László Halász

Who's afraid of Siegfried Schmidt?

When Fechner (1897) established experimental psychology and experimental aesthetics practically simultaneously, he sharply confronted inductively-empirically based aesthetics with a deductive-normative philosophical one. In this frame, however, »empirical« is a wider notion than »experimental«, and for a psychologist dealing with arts it has meant an unambiguous engagement. It has meant a break with a speculative approach from an »armchair«, and has connected theorizing with a systematic data-gathering verifiable by others.

Thus, for me as a (social) psychologist interested in studying literary understanding, »studying« has always been obviously equivalent to investigating the phenomena in question empirically based on some theoretical considerations and putting forward a theoretical construction based on some empirical considerations. But during my research work I soon had to face up to the evidence of my own eyes, evidence of which literary scholars were unaware. And I had to see at best they did not understand the role and possibilities of the empirical study of the psychology of literary understanding owing simply to their lack of relevant knowledge. Usually their attitude was categorically negative or mistrustful they denied the need for such an area of science and/or thought it empty play. They were certain that what such psychological studies could discover was already known to them, or that it was not worth knowing.

It is easy to imagine how much I was encouraged when nearly two decades ago I first read the attractive phrase: »empirical study of literature«, coined by Siegfried Schmidt. At that moment without knowing anything definitive as to what the phrase really meant, I felt just the opposite of Moliere's famous protagonist who was unaware that he had been speaking in prose his whole life. My idea was that I had been working - of course, not alone - on the empirical study of literature even before its foundation. By the way, this idea was correct. At the time of ESL's founding, Klemenz-Belgardt's paper (1981), which referred to more than a hundred items based only on studies in the US, gave a picture of the wide spectrum of related activity.

No doubt that there were some psychologists, sociologists and linguists dealing with literature who first thought, as I did, that ESL was practically an interdisciplinary legitimacy of the approaches and efforts they had tried mainly in the previous four decades. Naturally, we all recognized that systematic interdisciplinarity might contribute to enlarging knowledge of both literary scholarship and psychology, sociology and linguistics.

But on reading Schmidt's main ideas it quickly became clear that he required something different. ESL is not »an empirisation of literary studies which tries to solve old questions with new (...) methods« ESL »tries to look through new glasses at a much broader and partially innovative field of problems (in the LITERATURE-system) which (...) intends to solve as far as possible and reasonable with empirical methods« (Schmidt, 1981, 324).

I am afraid that the consequences of this cautious but clear proposition on how to deal with empirical methods were more significant than Schmidt could have realized at that time. Considering the fact that nobody could predict where the limits of possibility and reasonableness were, »empirical men« like myself have tried to show heightened interest in Schmidt's ideas and Schmidt has to some extent reciprocated. To avoid any exaggeration, I mean to say that his interest was at least greater than that of the majority of literary scholars who followed »a hermeneutic tradition (in the broad sense)«, to use his phrase (Schmidt, 1997, 140). This reciprocity was more than enough for these literary scholars not to accept ESL. And we should not forget that this was only one factor (see below). Thus, ironically the interdisciplinary legitimacy within ESL, instead of contributing to the organic embedding of the empirical sociological, psychological and mathematical-linguistic studies of literature on the one hand, and of Schmidt's (and his closest colleagues') work on the other hand, into literary scholarship, has institutionalized the sharp borderlines.

When I began my research work I was rather uncertain about many issues but there was one point on which I had no doubt. Namely, I thought that I knew what a literary work was. (Nowadays - with some exaggeration - I could say that this is the very point on which I have nothing but doubts.) I thought my task was to find reliable ways to ascertain how a reader could identify the meaning of it, or if he/she could not, then what sort of (personality) traits could be shown. Within some years I had to realize that this was not a fruitful idea. It was more instructive when by systematic falsifying-simplifying an unambiguously literary (according to judgment of experts) text from different semantic, structural and stylistic points of view I made a whole series of texts of the same topic and length partly along the dimension of original - banal, and partly along the dimension of fictive-nonfictive. I could study how a reader to whom the name of the author were not given and to whom all the versions (including the original) were unfamiliar was able to understand, remember and evaluate each of them separately, then compare them. During the seventies, to a certain degree I could see within the same frame of reference and within the same context how a highly educated expert group and a less educated, but not completely naive group of readers constructed meanings and values, or if you wish, constructed their own texts. This process and result were not only characteristic of the nature of their understanding but of that of the text-types themselves. (Following their publication in Hungarian in 1970s, I published them in English: Halász 1980, 1986 and in German: Halász 1993.)

That is why I was rather receptive to a specific consequence of Schmidt's proposition that ESL »is not a plea for an obsolete positivism or empiricism. The concept of ›empirical‹ is not at all a subjective, but is based on the intersubjectivity provided by the common language of a group of individuals (...) The living system constructs knowledge according to his organization and to consensual processes in his social group« (Schmidt 1981,320-321). The idea itself that our psychological processes are constructions is close to modern psychology. First, one may think of Piaget's constructive structuralism (1970), Neisser's cognitive constructionism (1976), and as a somewhat later development, the social constructivist point of view of emotions (Harré 1986).

But none of these said directly anything relevant about text understanding. This was precisely what Schmidt did (l981:323): »texts are considered as components of actor-text-syndrome«, which meant an attempt to overcome text-reader dualism. Schmidt (1981, 322) emphasized that »›literariness‹ as well as ›meaning‹ are not intrinsic qualities of the text but the result of cognitive operations on the part of the recipient.« These operations are regulated by aesthetic and polyvalence conventions. It seemed to me, and I highly appreciated it as Schmidt would have said that it was hopeless to see a literary text as a rigid and clear entity without studying the conditions and course of the meaning assignment there could be no literary scholarship.

Quite independently of his radical constructivist epistemology, in that respect Schmidt was not alone. He was not even the first to oppose the idea that a literary text as such could be studied. He had different »associates« who also opposed the idea conceptualized by Jakobson in its most categorical form. »The object of study in literary science is not literature but ›literariness‹, that is, what makes a given work a literary work« (Jakobson 1923, quoted in Pratt, 1977, 27). As Van Dijk (1979, 593) emphasized: »... there are no decisive textual properties which alone make a text literary«. Any text's intrinsic properties are »the manifestations of its function«, said Pratt (1977, 31), who stressed that it depended on the reader's decisions whether to read a work as literature or not. As Fowler put it (1979, 544), »...›literariness‹ should be shifted from texts themselves and re-identified in terms of the various productive and receptive behaviours of people in different literature-using societies.«

All these statements are double-edged. They emphasize how many kinds of texts may have exerted literary effects - at least for about two hundred years (see Sell 1991, Schmidt 1991). By all means, present-day readers are willing, in accordance with the view of our narrative culture, to process a wide variety of texts as stories that express human fate in emotional states and can be interpreted in aesthetic and polyvalent ways. It is precisely the consequence of this willingness that under given circumstances we may take even a laconic and simple text to be rich in meanings/meaningful structures (Halász 1986, 1993). It may be surprising but this empirical finding is in full accordance with Hirsch's proposition (1976, 134) as well: »Literature comprises any linguistic work, written or oral, which has significant aesthetic qualities when described in aesthetic categories.« Ironically speaking, if a text is characterized in such a way owing to our (in)competence, remarkable aesthetic qualities may be disclosed. Or as Short and Candlin (1989:202) put it: »... if the readers feel some need to process a text as a literary artifact (...) they will attempt to apply a set of special interpretative conventions.«

How can we explain why such statements and conclusions - which are basically quite close to Schmidt's standpoint that »... literature is what observers deem ›literary‹ according to the concepts and values by which they establish and handle the difference between the literary and the non literary. These concepts are acquired by experiencing prototypical cases of both sides of the difference in the course of literary socialization« (Schmidt 1997, 144) - have not aroused all in all as much resistance as Schmidt's ideas themselves?

In addition, even in his Foundations (1982, 56) Schmidt tried to fend off the expected charge of ontological fallacy, emphasizing that people cannot »treat art works with complete subjective freedom. In all societies, there are highly conventionalized rules for assigning meaning and relevance to the means of communication.« He failed in this attempt. He had to repeat his arguments (Schmidt in Schmidt et. 1989, 41): »I do not deny the existence of texts (...) Text-features are intersubjectively (= consensually) attributed to texts are no objective (=ontological) ingredients of text-objects.«

I myself also judged Schmidt's standpoint to be more provocative and risky than the ideas of the other authors presented above (Halász 1994). The common trait of the other ones was functionalism, which required a socio-psychologically penetrated cognitive constructionism, but which did not imply a radical constructivism to construct an all-embracing literary theory to avoid text-reader dualism. I pointed out that a consequent satisfaction of this demand opened the way to a complete psychological relativism. If essentially everything depended on the reading practice determined by conventions, then the number of possible and acceptable interpretations - although they were not arbitrary, that is, not totally at random - would be infinite and equivalent in principle. I knew that this accorded with postmodern requirements, and it was true that a literary reader could not, of course, test his/her interpretation(s) by falsification, but all this did not alter the fact that what was to be interpreted should never be lost from sight.

No matter how enigmatic and ambiguously or loosely structured a literary text might be, and how little it required specific formal linguistic characteristics, I realized that the problem was never that it had no meaning at all like a projective test, but that it had a circle of meaning potential or meaning set. Some cues might be easily available, while others were more obscure and hidden, so that the reader's interpretive activity assumed a greater role. It was clear to me that if the reader was able to process only traditional narrative texts, owing to his/her complete incompetence a non-traditional literary text would lose its communicative potential and function for him/her as a mere projective test. The reader responded by projecting his/her own needs, desires, idiosyncrasies if given the standard instruction of a projective test. But there was some limitation even in this case. Because of differences in provocative or evocative strength the characteristics of a test-stimulus could not be discounted, as it could not even in case of the most robust projective test. The subject's possibilities were always limited (if you like: conventionalized) in a rather broad way by the characteristics of the stimulus.

Supposing, but not conceding, that in the course of using conventions the reader's non-intersubjective (non-consensual, par excellence subjective) meaning-assignment processes were eliminated, even in that case the reader's relationship to the text would not be dissimilar to that of a subject who responded to a projective test in accordance with his/her group's attitudes: in some respects with tacit conventions about the (meaning of) the given stimulus. That means that emphasizing the use of highly conventionalized rules in a frame of radical constructivism and claiming to overcome text-reader dualism is unfortunately no defence against the dangers of psychological relativism.

There is no doubt that the receiver of an expository text confronts a clearly defined problem, one cannot say the same with the text which a given culture, and especially its expert opinion, judges literary. In literary text the goal-state is not clear, the problem is »ill-defined« (cf. Reitman 1965). In spite of this we need not conclude that the role of a text is unimportant compared with that of »interpretive communities«, as Fish (1980) does, and Schmidt (1982:165) is more than willing to agree with him. We cannot identify the degree of freedom of a writer and a reader when constructing a possible world. We cannot forget what the one has already created (formed, constructed, organized) is given for the other.

Our findings (Halász, 1996) showed that the general meaning assignment strictly followed what the text was communicating about, and even in the personal meaning processing these connections were not insignificant. That is why I think it too categorical to say that »the text does not strictly or automatically trigger the respective cognitive process in a foreseeable manner nor does it determine their results« (Schmidt 1991a, 276). Without supposing, of course, any strict or automatic triggering effect, the role of the text in the reader's cognitive processes and their results were foreseeable to a certain degree. At the same time, these findings suggested that a literary text experienced as literary by the more or less naive readers as well was certainly more appropriate than any other text on the same topic to learn how to construct personal responses.

No matter how it is, we cannot claim that the main difference between a page of »Paradise Lost« and a Rorschach blot is not in these themselves but in their communal interpretation (cf. Scholes 1986) or in cognitive operations regulated by some conventions, although there is a special kind of communion between letters on the paper becoming literary text and its reader. »Each becomes in a sense environment for the other. A two-way, or better, a circular process can be postulated, in which the reader responds to the verbal stimuli offered by the text, but at the same time he must draw selectively on the resources of his own fund of experience and sensibility to provide and organize the substance of his response. Out of his new experience, the literary work, is formed« (Rosenblatt, 1978:43). I find this transactionalism as a middle-level theory puts firmly text-reader dualism into brackets.

I realize what is enough for me it cannot be enough for Schmidt. Founding ESL, he has cut a wide road to unite all the possible metatheoretical and empirical - philosophical, historical, linguistic, sociological and psychological - researches systematically, to present a multidisciplinary approach. And soon he has found that even so the road was not wide enough. His ideal is to have a comprehensive theory, which means in his eyes as comprehensive a theory as possible. Considering that »the literary system is never determined within the literary system alone« and literary socialization is »embedded in the more complex process of media socialization«, literature is only one component of the media system (Schmidt 1992, 233). Thus, ESL is a special part of EMS (empirical media studies). The main issue is where the point is. In my studies to obtain the necessary frame of reference for the data of reader's reception and postprocessing I have also been investigating the subjects' attitude towards theatre, movies, museum, radio, TV, press, computer, the Internet, but that is no more than background and the form, the foreground is literature, and nothing else. That is ESL is sufficient why to identify these works.

But independently of this extra problem, I considered that radical constructivism was far from being indispensable to solve a (psychologically or sociologically) fundamental problem of ESL, but some solutions might have been somewhat easier without that theory than with it. I was especially afraid that to following his own efforts after a grand theory, when developing it Schmidt would incline to a premature renunciation of the use of empirical methods (Halász 1994). Now I have to say that I was not right. Discussing Von Foerster's proposition about »the methodologically controlled production of data as trivialization«, Schmidt (1997, 146-147) unambiguously sided with the scientific empirical research in the sense that I also do.

At the same time, I have understood better than I did earlier both Schmidt's and his rigid adversaries' attitudes. Schmidt is unwilling to give up (only to refine) radical constructivism as a philosophical basis, just as he is unwilling to give up the empirical method as a scientific research means of ESL, better to say EMS. By this particular composite the multiplied deterrent effect is, however, guaranteed for most literary scholars and experts in literary/media education simultaneously. First, it is not a negligible minority among them who feel at home without any philosophy and any scientific research. But even if we focus on those who realize the adequacy of philosophy, they certainly do not think of radical constructivism, and even if they were willing to accept it, they would be shocked by the demand of studying other media as literary scholars or doing the same with literature as media experts, and the rest would be put off by, and averse to, the role of scientific research method. As a counter-proof you can start from the opposite direction and first focus on those who as literary and/or media scholars realize the adequacy of scientific research method, then you go on up to radical constructivism as a terminus, the result will be the same. If Schmidt would say that all these show not so much comfort but the strength for defending acquired positions, that is, for preserving power in literary scholarship, I would hardly contradict him.

Still there is another issue which I have to talk about. ESL, if you prefer, EMS being as comprehensive a theory as it is, cannot consist of directly testable hypotheses but rather it provides us with a frame for further research directions. Schmidt's main aim - I think I diefine it correctly - has been to reach a focal shift and launch studies on the text-thematizing activities of producers, mediators, recipients and postprocessors of literary and other media phenomena. This would require a systematic and possibly simultaneous historical, sociological, linguistic and psychological study of the writer (and other author) and his/her activities as creative process, of the work itself, of the reception and impact of the work, of the mediation process (primarily literary and media education). Based on the presented resistance against ESL it is obvious why a wide circle of literary and media scholars has not accomplished such studies.

But if we remain in our circle we have to see that the focal shift has been realized rather fragmentarily even here. Unfortunately, we do not have the mentioned systematic studies in abundance at all. The writer's and other author's activities especially have not yet achieved sufficient prominence. Although the reading process has been relatively well-explored, it is rather peculiar that we can say more about expository text reading than about poem or novel reading. Without studying them using scientific method the ecological validity of ESL is doubtful. Yet as I have already pointed out (Halász 1995) this validity is of primary importance from the point of view of the efficiency of the studies of post-processing and mediation process. Thus, it is rather sad that there is no close connection between the works on different areas of ESL, and even the coherence between the works within the same area is weak.

Although my argumentation was rather different from Schmidt's (1997, 147) when at the end of his exposition he questioned whether IGEL as a representative gathering of ESL was more than »a biannually meeting flock of scholars«, as an old member of the flock I could not say that I did not understand his bitterness. Nevertheless, I have not shared it absolutely. Not only for practical reasons because some ESL colleagues' works - sometimes influenced by Schmidt - have been giving me direct or indirect stimulus during my research activity. In general, I am moderately optimistic about the future of ESL. I suppose what has not been reached by a sudden paradigm shift can be reached slowly step by step, even if these steps are not optimally coordinated. Maybe my imagination is too bold, but in the long run I cannot imagine literary scholarship without solid empirical basis.


Fechner, G.T. (1897), Vorschule der Aesthetik, Leipzig: Breithopf und Haerted.

Fish, S. (1980), Is There a Text in This Class? The Authority of Interpretive Communities, Cambridge, Mass.: Harvard University Press.

Fowler, R. (1979), »Preliminaries to a sociolinguistic theory of literary discourse«, Poetics (ó8. 531-556.

Halász, L. (1980), »Organization of judgments of works of art and psychological characteristics of value recognition,« in: Kardos, L., Cs. Pléh and Gy. Hunyady (eds.) Attitudes, Interaction and Personality, Budapest: Academic Publishing House.127-149.

Halász, L.(1986), »Remembering literary works and non-literary variations«, in: SPIEL 5, 23-48

Halász, L. (1993), Dem Leser auf der Spur: Literarisches Lesen als Forschen und Entdecken. Zur Sozialpsychologie der literarischen Verstehens, Braunschweig, Wiesbaden: Vieweg.

Halász, L. (1994), »Die Grundlegung einer Empirischen Literaturwissenschaft« - von einem psychologischen Standpunkt aus betrachtet, in: Barsch, A., G. Rusch und R. Viehoff (eds.) Empirische Literaturwissenschaft, Frankfurt M. Suhrkamp, 86-103.

Halász, L. (1995), »IGEL - Seven years after«, in: Rusch, G. (ed.) Empirical Approaches to Literature, Siegen: LUMIS, 10-15.

Halász, L. (1996), »General and personal meaning in literary reading.« in: Kreuz, R.J. and M. Sue MacNealy (eds.), Empirical Approaches to Literature and Aesthetics, Norwood, N.J.: Ablex, 397-418.

Harré, R. (1986), The Social Construction of Emotions, Oxford-New York: Basic Blackwell.

Hirsch, D.E.(1976), The aims of interpretation Chicago: Chicago University Press.

Klemenz-Belgardt, E. (1891), »American research on response to literature: the empirical studies«, in: Poetics 10, 357-380.

Neisser, U. (1976), Cognition and Reality, San Francisco-Oxford: Freeman.

Piaget, J. (1970), Structuralism, New York: Harper.

Pratt, M.L. (1977), Toward a Speech Act Theory of Literary Discourse, Bloomington: Indiana University Press.

Rosenblatt, L.M. (1978), The Reader, the Text, the Poem: The Transactional Theory of the Literary Work, London-Amsterdam: Southern Illinois Press.

Sell, R.D. (1991), »How can literary pragmatics develop empirical methods? The problem of modal and evaluative expressions in literary texts«, in: Ibsch, E., D. Schram and G. Steen (eds.) Empirical Studies of Literature, Amsterdam: Rodopi, 137-145.

Schmidt, S. J.(1981), »Empirical studies in literature: introductory remarks«, in: Poetics, 10. 317-336.

Schmidt, S. J.(1982), Foundations for the Empirical Study of Literature, Hamburg: Buske.

Schmidt, S. J.(1991), »Literary systems as self-organizing systems«, in: Ibsch, E., D. Schram and G. Steen (eds.) Empirical Study of Literature, Amsterdam: Rodopi, 413-424

Schmidt, S. J.(1991a), »Text understanding. A self-organizing cognitive process«, in: Poetics 20. 273-281.

Schmidt, S. J.(1992), »Why literature is not enough or Literary studies as media studies«, in: Cupchik, G.C. and J. László (eds.) Emerging Visions of the Aesthetic Process: Psychology, Semiology, and Philosophy, Cambridge: Cambridge University Press, 227-243

Schmidt, S. J.(1997), »The empirical study of literature: why or why not?« in: Tötösy, S. and I. Sywenky (eds.), The Systemic and Empirical Approach to Literature and Culture as Theory and Application, University of Alberta and Siegen University, 137-153

Schmidt, S. J. and N.Groeben (1989), »How to do thoughts with words: on understanding literature«, in: Meutsch, D. and R. Viehoff (eds.), Comprehension of Literary Discourse: Results and Problems of Interdisciplinary Approaches, Berlin-New York:de Gruyter, 16-46.

Scholes, R. (1986), Textual Power, Literary Theory and the Teaching of English, New Haven and London: Yale University Press.

Short, M. and Ch. Candlin (1989), »Teaching study skills for English literature«, in: Short, M. (ed.), Reading, Analysing and Teaching, London and New York: Longman, 172-202.

Van Dijk, T.A. (1979), »Advice on theoretical poetics«, in: Poetics 8. 569-608.