The paper tries to suggest parallels between the study of narrative organization in psychology

and the philosophical trends towards a decomposed view of the human mind. It starts

off from an analysis of narrative organization in modern memory research. The substantial

message of this research from the point of view of narrative studies is threefold: it emphasizes

the importance of schematization in memory, as opposed to mere associative

structures; it shows that among the possible schemata narrative organization is the most

available and most universal one; as to the content of narrative schemata it shows that they

are closely tied to our naive theories of human action. In a psychological sense, the cohesion

of narratives is tied to their use of intentional attribution. We apply our schemata of

human action to understand the plot of narratives. The Hume-Mach style empiricists, and

later on modern novelists have been struggling for a long time with the place of Subjects

in a totally decomposed vision of the world and the mind. This modern emphasis on a lack

of coherence is recently becoming connected to the issue of narrativity in non-trivial ways

by philosophers like Daniel Dennett. This trend accepts the idea of decomposing the self

and other unifying constructions. However, the human need for coherence is accepted by

them, and rather than proposing cohesion based on solid Egos, they propose different varieties

of narrative theories regarding the self. These proposals have a strong Humeian flavor

with their emphasis on the constructed but useful nature of the self concept. Their intellectual

novelty is, however, that they try to find the sources for constructed coherence

in narrativity.

The paper argues that the philosophical and psychological narrative theories of the self

have relevance to the study of literary narratives. Part of modern literature in this regard

can be seen as a human experiment in facing the lability and soft construal of human integrity.

0324–4652/2003/$20.00 Akadémiai Kiadó, Budapest

© Akadémiai Kiadó Kluwer Academic Publishers, Dordrecht

Neohelicon 30 (2003) 1, 187–205

* The presentation here borrows heavily from an earlier version (Pléh, 2002). I would like to express

my thanks to Ernõ Kulcsár Szabó to invite me to present my ideas to a literary audience.

While working on the paper I was supported by a Hungarian research grant NKFP 5/ 0079/2002.

All novels, of every age, are concerned with the enigma of the self.

Milan Kundera: The art of the novel, p. 23


Modern humanities and social sciences had come several times to the conclusion that

narration is a very special feature of human nature, and therefore, it is somehow a key

to understanding humans. At first, this train of thought was mainly psychological and

concentrated or rather limited its attempts to the study of the relationships between

memory organization and narrative patterns. Provocative ideas were developed vis a

vis the dominant world view, regarding elementary associationism especially.

In classical psychology, the issue of narration first emerged in connection with a

critic of association as an explanatory concept. The British experimentalist, Sir Frederick

Bartlett (1932), the French clinician Pierre Janet (1928), and the Russian educational

psychologist Pavel Blonskij (1935) have all advanced narration as a basic

non-associative organizing principle of the human mind. For Bartlett, narratives were

the key moments in schematic memory which was already characterized by Alfred

Binet (Binet and Henry, 1894) as the mirror of thought. Blonskij (1935) also believed

that narrative organization is the key to “logical memory”. The French clinical theorist,

Pierre Janet (1928) even claimed that logically organized text was a key to rationality,

since the origin of rationality should be looked for in interpersonal coordination.

More distant intellectual antecedents can also be traced since this first narrative

trend in psychology was part of a dissatisfaction with elementaristic experimental

psychology at the turn of century. The dual memory system proposed by Bergson

(1896) can be interpreted as a duality between the memories of the body (habits) and

meaningful, personalized memories. The human brain is a storehouse of habits, but it

only provides an access route to personal souvenirs, which themselves are not identical

with any “memory traces”. In order to interpret and personalize something, a royal

way is to assimilate it to a schema, provide a story-like narrative structure to the random

data of our memory system. (For an interpretation of Bergson along these lines

see Pléh, 1989).

The French social school of psychology, partly reflecting on Bergson, went further.

For them, the personally interpreted experiences of Bergson turned into socially

constructed experiences. They moved to a position that denied the existence of “raw,

uninterpreted experiences” altogether. In this process, narrative organization became

a key element of human life. Maurice Halbwachs (1925, 1950, 1992, 1994) in his

work on collective memory, that still is the standard reference point for studies on historical

memory, presented narratives as being responsible for the social nature of our

memory in two regards. Social groups build up their community life by inventing and

sharing stories, thereby creating a common interpretation of experience. The “substance”

of community life is inherent in stories. The social aspect, however, has a


structural side, as well. Narrative organization prevails even in our most private memories,

in our dreams and daydreams as well. Structurally, all our experience is socially

organized. There is no such thing as socially non-interpreted mental content. Whatever

comes to our mind, becomes intelligible through the intervention of the “social

moment”, and this social moment is provided by narratives.

The social aspect for Halbwachs was an “intentional issue”, and not that of the relationship

between stable external and internal objects. In this way, with the rejection of

the “objecthood” of memory, Halbwachs opened a road towards constructive theories,

both regarding memory and in constructing the person (Pléh, 2000).

Their philosophical roots notwithstanding, these first narrative theories failed to

become paradigm setting alternative approaches to psychology and in the humanities.

They had relatively little affinity with the philosophies of their time, and while they

had a clear conception about the role of schemata and their social origin, they did not

succeed in winning the interest of epistemologists, and did not succeed in trying to

propose a general theory on the origins of Self and personhood based on the notion of

narrative schemata. Interestingly enough, the strong alternative paradigm at the time

of the first narrative theories was articulated in the domain of the psychology of perception

as Gestalt theory.

The first narrative models were also insensitive to the dramatic changes going on in

their own time in the narrative patterns of modern European literature. On the other

hand, it is true as well that early constructionist theories of selfhood in philosophy

were also uninterested in the issues of narration. The recent affinity between these

three areas indeed indicates a definite change in cultural climate. This is rather a present

day development, as well as the search for parallels between changes in contemporary

prose writing and our conception of identity and the Self.


The high time of narratives in psychology has come back from the late sixties on.

Now, however, the narrative patterns were imported to psychology from folklore, anthropology

and literary studies, and while they infiltrated psychology, they had soon

reached a level of generality touching upon philosophical issues such as the relationships

between story telling practices and our naive notions of personhood.

Through the rediscovery of the works of the Russian folklorist Vladimir Propp

(1928/1958), anthropologists realized that strict rules or regularities are hidden behind

the fantasy rich world of our tales. Tales are by far not the terrain of irregularity and

unconstrained fantasy. Rather, they are characterized by a limited number of “roles”

and “functions”, they have a culturally constrained repetitive structure.

In this new trend, narratives started to carry two basic notions for social sciences

and the humanities: order and “surplus value”. Order implies that connected discourse

is no more a mirror of the chaos of the world, but rather, follows a set of constraints.

These constraints embody an underlying pattern that goes beyond the mere

concatenation of propositions. In the new, linguistically minded trend this suggested


to look for a pattern that goes beyond individual sentences to find the real patterned

meaning in discourse. In actual research, this would both mean an emphasis on the

“surplus value” of discourse as compared to a set of individual sentences, and an ideology

which sees a world view hidden in the organization of text (Greimas, 1966,

Todorov, 1969). The new structural approach to stories had some resemblance to the

previous starting from social schematization and constraints, but with a stronger linguistic

emphasis. It also has similarities to earlier Gestalt approaches, but now, with

the linguistic turn in full force, the “enemy” is not the idea of elementary sensation,

but atomic sentences.

Psycholinguists have come in touch with this structural approach to narrative organization

through revitalizing the concept of schemata. Narrative patterns turned into

something crucial for the psychologists, because they promised to provide a substantial

anchoring point for the otherwise elusive concept of schemata. Schemata interpreted

in a narrative way are not simple products of the constructive powers of the individual

mind, rather, they are interpreted as being anchored in an external social order.

The patterning inherent in narration promised to give a really interpreted view of

schematization, that had unexpected consequences for our conceptions about ourselves.


Out of the mysterious and chaotic fabric of life, the old novelists tried to tease

the thread of a limpid rationality;

in their view, the rationally accessible motive gives birth to an act,

and that act provokes another. An adventure is a luminously causal chain of acts.

Milan Kundera: The Art of the Novel, 58.

In modern psychological research on discourse memory usually two types of processes

are highlighted: integration and selection. Whatever one remembers from a

text is at the same time more and less than the text itself. Bridging informational gaps

between the different propositions becomes part of the representation, which is a coherent

causal chain (Schank, 1975). Complex and unclear texts, on the other hand are

simplified and converted into a more logical version. In this process we act as if we

had found the leading propositions underlying the text.

These general principles are valid for remembering all sorts of texts. Why are narrative

texts so special, why are they for examples so much easier to understand and remember

than other texts? (See a detailed argument for this ease in Zinchenko, 1961.)

Present day psychological studies look for the explanation of this superiority in the

general organization of narrative patterns. While the organizing moments are easy to

mobilize, they can turn into overall mental models at the same time, keys to the understanding

of other processes such as the interpretation of real events.



(A) They represent a temporal organization where the order of some (the critical)

propositions is assumed to be the order of events in real life. Due to this feature, narratives

are apt to be used and treated as causal models of events and actions. Labov and

Waletzky (1967) used a distributional analysis to formalize this in spontaneous narratives.

Narrative clauses in spontaneous story-telling are partially ordered thus providing

the basis for “narrative time”. This has an interesting developmental aspect, as

Bruner and Luciarello (1989, Bruner, 1996) highlighted it: temporal perspective in the

child’s mind is formed in the process of story telling. One can of course go further,

and claim as Janet (1928) already did, and as Ricoeur (1965) and others do today, that

the psychological time dimension comes from narration altogether (see about this

Modell, 1993).

(B) Stories as a special type of narration require a hero, who has a system of goals,

as well as a perspective. Perspective (who’s vision of the world do we use) characterizes

not only higher literary forms but all narratives. The hidden philosophy of stories

is given by the problem solving path of the hero within a motivational field created by

the goal system. The essence of this “philosophy” is the coherence in the actions of the

hero. A further step is provided by the idea that heroes are intentional agents, and that

human actions performed by these agents form causal chains based on reasons and

causes (Schank, 1975, Graesser, 1996). From a developmental perspective this suggests

that by distinguishing between outside (“real life”) events, the inner life of the

hero, and the reactions of the narrator, story telling practices foster the distinction between

objective reality and mental reality. (See the volume edited by Bruner and

Hastie,1987 on this process of differentiation.) This aspect of stories has the challenging

implication that narration is somehow intimately tied to our models of personhood

and Self. The world of narration would be making the connection between the real

world and our inner world (our Ego?). Narratives provide us, rather than being mere

reports, also perspectives, the help to “give meaning” to whatever happens to us

(Bruner and Luciarello, 1989).


What sort of explanatory models do predict memorial schematization observed in real

life and in the psychological laboratory? To make a long story short, several decades

of experimentation basically came to the conclusion that our cognition is regulated

here as well as in person perception, action interpretation, and many other areas of social

psychology basically by an anthropomorphic naive psychology about the causes

of human action. Complicated events of real life, and the stories narrating them are

made coherent and understandable by relying on a system of expectations using a

model of human action.


Recall of narratives in this view is a selective and integrative reconstruction based

on our action schemata. Narrative schemata are easy and they appear early in life, because

they do rely on an early model of human intentional action that is supplemented

by a perspective providing pattern of narration. Contemporary research on memory

and psycholinguistics of text attempted to provide a more explicit version of these

schemata in several forms. The different models can be sorted into two types, form

based and content based models of narrative schematization. Initially, due to the enthusiasm

for the structural models discovered in Propp (1958) psychologists also

looked for the sources of schemata in the text itself, following the “linguistic turn” of

philosophy in psychology as well. Structural models proposed for a given corpus of

texts by Propp (1958) were extended and rewritten as a generative system of rules able

to produce an open and infinite set of stories. The most elaborate example was the

grammar for Eskimo folktales proposed by Colby (1973). According to the processing

application of this model by David Rumelhart (1975), recall would be directed by

these rules. Basically, more embedded episodes would have a lesser chance to be recalled,

and some types of grammatical nodes would have a higher chance to be recalled.

Several models of story grammars were proposed following Rumelhart. Essentially,

all of them treated stories as a series of embedded attempts trying to achieve a

Goal, as the simplified higher level structural rules indicate.

(1) StorySetting + Episode

(2) EpisodeState + Attempt

(3) AttemptPlan + Action + Outcome

Two aspects are rather relevant for our concerns here. First, these seemingly formal

(syntactic) rules are also based on the organizational principles of human action,

as a careful reading of the “morphology” of Propp shows it. An action-based pragmatic-

semantic model stands behind the apparent “syntax” of these models. What else

does rule (3) represent than the fact that according to our folk psychology Actions are

preceded by Intentions, and Actions are interpreted as successful or unsuccessful in

relation to the Intentions or Plans?

Beside this category analysis, some content-based models were also formed that

reconsidered the seemingly “grammatical” relationships between events and actions

in a narrative as motivational relationships between “naive psychological categories”.

“Who Done What Why” is the organizing principle as it was phrased most clearly in

the Causal Chain model of Roger Schank (1975, Schank and Abelson, 1977).

All of these models received empirical support from experimental studies. “All”

implies here that both types of models had a half dozen varieties in the heyday of story

grammars. Grammar-like models received their psychological relevance from higher

recall of structurally higher nodes and a better recall of Attempts, than Plans or States.

Regarding the content-based models, higher recall of the main causal chain was observed

and also more recall of Causes in physical actions, and higher recall of reasons

regarding interpersonal scripts (László, 1986). The real test came, however, when pre-


dictions of the different models were compared over the same experimental material.

Black and Bower (1980) and Pléh (1987) showed using multiple regression models

for recall patterns based on the different story structure proposals, that the models that

relied on the action system of the hero in assigning structure to the stories had a higher

predictive power than the ones relying on a purely formal model. The kind of narrative

research based on this naive social psychology has been flourishing ever since.

Graesser (1992, 1996, Graesser and Clark, 1986) even developed a special on line

questioning model to analyze how immediate is the construction of causal chains during

reading texts.

These results coming from cognitive psychology imply that the key for the simplicity

of simple stories , the special schema we looked for (the schema that is so easy,

appears so early in life, and is so much universal) should be looked for in the naive

psychology of human action. In understanding stories, we mobilize our naive social

psychology about the structure of human action and about the usual motives for action.

Coherence is found by the hearer-reader through the projection of these motivated

action schemata to the story. The specificity of traditional simple stories lies in

the fact that due to the prototypical motivations in a given culture, and due to the simple

transparent narrative point of view, this action organization can be revealed easily

and unequivocally on the part of the understander (see about this László, 1986, Halász,

László and Pléh, 1988).

There is a further question regarding the origins of these interpretation patterns.

The initial questions regarding what gives pattern to simple stories, find an answer in

“naive social psychology”. One has somehow to answer the upcoming next question:

where do patterns of naive social psychology originate from? Through its intimate

connection with the issue about the origin of a “theory of mind” in humans, there are

rival solutions here. One of them would basically state that some kind of intentional

and earlier even teleological attribution is a modular feature of the human mind developing

very early on (Gergely, Nádasdy, Csibra, and Bíró, 1995, Csibra and Gergely,

1998), while others would claim in their similarly non-essentialist approach that this

naive theory develops as a very result of experience with narrations (Bruner, 1985,

1990, 1996), I should add, also carrying a strong social emphasis about the origin of

our attributing schemata. Thus, a search for coherence underlies our schematization of

stories, and this coherence is basically found by “turning on” our machinery of intentional

attributions, and thereby reconstructing a causal chain that consists of causes

and reasons that lead to these events. This is the view of the classical writer, too, as the

Kundera motto indicates.

The empirical research on remembering narratives that seemed to be originally a

rather “down to earth” empiricist project, through the seemingly innocent notion of

“connected discourse” and “schemata” did become tied to issues touching upon the

“frame problem” of the entire enterprise of human cognition.




That is the point where the narrative frame issue becomes intimately tied to the crisis

of modernity and to the problem of the relations between the changes of narrative patterns

and a crisis in our view of ourselves. In present day intellectual contexts

narrativity is entertained not only as a “low level, down to earth” theory about actual

narrations, but also as narrative metatheory. There is a remarkable basic similarity in

the way narratives become central in experimental psychology, in the study of development

and in the cultural and philosophical theorizing about the centrality of narration.

At the same time, new patterns of narration take shape. The extended narrative

conceptions define themselves as contrasted to the essentialism of the classical view

of man (see about this opposition Bruner, 1985, 1990). Essentialism in this regard is

the belief in a postulated stable Ego as a starting point, and a stable world of objects.

This is replaced by a world that is socially constructed through our narratives, and

with a constructed Ego, rather than the Ego being a Cartesian starting point (Dennett,


The narrative philosophical interpretations take a special look on the mind and

personhood. They attempt to construct a new vision of Consciousness and the Self on

this basis. This line of thought sets off from the concept of dissolution and distribution.

When it looks for metaphors of the mind, it compares it to the issue of integrating

large empires like the British Empire where due to communication problems things

seem to be disintegrated, there is no real center. Human mind is likewise characterized

by spatial and temporal disintegration or distribution. The narrative metaphor in this

argumentation becomes an integrating tool. The disorganized events in our mind get

organized through the mediation of stories we tell to ourselves. The unity of consciousness

and self disappears as a first step and comes back through the back door as

narrative integration (Dennett, 1990, 1991, Dennett and Kinsbourne, 1992). This type

of metatheory is among the first ones to make a close connection between two sorts of

“crises” regarding coherence going on within European culture. One is the crisis of

narration, the other the crisis of the Self-concept.

In the 20th century, there were several waves of feelings to the effect that not only

our outer world fell apart into pieces and became incoherent, but our internal world as

well. We could start from a “Once upon a time Order”, the harmony between the

world of God, the stable Self and the stable patterns of Narration. This age is characterized

by transparency. In the good old days of early modernity, Cartesian unity and

transparency governed everywhere (Latour, 1993). The intentional stance (Dennett,

1987) could be taken regarding all the three realms that are of interest to us. The World

was created by an intentional God (the argument from design), our internal world of

experience was transparent to a self-conscious intentional agent, the Mind, Soul, Ego,

or Self, and the Stories we wrote were also about intentional Heroes who carried

clearly identifiable Plans. The whole world was to be characterized by stable intentional

agents. (See about this heritage and its multiple crises Toulmin,1990, and especially

the volume edited by Dorothy Ross, 1994.)


The dissolution of this unity and the birth of a “distributed or deconstructed view”

of the human mind was first proposed by the most courageous of the empiricists, David


What we call a mind, is nothing but a heap or collection of different perceptions, united together

by certain relations. ... The mind is a kind of theater, where several perceptions successively

make their appearance.

D. Hume: A Treatise on Human Nature. Vol. 1. 436, 438.

He clearly realized that this vision had dreadful consequences for the notion of personal


There are some philosophers who imagine we are every moment conscious of what we may

call our self ... for my part, when I enter most intimately into what I call myself, I always stumble

on some particular perception or other, ... I never can catch myself at any time without a

perception. ... And were all my perfections removed by death, and could I neither think, nor

feel, nor see, nor love, nor hate after the dissolution of my body, I should be entirely annihilated.

D. Hume: A Treatise on Human Nature. Vol. 1. 436, 438.

This attitude was never trivial to take. Following Hume, the questioning of unity and

its reassertion came usually from rival ideological camps and therefore from different

sources. As a bold generalization one can claim that the proponents of disintegration

were usually progressives who believed in the power of empiricist attitudes and science,

by far not being pessimistic about their conclusions, and the trends reemphasizing

unity were usually closer to church powers and to religious ideas about an immortal

soul (Goldstein, 1994).

In the debates of nineteenth century on the issue of a substantial Self there was a

constant give and take between the “religious” and the “empiricist” critical theories

(Carrithers, Collins, and Lukes 1985). The idea of a disunited Self came along several

times during these debates. It showed up in the French empiricist epistemology of

Condillac and others, and in the radical positivist views of people like Mach (1897,

1910). There was always a strong counter reaction. In French intellectual life, the radical

deconstruction proposed by the body image theories of Condillac were opposed by

the mid-century school philosophy of Cousin that reinstated “Moi” as a starting point

(see Goldstein, 1993). Seen in this context, the psychopathological ideas of Charcot

and Janet about the dissolution of personality, as well as those of Freud at the other

end of the continent were less radical proposals than seen at first sight: the contemporary

and earlier empiricist deconstructions were much more radical in their intent.


Our present intellectual world can be characterized by two types of dissolutions (or, if

you prefer, crises). The first one is the dissolution of the stable Ego that was already

characteristic of the late 19th century philosophy and psychology that became with the


words of the Hungarian philosopher Kristóf Nyíri (1992) “impressionistic” in its

search for stable reference points. The same goes on in the most up-to-date

connectionist approaches to the mind (McClelland and Rumelhart, 1986). Their emphasis

is laid on a “nothing but” approach to the mind (knowledge is nothing but a pattern

of activation in a neural network), and on overwhelming parallel processing and

distributed representation all question the conventional Cartesian unity of the mind. It

is still the best introduction to this radical connectionism. There is not too much space

left in this for the Ego: it is in a way dissolved in the multiplicity of the parallel and in

themselves meaningless multiplicity of computations and connections. Churchland

(1995) gives a comprehensive radical interpretation for all of this by stating that all aspects

and issues of human inner mental life would on the long run be simply identified

with neural models.

The other, parallel dissolution or disintegration went on in the realm of culture.

One dominant aspect of this in 19th century had been a dissolution of traditional patterns

of narration. There are interesting parallels here between artistic practice and

philosophy. Kristóf Nyíri (1992) analyzed the affinities between the elementaristic

theory of mind proposed by Ernst Mach (1897), and the school of impressionistic

painting. The strong drive to liberate yourself from anything secondary, knowledge

based (top-down), anything schematic and a search for undeniable, original certainty

lead to pictorial and epistemological impressionism: the real raw stuff of both would

consist of patch-like pieces of experience. There was, however, a similar trend in

questioning the validity of traditional narrative schemata and the underlying naive application

of the intentional stance as well. There are interesting parallels between giving

up the idea of a causal chain in the outside social world of the novel, and questioning

the presence of an integrative Ego in the inner world of the novel. (Kundera, 1986

gives an interesting survey along these lines.)

In the new types of narrations taking shape in 20th century the God-like image of an

author with all encompassing knowledge is replaced by either a direct presentation of

the inner world, or with a description of external behavior with no preassigned perspectives.

Narration dominated by the intentional stance is replaced by a presentation

of internal mosaics like already in Proust or Joyce, or in the French Nouveau Roman,

or in the theoreticians and practitioners of postmodern literatures. Among the later

ones, Hassan (1987, 1990) tries to make a connection between new writing and the

“deviations of the self”. He lists the slogans about the dissolution of the Self that have

a tragic connotation (divided Self, downgraded Self, dissolved Self, and so on), but after

this exercise, he makes a postmodern turn. He arrives to the conclusion that by

questioning the selves of the hero and the writer, and hereby by playing with the narrative

point of view, new playful narrative structures may emerge.

With the birth of the modern novel in Proust, Joyce, and Musil, writers show that

Kundera is right: they were experimenting with knowledge structures, and prefigured

a narrative concept of identity (including with all of its crises) well before it was formulated

as a theory of mind by philosophers.


The relationships between the issue of Self and narrative structures was already realized

by some of the classics of narrative research in psychology, without relating

them to the issue of literature. Frederic Bartlett (1935, p. 311) himself said:

There may be a substantial Self, but this cannot be established by experiments on individual

and social recall, or by any amount of reflection on the results of such experiments.

Samuel Beckett (1987) gives a similar interpretation for the importance of the multiple

and non-conscious construction of the Self in Proust:

But here, in that ‘goufre interdit a nos sondes’ is stored the essence of ourself, the best of our

many selves and their concretions that simplists call the world.

S. Beckett: Proust. 31.

There are tragic, and ambiguous, ironical overtones as well in the literature regarding

the dissolution of the Self. Both do leave one central issue open, however. When we

dissect the Self into elementary experiences and their relationships, and narration into

narrative morsels, do we make them disappear by this very act? Did the Self really disappear,

or do we only claim that compared to the primary stuff of experiences it is only

secondary? (That is the way, for example, Beckett interprets Proust.) Did narration

disappear, or is it only a secondary organization compared to the primary thread of

discourse? Did we manage to radically eliminate coherence, that is usually accounted

for by the Self and by narration, or we only made it secondary rather than using it as a

starting point?

Narrative metatheory as a non-essentialist view of coherence takes the second option.

Rather than postulating a substantial Self, the coherence of our internal world comes

around by milder means, by story telling.

The issue of coherence in communicative terms implies that the partners, A and B

have to follow a mutual, joint model. They have to allow each other to reconstruct

similar relationships between the individual propositions. This is referred to as the

maxim of relevance by the communication model of Paul Grice (1975), and as the issue

of higher order models of intentionality by Dennett (1987).

Seen from this perspective, traditional narrative schemata with their mobilization

of intentional action interpreting moduls are rather powerful coherence-building devices.

One of the clearest aspects of the transformation of these in modern “high literature”

concerns the changes in the comprehensive Plans of action from the point of

view of the Hero and/or the Narrator. Its presence gives coherence to classical narratives,

be it fairy tales – he wants to marry a king’s daughter, sets out into the world, and

through many obstacles gets her, or bourgeois novels – he comes to the big city, wants

to make a career, relying on relatives and women, reaches his goals. The comprehensive

message of the literary work is tied to the intentional system of the hero. The final

meaning (the life philosophy embedded in classical narration) of the work is the idea

that there is a continuous, intelligible life, with initiatives of human Agents, who are

full of Plans. These Plans give coherence of the man and of narration, as Kundera

(1986) shows.


Of course, classical narration had many varieties, but they were sharing these underlying

assumptions. Classical detective stories hide the plan that makes the events

coherent, and it is the task of the reader, or Mr. Holmes to reveal them. But it is possible

to reveal them because they do exist. Something different is happening in modern

literature. It spoils this “super plan”, this goal system encompassing the entire novel.

Kundera (1986) sensitively presents how this type of goal coherence is ruined in the

novels and realities of Franz Kafka. The hero is subjected to non-transparent Plans of

others, and these Plans do not become clear even till the end of the story. The continuous

goal system disappeared before Kafka as well. It is replaced by the world of inner

experience in Joyce, and in Proust, as analyzed by Beckett the action based logic of

narration and the actions of the hero are replaced by an undifferentiated network of experience,

imagery, and souvenirs.

The identification of immediate with past experience, the recurrence of past action or reaction

in the present, amounts to a participation between the ideal and the real, imagination and direct

apprehension, symbol and substance. Such participation frees the essential reality that is denied

to the contemplative as to the active life. What is common to present and past is more essential

than either taken separately.

S. Beckett: Proust. 74.

The whole work of Proust is not merely an essay on memory, but a practice in the relationship

between unconstrained unintentional memory, and “voluntary memory”, the

“uniform memory of intelligence”. “…his entire book is a monument to involuntary

memory and the epic of its action” (ibid., 34).

The comprehensive Plan disappears not only in the impressionistic presentation of

the internal world but also on the level of behavior. In this third type of modern writing

the external behavior is not characterized by clear Plans. Rather, things just happen to

the hero, and he acts reactively, and tries to give meaning to the actions only afterwards.

The continuous world of intentions is replaced not by an inner world of experience,

but by the world of behavior. Think of some of the acts of Mersault in The

Stranger, or to the beginning acts of Belmondo in Au Bout de Souffle. The reader and

the viewer are immediately presented by pieces of behavior, without enough preparation

for the setting, and without a possible intentional interpretation. The individual

experiences and acts are not presented as parts of an encompassing Plan. They can

only be given a local interpretation. He shot the cop asking for his papers, but this happened

so fast that neither he (the hero, Belmondo), nor we, the viewers had any chance

to build up a plan to motivate the deed (Au Bout de Souffle). In a secondary way, we

give interpretation to something that already happened. We make a story out of it like

psychoanalysts, but the unique uninterpreted act preceded the story, while in classical

narrative patterns the starting point is the story with its intentional layout, and unique

events fill the slots in a secondary way. Classical narration, while it certainly uses a

narrative pattern, treats the pattern in an essentialist way, with a belief in the integrative

Self, the events being only manifestations of this. This is in accordance with the

top-down style of writing, and with the ideal of an omnipotent and omniscient writer.


The key scene from The stranger illustrates this lack of narrative buildup relying

on intentions:

Then everything began to reel before my eyes, a fiery gust came from the sea, while the sky

cracked in two, from end to end, and a great sheet of flame poured down through the rift. Every

nerve in my body was a still spring, and my grip closed on the revolver. The trigger gave,

and the smooth underbelly of the butt jogged my palm. And so, with that crisp, whipcrack

sound, it all began.

Albert Camus: The Stranger. 76.

The murder by Meursault is rather different from that of Lafcadio in the Caves of the

Vatican by Gide. His act is quoted as the classical example of action gratuite. This is

an act of “no motive” however, only in the sense of bringing no utility to the actor.

Otherwise, Gide makes it sure that we see it as planned, intentional, premeditated action.

Lafcadio even laughs in advance how much trouble will the police have in dealing

with a crime sans motive, with an unmotivated crime.

It’s not so much about events that I’m curious, as about myself. There’s many a man thinks he

is capable of anything, who draws back when it comes to the point... What a gulf between the

imagination and the deed !... And no more right to take back one’s move than at chess. Pooh! if

one could foresee all the risk, there’d be no interest in the game!

A. Gide: Lafcadio’s Adventures. 186.

Of course, there are plenty of more recent examples for the dissolution of intentional

coherence. In this respect, Christine Brooke-Rose (1986) presented a rather interesting

outline for the changes in writing so typical of modern (and of course postmodern)

literature. First came the defocalization of the hero. That was already present in the

19th century. Think of the well known comparisons regarding the Waterloo battlefield

descriptions by Victor Hugo in Les Misérables where you have an epic enumeration

combined with a panoramic view and a clear presentation of the scenery, with the

scene of Fabricio del Dongo being part of the great battle in Stendhal’s Chartreuse de

Parme without really knowing it. The entire scene is defocussed: we see the hero as

being entirely out of the intentional plans of the agents, unaware of their plans, and

even of them being agents. He does not even realize he is seeing the great man he came

for. He is part of the battle without knowing he was there.

This is defocussing of the intentional plans, indeed. But this was further combined

with a defocussing of the “survival value” of the hero. Present day heroes are no more

close friends of ours as Madame Bovary or Anna Karenina, or Rastignac, to that effect.

In the vision presented by Brooke-Rose (1986), the postmodern riot against character

and stability actually eliminated anything that would contribute to the coherence

i.e., anaphoric and referential signs, and only the egocentric or deictic signs would

work. There are, however, signs of hope (if you like) according to her. Character shall

come back through the back door, through science fiction and stabilization of the

postmodern play with fiction.



The concept of communicative coherence allows us to look for inner coherence in a

non-essentialist way and to avoid the usual pendulum-like shifts between disintegrated

and essentialist views of the Self. The notion of coherence might be a help in

finding some peace in the chaos of the world without necessarily committing ourselves

to another round of essentialism. The philosophical system proposed by Daniel

Dennett (1987, 1991, 1994, 1996) has some intellectual promise here, especially since

he consciously connects the two types of dissolutions, that of the Self, and of the narrative

patterns. For postulating coherence, he does not need a hypothesized subject.

The coherence of inner life (the Self, if you like) will be a “soft notion” for him, a

“gravitational point” of all the stories we tell ourselves.

For classical empiricist theories like Condillac and Mach, body image was the central

source of coherence. Beside emphasizing our general tendency to attribute continuity

to things we name, the other anchoring point is found by Mach in the concept of

the body. Here Mach reiterates and modernizes an idea already apparent in the work of

Condillac, and later on taken up by several theories both in comparative and in clinical

psychology (Henry Head, Frederick Bartlett) that take the body schema, the representation

of our body as central to the Self concept. “As relatively permanent, is exhibited,

further, that complex of memories, moods, and feelings, joined to a particular

body (the human body), which is denominated the ‘I’ or ‘Ego’” (Mach, 1897. p. 3).

This centrality of the notion of body image did not disappear in present day discussions.

It is certainly textbook material in psychology and it is seen in new neuropsychological

theories as well. Antonio Damasio (1994) in his work on the representation

of emotions and their role in the shaping of our goal system claims that the emotionally

laden representation of our body is the central core to our Self concept.

Dennett (1991) is his view on consciousness also gives a central role to the body image

in the reduction of our notion of the Self. However, his new original point is the

use of the narrative metatheory.

Dennett basically claims for a soft and constructed theory of Self that is even softer

then the body image notion.

A self, according to my theory, is not any old mathematical point, but an abstraction defined

by the myriads of attributions and interpretations (including self-attributions and self-interpretations)

that have composed the biography of the living body whose Center of Narrative

Gravity is.

Daniel Dennett: Consciousness explained. 426–427.

Without a narrative excursion this very metaphor was already present in the philosophy

of arithmetic proposed by Frege (1884) who already uses the metaphor of soft notion,

and the very image of gravitational points.

I distinguish what I call objective from what is handleable or spatial or real. The axis of the

earth is objective, so is the center of gravity of the solar system, but I should not call them real

in the way the earth is real.

G. Frege: The Foundations of Arithmetic. 35.


Dennett is clearly aware of the ethical implications of a system based on soft notions.

This issue was raised many times since the time of Hume, and it was indeed one of the

main resources of the essentialist moralistic arguments (see Perry, 1985).

The task of constructing a self that can take responsibility is a major social and educational

project... The only hope… is to come to understand, naturalistically, the ways in which brains

grow self-representation. ...

If you think of yourself as a center of narrative gravity, on the other hand, your existence depends

on the persistence of that narrative (rather like the Thousand and One Arabian Nights,

but all a single tale), which could theoretically survive infinitely many switches of medium…

and stored indefinitely as sheer information.

Dennett, ibid., 429–430.

[Exegi monumentum aere perennius...]

Horatius, Carm III. 30, 1.

This is similar conception though with some ironical twist to the one proposed by

Sir Karl Popper (1972) about World III, namely that culture can in principle survive

the destruction of its objective carriers. Here, the similar view is claimed for the relationships

between bodies, programs, selves, and narrative constructions. (See about

this Pléh, 1999.)

Dennett has his own ironies, of course, what would you expect from a constructionist

theory. He quotes a novel mocking deconstructionism by David Lodge.

According to Robyn [she is the postmodern theorist] (or, more precisely, according to the

writers who have influenced her thinking on these matters) there is no such thing as the “Self”

on which capitalism and the classic novel are founded – that is to say, a finite, unique soul or

essence that constitutes a person’s identity; there is only a subject position in an infinite web

of discourses – the discourses of power, sex, family, religion, poetry, etc. And by the same token,

there is no such thing as an author, that is to say one who originates a work of fiction ex

nihilo. ... As in the famous words of Jacques Derrida ... “il n’y a pas de hors – texte”, there is

nothing outside the text. There are no origins, there is only production, and we produce our

“selves” in language. Not “you are what you eat”, but “you are what you speak”, or rather

“you are what speaks you”, is the axiomatic basis of Robyn’s philosophy, which she would

call, if required to give it a name, “semiotic materialism”.

David Lodge: Nice Work, 22.

In the view of Dennett, there is no internal agent in a Cartesian Theater who would

make things coherent. Coherence comes as a relaxation point in forging intentional sequences

out of the events coming to us. The use of the intentional stance is the macroscopic

side of the constructing work of our brain. There is microscopic one, too, that is

valid over shorter periods. Dennett and Kinsbourne (1992) apply the narrative metatheory

to the order of milliseconds as well to create a theory of consciousness. Essentially

they claim that there are no central moments of consciousness. In our brain we

are making Multiple Drafts of every incoming event (another narrative metaphor),


and there is one of these that under normal circumstances is treated as being a conscious

stage in information processing. But that delimitation is also a constructive and

slightly artificial notion.

For the same sequence of events several “stories” are created. There is none being

more basic than the others.

How can you reconcile this fragility with a theory that otherwise claims taking the

intentional stance as basic? Dennett protects himself against this challenge by allowing

two different levels. You are entitled to use the intentional stance for the entire

system, but not to its individual operations. But always remember, the application of

the intentional stance is an evolutionary and economic hypothesis, that does not allow

or require to postulate real agents in the brain.

This is a two level theory, like the one proposed by Bergson (1896) a century ago.

The novelty of Dennett is twofold. For him, the second level, the level providing us

with meaning and coherence, does not require a disembodied mind. This level is set

into a narrative and intentional model that in principle will have an evolutionary story

to it (see Dennett, 1994, 1996).

But if Self and all these things are soft notions, why can we talk about their crisis?

The evolutionary argumentation might argue for simplification. We have essentialized

our stances for thousands of years using notions such as soul, Self, mind, etc.

When we are replacing them with narrative constructions, we do not have to be so passionate

as the traditional empiricists were. The skeptical seriousness of finding the

emperor naked should or could be replaced by a lighter and more narrative attitude to

this newest questioning of the Self as a starting point.

Dennett of course, has his own critics. One of the most telling examples from the

“neuro essentialist camp” is Paul Churchland (1995, p. 266) who criticizes him

sharply for reintroducing in his narrative model a linguistic type of model for human

consciousness. Dennett in his view pulls back the wrong prototype for cognition

namely language, then “(2) makes it the model for human consciousness, (3) gives

parallel distributed processing a cursory pat on the back for being able to simulate a

‘virtual instance’ of the old linguistic prototype, and (4) deals with his theory’s inability

to account for consciousness in nonlinguistic creatures by denying that they have

anything like human consciousness at all”. This criticism is rather misfired. Dennett

(1987, 1994) always stands for continuity in his speculations about the origins of human

cognition. While he has a slight linguistic bias, this is not necessarily a disadvantage,

if you consider us a speaking, and even story telling species.

Psychoanalytic theory also phrased an interesting criticism, not directed specifically

against Dennett, but more generally against exclusively narrative approaches to

the Self. Narrative metatheory of course is a driving force in some reinterpretations of

psychoanalysis as well. Starting from Ricoeur (1965) an approach has taken shape

that sees in the psychoanalytic discourse not an unfolding of an internal essence, not a

symptom of some natural sequence of libidinal development, but a series of constructed

narratives, and psychoanalysis is a joint reconstruction of a new narrative between

the analysts and their clients. As appealing as this approach is with all of its hermeneutic

overtones, it is interesting to consider what Arnold Modell (1993), himself a


research oriented psychoanalyst, thinks of it. In his view, the narrative approach can

only reveal and touch upon the temporal aspect of the Self, the other aspect remains

tied to the representation of the body itself. “The literary analogy cannot be followed

very far, because the coherence of the psychoanalytic narrative is ultimately derived

form the bodily Self and its affectionate memories” (Modell, 1993, p. 182).

Dennett himself is not entirely exclusive in his narrative metatheory. At least as far

as the concept of Self goes, he is happy with the representation of the body but wants

to supplement it with the narrative metaphor.


Bartlett, F. (1932): Remembering. Cambridge: Cambridge University Press.

Beckett, S. (1931): Proust. Cited after the edition of 1987, London: J. Calder.

Bergson, H. (1896): Matière et mémoire. Paris: Alcan. References are given to the critical edition:

Oeuvres 1959, Paris: Presses Universitaires de France.

Binet, A. and Henry, V. (1894): “La mémoire des phrases (Mémoire des idées).” Année Psychologique,

1, 24–59.

Black, J. B. and Bower, G. H. (1980): “Story understanding as problem solving.” Poetics, 9,


Blonskij, P. P. (1935): Pamjat i mislenie. Moscow: Ogiz-Socekgiz.

Brooke-Rose, C. (1986): “The dissolution of character in the novel.” In: Heller, T. C., Sosna, M., and

Wellbery, D. E. (eds): Reconstructing Individualism: Autonomy, Individuality, and the Self in

Western Thought. Stanford, CA: Stanford University Press.

Bruner, J. (1985): Actual Minds and Possible Worlds. Cambridge: Harvard University Press.

Bruner, J. (1990): Acts of Meaning. Cambridge, MA: Harvard University Press.

Bruner, J. (1996): The Culture of Education. Cambridge, MA: Harvard University Press.

Bruner, J. and Hastie, H. (eds, 1987): Making Sense: The Child’s Construction of the World. London:


Bruner, J. and Luciarello, J. (1989): “Monologues as a narrative recreation of the world.” In: Nelson,

K. (ed.). Cambridge: Harvard University Press, 234–308.

Camus, A. (1954): The Stranger. Trans. by Stuart Gilbert. New York: Vintage Books.

Carrithers, M., Collins, S., and Lukes, S. (eds, 1985): The Category of the Person: Anthropology,

Philosophy, History. Cambridge, MA: Cambridge University Press.

Churchland, P. M. (1995): The Engine of Reason, the Seat of the Soul: A Philosophical Journey into

the Brain. Cambridge, MA: MIT Press.

Colby, B. N. (1973): “A partial grammar of Eskimo folktales.” American Anthropologist, 75,


Csibra G., & Gergely, G. (1998): “The teleological origins of mentalistic action explanations: A developmental

hypothesis.” Developmental Science, 1, 255–259.

Damasio, A. (1994): Descartes’s Error. New York: Basic Books.

Dennett, D. (1987): The Intentional Stance. Cambridge, MA: Bradford Books/ MIT Press.

Dennett, D. (1990): “The interpretation of texts, people and other artifacts.” Philosophy and

Phenomenological Research, 1, Supplement, 177–194.

Dennett, D. (1991): Consciousness Explained. Boston: Little Brown.

Dennett, D. (1994): Darwin’s Dangerous Idea. New York: Simon and Schuster.

Dennett, D. (1996): Kinds of Minds. New York: Basic Books.


Dennett, D. and Kinsbourne, M. (1992)· “Time and the observer. The where and when of consciousness

in the brain.” Behavioral and Brain Sciences, 15, 183–247.

Frege, G. (1884/1950): The Foundations of Arithmetic. Trans. by J. L. Austin. New York: Philosophical


Gergely, Gy., Nádasdy, Z., Csibra, G. and Bíró, S. (1995): “Taking the intentional stance at 12

months of age.” Cognition, 56, 165–193.

Gide, A. (1914/1953): Lafcadio’s Adventures. Trans. by Dorothy Bussy. New York: Vintage Books.

Goldstein, J. (1994): “The advent of psychological modernism in France: An alternative narrative.”

In: Ross, D. (ed., 1994): Modernist Impluses in the Human Sciences. 1870–1930. Baltimore:

Johns Hopkins University Press, 190–209.

Graesser, A. C. (1992): Questions and Information Systems. Hillsdale, NJ: Erlbaum.

Graesser, A. C. (1996): Models of Understanding Text. Mahwah, NJ: Lawrence Erlbaum Associates.

Graesser, A. C. and Clark, L. F. (1986): Structure and Procedures of Implicit Knowledge. Norwood,

NJ: Ablex.

Greimas, A. J. (1966): Sémantique structurale. Paris: Larousse.

Grice, P. (1975): “Logic and conversation.” In: Cole, P. and Morgan, J. L. (eds): Syntax and Semantics.

Vol. 3. Speech Acts. New York: Academic Press.

Halász, L., László, J. and Pléh, Cs. (1988): “Cross-cultural studies in reading short stories.” Poetics,

17, 287–303.

Halbwachs, M. (1925): Les cadres sociaux de la mémoire. Paris: Alcan.

Halbwachs, M. (1950): La mémoire collective. Paris: Presses Universitaires de France. 2nd ed.:


Halbwachs, M (1992): On Collective Memory. Chicago, IL: University of Chicago Press.

Halbwachs, M. (1994): Les cadres sociaux de la mémoire. Paris: Albin Michel. New edition of

Halbwachs, 1925.

Hassan, I. H. (1987): The Postmodern Turn. Columbus, OH: Ohio State University Press.

Hassan, I. H. (1990): Selves at Risk. Madison, WI: University of Wisconsin Press.

Janet, P. (1928): L’évolution de la mémoire et de la notion du temp. Paris: Alcan.

Kundera, M. (1986): The Art of the Novel. New York : Grove Press.

Labov, W. and Waletzky, J. (1967): “Narrative analysis: Oral versions of personal experience.” In:

Heim, J. (ed.): Essays on the verbal and visual arts. Seattle: American Ethnological Society,


László, J. (1986): “Scripts for interpersonal situations.” Studia Psychologica, 28, 125–135.

Latour, B. (1993): We Have Never Been Modern. New York: Harvester.

Lodge, D. (1988): Nice Work. London: Sexker and Warburg.

McClelland, J. L. and Rumelhart, D. (eds, 1986): Parallel Distributed Processing: Explorations in

the Microstructure of Cognition. Vol. II. Cambridge, MA: MIT Press.

Mach, E. (1897): Contributions to the Analysis of Sensations. Trans. by C. M. Williams. Chicago,

IL: Open Court.

Mach, E. (1910): Popular Scientific Lectures. 4th ed. Trans. by Thomas J. McCormack. Chicago,

IL: Open Court.

Mach, E. (1976): Knowledge and Error. Sketches on the Psychology of Enquiry. Trans. by Thomas

J. McCormack and Paul Folkes. Dordrecht: D. Reidel.

Modell, A. H. (1993): The Private Self. Cambridge, MA: Harvard University Press.

Nelson, K. (1989, ed.): Narratives from the Crib. Cambridge, MA: Harvard University Press.

Nyíri, J. K. (1992): Tradition and Individuality. Dordrecht: Kluwer.

Perry, J. (ed., 1985): Personal Identity. Berkeley: University of California Press.


Pléh, Cs. (1987): “On formal- and content-based models of story memory.” In: Halász, L. (ed.): Literary

Discourse. Aspects of Cognitive and Social Psychological Approaches. Berlin: De

Gruyter, 1987, 100–112.

Pléh, Cs. (1989): “A Hungarian Bergsonian psychologist: Valéria Dienes.” Hungarian Studies, 5,


Pléh, Cs. (1999): “Ernst Mach and Daniel Dennett: Two evolutionary models of cognition.” In:

Fleissner, P. and Nyíri, J. C. (eds): Philosophy of Culture and the Politics of Networking. Vol. I.

Insbruck–Wien: Studien Verlag, 13–25.

Pléh, Cs. (2000): “Remembering the collective memory of Maurice Halbwachs.” Semiotica, 128,


Pléh, Cs. (2002): “Narrative construction of texts and the Self.” In: László, J. and Stainton Rogers,

W. (eds): Narrative Approaches in Social Psychology. Budapest: New Mandate, 90–109.

Popper, K. R. (1972): Objective Knowledge. An Evolutionary Approach. Oxford: Clarendon Press.

Propp, V. (1928/1958): Morphology of the Folktale. Bloomington: Indiana University.

Ricoeur, P. (1965): De l’interprétation. Essai sur Freud. Paris: Fayard.

Ross, D. (ed., 1994): Modernist Impluses in the Human Sciences. 1870–1930. Baltimore: Johns

Hopkins University Press.

Rumelhart, D. E. (1975): “Notes on a schema for stories.” In: D. G. Bobrow and A. N. Collins (eds):

Representation and Understanding. New York: Academic Press, 211–236.

Schank, R. (1975): “The structure of episodes in memory.” In: D. G. Bobrow and A. N. Collins

(eds): Representation and Understanding. New York: Academic Press, 237–272.

Schank, R. and Abelson, R. P. (1977): Scripts, Plans, Goals, and Understanding. Hillsdale:


Todorov, T. A. (1969): Grammaire du Décaméron. Paris/The Hague: Mouton.

Toulmin, S. (1990): Cosmopolis: The Hidden Agenda of Modernity. New York: Free Press.

Zinchenko, P. I. (1961): Nyeproizvolnoje zapominanyije. Moszkva: Izd. APN.