The Ambiguities of Education Research or
Can a Rabbit Catch the Fox
Tom Heaney
I prepared an earlier version of this paper for the Adult Education Research Conference in 1981. In this paper I suggested a connection between the theory of sets which developed in the field of mathematics and social research. Specifically, I proposed for discussion four postulates which underlie my own conviction that all research--in fact, all explanations of reality--are paradigmatic; that is, that there exist multiple sets in which the criteria and methods of educational research are to be defined.
Social Science: Interpreting the World
In order to understand the world, we first divide it so that we can establish relationships between its various parts. Our difficulty is this task derives from the fact that the world can be divided in unlimited ways, each way influencing both what is seen and what can be seen. In the classic example of two observers, one of whom sees the donut while the other sees the hole, neither sees the whole donut. This division of the world into discrete realities is an inevitable consequence of what we have come to call scientific observation. Perception imposes divisions upon the world which conceptually lead us to the creation of separate, yet simultaneous worlds, each with its own rules for interpretation, each with its own hermeneutic.
Like the figure-ground worlds of Escher, we can only jump from one perceptual world to another by abandoning our earlier perception. On the horizon where birds and fish merge and alternately become the background for the other, we can only perceive the fish by obliterating the birds. Once identified, the objects of perception--the birds, for example--are able to move and the figure-ground separation is complete. This is one, but hardly the only way the world gets divided. Sometimes we observe worlds within worlds, as in the Escher print. Other times we might establish a discrete frame through which to view the world. We inventory all that appears within the frame and thus, by exclusion, identify all that stands outside.
What does all of this have to do with rabbits and foxes, you might ask? Well, as an example of these discrete worlds, we can recall one of the tales of Uncle Remus in which Br'er Fox has captured Br'er Rabbit with a tar baby. The fox deliberated on the fate of his prisoner; first he proposed hanging, then drowning, and still other indescribable tortures. To each proposal the cunning rabbit replied, "I don't care what you do with me, long as you don't throw me in that brier patch." You'll recall that the pleas of the rabbit were finally too much for the fox, who grabbed the rabbit, swung him high in the air, and threw him headlong into the brier patch. Well, the fox waited for the screams of anguish he was certain would follow, only to hear the taunting voice of Br'er Rabbit singing in the distance, "Born and bred in the brier patch, Br'er Fox. I was born and bred in the brier patch" (Harris, 1955, pgs. 12-13).
The point of this story is that there are two kinds of people: those who are born and bred in brier patches and those who are not. This might not seem to be an important distinction to make, but if you're a fox attempting to do in a rabbit, it can be an important distinction indeed. The fox's great mistake was in accepting an interpretation of the brier patch world based on his own, limited, non-brier patch experience. The fox lacked clarity about his perceptual frame and was thus deceived into thinking that the environmental impact of the brier patch on the rabbit would be the same as on other sensitive, furry creatures like himself. Researchers face a similar difficulty when they fail to ask two questions: First, what manner of world am I dealing with--that is, what falls within my frame of reference? Second, are my assumptions and my tools of investigation appropriate to such a world?
Finding a Point of View
In urging the importance of these questions, I am suggesting that research methodologies are "world-specific"--that is, they assume frames of reference which exclude as well as include reality. The point is, in observation we use division in order to interpret what we see, but by dividing the world in one way rather than another we have already imposed an interpretation on the reality observed. According to the uncertainty principle of Werner Heisenberg, the observer of any paradigm is never outside the paradigm observed and, therefore, can never claim objectivity. The observer cannot be separated from the plane of observation. Mathematicians have a word for these discrete planes of observation and the rules appropriate to each; they call them "sets." The theory of sets began, ironically, when Girolamo Saccheri attempted to establish the once-and-for-all validity of Euclidian geometry. He did this in 1733.
Instead, he unwittingly brought the major weakness of Euclid's theories to the attention of the mathematical world. The weakness had to do with Euclid's theorem about parallel lines. According to Euclid, such lines, if extended to infinity, could never intersect. However, it had been the genius of Renaissance artists in their use of perspective to discover that parallel lines do converge on the horizon. It remained for mathematicians after Saccheri to develop non-Euclidian, hyperbolic geometry for the measurement of a non-flat world spinning in curved space (Hofstadter, 1979).
Forgive this apparent digression into mathematics, but it is important to note that even in the reputedly more rigid "hard sciences," there are dichotomous world views, each demanding their own principles of investigation and research. Furthermore, until recently American research in the social sciences in general, and in education in particular, had largely assumed a statistical/mathematical form similar in stature and acceptance to Euclidian geometry prior to the 18th century. The trend toward quantification and complex statistical manipulation in adult education research has been well documented by Gary Matkin (1979), giving the use of a mathematical analog greater relevance.
Four Postulates for Social Inquiry
Self-definition
On what assumptions is a theory of sets advanced in social research? Four postulates are proposed for discussion and further inquiry. The first of these is this: The aim of social research is self-definition. There was a mathematician who once attempted to discredit a bumble bee by demonstrating with sophisticated and precise mathematical arguments that the bee's body mass was too large and wing span too short, and that therefore the bumble bee could not fly. Brought before the tribunal, the bumble bee listened patiently to all the evidence against him, and when he got up to give oral arguments on his own behalf, he didn't say a word. He simply flew away.
The point is, we ultimately judge any explanation or social theory by whether or not that theory "fits," by whether it satisfactorily explains us to ourselves. "Self-definition" is being used here, of course, in a social sense. If a community, a social group, or the whole of humankind cannot find itself and its experiences well expressed and plausibly explained, the theory is rejected as, if not false, at least useless. Such a theory has no utility.
Understand, I am not proposing self-definition as an epistemological principle--as a criterion for "truth"--but as a criterion for relevant and useful social knowledge. However, that which might not have been true in social research is likely to become true once it is accepted as self-definition. Once a theory gains social acceptance, it becomes a self-fulfilling prophecy, and thus we will see, when we get to the third postulate, that social research enters into and becomes an interactive factor in the social world it has chosen to investigate. We have only to look at Marx in the 19th century to see that the function of theory is not limited to explaining the world; theory can change the course of history as well. Social theory tends not to identify the stasis of society, but rather its aims. Social research can be said not so much to identify truth as to precipitate it.
Change
This leads to the second postulate: Change characterizes social reality. There are, I am told, animals whose eyes can only perceive motion. The frog is one example. Spiders and flies are in no danger of becoming the frog's dinner until they move. The social researcher is in a similar perceptual bind. Relationships--the cement of the social order--can only be observed in the dynamic, ever-changing actions of members of that order. It has long been observed that societies never stand still, but are constantly in the process of growth or diminishment or both growth and diminishment in regard to different aspects of the social relationship. It is this motion, this direction, that is the object of social research.
The exception that proves this rule is descriptive or survey research, which I would contend is not research at all. It is merely a partial measurement of social reality at a given place and time, much as a carpenter applies a yard stick to a piece of lumber or a mechanic uses a caliper to determine the bore of a cylinder. Such measurements are not yet research, but might well be a necessary prelude to classical experimental design which inevitably has a before (which is measured) and an after (which is measured) and a change-inducing treatment in between. Whether a survey is an appropriate yardstick we will discuss later. But here it is merely sufficient to note that a survey does not constitute research and the reason is that a survey does not, of itself, account for change. Since change characterizes social reality, surveys do not account for social reality. Research into learning theory, into participation in formal schooling, into the effects of lifelong schooling or into any of the myriad questions of concern to adult educators, is ultimately an attempt to comprehend and explain the rationale for change.
Manipulation and transformation
In going about this business of research, the researcher not merely observes change, but actually induces it. This is the third postulate: All investigative initiatives manipulate and transform reality itself. Philosophers of science have long debated the non-neutrality of observation and, more recently, have specifically begun to question the impact which observation has upon the observed (Bastin, 1978; Hamlyn, 1978). Noted physicist, John Archibald Wheeler, has concluded as a result of work in quantum physics that the eye physically shapes and alters what it sees (Gardner, 1979, p.40). What is being stated here is not that the image of the object is affected by observation or otherwise distorted, but that the object itself is changed by reason of having been observed. This is clearer, perhaps, when observation is mediated by an instrument--a microscope, for example. The usefulness of the microscope to reveal the characteristics of an amebic substance is enhanced not only by perfecting the image-producing characteristics of the lenses, but also by flattening the living matter between two glass slides to bring it into greater conformity with the two-dimensional limitations of the medium of observation. As the fingers alter the surface of an object they explore, so do other research instruments adapt and refine reality to the demands of the tools used.
Survey research achieves a similar adaptation of reality when asking a subject for an opinion on a matter about which s/he has not reflected previously. Being confronted with a question, a respondent is pressured to have an opinion. Even when the option of "no opinion" is provided, the subtle pressure to avoid appearing uninformed or dull frequently remains. Multiplying the number of respondents and aggregating the responses does not eliminate or even minimize the reality-transforming effect of the method. In fact, aggregating discrete date--as in national opinion polls--further transforms the reality under study by proclaiming as social reality opinions which were not the result of a social process, thus creating another form of self-fulfilling prophecy. Research on such polls has clearly demonstrated this "bandwagon" effect (Katz, 1972; Navazio, 1977).
Such a critique of research method is commonplace, and is an example of the transformation of reality that occurs as any researcher pursues "objectivity." This is also to say that we are never neutral observers; our perspective and analytic frame of reference are always shaping the reality we observe, as well as being shaped by that reality. By taking on the stance of a researcher I intrude upon reality, I stand over and against it critically observing and forcing the reality before which I stand to yield its meaning. If you are the object of my study, I could enter the room in which you are now reading this paper. I could find an advantageous position form which to observe your reactions, your tendency to distraction or to doze off, and any opinions you might reveal should another person interrupt your reveries. You might, at first, be vaguely aware that I am watching you. You could choose to ignore this, but my persistent stare would be likely to become a source of irritation. I might observe that you are becoming self-conscious. If you find it important to be judged well by me, you might adopt a posture and role which you feel would reflect well on your intelligence and social decorum. Or perhaps the irritation becomes anger or even fear, in which case I might evoke and thereby observe other dimensions of your personality. The important consideration is that much of what I observe is likely to be the result of my being present to you. I have, in fact, entered into dialogue with you; I receive messages from you, but at the same time I am transmitting messages which impinge on your reality and to some extent alter it. My perception of you is not thereby distorted; it is rather an adequate perception of you as perceived by me.
This does not mean that my perception is "subjective"--at least not in the usual sense of that term. It means that all perception of social reality is intersubjective. I cannot avoid this intersubjectivity by hiding my observation from you, as behind a one-way glass. Even then I have intruded upon your space and time, hidden a portion of your world from you. This artificial, laboratory world I have given you is my word to you, influencing your response, placing you on a glass slide and shaping you to the demands of my microscope. Such is a phenomenological view of social research.
Methodological skewing
Given that the social world is characterized by change and by a direction, and given that research impinges on the world in such a way that the rate or possibly the direction of change is affected, a fourth postulate would follow: Appropriate research method compensates for the manipulative/transformative effects of observation by directing its skew in a direction similar to the commonly, self-defined direction of the observed social reality. An example can be found in the way a film-maker observes a horse race. A stationary movie camera at close range would merely record a blur or possibly nothing at all. So generally the camera will be used in one of two ways. It will be mounted on a vehicle running a parallel course to the horse and approximating the horse's speed, or it will be mounted at a distance so that it can pan the track, maintaining at the point of contact (that is, the horse itself) the same forward motion. Either of these strategies will yield considerable visual information about the race. The researcher faces a similar problem in observing the social milieu, characterized as it is by change. To be sure, change and direction in the social world are not generally physical, as in the case of a horse race. More usually the changes of interest to the researcher concern ideological or paradigmatic shifts, the transformation of values, determinants of status, and thousands of additional conceptual categories--each a partial measure of the social phenomenon. Just as motion can only be observed by approximating--either directly or instrumentally--the motion of the moving object, so do other social measures require an analogous approximation of caring commitment on the part of the researcher. Stationary, neutral observers of society have hazy and blurred vision.
Toward a More Progressive and Participatory Social Science
It follows from these four postulates that social research is paradigmatic--that it evolves out of and has direct applicability to particular historical practices and the systems to which these practices give rise. Each research set derives its values and assumptions from a context which is both historical and concrete. To be sure, paradigms or sets in research overlap each other, as do the historical contexts which give them birth. While these sets are not mutually exclusive, no set is inclusive of all reality and thus all research is limited in its ability to define our social reality.
All research is an attempt to explain the world, and all explanations are paradigmatic, based on a point of view--a frame of reference. While we might be historically or socially determined to use one frame rather than another, choices have been made--if not by us then for us, if not consciously then unconsciously. All explanations are ultimately moral positions--that is, explanations based on certain conscious or unconscious, individual or collective notions of choice and value (Rousseau, 1979). The epistemological question which precedes and informs the selection of any research method seeks evidence that the method is a guarantor of congruence between the "reality" which the research investigates and the codifications which the research hopes to produce. Ideally, the chosen methodology will maximize congruence between reality and codification in order to insure the subsequent congruence between what is known as a result of the research and what we are thereby enabled to do with that knowledge.
In other words, the validation of research method guarantees the congruence between comprehensibility and replicability; what we know as reality we are thereby empowered to recreate. This "cycle of double-congruity" underlies the validation of any progressive research method and is critical in the selection of appropriate tools for the comprehension and codification of experience. Simply stated, the cycle is this: congruence between the phenomenal world and its codification is demonstrated by the extent to which the codification makes possible the predictable manipulation and control of experienced reality. This principle embodies two criteria commonly upheld by researchers: First, valid research produces meaningful results. Second, results are meaningful to the extent that they are verifiable--that is, able to be tested in the world and ultimately make a difference.
This cycle of observation-reflection-action is central to an approach to research which directly involves those from whom knowledge is to be derived and for whom the knowledge is ultimately produced. Research with people, rather than for people--participatory research--is an approach wholly consistent with this understanding of research and with adult education.
References
Bastin, T. "The Effects of Consciousness on the Physical World," in The Re-Evaluation of Existing Values. International Cultural Foundation (New York 1978).
Gardner, M. "Quantum Theory and Quack Theory," New York Review of Books (May 17, 1979, pgs. 39-40).
Hamlyn, D.W. "Perception and Agency," The Monist 61:4 (October 1978).
Harris, J.C. The Complete Tales of Uncle Remus. Houghton Mifflin (Boston 1955).
Hofstadter, D.R. Godel, Escher, Bach: An Eternal Golden Braid. Basic Books (New York 1979).
Katz, D. "Attitude Formation and Public Opinion," in Nimmo, D., and Bonjean, C. (eds.) Political Attitudes and Public Opinion. McKay Publiching (New York 1972).
Matkin, G.W. "Theory, Method, and Appropriateness in Adult Education Research." Unpublished paper presented at Adult Education Research Conference (December 1979).
Navazio, R. "An Experimental Approach to Bandwagon Research," Public Opinion Quarterly 41:1 (Summer 1977, pgs. 217-225).
Rousseau, G.S. "Multiplications," Humanities in Society 2:3 (1979, pgs. 245-272).
Saccheri, G. Euclides ob Omni Naevo Vindicatus. (Milan 1733).
Adult Education Research Conference: 1981
Contact: thea@chicago1.nl.edu
Entered: 20 Jun 95.
