practice, Queensland, Australia: eContent Management, December, Vol. 6, No. 3, pp.404–415.

Jackson, M.C. (1982) ‘The nature of soft systems thinking: the work of churchman, ackoff and checkland’, Journal of Applied Systems Analysis, Vol. 9, pp.17–28.

Jackson, M.C. (2000) Systems Approaches to Management, New York, USA: Kluwer/Plenum.

Johannessen, J-A., Olsen, B. and Olaisen, J. (2005) ‘Intellectual capital as a holistic management philosophy: a theoretical perspective’, International Journal of Information Management, Elsevier, Vol. 25, pp.151–171.

Maani, K. and Cavana, R. (2000) Systems Thinking and Modelling, London, UK: Prentice Hall.

Menand, L. (1997) Pragmatism: A Reader, New York, USA: Vintage Books.

Menand, L. (2001) The Metaphysical Club: A Story of Ideas in America, New York, USA: Farrar, Straus and Giroux.

Mingers, J. (2001a) ‘Multi-methodology – mixing and matching methods’, in J. Rosenhead and J. Mingers (Eds.) Rational Analysis for a Problematic World Revisited: Problem Structuring Methods for Complexity, Uncertainty and Conflict, Chichester, UK: Wiley.

Mingers, J. (2001b) ‘Combining IS research methods: towards a pluralist methodology’, Information Systems Research, Linthicum, USA, September, Vol. 12, No. 3, pp.240–259.

Morton, A., Ackermann, F. and Belton, V. (2003) ‘Technology-driven and model-driven approaches to group decision support: focus, research philosophy and key concepts’, European Journal of Information Systems, Basingstoke, UK, Vol. 12, pp.110–126.

Nonaka, I. and Takeuchi, H. (1995) The Knowledge Creating Company, Oxford, UK: Oxford University Press.

Porter, M.E. (1990) The Competitive Advantage of Nations, New York: The Free Press/Macmillan.

Ropohl, G. (1999) ‘Philosophy of socio-technical systems’, Philosophy and Technology, Vol. 4, No. 3, pp.59–71.

Rosenhead, J. and Mingers, J. (Eds.) (2001) Rational Analysis for a Problematic World Revisited: Problem Structuring Methods for Complexity, Uncertainty and Conflict, Chichester, UK: Wiley.

Senge, P. (1990) The Fifth Discipline, New York, USA: Currency Doubleday.

Senge, P. (1994) The Fifth Discipline Fieldbook, New York, USA: Currency Doubleday.

Sheffield, J. (2004) ‘The design of GSS-enabled interventions: a Habermasian perspective’, Group Decision and Negotiation, Dordrecht, Vol. 13, No. 5, pp.415–436.

100 J. Sheffield

Sheffield, J. (2005) ‘The evaluation of GSS-enabled interventions: a Habermasian perspective’, Proceedings of the Thirty-Eight Annual Hawaii International Conference on System Sciences, January, IEEE Computer Society Press, ISBN 0–7695–2268–8.

Sheffield, J. and Gallupe, R.B. (1994) ‘Using electronic meeting technology to support economic policy development in New Zealand: short term results’, Journal of Management Information Systems, Armonk, USA, Winter 1993–1994, Vol. 10, No. 3, pp.97–116.

Sheffield, J. and Gallupe, R.B. (1995) ‘Using group support systems to improve the New Zealand economy part II: followup results’, Journal of Management Information Systems, Armonk USA, Winter 1994–1995, Vol. 11, No. 3, pp.135–153.

Ulfelder, S. (2001) ‘The dirty half-dozen: six ways IT projects fail and how you can avoid them’, DarwinMag, June, http://www.darwinmag.com/read/060101/dirty.html

Weick, K. (1979) The Social Psychology of Organizing, New York, USA: Random House.

Weick, K. (1995) Sensemaking in Organizations, London, UK: Sage.

Weick, K. (2001) Making Sense of the Organization, Oxford, UK: Blackwell.

Appendix

Worked Example 1: Planning processes

How should a three-step planning process vision, action and outcomes be placed in a V-model?

Answer:

• Vision

The reasons why some future state is desirable (higher level of inclusiveness/abstraction + the first step → top left of the V).

• Action

What will be done to achieve that state, and how it will be done. (Lower level of inclusiveness/abstraction + second step → bottom of V).

• Outcome

The situation that is being enjoyed now that the vision has become reality (higher level of inclusiveness/abstraction + last step → top right of the V).

Worked Example 2: Presentations

How should a three-part presentation tell them what you are going to tell them, tell them, and tell them what you told them be placed on the V-model?

Answer:

• Tell them what you are going to tell them (higher level of inclusiveness/abstraction + the first step → top left of the V).

• Tell them (lower level of inclusiveness/abstraction + middle step → bottom of V).

• Tell them what you told them (higher level of inclusiveness/abstraction + last step → top right of the V).

Systemic knowledge and the V-model 101

Exercise 1: Quantitative

A quantitative executive information system ‘drills down’ to aggregate costs at three levels (programme, project and work order). Costs at each of these levels is the responsibility of a particular person (programme leader, project leader and team leader). Each item (viz, programme, project or work order) has a start date and a finish date. The person responsible maintains business rules such as the following:

• The start date of a programme must precede that of its constituent projects.

• The start date of a project must precede that of its constituent work orders.

• The start date of a work order must precede the charging of costs to that work order.

• The finish date of a work order must follow the charging of costs to that work order.

• The finish date of a project must follow that of its constituent work orders.

• The finish date of a programme must follow that of its constituent projects.

Describe the business rules in this executive information system using the V-model.

Exercise 2: Qualitative

Technical practices may be very different from managerial practices, yet each must be closely integrated for system (and career?) success. Separating out these issues via the concept of inclusiveness/abstraction and examining the linkages between adjacent levels is important to the learning, understanding and remembering of information systems. Courses that are commonly part of the information systems curriculum are listed below and may be described using the concepts of the V-Model. For example:

• INFORM 703 Information Systems Management, examines managerial practices that, in totality, are more inclusive/abstract (and higher on the V) than

• INFORM 702 Information Systems Analysis and Design, examines system development activities that are more inclusive/abstract (and higher on the V) than the objective knowledge of technical issues covered in

• INFORM 701 Information Systems Engineering.

Describe some of the conceptual linkages among these three information system courses using the V-model.