Journal of Operations Management 27 (2009) 339343

Contents lists available at ScienceDirect

Journal of Operations Management

journal homepage: www.elsevier.com/locate/jom

Too much theory, not enough understanding

Roger W. Schmenner a,*, Luk Van Wassenhove b, Mikko Ketokivi c, Jeff Heyl d, Robert F. Lusch e
a

Indiana University, Kelley School of Business, 801 W. Michigan Street, Indianapolis, IN 46202, United States
Insead, France
Helsinki University of Technology, Finland
d
Lincoln University, New Zealand
e
University of Arizona, USA
b
c

A R T I C L E I N F O

A B S T R A C T

Article history:
Received 15 June 2009
Accepted 15 July 2009
Available online 24 July 2009

This essay and the following commentaries address the use of theory in operations
management. While much is said about theory in the typical journal article, theory, as
science denes it, is not at the center of much of our research. The discipline had fallen into
some bad habits. This essay and its commentaries appeal for more attention to what
theory can mean for our understanding of operations management.
2009 Elsevier B.V. All rights reserved.

Keywords:
Theory
Operations management
Empirical research

Ten years ago, Morgan Swink and I published an article

in JOM entitled On Theory in Operations Management
(Schmenner and Swink, 1998). It was meant to counter the
concerns of many at that time that operations management did not have theories of its own that could ground
our discipline. Now, 10 years on, what is the status of
theory in our eld? Has it advanced? Are we using theory
in productive ways to advance our understanding? How
have our theories changed in response to empirical
investigations? Which theories have been abandoned
and which ones have been developed in their places? I
am afraid that my responses to these questions do not ring
with contentment. It is time to reassess the role of theory in
operations management.

Editors Note: This essay was submitted by Roger Schmenner in early2009. The responding comments were posted by the respective online
OSM forum. We invited Robert F. Lusch and Luk Van Wassenhove to
submit their commentaries in order to enhance the overall discussion.
* Tel.: +1 317 274 2481.
E-mail addresses: rschmenn@iupui.edu (R.W. Schmenner),
Luk.Van-Wassenhove@insead.edu (L.V. Wassenhove), mikko@ketokivi.
(M. Ketokivi), heylj@lincoln.ac.nz (J. Heyl), rlusch@email.arizona.edu
(R.F. Lusch).

0272-6963/$ see front matter 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jom.2009.07.004

1. What is theory and how should it work within a

discipline?
Our paper 10 years ago pointed out a number of
important features concerning the use of theory, many of
which were attributed to Carl Hempel, a renowned
philosopher of science (Hempel, 1965, 1966).
1. Theories explain facts and provide stories as to how
phenomena work the way that they do. They can, and
should, be used to make predictions.
2. Theories are not built; they are invented. That is to say,
theories cannot be systematically constructed or
deduced from facts. Theories require inspiration and
creativity. Facts and the regularities among those facts
may exist for generations before an adequate theory is
invented to account for them.
3. Theories can be disproved by ndings that run counter
to their predictions or explanations. On the other hand,
theories cannot be proved. They can only be supported
by other evidence.
4. The building blocks of understanding are hypotheses and
their tests. Hypotheses do not need to be based on any
theory; they can be mere guesses. When hypotheses are
tested, we gain facts with which we can confront theory.

340

R.W. Schmenner et al. / Journal of Operations Management 27 (2009) 339343

In a number of disciplines, these features of theory and

their use are fundamentally important. Recently, for
example, there was much ado about the newly built,
27 km circumference underground large hadron collider
near the Geneva airport. This expenditure of $8B has been
made specically to provide more facts against which to
evaluate the so-called standard model of sub-atomic
physics. The standard model has been able to handle the
facts as they are currently known, so the entire physics
community has rallied behind this construction of a new
and more powerful accelerator that, it is hoped, can
provide new facts with which to confront the standard
model. It is also hoped that new facts can resolve the
usefulness of string theory as well. String theory is elegant,
to its credit, but there are an increasing number of skeptics
with regard to its usefulness. The physics community is
thus united in the importance of generating new facts and
of showing the weaknesses of prevailing theories with
those facts. The goal is to topple a decient theory (and all
theories can be decient) and to erect a new one that does a
better job with the facts as they are known.
Unfortunately, we in operations management do not
appreciate theory as the physicists do. We do not get
excited about tearing down prevailing theories and
erecting new ones that can handle the facts better. We,
unfortunately, conform to the indictment of behavioral
science by Abraham Kaplan, another leading philosopher
of science. As he put it:
It might well be said that the predicament of behavioral
science is not the absence of theory but its proliferation.
The history of science is undeniably a history of the
successive replacement of poor theories by better ones,
but advances depend on the way in which each takes
account of the achievement of its predecessors. Much of
the theorizing in behavioral science is not building on
what has already been established so much as laying
out new foundations, or even worse, producing only
another set of blueprints. (Kaplan, 1964, p.304)
And, Kaplan argues, behavioral science often has an
unhealthy xation on methodology:
Many behavioral scientists, I am afraid, look to
methodology as a source of salvation: their expectation
is that if only they are willing and obedient, though
their sins are like scarlet they shall be as white as snow.
Methodology is not turned to only as and when specic
methodological difculties arise in the course of
particular inquiries; it is made all encompassing, a
faith in which the tormented inquirer can hope to be
reborn to a new life. If there are such illusions, it has
been my purpose to be disillusioning. In these matters,
the performance of the ritual leaves everything
unchanged, and methodological precepts are likely to
be as ineffective as moral exhortations usually are.
There are indeed techniques to be mastered, and their
resources and limitations are to be thoroughly
explored. But these techniques are specic to their
subject-matters, or to distinctive problems, and the
norms governing their use derive from the contexts of
their application, not from general principles of

methodology. There are behavioral scientists who, in

their desperate search for scientic status, give the
impression that they dont much care what they do if
only they do it right: substance gives way to form. And
here a vicious circle is engendered; when the outcome
is seen to be empty, this is taken as pointing all the more
to the need for a better methodology. The work of the
behavioral scientist might well become methodologically sounder if only he did not try so hard to be so
scientic! (Kaplan, 1964, p.406)
2. What have we done in operations management?
Think now about the typical paper published in operations management. At the risk of straining the archetype,
allow me to identify some key features of such a paper.
1. The paper is an empirical piece where direct inuences
on the variables of interest may be posited but where
indirect, or moderating, effects are often posited as well.
2. Although the piece is empirical it is typically positioned
as useful for building theory, and not simply as a source
of new facts that may need to be explained.
3. The hypotheses to be tested allegedly derive from
theory. The hypotheses are not merely guesses but
predictions from one or another social science theory. A
great literature search has been accomplished to isolate
these hypotheses and to defend them.
4. Great attention is paid to the reliability of the measures
used.
5. Although great attention has been paid to the variables
measured, somehow the effects of signicance to
managers are latent ones and thus structural equation
modeling is engaged in.
6. The empirical results are often mixed. Some of the
hypotheses are supported but others are not. The lack of
support for the theory standing behind these hypotheses,
and for which so much searching in the literature was
done, does not lead to any criticism of the theory nor any
claim that it has been overturned by the empirical results.
What is wrong with this archetypal approach?
a. Theories come, but they never go. The goal, it seems, is
never to knock down the theory that suggests an effect,
even if the empirical results contradict it completely.
The theory is thus never treated as a fully edged theory
that lives and dies with its ability to explain current facts
and to predict new ones. It is a vacuous theory that
shamelessly escapes to be cited another day. While the
typical claim is that the paper is grounded in theory, it
really is not. Theory is on the periphery and not at the
center of the inquiry.
b. The literature search is next to meaningless. The
literature search is done to provide cover for the
researcher and not to isolate an applicable theory and
provide a good test of it.
c. Hypotheses are tied to theory and they need not be. We
do ourselves a disservice by asking authors to base their
hypotheses on some theory or another. By doing so, we
may stie the inspired guess.

R.W. Schmenner et al. / Journal of Operations Management 27 (2009) 339343

d. The ambition of the paper is either limited or

misdirected. Often, the best claim that the paper thinks
it can make is that it is a building block for theory or that
it is a way station towards a theory. Of course, this
loses sight of the fact that theories are invented and not
built. In fact, the paper need not develop a theory itself.
Lots of good science is done without the need to
proliferate theories. In reality, the paper can simply
provide us with new facts. Sometimes those facts can
provide evidence that contradicts an existing theory and
forces us to rethink it. Sometimes it can provide support
for a theory. Other times, it simply provides us with
more facts to ponder. This is worthy enough.
e. The research pulls punches. One of the strengths of
operations management is that it lives in the real world.
We can frequently go out and measure directly lots of
the important aspects about processes of all different
types. We can expect that a lot of the relationships that
govern those processes are direct ones. The need for
moderating inuences or for latent variables is much
diminished in our eld. Labeling something as a latent
variable should be a last resort when we cannot be clever
enough to procure a direct measure. Wherever possible
we should avoid convoluted cause-and-effect or having
to rely on the second-rate power of structural equation
modeling. We would be better off spending more time in
the eld going after clever, effective, direct variables that
can save us from such potentially risky techniques.
3. What should be done?
Several things that can tie us better and more fully to
theory and to the good practice of the philosophy of
science come to mind:
1. Good empirical work does not need to be based on
theory. It can be independent of theory. Some of the
more important touchstones of operations management
came into existence without theory; any theory that
now stands behind them came later. Think, for example,
of just-in-time/lean manufacturing, the bullwhip effect,
cellular manufacturing, factory focus, and much of
scientic management. The useful early work on these
topics was heavy with observation. What we need are
well-reported and well-dened facts to explain.
2. We need to go the extra mile to uncover the most telling
situations and the most appropriate measures. The
Nobel Prize in Physics is often given to the clever
experiment that is stunningly simple in concept. Simple,
straightforward tests are almost universally preferred to
convoluted ones.
3. Have the courage to debunk what our colleagues,
revered as they may be, have done. We need to knock
down more theories and explanations and invent better
ones in their place. We also need to avoid treating
vacuous theory as theory. If a theory is so squishy
that it cannot be invalidated straightforwardly, then we
need to abandon its use. We will not advance our
understanding by using it.
4. We in operations management need to teach our
doctoral students differently and we need to review

341

one anothers papers with less of an eye to methodology

and more of an eye to creativity, insight, and understanding.
The excerpt from Kaplan suggests that we, as a
discipline, should not get carried away by the form of
our research. Methodology is not knowledge. Instead, we
need to focus on what appears to be important to know,
what we understand about it, and what we can do to
understand it better.
4. Commentary: Luk N. Van Wassenhove, Insead
The goal [in physics] is to topple a decient theory
(and all theories can be decient) and to erect a new one
that does a better job with the facts as they are known.
And Much of theorizing in behavioural science is not
building on what has already been established so much as
laying out new foundations, or worse, producing only
another set of blueprints. And Behavioural science often
has an unhealthy xation on methodology.
Strong words but it is unfortunately true that
grounded in theory is often an empty claim. Theory is
indeed frequently on the periphery and not at the centre of
the inquiry.
The author should be commended for ringing the bell
again. It is very sound for a discipline to do some honest
soul-searching at regular intervals, provided it also has the
strength to make the necessary corrections.
Operations management is a broad discipline. It is not a
hard science like physics so it is more difcult to have
theories that can be proved wrong (or incomplete) by new
experiments. We are dealing with business contexts in
operations management and that allows for people to
continuously add new theories without disproving older
ones or even without holding the new theories to the test
of reality. Of course, that should be avoided. Unfortunately,
as the author correctly points out, papers with highly
powered methodology have a high probability of being
published, irrespective of how futile the problem is they
are dealing with. It may indeed be healthier to accept
papers that present new facts that can provide evidence
that contradicts an existing theory or that provide support
for a theory. This is worthy enough. No need for more socalled theories and sophisticated methodologies. After
all, we want a better understanding of the business world
and the latter is constantly changing.
The author is focusing on empirical research which is of
course typical for JOM and a large branch of OM. The same
arguments could be made for analytical OM. There too
technical acrobats deal with slight modications or generalizations of existing models, beating the subject to death and
lacking any relevance to real problem solving. And they get
published. Nothing wrong with this as long as this work is
counter-balanced by some refreshing new problems taken
from actual practice, with an original and simple analysis
providing interesting and useful new insights. Unfortunately, there are few of the latter type analytical papers
being published. Often, their math is too simple. . .
What irritates me enormously is the current political
correctness. Journal editors all require their authors to

342

R.W. Schmenner et al. / Journal of Operations Management 27 (2009) 339343

write a managerial summary and to discuss the managerial relevance of their papers. As a result many papers start
with completely fake descriptions pretending to link the
papers intricate model and analysis to a comment in the
Financial Times or to I have a friend who has an uncle who
works at company X and he once mentioned this problem.
Only to follow immediately with Lets assume, for
analytical tractability, that. . . The assumptions made
are frequently completely unrealistic and unjustied for
the context the authors are pretending to model but they
get away with it. First, there are ethical issues with this
behaviour. It is fundamentally dishonest. Second, by
pretending to solve real problems we are alienating the
few remaining practitioners who still read our technical
journals. They do not buy that stuff since they know the
models are fake and therefore useless to them.
Fortunately, there is a movement to go out and collect
real data and to do empirical analysis to support some of
the assumptions we subsequently make in our analytical
papers. Indeed, as Roger Schmenner states: We would be
better off spending more time in the eld going after
clever, direct variables that can save us from such
potentially risky techniques.
So do we want to live in a fake world where we can fool
ourselves? Does the emperor have clothes or not? Is our
customer uniquely the peer review process or do we also
want to understand the changing world around us and
provide some solid support to practitioners? Do our
journals want to take a more risk prone attitude instead
of an increasingly risk averse stand? Why not publish
controversial papers that challenge existing theories? Let
the market decide.
Finally, why make up problems when the world around
us is full of fascinating and crucially important problems
that beg for some elementary insights? OM and sustainability should be twins. If our discipline cannot guide
managers into how to deal with low carbon economies,
poverty, aging population, access to water, and the like,
what is its legitimacy? We need answers to pressing
problems, not more theories or methodological scrutiny.
The eld is called operations management, not mathematics.
5. Commentary: Mikko Ketokivi, Helsinki University
of Technology
What Schmenner is suggesting, I think, is that in order
to understand a real-life phenomenon or managerial
problem, we may have to leave the connes of existing
paradigms and theories. Real-life problems and puzzling
new phenomena in particular seldom map onto specic
paradigmatic domains and thus, trying to understand a
novel phenomenon using existing paradigms is akin to
trying to play a new game with the old rules.
As long as OM research evolves around theoretical
discourse taking place within paradigms dened by us
academics, our understanding of novel and emerging
phenomena in particular will be hamperedthis is only a
natural consequence. But to shift to a different course
requires a fundamental shift in professional identity and
editorial policy, the enormity of which cannot be exagger-

ated. Paradigms help us make our scientic contributions

explicit. Paradigms help us demonstrate to our peers we
have done something that complements, completes, or
challenges extant research within the paradigm. This is how
I was trained as a doctoral student, and this is how I train
doctoral students now. If there is another way of becoming a
tenured professor in OM, I am not familiar with it:
paradigmatically well-dened discourses are an effective
instrument both in evaluating the novelty and contribution
of ones argument as well as in developing ones research
and publication strategy. I do not think it is an overstatement to say that paradigmatic (theoretical) contributions are in many ways our most important institution.
Contrasting this with Schmenners argument, the situation
does look like the all-too-familiar dilemma: theoriescant
live with them, cant live without them.
Theories are both in OM and the management literature
in general perhaps the most important instrument through
which researchers become to understand one anothers
arguments. Whether theories also help us understand reallife phenomenathat is an entirely different question. To be
sure, it is highly unlikely that theories formulated by
academics automatically coincide with managerial interests. Most of us are literature-driven not problem-driven in
our research.
If we wish to reduce the importance of existing
paradigms and theories in the future, something must
be offered in replacement for a scientic community and in
particular, an effective peer-evaluation process to survive
intact. We cannot just criticize existing notions of theory
without offering feasible alternatives. I do not think we are
anywhere near a realistically operational alternative, but I
did interpret Schmenners essay as an invitation to engage
in rigorous debate on the issue.
6. Commentary: Jeff Heyl, Lincoln University
I rst want to say that I essentially agree with all of the
points you have made in your essay. We do have a awed
approach to theory, attempting to build new wings on
existing structures that should have been long ago
demolished while applying the newest methodologies to
address issues of marginal interest or importance. This
situation was made startling clear to me some two decades
ago when I attended a UK professional conference. The
difference between European research and the US research
to which I was accustomed was dramatic. My observation
at the time was the European research dealt with much
more interesting questions but used weak (or occasionally
inappropriate) methodologies. In other words, the European approach was interesting and relevant. Sadly, I
believe that the globalization of our profession has
conspired to reduce this divergence with the world
adopting the US academic approach. But unfortunately,
Roger, I fear you are tilting at windmills.
I see two reasons why the changes you call for are
unlikely to be implemented. The rst is the increasing
pressure to publish in refereed journals. Blame this on rating
systems, accreditation requirements, or your driver of
choice, but the pressures are deantly increasing. However,
with quantity comes some sacrices. In the business world

R.W. Schmenner et al. / Journal of Operations Management 27 (2009) 339343

one need not look further than Toyota. According to Jim

Womack, much of the responsibility for Toyotas current
troubles derive from their quest for growth (lean.org, 1/1/
09). But as Cervantes wrote, Take away the motive, and you
take away the sin. There are authors out there who care not
in least about their actual contribution, they are only
concerned about the number of articles published. Again
from Cervantes, There are men that will make you books,
and turn them loose into the world, with as much dispatch
as they would do a dish of fritters. So, less pressure to
publish, less questionable work.
The second reason why change is unlikely is the, shall
we say, pride of the author. Most people like to see their
work published and gain some intrinsic reward from the
recognition. No problem with that, but once again from
Cervantes, No fathers or mothers think their own children
ugly; and this self-deceit is yet stronger with respect to the
offspring of the mind. In todays environment the fact of
publication seems more important than the contribution
made, and everyone thinks their work is of signicant
importance!
So I have little hope that the reforms you call for will
actually come to pass. There are severe pressures working
against this proposal, and only the courageous (in the Don
Quixote sense) are likely to be riding their trusty steed
across the countryside in pursuit of worthy targets. It takes
quite an effort to change the course of a river, and I believe
as you do that we have a strong, if misguided one, owing
around us today. But without change, the outlook is not
promising and we will continue to see the same sort of
publications in the future we see today. One nal bit of
Cervantes wisdom from 400 years ago, Never look for
birds of this year in the nests of the last.*****
7. Commentary: understanding the dominant logic,
Robert F. Lusch, University of Arizona
All sciences have a history. This history has a strong
inuence on the dominant logic of the scientic discipline.
A dominant logic captures the central and taken for
granted beliefs, concepts, and lexicon of a discipline; some
refer to it as a world view or a paradigm. Schmenner
argues: we need to focus on what appears to be important
to know, what we understand about it, and what we can do
to understand it better. To accomplish this I suggest a
reassessment of the dominant logic of operations management would be fruitful because the dominant logic
inuences what is important to know and understand.
In the mid-1990s Stephen Vargo and I began an
assessment of the dominant logic of marketing. This
assessment took us back to when the wealth of nations
was determined by the production and exporting of tangible
goods. We observed that relatively recently marketing was
evolving to a new dominant logic; one that was away from
its goods-dominant (G-D) logic that was grounded in
neoclassical economics. This dominant logic quickly became
known as the service-dominant (S-D) logic. G-D logic
focused on the maximization of prot from the sale of
output (either tangible or intangible products) whereas S-D
logic focused on the process of serving or using one actors
resources for the benet of another and how over time

343

actors could learn to serve each other better. Thus the focus
was away from an output and maximization logic to a
process and learning logic. Under G-D logic the rm and the
customer were separate; the rm produced value and the
customer consumed and destroyed value. However, with SD logic they were co-creators of value in a value network.
Ten foundational premises capture (S-D) logic (Vargo
and Lusch, 2004, 2008) and are developed in over 20
publications (www.sdlogic.net). Premises are similar to
axioms and are assumed true and used for theory building
in a deductive or inductive manner. S-D logic comprises 10
foundational premises:
 FP1. Service is the fundamental basis of exchange.
 FP2. Indirect exchange masks the fundamental basis of
exchange.
 FP3. Goods are distribution mechanisms for service
provision.
 FP4. Operant resources the fundamental source of
competitive advantage.
 FP5. All economies are service economies.
 FP6. The customer is always a co-creator of value.
 FP7. The enterprise cannot deliver value, but only offer
value propositions.
 FP8. A service-centered view is inherently customer
oriented and relational.
 FP9. All economic and social actors are resource
integrators.
 FP10. Value is always uniquely and phenomenologically
determined by the beneciary.
S-D logic has not only been recognized as a signicant
advance in our understanding of the basic nature and scope
of marketing and markets and what is important to know, it
has also been recognized by Jim Spohrer, Director of the IBM
Almaden Lab Services Research program, as the basis of the
conceptual foundation for service science engineering and
management (SSME). Furthermore, Robert Ford and David
Bowen, two management scholars, argue that the discipline
of management needs to adopt a service-dominant perspective. Perhaps at least some of S-D logic may be useful in
reframing the dominant logic of operations management
since we are increasingly nding that S-D logic applies
broadly to understanding business and society. From the
outset we argued that we do not own S-D logic and that it is
an open-source collaborative effort to co-create a new
dominant logic and thus we hope scholars in the operations
management eld will join this effort.
References
Hempel, C.G., 1965. Aspects of Scientic Explanation. Free Press, New
York, NY.
Hempel, C.G., 1966. Philosophy of Natural Science. Prentice-Hall, Inc.,
Englewood Cliffs, NJ.
Kaplan, A., 1964. The Conduct of Inquiry: Methodology for Behavioral
Science. Chandler Publishing Company, San Francisco.
Schmenner, R.W., Swink, M.L., 1998. On theory in operations management. Journal of Operations Management 17 (December (1)), 97113
(Elsevier).
Vargo, S.L., Lusch, R.F., 2004. Evolving to a new dominant logic for
marketing. Journal of Marketing 68 (January), 117.
Vargo, S.L., Lusch, R.F., 2008. Service-dominant logic: continuing the evolution. Journal of the Academy of Marketing Science 36 (Spring), 110.