Last changed 9 Sept 2006 ............... Length about 6,000 words (39,000 bytes).
(This document started on 23 Aug 2006.) This is a WWW document maintained by Steve Draper, installed at You may copy it. How to refer to it.

Web site logical path: [] [~steve] [this page]

Narrative pedagogy

by Steve Draper


These are very rough notes I mustn't spend any more time on, but wanted available for friends. They were triggered by the NILE06 conference, and the suggestion of a multi-author paper in j.ILEs on "towards a narrative pedagogy".

Part A: my ideas, especially of what do listeners, and separately, what to tellers, value in stories.
Part B: I offer notes from Paul Hazel's review
Part C: notes on applications to learning and teaching especially in HE.

Contents (click to jump to a section)

  • Part A. My base notes
  • Part B. Paul Hazel: ideas from the literature
  • Part C. Two modes of thought
  • Part D. What is not a story
  • Part E. Practical applications
  • Some comments from Paul Hazel
  • References

    Part A. My base notes


    Lots of people think "narrative" can help develop new and effective ways of teaching and learning. What could we say theoretically about this?

    In what follows, I'll put down ten or so different points that are about all I can think of about narrative pedagogy.

    I (unlike Rose Luckin) will use "pedagogy" here sloppily to mean instructional methods and the theory underpinning them, assuming that in the end there isn't enough difference in androgogy etc. i.e. teaching well is the general problem, regardless of whether the learner is 40, 18, 14, or 8 years old.

    On the other hand I think the differences in what people seem to mean by "narrative" are too important to pass by: that is my first discussion point.

    What do we mean by narrative?

    Even in the literature related to learning and teaching, never mind the wider literature, the term "narrative" is used to mean things so far apart as to undermine the logic of many arguments used: i.e. it is too extreme to ignore. Here I'll sketch four meanings A-D, that can be ordered on a spectrum, but for convenient contrast I'll introduce in another order.

    NarrativeA: Everyday stories. I once heard a talk by Doris Lessing in which she mentioned that she'd become fascinated by the stories ordinary people tell every day, that you can overhear in cafes, listen to in family homes. This is narrative as a universal human activity: something everyone can and does do. This fits with grand claims of narrative as a defining human capacity and activity. These are stories viewed as an author's (teller's) behaviour. (N.B. Labov & Waletzky (1997) define it like this, but qualified as oral versions of personal experiences: "We can now define quite simply those sequences of clauses which we will consider as narratives. Any sequence of clauses which contains at least one temporal juncture is a narrative.")

    NarrativeD: High culture. Persistent narratives. Some stories have come down to us from thousands of years ago, and are still retold widely. Of course only an infinitesimal fraction of stories persist in this way: most are told only once or twice then forgotten. We could ask, as students of culture and literature should and do, what makes a story persist. Closer to our time, we can ask what makes one soap opera attract an audience of tens of millions, but another only millions. This leads to theories of story structure. This is a theory from the listeners' viewpoint: what makes stories so successful that they attract people to listen, live on and are retold again and again. This does not tell us anything about a universal human capacity for story telling: most stories, most tellers, are not successful and remembered in this way, although it could tell us something about universal properties of human audiences for narrative.

    NarrativeC: Persuasive narrative. Some narratives are used to persuade other people: for instance in law courts, or when a child tries to persuade their parents that some mishap was not their fault. However just because persuasion is a universal human activity, and narrative is widely used for persuasion, does not mean that persuasion is an essential aspect of it. On the contrary, at least since the end of the middle ages, it has been clear in our culture that the best literature does NOT have a crude moral or point to each story, but in fact has the opposite property, that the best stories mean different things to different listeners, and only the third rate stories have a single meaning apparent to all. [Preface to 2nd edition of Doris Lessing's Golden Notebook] The same point can be made of the scientific literature: that while many papers argue a conclusion, science cumulates knowledge exactly because a reported observation can mean something different to a later reader than it meant to the original observer: that observation and interpretation are separated in a productive way, and that scientific papers are not mere rhetoric i.e. about persuading the audience to adopt the author's views.

    Narrative B: Recounting specific true cases. In reflecting on what makes science fascinating, an important element is often specific cases that relate the theory to our experience. But is a case study or application or instance, always a narrative? Here are some examples of this category. [Lisa] Using the loss of the submarine Kursk as a framework for some chemistry lessons on how long the air would last for the trapped crew. Seeing that stainless steel screws put into an aluminium engine block on a motor cycle caused corrosion. In a lecture full of demonstrations of explosions, one of the loudest came from perhaps half a teaspoon of water sealed in a glass tube, hung over a candle flame, and linking that to the assertion that the biggest explosion in human history was not a nuclear bomb, but the explosive boiling of seawater during the volcanic eruption of Krakatoa. (N.B. current geology views of Krakatoa are now somewhat different from this.) In a Royal Institution christmas lecture on geology, a still photo of some gold mine workers linked to the statement that the deepest mines in existence are so far below ground that the tunnel walls are at 70 degrees Celsius: hot enough to burn you, and such that a failure of the ventilation and cooling system would kill the miners horribly long before they could walk out. In a physiology course, the lecturer puts a glass of water and a drinking straw on the bench, then performs a handstand on the bench and drinks the water, thus demonstrating that swallowing does not depend upon gravity but on peristalsis.

    These are all vivid episodes that have stuck in my mind. They link the underlying science to particular cases. But are they narratives? Since we are interested in pedagogy, we should look at what they have in common: but it isn't clear that it is narrative in the sense of most of the literature on narrative. They are all about the particular (the episodic). They are all vivid in some sense. They do not all depend really on human emotion or motives, nor on a plot with a difficulty and a resolution. And unlike stories it is an essential aspect of these that they are actually true and believed to be so by the audience. The deep mine example is interesting in that the Royal Institution prides itself on highly engaging lectures with elaborate demonstrations and lots of audience participation. However this episode that stuck in my mind relied only on a still photo: it reminds us that both narrative in general and effective science teaching rely on words and the imagination, not images: i.e. just as constructivism says, it is the hidden and personal mental constructions of the individual learner that are the locus of learning not high tech. manipulation by mulitmedia. The peristalsis case has elements of high showmanship: undergraduates don't think lecturers are likely to be capable of doing a handstand and they certainly don't expect them to do one in a lecture: this is vivid and memorable. But it is not mere entertainment, not even a mere illustration, but direct experimental proof that swallowing does not depend on gravity but on some invisible internal mechanism. Its important features for educational effect are far more than generic features of narrative.

    Another of the difficulties in discussions of "narrative" is that there is usually no attempt to distinguish features of narrative from features of any communication system that could exist, whether or not is used for narrative. For instance the story grammar work by Rumelhart pointed out that simple stories are essentially accounts of how the event recounted differed from what you would otherwise assume as the normal case. In other words, they work by assuming a lot of common knowledge between teller and audience, and the telling points out only the unusual. But all communication is in fact of this kind. In information theory, information itself is defined as differences from what the audience already knows or expects. It is true that this is worth pointing out because poor story tellers, and in general poor communicators, often fail to act on this, and say large amounts of stuff that is either irrelevant or what the hearer already knows: but this is not a property of narrative, but of communication. Furthermore it is not a property of humans nor of human languages, but of all conceivable agents that communicate.

    Just as the issue of the complication being the heart of a story is really a feature of any possible communication, so we should ask similar questions about theory of mind, and suspension of disbelief. Famously, narrative requires the suspension of disbelief. But this could be seen as just a form of a general "reason maintainence system" from Artificial Intelligence. Any attempt to model a mind that can deal with the way that new information may challenge its existing beliefs requires not just a simple database of what is true and false, but instead the maintainence of beliefs as supported by some things and challenged by others. This of course is the apparatus underlying "critical thought", so suspension of disbelief is also a function necessary for minds whether human or not, not a special property of narratives. Similarly stories often require the use of a "theory of mind" capability: the ability of an agent to track the fact that other agents may have different information from them. If John leaves the room, and you move a toy from one drawer to another, when he returns an agent needs to deal with the fact that the agent knows the toy is in one place while John will believe it is in another. Autistic people are thought to lack this, but any agent (not necessarily human) that can deal with other agents needs this; and we also need it for most stories which routinely deal with differences in what different characters know, and the way this leads them to act in different ways.

    Bruner (2002) has a central argument that is incoherent. It makes grand claims of universality that are sustainable for narrativeA, but studies only very unusual stories of type C and D, that the vast majority of people never create themselves, and a majority may not even ever hear. The same kind of criticism applies to many other theories about narrative. Labov & Waletzky (1967/1997) assert that all narratives have 6 structural features, and went to the trouble of moving away from literature to oral stories. But as they say, not all stories exhibit all of these features. So this is not a universal structural truth like the assertion that all humans have one thigh bone and two bones in the lower leg, but an idea about the functions needed for narrative use, but which may be performed by other aspects of the context and not by the words of the narrative itself.

    I return to what a narrative is and is not in Part D.

    The two natural modes of learning and teaching

    Willie Dunn once told me that in his view there were two natural modes of teaching and learning. One is what I would call "learning by exploration", where the teacher is at most in the background, and the learner tries things out in multiple ways. The other he called narrative, or story telling.

    One of the things that critics of the use of lectures usually choose to forget is that story telling seems to be a pervasive and natural human activity, and that perhaps we should see lectures as an attempt to harness that for pedagogy. On the other hand, stories have features that are perhaps not good for academic learning. (1) They essentially revolve around suspension of disbelief. People do not process them primarily around truth: we just monitor internal coherence/consistency. The classic problem with science teaching and learning is that many learners learn the "story" without relating it to their own experience, happily maintaining what to scientists are huge inconsistencies between the science story and their beliefs about everyday reality. This is a central property of stories: they frequently aren't true, and their truth doesn't even need to be considered to understand them i.e. their truth doesn't matter. (2) And they needn't matter exactly because they are essentially disconnected from personal action by the audience (in contrast to learning by example which is intrinsically and essentially connected).

    Should we teach narrative?

    In "Mindstorms", Papert in passing makes a profound point about learning. He points out that how people learn a second language is quite different from how they learn their first language. Almost everyone learns a first language, but they do so without any formal teaching, and without any intention to learn, while second languages generally are learned in a conscious, effortful way that requires teachers, and that many fail at. This is still more markedly the case for maths: Papert equates Piagetian conservations as deep maths that is learned unconsciously, unintentionally by almost every child, without teaching. In contrast school maths leaves a majority of learners convinced they can't do maths.

    So just because listening to, and telling, stories is an almost universal ability and practice, it does not follow that we should attempt to teach it explicitly nor that we can assume that any such teaching to work for most learners.

    Why do people tell stories? remember stories?

    Why do people tell stories, want to tell a particular story?

    Of course, as social beings, one big influence is the effect on the hearers. We tell the police stories designed to show we are innocent, fellow gang members stories designed to show we are bad, a child a story to soothe them to sleep, a paying audience a story to surprise and entertain them, for an audience of girls we don't pick a story of a football match or a race car, for an audience of boys we don't pick a story of celebrity dieting. For most audiences we won't tell a story of ordinary sex because it's too intimate, nor of finding a decaying child's body by the canal, nor of the smell of an aircrash site because it's too upsetting. The many constraints also make most of us less likely to tell a story at all to relative strangers than to close friends.

    But when we are among friends, why do we tell stories? The teller often finds it satisfying. The institution of counselling and perhaps psychotherapy is about creating a situation where the teller need not consider the audience's preferences at all, and so maximise the benefits for the teller. That is, besides benefits for an audience, telling a story seems to have independent value for the teller.

    My suggestion is that we tell stories until we feel we understand them, and no longer need to remember the episode. Why remember incidental circumstances if we are confident that a general rule covers the only important lesson? Telling a story helps us to digest it, and most incidents that weren't instantly assimilated at the time are digested in a few tellings. This idea is consistent with the cognitive psychology distinction between episodic and semantic memory; with the observation that most stories e.g. of what we did today in the office, are only told once or twice then forgotten by both teller and audience, although a few stay in the mind and might be told years later; with the way the most admired stories in literature are those that seem to mean different things to different people, and to throw up new meanings when the same person re-reads them. It is also consistent with the idea that science is based on induction: on collecting cases which at a later stage will be subsumed into a general law or concept that describes them all. Before the rule emerges, what we have is instances, stories, that represent important and so far problematic cases. Because we don't know yet which aspects of them are important, we preserve many details.

    I made this point by asking what value a story has for the teller independently of any value for or effect on the audience. But similarly, audiences remember and perhaps re-tell stories independently of any purpose the teller may originally have had. In both cases, the point here is to get away from the observation that stories may be used to try to influence others, as means to personal ends, a method of social action; and instead to keep in mind that they also have a value (in fact two values) independent of that.

    Accounts of ourselves and others

    Another common (but not essential or universal) aspect of story is the accounts we give of ourselves. Explaining ourselves to our fellow beings. One part of Social Psychology sees this as central to human activity. And this is probably closely related to the way stories are associated with people's sense of identity, to origin myths, to creation myths, even to phrases like "What's your story?". This is not (directly) rhetorical persuasion but is about how stories are very important in affecting and changing our relationship with other people. Like persuasion, it is a common motive behind story telling, but in the end is one instrumental reason for story telling, not an intrinsic aspect of all stories.

    Our dealings with other people also have the aspects of partial knowledge characteristic of stories: taking them as they are or rather as we can perceive them rather than as a full knowledge of them might show them. That is, our view of them has internal coherence (putting together all the little we know of them), without (perhaps) external truth and/or coherence. Alan Bennett's "The Lady in the van" illustrates this particularly well. And part of that masterpiece is how in the second part of it, after many years of interaction, a much different and fuller picture of her (the lady in the van) comes into his (Bennett's) ken: and that is both deeply interesting as an addition to the story, yet also was so clearly not necessary, NOT a resolution to the complication (in narrative theory jargon).


    Part B. Paul Hazel: ideas from the literature

    Read Paul Hazel (2006), which contains both a review of ideas about narrative, and some proposals for applying them to teaching.

    Labov & Waletzky (1967) assert there is a universal set of functions needed for narratives: but these are unlike, say, the bones in a human body, the elements are not always there. These are functions not structures: what a listener needs somehow, that must be supplied somehow, but not necessarily in the story as opposed to some other way (e.g. they may be common knowledge) and not necessarily in that order. Does not apply to news stories.

    1. Abstract. Greek prologue. What is told, why it is told.
    2. Orientation. Setting the scene if necessary: person, place, time, behavioural situation.
    3. Complication. The main business.
    4. Evaluation: the attitude of the teller to the events
    5. Resolution. The rest of the narrative, if any, after the evaluation.
    6. Coda. Transition back to the present from the story. E.g. "And they lived happily ever after", "And I've never been back", "And that was that".


    Mishler (attributing it to Katz, as reported by Hazel):
    If we take an academic narrative (case study, anecdote), can it be scholarly? Test it by four evidentiary questions: Hazel is suggesting using these as assessment criteria for student stories.

    Hounsell for history essays, but perhaps for science papers as well?:

    1. All have 3 core elements: data, organisation, interpretation.
    2. At least 3 essay types, that configure these differently:
      • Argument: as in Perry, supports the interpretation from the data.
      • Arrangement: interpretation not founded on the data and organisation.
      • Viewpoint: data not related to viewpoint and organisation. Merely subjective?

    It seems to me, that an empirical scientific paper focuses on the data, and the careful organisation of it, and squeezes down the interpretation.

    But it also makes me think about alternative types of ways to present material. E.g. tree structured bottom up logic (rather than narrative sequence). Narrative as a type of binding, but others are possible especially since we are trying to convey NOT episodic info BUT semantic.

    Part C. Two modes of thought

    I've come across a set of comparable but probably conflicting ideas, each a version of a claim that there exist two contrasting modes of thought. The contrast is between what is taken as true, and what is taken as a story which may or may not be true.

    Jerome Bruner takes as a deep duality or contrast the mental world of "paradigmatic" thought (basically what we take to be factual reality) and narrative: the world of possibility, imagination, and potential. He seems to say humans have these two different modes of thought.

    My version of this might be to say that humans, or in fact any agent capable of planned action, needs two modes of thought. One is the mode of planning and selecting goals: this requires imagining possible, but not actual, ways the world could be, just as fictional stories do. Even commonplace plans such as what you might have for dinner today are nevertheless possible (or so they seem to you), not actual. The other is the the facts about the actual world, on which many details of plans depend: e.g. what you have in the fridge, what the supermarket has in stock today, and so on. We are so familiar with this it doesn't sound like an insight worth noticing. Yet there are occasions when people get into trouble by not distinguishing them: when someone has set their heart on an imagined event such as their daughter getting married, getting the job they have planned on, etc. they can have trouble believing it hasn't happened. In fact the whole tendency to shoot the messenger, of denial being a common first reaction to sudden bereavement, are versions of confusing the two modes.

    Another kind of trouble from failing to maintain the distinction occurs when ethics committees, but still more often complaints they receive, exhibit crude misconceptions about the impropriety of controlled human experiments, basically arguing that since we now know that the treatment works it was then immoral not to give it to the control group. I suppose we are so imbued with the idea of truth being eternal, that we often have trouble reasoning about how our knowledge of eternal truth is however itself time dependent.

    A third version of the contrasting modes is put forward by Kertesz particularly in his book "Fatelessness", which is a story of his experiences in concentration camps. In this version, the contrast centres on the difference between the human perspective of looking back, when hindsight tells you what will happen and you see it all as inevitable (fated), vs. when you are looking forward (from your place in the queue for induction at Auschwitz say) and you don't yet know what will happen, are still full of hopes and fears, still seizing any opportunity to improve your chances. Kertesz' protagonist on his return in 1945 immediately quarrels with his acquaintances because of their framing of his experience as fated, out of his control, to be left behind as meaningless: which was not (he argues) at all what the experience as lived was like. To regard it as fated, to "put it behind you", is to see it as without meaning. Certainly to see it as having a single emotional value. But that flatness of post hoc judgement is not how it was to live through it. As he writes at the end of the book, "For even there, next to the chimneys, in the intervals between the torments, there was something that resembled happiness. Everyone asks only about the hardships and the 'atrocities', whereas for me perhaps it is that experience which will remain the most memorable. Yes, the next time I am asked, I ought to speak about that, the happiness of the concentration camps."

    One problem with almost all history books is that they can only speak about the retrospective view of it as reality, fact, inevitable; whereas the characters whose behaviour in part made historical events happen could only see the situation from the other perspective of alternatives and uncertainty.

    The contrast between these modes of thoughts is not the logician's one of truth vs. falsity, or fact vs. fiction; between whether a story is true or not. It is between what is taken as reality (even though we know that any part of it could in principle turn out not to be what we believed after all), and something where you don't care whether or not it is true but that it is internally coherent. Whether something is true or not does matter to us, but is a different distinction from being a story or not. Two distinctions.

    Part D. What is not a story

    What is the boundary between stories and other uses of language? If there is no boundary, then there is no theory of narrative or narrative pedagogy other than general language use.

    A good context to think about this is the "news", which is often referred to as "news stories". Yet the whole point of news is that it should be fact not fiction, although not any fact but "new" and surprising ones. Many news items are clearly stories: the more they are about familiar characters (celebrities), or chosen for a particular emotional effect, the more they are like stories. Yet at the other end, announcements of daylight saving ("the clocks go forward tonight") is not a story; anymore than telling your partner what your new mobile phone number is, is a story.

    The interesting thing about the news is on the one hand the relentless pressure to turn facts into stories with ever stronger narrative elements; and on the other, that it is exactly when there are unscripted huge events that the most people tune into the news.

    In fact personally, I find the markers of "bad" or "non" story-telling the most gripping and moving. This is so whether in my research area it is how I learn from participants' open ended remarks (not when they tell a familiar story that suits them, or is what they think I expect to hear); or when you look at the first accounts from journalists in Time/Life magazine of concentration camps, where the language itself shows how utterly beyond their normal resources for writing "stories" they were; or in a free weekly newspaper that gave such huge space to its journalists' featured story that you could see when they ran out of the aspects that fitted their view and had then to include the other things they'd learned that didn't really fit their view, but by the same token conveyed a sense of true reporting not commentating that is wholly absent from most journalism. In showing this preference I am perhaps merely preferring accounts of personal experience that are markedly disconnected from persuasion or other self-interested goals of the teller; possibly even a preference for truth over narrative.

    Perhaps in the end stories, narratives are what I call narrativeA, and which Labov & Waletzky simply defined as containing temporal sequences i.e. any account of past experience: episodic memory by definition.

    However Gladwell (2002, p.118) says (following Bruner) that narratives are particularly important to young children (pre-schoolers); and that they don't like TV advertisements as much as was once thought because ads don't tell stories. The original Sesame Street was anti-narrative and had to change.

    This, if true, means: that narrative is indeed important for learning and teaching; that many things that seem to hold children's attention are actually ineffective at least for learning and retention; but msot importantly here, seems to imply that "real" narrative in the sense of what holds children's attention and supports learning is somewhat larger and more sustained and connected than a single temporal juncture as in Labov & Waletzky; or a single TV sketch. That is, real "narrativeA" is a chain more than 2 links (events) long; and not a single sketch or single cause-effect item. This assertion is built on recording the soliloquy narratives of young children (Nelson, 1989).

    Part E. Practical applications of narrative to teaching

    Relationship of narrative to educational theories

    Why are (some features of some) narratives a good match to existing educational ideas?


    Use Labov & Waletzky to review course or lecture structure: have all 6 elements been provided explicitly and well enough (Hazel's proposal)? This approach is consistent with seeing Labov & Waletzky as an analysis of what hearers want and perhaps need.

    Requiring Learners to write narratives

    Paul Hazel for one gets learners to write narratives. This can be valuable in disciplines that normally exclude this. A corrective to, say, lab. reports. A narrative format can be a good match to at least two important educational objectives in any subject.

    Case studies

    Case studies (Stake, 1995) are one form of data presentation: they are essentially stories, and the format is appropriate where the cases are few, rich, but do not by themselves support reliable generalisations.

    Some comments from Paul Hazel

    1. In the section on 'Natural Modes' you seem to be equating lecturing directly with storytelling. I believe this is an over-simplification. My guess is that the verbal content of most lectures is what I would call 'exposition', a rather more formal and abstract type of discourse. It may be that the lecturer might use stories (often to relate personal experiences) but the vast majority of the time what is said is not narrative in form. (Unless something like History, essentially a narrative discipline, is being taught). Of course this relates back to what I say in the NMHE 'paper', that lectures are a poorly understood phenomenon.

    2. In Part B you touch upon the issue of academic papers. Read Harré: he argues that these are narratives with the subjective element leeched out of them, i.e. it's a rhetorical or stylistic device to make them appear more 'objective'.

    3. In Part C you make an interesting observation about the opposition between the retrospective "inevitable" view offered by History and the forward-looking "perspective of alternatives and uncertainty" experienced by the protagonists. I'm not sure this is true. Many historical accounts emphasize chance elements played, the uncertainty of the protagonists, or happy/unhappy accidents that led to certain outcomes. Also, much History is speculative.

    4. In Part E you say that stories are "very frequently" about people. OK, yes. But I believe this is one of THE major fallacies about narrative, probably derived from most narrative analysis being literature-based. As I say in the NMHE notes: "A narrative does not by definition have to be centred on an individual. Many historical accounts are not: there are narratives about plagues, wars, art movements, institutions, products, and even raw materials such as paint and glass."


    Alan Bennett The Lady in the Van (1999) (Profile Books)

    Gladwell,M. (2000/2) The tipping point: How little things can make a big difference (Little, Brown and co.: New York)

    Hazel, Paul (2006?) "Narrative, Meaning, and Higher Education" (unpublished so far: Paul Hazel, Swansea Institute of Higher Education,

    Hounsell,D. (1984/ 1997) "Contrasting conceptions of essay writing" in Marton,F., D.Hounsell & N.Entwistle (eds.) The experience of learning (Edinburgh: Scottish academic press)

    Kertesz, Imre Fatelessness Hungarian publication 1975 English translation, 2004. (Random House)

    Labov, W., & Waletzky, J. (1967) "Narrative analysis: Oral versions of personal experience" pp.12-44 in J. Helm (Ed.) Essays on the verbal and visual arts Seattle, WA: University of Washington Press.

    Labov, W., & Waletzky, J. (1997) "Narrative Analysis: Oral Versions of Personal Experience" Journal of Narrative and Life History vol.7 no.1-4 pp.3-38

    Mishler, E.G. (1986) Research interviewing: context and narrative (HUP)

    Nelson,K. (ed.) (1989) Narratives from the crib (Harvard University Press)

    Papert, S.A. (1980) Mindstorms: Children, computers and powerful ideas (Basic Books: New York).

    Perry, W.G. (1968/70) Forms of intellectual and ethical development in the college years (New York: Holt, Rhinehart and Winston)

    Stake,R.E. (1995) The art of case study research (Sage publications)

    Web site logical path: [] [~steve] [this page]
    [Top of this page]