JLT Articles, Vol 1, No 1 (2007)

 Tom Kindt

Uri Margolin


More than a quarter century ago, I had the audacity to publish a long article (Margolin 1979) in which I discussed the nature of literary theories and their relation to theories in other disciplines in the humanities. The kind invitation of the editors of JLT affords me a golden opportunity to revisit these perennial crucial questions. Now while my 1979 essay appeared in the very last issue of PTL, the present one appears in the first issue of a new journal, which, I trust, has many years of fruitful intellectual activity ahead of it.

How Theoretical Is Literary Theory?

The Editors’ twin questions to the participants concern the perspectives and challenging problems of literary theory (LT). I would like to reformulate them as »what can and should LT do?« Now in English usage »LT« means three interrelated things: a particular LT, the totality of LTs available, and the activity of theorizing or theory construction. I will focus my remarks on the activity of theorizing, since its capabilities and limitations determine the nature of the resultant theories. But what is a LT? A minimum definition would probably look something like the following: A LT is a theory (1) intended to account for one or more aspects of the literary system – either as semiotic system or as social action one – (2) as this system is demarcated in current pre-theoretical cultural awareness (in other words, (3) it is a theory with initial and intended domains of application covering at least part of the literary system), and usually, but not always, (4) formulated within the disciplinary confines of literary studies. All literary theories contain a factual component since they seek to account for actual (and sometimes also possible) space-time anchored phenomena: objects (texts and text elements), events and processes (literary change), and, of course, activities of literary production and reception. But how do we assess what such a theory can and cannot do? A sensible way would be to scrutinize actual LTs against a roster of components, tasks and roles theoretical endeavours in any field are ideally supposed to fulfill according to philosophy of science. The image of scientific theorising I will be relying upon is the synthetic one presented in Mario Bunge’s comprehensive treatise (Bunge 1998). I will treat this image as an ideal type or regulative idea in the Kantian sense and check how far (existing) LTs, especially text oriented ones, (can) go in matching it, the results being treated as value neutral.

One overriding goal which most LTs seek to achieve is to define a series of regularities or generalities, determinate features or patterns with respect to a certain range of phenomena in the literary domain, from metrical forms to the social constraints on literary production. Some theorists go further and look for literary invariants or universals as regards text types, basic plot structures (Frye, Shklovski), the structure of verse (Jakobson’s law of equivalence), underlying emotions (Hogan), or the laws governing the relations between two literary systems (Even-Zohar). The generality requirement is in fact the yardstick that tells apart poetics (theoretical, descriptive, diachronic) from textual interpretation. The quest for regularities has rationality as its boundary condition. Informally, rationality in theorizing can be characterized as a set of requirements for explicit formulation of both problems and the solutions offered, intersubjectivity, that is, proceeding in a way which can be both learned and taught (›lern- und lehrbar‹), respect for the rules of logic, consistency, and, in our context, formulating claims which can be tested notionally or empirically (Hauptmeier/Schmidt 1985, 13 and 118).

As far as content is concerned, LTs, like all other theories, contain philosophical presuppositions of some kind; background theories they take as valid knowledge to be drawn upon; goals, including the kind of knowledge they want to attain; methodological norms, i. e. rules concerning the procedures to be employed to this end (e.g. top down or bottom up), and the way claims should be formulated (e. g. formalized or not). While none of these components may be explicitly stated by a given theory, its logical reconstruction will reveal their existence and nature.

The building blocks of theorizing proper begin with a set of concepts, expressed in a theoretical vocabulary, whose elements, in the case of LTs, being either deliberately coined terms (e. g. free indirect discourse), or ones consisting of an explication and sharpening of pre-theoretical terms (e. g. plot). This vocabulary is employed in order to designate the objects and features a theory is concerned with. The terms occurring in a given theoretical project are of course interrelated, to form a coherent network that provides the prism or manner of seeing (›Sehweise‹) to be employed in theorizing the targeted range of phenomena. An extensive theoretical vocabulary is in fact the basis for developing a systematic description of phenomena in terms of determinate features and their combinations, and for developing a set of distinctions and categories for any literary aspect. This has been recognized by poeticians from Aristotle to Genette. Classifications, taxonomies and typologies in their turn are the first step in mustering the plurality of data and subsuming them under a small number of headings or common denominators referring to entities, properties or relations. The categories themselves may be defined in terms of binaries, gradients or prototypes, and their whole system graphically represented as continua, matrices, trees, box diagrams or circles (e.g. Stanzel’s typology of narrative situations).

Another component widespread in literary theorizing consists of models, that is, simplified schematic representations of selected aspects of complex real objects and the interrelations of these aspects. Such models (model objects or theoretical referents) are formulated in terms of the theoretical vocabulary employed by the scholar and provide a basic way of seeing the object and thinking about it. We thus have the semiotic model of the literary work as a complex sign (Barthes, Corti), Ingarden’s model of it as a stratified system, and Tynjanov’s model of the verse text as a dynamic system of systems. In narratology, the text is often modeled as consisting of several stacks or embedded circuits of communication, as a dual narration-narrated system and so on. Much of what is termed »literary theory« is in fact the formulation of concepts, identification of recurrent features, and the setting up of typologies and models, all of which deal with existential and compositional issues (see Soerensen 1987, chapter 5). These operations do not provide, however, any claims which can be supported or rejected but rather directions what and how to observe in texts, and they are accepted if professionals feel they serve as illuminating cognitive instruments, if they help us subsume, unify and integrate numerous textual features.

Categories and model objects are the first part of any theoretical project, but are not yet theories sensu strictu, since theories are basically sets of claims serving as answers to questions about model objects, or solutions to problems concerning them. So questions need to be formulated to guide theory construction. Questions consist of logical operators such as what, which, how many, how, why and what for, and of substantive referents and predicates specified by a theoretical vocabulary. LTs abound in questions such as ›what are the relations between the different levels of the literary text‹, which can be further specified into many sub-problems such as sound-sense relations, syntax and metre and so on. A different kind of question would be ›what are the major ways of representing consciousness in narrative, and do they form a system and/or a clear historical sequence‹. Problems tend to cluster into sets or problem systems, partially ordered by the relation of logical priority. Such problem systems define what a theory is about and what its goals are in terms of knowledge production. Once again, while not all literary theories formulate their questions explicitly and systematically, they are all guided by the desire to answer specific questions.

Questions require answers, and these are provided by sets of claims/hypotheses about the model object in focus. As already indicated, a theory is minimally a set of interconnected generalized claims about a certain domain of phenomena, formulated in a specific theoretical vocabulary, such that some of the claims and the concepts they employ are more basic than others, forming the theory core. Theories are meant to first describe and then explain, and when possible predict, regularities in a domain. LTs, especially in descriptive poetics, synchronic or diachronic, provide a large number of sets of interconnected descriptive or observational generalisations of different coverage or scope, scope being defined as range X accuracy (Bunge 1989, vol. 1, 575). Such theories are usually referred to as mid-range theories, located between high level abstract ones (e. g., about the general mechanisms of literary change) and individual data (McHale 1994). Midrange theories deal with the occurrence and correlations (co-occurrence, co-variation, succession) of textual elements or patterns (e. g., genres), looking of course for regularities in any of them. They typically answer questions of the ›what‹ and ›which‹ variety. Ideally one would want to formulate empirical generalizations of the widest possible scope, hence indifferent to time and place. But very few such universal claims exist in literary studies. Mid-range theories have non-universal domains and are typically constructed around various paradigm or standard cases that serve as prototypes. Literary studies and many other disciplines, including most of biology for example, consist primarily of this kind of theories. The rapid change and inner complexity and diversity of literary phenomena go a long way towards explaining this situation. Although most descriptive generalizations in literary studies are qualitative, quite a few admit of quantification, hence of operations such as counting, comparing, assigning numerical values to a variable, calculating probabilities and frequencies (statistics) and plotting the results on graphs. This is true in metrics and stylistics, but also where the historical distribution of genre patterns for example is concerned.

But how does one check the validity of any purely qualitative descriptive generalization in a LT? Suppose one proposes a systematic typology of kinds of focalization. Can such a typology be corroborated or refuted, and, if so, how? Since such claims do not form part of a formal system, a deductive decision procedure or proof is not available. On the other hand, such claims are not open to experimental testing based on drawing predictions from them and then observing how they turn out in actuality. What is more, we have not agreed upon standards for deciding whether or not an individual utterance is an instance of a particular kind of focalization. In the absence of both deduction and experimentation we must look at these claims as not strictly true or false but rather as useful, illuminating and insightful or not, plausible, supported by good reasons, possessing warranted assertibility or not, all such decisions being based on a dialogue within a scholarly community governed by some consensual rules of informal argumentation, and of course by rationality. And the same goes for citing instances that are claimed not to be subsumable under a proposed generalization, i. e, counterexamples. Who and how decides that X cannot be fitted under a given pattern? This decision too can result only from a dialogue inside a scholarly community, and not by the discovery of a logical inconsistency or of experimental results that run counter to a prediction based on a generalization.

And how about the explanation of such occurrence or correlation? In most disciplines, sets of descriptive generalizations are considered low level theories, to be grounded in higher level, deeper or more basic, hence more powerful, ones, containing more basic entities. The theoretical inventory of any discipline thus forms a hierarchy of deeper and shallower theories, reflected in LT in the relation between theoretical and descriptive poetics. (For a detailed systematics of this sort see Ingarden 1976.) Ideally, the higher level theories provide the nomological basis (general laws) from which lower level regularities can be inferred. Such inference is often considered to be an explanation of the observed regularities, in the sense of an answer to the ›why‹ question. Another kind of explanation is of the ›how‹ kind, providing an underlying mechanism, that is, as set of entities and activities and their modus operandi such that these operations realize or give rise to the descriptive regularities in question. Does literary theorizing possess explanations of either kind? And does it possess other patterns of explanation such as the means-end or teleological one? There is clearly an abundance of ›basic‹ theories in literary studies, such as those about the nature of literature. But most of them in their turn depend heavily on theories in other disciplines formulated for the treatment of wider or more basic domains (see section 2 below). ›Basic‹ literary theories vary greatly in terms of their degree of formalization, conceptual clarity and observational operationalisibility. Mid-range theories cannot consequently be derived from the basic ones, since most of these basic theories provide only a ›Sehweise‹ which suggests aspects to be stressed, kinds of questions to be asked and types of descriptive generalizations to be looked for. They provide grounding, overarching principles and conceptual foundations for mid range theories, but not laws from which the more superficial or specific theories can be derived. The degree of vertical integration of literary theorising is hence quite low. Such basic theories also have very low empirical content, straddling the border between theories responsible for actual facts and objects in space and time and purely conceptual analysis, as in the philosophy of art.

There are, nevertheless, some procedures in literary studies which could be

considered similar to the why explanations in other fields. One example of derivation

of descriptive regularities from higher level principles is provided by the

structuralist procedure of constructing an exhaustive calculus of possibilities for, say,

metrical forms or tense aspect and modality. Here,

instead of supplying an empirically obtained list of categories, [one] establishes the most general logically possible pattern thereof. The [scholar], in approaching the design of some range of phenomena, must single out the simplest [most basic] items underlying these phenomena and then, by combining them in all possible ways, construct the most general universal pattern for the totality of observable data. Such a calculus further guides the scholar in search of new, not yet identified, categories which it predicts, (Melcuk 1985, 181f.; for an example concerning narrative time and modality see Margolin 1999)

The question why the descriptive system includes this particular set of possibilities and no other(s) can then be answered (explained) by the features and combinatorial possibilities of the underlying elementary units. Another kind of claim that provides a partial explanation of phenomena, short of a universal cause and effect one, is a claim regarding dependency between variables, pointing out the free and the dependent ones. Thus, the division of a verse text into lines is deemed to have a crucial effect on the resultant semantic structure of this text, and the choice of kind of narrator is decisive for the kind of valid information one can possess about the story world. The underlying high level generalization is obviously that forms of expression determine (or at least have a decisive influence on) forms of content, but no specific forms can be derived from it.

Text grammars and story grammars (trees, rewrite rules, derivations, transformations), where the final surface product or individual text emerges through a sequence of well defined operations, are more like generative grammatical or semantic rule systems, in that they provide production mechanisms, answering the ›how‹ question. AI computer programmes for story generation (Ryan, Meister) are meant to describe how exactly a given simple story sequence or story schema comes into existence from a set of basic semantic elements and patterns plus specified moves or an algorithm. Such programmes can also generate new, hitherto non-existent stories.

Yet in either case the value of the grammar or programme, especially when new stories are generated, is determined not by a story’s formal well-formedness but by its acceptability to members of a cultural group, just like sentences produced by any TG. Furthermore, even if a given pre-existing actual story or story schema is successfully generated by a particular algorithm, we may still ask why does such a schema exist and why is it widespread or not, culture bound or not, time bound or not. And to answer this kind of ›why‹ question we must go beyond semiotic objects to human agency, thus to questions of cognitive, cultural and possibly social regularities and mechanisms, and to construct multi level theories involving elements of at least two different kinds, such as textual and cognitive. Cognitive narratology and stylistics are engaged precisely in this kind of project, but it is still early days. Similarly, in the study of literary change (diachrony) one can describe regularities in the sequence of stages involved in the fate of any literary convention: from innovation to repetitiousness, from periphery or marginality to center or dominance and back again, from sub-literature to canonization or vice versa etc. But to explain why change occurs at a given point, direction and intensity one needs a deeper level of theorization, such as diminishment of effectiveness of devices as they become familiar (psychology of perception), changing cognitive needs and values of a group (social psychology), and social desire for innovation as means of distinguishing oneself (sociology). (On these explanations see Fokkema/Ibsch 2000, 83–85.)

Functional explanations of the means-end or teleological (in order to) variety abound in LTs, from classical rhetoric to Russian Formalism, and correlate textual elements, devices, procedures and structures to particular effects, be they semantic, affective or aesthetic. The choice and deployment of textual elements is thus motivated by their ability to create certain effects on the recipient (Verfremdung, surprise, suspense, freshness of sensation etc). This mode of explanation makes perfect sense since works of literature are messages (›Kommunikate‹) directed to someone and meant, like all communicative acts, to modify the receiver’s cognitive, emotive or normative set (›Einstellung‹) in some particular way(s). But two fundamental difficulties remain: as Meir Sternberg has convincingly shown, the device-effect relation is many to many and context dependent, so universal claims are impossible. One can at most claim that under certain conditions a given device D tends to be correlated with effect E. Secondly, who decides what is the effect of a given artistic device to begin with? Members of a homogeneous professional community sharing the same implicit assumptions may claim that device D necessarily creates effect E on »the reader«. But by »the reader« they actually mean themselves. A more fruitful move would be to regard any such claim as a causal hypothesis or prediction and then run tests on groups of non-professional readers to test its validity. Here once again we see the need to move from the purely semiotic to the empirical psychological dimension (Fokkema/Ibsch 2000, chapters 1 and 2).

LT contains at least one other kind of theory specific to the human sciences, namely theories whose objects, are not literary products but rather the informal »theories« about the nature of literature, its kinds, elements, functions and effects held by members of a given cultural community. This kind of study is referred to as theory-theory in cognitive studies, and its rationale is that the literary behaviour of members of a community will be decisively influenced by their beliefs about literature, no matter whether or not these beliefs are upheld by the literary theorist (the same way that human behaviour is influenced by folk beliefs about the working of the human mind, regardless of whether or not such views are upheld by psychology or sociology) and that any local historical explanation of the nature and change of literature at a given time and place must therefore take these »theories« into account.

Finally, theory assaying. As we have seen, experimentation is hardly a feasible way of assessing the merit of semiotic theories of literature, unless claims about effect or impact on actual people are involved. But numerous conceptual criteria formulated in the philosophy of science (Bunge 1989, vol. 2, 394–400) do apply in this field too. One could mention well formedness, consistency and valid mode of argumentation as formal criteria; linguistic exactness, conceptual coherence (the predicates expressed by the theoretical vocabulary are semantically homogeneous and interconnected) and eventual observational interpretability as semantic ones. Among epistemological criteria, consistency with much of the established knowledge in literary studies and related disciplines, ability to answer many of the underlying questions, depth, unifying power and ability to suggest further research are clearly good making features of any LT. Another important feature would be a theory’s elasticity or ability to accommodate, sometimes through internal modification, new unforeseen cases or ones initially ignored. This is most important in our field, since the object itself is subject to frequent major changes.

How Deep and Independent Is LT?

Literary theorizing is defined by its responsibility for a culturally demarcated domain of phenomena, »literature«, and not by any particular philosophical framework, set of presuppositions, theoretical vocabulary, concepts or methodology. In addition, everybody agrees the object itself is multi-aspectual and involves several levels of organization. There is consequently no prima facie restriction on the number and nature of theoretical frameworks that can legitimately be employed in the study of literature or any of its aspects. This by itself ensures a permanent multiplicity and diversity of approaches in literary theorizing, all of which possessing some initial justification i.e. pluralism. The different approaches coexisting at any point often differ in their basic theoretic terms, model objects and claims, or they refer to entirely different bodies of data (are incommensurable). The field of LT as a whole will thus inevitably be heterogeneous, often consisting of theories that are not inter-translatable. ›Synthesizing‹ them into one grand theory is hence practically impossible. The complexity of the object further implies that no single global theory could account for all observationally given aspects of literature – especially if we want to include both semiotic and social system ones – and that a number of partial theories will be needed for this purpose. Most of the approaches or paradigms brought to bear on literary questions or problems originate in numerous other disciplines, including philosophy. Literary theorizing as a whole is hence essentially multidisciplinary. Since new approaches and perspectives constantly arise in the sciences of culture, mind and society, and since many of them are relevant and potentially fruitful for some literary problem, the field of LT is one of frequent paradigm shifts, hence radical discontinuity. Theoretical understanding will be continuous, cumulative or improvable only within each given paradigm.

As crucial is the realization that the extension or domain of objects to be accounted for by LT always forms a subset of a wider domain, no matter what model object (›definition of literature‹) one employs: cultural artifact, message, media offering, work of art, work of the imagination/fiction, secondary modeling system etc. In traditional terms, it is but one species of a wider genus. Literary theorizing, unlike linguistics or psychology, does not consequently possess any natural domain, such as language or the human mind, and is in this sense non-fundamental. Now if literature under any given perspective is but one area of a wider domain, then all generalizations formulated in some other discipline(s) about this domain as a whole are eo ipso relevant and many are (potentially) applicable to literature as well. In other words, much theoretical knowledge about literature is (potentially) contained in the work of other disciplines dealing with the domain as a whole (linguistics, semiotics, cognitive science etc.) What of all this is immediately relevant to the work of a particular theorist will of course depend on his general perspective and on the questions he is dealing with. Conversely, many of the concepts employed in LT and the regularities formulated in it turn out to be applicable to a wider domain. Even-Zohar has gone so far as to claim that every single high level law formulated in LT on the basis of observation of literature, and having literature as its initial and intended domain, validly accounts for larger semiotic phenomena, and that the literary specificity of such laws resides only in their manifestation through particular materials and variations (Even-Zohar 1986, 79). In other words, the specificity of LTs is confined to the level of corpus- specific descriptive generalizations. Be that as it may, it is cleat that LT is hence an importing as well as exporting activity: borrowing, and checking for applicability to its own corpus, of wider-scope generalisations made in other disciplines (»All fictions are«, »All texts contain«), and offering these disciplines its own concepts, categories and claims for testing their wider applicability. In addition to this vertical process there is a constant two way horizontal transfer of concepts, models and claims between same level domain-specific disciplines, such as literary and film narratology.

In terms of theory hierarchy, one can distinguish three levels according to one view. Level I theories are on this view mid-range, basically descriptive, and deal with one or more aspects, such as kinds of narrators, of a specific literary corpus. The claims made in them are empirical in the sense of open to textual observation. Level II theories operate with higher level concepts and theoretical constructs, such as the nature of the narrative function or the demarcation of narrative texts from other text types. While level I theories are specifically literary, Level II ones may have wider application as we have just seen, and in this sense may be termed generic theories. Level III theories are generic semi interpreted ones such as communication theory, semiotics, and general action theory. They provide a framework helping us to think of a whole class of entities in a variety of domains, but have no empirical content and solve no problems, but help us discover and clarify basic ideas. In other words, they supply the literary scholar with general research orientations or points of view (on this particular hierarchy see Soerensen 1987, 146–154). But level hierarchy can be drawn in various ways. Siegfried Schmidt, for example, has proposed the following four term hierarchy: theory of action; theory of communicative action; theory of aesthetic communicative action; theory of verbal aesthetic communicative action.

As a matter of principle, specifically literary theories are non fundamental and essentially dependent on higher level ones to provide their foundation and placement in the wider field of humanistic enquiry. For example, most problem clusters occurring in LTs are specific cases or variants of problems formulated in general terms in higher level theories, such as those of text theory or communication theory. And the same goes for strategies, conceptual and empirical, employed by literary studies to solve these problems. Since most of such higher level theories are formulated in fields other than literary studies, literary theorizing is essentially dependent on more powerful or deeper theories formulated in other disciplines. On the other hand, LTs seem to have domain specific claims that cannot be derived from claims occurring in higher level theories, even if all the terms and concepts of the LT can be translated into or subsumed under those of the higher level theory. In this sense, LTs are not reducible to higher level theories in other disciplines, be they linguistics, cognitive science or sociology.

How Should Literary Theorizing Proceed?

As we know from Kant’s critiques, one first asks »what can I know«, and, in light of the answer given, one proceeds to ponder »what should I do?« So here are some suggestions:

  • Literary theories as text theories have inherent limitations as regards causal explanation, prediction and experimentation. Whoever insists on having these theory components needs to move to the literature as social action paradigm.

  • Any quest for a grand unifying theory in literary studies is probably unwarranted. A series of well formed partial theories is all we can expect. But this is true of most of the social and even natural sciences, so it need not be viewed as a deficiency. By the same token, any attempt to create a super theory by conjoining incommensurable theory paradigms is logically suspect.

  • All major problem systems and areas of enquiry in literary studies are multidisciplinary (treated by LT plus at least one other discipline). Moreover, the most basic models and claims about them are also formulated in other discipline(s). A good literary theoretician consequently needs to constantly follow work relevant to his area of enquiry in at least one other discipline, and different kinds of problems in LT will require awareness of different discipline( s). Conversely, it is counterproductive for literary scholars to invent ab initio theories of language, cognition, society or culture where tremendous amounts of valuable work on these subjects are already available in other disciplines. In most cases, the best result would be a reinvention of the wheel. Only in rare cases, and after having acquired a thorough familiarity with available work, will it be possible for a literary theorist to formulate such a higher level generic theory.

  • In view of the above, a literary theorist must ask himself: do I want to move to a higher level of theorizing, which may eventually enable me to construct better, and more vertically integrated LTs? Van Dijk moved from literary to general text theory, Schmidt to media studies, and the Moscow Tartu School to general theories of cultural systems.

  • On the other hand, we have seen that many of the concepts and models developed within LT apply to other domains as well. So another important question for the literary theorist is: do I want to extend my work to distinctly non-literary, sometimes even non verbal corpora as long as my theoretical apparatus applies to them? (horizontal extension). Mieke Bal, for example, has moved to pictorial narratology, and Chatman to cinema narratology.

  • Multiple levels of units and of organization are necessary for a powerful LT. But one cannot skip levels, proceeding directly from evolutionary biology to the portrayal of human consciousness in a novel, for instance. What one ends up with are pure analogies with no mediating mechanisms. The same is true of premature level reduction, declaring narrative to be »nothing other than« a cognitive mechanism, for example.

  • Current LT, especially in the United States, is characterized by the overabundance of general speculation (»Theory« with a capital T) and dearth of midrange theories or systems of descriptive generalizations. The neglect of descriptive poetics leads to grand theories with a poor knowledge base and, lacking the intermediate level, unable to contribute even indirectly to the conceptualization of data (McHale). The balance obviously needs to be redressed, especially since literature itself is after all a concrete historical phenomenon.

  • A wealth of still valuable low and mid-level theories (descriptive poetics of genres, styles, movements etc.) was produced in the first half of the twentieth century, especially in Germany. This work has been neglected since 1968 because of its low level of theoretisation, yet its retrieval is essential if we want theories that are not only formally strong, but which also possess empirical content and are able to account for more than just some 19th and 20th century works or genres.

In this essay I have cited briefly and for illustrative purposes various models, procedures and methodological norms, and kinds of claims occurring in LTs. This was an essential part of my claim that much of what is referred to as literary theorizing is a rational, inter-subjective and repeatable activity. But the truth of the matter is that, to my knowledge, nobody in recent years has taken systematic, not to say exhaustive, stock of the range of these three actually occurring in literary theorizing and its various branches, from textology to the study of literature in culture. One crucial part of theoretical self-consciousness would certainly consist of doing this. In view of the amount of work involved, and the multilingual nature of the project, one would hope that somewhere a group of qualified and dedicated researches, similar to the Hamburg Narratology group, will arise and take up the challenge to the benefit of all literary theorists.

Uri Margolin

Department of Modern Languages and Cultural Studies

University of Alberta


Mario Bunge, Philosophy of Science, New Brunswick 1998.

Itamar Even-Zohar, The Quest for Laws and its Implications for the Future of the Science of Literature, in: Gyorgy M. Vajda/Janos Riesz (eds.), The Future of Literary Studies, Frankfurt a.M. 1986, 75–79.

Douwe Fokkema/Elrud Ibsch, Knowledge and Commitment, Amsterdam 2000.

Helmut Hauptmeier/Siegfried J. Schmidt, Einführung in die empirische Literaturwissenschaft,

Braunschweig 1985.

Roman Ingarden, Gegenstand und Aufgaben der Literaturwissenschaft, Tübingen 1976.

Brian McHale, Whatever Happened to Descriptive Poetics?, in: Mieke Bal/Inge Boer

(eds.), The Point of Theory, Amsterdam 1994, 56–65.

Uri Margolin, The (In)dependence of Poetics Today, PTL 4 (1979), 545–586.

–, Of What is Past, Passing or to Come: Temporality, Aspectuality, Modality and the Nature of Narrative, in: David Herman (ed.), Narratologies, Columbus 1999, 142–166.

Igor Melcuk, Three Main Features, Seven Basic Principles, and Eleven most Important Results of Roman Jakobson’s Morphological Research, in: Roman Jakobson, Verbal Art, Verbal Sign, Verbal Time, Minneapolis 1985, 178–200.

I wish to express my gratitude to Professor Willie van Peer (München) for a detailed discussion of this paper, which provided me with several valuable insights and saved me from some major errors.

In: JLT 1/1 (2007), 196-207.


JLTonline ISSN 1862-8990

This work may be copied for non-profit educational use if proper credit is given to the author and JLTonline.

For other permission, please contact JLTonline.

How to cite this item:

Uri Margolin, Response.

In: JLTonline (20.03.2009)

URL: http://www.jltonline.de/index.php/articles/article/view/58/244

A Persistent Identifier can be found in the PDF-Version of this article.