Toward reflexive methodologies
Gergen K. J. & Gergen M. M. (1991) Toward reflexive methodologies. In: Steier F. (ed.) Research and reflexivity. Sage Publications, London: 76–95. Available at http://cepa.info/2752
Table of Contents
Enlightenment without Observation
Expanding Understanding through Relational Reflexivity
Reflexive Elaboration of ‘The Event’
Narration and Image: An Emergent Understanding (by Mary Gergen)
Summary and Conclusions
Scientific practitioners have traditionally ignored the role they play in shaping the outcomes of their research. So long as their personal passions can be suspended, they presume that the rigors of scientific method will more or less guarantee an objective account of the world independent of themselves as observers. Natural scientists, such as Heisenberg, have raised difficult questions concerning the disturbance of the object (such as atomic particles) by the very attempts of scientists to observe their nature. However physicists have responded to Heisenberg’s challenge by developing techniques to convert ‘observer created error’ to a constant – thus allowing systematic correction to be made of observer estimates and restoring the observer-independence of the object. Social scientists have also been concerned with experimenter artifacts and situational demand characteristics as they bias research outcomes (see reviews by Rosenthal and Rosnow, 1966, 1975). However, as in the natural sciences, social science practitioners have attempted to develop safeguards (for example, ‘double blind’ techniques, field experiments, unobtrusive measures) to contain or eradicate such biases. The presumption thus remains broadly shared that with proper caution scientists can safely avoid disfiguring the picture of nature with their own fingerprints.
Yet, in recent years a number of new voices have been registered. And, where traditionalists view ‘observer effects’ as mere annoyances to be subverted, the new voices are more disquieting. For them, there is no means of removing the observer from the production of scientific accounts. There is no means of achieving an ‘observer free’ picture of nature. These new voices are heard across the spectrum of the social sciences. Thus, anthropologists explore how Western research methods and techniques of writing inevitably guide our characterization of other cultures (Clifford and Marcus, 1986; Tyler, 1986, 1987).
Historians question the ways in which Western practices of storytelling (narratology) influence their accounts of the past (White, 1978). Critical theorists and feminists elucidate the means by which value suppositions guide the framing of theory and fact (cf. M. Gergen, 1988; Habermas, 1971; Harding and Hintikka, 1983; Martin, 1987; Reinharz, 1985). And sociologists of knowledge portray how social relationships among scientists determine what comes to be taken as ‘the factual world’ (Kuhn, 1970; Latour and Woolgar, 1979; Knorr-Cetina 1981; Knorr-Cetina and Mulkay, 1983). In none of these cases do the critics find room for scientific accounts of the world independent of scientists themselves. In no case can one separate what is ‘subject’ from ‘object’, ‘knower’ from ‘known’.
If we accept the implications of these various arguments-that accounts of objects are never independent of the observer – what steps lie before us? Are we to dismantle the scientific apparatus, declaring all attempts at ‘objective’, ‘authoritative’ knowledge to be fatuous? Are we to conclude that because we are each locked into our subjectivities we cannot even be certain that there is a ‘world out there’, or that we are truly communicating with other persons? Is there nothing left but to reflect on our own subjectivities, and then to reflect upon the reflection in an infinitude of self-reflexive iterations? These are all dolorous conclusions, indeed, and one would scarcely wish to pursue lines of thought for which these are the inevitable consequences. However, the consequences of obliterating the subject-object dichotomy largely depend on how we understand or interpret the problem. It is our view that if a social constructionist view is taken toward the issues, none of the above conclusions need follow. On the contrary, new vistas of research are opened for exploration.
To appreciate the force of these remarks, let us first consider an alternative to the constructionist orientation, an approach to the subject-object problem typically termed constructivist. For the constructivist, ‘observer biases’ are primarily outcomes of the scientist’s cognitive processes. As it is proposed, scientists bring to their research sites cognitive (or other forms of mental) predispositions, and it is only through these internal means that they can apprehend the world. Because all that is taken as ‘observation’ must necessarily be a product of observational processes within persons, there is no way of separating subject from object. This view was made apparent to many in Kuhn’s (1970) analysis of paradigm shifts in science. As Kuhn put it, changes in the nature of what we take to be knowledge are akin to ’Gestalt shifts’ in perception. One simply comes to see the same phenomenon in a different way. This orientation is elaborated more fully in other works. For example, as von Glasersfeld (1984: 4) states the case,
Knowledge cannot be a commodity that is found ready-made but must be the result of cognizing subjects’ construction What we ordinarily call “reality” is … the reality of the phenomenal, the reality of the relatively durable perceptual and conceptual structures which we manage to establish, use, and maintain in the flow of our actual experience.
Further support for this view is often drawn from the works of Bateson, Maturana, and Varela.
At times the constructivist approach is elided with that of constructionism.[Note 1] However, the differences are substantial and a contrast between the constructivist view of the subject–object dichotomy and the social constructionist stance (cf. Gergen, 1985) proves useful. From this latter perspective, it is not the cognitive processing of the single observer that absorbs the object into itself, but it is language that does so. Accounts of the world (in science and elsewhere) take place within shared systems of intelligibility – usually a spoken or written language. These accounts are not viewed as the external expression of the speaker’s internal processes (such as cognition, intention), but as an expression of relationships among persons. From this viewpoint, it is within social interaction that language is generated, sustained, and abandoned. It is this view that also unifies the multidisciplinary critique discussed earlier. In each case the analysis draws attention to the manner in which conventions of language and other social processes (negotiation, persuasion, power, etc.) influence the accounts rendered of the ‘objective’ world. The emphasis is thus not on the individual mind but on the meanings generated by people as they collectively generate descriptions and explanations in language.
In our view there are distinct advantages of a social constructionist as opposed to a constructivist orientation to the problem of knowledge. At the outset, constructionism does not render science impracticable. Rather, it recognizes the distinct practical achievements of scientific activity, while reconceptualizing the language of science. Rather than defining language as a device for mapping or picturing the world, as traditionally supposed, the constructionist centers on the performative function of language in science. Specifically, this language is essential in coordinating the activities of scientific communities around mutually agreed upon problems (for example, of prediction and control). In this sense what we take to be knowledge is not placed within individual minds; nor is it contained within abstract descriptions and explanations. Rather, from the constructionist standpoint, knowledge (as it is represented in language) is part of the coordinated activities of individuals, which are used to accomplish locally-agreed-upon purposes concerning the real and the good. The focus is thus on inter-dependence rather than independence (See K. Gergen, in press, for further elaboration.)
Regarding the problem of reflexivity in research, a constructivist approach tends to lead inward. For if there is no exit from personal subjectivity, then we find individuals looking back at themselves in an infinite regress of cognitive dispositions. Each attempt at self- understanding (or of understanding one’s research efforts) can only lead to a replication of the same dispositions one attempts to transcend through self-reflection. This is so because a system designed to carry out a specific set of operations (in this case cognitive), can only deal with itself in terms of the same set of operations. In contrast, a social constructionist view invites the investigator outward into the fuller realm of shared languages. The reflexive attempt is thus relational, emphasizing the expansion of the languages of understanding. The aim is to realize more fully the linguistic implications of preferred positions, and to invite the expression of alternative voices or perspectives into one’s activities. In effect, this is to agree with Feyerabend’s (1982: 17) contention that ‘reflexive consideration of various forms of discourse should be a continuous undertaking in which all subcultural enclaves should be invested’.
It is this latter perspective we attempt to explore in the remainder of this chapter. In the work that follows we sketch out three forms of reflexively oriented research that derive specifically from a constructionist perspective. The attempt here is not at all to be inclusive; rather the hope is to open a space for a new range of research endeavors specifically cognizant of the scientist’s enmeshment within a culture of multiple languages. The first sketch speaks primarily to the traditional experimentalist. We argue that not only are experimental attempts at ‘testing’ general theoretical ideas wasteful of effort and resources, but that by engaging in reflexive techniques a scientist can learn more about a theoretical position in a brief period than research can ever ‘demonstrate’. The second sketch expands the focus from the single researcher exploring his or her language capabilities to the possibilities for dialogic expansion of understanding. Here we demonstrate the advantages of allowing a multiplicity of voices to speak to the research issues of concern. In the third and more tentative sketch, we challenge the traditional presumption of a ‘subject to be explored’, and discuss ways in which an investigator may begin with but a provisional stance toward a subject, and progressively elaborate ‘the nature of the problem’ as it is refracted through the intelligibilities of others.
Enlightenment without Observation
The traditional empiricist has always had problems accounting for the development or emergence of theoretical ideas. Ever since Popper’s celebrated The Logic of Scientific Discovery, it has been apparent that no logical means exists for deriving abstract theoretical propositions from observation. Popper himself simply abandoned the problem of theory construction (assigning it to the somewhat mystical domain of ‘context of discovery’), and devoted his major effort to elucidating the logic of justification. Hanson’s Patterns of Discovery represents perhaps the most widely acclaimed attempt to solve the problem of origins. Hanson’s solution was to argue that ‘the facts of the matter’ are largely a product of the categories that the scientist brings to the situation. In effect, theory precedes observation. Yet, similar to the problems confronting constructivism, Hanson cannot give an account of the origin of theoretical categories. For if the facts can tell us only as much as the a priori categories, we are still left with the problem of explaining the origins of the a priori.
Within a social constructionist framework, we locate new vistas for solving the problem of origins. More specifically, we abandon the problem of the origin of ideas within the head, and shift concern to the emergence of language within communities. We may agree with both Hanson and the constructivists that observation does not produce our categories of understanding, and that we bring to the world forms of intelligibility by which the world is made meaningful. However, we differ from the constructivists in that what is imported into the situation is not a ‘state of mind’ but an array of linguistic capacities. These capacities emerge primarily as we acquire the language of the culture. In this sense, scientists approach their problems with a range of linguistic predispositions already at hand. For them to generate understanding is to apply the existing language to the problem at hand. Each must ask ‘What do I already know about the subject in question?’ (Even to name a ‘subject’ is already to suggest a range of associated linguistic practices.) In effect, the scientific investigator must bring to bear language forms acceptable by current standards of intelligibility within the profession.[Note 2]
Consider, for example, the theorist who wishes to ‘understand’ the nature of jealousy. By theoretical understanding generally the scientific establishment means the production of a series of propositions of the ‘if . then’ variety. These propositions should describe the central features of the ‘phenomenon’. For the constructionist, however, jealousy is not a feature of the world independent of a language system. Instead, Jealousy’ is a linguistic integer enmeshed in cultural codes of communicating (and acting). Whether or not one uses the term ‘jealousy’ does not depend on what is ‘actually the case’, but on local conventions of naming or indexing patterns of events. By the same token, the ‘determinants’ or ‘antecedents’ of jealousy are not read from nature, nor derived logically from observation. They, too, are constituents of an elaborate code of intelligibility. Thus, an important step in a reflexively oriented inquiry into jealousy would be to formalize the understandings already contained in the common conventions for talking about jealousy. (See also Smedslund, 1988.)
A variety of local expressions may be combined to produce a definition of jealousy as ‘a negative emotional state generated by the attraction of a loved person to some other person’. From this rudimentary definition one may then through definitional unpacking (rather than empirically) derive the following propositions:
1. The greater the attraction to a loved one, the greater the jealousy upon his or her attraction to another.
2. The greater the attraction of the loved one to another, the greater the jealousy.
The definition also tells us by implication that jealousy is a ‘state of loss’ (or impending loss). By general cultural convention, people who possess much attraction from others need not worry so much about loss. Thus, the additional proposition:
3 People who are loved by few others are likely to experience more jealousy, other things being equal, than those who are loved by many others.
And, given common cultural understandings, people who are low in feelings of being loved are also low in self-esteem. Combining the above formulations, one can thus derive the proposition:
4 People who are low in self-esteem will be more highly prone to jealousy than those who are high in self-esteem.
This last proposition enables the investigator to move from the simple level of definition and homily to something on the order of a ‘theoretical insight’ with important practical implications for interpersonal relationships. It must be realized, however, that the entire formulation is based on an elaboration of definitions already embedded within the cultural code. Such propositions stand intelligible without a single observation of persons in action.
By traditional empiricist standards formalizations such as these have been sufficient to unleash substantial efforts and resources to establish truth or falsity (in Popper’s ‘context of justification ‘). Yet, from a social constructionist perspective such efforts are largely unnecessary. The confirmations (or disconfirmations) of hypotheses through research findings are achieved through social consensus, not through observation of the ‘facts’. The ‘empirical test’ is possible because the conventions of linguistic indexing are so fully shared (‘so commonsensical’) that they appear to ‘reflect’ reality. Thus, for example, we can treat the proposition ‘John was in class this morning’ as empirically verifiable because the meaning of the terms of the proposition are so broadly shared that they seem to be ‘mirrors’ of the world. (With subtle shifts in convention and contextualization, John might declare that his day at the shore was ‘truly instructional’ while the students in class were so bored that they would declare they were ‘on vacation’.) Thus, whether or not John was actually in class depends on what one is willing to call ‘John’, ‘class’, ‘in’ and the like and not on what is given to observation.
Yet, colleagues from the empirical domains might respond, ‘We do learn from observation. Our research findings do not always come out as we anticipated, and we are often forced by such results to reconsider our theoretical assumptions. Further empirical research is thus required.’ From a social constructionist standpoint the experimentalist’s argument is misguided. For it is not empirical research that is required, but a more concerted confrontation with the theory itself. Experimental procedures may furnish experimentalists with a sense of ‘objective hypothesis testing’, but this is primarily because the procedures (measures, settings, instruments, participants and the like) have become so fully saturated by the theoretical language that the scientist virtually ‘sees’ events in these terms. For example, as certain configurations are consistently termed ‘reinforcement’ and ‘response’, investigators come to see the experimental setting in these terms. And once this localized ontology is accepted, it becomes very difficult to reconceptualize the same ‘events’ in other terms, such as ‘enticements’ and ‘voluntary acts’ or more radically ‘moral goods’ or ‘quests for eternity’. Thus, further experimental work does not bear one way or another on the supposed ‘truth value’ of a theory already embraced. Events that seem to disconfirm theoretical relationships simply force the investigator to elucidate or make clear the up-till-then suppressed features of the theoretical forestructure.
From a social constructionist perspective, it should be possible to expand the domain of linguistic implicature by other means – far less costly of time and resources than experimentation. One significant means toward such an end is through what may be termed hypothetical data rotation. As will be described, through this self-reflexive methodology investigators can force themselves to explicate the ‘known but unsaid’. In the first step of this procedure the investigator undertakes the traditional preparatory steps in generating an idea for a laboratory study. Theoretical preferences are singled out, hypotheses are formed, a research design is elaborated, and procedures (setting, subjects, etc.) are envisioned. However, at this juncture the standard research procedure is terminated. No funds are spent on equipment, subjects, data collection, data analysis and the like. Rather, the researchers may lay out the research design and arrange the predicted pattern of results in a matrix. At this point they are poised for the expansion of understanding.
At the outset it may be ventured: Nothing will be learned about one’s theoretical commitment by virtue of a ‘confirmation’ of the hypothesis. If results are as predicted, the experiment simply leaves the investigator in the same theoretical position (or discursive space) as he or she was at the beginning. The theory is expanded and its hidden features explained only when a deviant pattern of results is confronted. However, the effect of a deviant pattern may be achieved conceptually by rotating hypothetical patterns of results through the research design, and at each iteration inquiring into the theoretical implications of the configuration. With each pattern the investigators are forced into reflexive elaboration (theoretical explanation) that reveals the unspoken potentials of the theoretical position.
A rudimentary illustration is useful. Based on the discussion on jealousy, our imaginary investigator decides to ‘test’ the simple hypothesis that persons low in self-esteem are more vulnerable to jealousy than those who are high. A research design is developed in which subjects will work together in pairs on a complex task. During an interim period each will be given information designed (at random) either to raise or lower self-esteem. Before returning to the task, each learns (at random) that his or her partner has opted to leave the pair and join another group. Subjects are then given an opportunity to evaluate their partner. Based on the reasoning developed above the following predictions would be made: Low self-esteem subjects will evaluate the partner more negatively when they decide to leave than when they stay; for high self-esteem subjects there will be no difference in evaluations of the partner. This is shown in diagram form in Figure 5.1.
By traditional standards the investigator would now take to the laboratory. Months of effort might be devoted to collecting and analyzing data. From a constructionist perspective virtually all such efforts are expendable. Should the prediction he verified the investigator can say no more than he or she could at the outset. But now, consider what a hypothetical data rotation might achieve. Here the investigator generates alternative patterns of findings within the matrix and asks, reflexively in each case, for a theoretical accounting.
For example, consider the reverse of the anticipated results, as shown in diagram form in Figure 5.2. Here we find that subjects high in self- esteem evaluate their partners lower when they leave than when they stay. For subjects low in self-esteem, there is no difference. In this case the low self-esteem groups remains unmoved, while the high group shows the ‘jealousy effect’. How can this be explained? The investigator might, for example, consider that low self-esteem is, under certain conditions, depressive – at which point individuals become lethargic and unreactive to social conditions. In contrast, it might be reasoned, high self-esteem carries with it a higher sensitivity to certain forms of rejection. Rejection is considered a form of stupidity. At this point the theoretical position is now rendered more complex. The theoretical account of the reactions of high and low self-esteem persons to jealousy arousal is more richly laminated.
Consider a further rotation of the pattern of results, for example the possibility of ‘no rejection effects’ and only a main effect for self- esteem. Here we find the result shown in Figure 5.3. In effect, high self- esteem persons are more hostile to the partners than the low, regardless of whether they depart or remain. The theoretical forestructure is again forced into articulation. In this case the investigator might propose, for example, that high self-esteem persons look at others in a more negative way than lows; they tend to feel superior. An opposite result might lead to the conclusion that low self-esteem persons look up in admiration to others.
Other rotations of the data should more fully flesh out the hidden assumptions about the nature of self-esteem and jealousy. These views would demand further formalization and synthesis and the rudimentary propositions with which the research began would eventually he replaced by a richly elaborated theoretical superstructure.
High self- esteem
Low self- esteem
Low self- esteem
High self- esteem
Low self- esteem
High self- esteem
Expanding Understanding through Relational Reflexivity
In the foregoing case the strategy for generating understanding capitalized on the potential of the individual to articulate and expand on the linguistic conventions in which he or she is already enmeshed. The reflexive move simply activated the latent language potentials in his or her linguistic community. We must now expand the social horizons. Not only is the single individual limited in terms of his or her favored languages of understanding, but on any given occasion (such as when a theoretical approach is taken) one may be severely limited in the languages that can be accessed. The use of social-dialogic procedures for the generation and the expansion of intelligibility is thus invited.
In traditional scientific enterprises the theoretical meaning of events is almost wholly controlled by the principle investigator. Subjects serve as reactive pawns for manipulation, control or observation. They are not encouraged to reflect on their situations within the study, nor to offer their interpretations of events. They are simply used as vehicles to enhance the power of the investigator’s voice. This voice remains effectively closed. By taking a reflexively dialogic approach to research, a new form of scientific work can be developed. Precedents for this participatory method are found most abundantly in the works of feminist social scientists (see, for example, Belenky et al., 1986; Elden, 1981; Henriques et al., 1984; Hollway, 1989; McNamee, 1988; Roberts, 1981; Wilkinson, 1986). The foremost feature of this type of work is the sharing of power between researchers and subjects in order to construct meaning. ‘Subjects’ become ‘participants’, and the number of interpretations (or theoretical possibilities) generated by the research is expanded rather than frozen.
To illustrate, a study was developed by Mary Gergen on women’s construction of menopause and their images of themselves as women growing older. (See M. Gergen, 1989, for a more extended treatment.) The purpose of the research was to counteract prevailing views in the culture (including the medical profession) by exploring alternative views of menopause and ageing with a group of peers in a dialogic procedure. The topic of menopause was particularly apposite because this process encompasses the major negative stages associated with the passage of women through middle age (such as physical deterioration, loss of function as mothers, loss of beauty, diminishing desire). The major component of the research process was a discussion concerning menopause among eight participants. While M.G. served as moderator, she did not strongly control or direct the major share of the discussion. In this regard she endeavored to reduce her power as researcher over the participants.
The major contribution of the researcher to the discussion was to introduce the notion that menopause was a socially constructed event, serving as a rite of passage for women in our culture. As it was reasoned, the occasion and the events that surrounded this passage are based on a medical model, and as such, are fraught with difficulties and losses that have been part of the whole social construction of menopause. Thus, if women could reframe their understanding of what menopause is, they could avoid certain problems and worries surrounding the event. If the group could come to a more optimistic outlook regarding the construction of menopause, the investigator’s views would be sustained. However, by generating discussion of these views and the process of menopause more generally, the field was open to the generation of possible alternatives.
Participants filled out questionnaires before and after the research, including both open and closed ended questions concerning their views toward menopause. In addition, the discussion procedure was itself recorded. As indicated by the various results, the investigator’s efforts to alter the discourse patterns of these co-participants about the social construction of menopause were only partially successful. While a constructionist view of menopause entered into the discussion, and was revealed on the questionnaires, the dominant theme was otherwise. And too, this alternative perspective, generated through dialogue, proved surprising to the investigator.
As the participants slowly built the case, menopause was not for them the threat that it was to people more generally. Their defense, as they elaborated, lay in their own capacities for control. Through diet, exercise, good habits and the relief of child care, they were going to ‘beat’ the consequences of ageing. In this way they need not reconsider the definition of ageing, perhaps a difficult task in light of broadly held constructions, but could take active, self-directed steps to avoid its consequences. In this way they could also protect themselves against the sad portraits painted for them by the medical profession.
In general the reflexive results of the research included an enrichment of the conception of menopause and ageing – both in terms of what the investigator could communicate to the professional audience more generally, and for the participants themselves. The opportunity to talk about menopause, an almost taboo topic for many women, lifted some of the negativity surrounding it. In effect, by allowing the participants to share in the development of theoretical conceptions, more useful and significant results emerged, for the members of the dialogic circle, and the broader community as well.
Reflexive Elaboration of ‘The Event’
In traditional research it is more or less presumed that there are real- world objects or processes to be elucidated. In the same way that the astronomer explores the solar system, for example, social scientists can study economic processes, drug addiction, and divorce. Yet, from the constructionist perspective terms such as ‘solar system’ and ‘divorce’ are fundamentally constituents of socially shared discourse. It is a mistake to presume that the language is a reflection or map of a reality independent of it. The discourse gains its ‘sense of reality’ as it is used in various social and scientific practices. In this respect, the two preceding exercises in reflexivity occupied a tenuous position: they treated as real for all practical purposes the subjects of ‘jealousy’ and ‘menopause’ while simultaneously viewing them as constructions within discourse. In this third line of reflexive inquiry the tension is broken. The attempt is to launch oneself into discourse around a given set of events and to expand continuously on its meaning through dialogic procedures. With each new lamination of meaning, the attempt is to break the hold of the ‘ostensibly real’ created by the preceding accounts. In effect, the hope is to generate a process of continuous reflexivity, enabling new forms of linguistic reality to emerge, while simultaneously retaining the previous increments in linguistic (and thus social) resources.
To elaborate, in the initial phase the investigator generates a series of tentative interpretations (descriptions, explanations, meanings) of a given range of ‘phenomena’. (To specify what the ‘phenomena’ are, or indeed that they ‘exist independently of the observer’, becomes problematic, for to do so already commits the investigator to a given ontology. Yet, one does play out life within the confines of shared ontologies, and they must be used – even if in quotes – if we are to speak at all.) In the second phase, the investigator opens ‘the phenomenon’ to inspection by another (or others). The interpretations emerging within this dialogue are then used to expand upon or to question the initial, tentative understanding – demonstrating its potentials and limitations. In succeeding phases, still other voices are added to the emerging body of interpretations. Some views may be combined or synthesized, others may open new vistas of understanding, and still others may exist as a critical sub-text. With each new encounter, the hope is to increase the laminations of understanding. While there is no principled endpoint to such research, the attempt is similar in its aim to the earlier study of menopause: to expand and enrich the vocabulary of understanding.
We cannot offer at this juncture a fully elaborated example of reflexive elaboration. However, an initial attempt to ‘feel through’ the contours of possibility has been undertaken by Mary Gergen, and is useful for purposes of stimulating further dialogue.
Narration and Image: An Emergent Understanding (by Mary Gergen)
Over the past ten years Kenneth Gergen and I have been working on the importance of narrative forms within psychology (Gergen and Gergen, 1988, provides a summary). One of the chief propositions we have advanced is that the dominant narratives or stories within the culture control, contain and construct our understandings of personal history. While travelling in South East Asia – encountering much that was new and absorbing – I began to muse about the extent to which narrative forms guide our understanding and appreciation not only of self, but ‘all that we survey’. Even more, I wondered if I could show how our stories form a lorestructure of understanding’ out of which we make sense of the world and thus, in an important sense, determine what we find interesting, exciting, enchanting or forbidding.
Prompted by these under-articulated musings, I developed a goal of taking a series of ‘unstoried’ photographs during a circumscribed period of time; in effect, I wanted to produce a group of pictures without the narrative understanding of the photographer guiding the assemblage of the pictures themselves. Would this not eventuate in a series of pictures without a narrative to support it? Roland Barthes has proposed that ‘pictures are more imperative than writing, they impose meaning at one stroke’ (1972: 110). In contrast, it was my hunch that by taking the narrative form away, photographs would have very little ‘meaning’ at all. Though it is said that a picture is worth a thousand words, I suspected that pictures without words are practically mute in terms of signification. While enthusiasm for the project was high, my rationale was not well developed. Much seemed bubbling to get expressed from my private speech world, but I needed both action and interaction to reveal its possibilities.
I chose the day of September 15, 1988 to generate the photographs. In terms of broad cultural meanings it was a day with a ‘natural story line’, and thus a conservative context in which to explore my views; the story seemed almost built into the day. It was a day in which we were to travel by car from Fraser’s Point, Malaysia, a mountainous resort, down through a jungle road (sometimes crossed by Communist guerillas) to Kuala Lumpur, and then to fly to Bangkok. In effect, I was giving more advantage to the Barthesian view than if I had chosen a day of reading novels or sitting by the seaside.
The schedule for the photographs was confined to 12 hours of daylight, with one shot required at each 20 minute interval (for a total of 36 pictures). The most difficult decision for me was where the focal point of the camera should be. If I took a photo of my feet every 20 minutes, my point would be made, but no one would be interested. So I decided to take whatever seemed pictorially interesting to me at the required moment directly within my field of vision. On the following day I wrote a rendition of the day’s events in a log book. This narrative account could be regarded as the ‘normal’ telling of the day from the photographer’s social standpoint. When again in the home environment – approximately one month later – the film was developed.
I was startled when I saw the photographs for the first time. They seemed tedious, redundant, ugly, simple, bland, and uninformative – just one damn thing after another. It then occurred to me that the photographs now stood totally outside of the story in which I was so deeply embedded at the time, a story of happy times at Fraser’s Point, playing golf and enjoying natural beauty, sadness at leaving our beautiful bungalow, excitement at driving the mysterious road, anxiously watching for guerillas in the jungle, getting lost in frantic traffic jams on the way to the airport, being frustrated at the flight delays, enjoying an Indian meal, relaxing over drinks and news of the Olympics when safe at last on the plane, and looking down from the sky at the lights of a new and exotic city. The story, or ‘what really happened’, had become disconnected from the pictures, leaving them flat and empty. I also asked K.G. to write a description of each photograph on a separate piece of paper and to ‘react’ to the experience. Among his reactions: ‘The photos do bring back a lot of dull moments – driving, packing, flying, etc. – that I would prefer to forget in the desire to make looking backward a fun thing to do (or at least interesting) (because what is there about the past except tellings in the present?)’
Two weeks later a non-academic friend, Sally Lyle, came to visit. She was conscripted as the first independent participant, an uninvolved spectator who had little knowledge of our travels or personal experience in Malaysia. The question was: what kind of sense could she make of the photos? Without any narrative structuring, except the benefit of chronology, what meaning would the pictures have for her? I took notes as she commented on the 36 photographs. The first picture had a portion of a bus in it. She began by trying to create a story: ‘You came here on a tour bus’. I said ‘no’. She then looked at the next several pictures, and described the landscapes as ‘flat, peaceful looking, bucolic’, which surprised me since the photographs had been taken in a valley between mountainous hillsides. In addition, despite the fact that there were some obvious (to me) indicators that this was a series of shots of a golf course, including one of Ken pulling a cart with golf clubs, she did not recognize it as a golf course, nor as Ken playing golf until 8 photos had passed. On number 9 she asked, ‘Is that Ken?’ We seldom play golf, so she did not expect to find us on a golf course, especially in Malaysia. She also did not recognize such things as the inside of a car, an Indian meal (she called it Chinese), and an airport ticket counter (despite the sign ‘British Air’ over it). By picture 20 she said, ‘This is really boring. These are pictures only people who were there could love.’ At photo 34 she asked, ‘Are these supposed to be meaningful?’
In this phase I was able to validate my impression that without a narrative line, the photographs provided little means for ‘making sense’ of the day. What was especially curious, and what I did not know before Sally made her assessment, was how much even the knowledge of what is being pictured is a function of the narrative. It seems that our attention is primarily directed by our prior sharings with others concerning the nature and significance of events. Thus, while it is often held that pictures speak for themselves (for example, a photo of a tourist standing besides the ‘Old Faithful’ geyser at Yellowstone National Park), pictures only speak when we have been embedded within social relationships in which meaning has already been established. (Americans are often exposed to photographs and tellings of travelers at Yellowstone National Park.) Thus we learn to see photographs as parts of stories already shared by members of social communities.
By December, the project had taken on new shape. The enterprise had become self-consciously reflexive. Like a geometric pattern, the photo series had become the point of origin, and each reflexive phase was a loop, circling out and then returning to the core. Each circle took on a new direction – and with each line the rosette became increasingly complex, rich, and interesting. Clearly, the configuring could go on for as long as attention was given. Even in this moment as I write, new lines are circling the point.
In the next phase I approached Brian McGuinness, a British philosopher and biographer. He began to talk about how biographies, which are supposed to be truthful accounts, sometimes are less ‘revealing’ of the lives of the central characters than fictional accounts. Fictions can include private and socially problematic events that the author may always discount as ‘merely fictional’. Describing the life of the poet John Berryman, he revealed how Eileen Simpson, Berryman’s ex-wife, wrote a biography in which many intimate details of their life were excluded. However, in a later novel, The Maze, she exposed significant aspects of their marriage and divorce that were missing from the non-fictional account. Within this conversation the distinction between ‘true story’ narratives and ‘fictional’ narratives become blurred. The ‘truth’ of the story line from our day in Malaysia, seemingly embedded in the photographic ‘account’ became even less credible. From his perspective a short story about the day might do better to shape the day than any supposedly accurate chronology of events. His commentary also suggested that ‘the facts’ out of which history is constructed are analogous to my photographs; they are meaningless without the story forms in which the events are embedded.
In the next phase I discussed the project with Robert Cohen, an historian from the University of Haifa. His views placed a wholly new light on the project. As he pointed out, I had successfully eliminated many of the linguistic elements that provide a narrative line, but I had introduced into the project the strictures set by the camera itself. By the nature of its construction the camera limits what can be recorded; for example, it presents a single point of focus, has a limited angle of exposure, depends on the existing lighting conditions, and reduces the spectrum of color. His point was that I had shaved off one important dimension, narrativity, but I had imposed another: camerativity. Now the implications of Brian’s concern with the difference between biography and fiction became amplified in a different way. For even the camera could not furnish an accurate picture of what there is. Perhaps the very idea of pictures as valid reflections of reality is specious and the same may be said of ‘statements of fact’.
Such thoughts later moved me to explore literary deconstruction theory as it related to the project. For, as the deconstructionists argue, a text is not about the world, but gains its meaning and importance only by virtue of its relations to other texts. Rather than thinking of a photograph as a mirror of nature, as a representation of reality, the most significant relationship is between the photograph and the narrative line. Pictures gain meaning only when they are a part of broader narrative, and the narrative gains its meaning by virtue of the embedded images. At this point, I realized that the focus of my concerns had changed. Where I had initially been interested in the relationship between narratives and our experience of the world. I had now become more involved with the nature of visual and linguistic representation.
On February 6, however, the developing logic now began to circle again toward its beginning. I read Elizabeth Bruss’s (1980) essay, ‘Eye for I: Making and unmaking in film’. From this reading the project again shifted, taking on a broader context. As she writes (1980: 303), ‘Many film theorists have claimed that what is most characteristic about film is precisely its power to constrain human agency, to limit selectivity, temper will, and blunt authority … the automatic undoes the autobiographic’. As Robert had reminded me of the limits of the camera as a machine, Bruss now remarked upon the power of the film itself, especially in its inability to edit itself. The human does not see ‘democratically’. Perhaps our ‘self-reflexivity’ – our discourse about who we are – depends on our expertise in narrativizing. The construction of narratives, and thus the meaning we give to life events, is also the means by which it comes to be represented (in stories, poems, songs, pictures or gestures). These, in turn, influence the future unfolding of lives themselves. Narrative, camera and action all exist in a state of mutual interdependence, as does the dialogue that spices and splices the disparate segments of ‘self-understanding’ together.
What phase is the project now in? The present writing is but one further iteration of its completion. Yet, more might well be done on just this group of photographs alone. A future exposition, with photographs, for example, is sorely needed so that readers can join writers and friends to become authors within the reflexive circle. In principle the spiral knows no boundaries. With socially reflexive research one need never say ‘goodbye’.
Summary and Conclusions
Where have we arrived in this discussion of self-reflexive research strategies? From a social constructionist position, in which language practices are central, and where meaning is derived from social interactions, we have explored several possibilities for expanding on the immediate, taken-for-granted world of the researcher, through reflexive elaboration of the initial language commitment. Discussion was opened on three forms of ‘self-reflexive’ inquiry. In the first, we described how imaginary data configurations could enrich one’s theory more effectively than actual research data. By a hypothetical data matrix rotation, patterns of means, for example, could be revised to suggest new hypotheses about jealousy within social relationships. In the second exploration, a research endeavor with the valuational goal of reducing the agony of menopause among women in their forties was enhanced through self-reflexive methods. In this case dialogic participation was used to increase the range of possible constructions of menopause in society. Self-reflexive methods became interdependently reciprocal, and the term ‘reflexive’ applied not to one researcher, but to relations between and among investigator and research participants. In the final inquiry, the critical import of relational reflexivity was demonstrated as it may shape the very understanding of the ‘domain of study’. A project tentatively examining the effects of narrative meaning and pictorial representation spawned a rich range of new understandings as the project was refracted through the views of others.
This focus on reflexive techniques is intended to expand on the range of research strategies within a social constructionist framework. And it should be clear that the present undertaking is a beginning rather than a considered conclusion. A full demonstration is required to flesh out the possibilities of the initial proposal. A more rigorously systematic inquiry is required to realize the potentials. Further questions of research criteria, standardization, ethical implications, authorship, and practicality all stand open for debate. With such debate the present undertaking itself will benefit from socially reflexive work.
Barthes, Roland (1972) Mythologies, trans. Annette Lavers. New York: Hill & Wang, (Originally published in 1957.)
Belenky, Mary, Clinchy, Blythe M., Goldberger, Nancy R. and Tarule, Jill M. (1986) Women’s Ways of Knowing. New York: Basic Books.
Bruss, Elizabeth W. (1980) ‘Eye for I: Making and unmaking autobiography in film’, in J. Olney (ed.), Autobiography: Essays Theoretical and Critical. Princeton, NJ: Princeton University Press.
Clifford, James and Marcus, George (1986) Writing Culture. Berkeley, CA: University of California Press.
Deaux, Kay (1985) ‘Sex and gender’, Annual Review of Psychology, 36: 49-81.
Elden, Max (1981) ‘Sharing the research work: Participative research and its role demands’, in P. Reason and J. Rowan (eds), Human Inquiry: A Sourcebook of New Paradigm Research. London: Wiley.
Feyerabend, Paul (1982) Science in a Free Society. London: NLB.
Gergen, Kenneth J. (1985) ‘The social constructionist movement in modern psychology’, American Psychologist, 40: 266-75.
Gergen, Kenneth J. (1989) ‘Social psychology and the wrong revolution’, European Journal of Social Psychology, 19: 463-84.
Gergen, Kenneth J. (in press) Construction, Critique and Community. Chicago: University of Chicago Press.
Gergen, Kenneth J. and Gergen, Mary M. (1988) ‘Narrative and the self as relationship’, in L. Berkowitz (ed.), Advances in Experimental Social Psychology, vol. 21. San Diego: Academic Press.
Gergen, Mary M. (ed.) (1988) Feminist Thought and the Structure of Knowledge. New York: New York University Press.
Gergen, Mary M. (1989) ‘Talking about menopause: A dialogic analysis’, in L.E, Thomas (ed.), Research on Adulthood and Aging: The Human Sciences Approach. Albany: SUNY Press.
Habermas, Jiirgen (1971) Knowledge and Human Interests. Boston: Beacon Press.
Hanson, Norwood R. (1958) Patterns of Discovery. London: Cambridge University Press.
Harding, Sandra and Hintikka, Merrill (eds) (1983) Discovering Reality. Dordrecht, Holland: D. Reidel.
Henriques, Julian, Hollway, Wendy, Urwin, Cathy, Venn, Couze and Walkerdine, Valerie (eds) (1984) Changing the Subject: Psychology, Social Regulation, and Subjectivity. London: Methuen.
Hollway, Wendy (1989) Subjectivity and Method in Psychology. London: Sage.
Knorr-Cetina, Karin, D. (1981) The Manufacture of Knowledge. Oxford: Pergamon.
Knorr-Cetina, Karin D. and Mulkay, Michael (1983) Science Observed. Beverley Hills, CA: Sage.
Kuhn, Thomas S. (1970) The Structure of Scientific Revolutions (2nd rev. edn). Chicago: University of Chicago Press. (Originally published 1962.)
Latour, Bruno and Woolgar, Stephen (1979) Laboratory Life: The Social Construction of Scientific Facts. Beverly Hills, CA: Sage.
McNamee, Sheila (1988) ‘Accepting research as social intervention: Implications of a systemic epistemology’, Communication Quarterly. 36: 50-68.
Martin, Emily (1987) The Woman in the Body: A Cultural Analysis of Reproduction. Boston: Beacon Press.
Popper, Karl R. (1968) The Logic of Scientific Discovery. New York: Harper & Row.
Reinharz, Shulamith (1985) ‘Feminist distrust. Problems of context and content in sociological work’, in D. Berg and K. Smith (eds), Exploring Clinical Methods for Social Research. Beverly Hills, CA: Sage.
Roberts, Helen (ed.) (1981) Doing Feminist Research. London: Routledge & Kegan Paul.
Rosenthal, Robert and Rosnow, Ralph (1966) Artifact in Behavioral Research. New York: Academic Press.
Rosenthal, Robert and Rosnow, Ralph (1975) The Volunteer Subject. New York: Wiley-Interscience.
Smedslund, Jan (1988) Psychologic. New York: Springer-Verlag.
Tyler, Stephen (1986) ‘Postmodern ethnography: From document of the occult to occult document’, in J. Clifford and G. Marcus (eds), Writing Culture. Berkeley: University of California Press.
Tyler, Stephen (1987) The Unspeakable. Madison: University of Wisconsin Press.
von Glasersfeld, Ernst (1984) ‘Steps in the construction of “others” and “reality”: A study in self-regulation’. Paper presented at the 7th European Meeting on Cyber‑netics and Systems Research, Vienna, April.
White, Hayden (1978) Tropics of Discourse: Essays in Cultural Criticism. Baltimore: Johns Hopkins University Press.
Wilkinson, Sue (ed.) (1986) Feminist Social Psychology. Milton Keynes, England: Open University Press.
For purposes of conceptual clarity we are making the distinction between constructivism and social constructionism in its strongest form – that is, to demark the boundaries between a wholly cognitive ontology in the first instance and a micro- social one in the second. Many theorists who view themselves as constructivists are indeed concerned with language and social interchange, and many social constructionists are concerned with the cognitive basis of language. At this point there is virtually no important difference to be made between the perspectives. However, to make this form of compromise – including both a subjective level of cognition and a public level of discourse – is not only to recapitulate the traditional dualism of Western culture, but to lose the advantage of the epistemological critique of empiricism (itself based on a dualistic foundation). Further, if closely examined, dualistic views of human action lead to a conceptual impasse (see K. Gergen, 1989).
The critic might again press the question ‘What then are the origins of various discursive practices within a community?’ From the social constructionist standpoint the answer would be in terms of micro-social processes, and would perhaps emphasize the functional bases of language within ongoing relationships.
Found a mistake? Contact corrections/at/cepa.infoDownloaded from http://cepa.info/2752 on 2016-05-28 · Publication curated by Alexander Riegler