Department of English, University of Alberta
Paper published in Frame. Utrecht Journal of Literary Theory 14.2-3 (2000), 43-59.
© Frame
Abstract
Current theoretical divisions in literary studies suggest the need to establish the empirical foundations of the discipline. While it has not yet achieved paradigmatic status, empirical study of reading has shown its value, bringing new insights to fundamental questions about the canon, stylistics, and narrative response.
[43] {page numbers are inserted at the top of each page}
1. Why empirical studies?
Empirical studies of literary reading have not yet appeared over the horizon of most mainstream literary scholars. Involved as most now are in theoretically self-conscious practices of historicist or deconstructive research, the very assumptions of empirical research would, if they came to attention, seem irrelevant, or even improper. This point has been made several times, for instance, by Jonathan Culler. When enquiring about literary competence, he argues, "The question is not what actual readers happen to do but what an ideal reader must know implicitly in order to read and interpret works in ways which we consider acceptable, in accordance with the institution of literature" (Structuralist Poetics 123-4; cf. Signs 129). Similarly, Culler refers to "the dangers of an experimental or socio-psychological approach which would take too seriously the actual and doubtless idiosyncratic performance of individual readers" (Structuralist Poetics 258).
Such statements reveal two common assumptions: first, that the nature of literary reading is necessarily decided by the theorist, who determines in advance what is to count in "the institution of literature"; second, that actual readers are too wayward in their readings to justify serious attention. Both these assumptions deprecate the value of an empirical approach, yet both presuppose the answers to questions that call for empirical study. Are readers idiosyncratic? How are we to decide what qualifies a reading to be "literary"? At the moment we have almost no attempts to examine these questions. But I will suggest that [44] empirical studies have the capacity to take the primary place in defining literary studies, and that this is an approach that would help to clarify the aims and unify the divided nature of current scholarship.
Thus at the present time empirical studies might be seen as the Cinderella in the family of literary disciplines. Mostly disregarded, and deprecated when noticed at all, as the example of Culler shows, no prince will arrive to exalt her at the end of the day. But by her own efforts empirical study will come to dominate the literary field by providing a matrix for evaluating theoretical proposals and for rethinking the nature of literary reading and its cultural place. It is destined to play this role in literary studies, just as over the last two centuries the sciences have emancipated themselves from theological control or superstition by subjecting themselves to validation by empirical methods. As astrology was replaced by astronomy, or alchemy by chemistry, or as evolutionary theory has replaced creationism, despite last ditch defences in several jurisdictions in the United States, our understanding of literary reading will be recast in the light of evidence gathered from real readers. Rather than debate the (often conflicting) assertions of reader response theorists such as Wolfgang Iser or Stanley Fish, literary scholars will formulate their claims as hypotheses and set out to design empirical studies to assess their validity. As Colin Martindale, one critic of the conflicted state of literary studies has argued, "Literary theorists, for whatever reason, often take as axiomatic what are really empirical questions" (Martindale 349). While the place of theory will remain central, it will become possible to arbitrate between alternative positions; theory will no longer remain the interminable and inconclusive mode of debate that we currently witness; nor will it remain divorced from the interests and concerns of those outside the academy who continue to be engaged in reading literary texts. Such readers are currently excluded by the irresolvable conflicts that rack the discipline, and the excessive claims being made by its most prominent exponents. As De Beaugrande has put it, "Only empirical studies can resolve this state of affairs by freeing these [theoretical] claims from their absolute dependence on the personal eloquence or effrontery of the individual theorists and by providing progressively more reliable and intersubjective grounds for preferring any set of claims over any other" (1989: 10).
One major shift in perspective will be required of literary scholars in order for this development to occur. The last thirty years has been [45] dominated by what, following Ricoeur, has come to be know as the hermeneutics of suspicion (Ricoeur 32). As Ricoeur's way of characterizing Freud (his other examples were Marx and Nietzsche), it was intended to suggest that the evidence of consciousness, whether the recollection of a dream or a slip of the tongue, was not to be taken at face value; the mode of suspicion is a way of understanding the aims of Freud's forensic approach. As an approach to literary texts it has been generalized to become the predominant paradigm: all texts are considered to conceal latent content at odds with their surface claims. Whether this is approached in the light of Derrida's différence, Foucault's epistčme, or Barthes's intertextuality, a text is not what it seems; moreover, it has designs on its readers. A particularly clear example is provided by the latest issue of an online journal for Romantic studies, whose editor refers to the influence of the aesthetic of Wordsworth or Shelley on the critics of thirty years ago. There is, he says, "a recognition shared more broadly by many in the profession who can now only regard with irony the claim (did anyone really believe this?) that poetry 'can set one free of the ruins of history and culture', to borrow McGann's famous phrase." Advocating a renewal of historicist methods, he suggests that, on the contrary, "history will save us from poetry" (Kooy).
Suspicion is thus directed at the text, at whatever effects, feelings, or insights the text may be designed to elicit. One critic has renamed this approach, with some justice, "the hermeneutics of disparagement" (Alan Grob). As a departure point for literary studies, it precludes attention to readers' modes of engagement. Whether or not a given aesthetic response should be called into question -- and this itself is an empirical question -- the reader's experience of that response is precluded as a focus for study. In fact, it is more likely, as I will argue in detail below, that a literary text stands in relation to its reader as Freud to his client, and that to read appropriately is to have one's own assumptions and way of life called into question. The reader, in other words, is invited by a literary text to place her experiences under the aegis of suspicion, to reconsider or modify her attitudes, feelings, and conceptions. This does not eliminate literary studies under the sign of Derrida or Foucault, but it displaces them from their current central place and turns each critical thesis generated by their approach into an empirical question: i.e., what, given sufficient evidence, can we trace of this supposed property of the text in the responses of actual readers. To [46] continue otherwise, allowing the present hermeneutics of suspicion to assert its domination of literary studies, is to leave out of account the most important question of all: why do people read? It is like studying food in terms of its appearance, customs, and history, while failing to pay attention to its nutritional function.
In this essay I will offer three more specific reasons for this claim (each, by the way, being empirically testable), drawing on the work of our research group and other published research studies. I will discuss the canon, stylistics, and empathic identification in narrative. Empirical studies have already reinvigorated some areas of research that had become largely disregarded, such as stylistics, or emotional aspects of reading; it has also broached some new ways of thinking about literary reading, from its inroads on discourse processing (cf. Zwaan, et al.) to the electrophysiological studies of Hoorn. It has led to some rethinking over what literature is, and how we might delimit it. In these and other ways empirical study represents a significant alternative framework for understanding literature, but it is one that in itself has yet to achieve paradigmatic status. There is no consensus among its practitioners what its theoretical framework should be, or what would count as exemplary experimental methods. I will next indicate briefly what some of the main causes of dissension are that must be resolved before empirical study becomes coherent and programmatic.
Overall, not surprisingly, some of the same theoretical problems that have focused mainstream debate have preoccupied a number of empirical scholars. Several typical postmodern claims appear rather often in the empirical literature. For example, Bürger has argued that after Duchamp (with works such his urinal) we cannot argue that there is an essential nature to works of art: "What seemed to be an incarnate 'nature' turns out to be a pure construction." We now know that "the institution of art is, in the truest sense of the word, groundless" (Bürger 52). The canon of esteemed literary works is decided institutionally, says Shavit: the dominant institution gains the mandate, which "has nothing what-so-ever to do with 'poetic justice' nor with the question of the value of the texts. . . . A text gains a high status not because it is valuable, but because . . . someone has the political-cultural power to grant the text the status they believe it deserves." (Shavit 233). Similarly, literature has no defining elements distinguishing it from what is non-literary. While conceptions of literature play a role in enabling readers to perceive and name the textual [47] elements of their reading, according to Verdaasdonk, "The normative and imprecise nature of conceptions of literature rules out, however, that the labels readers attach to textual elements have descriptive value" (Verdaasdonk 89). The programme of research initiated by Schmidt, the so-called Empirical Science of Literature, is based on two foundational conventions, the aesthetic and polyvalence conventions which, once again, are said to be institutionally determined. Other parallel examples could be indicated. Each such position must be confronted by the question that Elrud Ibsch puts to Schmidt: "Does this construction permit empirical testing or is it used as a hard core which does not need corroboration?" (Ibsch 399). In this mode of argument, she points out, such theorists seem unreceptive to Popperian scientific procedure, since they have no place for "the ethical impulse inherent in the search for the counter-example." Assuming validity, rather than demonstrating it, a theory of this status calls for "immunization procedures instead of a strengthening of methods for testing" (Ibsch 402).
The testing of such claims must be at the heart of the empirical programme. So far, however, we have no unequivocally accepted paradigm of empirical method, no foundational studies. The problems facing the empirical scholar have been well stated by Graesser and his colleagues. First, there is no agreed essential defining property of the aesthetic experience: this could reside in good form, prototypical aspects, novelty, formal devices, or somewhere else; second, the components of response are relatively inaccessible to consciousness, thus hard to track empirically; and third, we lack agreement on a set of theories or mini-theories that would direct our research (Graesser, et al.). Thus, in the three domains I discuss below, I try to show how the problems of the field identified by Graesser and his colleagues might be addressed. Taking issues that have commonly been debated in mainstream scholarship, I show how each has been amenable to empirical study.
2. The renewal of the canon
A qualitative basis to the canon has been called into question by most recent literary theorists, who now take the view that it is a sociological construct. But I will suggest that this is properly an empirical question. The institutional perpetuation of the canon is one possible explanation [48] for the survival of eminent works such as those by Homer, Shakespeare, Wordsworth, or Dickens; but other explanations should also be considered. In particular, the canon appears to be subject to renewal: the canon renews itself through being re-experienced and reinterpreted (the main feature I examine below), and the canon is itself renewed by the appearance of new works that gain canonical status.
In the first case, the literary academy itself gives evidence of the power of the canon: it does so by its continual provision of new interpretations of the notable canonical works. This has been seen as unfortunate by some critics: "Each new generation feels the urge to produce new interpretations, new evaluations, new concepts of literature and new literary histories," remarks Fokkema: "is it necessary to start time and again from scratch?" (Fokkema 532). Moreover, new interpretations generally compete with extant rival views, suggesting widespread disagreement over what a given literary text "means." This phenomenon argues against institutional determination of the literary canon (since no institution worthy of the name would tolerate such an array of incompatible readings), but the renewal of interpretation itself points to the need for continual repositioning of a given text in relation to contemporary historical and social conditions; our Hamlet is not that of the 1960s, and this is in turn differs from that of the nineteenth century. Reinterpretation is thus a sign of the inexhaustible vigour of the canonical texts at issue, not of our weakness as critics. It is, on a large scale, a parallel to the dehabituation requirement that I discuss below in the context of literary language.
The canon also seems to be renewed in another way that falls outside institutional control. While standard histories of reading argue for the emergence of the literary canon in the eighteenth century as a vehicle for the rising bourgeoisie to identify itself and its interests (e.g., Terry), there is a small but growing volume of evidence for the impact of canonical works on readers outside this class. Jonathan Rose, summarizing his research on British working class readers of the nineteenth century, reports that such workers frequently discovered classical texts for themselves, picking them up from trash heaps or buying them from penny bookstalls, but were then influenced by them to imagine different or new worlds. Moreover, as Rose puts it in the case of Dickens, who appears to have had a particularly pervasive influence, "Dickens provided working people the inspiration and the generic literary conventions they needed to tell their own stories" (Rose 61), an [49] impact that intentionally popular fiction writers of the time failed to achieve. Rose's evidence shows repeatedly that the texts that genuinely influenced their working class readers were those we now regard as canonical. Far from being instruments of social control, as critics such as Eagleton have asserted, Dickens or Hardy were avenues to emancipation for many of their working class readers: they saw new worlds, recognizing their common humanity in the figures of David Copperfield or Tess D'Urberville, and were empowered as a result to change their lives and the lives of those around them (Rose 64). Thus, rather than theorizing about the impact of the canonical text, or supposing it to be an instrument of social control, Rose has collected empirical evidence from the autobiographies or comments of the working poor, evidence that seems to provide strong support for the central values of the canon.
A contemporary example of the power of reading is a programme for sentencing criminals called "Changing Lives through Literature." Now available in a few jurisdictions -- Massachusetts, Texas, and imitated in Manchester, England - the programme is an alternative to jail time. Offenders are required to read novels and short stories, and texts by Plato and Shakespeare, and attend discussions of them at regular intervals. According to one study, reoffending rates of those in the programme were 19% compared with 49% in a matched group of offenders (Fitterman). Again, most of the texts being read are canonical. These are texts with the power to make a difference in their readers' lives.
Jack Gold offers a personal view of such an experience. Reading a novel by David Lodge that describes air raids in London during the war, an experience Gold lived through as a child, he reports "I was strongly moved by it, but more, I was grateful for it. The expression, the novel, sometimes gives a shape, a form, to experience that we recognize as our own. The novel is then a gift, a creating of the reader's reality, existence, history. The pieces of my past, my life, that were lying around in a puzzling mess - unexpressed, unformed, vaguely felt -- are gathered together and given recognizable and storable shape. This is a priceless gift . . ." (Gold 176-7).
In this way, the canon renews itself in the experience of individual readers. While this process may of course be assisted (or endangered: Miall, 1996) by the institutions in which most of us learn to read - the family, school, college - it seems improbable that the canon would be [50] perpetuated if readers had to be persuaded of the value of the literary texts they read, rather than discovering it for themselves at first hand. That experience seems to matter, in the first place, because it often appears to be as vivid as our lived experience. Janos László, for example, made a study of image production in readers. He found that images generated in response to literary texts were as strong as those derived from experience, whereas images generated in response to a newspaper article were generic in nature, similar to images of social categories. In the second place, literary reading can often be a powerful emotional experience. These two features are linked: as Goetz and Sadoski have shown, imagery in response to literary reading is usually associated with emotion, suggesting that imagery may provide a matrix for representing emotions. Through imagery and emotion, in other words, a literary text engages with the reader's own experience and, as Gold suggests, helps the reader to think about it afresh, even to reconfigure it and understand it in a new light. The two principal features specifically responsible for this process appear to be the dehabituating power of literary forms, and empathic projection into the lives of others through narrative.
3. Dehabituation
Everyday experience is governed largely through standard and familiar concepts that provide economical and efficient ways of dealing with the world. Cognitively speaking, this aspect has been enshrined in schema theory (other common terms being scripts, or frames): schemata are those stereotyped processes of behaviour by which we orient ourselves and know what to expect -- as, for example, when we enter a restaurant. However, it is also critically important to be able to question our familiar, everyday behaviours; literary reading provides one vehicle for going beyond the customary, familiar world, and for reconceiving our role within it. Through literary reading we dehabituate, that is, we are enabled to contemplate alternative models for being in the world. Such reading prepares us for being more adaptable: it is an "offline" way of experimenting with emotions or experiences that might have dangerous or unpleasant consequences in the real world, gaining insight into their implications so that we know better how to act when similar situations occur in reality.
The dehabituating aspect of literature has been described in several ways. For the British Romantic writers it was a central part of the theory [51] of imagination: for Coleridge, writing in 1817, the imagination at its most powerful "dissolves, diffuses, and dissipates, in order to re-create" (Coleridge I.304). Among empirical scholars, several terms have been used. John Harker calls attention to the "reattentional" activity required of the literary reader: this is because "The literary text does not simply distort or blur reality; it refocuses it, instructing the reader in the new ways of knowing that to its author are more authentic and real" (Harker 650). Similarly, De Beaugrande (1983) argues that the most important trait of literary communication, "the only one that seems to apply to all cases -- can be termed the ALTERNATIVITY PRINCIPLE: that the world-model evoked by a literary text is free, though not obliged, to present or imply alternatives to the socially established model of reality." (1983: 91).
A predominant way of achieving this is through special uses of language. Vipond and Hunt studied what they termed "evaluations" in narratives, defined as elements that depart from the local norm of the text. These might be discourse evaluations (the way something is told), story evaluations (an event that is surprising), and telling evaluations (that something is mentioned at all or that moment) - such evaluations are likely to be noticed by readers and prompt particular forms of attention. These are all connected with "the idea of incongruity, distinctiveness, or surprisingness" (Vipond and Hunt 157). Other research, such as that of Van Peer and our studies (Miall and Kuiken, 1994), has seen literary language as distinctive in comparison with the uses of language in everyday discourse, as in a news report. In this respect, following Mukarovský, we and Van Peer described such features as foregrounded, since they stand out against a background of common usage. Such features as assonance, metre, syntactic inversion, or metaphor, are effective in attracting attention; they serve to defamiliarize the reader, who (as our studies showed) lingers a little more over such features, and usually appears to attribute greater feeling to them; they may also provide the germination point of alternative interpretations that will emerge later during reading (Miall and Kuiken, in press); or, in De Beaugrande's account, "Deviations act as intermittent cues to apply the alternativity principle to the entire text, including elements that could occur in the same form in ordinary discourse" (1983: 92). In our studies we have found a high degree of consistency between readers in noticing and being influenced by the same set [52] of foregrounded features, even though individual interpretations might vary widely.
The "literariness" of foregrounded language raises the question whether this is a defining characteristic of literature, as an earlier generation of scholars appeared to believe (e.g., Jakobson). The recent consensus among empirical scholars has been against this position. De Beaugrande, for example, argues that literariness can be defined "only as a processor disposition, rather than as a text property"; there is "no one manifest property" than is necessary for a literary text (1983: 91-92). Similarly, Halász rehearses the arguments for defining literariness in terms of the formal qualities in texts, but concludes "There is no literary object, there is only literary function that any sort of written text can have" (1989: 31). This is to claim that literariness is an outcome of a reader's cognitive operations, opening the door to the claim that any text can be considered literary and that any interpretation is valid. But Halász stops short of such total relativism; a literary text is not a projective test, and possible interpretations are constrained by certain characteristics of the text. In this respect, the differences between readers appear to be less salient than the commonalities, and this finding appears to be due principally to the text.
The issue remains unresolved (and is perhaps unresolvable), but the question of the background against which foregrounded features would be perceptible is broached by at least one supporting study. Frey found that at the level of word frequency, people show a high degree of accuracy in judging how often a set of words occur, and that their rank orders agreed closely with word norm data from several sources. Participants also gave different but consistent ranks when invited to consider the frequency of the same words in different situations (e.g., among workmen building a house; on the evening TV news, etc.). This provides evidence for the existence of a "background" against which deviations in a literary text will be perceived (i.e., the use of an unusual word in a given context); and if words are amenable to such judgements, then we might also expect a similar capacity for discrimination of sound patterns, the background against which metre, assonance, and alliteration are perceived. In the absence of further studies, it seems premature to claim that "literariness" is definable in terms of objective features in texts; but neither should this possibility be prematurely dismissed. What is clear from existing studies, is that dehabituation is a prominent, perhaps even the most significant, aspect of readers' responses to [53] literary texts, and that as an agent for initiating this response identifiable formal features of the language of texts often play a key role.
4. Decentering
Another major component of literary reading is our interest in the fate of the characters we encounter in narrative, especially the protagonist in whose predicament we come to empathize. While a novel or short story often draws us to view the world from the protagonist's perspective, that character is not us, and does not share our experience; thus, in addition to the close interest evoked by our reading, we may also experience a decentering, a shift away from attending to our own daily concerns. As Birkerts puts it, during such reading the self is "suspended in the medium of language, the particles of the identity wavering in the magnetic current of another's expression" (Birkerts 78); and, he adds later, "Our awareness, our sense of life, gets filtered into the character, where it becomes strangely detached from us" (93). To emerge from the absorbed state of such reading can seem a distinct change of state, like awakening from a dream; there is a momentary and disorienting sense of engaging the gears of our own daily life once again. Yet, perhaps our interests are reflected at some deeper, transmuted level during reading.
Halász (1996) examined the how far personal meanings were invoked by a literary in comparison with a non-literary text. His studies employed two texts, Kakfa's "The Vulture", and an expository text called "The Peregrine Falcon"; these were presented to readers without information about their authors or that one was literary. During reading participants were asked to pause three times and to comment on the "accepted" (i.e., impersonal) meaning and the personal meaning of the text. In counting the frequency of impersonal and personal meaning units, Halász found that the expository text produced three times as many impersonal to personal meaning units; the literary text produced almost the same number in both categories, showing that the literary text enabled readers to generate a much higher proportion of personal meanings. Among the personal meanings, the predominant types were actions, feelings, evaluations, and "cognitive qualities" (intuitions, imaginative or daydream-type responses).
Similarly, a study by Sielman and Larsen explored the kinds of memories that occurred to readers while reading a literary and a non-literary text. They proposed that when comparing responses to a literary [54] and an expository text, the literary text would involve more memories of the reader as an actor than as an observer. Using two texts, a short story and a text about population growth (each of about 3000 words), they found that although a similar number of remindings was elicited by both texts, twice as many actor-perspective remindings were elicited by the literary text, while the expository text elicited more receiver remindings (memories of things read or heard about). Thus, they suggest, literary reading "seems to connect particularly with knowledge that is personal in the sense that one is an agent, a responsible subject interacting with one's environment" (Sielman and Larsen 174). This seems to evoke the possibility of readers' complicity with the actions of the characters in a narrative, or more radically, as we have been finding in our own recent studies, an emerging convergence between the interests of the reader and that of the main character (cf. Miall & Kuiken, 1999).
Just as literariness at the level of language has been a focus for dispute, however, so too has the nature of fictionality and its effects. According to Siegfried Schmidt fictionality belongs to the level of discourse not text, and is attributed to texts "by judgements of agents according to conventions regulating fictional discourse" (Schmidt 539); "whether a linguistic event is or can be treated as fictional cannot be decided except on the level of discourse and using information provided by the non-verbal social context of discourse" (540). Similarly, Richard Gerrig has rejected the notion that our encounter with fiction causes us to "suspend disbelief"; the cognitive operations involved in reading literary and non-literary texts are the same, in his view, and judgements about fictionality follow later. This is questionable, however, when the devices available to the fiction writer are considered. These are precisely the devices capable of eliciting the decentering response of empathic projection. At the sentence level, for example, it is possible to demonstrate the omniscient (or limited omniscient) narrative style that provides privileged information about a character's mind; since this is unavailable in any non-fictional mode of discourse, it is necessarily fictional. For example, the use of free indirect discourse enables us to be an intimate witness of the thoughts and feelings of a fictional character. As with foregrounding, then, omniscient narratives contain unequivocal marks of literariness at the level of language, although these are not defining features of all literary texts. If this is accepted, it is evidence against Schmidt's claim that "a concept like 'literariness' must primarily be defined pragmatically and historically; [55] only after that the semantic and syntactic features of literary objects are discoverable and describable" (Schmidt 544).
5. Conclusion
Thus we return to the opening question. The concept of literature itself has been called into question, either through a hermeneutics of suspicion that attempts to forestall the effects of literature on readers, or by a dismissal of literariness except as a construct of recent (and unacceptable) ideologies. In contrast, I have tried to show that on three grounds literature appears to have innate powers: it renews itself through canonical texts whose effects are discoverable regardless of class or education, and it exerts local effects at the level of foregrounding and narrative through features that are probably objectively demonstrable and available to all readers. These claims, however, are not the outcome of purely speculative consideration, as has been the case with the positions they oppose. They are based on empirical evidence gathered from readers, whether Rose's historical witnesses or the participants in experimental studies of reading by scholars such as De Beaugrande, Halász, or Van Peer. This work is too little known, and so far lacks paradigmatic status, yet as I have tried to show it has the potential to reorganize our understanding of literary studies root and branch. Hence, I argue, empirical studies of literature are necessary. They are, perhaps, even inevitable, although only time will show whether in the present state of our discipline this prophecy is capable of being fulfilled.
References
Birkerts, Sven, The Gutenberg elegies: The fate of reading in an electronic age, New York: Fawcett Columbine, 1994.
Bürger, Peter, 'Interpretation after Duchamp,' in: Comprehension of literary discourse: Results and problems of interdisciplinary approaches, D. Meutsch, & R. Viehoff (eds.), Berlin & New York: Walter de Gruyter, 1989.
Coleridge, Samuel Taylor, Biographia Literaria, J. Engell & W. J. Bate (eds.), London: Routledge & Kegan Paul, 1983. (Original work published 1817.)
Culler, Jonathan, Structuralist poetics: Structuralism, linguistics, and the study of literature, London: Routledge & Kegan Paul, 1975.
Culler, Jonathan, The pursuit of signs: Semiotics, literature, deconstruction, London: Routledge & Kegan Paul, 1981.
De Beaugrande, Robert, 'Surprised by syncretism: Cognition and literary criticism exemplified by E. D. Hirsch, Stanley Fish, and J. Hillis Miller,' Poetics 12 (1983), 83-137.
De Beaugrande, Robert, 'Toward the empirical study of literature: A synoptic sketch of a new "society",' Poetics 18 (1989), 7-27.
Eagleton, Terry, Literary theory: An introduction, Oxford: Basil Blackwell, 1983.
Fitterman, Lisa, 'Book `em!' National Post, Sept 25 1999, p. B11.
Fokkema, Douwe Wessel, 'On the reliability of literary studies,' Poetics Today 9 (1988), 529-543.
Frey, Eberhard, 'Subjective word frequency estimates and their stylistic relevance in literature,' Poetics 10 (1981), 395-407.
Gerrig, Richard J., Experiencing narrative worlds, New Haven: Yale University Press, 1993.
Goetz, Ernest T., & Sadoski, Mark, 'Imaginative processes in literary comprehension: Bringing the text to life,' in: Empirical approaches to literature and aesthetics, R. J. Kreuz & M. S. MacNealy (eds.), Norwood, NJ: Ablex, 1996.
Gold, Jack, Read for your life: Literature as a life support system, Markham, ON: Fitzhenry and Whiteside, 1990.
Graesser, Arthur C., Person, Natalkie, & Johnston, G. Scott, 'Three obstacles in empirical research on aesthetic and literary comprehension,' in: Empirical approaches to literature and aesthetics, R. J. Kreuz & M. S. MacNealy (eds.), Norwood, NJ: Ablex, 1996. Norwood, NJ: Ablex, 1996.
Grob, Alan, 'William and Dorothy: A Case Study in the Hermeneutics of Disparagement,' ELH, in press.
Halász, László, 'Social psychology, social cognition, and the empirical study of literature,' Poetics 18 (1989), 29-44.
Halász, László, 'General and personal meaning in literary reading,' in: Empirical approaches to literature and aesthetics, R. J. Kreuz & M. S. MacNealy (eds.), Norwood, NJ: Ablex, 1996.
Harker, W. John, 'Toward a defensible psychology of literary interpretation, in: Empirical approaches to literature and aesthetics, R. J. Kreuz & M. S. MacNealy (eds.), Norwood, NJ: Ablex, 1996.
Hoorn, Johan, 'Psychophysiology and literary processing: ERPs to semantic and phonological deviations in reading small verses,' in: Empirical approaches to literature and aesthetics, R. J. Kreuz & M. S. MacNealy (eds.), Norwood, NJ: Ablex, 1996.
Ibsch, Elrud, '"Facts" in the empirical study of literature: The United States and Germany -- A Comparison,' Poetics 18 (1989), 389-404.
Kooy, Michael John, 'Special issue: After romantic ideology,' Romanticism on the Net, 17 (February 2000) http://www-sul.stanford.edu/mirrors/romnet/ (visited April 2000)
László, Janos, 'Images of social categories vs. images of literary and non-literary objects,' Poetics, 19, (1990), 277-291.
Martindale, Colin, 'Empirical questions deserve empirical answers,' Philosophy and Literature, 20 (1996), 347-361.
Miall, David S., 'Empowering the reader: Literary response and classroom learning,' in: Empirical approaches to literature and aesthetics, R. J. Kreuz & M. S. MacNealy (eds.), Norwood, NJ: Ablex, 1996.
Miall, David S., & Kuiken, Don, 'Foregrounding, defamiliarization, and affect: Response to literary stories,' Poetics 22 (1994), 389-407.
Miall, David S., & Kuiken, Don, 'What is literariness? Three components of literary reading,' Discourse Processes 28 (1999), 121-138.
Miall, David S., & Kuiken, Don, 'Shifting perspectives: Readers' feelings and literary response,' in New perspectives on narrative perspective, W. Van Peer & S. Chatman (eds.), New York: SUNY Press, in press.
Ricoeur, Paul, Freud and philosophy: An essay on interpretation, D. Savage (Trans.), New Haven: Yale University Press.(1970).
Rose, Jonathan, 'Rereading the English Common Reader: A preface to a history of audiences,' Journal of the History of Ideas 53 (1992), 47-70.
Schmidt, Siegfried J., 'Fictionality in literary and non-literary discourse,' Poetics 9 (1980), 525-546.
Seilman, Uffe, & Larsen, Steen F., 'Personal resonance to literature: A study of remindings while reading,' Poetics 18 (1989), 165-177.
Shavit, Z., 'Canonicity and literary institutions,' in: Empirical Studies of Literature: Proceedings of the Second IGEL-Conference, Amsterdam 1989, E. Ibsch, D. Schram, & G. Steen, G. (eds.), Amsterdam & Atlanta, GA: Rodopi, 1991.
Terry, Richard, 'Literature, aesthetics, and canonicity in the eighteenth century,' Eighteenth-Century Life 21 (1997), 80-101.
Van Peer, Willy, Stylistics and psychology: Investigations of foregrounding, London: Croom Helm, 1986.
Verdaasdonk, Hugo, 'Conceptions of literature as frames?' Poetics 11 (1982), 87-104.
Vipond, Doug, & Hunt, Russell A., 'Literary processing and response as transaction: Evidence for the contribution of readers, texts, and situations,' in: Comprehension of literary discourse: Results and problems of interdisciplinary approaches, D. Meutsch, D., & R. Viehoff (eds.), Berlin & New York: Walter de Gruyter, 1989.
Zwaan, Rolf A., Magliano, Joseph P., & Graesser, Arthur C., 'Dimensions of situation model construction in narrative comprehension,' Journal of Experimental Psychology: Learning, Memory, and Cognition 21 (1995), 386-397.
Document prepared July 21st 2001