Operational Research Society

Palgrave Macmillan Journals

O.R. and the Systems Movement: Mappings and Conflicts

Author(s): Peter Checkland
Source: The Journal of the Operational Research Society, Vol. 34, No. 8, Systems in O.R. First
International IFORS and O.R.S. Meeting. Discussion Conference at Henly, U.K. 9-11 May 1983 (
Aug., 1983), pp. 661-675
Published by: Palgrave Macmillan Journals on behalf of the Operational Research Society
Stable URL: http://www.jstor.org/stable/2581700
Accessed: 25-10-2015 05:04 UTC

REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.org/stable/2581700?seq=1&cid=pdf-reference#references_tab_contents

You may need to log in to JSTOR to access the linked references.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact [email protected].

Operational Research Society and Palgrave Macmillan Journals are collaborating with JSTOR to digitize, preserve and extend
access to The Journal of the Operational Research Society.

http://www.jstor.org

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions
J. Opl Res. Soc. Vol. 34, No. 8, pp. 661-675, 1983 0160-5682/83$3.00+ 0.00
Printedin Great Britain.All rightsreserved Copyright<C)1983OperationalResearchSocietyLtd

O.R. and the Systems Movement:

Mappings and Conflicts

PETER CHECKLAND
University of Lancaster

The 'crisis' debate in O.R. expresses concern at the divergence between textbook O.R. and what
practitioners actually do. The debate is examined by comparing O.R., systems analysis and systems
engineering. They are all wedded to logic in situations in which logic may not be paramount. The
science in O.R. applies only to aggregate results, but the practitioner must deal with a specific situation.
The tradition of systems thinking which emerged from organismic biology is described. It leads to a
way out of the O.R. 'crisis', by providing a formal structuring of a paradigm of learning rather than
optimization. O.R. can aspire to match natural science, and pass the problems by; or it can close the
textbook/practitioner gap by changing its concept of 'being scientific'.

Key words: methodology, management, philosophy, systems theory

INTRODUCTION

In 1976 an ad hoc but prestigious panel, assembled by the Governing Board of the
National Research Council ofthe U.S.A., produced a remarkable report1 on "one proven
method of helping the administrators in public or private organisations make their
decisions". The "proven method" was something called S.A./O.R.; "the notation is
clumsy, but there seems to be no better alternative", claimed the report.
In its first chapter we read:

"Over the past 35 years there has been far-reaching development in the techniques
for applying the ways of thinking and working commonly used by scientists to the
problems confronted by decision makers in government, business and other institutions.
These techniques are called, variously, Operations Research, Systems Analysis, Manage?
ment Science, and Cybernetics. All these terms mean much the same thing and therefore
are combined here under the designation Systems Analysis/Operations Research
(S.A./O.R.)... The S.A./O.R. workers are technicians. They do not determine policy;
that is the task of the elected or appointed officials ... to whom the S.A./O.R. workers
report."
I have called the report "remarkable" because 1976 is rather a late date to come upon
apparently apodictic assertions of the kind quoted above. By the second half of the 70s
many knowledgeable people were not prepared simply to accept as unproblematic that
O.R., S.A., M.S. and cybernetics "mean much the same thing", that work in them is
properly described as "technique", that the professionals' body of knowledge is simply at
the disposal of "elected or appointed officials" or, indeed, that there is no problem at all
in dubbing these fields "scientific".
This paper concerns some of these problematical issues. It tackles them by examining
the relation between O.R. and what will be defined as 'the systems movement'. It considers
the nature of the two and hence is able to establish similarities and differences. Doing so,
it will be argued, enables the often polemical debate of recent years about the nature and
status of O.R. to be seen in a new perspective, one which argues for a specific direction
for the future development of the approach and has both academic and practical
consequences.
To accomplish these ends it will be necessary first to examine the 'crisis' in O.R., then
to discuss briefly the history and nature of both O.R. and the systems movement in order
to appreciate the differences between them. Comparing and contrasting them will then
yield its insights relevant to O.R.'s future.
661

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions
 Journal of the Operational Research Society, Vol. 34, No. 8

THE 'CRISIS' IN O.R.

One man's 'crisis' is another's 'business as usual', hence I would not wish to waste effort
in discussing whether or not the situation in O.R. constitutes a 'crisis', a 'paradigm shift',
a 'revolution', a 'debate', a 'schism' or 'unrufHed normality'. All these labels would no
doubt have their defenders. What would probably be agreed is that for some years,
especially in Great Britain and the U.S.A., there has been a marked readiness to discuss
the nature of O.R. and how it is changing or how it should change. Where a quizzical
professional chemist would have difficulty in finding amongst his colleagues an intellectual
debate on "What is Chemistry?", an operational researcher who notes the literature and
conference activity of his subject could hardly avoid coming across fundamental discussion
of O.R. itself. Chemists simply get on with doing chemistry; but operational researchers,
in doing O.R., have been asking themselves what it is they are doing, what is the scope
of O.R., its present limits and its future.
This questioning has arisen not in the wake of the many successes of O.R.; rather, its
origins lie in a feeling that the achievement has in fact been less than the early promise
suggested. Surveying the Journal of the O.R. Society over the period from the sixties to
the late seventies, Dando and Bennett find the "widespread mood of optimism" ofthe late
1960s converted by the late 1970s into "a quite widespread and often pessimistic debate
about the practical success of O.R.".2 When the Operational Research Society of Great
Britain held a public celebration to commemorate 30 years of O.R., the Financial Times
said "the keynote was one of disillusionment", and it was the then President ofthe Society,
not feeling affronted, who commented: "perceptive F.T.".3
Perhaps the most celebrated part of the questioning, certainly the most polemical, is the
reaction of some of the American pioneers of O.R. to the way the subject has developed.
Ackoff and Churchman were, with Arnoff, author/editors of the first major textbook of
O.R. in 1957,4 a collaborative effort by 15 people. In the first chapter of that book,
describing O.R.'s goal as "an over-all understanding of optimal solutions to executive-type
problems in organisations", the editors write:

"The comprehensiveness of O.R.s aim is an example of a 'systems' approach, since

'system'implies an interconnected complex of functionally related components. Thus a
business organisation is a social or man-machine system." (Page 7.)

Twenty-odd years later, Churchman takes a baleful view of the realizations of those
early aspirations. Where a useful recent account ofthe intellectual and institutional history
of O.R.5 describes the 1960s as the decade which saw "great strengthening of theoretical
foundations and academic programs", Churchman describes the decade as "The Dreary
60s".6 He sees it as the decade in which the theoretical developments separated from the
applications and the subject became academic in the worst sense:

"During the 1960s in a large variety of departments, O.R. academically became

'modelling'; not really modelling at all, but a study ofthe delights of algorithms; nuances
of game theory; fascinating but irrelevant things that can happen in queues...
"Meanwhile, the practice of O.R. was presented to managers as 'solving specific
problems by models', exactly the opposite of our original intent." (Page 19.)

In the debate of the last five years, this divorce of theory from practice (which personal
observation endorses) is no longer taken as requiring proof; it is taken as a given. Machol,
after two years spent visiting "some of the O.R./M.S. people in most of the countries in
Europe and North Africa", says:7
"I find, as everyone else has found, that the practice of O.R./M.S. differs substantially
from what is taught in universities, but that both the practice and the teaching are more
practical in Europe than in the U.S." (Page 209.)

This situation of a "great and growing divergence between what is published and taught
on the one hand, and what is actually done on the other"' had already evoked a sharp

662

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions
 P. Checkland?O.R. and the Systems Movement

response from Ackoff. The apocalyptic language of the titles of the two plenary papers
he gave to the 1978 Annual Conference of the O.R. Society (The future of operational
research is past;8 Resurrecting the future of operational research9) was itself a further
stimulus to the debate which had already begun in many guises.10 Ackoff's complaint goes
beyond the gap between theory/teaching on the one hand and practice on the other, to
which Machol and many others refer. Ackoff's argument is that what he is dealing with
is "the death of O.R.",8 and that what is needed is "an alternative paradigm",9 one which
focusses on "planning for and design of systems"9 rather than "optimization and (the)
pursuit of objectivity".8
Kuhn's concept of paradigm11 is a slippery one, at once sociologically insightful and
philosophically obscure, with Kuhn himself using the word in "not less than twenty-one
different senses".12 Nevertheless, most operational researchers, systems analysts, manage?
ment scientists and cyberneticians would agree broadly that early O.R. studies make sense
according to the paradigm of experimental natural science, but with attention not on the
natural world but on human (initially military) operations. Civil applications then
produced textbook O.R. after the Second World War, but as the experience of these
applications accumulated, so did the unease, the feeling that purposeful activity in
organizations is much more complex than the analytical methods allow. Hence arises the
debate about the adequacy ofthe paradigm of (natural) scientific objectivity, the discussion
about the need for a different-possibly a systemic?paradigm.2
The rest of this paper will be concerned with this possible paradigm shift. I shall try to
show that the 'crisis' debate has
been characterized by a lack of knowledge of the
alternative (systemic) paradigm, not and
only by those who adhere to the scientific
paradigm. I shall argue that the systems movement is itself struggling with a paradigm
shift, and that this displacement sheds light on the future for O.R. or any other attempt
rationally to intervene in human affairs.

O.R. AND SYSTEMS ANALYSIS

Remembering that knowledge of the world outside ourselves is indivisible, and that
convenient division of knowledge into 'subjects' or 'disciplines' is an arbitrary human act,
it may be difficult to take seriously a discussion ofthe notions 'O.R.' and 'systems analysis'
and whether they are identical, or how they are different. Nevertheless, examining the
admittedly arbitrary distinctions which are the source of argument can teach us much
about the current state of knowledge. In the present case the thin vein of discussion over
the years of the relation to each other of O.R. and system analysis is relevant to
understanding the present 'crisis' but is hardly rich enough to resolve it.
Although the vein is not a rich one, it is nevertheless virtually continuous over the
history of O.R., from the Flagle, Huggins and Roy volume of 1960 {Operation Research
and Systems Engineering12) to discussion in the 1980s of progress in both areas taken
together, written for the Centennial Issue of Science,14 or discussion of the question
whether 'systems analysis' represents a new vocabulary for O.R. or a new conceptual
framework.15 In between we find, for example, a discussion of system theory "from an
operations research point of view",16 and discussion of O.R. "from the viewpoint of
General Systems Theory".17
Perhaps surprisingly, these latter two papers do little to illuminate the differences and/or
similarities between O.R. and systems-oriented activity. Sengupta and Ackoff16 argue that
the distinction is that where systems theory may be either descriptive or normative, O.R.
is prescriptive:

"It is primarily concerned with how systems ought to behave and only with how they
do to the extent that such knowledge is necessary to arrive at prescriptions." (Page 9.)

Sagasti and Mitroff17 offer a systems model of the O.R. process in which a conceptual
model of a problem situation leads to a scientific model which "should be capable of being
manipulated and of generating solutions." (Page 698.)
663

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions
 Journal of the Operational Research Society, Vol. 34, No. 8

Most of this vein of the literature, however, worries away at the difference, or similarity,
between O.R. and systems analysis or systems engineering. The general view expressed is
that the difference is one of scale, the difference being that between tactics and
strategy.1415,18'19 The point is sometimes made polemically, even eristically; Miser, however,
makes it coolly:14

"... the wartime work in operations research... began in tactics but grew into
planning and strategy ...
"Thus, the success of operations research workers in developing scientific theories
describing important classes of phenomena occurring in man/machine operating systems
and in using these models to solve problems arising in these systems has inevitably driven
them to study larger and larger systems: in other words, to what has come to be called
'systems analysis'." (Page 144.)

Not surprisingly, operational researchers who do not wish to be tied to what a hostile
critic terms "'cookbook' approaches to narrow and highly specific problems"19 sometimes
join O.R. and systems analysis together, as in the quotation at the start of this paper
concerning "S.A./O.R.",1 or in Tomlinson's "ORASA".20
But discussion which asserts the identity of O.R. and systems analysis, or uses a
historical perspective simply to differentiate them by scale, neglecting the relation of either
to systems thinking, precludes the learning which might accrue from a careful examination
of the fundamental nature of the two fields and which relates them both to the parallel
development of systems thinking in its various forms. The rest of this paper constitutes such
an examination.

THE FUNDAMENTAL NATURE OF O.R., SYSTEMS

ENGINEERING AND SYSTEMS ANALYSIS

Operational research
Human groups which develop a corporate identity and so survive and grow are always
shaped by stories concerning their origin, by the creation of their myth, using the word
as ethnologists use it, not in the sense of "made up' or 'untrue'.21 There would be no
disagreement that the core component in the O.R. myth is that its origin lies in empirical
natural science. Professional operational researchers make sense of their world by seeing
themselves as the heirs of the natural scientists who, in the late 30s in the U.K., worked
with officers and men of the Royal Air Force to create and measure the effectiveness of
radar-based air defences. A. P. Rowe called this work "operational research", and in
war-time it quickly spread to the other services and to the American armed forces.5
The work was scientific because it depended upon collecting and analyzing evidence and
making predictions which could be publicly tested. What made it novel was the fact that
these scientists were not playing the normal game against unchanging Nature, which goes
on within laboratories; they were studying real-world operations and doing so outside the
laboratory. This meant that they could not manipulate the research object in the same way
that, for example, the laboratory physicist studying magnetism can decide what to include
in his experiment, what to leave out, which variables to change and which to keep constant.
The stratagem was to construct a surrogate for the real-world situation, a model of it which
could be manipulated where the research object itself (the real-world situation) could not.
Experiments could then be done on the model and the results hopefully transferred to the
real situation.
The ease with which scientific results could be achieved when the research object was
a real-world operational situation surprised the pioneers of O.R. Blackett's celebrated note
on the methodology of O.R.,22 written in the early 40s, records that:

"... many more useful quantitative predictions can be made than is often thought
possible. This arises to a considerable extent from the relative stability over quite long
periods of time of many factors involved in operations. This stability appears rather

664

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions
 P. Checkland?O.R. and the Systems Movement

unexpected in view of the large number of chance events and individual personalities
and abilities that are involved in even a small operation. But these differences in general
average out for a large number of operations, and the aggregate results are often found
to remain comparatively constant." (Page 178.)

What is being said here is that O.R. had discovered the logic of situations. Although any
particular situation may be dominated by Blackett's "chance events" and "individual
personalities and abilities", a group of many examples will exhibit a situational logic which
can be scientifically investigated using the ruse of working with a model of the logic of
the situation rather than with the situation itself.
This discovery, or insight, leads inexorably to the formal methodology of an O.R. study.
From Churchman, Ackoff and Arnoff's version in the 1950s4 to the version in the
Handbook of Operations Research in 197823 the same pattern emerges: formulate the
problem as an objective to be achieved; build a model of the situation (these casual words
are frequently used. What should be said is: build a model ofthe logic ofthe situation?any
situation is much richer than its logic!); derive from 'experiments' on the model the solution
which best achieves the objective; implement the solution.
This was an intellectual breakthrough. Its thorough exploitation followed rapidly in
peacetime O.R. It was discovered that industrial and commercial operations throw up a
number of structured problem situations which recur, problems concerning, for example,
allocation, inventory, replacement, queueing, routing, location and bidding. The logical
algorithms concerning these situations were worked out and, being highly teachable,
passed into university courses in O.R. Sometimes they became the very substance of those
courses, the situation Churchman deplores as characterizing "the dreary 60s".6 The typical
textbook of O.R. emerged as a brief account of the war-time emergence of O.R., a bald
statement of its methodology, and a series of chapters considering in turn the logical
structure of the problem situations which recur: the investment portfolio problem; the
queueing problem; the depot location problem etc. (A major university text in the late
1970s takes but 22 of its 1000 pages to discuss the origins, overall methodology and
implementation of O.R.; the rest discuss the algorithms.24)
We can see from this picture of the emergence and growth of O.R. that the "divergence
between what is published and taught on the one hand and what is actually done on the
other",7 and the 'crisis' debate which was initiated by the divergence, are wholly
unsurprising. For many academics, research on the algorithms which describe the logic of
a class of situation was more attractive than "action research"2526 in specific situations?
and yielded more readily-published results. Researchers who followed that line were bound
to become separated intellectually from practitioners who were concerned not with the
logic of a class of situation, but with a unique, idiosyncratic situation much affected by
Blackett's "chance events" and "individual personalities". The 'crisis' debate was bound
to follow.
In summary, then, O.R. seeks to transfer the empirical method of natural science to
real-world operations. It does this by defining the objective to be achieved in a real-world
activity and explores how the objective might be achieved by manipulating a model of the
problem situation. A number of ready-made manipulations are available for certain
well-structured problem situations which recur.
The weaknesses of this approach are discussed later in this paper, together with the
weaknesses of systems engineering and systems analysis.

Systems engineering and RAND systems analysis

Having discussed at length elsewhere27 the origins and nature of both systems en?
gineering (S.E.) and RAND-style systems analysis (S.A.), I shall here only outline that
argument.
In the 1950s the scientists and engineers of Bell Telephone Laboratories hoped to
establish standard procedural mechanisms by means of which new research knowledge
would be translated into applications meeting human needs. In the classic account of

665

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions
 Journal of the Operational Research Society, Vol. 34, No. 8

systems engineering methodology, Hall28 sees it as part of "organised creative technology".

By generalizing from case histories, Hall outlined a problem-solving approach which
closely resembles that of O.R. Concerning differences between the two, Hall points to
S.E.'s origins in "research and development organisations", as compared with O.R.'s
military origins, and suggests that while "operations research is usually concerned with the
operation of an existing system ... systems engineering emphasises the planning and design
of new systems".28 (Page 18.) Such differences could hardly lead to significant intellectual
differentiae, and Hall's methodology is highly recognizable to an operational researcher.
It involves: expressing the problem situation as the definition of a need; choosing
objectives; creating alternative systems (intellectually); analyzing the alternatives via
models; selecting the most promising according to criteria related to the objectives; and
realizing the system in practice.
Just as operational researchers were quick to spread the approach from military to
civilian applications, systems engineers soon claimed that their field included not simply
systems of hardware but the planning, design and realization of the combined concrete and
abstract systems which deliver such societal needs as health care, communication, law
enforcement and education.
Simultaneously with the development of S.E. in the 1950s, there emerged the strand of
methodological thinking known as 'systems analysis', a development associated especially
with the RAND Corporation. RAND (an acronym for "R and D") began as a
Government-funded project with the Douglas Aircraft Company in 1946. The aim was "to
assist in military planning, and particularly in co-ordinating planning with research and
development decisions".29 The project became the non-profit RAND Corporation in 1948,
and during the 1950s the pattern of a RAND-style systems analysis became clear. The
work consisted of a broad economic appraisal of the costs and consequences of various
alternative ways of meeting a defined end. It was a refinement of the 'requirements
approach', developed in the American Government since the 1930s, in which officials
defined 'a requirement' whose provision would solve a problem, the requirement being a
piece of equipment, a task or a complete system. Feasibility and performance character?
istics of alternatives would then be checked and the analysis passed to Government
decision-makers for a decision on whether the necessary budget could be obtained.
During the 1960s, Secretary McNamara made RAND-style systems analysis/cost-benefit
analysis/program budgeting standard practice within the Pentagon, and by 1965 there was
an "Assistant Secretary of Defence for Systems Analysis". McNamara's Comptroller was
Charles Hitch, who in a RAND report of 195530 had given an account of S.A. methodology
very similar to that of O.R. and S.E. According to Hitch, the elements ofthe S.A. approach
are: an objective we desire to achieve; alternative systems for achieving it; costs or resources
required by each system; models showing the inter-dependencies of objectives, systems,
resources and environment; and a criterion for choosing the preferred alternative. The
historian of RAND describes this methodology as a "sister discipline" of O.R. "less
quantitative in method and more oriented toward the analysis of broad strategic and policy
questions".29 (Page 8.)
This brief examination of the history of the emergence of O.R., S.E. and S.A. makes
it clear why it is normally stated that the difference between O.R. and the other two
approaches lies in O.R.'s tactical focus and its concern for existing rather than future
systems. O.R. did begin its life at a tactical level, concerned, in wartime, with improving
existing operational systems; S.E. and S.A., with origins in research projects and in
planning respectively, were from the start concerned with strategic issues and not-yet-
existing systems. But these are trivial differences when set alongside the much greater
similarities.
Analysis of a dozen expositions of S.E. and S.A. published between 1955 and 1976
(described in detail in Checkland27) shows that fundamentally they all reduce to the same
intellectual proposition. It is clear that a dozen expositions of the overall methodology of
O.R. could be added to that analysis without changing its outcome. O.R., S.E. and S.A.
are all predicated upon the belief that an important class of real-world problems can be

666

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions
 P. Checkland?O.R. and the Systems Movement

formulated as a search for an efficient means of achieving objectives known to be desirable.

The search can be conducted systematically, according to this proposition, by defining the
objective to be achieved and manipulating models of the situation or of alternative forms
which it might take. All of this can be done scientifically because, in spite of chance events
and individual personalities and abilities, real-world operations do in aggregate reveal the
working out of the logic of situations.

THE FUNDAMENTAL WEAKNESSES OF O.R.

SYSTEMS ENGINEERING AND SYSTEMS ANALYSIS

Consider now the fundamental weaknesses of these formal, 'official', as-taught, versions
of O.R., S.E. and S.A. methodology. (In doing this I shall not be denying or belittling the
ability of skilled individual practitioners to achieve useful results; my concern here is the
deficiencies of the generalized, the in-principle practitioner-free versions of what I have
elsewhere designated "'hard' systems thinking",27 by which I mean systematically-ordered
thinking concerned with means-definition in well-structured problems in which desirable
ends can be stated.)
Obviously S.E. and S.A. will be found wanting intellectually in problem situations which
cannot be formulated as a search for a way to a known-to-be-desirable end. In a decade
of action research in which, initially, the methodology of 'hard' S.E. was applied in
unsuitably 'soft' (ill-structured, 'wicked') problems, my colleagues and I found both that
well-structured problems are extremely rare in human situations and that the systems
methodology had to be not simply modified but radically transformed into what is now
called "soft systems methodology"27 if it was not to pass the problems by. Soft systems
methodology is a methodology of inquiring and learning, rather than optimizing, and in
the remainder of this paper I shall try to apply lessons from its development to the 'crisis'
in O.R. But first, consider the weaknesses of the formal textbook methodology of O.R.
Firstly, as with S.E. and S.A., O.R. will face difficulties in principle if the problem
situation cannot be expressed in terms of an objective to be achieved. This, we can surmise,
will always be problematical because there will always be several answers to the question:
whose objectives? The RAND analyst and the systems engineer have answers to this
question: respectively, the real-world decision-taker who has the decision problem and the
originator ofthe specification for the systems engineering team. The Panel ofthe National
Research Council ofthe U.S.A., quoted in the introduction to this paper, import the same
solution into O.R. The S.A./O.R. workers, they assert, "are technicians" who report to
"elected or appointed officials".1 But these are political solutions to what is really an
intellectual problem. Intellectually there is no reason at all why the expertise should not
be equally at the disposal of the likely victims of the system in question, those who decree
its creation, its designers or, indeed, any other group with any interest in the outcome of
the study. So it is a weakness of the formal methodology of O.R. that it does not contain
any response to this intellectual problem.
Secondly, even if the objectives problem can be resolved, it is extremely unlikely that
the real-world problem situation will map neatly the well-structured situations with which
the algorithms deal. It will be less pure, more what Ackoff31 terms "a mess" and Rittel32
"a wicked problem", a problem which defies formulation, has no stopping rule, is
essentially unique. Thus, although the concept 'depot location problem' is a meaningful
one, to which network and transportation algorithms are relevant, any actual depot
location problem in an industrial situation will resist isolation from problems of other
kinds: capital investment, inventory policy, marketing policy, organizational politics and
personnel, for example. O.R. methodology has a weakness to the extent that it imagines
that pure problem forms map real-world problems.
Thirdly, because O.R. is putative science, i.e. it seeks to establish testable public
knowledge, and because the currency of science is rational thought, O.R. will be able to
deal with real situations to the extent to which they exhibit rational behaviour or are
susceptible to rational analysis. Idiosyncratic irrational behaviour will be difficult to bring
667

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions
 Journal of the Operational Research Society, Vol. 34, No. 8

within O.R.'s universe of discourse. Thus, where in a real situation the depot location
algorithms based on economics may indicate, say, a warehouse near Coventry, the actual
warehouse may get built at Macclesfield because the Marketing Director has a mistress
in south Manchester, something the cost-based algorithms missed. O.R.'s commitment to
science, historically understandable, creates the weakness that it deals excessively in
rationality, "excessively", that is, for human situations, situations which normally exhibit
a spendid mix of rationality, farce and general 'cock up'.
These weaknesses of formal O.R. can all be traced back to the quotation from Blackett's
wartime memorandum quoted earlier.12 Blackett expressed the surprise which the scientists
felt when it was found that in spite of "the large number of chance events and individual
personalities and abilities that are involved in even a small operation... the aggregate
results are often found to remain comparatively constant". In other words, in the
aggregate, the logic of situations dominates and is capable of being scientifically
investigated. Blackett's phrase about the aggregate results is crucial. It signals that
academically-inclined researchers can set about developing the algorithms of situational
logic. But it also signals the problem which the 'crisis' debate has at least revealed starkly:
that the practitioner is not dealing (as the wartime submarine hunters were33) with
aggregates of similar operations. The practitioner is dealing with a unique situation with
its own possibly crucial chance events and its own individual personalities; to such
situations the aggregate results do not apply.
Even as Blackett was pointing out that operations could be researched scientifically,
Freeman Dyson as a young operational researcher in Bomber Command was finding that
his results?scientifically impeccable?were simply unacceptablel Dyson had demonstrated
that, contrary to the general belief in Bomber Command, it was not true that a crew's
chance of surviving a mission increased with experience.34 Alas for his science, it countered
the official mythology?important for maintaining morale?that after the firs* few
missions, survival chances increased; the science was discounted, even though it pointed
to counter measures against German upward-firing guns which, we now know, would have
saved RAF lives.
In a sentence, the weakness of O.R., not less than that of S.E. and S.A., is that it is
wedded to logic in situations in which logic is not necessarily paramount.

SYSTEMS THINKING AND THE SYSTEMS MOVEMENT

Introduction
So far this paper has not questioned the assumption that O.R. in some way embodies
'a systems approach'. It has been taken as given that in seeking an optimal system to
achieve the objectives extant in the problem situation, O.R. adopts this approach. As the
Churchman, Ackoff and Arnoff text asserted in 1957: "The comprehensiveness of O.R.'s
aim is an example of a 'systems approach'".4 Is it?
It would be difficult to deny that O.R., S.E. and S.A. are all, in fact, systematic?that
is to say: rational, methodical; not random but carried out according to a plan.
But there is another adjective from the noun 'system', namely 'systemic'?of, or
pertaining to, the whole.35 It is much less obvious that O.R. (or S.E. or S.A.) is systemic.
If they were, the rudiments of a solution to the 'crisis' in O.R. might be apparent.
The adjective 'systemic' used to refer only to (and is still often defined in dictionaries
only in relation to) a strand of thinking in biology and medical science which originated
in the second half of the 19th century. By the 1920s, the so-called organismic biologists
were arguing (in books such as J. H. Woodger's Biological Principles of 1929) that the
reductionism of natural science was probably not the best approach to the understanding
of specifically biological phenomena. In the 1940s biologists like von Bertalanffy gener?
alized organismic thinking ("the system theory of the organism", as Bertalanffy called it)
into thinking concerned with wholes of any kind. Systems thinking of this systemic kind,
emerging in the late 1940s and spreading into diverse fields, has been virtually ignored in
the literature of the 'hard' (systematic) approaches: O.R., S.E. and S.A.
668

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions
 P. Checkland?O.R. and the Systems Movement

I have argued that if this mode of thinking is stripped to its core ideas, two pairs of
concepts remain: emergence and hierarchy, communication and control.27 The most basic
of these concepts is that of emergence, the idea that whole entities have so-called 'emergent
properties', properties which are meaningful in relation to the whole entity, not in relation
to its parts. The smell of ammonia is an emergent property of that gas which has no
meaning at the level of the nitrogen and hydrogen which make up ammonia molecules.
This idea has become part of everyday language in the expression "The whole is more than
the sum of its parts". My student had understood the idea exactly when I heard her say
to a fellow student: "You are certainly more than the sum of your parts, you're an idiot!".
She was saying that the description "idiot" has no meaning at any level below that of her
companion regarded as a single entity.
On the other hand, levels below that of the whole entity may themselves contain
sub-wholes (with emergent properties at the new lower level); thus wholes may contain
sub-wholes in a hierarchial structure and may themselves be sub-wholes of higher levels.
Such wholes, according to the basic systems image or metaphor, may survive in
environments which change by adapting to these external changes. Such adaptation
requires processes of communication and control (the latter in the control engineer's sense
of managing variables which govern the behaviour of the whole).
These ideas have emerged since the 1940s as the core of explicit systems thinking. Any
use of such thinking, systemic and systematic, in any field of endeavour is part of 'the
systems movement', a diverse and ill-defined intellectual enterprise. Try looking for
systems books in university libraries: they will be found in every section?in biology,
geography, economics, history, politics, psychology, sociology etc. etc. This means that
understanding the developing systems movement as a whole is not easy.

The shape of the systems movement2736,31

In spite of the problem of its diversity, it is possible to make sense of the systems
movement as a whole by making four distinctions which define different kinds of
systems-oriented activity. These yield not exactly a map of the systems movement, since
any piece of systems work may cut across the boundaries, but rather a map which enables
us to make sense of, or place, any example of systems thinking in the context of the
movement as a whole.
The distinctions made to create the schema are as follows:

o.r.s.34/8?b 669

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions
 Journal of the Operational Research Society, Vol. 34, No. 8

Examples of work which fits fairly tidily into these categories (although any specific
example may be less tidy) would be:

2.2 The 1970s systems revolution in geography.38

3.1 Work on general systems theory (G.S.T.)39 or cybemetics.40
4.1 S.E.28
4.2 RAND S.A.,30 O.R.
4.3 Application of soft systems methodology.27
Our concern here is specifically the distinctions between categories 4.1-4.3 within category
3.2?the development of systems ideas by using them in problem situations outside the
laboratory. The argument is that the bodies of work in 4.1 and 4.2, arising in different
historical circumstances, are nevertheless intellectually very similar indeed. In a sentence,
they embody a systematic approach. Category 4.3 arises from the failure to apply 'hard'
S.E. in 'soft' problems of management: it tries explicitly to add systemicity to the
systematic approach.

Systems of types 1, 2 and 3

Perceiving the systems movement as a whole by means of the schema above, and
examining the work which goes on within it, draws attention to the fundamentally different
types of entity (things, phenomena or situations) which the researcher or practitioner may
deem to be 'systems'. Three fundamentally different types may be discerned:41

"Type 1: situations or phenomena characterized by interconnections which are part of the

regularities of the universe. Examples would be frogs, foxgloves, ecological systems,
systems of chemical reactions.
Type 2: situations characterized by interconnections which derive from the logic of
situations. Examples would be arrangements to manufacture or assemble products, or
situations dominated by a decision about to be taken to achieve a known objective.
Type 3: situations in which interconnections are cultural, situations dominated by the
meanings attributed to their perceptions by autonomous observers. Most real-world
problem situations are of this type, both on the small scale (e.g. how should we behave
towards aging parents?) and on the large (e.g. should the nuclear deterrent be
abandoned?)." (Page 12.)
In this perspective it is clear that the hard systems approach was developed in, and is
suitable for use in, situations of types 1 and 2. Natural science deals with phenomena of
type 1. Soft systems methodology was developed for use in situations of type 3. Most
individual situations entailing human perceptions are of this type, and the regularities
which derive from the logic of a class of situation may not be exhibited in the specific
example with which the practitioner is trying to cope. This analysis gives us another way
of describing the theory/practice split in O.R.: it is the result of the theory covering
situations of type 2, while the practice is immersed in type 3 situations.

The concept 'human activity system'

Situations of type 3 derive from the autonomy of the human observer, from his or her
freedom to attribute meaning to perceptions. This gives special importance to a particular
kind of system highly relevant to would-be-problem-solving interventions in human affairs:
the 'human activity system'.27,42 Learning to cope with the special characteristics of this
systems concept is what transformed S.E. into soft systems methodology.
The systems thinker may perceive the world as containing many wholes which are
natural systems; in addition, the man-made world may be perceived as containing many
designed wholes, some concrete (tramcars, computers, chemical plant, etc.) some abstract
(mathematics, positivist philosophy, existentialism, etc). In addition, it is clear that a
structured set of connected activities which together comprise purposeful activity may itself
be regarded as 'a system', since it may be convincingly described in terms of the basic
systems metaphor of emergence, hierarchy, communication and control. Voting in a
670

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions
 P. Checkland?O.R. and the Systems Movement

democratic election, selling second-hand cars, playing the Cup Final, manufacturing nitric
acid?any purposeful activity of this kind may be described as a human activity system.
However?and here is the subtle nature of this system concept?where a description of
a natural system (such as a bee) or a designed system (such as a bicycle) will be publicly
testable, a description of a human activity system will not. The set of linked real-world
activities which one observer describes as "terrorism" or "a terrorist system" will be
described by another as "a freedom-fighting system". Each observer is perceiving the same
human activity; and each can give a valid account of it from his point of view. But the
descriptions will be very different because of the very different images of the world, or
Weltanschauungen, which the different observers take as given. This is always the case
where cultural artefacts are concerned. A large number of systems descriptions of a prison
could be given, deriving from different Weltanschauungen: punishment, rehabilitation,
education, protection of society, 'university of crime', etc.
Thus, when using the concept 'human activity system' (and this is a concept highly
relevant to real-world problem situations), it is always necessary to name not only the
system but the Weltanschauung which is being taken as given. (There are certain
circumstances in which it is worth striving for as Weltanschauung -free a description as
possible of human activity, but this will not be discussed here. See Checkland and Wilson,43
Wilson.44) It is doing so which points the way to getting past the logic of situations (which
applies only to the aggregate) to the unique features of the special case with which the
practitioner deals.

The two systems paradigms

The line of argument concerning systems thinking here leads to one more result which
is relevant to placing the 'crisis' in O.R. in a new perspective.
The special nature of the concept 'human activity system', that there is no once-and-
for-all publicly testable systemic description of human activity, only descriptions valid for
a particular world view, reminds us that in serious discourse we must be extremely careful
about casually referring to things in the real world as systems. We do this all the time in
'
everyday language: 'the education system', 'a transport system', a political system', etc,
and for everyday purposes it does not matter. But for serious purposes we need to use
language with greater care, and need to remind ourselves that we have no access to what
the world is, to ontology, only to descriptions of the world (some of which may survive
severe tests), that is to say, to epistemology. Newton's Principia describes "the frame of
the system of the world", and no doubt Newton thought he was describing what the world
is. But the 20th century has reminded us that Newton's account is only an account of a
description of the world, one which survives less severe tests than Einstein's?which will
itself ultimately be replaced.
Thus, systems thinking is only an epistemology, a particular way of describing the world.
It does not tell us what the world is. Hence, strictly speaking, we should never say of
something in the world: "It is a system", only: "It may be described as a system". (Of
course, keeping to that rule is tedious!)
The status of systems thinking as epistemology, not ontology, is made very apparent
in the systems thinking which consciously uses the concept 'human activity system',27 and
the learning associated with that development has suggested that within the systems
movement there are two basic intellectual stances, here called paradigms I and II.45 The
first paradigm does not recognize the primacy of epistemology over ontology; the second
paradigm does.
Consider the human observer who, although himself part of reality, can by a familiar
mental act think about his own intellectual engagement with the rest of reality, R, in which
he uses particular mental processes or methodology, M. One common stance, paradigm
I, makes the following assumptions:

(i) R is systemic; the world contains systems;

(ii) M can be systematic.
671

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions
 Journal of the Operational Research Society, Vol. 34, No. 8

In this paradigm the world is assumed to contain systems, and understanding or

engineering them is assumed to be possible via systematic processes. This is the paradigm
of RAND S.A. and systems engineering, and of much natural science. Its implicit social
theory is that of functionalist sociology (the notion that social systems can be treated as
if they were organisms,46,47 and its underlying philosophical base is that of positivism (the
philosophical stance which concedes primacy to the given world as known through
empirical evidence48). This is the systems paradigm of the 50s and 60s, the period during
which it was assumed without question that to import systems thinking into the social
sciences was to adopt functionalist analysis.49
The alternative paradigm, paradigm II, has emerged in systems work in real-world
problems in the 70s.27 In this stance the corresponding assumptions to those of paradigm
I are as follows:

(i) R is problematical: we cannot know it ontologically;

(ii) M can be systemic.

In this paradigm the world is taken to be very puzzling: there is no reason why we should
have evolved brains beyond those needed to survive together on the planet's surface; there
is no reason to suppose our senses and brains can tell us what reality is really like. But
the process of engaging with it, of trying to understand and/or engineer it, can be systemic.
This is the paradigm of soft systems methodology,27 of what Churchman calls "Singerian
inquiring systems",50 of Vickers' "appreciative systems".51 Its implicit social theory is that
of interpretive rather than functionalist sociology,52 and its underlying philosophical base
is that of phenomenology53,54 (the philosophical stance which concedes primacy to the
mental processes of observers rather than to the external world).
The important feature of paradigm II, as compared with paradigm I is that // transfers
systemicity from the world to the process of inquiry into the world. The paradigm of
optimizing becomes a paradigm of learning. (In soft systems methodology the process is
formally embodied in a naming of relevant human activity systems in 'root definitions',55
a building of models from those definitions,56 and a comparing of models with perceptions
of the real-world situation; the comparison yields changes simultaneously systemically
desirable and culturally feasible.27)
In the terms of this analysis, traditional O.R. has been rooted in paradigm I. And yet
another way to describe the 'crisis' debate in O.R. is to say that it is the debate occasioned
by increasing awareness of the inadequacy of paradigm I in the face of the extreme
complexity of real-world affairs.

Systems thinking: conclusion

This section has tried to outline briefly some features ofthe tradition of systems thinking
as it emerged from organismic biology and has developed over the last several decades.
The material has been summarized here because, on the whole, this strand of thinking is
not part of the mental furniture of operational researchers and is not normally referred
to in the literature of O.R.
The final section will argue its relevance to the future of O.R.

TOWARDS SYSTEMIC O.R.?

The didactic nature of the previous section has been necessary because of the scarcity of
discussion of systems thinking in the literature of O.R.
Probably for most operational researchers the phrase 'systemic O.R.' would call to mind
Beer's work,57,58 an impressive but lonely edifice based on the notion that cybemetics is
"the embodiment of science in O.R.".59 There is not space here to discuss this body of
work, beyond pointing out that it assumes a functionalist model of an organization and
pays great attention to the engineering of its control systems at many levels. This mix of
functionalism and control engineering60 means that its intellectual arena is, to use the
language of the previous section, a paradigm I approach to systems of type 2.
672

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions
 P. Checkland?O.R. and the Systems Movement

This conclusion holds for much of the rest of the sparse discussion of these matters in
the O.R. literature.
Bryer61 and Bryer and Kistruck62 have been concerned at what is seen as the inadequate
treatment of social reality in O.R., but in criticizing the "major attempts to apply systems
concepts to social organizations" the two authors assume that the only way to apply
systems concepts is to treat organizations as systems, with departments and sections as
sub-systems or sub-sub-systems. Never entertained is the notion of transferring systemicity
from the organization or the social situation to the process of inquiring into its (possibly
unique) social reality. The same lacuna occurs elsewhere. For example, Dando with his
collaborators has usefully discussed whether O.R. is in a Kuhnian crisis2 and has
emphasized the importance of Weltanschauung -analysis in decision-making situations.63
But when Dando and Eden,64 discussing O.R. as reflected in the Euro II Congress ("yet
another demonstration of the technically-bounded perspective which pervades so many
O.R. conferences") mention the systems perspective, they assert in a footnote that this "is
well known amongst O.R. workers" and end:

"Generally speaking we would expect to see a correlation between the systems

perspective and a dominance of Economics and Mathematics in O.R.!" (Page 267,
present author's exclamation mark.)

Of course, they are equating the systems perspective with the positivist, functionalist
stance of paradigm I. Pidd, similarly, discussing systems approaches and O.R.65 and finding
that "a conscious shift towards a pragmatic systems perspective might be of benefit to the
practice of O.R.", nevertheless describes the systems framework as having "severe
limitations". Again, these turn out to be the limitations ofthe positivist systems paradigm
of the 1950s and 1960s. All these authors, in fact, quote Silverman's The Theory of
Organisations49 to provide their definition of the systems view of social reality. This
admirable and influential, but now dated work of 1970, contrasts the systems framework
(positivist, functionalist) with "the Action Frame of Reference" (phenomenological) but
does not imagine that systemic thinking could pervade the latter. It now can and does.
If the outcomes from the tradition of systems thinking outlined in the previous section
are to be useful to O.R., then the major change in thinking is that operational researchers
would extend their concept of systemicity. The methodology of O.R., like the older systems
methodologies, S.E. and S.A. assumes that the model has to be a would-be representation
of some aspect of reality. The newer systems methodology uses models of human activity
systems to structure a process of inquiry. This can end the separation between the
practitioner struggling with the unique human situation and the theoretician limited to the
logic of situations which is revealed only in the aggregate?for example, by application
of such methodological rules as: never build a quantitative model without also building
several models (based on different Weltanschauungen) of the (human activity) system
which will use the quantitative model.
The intellectual cost of bridging the theory-practice gap is, of course, not negligible. For
S.E. and S.A. the cost of achieving high relevance to real-world problems (not just to their
logic) is giving up the idea of engineering the optimum system or finding the optimum
decision. For O.R. the cost would be some disengagement from the notion that O.R. is
scientific in the full natural-science sense.
When Eden and Jones published in this journal an account of a case study which
involved use of their methods of cognitive mapping66?something highly relevant to getting
beyond the logic of a situation to its specific human characteristics, they were told
magisterially by Machol67 that "there is no content in any of this", and that "this kind
of article is not operational research". This exchange illustrates a considerable difference
between the 'appreciative systems'51 of the two sides.
The issue is this: should a fully scientific O.R., an O.R. as scientific as physics, aspire
to deal not with the aggregate logic of situations but with the full complexities of "complex
problems arising in the direction and management of large systems of men, machines,
materials and money in industry, business, government and defence?".68 Or should O.R.,

673

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions
 Journal of the Operational Research Society, Vol. 34, No. 8

learning from the system thinking tradition it has so far largely ignored, aspire to
orchestrate a systemic learning process which, while not daunted by the full complexities
of human situations, gives up the idea of 'optimum solutions'?
The first approach implies somehow finding a solution to the problem which has
defeated all the social sciences for a hundred years. Even the shape of such a solution is
not currently apparent. The second approach implies a more realistic but less heroic target.
Those three famous references to science in the Operational Research Society's official
definition might be appropriately re-phrased in terms of working within the spirit of
science.

SUMMARY

In summary, this paper has argued:

?that the 'crisis' debate in O.R. can be examined by studying O.R.'s relation to a systems
approach;
?that O.R., S.E. and S.A. all represent a systematic search for an efficient means of
achieving a defined objective;
?that they are all wedded to logic in specific situations in which the aggregate logic
expressable for the class of situation does not apply;
?that the tradition of systems thinking emerging from organismic biology provides a
systemic approach which gives a new view of the 'crisis' debate;
?that this tradition provides a means of overcoming the theory/practice gap;
?that it can do this by structuring a systemic process of inquiry, rather than by providing
means of optimizing;
?that this sharply defines the 'crisis' in O.R.: continue to aspire to the status of natural
science?and so pass the problems by; or adopt a less heroic aspiration and heal the
theory/practice gap by transferring the notion of systemicity from the situation faced
to the process of inquiry into the problem situation.

REFERENCES
1
Systems Analysis and Operations Research: a Tool for Policy and Program Planning for Developing Countries
(1976) Report of an ad hoc panel, National Academy of Sciences, Washington DC.
2M. R. Dando and P. G. Bennett (1981) A Kuhnian crisis in management science? J. Opl Res. Soc. 32, 94-103.
3S. Beer (1970) Marlow Seventy. O.R. Society, London.
4C: W. Churchman, R. L. Ackoff and E. L. Arnoff (1957) Introduction to Operations Research. Wiley, New
York.
5H. J. Miser (1978) The history, nature and use of Operations Research. In Handbook of Operations Research
(J. J. Moder and S. E. Elmaghraby, Eds), Vol. 1. Van Nostrand Reinhold, New York.
6C. W. Churchman (1979) Paradise regained: a hope for the future of systems design education. In Education
in Systems Science (B. A. Bayraktar et al, Eds). Taylor & Francis, London.
7R. E. Machol (1981) O.R./M.S. in Europe?an American's impressions. In Operational Research '#7,
Proceedings of the Ninth IFORS International Conference on Operational Research (J. P. Brans, Ed.). North
Holland, Amsterdam.
8R. L. Ackoff (1979) The future of operational research is past. J. Opl Res. Soc. 30, 93-104.
9R. L. Ackoff (1979) Resurrecting the future of operational research. J. Opl Res. Soc. 30, 189-199.
10The
King is Dead: Long Live the King, Conference organised by the Operational Research Society, University
of Aston, December 1977.
llT. S. Kuhn (1962, 1970) The Structure of Scientific Revolutions. The University of Chicago Press.
12M. Masterman
(1970) The nature of a paradigm. In Criticism and the Growth of Knowledge (I. I. Lakatos
and A. Musgrave, Eds). Cambridge University Press.
13C. D.
Flagle, W. H. Huggins and R. H. Roy (1960) Operations Research and Systems Engineering. Johns
Hopkins University Press, Baltimore.
14H. J. Miser
(1980) Operations research and systems analysis. Science 209, No. 4452 (July: Centennial Issue),
139-146.
15J. Lesourne
(1981) From operations research to systems analysis: a change in vocabulary or a new conceptual
framework? In Operational Research '81, Proceedings of the Ninth IFORS International Conference on
Operational Research (J. P. Brans, Ed.). North Holland, Amsterdam.
16S. S. Sengupta and R. L. Ackoff
(1965) Systems theory from an operations research point of view. IEEE
Trans Syst. Sci. and Cybern. 1, No. 1.
17F. R, Sagasti and I. I. Mitroff
(1973) Operations research from the viewpoint of general systems theory.
Omega 1, 695-709.
18R. H. Roy
(1960) The development and future of operations research and systems engineering. In Flagle et
al}\

674

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions
 P. Checkland?O.R. and the Systems Movement

19R. Lilienfeld
(1978) The Rise of Systems Theory. Wiley, New York.
20R. Tomlinson
(1981) Some dangerous misconceptions concerning operational research and applied systems
analysis. Eur. J. Opl Res. 7, 203-212.
21D. Klein and P. Butkovich
(1976) Can the profession of operations research/management science change and
survive? Interfaces 6, 47-51.
22P. M. S. Blackett
(1962) Studies of War: Nuclear and Conventional. Oliver & Boyd, Edinburgh.
23J. J. Moder and S. E. Elmaghraby
(Eds) (1978) Handbook of Operations Research: Volumes I and II. Van
Nostrand Reinhold, New York.
24H. M. Wagner
(1975) Principles of Operations Research. Prentice-Hall, London.
25G. Susman and R. D. Evered
(1978) An assessment ofthe scientific merits of action research. Admin. Sci. Q.
23, December.
26M. Hult and S. Lennung
(1980) Towards a definition of action research: a note and bibliography. J. Mgmt.
Stud. 17, 241-250.
27P. B. Checkland (1981) Systems Thinking, Systems Practice. Wiley, Chichester, U.K.
28A. D. Hall
(1962) A Methodology for Systems Engineering. Van Nostrand, Princeton, New Jersey.
29B. L. R. Smith
(1966) The RAND Corporation: Case Study of a Non-profit Advisory Corporation. Harvard
University Press.
30C.J. Hitch (1955) An appreciation of Systems Analysis. In Systems Analysis (S. L. Optner, Ed. 1973). Penguin,
Harmondsworth, U.K.
31R. L. Ackoff
(1974) Redesigning the Future. Wiley, New York.
32H. W. J. Rittel and M. M. Webber
(1974) Dilemmas in a general theory of planning. In Systems and
Management Annual 1974 (R. L. Ackoff, Ed.). Petrocelli, New York.
33C. H. Waddington
(1973) O.R. in World War 2: Operational Research against the U-boat. Elek Science,
London.
34Freeman Dyson, Disturbing the Universe. Harper & Row, London.
35P. B. Checkland
(1972) Systemic, not systemic. (Review of L. von Bertalanffy, General System Theory,
Penguin, 1972) J. Syst. Engng. 3, 149-150.
36P. B. Checkland
(1979) The shape ofthe systems movement. J. appl. Syst. Analysis. 6, 129-135.
37P. B. Checkland (1980) The systems movement and the "failure" of management science. Cyb. Syst. 11,
317-324.
38W. K. D. Davies
(Ed.) (1972) The Conceptual Revolution in Geography. University of London Press.
39G. J. Klir (Ed.) (1972) Trends in General Systems Theory. Wiley, New York.
40W. R. Ashby
(1956) An Introduction to Cybemetics. Chapman & Hall, London.
41P. B. Checkland
(1981) Rethinking a systems approach. J. appl. Syst. Analysis 8, 3-14.
42P. B. Checkland
(1971) A systems map of the Universe. J. syst. Engng 2, 107-114.
43P. B. Checkland and B. Wilson
(1980) 'Primary task' and 'issue-based' root definitions in systems studies.
J. appl. Syst. Analysis 7, 51-54.
44B. Wilson
(1980) The Maltese cross: a tool for information systems analysis and design. J. appl. Syst. Analysis
7, 55-65.
45P. B. Checkland
(1983) The application of systems thinking in real-world problems: the emergence of soft
systems methodology. Presentation to International Symposium on Industrialand Systems Engineering, Mexico
City, March 1983.
46W. E. Moore
(1979) Functionalism. In A History of Sociological Analysis (T. Bottomore and R. Nisbet, Eds).
Heinemann, London.
47R. K. Merton
(1957) Social Theory and Social Structure. Free Press, Glencoe IL.
48L. Kolakowski
(1972) Positivist Philosophy. Penguin Books, Harmondsworth.
49D. Silverman
(1970) The Theory of Organisations. Heinemann, London.
50C. W. Churchman
(1971) The Design of Inquiring Systems. Basic Books, New York.
51G. Vickers (1965) The Art of Judgement. Chapman & Hall, London.
52G. Burrell and G. Morgan
(1979) Sociological Paradigms and Organisational Analysis. Heinemann, London.
53K. H. Wolff
(1979) Phenomenology and sociology. In A History of Sociological Analysis (T. Bottomore and
R. Nisbet, Eds). Heinemann, London.
54T. Luckmann (Ed.) (1978) Phenomenology and Sociology. Penguin, Harmondsworth, U.K.
55D. S. Smyth and P. B. Checkland
(1976) Using a systems approach: the structure of root definitions. J. appl.
syst. Analysis 5, 75-83.
56P. B. Checkland
(1979) Techniques in 'soft' systems practice, Part 2. J. appl. syst. Analysis 6, 41-49.
57S. Beer
(1972) Brain ofthe Firm. Wiley, Chichester (2nd enlarged edition, 1981).
58S. Beer
(1979) The Heart of Enterprise. Wiley, Chichester.
59S. Beer (1959) What has cybemetics to do with O.R.? Operat. Res. Quart. 10, 1-21.
60P. B. Checkland
(1980) Are organisations machines? A review of S. Beer The Heart of Enterprise.5*Futures
12, 421-424.
61R. A. Bryer
(1979) The status of the systems approach. Omega 7, 219-231.
62R. A. Bryer and R. Kistruck
(1976) Systems theory and social science. Working Paper 735/76, Warwick
University, School of Industrial and Business Studies.
63A. Berresford and M. R. Dando
(1978) Operational research for strategic decision making: the role of
world-view. J. Opl Res. Soc. 29, 137-146.
64M. R. Dando and C. Eden
(1977) Reflections on O.R.: a report from the Euro II Congress. Omega 5, 255-270.
65M. Pidd
(1979) Systems approaches and operational research. Eur. J. Opl Res. 3, 13-19.
66C, Eden and S. Jones (1980) Publish or perish-a case study. J. Opl Res. Soc. 31, 131-139.
67R. E. Machol
(1980) Comment on "Publish or perisrT. J. Opl Res. Soc. 31, 1109-1110.
68From the
Operational Research Society's definition of O.R.

675

This content downloaded from 132.203.227.63 on Sun, 25 Oct 2015 05:04:43 UTC
All use subject to JSTOR Terms and Conditions