[Note: I found this short paper of mine, written almost a quarter-century ago, while rummaging around in some old 3½-inch diskettes. The paper contains an early statement of what I now call transformative social research methodology.
I wrote ithis paper when I was a doctoral student in the US, trying to understand how I could do research and contribute to making the world a better place at the same time--since this was indeed the mission of the university department of which I was a part.]
(Notes written May 24, 1985 for students and faculty at the Department of Social Systems Sciences (S3, or S-cubed) at the Wharton School of the University of Pennsylvania, as I was preparing the proposal for my Ph.D.-dissertation.)
Over the past semester or two, as I have sat through presentations of fellow S-cubers' thesis proposals and have been struggling with my own, I have been led to reflect on the nature of an S-cubed dissertation. Let me say right away that I believe one of the advantages of doing a dissertation in S-cubed is our history of enormous diversity in topics and approaches, and the flexibility and trust demonstrated by our own and outside faculty in accepting such a variety of dissertations.
The reflections I am about to share with you all, for what they are worth, are not intended to reduce that variety. But I think that at the very heart of S-cubery there is an opportunity to do research in a radically new way, a way that is discernible in many dissertations currently in preparation and in publications by S-cubers. In these notes I try to identify, articulate and bring into public debate this opportunity, in the hope they will spark further discussion and clarification of who we are and how we do things.
One of the things about S-cubed that has been the most demanding for me, in the preparation of my thesis proposal and otherwise, is how to think about the relationship between research and planning. Although the department is called Social Systems Sciences, everything we do is ultimately directed toward something mainstream social scientists think has nothing to do with science, namely effecting social and organizational change. This is something which I think unites us all, in spite of our many intellectual, political and stylistic differences, and sets us apart from establishment social science: that we're not content with just trying to study and understand the world, we also want to help change it.
Now, what does that mean when it comes to the dissertation? I guess we all (or at least those of us who haven't been through it yet) have an idea that the dissertation is supposed to be this massive, scientific study of some corner of the world, investigated in accordance with the canon of detached objectivity and preferably identifying if not causal relationships then at least statistically significant correlations between painstakingly quantified and measured variables.
One way of integrating the S-cubed concern for planning into one's disssertation research is to make such a scientific study of a planning intervention, attempting to assess the effect of the intervention by criteria amenable to scientific analysis (quantified performance characteristics, responses to multiple-choice questionnaires, coded interviews with stakeholders, content analysis of the minutes of labor-management meetings, etc.).
This approach keeps the planning and research activities conceptually (but not necessarily temporally) separate. It is reminiscient of the often-heard conceptualization of the relationship between the two that goes something like this: "First we'll go in and do some research on the client's problem, and then we'll do planning."
Another way of combining research and planning in one's dissertation is to do an empirical study of some organization or social field, write up one's analysis of what's going on out there and add a chapter or two where one tries to come up with an idealized design for the system under investigation or at least some suggestions as to how things might improve.
In both of these approaches the distinction is fairly clearcut and not too hard to think about: Here is research (a description/explanation of what is) and here is planning or design (a prescription of what should be). Now I'm a researcher (when I'm trying to understand what's going on out there), and now I'm a designer (when I'm sitting at my desk trying to think up new ways of running the show).
Another group of dissertation writers muddy the waters by deliberately taking on the roles of researcher and change-agent at the same time. This is where things get difficult. How can you maintain the kind of scientific neutrality required for producing knowledge in the form of valid generalizations about the world when your express intention is to actively participate in the shaping of the events you study? How can that end up being anything but introspective stories about your adventures in the big world?
Several major schools of social research, at least one of which is prominent around here (action research in the vein of Argyris and Schon), have addressed and clarified this problem. I, however, would like to examine it in the light of interactive planning. Briefly stated, my argument is that upon closer scrutiny, interactive planning is not only a planning methodology, it is also a methodology for scientific research. To appreciate this potential affinity between planning and science, we must first explore the epistemological bases of each.
For convenience, I will assume that you all agree that naive realism -- the belief that there is a real world out there, ordered in fixed categories independent of our observing minds, a world which we may seek to map or represent as accurately as possible -- is untenable as a basis for scientific inquiry. A time-honored alternative is what we may call "process constructivism", a term expressing the idea that reality is in constant flux, like the flowing of a river, and that our sensible and "manifest" reality is created or constructed out of it, much like vortices and eddies form on the surface of the river.
The realist sees only the vortices, not the flowing movement of water through them, and believes the vortices to be stable objects, constituting a given and unchangeable reality. This is the kind of reality the reactive planner lives in. He takes too many things for granted and believes that problems have ontological status and come in pre-established categories and so on. The interactivist, on the other hand, recognizes the creative nature of reality and wants to play an active role in it, to create social worlds or futures that are more satisfying than the one the reactivist assumes to be given.
In my contribution in the last seminar of West's class a couple of weeks ago (the conversation on intelligence), which I called "Creactivity in Knowledge and Planning," I argued that such a process constructivism is an appropriate epistemology for interactive planning.
If we now take process constructivism to science (which I simply take to mean "the systematic production of knowledge"), what are the implications? Well, it is clear that if we are creating reality as we go along, then to "know" it cannot mean to map it by tiptoeing as close to it as possible without imposing our prejudices or categories on it, since it is by this very "imposition" of concepts or symbol-making activity that reality is created in the first place.
Rather, I suggest that we must understand science as an attempt to give actual form and structure to the activity or process that underlies reality, to throw light on what is hidden from view, to make the implicit explict, to identify and articulate possibilities and potentials, to create or construct the concepts and categories and symbols that make up our human and social world -- in short, science is in the busines of reality construction. A quote from Chris Argyris (Reasoning, Learning and Action, p.469) amplifies this point: " A complete description of reality requires not only a description of the universe as it is but also of its potential for significantly reformulating itself (its potential being part of what it is)."
We see that when equipped with a suitable epistemology, science and planning both turn out to be about constructing realities or creating futures, whichever way we want to put it. Having established that, we need to consider one more point in order to appreciate what it is interactive planning has to offer scientific method.
This point is well-known fact that in the social domain, the theories we hold about the world influences its behavior. For instance, no matter how much economists claim their discipline is an objective science, the assumptions about human nature and social behavior underlying it have subtle, but powerful normative effects on our behavior in society. Given the impossibility of an absolute separation of theorizing scientific observer and theorized-about human subjects, this seeping through of the values implied by the assumptions is an inescapeable condition of the social sciences.
Clearly, many of the assumptions we hold about the social world are self-fulfilling: if we act on the theory that life is nasty, brutish and short, we'll very soon find that that's how it is actually turning out. Knowing that, we might well ask ourselves what point there is in trying to explain the miserable state of the world today by assuming that people are egoistic utilty-optimizers, war-mongering powergrabbers, unthinking automatons, vehicles for selfish genes or whatever misanthropic assumptions are currently in vogue in the social sciences. That is, why take such a negative and pessimistic starting-point if those assumptions have the effect of slowly creating the world they imply? ("Because that's the way things are," the naive realist will say, to which we respond, "No, you create the world, so you better start taking responsibility for your creations, that is, for your assumptions.")
In interactive planning, this pessimistic starting-point and the world resulting from it correspond to the mess formulation. Given that things are as bad as they are now, how bad will they be 10-20-30 years into the future? With the reactivist, the mainstream social scientist will think, "Well, this is the way the world is and where it's going, and there's nothing we can do to change it, other than just react to it (says the planner) or study it (says the scientist)." The interactive planner, on the other hand, goes beyond the picture of the world as it appears to be and constructs an image of what she would like it to be.
It seems to me that the scientist could take a similarly constructive stance. Instead of trying to describe the world as it is, independent of herself (which is impossible anyway because she is actively involved in creating it, whether she recognizes it or not), she could describe the world as it could be, as she would like it to be, and then account for the current state of the world by referring to the characterististics of her "idealized design" that are missing in the current system. In other words, her idealized design would become a kind of negative or indirect "theory" of the present circumstances saying that the system behaves the way it does right now because it does not (yet) have the following characteristics, specified in the "idealized" or "ideal" theory.
Notice how curiously similar this is to natural science, where a theory specifies ideal or "pure" conditions which are then adapted to local circumstances to account for observed behavior. Recall Russ' remarks that the law of gravity is such a powerful law, not because it holds everywhere but because it holds nowhere. To be sure, the social sciences, especially economics, also advance theories in the form of ideal or simplified abstractions, like the law of supply and demand, which then, it is claimed, can be modified to account for the complexities of real-world behavior. But the problem is that the values implied by the assumptions that go into the theory spill over into peoples's understanding of themselves and the world. For example, the seemingly innocent economic axiom that three apples are better than two apples has disastrous consequences for our thinking. To mention just two: it lays in concrete a norm about material acquisitiveness that is currently depleting our resource base, and it leads us to think 3000 nuclear warheads are better than 2000.
Mainstream social scientists never take responsibility for the ethical or moral consequences of their assumptions, believing them to be either inscrutable primitives or appropriate mappings of the world as it is. With the realization that the world it not given, but taken, scientists creating ideal theories, that is, images of possible and desirable future and ways of bringing them about, will be fully responsible for their constructions, and they will presumeably propose assumptions that are maximally conducive to development.
The scientist's claim that her ideal theory does indeed explain the current system is immediately testable. If to "explain" a state of affairs does no longer mean "to describe it as it is", but, in accordance with our constructivist redefinition of science, "to identify its potential for developing into something more desirable", then we can put the ideal theory to the test by ascertaining whether attempts to bring out the identified potential, to realize the postulated opportunities, do indeed lead to development. This can be done in precisely the way suggested by interactive planning: by deriving the specifications for social change implied by the theory and subjecting them to continuous experimentation in concrete social situations and systems.
Arguing that our scientific binoculars be trained at the future rather than at the present is equivalent to asking who cares about the exact details of the present rut if we can find a way out of it? Applying this principle to the present world-wide rush to isolate carcinogens, we may well ask who cares how many thousand substances can be shown to cause cancer, if cancer is simply an organism's perfectly reasonable way of responding to a sick environment? With a minimum of resources (but a considerable change in attitude) we can create circumstances under which cancer would be much less likely to occur in the first place -- a balanced diet, exercise, no undue stresses, more trees. To be sure, lots of scientists already work in preventive medicine trying to identify ways of living that would not induce the kinds of diseases we see today -- but they are working against the norms of medical science and the received wisdom held by those concerned with scientific methodology, and that's the point.
Now, how do we know if our ideal theory is any good? That is, how do we know whether the more desirable state of the world, the potential, not yet actualized reality described by the theory has any relevance to the system under consideration and will help it develop, help it change itself in a direction more satisfying to its stakeholders?
An obvious answer is, "Develop indices of desirability, of satisfaction, of relevance, of reality construction, and measure them as the theory is tried out in concrete social systems. The theory that scores the hightest is the best." This approach may be feasible, but I'm afraid it may create as many problems as it solves. Another, more unusual approach is indicated in the title of a paper Russ wrote some ten years ago, "Does the quality of life have to be quantified?" The paper argues that development does not depend on our ability to quantify it and that there is another way to "measure" it: if the stakeholders don't know right away if their system has developed, it hasn't.
The implication of this for planning is straightforward: no matter what numbers experts come up with in their "scientific" assessment of the results of the planning effort, the stakeholders' judgments, arrived at in whatever way, ought to prevail.
The implication for science, if we take it seriously, is considerably more radical: there is a kind of knowing, a personal, intuitive, gut-level kind of knowing that we may develop and depend on in our assessment of scientific theories. (Of course, in the eyes of old science hands, reliance on such "subjective" modes of knowing immediately disqualifies the whole exercise and renders it opinion, fiction, art, etc. But, as is evident, I'm concerned with expanding the boundaries of what is considered scientific.)
Much has been said in the literature about this qualitative and hard-to-explicate way of acquiring knowledge, and I won't go much into it here. But I do want to make a plea that we take it seriously in our dissertation research and not, as Glaser and Strauss say somewhere, flee into quantifications out of insecurity because we don't trust our personal judgment of our subject-matter and don't believe others will unless we make our research look really "scientific" by sprinkling it with formulae and statistics. I have personally found my doctoral committee to be very willing to accept the kind of approach I'm arguing for here; most of my struggle was about clearing out my own preconceived notions of what "real research" is.
Let me now return to the question of how the notion of an ideal theory can be used in our (dissertation) research. I suggested that interactive planning holds up a promise of a new kind of scientific theory, an "ideal theory", that explains or speaks to the current state of the world not by using pessimistic and inevitably normative assumptions to account for a problematic present, but by inventing images of a desirable future and explaining the present mess as an imperfect and not yet fully developed version of the more ideal future.
As I see it, an ideal theory is different from a regular idealized design in that it is abstract enough to be applicable in many different social and organizational circumstances, yet rich enough in its implications that many fairly specific, lower-level directions for instituting change will be deriveable from it.
Another but related interpretation of what an ideal theory is is that it has more of a process than a content emphasis -- that is, the theory will point in very general terms to the avenues out of a mess available to a system, rather than to a specific goal, a specific future. The task of science would be to identify the ways in which the system could develop in whatever direction its stakeholders saw fit. If we take such procedures or methods useable by a system desiring to develop to be the meaning of the term "ideal theory", then interactive planning itself is an ideal theory, tested for its usefulness every time it is applied in a client's system. The refinement of the ideal theory is the refinement of the process by which systems develop toward more desirable futures.
This, then, is what we could do in our dissertations: create and test ideal theories, explanations of the present couched in terms of the present's transformation into the future. Instead of passively witnessing the inevitable fading away of today's reality and the consequent gradual invalidation of the theories so carefully tailored to it, we can look to the future and try our hand at constructing tomorrow's realities by engaging ourselves and the people we study in a collaborative attempt to give shape and meaning to ongoing social tranformations. Paradoxical as it may sound, this attempt may be considered a "test" of the ideal theory that guides it. The testing of the ideal theory would constitute our research, to be documented as faithfully and self-critically as possible in the written-up dissertation.
These were my thoughts. I hope you found them worth the read.