Scientific Study of Literature - Volume 1, Issue 1, 2011

What is meant by a scientific approach to literature? I suggest that this question raises several issues: the need to elucidate the object of study, that is, to examine readings that reflect the literariness of the text; also the question how literariness is to be identified, what may be distinctive about it. Then I suggest the importance of defining theories that draw from the literary domain, not outside it (such as psychology or linguistics), and that these should lead us to a set of hypotheses that can be subjected to empirical testing with readers. We should also have in the mind the possibility of learning what purposes literature fulfils for the individual or the wider community.

The present paper introduces corpus-based analytical techniques and surveys some of the specific ways in which corpus analysis has been applied to the study of literature. In recent years, those research efforts have been mostly carried out under the umbrella of ‘corpus stylistics’. Most of these studies focus on the distribution of words (analyzing keywords, extended lexical phrases, or collocations) to identify textual features that are especially characteristic of an author or particular text. Corpus-based grammatical and pragmatic analyses of literary language are also briefly considered. Then, in the concluding part of the paper, I briefly survey earlier computational and statistical research on authorship attribution and literary style. While that research tradition is in some ways the precursor to more recent work in corpus stylistics, it is also complementary to recent research in its application of sophisticated statistical and computational methods.

Everyone agrees that a computer could never understand and appreciate literature, but the fields of computational linguistics and discourse processing have made important advances in automatic detection of language and discourse characteristics. We have analyzed literary texts and political speeches with two computer tools, namely Coh-Metrix and Linguistic Inquiry Word Count (LIWC). Coh-Metrix provides hundreds of measures that funnel into 5 principal components: word concreteness, syntactic simplicity, referential cohesion, deep cohesion, and narrativity. LIWC classifies words on 80 categories, such as first person pronouns, negative emotions, and social words. This paper illustrates how computer tools can unveil new insights about literature and can empirically test claims by literary scholars and social scientists. Our approach offers a computational science of literature.

Through computerized text analysis, the psychology of literature is on the threshold of becoming a dominant force in psychology and the social sciences. The ways people use words in their writing and in everyday life reflect people’s social and psychological states. Whereas most text analysis research has focused on the content of people’s writings, the current paper demonstrates that almost-invisible function words can be psychologically relevant as well. Through the analysis of pronouns, prepositions, and other function words used in literature, several studies demonstrate how authors’ emotional states, aging processes, theories of mind, and the nature of their romantic and collaborative relationships are revealed through their words. The function word approach provides a glimpse of the rapidly expanding methods available to psychologists interested in tracking the social and psychological worlds of authors. With the upcoming release of data sets such as Google Books, the analysis of literature will likely serve as a foundational method used in the fields of psychology, linguistics, history, and other areas of the behavioral and social sciences.

This reflective essay argues that a major constraint on the effective teaching of literature in elementary and high schools is the challenge of articulating and empirically validating a model of literary reasoning that encompasses the following: (1) the multiple dimensions of the literary experience and how they interact (cognitive, emotional, dispositional, personal introspective, aesthetic, experiential); (2) developmental issues and trajectories impacting the growth of expertise in literary response; (3) the multiple sources of knowledge on which readers draw in responding to literature (knowledge of text structures, of rhetorical conventions, of literary traditions, of real world correlates to character types and motivations, settings, and events, and of moral and philosophical domains, including cultural variation among these). Articulating and validating such a model will require interdisciplinary collaborations across fields ranging from cognition, human development, linguistics, philosophy, and literary theory.

In the present editorial, we briefly describe some aspects of the domain of the scientific study of literature, the methods that have been used, and the nature of the theories that have been developed. We discuss some of the prior work that has been done on cognitive processing of and affective reactions to literary texts and how this interacts with the nature of the reader. We note that there is a need for further work on how the literary reactions vary with the reading context. We also describe some of the methods that have commonly been used, such as reading time, questionnaire responses, and protocol analysis. The potential for applying methods from cognitive neuroscience, such as the measurement of event-related potentials and brain imaging, is an exciting opportunity in the future. Finally, we identify some of the types of explanations that have been developed in the scientific study of literature, including variable relations and processing accounts. Other kinds of theoretical approaches, such as those based on complexity theory, might be needed in the future. Our conclusion is that although a great amount of further work needs to be done in understanding literature, there are a wide range of exciting possibilities.

We believe that the scientific study of narrative comprehension will move from using short, laboratory-contrived “textoids” to longer, naturalistic narratives. This move is being driven by technological developments and by theoretical interest in event comprehension mechanisms in reading. One mechanism in which we have been particularly interested is event segmentation, which is the spontaneous organization of incoming information into meaningful discrete events. Behavioral and neurophysiological studies suggest that similar principles govern event segmentation in reading and in the perception of movies and live action. Some of these studies contribute to a growing body of evidence that readers mentally simulate events as they read, including some of their perceptual and motor properties. Based on these trends we look forward to a scientific study of narrative comprehension that is increasingly integrated with broad general theories of perception and memory.

This aim of this brief paper is to promote the scientific study of creative writing and make it an integral part of the research agenda of the scientific study of literature. This paper focuses on the genre of poetry and is organized around three basic questions: (1) What do we currently know about poetry writing? (2) Why study poetry writing? (3) How can poetry writing be studied? Together the answers to these questions provide a rationale for the proposed research agenda and an overview of ways in which this project can be accomplished. A review of the literature on poetry writing reveals a process which involves extensive revision within a recursive writing process that reaffirms the characterization of poetry as a genre that combines attention to meaning with attention to form.

Readers often have quite different experiences of literary texts. In this article, I suggest that readers’ life experiences affect the knowledge they bring to bear on narrative experiences as well as the responses they encode as those experiences unfold. I use an episode from the novel Transmission to illustrate potential individual differences in a literary context.

The scientific study of literature raises a number of critical issues on the best methods to explore how people create, interpreted and are affected by literature and other media. One approach that is widely employed in psychology is to assess the behaviors of groups of individuals in some task, and from there infer underlying cognitive structures and processes that are the likely causal basis for the observed literary behaviors. Adopting this approach allows scholars to make broad scientific generalizations about the ways that people, most generally, interact with literature. But many scientific methods fail to account for individual differences in literary interactions, and as importantly, the unique individual character of literary experience. I explore this concern in the context of some of my own previous work on conceptual metaphor theory in reading poetry, and advocate a scientific approach to literary experience, based on dynamical, self-organizational theory, that may provide the conceptual tools for proper analysis of the individual in the scientific study of literature.

Looking into our individual differences, either as a group (e.g., women; Israelis who support boycotts of Israel) or as a particular human being is important and interesting. Despite assuming commonalities, the persistent quest for the uniqueness of the individual, however, is instrumental in obscuring the reality that we are all a lot more similar than different. In the same manner, the search for the uniqueness of poetic language may also blur the fact that both poetic and non-poetic linguistic uses follow, in most part, similar cognitive principles, and may have similar aesthetic effects, whether in production or in comprehension. Good science underlines that which we have in common even while looking at our differences; at the end of the day, when our idiosyncrasy is filtered out, our similarities stand out quite clearly. Clearly, studying our uniqueness only as well as studying both our idiosyncratic and shared characteristics are political choices. Even as scientists, we are always faced with a choice.

“Transportation into a narrative world” refers to cognitive, emotional, and imagery engagement in a story (Green & Brock, 2000). Transportation has been studied as a mechanism of narrative persuasion; individuals who are transported into stories are more likely to change their attitudes and beliefs in the direction suggested by the story. The current paper highlights the challenges and benefits from the scientific study of literature, and outlines promising avenues for future research. These directions include a greater understanding of ways to evoke transportation, and a fuller exploration of the outcomes of transportation, including impact on implicit attitudes, the persistence of narrative persuasion, and the effects of multiple narratives. We also highlight the role of individual differences, particularly motivation for mindreading (Carpenter & Green, in press), the extent to which individuals are willing to exert effort to understand others’ perspectives.

A rich base of research has investigated the types of responses that literature engenders in readers. In the current contribution, we consider other aspects of the reading experience that have, to date, gone relatively ignored in empirical study. We explore issues including authors’ decisions about literary constructions, readers’ responses to real-world texts, affective evocations (and their neural underpinnings), and how we know what people have learned from the novels they read. With each issue we recommend potential approaches and methods that might usefully support scientific investigations of readers’ engagements with classic and contemporary literature.

We argue that entertainment experiences as individual responses to literary texts have not been sufficiently studied in the past. Literary scholars have long regarded entertainment as an inappropriate response to literature. Likewise, psychologists and communications scholars have been hesitant to study entertainment as an effect of literary reading, arguably because those responses have been seen as too complex to fit into given explanations of entertainment. But entertainment theory has advanced recently and now tries to include more complex responses to media content. Those responses include affective states which are other than simply pleasure-driven.. The idea of entertainment as a response that can imply enjoyment, appreciation, or both allows and provides an explanation of readers’ responses to literary texts that goes beyond purely hedonistic motivations.

There must be more to reading than just text comprehension — stories are to entertain, Brewer and Lichtenstein (1984) famously proclaimed. Hence, they said, research should focus more on how stories do that. What are the implications of this statement? Could that knowledge be of interest to people outside academia? For example, could it result in guidelines for writing bestsellers and blockbuster scripts? Some would suggest that there is more to literary stories than just entertainment. One of the many things this journal could do for the world is find out what exactly this may be. To do this we should extend our knowledge about what distinguishes the processing of literary stories from that of narratives in other genres and media. Here it will be argued that the emotional and cognitive processes typically found in literary reading may offer profound improvements of our lives and the world, and more so than better results at the box office would.

There has been a growing understanding of how the mind and brain work in readers’ and writers’ engagement with fiction. This is worthwhile because fiction occupies much time in people’s lives and because it enables them to understand others and themselves. At the same time, the future of research in this area will contribute to psychology generally, with insights into the model-making function of mind, relating by means of conversation, empathetic theory-of-mind, imagination, and personal transformation.

Readers of a new journal in the scientific study of literature are undoubtedly aware of the potential benefits of a scientific culture in literary studies. However, they may be less sensitive to potential dangers. In order to enhance these benefits and avoid some of the dangers, this essay takes up the relations of authority and prestige that often accompany and distort the interconnections between humanistic and scientific research. Specifically, it considers how social and institutional conditions may place scientific and humanistic cultures in relations parallel to those between colonizing and colonized cultures. (This refers solely to the cultural relations. Clearly, there is no issue of violence or exploitation.) The parallel extends to forms of cultural response (e.g., “mimeticism”) that potentially distort both the humanist’s understanding of science and the scientist’s understanding of the humanities.

How shall we guide ourselves to ensure that research on creation and reception are ecologically valid, representative of natural processes in everyday life? Rather than romanticizing science, we need to be critically aware of the hidden assumptions which lie behind the empirical narratives that appear in journals. Francis Bacon awakened us to the “false idols” and Goethe encouraged us to place the careful observation of phenomena of the natural world ahead of theories which should be derived from observation and help focus our understanding of them. Following Bacon and Goethe, I advise that researchers should work from actual instances or episodes in which the phenomena of interest are manifested. Our goal should be to ensure that the parallel world of experimentation is not divorced from everyday life. This can be achieved by developing a large sample space of materials (poems, literary texts, films, and so forth) from which at least two must be chosen to represent each kind of stimulus. Further, we need to specify the orienting task set according to which participants are to approach these materials. Finally, I advocate the use of quantitative and qualitative kinds of data in a complementary manner to elucidate the underlying processes. By engaging in an empirical dialogue with ourselves and others, we can advance our understanding and explanation of the phenomena which fascinate us.

The migration of science into literary studies risks importing an over-simplified conception of “the” scientific method. To avoid this pitfall, it will be crucial to nurture a conception of scientific praxis grounded in historically established methods from a range of scientific disciplines, rather than in the physics-centered prescriptions of much 20th Century philosophy of science. Doing so may require reconsideration of the forms of explanation and investigation that are appropriate for semi-autonomous, agentic capacities (e.g., generating a meaningful interpretation). Because such capacities resist externalist forms of explanation and experimentation, it is important to reconsider the first- and third-person methods, the role of experimentation, and the importance of classificatory procedures in their descriptive explication. Moreover, acknowledging the conceptual precision that derives both from operational definition (in one form) and connoisseurship (in another) may support articulation of a multi-faceted conception of science that is apt for studies of literature.