What is literature for? The role of transformative reading

The question of what literature is for—if there is a purpose—is not new. Since the beginning of literary theory as a field of study, the debate has been long and complex and is still ongoing. This article offers a reflection on the concept of purpose in the development of literary theories up to the advent of the cognitive turn in the twenty-first century, when empirical studies of literary reading began to proliferate. The paper argues that discussions on the question of purpose have changed from no purpose to pragmatic and later to more existential purposes. It places transformative reading in the center of this debate and reflects on the results of the series of empirical studies conducted so far. The paper focuses on the implications and uses of transformative reading in social contexts. It concludes by discussing how empirical work in this area suggests new conceptual distinctions that could contribute to theorizing about purpose in literary studies more generally. It also indicates what lies ahead in terms of challenges while pointing at new research directions. Olivia Fialho ABOUT THE AUTHOR Olivia Fialho is a Post-Doctoral Researcher at the Department of Literature, Area Studies and European Languages, University of Oslo, and Assistant Professor of Leadership at Utrecht University. She holds a Ph.D. in Comparative Literature (University of Alberta), and has contributed to empirical studies of literary reading and education (e.g., see her co-edited Scientific Approaches to Literature in Learning Environments, 2016; Learning and Instruction, 2019). Her project “Transformative Reading” investigates the phenomenology, preconditions, and underlying processes through which literary narrative fiction deepens perceptions of self and others. While funded by the Netherlands Organization for Scientific Research (NWO) as part of the project “Uses of Literary Narrative Fiction in Social Contexts” (grant number 360-30240), she developed a theoretical-empirical model of transformative reading. In her current research, funded by the “Literature, Cognition and Emotions” initiative at the University of Oslo, she tests this model and researches how its cognitive underpinnings may contribute to a theory of literary reading. Her research has been implemented in education and business. PUBLIC INTEREST STATEMENT We say that literature changes lives. But is this statement supported by scientific evidence? This paper presents recent work on how readers might gain fresh insights into themselves and others through reading literature, in a process called “transformative reading.” It describes the forms of cognitive and emotional engagements involved: by vividly imagining stories and resonating with characters and situations in novels and stories, readers may reflect on how they would feel or act, what consequences their thoughts and behaviours might have for others, and who they may or may not be in the future. In light of a reflection on the concept of “purpose” in the development of literary theories, this paper holds that the purpose of literature lies in the experience itself—it transforms readers’ personal and social concepts—and in its uses in social contexts. It concludes by discussing how empirical work in this area suggests new conceptual distinctions that could contribute to theorizing about purpose in literary studies more generally. Fialho, Cogent Arts & Humanities (2019), 6: 1692532 https://doi.org/10.1080/23311983.2019.1692532 © 2019 The Author(s). This open access article is distributed under a Creative Commons Attribution (CC-BY) 4.0 license. Received: 01 July 2019 Accepted: 25 October 2019 First Published: XXXXXXXX Corresponding author: Olivia Fialho, Literature, Area Studies and European Languages, University of Oslo, Norway E-mail: o.d.c.fialho@ilos.uio.no Reviewing editor: Anezka Kuzmicova, Stockholms Universitet, Sweden Additional information is available at the end of the article


PUBLIC INTEREST STATEMENT
We say that literature changes lives. But is this statement supported by scientific evidence? This paper presents recent work on how readers might gain fresh insights into themselves and others through reading literature, in a process called "transformative reading." It describes the forms of cognitive and emotional engagements involved: by vividly imagining stories and resonating with characters and situations in novels and stories, readers may reflect on how they would feel or act, what consequences their thoughts and behaviours might have for others, and who they may or may not be in the future. In light of a reflection on the concept of "purpose" in the development of literary theories, this paper holds that the purpose of literature lies in the experience itself-it transforms readers' personal and social concepts-and in its uses in social contexts. It concludes by discussing how empirical work in this area suggests new conceptual distinctions that could contribute to theorizing about purpose in literary studies more generally.

Introduction
"Purpose" has different definitions in literary theory and criticism and it is a highly contested concept. For the Transformative Reading Program (henceforth, the TR Program), the purpose of literature lies in the experience itself; and this experience is transformative. According to TR, literary reading always implies both a text and a reader in a reciprocal experience at a particular time and place. In such a fluid exchange, both text and reader are mutually modified. Thus, from this perspective, although the purpose of literature is only one-to be transformative-it might have different expressions, or different forms.
The transformative purpose of literature lies furthermore in the unique meaning reading gives to readers' lives. But what kind of meaning? Some repeated themes that emerge from self-reports of readers (e.g., Coady & Johannessen., 2006;Osen, 2002;Rosenthal, 1995) indicate that such meaning (a) is associated with flow, as described by Csikszentmihalyi (1990); (b) implicates emotional response-especially merging childhood memories with strong sensory dimensions; and (c) involves pleasure in language play (Ross, McKechnie, & Rothbauer, 2006: 148-9). One of the participants in Toyne and Usherwood (2001) This statement reveals a typical mode of engagement, that is, an enactive form of identification. While commenting on David Lodge's novel Out of the Shelter, Joseph Gold (1990) eloquently describes a comparable experience: I lived through World War II in London, England … Mostly I have kept the experience to myself … I think that I buried a lot of it inside me somewhere … Lodge's novel … was just such an experience for me … 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-were gathered together and given recognizable and storable shape. This is a priceless gift-a gift to the reader of part of the reader's life. Now I can say, if you want to know some of how it felt to be me as a twelve-year-old in England in 1944 and1945, read Out of the Shelter. (quoted by Kuiken, Miall, & Sikora, 2004, p. 178) For Gold, Lodge's novel represents a means whereby his sense of self is modified. 1 This same kind of experience is described in the words of ordinary readers. In accounts of the relevance of the experience, it is frequent to find the awakening metaphor, such as in: Books help me clarify my feelings; change my way of thinking about things; help me think through problems in my own life; help me make a decision; and give me the strength and courage to make major changes in my own life; they … give me hope to rebuild my life. (Ross et al., 2006, p. 163) This statement explains in a way how reading can be woven into the texture of readers' lives. The experience seems to have transformative powers as it deepens our understanding of the position of the self in the world. Granted that the experiences of reading vary widely, these accounts illustrate how individuals often refer to the impact of reading on their self-awareness when they explain why they engage in fiction (Fialho, 2002;Miall & Kuiken, 2002;Radway, 1984;Toyne & Usherwood, 2001). Thus, the meaning literature makes in readers' lives is personal while being unique to each individual.
In the midst of studies that indicate the decline of Literary Studies and the Humanities in general (for a review, see Fialho, 2012;Fialho, Zyngier, & Miall, 2011;Schrijvers, 2019), and the number of literary readers (Leesmonitor.nl), there is no reason to be pessimistic and assume that the relevance of literature is waning. It is true that audiences may go for the many self-help books available today or popular magazines (Bergsma, 2008;Cuijpers, 1997). However, they also show a preference for literary texts (Koopman, 2016a), movies (Oliver & Bartsch, 2011), as well as television (Krijnen, 2007). Besides the pursuit of pleasure and amusement (hedonic motives), media audiences also search for texts that reflect on life's meaning, truths and purposes (eudaimonic motives; Oliver & Raney, 2001). In training students to see the meaning for themselves and how to teach others to uncover the importance of individual literary texts, the TR Program contributes to the relevance of Humanities in general.
The problem is that when it comes to an education in literature, such a purpose is not always expressed or have a place. Literature students may acquire the skills necessary to analyze narrative techniques, and to distinguish modernist from postmodernist novels, but they are not asked to read for personal relevance. Traditional literary studies have centered around questions such as "What is this text about?". In fact, why one would study literature in the first place is an issue that is notoriously ignored in curricula. Questions such as "What is this text about for you?", "What does it mean for your life?", "What can I learn from this story", and "How could this novel change the way I live?" are seldom considered (see Fialho et al., 2011;Fialho, 2012; for exceptions, see Schrijvers, 2019;Schrijvers, Janssen, Fialho, & Rijlaarsdam, 2016, 2019aSchrijvers, 2019;Schrijvers, Janssen, Fialho, De Maeyer, & Rijlaarsdam, 2019c). Outside academia, the relevance of literature for people's lives may be more common, such as programs like Changing Lives Through Literature 2 or Literature for Life 3 show (see also Skjerdingstad and Tangerås in this special issue). In bookclubs, or any other shared reading groups, literature is seen to bring an increased sense of life purpose (e.g. Longden et al., 2015;Trounstine & Waxler, 2005), but there the search for meaning remains largely uninformed by the insights from literary scholarship, nor is it guided by evidencebased principles. These are central questions investigated by the TR Program. As it provides an evidence-based program (see Section 3), TR is a scholarly endeavor that goes beyond everyday reading practices.
In placing TR within the debate of the purpose of literature and literary studies, this article initially asks where literary theory contributes to the question of purpose. Section 1 offers a reflection on the concept of the transformative purpose in the development of literary theories up to the advent of the cognitive turn at the turn of the twenty-first century. Section 2 focuses on the question of how the experiential purpose of literary reading comes about. It discusses the cognitive turn in literary studies, indicating how empirical studies of literature have been testing some of the hypotheses emerging from literary theory. Then, it places TR in the center of this debate and reflects on the results of the series of empirical studies conducted so far. Section 3 concentrates on the implications and uses of TR in social contexts. The paper concludes with a discussion on how empirical work on transformative reading suggests new conceptual distinctions that could contribute to theorising about "purpose" in literary studies more generally. It also indicates what lies ahead in terms of challenges while pointing at new research directions.

Transformative purpose in the development of literary theories
The question of the transformative purpose of the arts and of literature, more specifically, or the idea that the arts (including literature) is for transformation is not new. It has been present since human beings realized that they could influence others through discourse. In the course of the development of literary theories, however, opinions are divided as to (1) whether literature's transformative powers are desirable or not and as to (2) the aspects of life that literature changes.
In Poetics, Aristotle was already aware of the emotional effects drama produced on the audience. Plato also attributed to "good poets" the capacity to affect readers, and regarded emotions as "the element in us that the poets satisfy and delight" (qtd. in Burke, 1995, p. 20). The transformative effects of literature were considered dangerous because of their power to influence readers, which led Plato to ban poets from his Republic. Although the Greeks acknowledged the effects of literature and rhetoric, their focus was mainly on how effects in reception, such as catharsis, depended on the manipulation of verbal art.
After Romanticism, the question of the purpose of literature varied, from the preservation of the separateness of literature (especially in Kantian aesthetics) and its elevation to a cult object (Art for Art's Sake) through a claim for literature as a source of consolation and focus for human feelings, including its moral values. In his preface to The Picture of Dorian Gray, Oscar Wilde concluded (Wilde, 1891(Wilde, -1993 that "All art is quite useless". This notion that literature has no purpose other than being aesthetically pleasing was also shared later by critics and writers such as Virginia Woolf and T.S. Eliot, among others. Despite such possibilities entertained about whether or not literature would have a purpose, Matthew Arnold, in the 19 th century, for example, emphasized its transformative purpose, seeing literature as a substitute for religion and claiming that the purpose of literature was to change us for the better. In the 20 th century, the dominance of Formalist approaches to art and literature and New Criticism limited the efforts to understand the transformative aspects of reading, but it was with the opponents of formalist approaches that attention to its social function and purposes regained the weight. Raymond Williams (1977) criticized the formalists in his introduction to Marxist literary criticism, and Lukács (1948) had already argued that art cannot be understood as an autonomous entity which pleases without concept. Art has to be conceived as an historical fact within a social totality. Indeed, Lukács (1971) also claimed that literature should become a revolutionary practice capable of transforming society.
Later, Eagleton (1983) criticized the formalists' notion of literariness and argued that "special language" can be found not only in literary language but elsewhere. Sharing Trotsky's (1923) and Williams' views, Eagleton claimed that the formalists "pass over the analysis of literary 'content' for the study of literary form"-It was essentially the application of linguistics to the study of literature and concerned with the structures of a language rather than with what one might actually say. Content was merely the "motivation of form, a convenience for a formal exercise". Eagleton believes that it is the context that tells that something is literary and that the language itself has no inherent properties or qualities which might distinguish it from other kinds of discourse. He goes to the extreme of stating that "anything can be literature" if they are valued to be so. Also, prominent poststructuralist critics have argued that there are no special characteristics that distinguish literature from other texts. They seem to imply that any text, whether literary or not, depends on functions common to all texts.
The second point of agreement among opponents of formalism is the argument that it is not possible to have a "scientific" study of literature. Based as it is on a general and not specifically literary theory of signification, deconstruction ended up becoming synonymous with a particular method of textual analysis and philosophical argument involving the close reading of works of literature, philosophy, psychoanalysis, linguistics, and anthropology to reveal logical or rhetorical incompatibilities between the explicit and implicit planes of discourse in a text and to demonstrate by means of a range of critical techniques how these incompatibilities are disguised and assimilated by the text. The title of de Man's second book, Allegories of Reading, is suggestive. What began as a critique of methods and systems of reading can be legitimately accused of having succumbed to the normative methodization it criticized. All in all, it can be argued that in the evolution of literary criticism briefly outlined so far, the issue of purpose has centered around the question "what purpose?" The question "purpose for whom?" was only addressed by reader reception theorists.
Reader reception and response theorists in the sixties and the seventies renewed appreciation for the centrality of the role of the reader, now seen as an active participant in the creation of a text's meaning. In an early proposal, Rosenblatt (1938Rosenblatt ( -1995 already shifted the emphasis of textual analysis away from the New Critical focus on the text alone and viewed the reader and the text as partners in the production of meaning, although a rather solitary voice at the time. Reader reception and response critics then moved away from the text as a sole determiner of meaning to the significance of the reader as an essential participant in the reading process, one who is informed by the experiences he/she brings to the moment of reading: present historical circumstances, world knowledge, gender, race, class, age, education, personal experiences, feelings (1978, p. 127). This consideration of the reader as an active participant in the interaction readertext is what differentiates Rosenblatt and reader-response critics from other critical approaches that also considered aesthetic responses.
Despite sharing an emphasis on the role of the reader, reader reception and response criticism do not provide us with a unified body of theory or methodology, with each critic espousing a different approach to textual analysis and different models of readers, e.g., Eco's "model reader" (1979); Fish's "informed reader" (1970); Iser's "implied reader" (1974); Riffaterre's "super reader" (1959; for an overview, see Zyngier, 1994). These approaches brought readers to the limelight, but their models tend to be those of specialized and not of ordinary readers. It is true that some efforts towards valuing the ordinary reader were being made, for example, by de Beaugrande (1985) and Rabinowitz (1987). However, the mainstream in literary studies continued to be carried out within a purely theoretical framework, with a few exceptions (see Holland, 1975;Holland & Schwartz, 1975). Felski (2008) prolifically demonstrates how recent theorists and common readers are in more agreement about the purpose of literature than one would expect. She offers a well-articulated account of the various modes of textual engagement, including literature as a source of selfknowledge, revealing and concealing much of who we are (pp.77-104). She also argues that aesthetic value is inseparable from its use. Thus, her central argument is that literature can be a rich source of personal meaning. It might give us direction in our struggles to find out what we are in this world for, what our life mission is, how to lead our lives purposefully. Since literature reveals much about ourselves, what we do with it and how we use it also reveals its meaning and purpose. Aligned with such a view, from the TR Program perspective, the transformative purpose of literature is the essence of what literature is, which can be expressed by what readers do with it and by its uses.
The TR program contributes to reader response theories, and takes Felski's proposal a step forward, by putting the actual reader in the limelight. It is innovative in describing and explaining the detailed mechanisms by which literature can be meaningful for readers' lives, which revises the traditional takes on the notion of literariness.

The cognitive turn and the role of transformative reading
The cognitive turn in literary studies developed alongside a proliferation of empirical studies of the actual reader. They show, however, far more divergent reading practices and varied understandings of literature than, for example, Eagleton's account would allow (Miall & Kuiken, 1994, 1999. Based on the assumption that the purpose of literature is to gain a fresh experience of the self in the world, the cognitive turn has moved the debate to the interplay between cognition and emotion, and also addresses literariness, through the study of actual readers and the detailed mechanisms by which literature is self-implicating and self-modifying. In empirical studies, and more specifically, in TR research, literariness is not seen exclusively as a set of formal textual properties (Jakobson, 1960) or as only relying on a set of conventions (e.g., Culler, 1975;Fish, 1980), but as involving readers' refamiliarizing/reconceptualizing strategies (Fialho, 2007). The underlying assumption is that responses to literary texts combine verbal, emotional and cognitive elements that may account for the distinctiveness of the literary experience. Thus, TR reconceptualizes the notion of "literariness" and how the reader processes literary texts, and the results contribute to the debate on the purpose of literature, now seen from the perspective of the actual reader.
Drawing on the dehabituation theory of literature (Miall, 2006), TR is grounded on a notion of literariness that is not solely defined as a characteristic set of text properties or as residing exclusively "in what people do with the writing", as Eagleton suggested, but resides in the interaction between reader and text. Literariness is here seen as the product of a distinctive mode of reading that is identifiable through three key components of response to literary texts: (1) foregrounded textual or narrative features, (2) readers' defamiliarizing responses to them, and (3) the consequent modification of personal meanings (see also Miall & Kuiken, 1999), or "refamiliarizing/reconceptualizing strategies" (Fialho, 2007).
Research on the first key component of the dehabituation theory, what foregrounded textual or narrative features contribute to the second and third components is underdeveloped. In other words, it remains unclear what textual components or what it is exactly that causes defamiliarizing responses and the consequent modification of personal meanings, or transformative reading (see also Hakemulder, Fialho, & Bal, 2016, p.23). Other questions yet to be investigated involve the role of individual differences (Henrich, Heine, & Norenzayan, 2010;Ross et al., 2006) and the conditions of implementation (Schrijvers et al., 2016(Schrijvers et al., , 2019a(Schrijvers et al., , 2019b(Schrijvers et al., , 2019c, in preparation; see Section 3).
Research on the second and third components have been more illuminating. Insights into the second key component, readers' defamiliarizing responses, or the forms of cognitive and emotional engagement from the part of the reader in literary experiencing have been described in many ways. Among the theories and models put forward in the psychology of art, reading has been seen as a "simulation" of social phenomena, prompting readers to go "beyond the schema given" (Miall, 1989, p. 55). As Oatley and colleagues suggest, feelings may play a mimetic role. While empathizing with characters, readers take on their goals, feel their emotions, and draw upon the same social skills that enable them to understand others (Oatley, 1994, p. 66). Kneepkens and Zwaan (1994) elaborate upon the ways different feelings in literary response might interact and provide empirical evidence for such arguments. For instance, they show how readers become less and less involved in technical aspects and contextual information (A-emotions or Artefact emotions) and more involved in character and event descriptions in the story world (experiencing F-emotions, or Fiction emotions) as a story progresses (p. 134). Their contribution, however, resides much more in the kind of questioning their discussion generates, materialized in a number of empirically testable predictions (idem 136). The "transportation" model has argued that, in literary experiencing, readers are transported into narrative worlds ( (Green & Brock, 2000;). According to this model, in imagining what it would be like to be in the characters' shoes, readers sympathize or identify with characters positively or negatively, and share or reject their views of the world (Green, 2004;Hakemulder, 2001;Oatley, 1994Oatley, , 1999. Literary reading processes have also been described as coming about via a merging of boundaries between self and others, a result of "experience-taking", where readers take the experiences of the text as their own (Kaufman & Libby, 2012;Sikora, Kuiken, & Miall, 2011), a form of enactive engagement (Fialho, 2012). Despite the kind of metaphor used to describe forms of reader engagement, these models and theories seem to share a view on literary experience as a combination of emotional and cognitive processes.
As a development of the cognitive turn, some studies have focused on the third key component of the dehabituation theory, the consequent modification of personal meanings, or how readers perceive reading and how it changes the reader. Two centuries after Matthew Arnold assumed that literature made readers better persons (see Section 1), scientific efforts to check this assumption began to sprout, but empirical research on this issue is still rather scarce. So far, explorations on a form of reading that transforms the reader's self as well as his or her perceptions of others (Fialho, 2012) have generated two working hypotheses: reading literature impacts concepts of (1) "self" and (2) "other." These hypotheses have been examined experimentally (Bird, 1984;Djikic, Oatley, & Carland, 2012;Djikic, Oatley, Zoeterman, & Peterson, 2009;Kaufman & Libby, 2012) and qualitatively (Coady & Johannessen., 2006;Csikszentimihalyi, 1990;Fialho, 2002;Miall & Kuiken, 2002;Osen, 2002;Radway, 1984;Rosenthal, 1995;Toyne & Usherwood, 2001;Trounstine & Waxler, 2005). The experimental work indicates that literature may influence readers' outgroup perceptions (Hakemulder, 2000;Johnson, 1993) as well as their mentalizing abilities (e.g., Mar, Oatley, Hirsh, Dela Paz, & Peterson, 2006). Some studies suggest that for the effects to occur, readers need some form of engagement with the text through absorption in the story-world and appreciation of the style (Koopman, 2016b), sympathy (Johnson, 2012), empathy and absorption (Bal & Veltkamp, 2013). Qualitative studies have revealed that literature may potentially deepen readers understanding of the position of the self in the world. They also suggest that some form of emotional engagement might be a precondition for such an effect.
A better understanding of the reading experience itself and the interplay between the second and the third components of the dehabituation theory, namely, readers' defamiliarizing strategies and the consequent modification of personal meanings (or, readers' reconceptualizing processes, cf. Fialho, 2007), may provide more insight into what kinds of narrative and emotional engagements are involved (see Polvinen and Sklar in this special issue). Empirical (Hakemulder, 2000) and phenomenological studies have been quite illuminating here (see Bálint, Hakemulder, Kuijpers, Doicaru, & Tan, 2017). By offering a hybrid between qualitative and quantitative procedures, Miall and Kuiken (1995) proposed a system and a measurement to classify individual differences in readers' orientation towards literary texts. The items they describe refer to shifts in selfunderstanding and to changes in the reader's perceptions of less personal matters. Later, (Miall & Kuiken, 2002) attempted to describe the different forms of feelings in literary reading, which they categorized as evaluative, narrative, aesthetic, and self-modifying. Building on this initial distinction, (Kuiken et al., 2004) showed how literary reading has the capacity to deepen selfunderstanding and one's perception of everyday life, especially after a personal crisis, or what they called "expressive enactment." Trying to locate whether and how "expressive enactment" occurred within literary reading, Sikora et al. (2011) found that self-perceptual change occurs through a succession of evocative moments during the reading, which involve (a) aesthetic feelings, as well as narrative feelings in response to situations and events in the text; (b) blurred boundaries between the self and the narrator or story characters, suggestive of metaphors of personal identification; and (c) active and iterative modification of an emergent affective theme. For some readers, these iterative modifications move toward saturation, richness, and depth (Kuiken et al., 2004), which they called "selfmodifying feelings".
Drawing on these findings, the TR project examines how literary narrative fiction may deepen readers' self and social perceptions. The aim is to obtain a rich description of the phenomenon (i.e., what is transformative reading like? What are the components involved?). A second issue is how the different TR components are in a relationship with one another so that a theoretical-empirical model of transformative reading may be offered.
Looking specifically into how changes in the sense of self occurs as the reading experience unfolds, Fialho (2012) showed that "self-modifying reading" is not a monolithic phenomenon. Two types of experiences are articulated, each distinguished by different "modalities of consciousness 4 " and ways of embodied repositioning of the reader in relation to the text (Iser,  1692532 1974Merleau-Ponty, 2002). One type is mediated by the setting and another one by engaging with characters. Each of these two types seems to entail a different dimension of self possibly resulting from a different form of engagement with narratives. In the first, the reader recognizes aspects of the setting in self. In the second, social concepts (perceptions of others) are involved. Through engaging with characters, the reader recognizes aspects of the protagonist in self and others. Common to these two types is a temporal aspect that seems to be essential to its nature. These two types, their dimensions and forms of engagement have generated two models of reading processes: a form of reading that transforms the reader's self as well as his or her perceptions of others (Fialho, 2012). Fialho (2018, in preparation) now aims at gaining access to how readers describe their subjective experiences of TR and is exploring the moments in which changes in self and self-and-other constructs occur. Thirty thematic semi-structured in-depth interviews have been conducted with native speakers of English, in two sessions. In the first, participants reported their most memorable TR experiences. In the second, they selected a story that was reread with a focus on five evocative passages. As a result, an inventory has been offered. Currently, she is investigating how the different components interact, thus opening ways to designing empirical process models.
From a more quantitative perspective, (Fialho, Hakemulder, & Hoeken, in preparation) have been aiming to articulate the underlying dimensions of TR to find out their relationship. They have been developing and testing a scale that measures both the experiences during and after reading and charting the relations between them. Preliminary results further detail the types of reading feelings suggested by Miall and Kuiken (2002) and have also enabled the articulation of an empirical model of transformative reading.
So far, the theoretical-empirical model of transformative reading (Fialho, 2018) distinguished two outcomes-insight into oneself and into others-and identified six underlying components. Adult readers who participated in studies about reading experiences that had transformative impact on them, indicated that they vividly imagined the setting and characters in texts (imagery), recognized aspects of themselves or others in characters (identification), enacted and embodied the experiences of a character (experience-taking), evaluated characters positively or negatively (character evaluation), felt sympathy and compassion for characters (sympathy), and noticed which words, phrases or sentences were particularly striking or evocative to them (aesthetic awareness). For the readers investigated, these particular experiences preceded new or deeper insights into themselves and others (self-other insights). In light of the dehabituation theory, the six underlying components (imagery, identification, experience-taking, character evaluation, sympathy, and aesthetic awareness) are arguably readers' defamiliarizing strategies (see Fialho, 2007, for the description of other defamiliarizing strategies that are not "transformative"). The result of this complex engagement with texts (self-other insights) is where modifications of personal meanings are observed.
What the TR Program has shown so far is that transformation (defamiliarizing strategies and the consequent modification of personal meanings) does occur, but it seems to happen naturally and unexpectedly. And when readers eventually experience shifts in sense of self while reading, when they gain fresh understandings of who they are as individuals, their reading becomes meaningful, as they realize how they may be changed.
In the TR Program, it is proposed that the way texts are approached, or their uses, determines if readers will become aware of personal relevance. Experiencing reading (i.e., reading for personal response and affective resonance rather than for analysis and interpretation) has been shown to foster different forms of emotional resonance, of self-implication, and of self-reflection, and perceiving the text as a meaningful experience (e.g. Fialho et al., 2011;; for a review, see Hakemulder et al., 2016;Schrijvers et al., 2016). Changing insights about what the life of others might look like may depend on the degree to which readers make efforts to imagine themselves in Fialho, Cogent Arts & Humanities (2019), 6: 1692532 https://doi.org/10.1080/23311983.2019 the shoes of fictional characters representing these others (Hakemulder, 2008;Johnson, 2012). In this respect, the articulation of first versions of the TR model (Fialho, 2012(Fialho, , 2018 through qualitative and quantitative methods investigating the role of different modes of reading in deepening the perceptions of self and others have enabled the design of evidence-based approaches to literary narratives that help readers uncover personal meaning. It is important to stress here that TR is not a reason to read. In the interview studies conducted, readers do not usually report going to libraries, bookstores to look for books that will change their lives (Fialho, 2018, in preparation). It is not a "why" they read (hence, the title of the article). When they gain fresh understandings of who they are as individuals, their reading gives them a sense of purpose by adding meaning to their lives. In this sense, only during the act of reading, readers find out what art is for.

Uses of transformative reading in education and at the workplace
The purpose of literature-to change the reader (not necessarily for the better or for the worse)can be expressed by its uses (see Introduction). How literature changes the reader, or the essence of what literature is, has been operationalized by means of the theoretical-empirical model of transformative reading, describing its components and how its components relate to one another. In other words, the TR model describes how literature defamiliarizes perceptions and modifies personal meanings, or readers' reconceptualization processes (Sections 1 and 2).
The main research questions at the moment are how the TR model can be used in different social contexts and what the role of contextual factors (i.e., types of instruction) is to this form of reading. So far, the model has been applied in both the academic (students of literature) and nonacademic contexts (participants in reading workshops in business settings).
In the educational context, the model was adapted and applied to grades 10-12 young adult Dutch students (for a complete program, see Schrijvers, 2019). The Dutch Institute for Curriculum Development argues that literature education is important for broadening students' personal, social and cultural horizons. The aims of this study were to examine the impact of literature education on students' self-and social perceptions and to explore relationships between students' learning experiences and their teachers' classroom practices. First we asked whether, indeed, (1) adolescents gain personal and social insights through reading in the secondary literature classroom, and (2) how these perceived learning outcomes are related to their teachers' approaches to various aspects of literature teaching. Dutch students (N = 297, grades 10-12) and their teachers (N = 13) were assessed and findings showed that nearly all students (99%) reported to have learned something about themselves and others through literature education, mainly personal characterizations of oneself and others, learning about oneself and others as literary readers, descriptions and evaluations of people's behaviors, and lessons for life. In addition, teachers' reports of more classroom interaction and student autonomy were related to students' more frequent reports of personal and social insights, but this could also partly be explained by students being more familiar with fiction and having a more positive attitude toward literary reading (Schrijvers et al., 2016).
As a second step, we explored whether and how literature education may foster adolescent students' insights into self and others. A systematic review of 13 experimental and quasiexperimental intervention studies yielded instructional design principles on (a) text selection; (b) activating, annotating, and reflecting on the personal life and reading experiences in writing activities; and (c) verbally sharing these experiences with others in exploratory dialogues. Such review resulted in design principles for literature education to foster students' social behavior, their attitudes toward outgroups, their moral development and their personal responses to fiction. To this end, we concluded students must (a) engage in exploratory dialogues in which a variety of personal responses can be expressed and shared, based on (b) reading and writing activities that focus on noticing personal responses and connecting these to prior life experiences, with regard to (c) fictional novels, short stories, poems or passages containing social-moral themes (Schrijvers et al., 2019a).
The theoretical-empirical TR model, the explorative study in Dutch literature classrooms, and the three design principles identified in a review of previous intervention studies enabled the design of a literature classroom intervention for 15-year-old students in the Netherlands, which aimed to foster their insights into themselves, fictional others, and real-world others. A first intervention was developed in collaboration with teachers, tested, and redesigned into a second intervention, resulting in the design of a valid and practical domain-specific program titled Transformative Dialogic Literature Teaching (TDLT, Schrijvers, 2019;Schrijvers et al., 2019b; the complete TDLT instructions package is offered by Hakemulder et al., in press).
Another study, quasi-experimental, assessed the effects of the newly developed TDLT intervention on 15-year-old students. Six TDLT units centered around short stories about "justice and injustice". Students were stimulated to engage in internal dialogues with stories and in external dialogues with peers about stories and reading experiences. TDLT students (N = 166) were compared to students who received regular lessons (RLT) focused on the analysis of literary texts (N = 166). Analysis of quantitative and qualitative data indicated that TDLT fostered (a) students' insight into self, fictional and real others, (b) eudaimonic reasons for reading, (c) reported use of strategies to deal with difficulties in literary texts, and (d) motivation for literature education, whereas RLT did not (Schrijvers et al., 2019c;IN Preparation).
So far, the series of empirical studies here described have shown that literature education may be a promising domain for affecting adolescents' insights into self, fictional and real others. These studies suggest that TDLT may guide students toward developing such insights. The instructional differences between the experimental and control condition confirm previous insights (Fialho et al., 2011 and imply that we may indeed want to move away from formalist, knowledge-oriented instruction that may still exist in literature classrooms, as aptly described by Wilhelm (2016): Teachers […] may emphasize knowing and recognizing literary devices, getting at the "internal logic" of a text's construction […], and relating a work's central "organic" meaning to how this meaning was expressed. There may be an emphasis on "rightness" of literary interpretation. Interpretative questions about the text will be answered after reading […], and discussions mediated by the teacher, who acts as the authority on the text (p. 25, quoted by Schrijvers, 2019, p. 181).
If developing insights into self and others is acknowledged as one of the objectives of literature teaching, a formalist, knowledge-oriented approach appears not to be too helpful. In contrast, as in TDLT, instruction should encourage students to explore their personal responses in dialogic interactions with and about literary texts, by completing purposefully designed combinations of pre-, during-and post-reading tasks in which analysis of literary devices is a means to reason about reading experiences, themes, characters, and moral implications.
Several are the elements still needed to better understand what causes such effects for a more central function of the benefits of reading in this context: (a) the role of the text (i.e., specific narrative features); (b) the literary experience itself; (c) the reader (i.e., identity, general reading Fialho, Cogent Arts & Humanities (2019), 6: 1692532 https://doi.org/10.1080/23311983.2019 orientation); (d) the approach (i.e., instructions, educational approach, etc.). At present, the TR program can contribute with better insights into the role of the reading experience itself and the approach to reading.
In this context, adaptations and applications of the TR model are ongoing and still preliminary. One of such adaptations has resulted in Empathy Reading Workshops. These are evidence-based interventions that aim to (a) enable participants to experience, or live through texts (vs. analyzing, cf. Fialho et al., 2011Fialho et al., , 2012Rosenblatt, 1938Rosenblatt, -1995Schrijvers, 2019); (b) make participants aware of notions of empathy in reading and life at the workplace; (c) sensitize participants to relations among cognitive and affective processes involved in TR and in interpersonal relationships; (d) give participants the opportunity to practice such processes; (e) suggest how such processes might be applied in real-life situations (Fialho, 2017).
Intervention studies have not been carried out yet, but the TR components as described in the model (see Section 2) have been adapted into the form of TR activities. We have been testing the effects of each activity on scores on empathy, access to possible selves, changes in self and social (self-other) concepts. Current research findings have shown that embodied approaches to literary reading seem to be the most effective in fostering empathic engagements and identification with story characters, which seem to be essential components of TR (Fialho, 2017).
By applying TR to the workplace, the TR program has been shifting the research agenda to better understanding the causal mechanisms of the effects of reading fiction on social cognition (cf. Mumper & Gerrig, 2016). In fact, Mumper and Gerrig (2016)

may be correct: what is relevant is
perhaps not only what we read, but also how we read and how we invite readers to read. Designing embodied approaches to reading seems to be promising. So far, we have improved our understanding of how individual and shared reading can be transformative. Further studies could provide theoretical and empirical contexts for differentiating effects of the theory of mind and empathy.

Conclusion
This paper has reflected on the purposes of literature from the perspective of literary criticism. It has shown how throughout the development of the theories proposed, the debate has shifted from focusing on the text, the author, the contextual conditions, and/or the reader. This brief review has shown how, despite differing perspectives, literature can be seen as a source of self-knowledge, revealing and concealing much of who we are. It can be a rich source of personal meaning and what we do with it and how we use it also reveals its purpose.
The TR program helps bring the issues of what we do with literature and how we use it to the limelight. It investigates how readers refer to the impact of reading on their self-awareness while engaging in fiction and, through this process, find personal meaning in the act of reading. So far, its main result is that the purpose of literature lies in the experience itself, in its power to prompt us to connect deeply and conscientiously with our emotions, deepening our senses of who we are, what we are in this world for, and how we are in a relationship with others. Such findings become premises that inform the applications of TR in social contexts. The TR program starts from insights from literary scholarship and creates evidence-based principles for the experiential purpose of literature to unfold in different contexts. The TR workshops are about the experiential purpose of literature as they invite readers to experience texts and not to engage it from an analytical perspective. Readers have been seen to read for affective resonance and personal relevance. TR workshops are, thus, evidence-based approaches to literary narratives that help readers uncover personal meanings.
We are just in the early stages of this research program. However, the empirical work on transformative reading conducted so far suggests new conceptual distinctions that could contribute to theorising about "purpose" in literary studies more generally. To be precise, at present, transformative reading is conceptualized in light of the dehabituation theory of literature (Miall, 2006). This raises the question of whether purpose is theory-specific. In this sense, The TR model is both theoretical and empirical, offering a description of the components that characterize readers' forms of engagement and reconceptualizing strategies (Fialho, 2007). A question that follows regards the purpose(s) of each of the constructs articulated so far, namely imagery, identification, experience-taking, character evaluation, sympathy, aesthetic awareness, and self-other insights for a literary reading. Finally, research on transformative reading indicates a new strain in applying literary reading and literary theory to different, practical fields. An implicit question here is whether literature would serve different purposes in different contexts and the extent to which the learning goals in each learning environment would need to be context-dependent (see Burke, Fialho, & Zyngier, 2016).
In terms of future empirical work, many challenges lie ahead. Issues that still need to be investigated are: (1) the role of the text; i.e., a systematic comparison between literary and other forms of reading and what it is exactly in the text that causes the particular effects. Future research may benefit from making a clearer distinction between the effects of "literariness," "narrativity," and "fictionality." It may also benefit from looking into what in the text evokes embodied resonances from the reader (Caracciolo, Guédon, Kukkonen, & Müller, 2017;Kukkonen, 2014;Kuzmičová, 2014); (2) the role of individual differences: whether such effects occur for any reader, for avid readers (cf. Ross et al., 2006) or just for a small elite of highly educated students of English literature; (3) the conditions of implementation: the role of approaches to literature, for example, finding out whether instructions-assignments, educational approach, providing background information about authors and texts-can enhance these effects (see also Hakemulder et al., 2016).
The design of evidence-based interventions enables a multiplicity of uses and concretizations of the aesthetic values of literature. In line with Felski (2008), who claims that aesthetic value is inseparable from its use, the TR program might also be adapted to investigate, for example, how the search for meaning may be related to positive health outcomes (cf. Rieger, Reinecke, Frischlich, & Bente, 2014;Wirth, Hofer, & Schramm, 2012). Eudaimonic responses to movies may be associated with autonomy, relatedness and competence, the three constructs that Self Determination Theory predicts to be related to well-being. Here, the search for meaning can be studied from the perspective of positive psychology, testing hypotheses concerning positive health outcomes. In particular we might take "positive health"-"the ability to adapt and self-manage, in the light of the physical, emotional and social challenges of life" (Huber, 2014)-as one of the key outcome measures in future studies. Among the six main dimensions that constitute "positive health", Huber (2014) names "meaningfulness" as the most important dimension. This way, the TR program might add to the insights of studies in media psychology about the eudaimonic effects of media (e.g., Bartsch, Kalch, & Oliver, 2014) in contributing with a better understanding of how we uncover meaningfulness through engaging with the arts, a topic largely neglected by researchers outside philosophy and religion studies.
All in all, the TR program aims to contribute to the relevance of the Humanities in general. A central tenet is that reading literature means experiencing the world. It is true that, so far, TR has focused on transformations, but if art is a source for meaning-making by fostering selfrefection, and more conscientious awareness of how we engage with ourselves and others, it impacts social cognition, and such abilities as empathy. Literature has, then, both personal and social purposes.