Return to Enquiry Learning home
Return to Research Issues menu
Return to Research Resources menu
Narrative as a Vehicle for Research
Narrative-based research is not just telling stories. The purpose of this paper is to introduce the basic ideas behind it and why it is superior to statistical forms of research for many purposes. In particular, it is appropriate for professionals who want to reflect upon work practices and institutions in order to improve their practice. In this form of research, the self (as agent) is the key 'instrument' of the research. Through the analysis of narratives one can begin to understand the processes of change and development that are integral to personal and professional learning. The paper explores the issues under the following headings:
The basic tools of science are typically assumed to include measurement and the use of statistics. For some scientists, if something cannot be measured, then it is either illusion or impossible to conceive. Thus, in some versions of science the 'real' is also the measurable. If it cannot be measured, then it cannot be real. The most extreme forms of this can be found in the early exponents of what has come to be known as positivism (see reader unit 1). It is clear, however, that there are many experiences which each person would consider real but which do not easily, if at all, transform into measurable entities. There are the experiences of the emotions - love, anger, fear. While certain behaviours associated with such emotions may be observable and hence measurable, the quality of the experience itself resists any attempt at measurement. Most professionals will accept certain qualities associated with their chosen profession such as 'judgement', 'intuition', 'empathy', 'insight' even 'wisdom'. None of these are readily open to measurement. They are often said too subjective. Nevertheless, they are included in what we mean when we say of someone that he or she is a 'good professional'.
Even when we look at 'reality' we do not easily find the words of categories with which to describe it. What ever it is, it is very complex. Although attempts have been made, no one yet has come up with a mathematic model sufficiently sophisticated to tell us what the UK weather will be in detail during the next month. Similarly, there is no mathematic model capable of handling a national let alone a global economy. Nor is there a mathematical model capable of describing the personality, experience and action of a single person.
There is another argument to be made. The very assumptions which make statistical analysis possible are not appropriate for describing, explaining and predicting social and individual processes and experience. The power of statistics is that through its operations generalisations can be made from samples. To do this certain assumptions have to be made about the samples. Cohen (1944:134-5) pointed out that:
In the end, the truth of a generalisation from a sample depends on the homogeneity of the group with respect to which we wish to generalise. A single experiment on a new substance, to test whether it is acid or alkaline, is much more convincing than the result of a questionnaire addressed to millions of army men to measure their intelligence. For the latter is not a simple quality of a uniformly repeatable pattern. In this respect the methods of social statistics are gross compared with refined analysis, so that when our analysis is thoroughgoing, as it generally is in physics, one or two samples are as good as a million. If what we are measuring is really homogeneous, one is sufficient. In the social field, therefore, statistics cannot take the place of analysis ....
In this relatively early statement of the problem from the point of view of a philosopher of logic the issue is already clearly posed that the elements or 'units' of social analysis are not as readily amenable to the logic appropriate to statistical analysis as are the data of physics. Many reasons can be adduced for this. As Schutz (1976) pointed out, atoms do not make decisions or judgements about how or whether to act or not. Additional factors are due to the great complexity of human institutions, action and culture that result in no one individual or social unit being identical to another. Social facts are not stable in the sense of being capable of being rendered a 'pure' or refined substance like sulphuric acid. Rather than relations of identity (involving a one to one matching of features) there are relations of similarity (where a judgement call is made as to whether one member is sufficiently like another to be placed in the same class). The categories employed in social analysis are therefore unlike those employed in the analysis of non-social phenomena. Even a single individual does not remain identical throughout a lifetime. We change, we develop, we forget. Yet when we categorise we do it as if nothing changes.
The process of categorisation can be set out schematically. In the following diagram reality is designated by 'X'. This 'X' cannot fully be encompassed nor comprehended. The only way of thinking about it is as a complex, dynamic, 'something' which is initially undifferentiated by human perception and intelligence. That is, there are no boundaries which parcel up the 'real' into nice packages and boxes, except for the conceptual boundaries, 'packages' and 'boxes' made by people. The 'real' does not force its categories upon us. Rather, we make categories as a way of handling the 'real'. In the following diagram we place categories x1, x2, x3, x4 on the real. We could call this the light of reason at work: like a flood light which is projected out into the night 'capturing' a passing cloud or a plane in a circle of light. Our concepts help us to see entities. Our perceptions are composed by sensation and concept. The one cannot do its work without the other. Our perceptual grasp of 'X' is thus reduced to 'A'. Some of the entities captured perceptually can perhaps be adequately describe through language, say x2, x3 and x4. However, of these, perhaps only one, x3, is measurable. At each stage a reduction is taking place. The statistical reduction, in this analysis, is the most severe of all.
We know reality only through the interpretational categories that we project upon it, or like a net that we cast upon it, hoping to capture something of it. These categories are both our strength and weakness. They provide the power by which we manipulate the world and find ourselves positioned within it. They are also the veil which forever comes between us and the world.
Morgan (1986:132), for example, refers to accountants as "reality constructors":
They can shape the reality of an organisation by persuading others that the interpretive lens provided by the dollar should be given priority in determining the way that organisation is to be run. This, of course, is not to say that financial considerations are unimportant. The point is that thinking about organisation in financial terms is but one way of thinking about organisation. There are always others, and these are usually forced into the background as financial considerations gain a major hold on the definition of organisational reality.
If our experiences of reality are organised through the categories we employ, it is therefore critical that we explore the nature of these categories. These categories exist only as symbols that stand for aspects of the real. If the categories change, then the ways in which reality is perceived changes, in effect, reality for all practical purposes 'changes'.
Kuhn (1970) in writing about changes in scientific understanding and explanations gave many instances of perceptual transformations which lead to what he called 'paradigm' changes. These are scientific revolutions:
Led by a new paradigm, scientists adopt new instruments and look in new places. Even more important, during revolutions scientists see new and different things when looking with familiar instruments in places they have looked before. It is rather as if the professional community had been suddenly transported to another planet where familiar objects are seen in a different light and are joined by unfamiliar ones as well. Of course, nothing of quite tat sort does occur; there is no geographical transplantation; outside the laboratory everyday affairs usually continue as before. Nevertheless, paradigm changes do cause scientists to see the world of their research-engagement differently. In so far as their only recourse to that world is through what they see and do, we may want to say that after a revolution scientists are responding to a different world.
Even in so called 'objective' science, it would seem, the subjective dimension through which perception of the world is constructed is critical. How we categorise and theorise the world affects how we see the world.
It also affects how we work in the world. Handy (1989: 108) tells of an Indian business that he knew 'employing 20,000 people in a big shed in the middle of India making turbine generating equipment'. When he analysed its management structure he understood why it was clogged. There were far too many levels in the decision making process. When asked why this was so they replied that they had been much impressed by British appraisal systems. In order to ensure motivation following a good appraisal then their should be some promotion 'otherwise Indians lose face'. Hence generating many promotional steps was the only answer. There are two paradigms operating in Handy's account. The first is the pyramid-like hierarchy requiring many steps along the promotion ladder. The other, the one that Handy advocates is the 'flat' management structure which is team-like, (the federal organisation) where each team has its own decision making function. Handy calls his vision the shamrock organisation. The one is not compatible with the other. The 'world' operates differently under each form of organisation.
Referring to our diagram above, the Indian company had interpreted the UK appraisal approach through its own interpretational categories which led to the construction of an unwieldy, hierarchical management structure. They 'saw' a structure appropriate to motivation. Using his alternative interpretational categories Handy 'saw' a structure that was arthritic, incapable of making quick decisions.
A statistical or 'positivistic' approach chops experience up into classes and variables. Look at the typical questionnaire. There are questions relating to income, marital status, age, and so on. These are the 'measurable' variables, that is, they can be counted. The results are typically formed into percentages and represented as bar charts or pie charts. Referring back to diagram 1 the information is of the kind that has been reduced to level C. Can one take level C information and reconstruct what could be the 'perceptual grasp' (level A) of any one of these individuals who responded to the questionnaire? Their integrity as living human beings is clearly missing. Their lives, their experience, their qualities as distinct human beings have all been lost in the reduction to bits of information appropriate for statistical analysis. Such information may well be important for certain kinds of decision making. For example, it is important to know whether low levels of formal education relate to low levels of income and employability and whether it is related to poor economic performance. If it is, then a government might well decide to increase educational opportunities. However, if a government wants to know the best way of encouraging people to take educational courses of various kinds and the best ways of delivering these courses in order to produce the best outcomes for economic performance, this kind of information will not help. What is required is information about processes, about the real contexts of peoples lives, how education influences decision making at work, how businesses actually operate and how education may facilitate that operation.
It soon becomes apparent to any social researcher that no one organisation, nor single individual is identical to another. A solution that may work in one context may not work in another, Recall Handy's example of the Indian business. It drew its solution from a different operating context, UK culture and applied it in another context, Indian culture. It did not work. Cultural processes and contextual understandings matter. A method of analysis appropriate to researching process and context is necessary.
In statistics, the unit of analysis is whatever can be measured and what ever can be defined unambiguously. Each member of a given category is assumed to be identical to any other member of the same category. For example, one phial of sulphuric acid is identical to any other However, one human being is not identical to any other. The unit of analysis for the study of human beings in action is no longer the unchanging element that comprises the units for statistical analysis. Nor can it be an element severed from its context. Process and context are both vital dimensions in social and educational research. This means that the basic unit of analysis must have a narrative structure, that is, a temporal ordering where 'this' follows 'that'. The simplest narrative has the strure: "First there was this and then there was that ........". Our Indian example, also reminds us of the importance of context, particularly the cultural context. To understand what is happenning, it is not enough simply to watch. You have to ask people what the different events mean to them and why they are heppening. In asking such questions it quickly becomes clear that reference has to be made to previous occasions when similar things happened, or to whole systems of actions, beliefs and so on. Narratives always refer to other contexts and other narratives.
The narrative analyses involve the identification of 'sequences' (first this, then that) and contexts (this is embedded in that, e.g., individuals are 'embedded' in a social context. Sequences may be be of many different kinds. There may be the cause-effect sequences where a particular stimulus or input is associated with a particular response or output. Or, there may be a particular input but several possible associated outputs. Or, the relationship may be merely a correlation with no necessary causal relationship. The context is of two kinds, symbolic and material. A symbolic context involves all the values, norms, beliefs, linguistic and cultural forms of a given society or group. A material context refers to the resources, the nature of the physical environment, and the built environment of housing, roads and so on. The material context can, of course, have symbolic import just as the symbolic context can have a material form.
Narrative provides a way of organising the complex forms of experience in ways which can be told, which can be recounted and hence made predictable. It is by being able to tell a story again and again that a sense of stability, identity, recognisability and hence action and learning becomes possible. Despite the flux and the change that takes place, a story points to recurrent features and essential structures and processes so that it can be said that the Jo who did 'x' ten years ago is the same Jo who does 'y' today.
Look at Handy's (1989) strategy in writing this book. It is full of anecdotes. These snippets of stories are vehicles for not only a considerable amount of information but also for facilitating understanding, persuading, and learning. Handy has not used his stories in any rigorous manner. Nevertheless they are effective. Why?
It is increasingly being argued that the basic structure of knowledge is narrative (c.f. Danto 1985). Certainly, experimental science is constructed as a series of acts or performances which have to be carried out before a particular result occurs. Knowledge is constructed. To be specific, it is a social construction. For a laboratory to exist there has to be the social, economic, political and material systems that support it. Lab technicians are paid, scientists are educated in schools, colleges and universities, research funds are allocated by government and private sector organisations. There is therefore considerable social organisation that needs to be in place before anything can happen at all. Social science has as one of its objects the analysis and understanding of this narrative organisation of social reality and of scientific enterprises.
The effectiveness of the anecdote is that it echos this basic narrative organisation of experience. Reading Handy we can all either locate similar examples to his or we can draw upon a considerable degree of contextual knowledge which enables us to fill in the missing details in his examples. Take for example, the following extract from a story about Charlie and Jim, now in their 80s who work a farm in East Anglia near where Handy has a small cottage where he goes to do his writing (pp. 82-3):
They still pull the beet by hand for their cattle and slice off the leaves with a sickle as it used to be done 100 years ago. Jim passed by as I was writing.
'Down here for a bit of a holiday, are you?' he said.
'No, I'm working,' I replied, pointing to my papers.
'I'd call that scribbling,' he said with his gentle smile, 'not working.'
Of course, I reflected, the sons of toil have never respected the lily-white hands of the knowledge worker nor known many of them but they say that half the cottages down the lane are now owned by journalists. Perhaps things are changing even here. Then I reflected that until two years ago Jim used to clear his ditches with a scythe. Now young Stephan does it for him, for a fee, with his £20,000 Caterpillar digger, wearing ear muffs with walkman earphones stuffed inside - music while you work in the contractual fringe! Quite a skilled contractual fringe, too. Stephan does his own repair work and is starting a business in spare parts on the side.
Handy told this story as a 'parable' with the meaning that:
Until we look around us with fresh eyes we often don't notice what should be obvious. There were the knowledge workers, the highly capitalised contractual workers, the move to a knowledge industry in a spot which I thought was the unchanging heart of rural England. Open your eyes and ye shall see!
There is, however, more to reading the story than this.
This story is going to make more immediate sense to English readers than people of other nations and in particular to Western European readers than say Asian readers. As a person who also lives in East Anglia I can reconstruct a picture of Charlie and Jim's farm that is close to the typical East Anglian scene. It is flat, low lying and at certain times of the year it is quite bleak with a grey misty kind of monotony. Without this back ground knowledge I might have reconstructed the scene as a hillside farm, or a valley farm or perhaps a hot dry dusty farm. I can also hear something of Jim and Charlie's accent and perhaps construct an image of what they might look like. Fairly or unfairly, I can also locate them into other stories I have heard of rural people.
Indeed, Handy's story inclines the reader in the direction of stereotypical images of rural farmers and their resistance to change. The purpose is to challenge those stereotypes through its parable-like structure. What it also does is draw a crude distinction between 'sons of toil' and the 'knowledge worker'. This categorisation is a key organising devise of his whole book not just this story. it is also a key pedagogical devise, that is, he has a 'teaching' to convey. This teaching is about the way in which organisations are changing due to changes in information technology and the drive towards 'knowledge work'.
Handy is thus subtly positioning his reader. The reader is being positioned to identify with the knowledge-worker not the 'sons of toil'. In particular, by focusing upon the change that is taking place 'beneath the surface' we are being taught to read the everyday occurrences in a particular way, that is 'with fresh eyes' in order to see the changes. It is our perceptual grasp that is to be educated (or, is it re-schooled? see Schostak 1993: 134; 147; 152; 203).
Handy's stories while powerful for pedagogic purposes are limited for research purposes because they do not provide a means of inner challenge. They are so carefully crafted for a particular limited purpose that they do not provide a means of challenging particular interpretations or 'readings'. We need therefore a more systematic approach to developing and employing narratives for research purposes. One such approach is provided by Schostak 1985 which describes some principles underlying the development of narrative case records.
One way of thinking about what is required for the development of narratives appropriate to research purposes is to review the Handy narrative told in the previous section.
It is clear that Handy's story is one-sided. If we ask, whose voice is privileged in this account, it is none other than Handy's. The dialogue extracts for the farmer are chosen for Handy's purposes. To challenge the opposition that Handy creates between the 'sons of toil' and the 'knowledge worker' one would need to interview and observe in detail the actual work carried out by the farmer and the journalists living near by. What would they say? Of course, we can only guess. However, it may be that the clear distinction begins to dissolve if we discover through the interviews that the work of the farm is organised in terms of a highly sophisticated knowledge base concerning soils, fertilisers, growing seasons, pests and so on. It may also be that the journalists talk about the routine toil involved in obtaining and working on 'stories' and 'reports'. What is at issue here is not who is right, but rather that there are different 'realities'. An appropriate narrative should make available the multiple realities that the different actors bring to the 'story'.
Each actor will define his or her reality in a different way. Only by listening to each actor will we find out the underlying concepts through which reality is organised. Morgan (1986) describes the importance of the different kinds of metaphor that can be applied to an organisation. for example, an organisation can be described in terms of being a machine. If the machine metaphor rules, this has consequences for the form of organisation that results:
Mechanistically structured organisations have great difficulty adapting to changing circumstances because they are designed to achieve predetermined goals; they are not designed for innovation. This should come as no surprise, for machines are usually single-purpose mechanisms designed to transform specific inputs into specific outputs and can engage in different activities only if they are explicitly modified or redesigned to do so.
(p. 38)
Other managers within a given company may not employ a machine metaphor but say the metaphor of the organism. With this metaphor the organisation tends to locate itself within an environment, a kind of ecology, to which it seeks to adapt. it is a metaphor that allows change. However, a weakness of the organism metaphor is that in seeking to adapt to an environment it may overlook its role in transforming environments. Thus, the organism metaphor may result in an unnecessary conservativism.
Returning to Handy's story of the farmers and the journalists we need to know much more about how they talk about their experience and organise their work practices and how they define their context of operations, their environment before we can begin to make realistic comparisons. In each case, work has a narrative structure, materials (real or symbolic) that are transformed and so has an effect in the real world. There are three levels which must be explored in any account of such work:
Each of these three levels need to be represented in the development of narratives appropriate for research.
Referring back to figure 1, the task is to identify how each actor engaged in some social practice talks (B) about their perceptual grasp (A) of 'reality (X). If we want to know how an organisation operates, whether it is a small farm or a multinational company, by exploring accounts that key actors give, an increasingly detailed picture of organisational realities and activities can be built up.
Crudely, a work narrative can be defined as 'there is someone (agent) who acts on some materials (work) to produce something (product). This product may not be a physical object like a piece of furniture nor a symbolic product like the 'image' that is created for a car or a can of beer, but may be a new set of social practices, or new ways of doing things. This narrative structure can be represented as follows:
Figure 2
The agent can be thought of as a causative agent, that is, an agent who acts to bring something into effect (an event). There is, of course, no simple relationship between agent and event. Events do not always turn out as people expect, hope or intend. This puts a complicating factor into the narrative. The narrative loses its initial simplicity:
Figure 3
This complicating structure is rather like the plot structure of a novel. The complications can be because of natural phenomena (an earthquake!) or because of intrigues of various kinds (political, status games) or because of various personal reasons (fear, incompetence, particular interests).
The agent is always in relationship with other agents. These relationships in turn may be of many kinds (love, hate, helper, obstacle, competitor). The different agents can be engaged in different kinds of roles (leader, led, employer, employee, member of team, expert, informant, spy ... etc). Each agent may be perceived differently by other agents. Some may provide one impression of themselves but really be playing a secretive game to undermine the work of others. And so on.
The narrative of work can be developed by exploring:
What this results in is effectively a case study (see Reader unit 1). It provides an extremely rich data base capable of being used for a multiplicity of purposes. The major research purpose is the generation of knowledge about practical, social situations and processes in order to improve understanding and the quality of social action. It provides not the fragmented approach of statistical strategies but a holistic approach.
Contact the author at:
John SchostakCohen, M. R. (1944) A Preface to Logic, New York: Dover publications
Danto, A. C. (1985) Narration and Knowledge, New York, Columbia University
Press.
Handy, C. (1989) The Age of Unreason, Arrow Books
Kuhn, T. (1970) The Structure of Scientific Revolutions, (2nd edition), Vols.
I and II. Foundations of the Unity of Science, Chicago: University of Chicago
Press
Morgan, G. (1986) Images of Organisation, London etc: Sage
Schostak, J. F. (1993) Dirty Marks. The Education of Self, Media and Popular
Culture, London, Boulder: Pluto
Schutz, A. (1976) The Phenomenology of the Social World, tr. G. Walsh
and F. Lehnert, London: Heineman