Volume 1, Issue 2

This article takes the classic film Casablanca as its example in describing a multilevelled model of how spectators react emotionally to films. The model combines concepts drawn from film studies, reception psychology, and the study of emotions. The aim is to allow the connections between film structures, cognitive processing, and emotional responses to be described in a nuanced manner. Emotions are understood here as psychophysical phenomena shaped by both biology and culture, as multidimensional events in the mind and body of an individual that respond to external or internal stimuli. The emotional responses of spectators are influenced by various kinds of cognition – perception, association, imagination, and memory, for example – and thereby develop into episodes of varying complexity. My model distinguishes between four levels in the development of the emotional episodes concerned. 1. The direct perception of sounds and images stimulates perceptual affects , that is, sensations and moods that do not concern a specific object. Moods induced by music are an example of this. 2. On this basis, spectators develop mental models of the represented world that bring forth diegetic emotions such as suspense, sympathy, empathy, and emotional appraisals. 3. When spectators grasp indirect meanings on the basis of the represented world, thematic emotions are activated, often involving social norms or individual values. 4. Finally, on a level of reflection, spectators evaluate the communication process and its elements: their own responses, the responses other spectators are assumed to have, the film-makers and their intentions, and the shaping of the film. This is the source of communication emotions such as moral outrage, fear of social consequences the film might have, or pride in one's own expertise. These four levels interact with each other so that a complex web of emotional stimuli and processes takes shape. Individual emotional episodes can be confined to a single level or unfold across several levels, and several emotions can overlap in time. The upper three levels of emotionality, at least, also feature in the reception of literary works, so the model could, with some modifications, be transferred to the analysis of literature. In Casablanca , as in most film drama, diegetic emotions are foremost. This level is also the one that has been discussed most extensively and controversially, so it constitutes the focus of the article. First, two ways of describing diegetic emotions are distinguished. The description of the narrative emotions of suspense, surprise, and curiosity is relatively abstract and highlights the play with spectators' knowledge that unfolds in time. Implicitly, though, it assumes a more tangible answer to the question of how emotional responses to characters and other elements of the represented world come into being. There are at least five competing explanatory approaches to answering this question: it has been suggested that spectators evaluate characters according to objective moral or non-moral criteria, according to their abilities or attractiveness, for example; evaluate them subjectively in terms of their own interests (erotic interests, for example); empathize with them and experience similar feelings to them due to emotional contagion, identification, simulation, or projection; activate emotional memory and their own recollections; or evaluate the represented situation in terms of the interests of the characters with whom they sympathize. My model proposes that all these possibilities be considered and derived from the more general phenomenon of perspective-borne appraisals . Following multilevel appraisals in psychology, we can assume that characters, situations, and other represented objects give rise to emotion-triggering appraisals on various levels in spectators. The model proposed here places special emphasis on the fact that spectators adopt a particular mental perspective in the process: they grasp the object in question in a specific, qualitative way by means of mental processes of perception, thought, feeling, evaluation, or desire. As spectators, we ourselves adopt a particular perspective toward what is represented, but we can also attribute film-makers, narrators, or characters with a perspective and compare it with our own. Transfer of perspective or identification with characters can take place to various degrees in relation to various objects and aspects of them: we can, for example, share the perception and knowledge of a character whereas our desires and feelings can be different (and vice versa). Thus, although the explanatory approaches outlined above identify various ways in which characters and situations can be the object of perspective, they are not mutually exclusive alternatives but more a continuum of possibilities. Applied to the analysis of Casablanca , the model put forward makes it possible to compare various explanations of the film's emotional effect, to test their plausibility, and to combine some of them and add new refinements to them. In this way, we obtain results that are more comprehensive and nuanced than those of the approaches employed to date, which derive emotional responses to the film from, among other things, an accumulation of stereotypes, identification with Rick's Oedipal development or moral perfection, and strategies involving suspense and nostalgia. Discussion of the film's closing scene as a case study, on the other hand, points to a combination of the following potential emotion elicitors in first-time adult audiences in the present-day Western cultural area: – a smooth, melancholic underlying mood on the perceptual level, generated by frame composition and music; – activation of emotional memories that recall personal experiences of separation and are bound up with tenderness, nostalgia, and sadness; – emotional intensification that results from caring emotions of concern for others and involves the use of catchphrases and musical leitmotifs; – empathetic sadness put across by, among other things, the voices and facial expressions of the actors; – moral evaluation leading to sympathetic alignment with Rick and Ilsa, which activates emotional scripts and adds respect, acceptance, and hope to the sadness; – a liberating and morally satisfying resolution of suspense and curiosity; and – responses to overall thematic statements of the film (such as ›moral integrity is what matters most‹). The set of redundant stimuli means that it is probable spectators will experience such emotions – but those listed above do not even come close to covering the whole range of possible spectator emotions. The power structure involving Rick, who is in control, and Ilsa, who is helpless, for example, can play a role as well, with partial identification with both of the two characters being possible. The different interpretations of Casablanca suggest, moreover, that spectators who watch the film several times often tend to engage in more far-reaching speculation about the ambivalent motives of the main characters, about the nature of their love and how serious it is, for example. Such speculation is also relevant where emotions are concerned. The lasting success of Casablanca and the comments of many spectators indicate that it is not just the story of the film that resonates, but also its overarching meanings, that the conflict of ›love vs. virtue‹ (Renault) has the potential to move many people, and that admiration for the aesthetic achievement of the film-makers remains undiminished.

The Discussion of Emotions in the Study of Medieval Literature Work on the theory and history of emotions is one of several new approaches to medieval studies that are currently the topic of considerable discussion. Research on emotions carried out by medievalists working in the field of literary studies shows certain points of contact with earlier approaches such as historical anthropology, historical psychology, and the history of mentalities; but it differs from them in its methods and in the insights it pursues. The most important difference lies in the fact that it does not see textual analysis as a methodological tool that can be used to arrive at all-encompassing conclusions about the psychological dispositions of historical individuals. This article attempts to set out some basic features of work on emotions by scholars of medieval literature. A controversial issue is raised in the process, namely the problem of what insights such work sets out to achieve. At times, the impression has arisen that it attempts to derive a phenomenology of historical emotions. To provide clarification on this point, some differences between the approaches of historiography and literary studies are first explained. Then, several methodological premises and areas of interest in a particular kind of work on emotions in medieval studies are considered in more detail and explained with the help of selected textual examples. The line of enquiry chosen here links the study of emotions to that of performativity; Austin's approach in particular has proved to be relevant to work on emotions from the perspectives of performativity and constructivism. For a long time, the relationship between literary characters and their emotions in the hands of medieval poets was understood in the tradition of Elias as one of passivity. The expression of emotions was described using terms such as ›release‹ and ›overwhelming‹. Work on emotions by medievalists, however, has revealed that emotions can be staged, shown, and brought to the surface. From the perspective of literary studies, furthermore, the insights pursued by medievalists in their work on emotions concern the cultural semantics and (special) logic of emotions, above all in fictional literature. This includes the study of how emotion and body and emotion and language are related, and of the social dynamics associated with emotions. With the help of the latter issue – which involves the social functioning, or performativity of emotions – it is possible to understand conflict scenarios and contexts of action in medieval texts in an entirely new way. Modern approaches to the theory of emotions can contribute to and complement such work. Relevant here is a concept of emotion based on social constructivism, according to which emotions are brought into being on a neurophysiological basis, as reactions to social situations that they evaluate at the same time, as well as being given expression in culturally governed ways. The assumption that there are cultural – and thus historical – conventions and styles for the expression of emotions is of crucial significance for work on emotions in medieval studies. It does not, however, provide a self-sufficient theoretical foundation that could be employed in the analysis of literature. Methodological reflection on what is meant by a particular emotion from the perspective of emotion theory should be included in analytical practice. In many cases, such as religious and secular drama and meditative tracts, as well as where courtly literature is concerned, investigating the presentation of emotions in literature also requires reflection on the cultural practices in which it is embedded. Moreover, although narrative texts have been at the forefront of attention to date, all genres and text types of medieval literature are potentially relevant – even those anchored in performance situations. Work on emotions in literary studies refers to emotions as having the nature of signs and codes. This is a way of saying that emotions and their functions in literary texts have the status of signifiers. It does not mean that the signs are being studied as representing emotions located outside them. Literary expressions of emotion can also be understood as signs in pragmatic and functional respects. This is particularly true of medieval texts that focus on modes of symbolic and non-verbal communication. Medieval literary texts contain established ways of expressing emotions and interacting by means of emotions; they are subject to rules that are not immediately obvious but can be reconstructed and described by means of analysis and interpretation. Applying the term ›code‹ in such cases does justice to this fact. The study of literature can also provide, for example, knowledge of the cultural encoding of emotions in samples of period-and genre-specific material. An analysis of emotions based on functional pragmatics can show the extent to which expressions of feeling do more than make the inner state of a character externally visible. Expressions of feeling carry out a social function and have an effect in time and space; this can be understood as the role they have as actions and their performative dimension. The concept of performativity has the fundamental conceptual advantage of allowing the pragmatic functions of emotions to be treated on different levels. An expression of feeling, for example, is illocutionary and carries out an action in that it makes an appeal, and perlocutionary and followed by an action in that it affects someone. The performativity of literature thus includes the extra-literary performativity of effect. The hypothesis that emotions are not only represented but also produced by literature provides a foundation for describing an intradiegetic level of composition. The emotional affects of characters can be described on this level, which can be analytically distinguished from an extradiegetic level on which those affects have the potential to extend to the recipient. As a level of literature, it is relevant to the analysis of the performativity of literary texts, and also offers a way to conceptualize the long-established assumption in medieval studies that discourses and mentalities are interdependent. These questions do not amount to treating the literary text as a document of real emotions that exist independently of it and are able to provide evidence of that existence. It is often believed that medieval texts contain only ›simple‹ emotions, or affects; in opposing this view, it is possible to cite the variety of emotional experience that in courtly literature, for example, takes many different forms. The transitoriness of emotional processes, their lack of ambiguity, and not least the connection of rational and emotional factors all form another important area of analysis where medieval texts are concerned. Poetic language would seem to prove itself at precisely those points where it is necessary to make present emotional states and processes that are not transparent, or not fully transparent, as such.

Analytic philosophy is not what it used to be – and thank goodness. Its practice in the late Twentieth and early Twenty-first centuries is not grounded in a grand metaphysical design or a particular philosophy of language. If there is a core or central character or ›methodology‹ in analytic aesthetics, philosophy of art, and the perspectives they take on emotions, it is better revealed in current practices rather than in analytic philosophy's origins. One salient feature of the practice of analytic aesthetics – a feature it shares, not coincidentally, with much philosophy of mind, especially in the subfield of philosophical psychology – is the respectful role that is afforded to what psychology and related fields of scientific inquiry have to tell us about emotions. And though introspection has long been known to be inadequate as a psychological method, the highly counterintuitive results of numerous psychological studies of emotions undermine the viability of conceptual analysis or ordinary language alone as a philosophical methodology to explain the character of emotions, including those expressed in art and experienced by appreciators of the arts. One feature of analytic philosophy that has not changed is its focus on issues or questions rather than, for example, histories and genealogies. The logical implications of certain definitions of emotion are central to several sets of issues that have arisen within analytic aesthetics in the last decades of the twentieth century and continue to engage philosophers in the early years of the twenty-first. Section 2 introduces the debates over how to define emotions in light of a well-known puzzle in the philosophy of art, the paradox of fiction, which are fueled in part by conflicts within the field of psychology with respect to whether particular types of beliefs or other cognitive states, such as thoughts or imaginings, are necessary conditions for emotions. Section 3 examines whether emotions are, or can be, rational. Emotions may be generated by various thoughts or imaginings rather than beliefs. A belief in the truth of what one reads in a work of fiction may be unwarranted and for this reason irrational, but it does not follow that thoughts or imaginings will likewise be irrational. Imagination has always been problematic in relation to reason and art's potential value, and here again, psychological research has something to tell us about the matter, this time by exposing emotions' purported role in the evolution of the species and in the learning history of individual persons. Imaginings are thus not judged on evidential grounds, such as are beliefs, but in relation to their historical origins, such as in the appeal to evolution, or to their role in the development of more wideranging mental capacities, such as in the appeal to the learning histories of individuals. Empathizing and sympathizing with fictional characters are more specific applications of the resources of imagination, and they too have received more systematic attention in relation to how humans engage with fiction. I concentrate on empathizing with fictional characters in Section 4, and how it is similar to and different from empathizing with persons in real life. There are two plausible, commonly (but not universally) accepted necessary conditions for empathy. One is that the empathizer must end up in the same or highly similar emotional state as the one with whom one empathizes (whom I shall refer to as the target). To empathize with someone who is angry, for example, one must experience something akin to anger rather than, say, delight. Call this the similarity condition. It is plausibly and typically supplemented by another allegedly necessary condition, a (roughly) causal condition that one must come to be in that state in the same or a sufficiently similar way. On one account, one empathizes in the first instance with a hypothetical reader of fact, who may, in turn, empathize with a character in the work. Though this activity clearly calls for imagination, it also leaves out what is distinctive of literary appreciation, such as a responsiveness to the formal qualities of a work or the way a story or plot is constructed. In Section 5, I discuss connections between emotions and specific genres of art. Since Aristotle, we have worked with the idea that particular types or genres of art are supposed to evoke particular types of emotions, an idea that has returned to the fore since the concept of a general, generic ›aesthetic emotion‹ has become more discredited. Analytic philosophy of art contains the more specialized philosophies of the visual arts, of literature, of music, of film, and more, which themselves admit of subdivisions that emerge out of the practices of artists and of those who theorize about the arts. According to Aristotle, a good tragedy must evoke pity and fear in the audience. Comedies should amuse their audiences, and some popular or mass art genres, such as horror should produce an experience of what has been dubbed »art horror«. Though Aristotle argued for the beneficial effects of pity and fear, and perhaps also sympathy, in response to tragedy, it is not clear that all emotional responses to art are morally benign, especially where humor is concerned. Not only does one wish for an explanation of how one could feel pleasure along with the otherwise painful and disturbing emotions that are often specific to a genre, but one also needs to consider how it is possible for a moral person to feel such pleasures.

Encoding Emotions in Texts This is the first instalment of a two-part study that examines emotions from the perspective of linguistic semiotics. The underlying issue at stake is the relationship between linguistic forms and the meanings attached to them. This part of the study lays the theoretical foundations for a consideration of the issues involved. In semantic terms, emotions can be explicated as predications: they assign to an experiencer the property of having particular subjective psychological experiences and forms of motor behaviour in certain situational conditions . When emotions are encoded linguistically, however, it is not necessary for each part of the predications involved to be lexically realized. In this respect, the linguistic meanings encoded in the lexical configuration of a language's grammatical and textual structures can be treated as underspecified. This allows us to analyse grammatical and textual structures as structures that are stringently compositional in nature, and to consider the meaning of linguistic signs and their expression on various levels. Meanings can be determined by lexical factors, grammatical factors, or factors relating to textual structure. In the case of both spoken and written utterances (that is, in conversations and texts), it is possible to convert such meanings into interpretations by making recourse to areas of knowledge other than that of grammar. These interpretations are tied to certain referents in a text or conversation. Emotional attitude ( emotionale Einstellung ) and emotional scene ( emotionale Szene ) are introduced as descriptive concepts for explicating emotions. They permit a more detailed understanding of the semantic and conceptual aspects that are relevant to the linguistic encoding of emotions. Emotional attitude is used to refer to the evaluative connection between a bearer of emotion and a relevant concept. Emotional attitudes can be formalized by means of three quantifying functions. These quantifying functions assign emotional values to emotional attitudes . The emotional scene serves as a descriptive tool for explicating clustered situational conditions of emotions. An emotional scene consists of the following descriptive features: experiencer, stimulus, emotional values, and judgements about the conditions for emotional evaluations. This last descriptive feature is a complex one and is formed using various predicates, including a set of basic predicates that are described in more detail and denote the qualities of internal states of living things that can be perceived through introspection.

In the following pages, I will consider what might constitute a neurocognitive account of romantic love and its relation to literature. Like all emotions, romantic love has eliciting conditions, expressive and physiological outcomes, characteristic modes of cognitive processing, actional outcomes, and so on. Undoubtedly, some of these change culturally and historically (e.g., the actional outcome of sending one's beloved sweets on February 14 is not universal). However, many aspects of romantic love are constant – a desire not only to experience sexual union with the beloved, but to be in close physical proximity with him/her; a painful feeling when separated; characteristic cognitive processing that involves the imagination of long-term union; certain sorts of physiological arousal, both sexual and non-sexual, etc. The ubiquity of this emotion is attested to by the cross-cultural and trans-historical breadth of its literary representation. As the last point suggests, beyond the usual problems facing any account of emotion, the relation of romantic love to literature poses several further questions. First, why is it that romantic love is so pervasive and so idealized as a prototype for happiness in narrative? This question is raised by the cross-cultural prominence of romantic tragi-comedy – the story of a hero and a heroine who fall in love, are separated (usually by their parents), then reunited to live happily ever after – a genre generated by that prototype. Second, how can we reconcile this pervasiveness and idealization with the fluctuating and unstable nature of romantic love in real life? Indeed, one might ask here why romantic love itself is not stable, why it comes and goes, why it varies in intensity even when present. Finally, literary critics tend to place a further demand on literary theory. To be valuable in a specifically literary context, a theory should not only make valid generalizations (e.g., concerning genre). It should help us understand the particularity of individual works. It should have interpretive as well as explanatory worth. This demand may appear to pose particular problems for a theory of narrative generation such as that put forth in The Mind and Its Stories . By that account, writing a story is, like other activities, guided by prototypes. The crucial prototypes for writing stories are, of course, story prototypes. The central argument of The Mind and Its Stories is that three such prototypes – heroic, sacrificial, and, most importantly for our purposes, romantic – recur across unrelated traditions of verbal art. These prototypes underlie most (though never all) narratives preserved and widely appreciated in any given tradition. The interpretive challenge faced by such an account is one of moving from this level of commonality to the level of individual distinctiveness. The first section of the essay sets out a basic framework for a neurocognitive account of emotion, modifying and complicating an account presented in Cognitive Science, Literature, and the Arts . It begins with the view that there are distinct, brain-based emotion systems. These systems are shaped by both genetic and experiential factors. One set of emotion triggers – the ›innate‹ triggers – are, in effect, fully genetically specified. A second, developmental group of triggers are genetically guided, but sensitive to particular sorts of experience in critical periods or »critical circumstances«. The third variety of triggers, emotional memories, are purely experiential. The first sort are constant across cultures and time periods. The second and third sorts may vary by culture and by individual. However, any such variation is constrained, by human biology and other factors. The second section considers the relations among emotion systems. These systems are partially autonomous. However, they interact through mutual enhancement or inhibition. Thus one emotion (e.g., disgust) may inhibit the physiological expression of a second emotion (e.g., hunger), while enhancing the actional outcome of a third (e.g., fear, which shares with disgust the actional outcome of retreating before the object). This section also takes up six emotion systems – fear, anger, attachment, disgust, hunger, and lust. It argues that disgust inhibits, not only hunger, but also lust, and that fear inhibits lust as well. At the same time, attachment serves to inhibit both fear and disgust. This section goes on to argue that romantic love is not an emotion system per se . Rather, it is the integration of sexual desire with attachment, an integration in which attachment serves to inhibit emotion systems that would themselves inhibit sexual desire. The third section turns to literature. First, it considers why romantic love is likely to be unstable and why it is also likely to define one central prototype for happiness. As a prototype for happiness, it serves to generate the cross-culturally predominant genre of romantic tragi-comedy. This section goes on to consider the place of social determination in emotion and literature. First, there may be differences in social patterns regarding emotion, due to differences in developmental patterns or in common emotional experiences. However, at the same time, there may be social patterns that recur cross-culturally. One of these is the recruitment of romantic love toward social stability. This occurs prominently in the proliferation of marriage plots – the standard form of romantic tragicomedy. The fourth section considers a peculiar aspect of attachment. Most emotions are oriented toward classes of objects. Fear or lust is often particularly excited by one object. But, in those cases, the object has fear or lust triggers of particular intensity or salience. In other words, the object is a particularly good instance of more general patterns. However, this does not seem to be the case with attachment. As psychoanalytic writers have suggested, attachment propensities are to some extent shaped by early critical period experiences. However, attachments arise in later life, and not only in relation to properties fixed in the critical period. Rather, they appear to arise in particular circumstances – for example, following childbirth or bereavement. I refer to these as ›critical circumstances‹. The final section takes up the preceding arguments to analyze Hamlet not as a revenge tragedy, but as a play about attachment. The perennial attention to some unresolved critical questions – most obviously, »Why does Hamlet delay?« – results, in part, from assuming that the play is about revenge. If the play is, in fact, about attachment, then the issues change, for our understanding of Hamlet changes, as does our understanding of his relation to his father's ghost, to his mother, to Horatio, and to Ophelia. Specifically, there is no question of Hamlet taking revenge. Revenge is based on anger. But, from the beginning, Hamlet feels grief, the loss of an attachment relation. His efforts – with Ophelia, with his Mother, with Horatio – are all attempts at establishing a new attachment relation in the critical circumstances of his bereavement. Nonetheless, in terms of the standard romantic plot, Shakespeare does something peculiar here. Usually, romantic works point toward the importance of breaking with one's parents and uniting with one's beloved. Indeed, this is what we find in Shakespeare's own plays that are primarily romantic. However, in some of his heroic plots, such as Hamlet , he makes the conflict between parental and romantic attachment much more complex and destructive, much more difficult to resolve. In that way, the crucial question raised – and, I believe, answered – by the play is not »Why doesn't Hamlet kill Claudius?«, but »Why doesn't Hamlet marry Ophelia?«.

Literature and Emotions from the Perspective of Biology A heuristics based on evolutionary psychology is put forward here as a possible way of describing the emotional effects of literature. An approach based solely on textual analysis can examine only the representation of emotions in a text, but the approach of literary psychology allows for a correlation of text structure and the mind of the reader and, thereby, for hypothetical statements about emotional effects of a text. Evolutionary psychology has the particular benefit of providing access to certain very fundamental mechanisms of emotional effect that are not as a rule accessible via introspection. Thus, the evolutionary theory of biology allows us to assume that there are a variety of emotional programs, the existence of which can be made plausible by referring to their adaptive function in the process of human evolution. A specific schema of stimuli can be identified as the trigger of each such emotional program, and literary texts can then be analysed in search of these schemata. When an emotional program is triggered by a textual stimulus, the effect can be described as one of dummy stimuli , since the triggering object is not as a rule identical with the object in relation to which the emotional program in question developed in the course of evolutionary history – it merely fits the same schema of stimuli. The article begins by introducing the theory of emotions in evolutionary psychology that underlies this model of a literary ›dummy‹. Central here is the distinction between the trigger mechanism and the program that controls the course of an emotion. The trigger mechanism identifies the triggering stimulus, initiates early physical reactions, and activates a specific strategy of cognitive processing. The program controlling the course of an emotion, on the other hand, performs a second appraisal of the situation and configures a person's behaviour on the basis of the options available in the emotional program. This decoupling of stimulus and behavioural response by means of the intervening program explains, among other things, why we respond to fictional stimuli in a way similar to, but not identical with how we respond to real stimuli: fictions trigger the same emotions as those that present themselves in real situations, but they take a different course because the situational feedback is of a different kind. In normal circumstances, we will ›feel‹ in a similar fashion (we will register the same physical and cognitive activations) but act differently (more specifically, not at all in most cases). The second part of the article outlines the adaptive significance of imaginary worlds for the ontogenetic development of the brain. In order to build up innate mechanisms during individual development, keep them ready to function, and adapt them to the cultural environment present in any given case, reasons must be found for activating them repeatedly (and, as far as possible, in situations where pragmatic factors are not involved). Literature would seem particularly well suited to activating a range of cognitive and emotional adaptations and thereby ingraining them ontogenetically. Evolution motivates us to recognize such opportunities by what Karl Bühler has called »function pleasure«: the secretion of pleasure hormones that accompanies our actions ensures that we voluntarily accept this kind of superficially pointless (›playful‹) activity and pursue it with the necessary persistence. Evolutionary psychology thus provides a concept of function pleasure that explains why we enjoy the emotional effect of works of art and why apparently paradoxical phenomena such as the ›joy of grief‹ or ›agreeable horror‹ can occur. Taking the aesthetics of landscape as an example, the third part of the article presents evolved motivational systems (i.e. systems of aesthetic preference) with an evolutionary basis and illustrates them using the locus amoenus and locus terribilis of literature. From our ancestors' need to identify habitats that were safe and rich in food, we can derive a series of preferences for scenic features that once corresponded to these requirements but are today felt to be pleasing or beautiful for no apparent reason. Stimuli such as lush vegetation, a good field of view, and shelter from the elements are approach stimuli, or »appetitive« stimuli (Jeffrey A. Gray) that guide our behaviour without our having to think about it. It follows that their use in literature, as in the topos of the pastoral idyll, should lead to a positive emotional response. Conversely, a set of repulsive, or »aversive« stimuli of aversion can be identified; they bring forth behaviour of fearful avoidance or restraint. In literature, such stimuli are familiar above all from the motifs associated with the sublime. The fourth section of the article uses two pieces of work in film studies to describe two particular emotional responses together with the textual stimuli that correspond to them. The response of crying (the emotion of ›sentiment‹) is traced back to the »capitulation response« (Helmuth Plessner) as an innate response of helplessness and submissiveness in the face of the overwhelming. Crying is thus linked to resultative plot events in narrative texts. Whenever a plot reaches a point at which there is nothing left to hope or to fear, with the reader forced to accept that a given state (the death of a character, a proven virtue, and the like) is definitive, the sentimental reaction of crying is triggered on the basis of a structural isomorphism between the experiential unit and the evolutionarily grounded trigger schemata of the capitulatory reaction. Likewise, the reaction of laughter is then presented as a domain-specific psychological response program. Laughter as a signal of willingness to cooperate may have left a particular mark on human evolution in situations of mild social tension – situations in which social cohesion was threatened but could still be restored. Accordingly, it is possible to identify as suitable stimuli for the reaction of laughter those textual structures that, by means of ›nonsevere faults‹, present obstacles to understanding and thereby create a nonseriously stressful communication situation for the reader. In addition, the speaking entity of literature that is in this respect a humorous one functions as an appetitive social stimulus. Finally, the fifth section of the article deals with the social attractiveness of fictive characters and the speaking entities of literature. Here, a hierarchical typology of four levels of attachment is put forward; it allows well-known phenomena of quasi-social reader behaviour to be grouped together and linked to particular psychological mechanisms. The ability to recognize psychologically imbued beings can be seen as the lowest level of attachment; it can be traced back to our ability to distinguish between animate and inanimate objects, and determines what things our empathetic activity applies to in a literary text. The second level of attachment is called ›partiality‹ and refers to the recognition of a member of one's own group (of one's own community of values and the like; or, in our evolutionary past, of one's own clan). It determines who the reader's heart beats with. The familiarity with fictive people that constitutes friendship can be introduced as a third level of attachment. This familiarity explains the reader's bonds of affection towards the fiction, and, as one of their consequences, the pain of separation felt on finishing a book. The fourth level of attachment involves not so much ties of friendship as love for a particular individual. Forms of expression typical of love can be observed primarily in fan cults (for example, the Werther cult).

Empirical study of literature has primarily emerged from two sources: literary theory (especially theories of stylistics and narrative) and psychology (personality, discourse processing, emotion). In this paper I apply its insights to a study of the place of feeling in literary reading, focusing in particular on the experiences of the ordinary or »common« reader. Are feelings incidental to reading, or do they play a more fundamental role? In this essay it will be proposed that feeling is the vehicle for some distinctive and central aspects of literary response. Six aspects are discussed. First, the active and continuous nature of feelings in our thoughts and actions has been established by neuropsychologists such as Damasio. As Ellis points out, we actively seek avenues for the expression and development of our emotions. Literary reading provides an important vehicle for soliciting familiar and novel experiences of feeling. Separate from language and imagery, as Opdahl has proposed, feeling provides a third way of knowing, specifically in its anticipatory properties, its power to relate concepts across conceptual boundaries, and other processes which may be significant to literary reading. The richness of experiential meaning in response to the stylistic aspects of literature is also attributed to feeling, based on the finding that the pathways to the feeling centres in the brain convey input more rapidly than the parallel cognitive pathways. This indicates a reconstrual of the concept of appraisal that suggests it is affective; thus feeling is likely to play a central role in subsequent cognitive construals (while itself becoming open to later cognitive reconstrual). The architecture of the feeling and emotion system appears to involve three levels: bodily aspects, the experiential level, and prototypes. This model is supported by the independent theories of several other scholars in both philosophy and psychology. One of its implications is the distinction it enables between our recognition of the formal object of a feeling and a particular occasion of that feeling (a type/token distinction). It is shown that literary readers generalize the significance of feeling by moving from the token to the type. This distinction also provides a basis for considering the problem of empathy or identification with a character in a literary text, including the question how we can experience real feelings for fictional characters. Feelings and emotions have typically been considered self-referential. In the literary context, it is argued, feeling in particular evokes not only the concerns of the self but a larger sense of the self concept and how this might evolve. Feeling provides a context in which relevant issues can come to awareness. In addition, feeling provides a sense of the cultural context in which the self finds its place (and which might then be seen as open to modification). Alongside the cultural implications of feeling, however, we must also consider the intrinsic structures and processes of feeling given to us by evolution (from facial expression to hormonal responses), but especially the »laws« of emotion as proposed by Frijda, such as the »control precedence« of emotion (its capacity to take over and determine consciousness). Such laws may be challenged or opened to question by the manipulation of feeling during our experience of reading. Empirical examples of the work of feeling are considered. Response to foregrounding (the occurrence of stylistically distinctive features at the level of sound, syntax, or semantics) has been demonstrated neurologically by Hoorn: occurring in the first 400 msec, it initiates what he terms »the aesthetics of alteration «. Given that foregrounding tends to compel the reader's attention, as do various narrative aspects, it has also been shown that readers may develop strong feelings when they become absorbed in a literary text, an aspect that Green has studied under the term »transport«. In the case of a powerful text with a compelling narrative, it has been found that the origin of the text (whether based on fact or a fiction) makes no difference to the degree of transportation – a finding that Green refers to as the »hegemony of the text«. The feelings experienced by readers also include what has typically been called »empathy« or »identification with a character«. Oatley explains this as a process in which the reader simulates a character's experience, taking on her goals. It is argued that this is not required: what occurs is evocation of a feeling from our existing repertoire accompanied by a sense of the lawful nature of its processes and outcomes; thus a feeling of empathy enables us to recognize the implications of the feeling for ourselves. It is suggested that our sense of what is literary may come from engaging with feeling at the prototypical level (as we termed it earlier), which points to processes independent of our will; this may account for our sense of aesthetic disinterest. The ›truth‹ of feelings in this respect helps explain how we experience real feelings for fictional characters. The feelings that we contribute to our literary reading are in some degree responsible for our sense of the dynamic form of the literary text.

Ever since Plato people have thought that there is an especially intimate relationship between music and the emotions, but in fact there are several such relationships. In this essay I explain how music can express emotions and arouse emotions. And although strictly speaking, music cannot represent emotions, it can tell psychological stories that lend themselves to expressive interpretations. As a philosopher, my main aim is to analyze these different relationships between emotion and music, but I also illustrate my arguments with an array of musical examples. Some people have claimed that music can represent the passions. According to the Baroque doctrine of Affektenlehre , different movements of a suite or concerto should ›represent‹ distinct emotional states such as gaiety or melancholy. The emotion ›represented‹ was often a principal means of unifying the movement. Some Baroque composers also wrote ›character pieces‹ that portray different characters or temperaments, sometimes illustrating that of their friends or the notabilities of the day. But ›representation‹ in music is not strictly representation at all. A picture can identify a specific person or thing or event, but with some minor exceptions music without the aid of a title or program or the words of a song cannot do this. All it can do is present qualities, including emotion qualities such as »cheerful« and »melancholy«, that may or may not be attributed to or characterize some specific individual. In the Romantic era, it became a commonplace that music can express emotions, whether the emotions of a character or protagonist in the music or the emotions of the composer himself. Some theorists believe that musical expressiveness is a matter of the listener's experiencing music as resembling expressive human gestures such as vocal intonations and expressive movements and behavior. On this view when we say that a piece of music is expressive of sadness, we are not saying that there is anybody around who is actually expressing any sadness. It's just that the music is experienced as sounding like or moving like a person who is sad. Others believe that when we hear music as expressive of emotion, we hear or imagine an agent or persona in the music, the ›owner‹ of the states expressed. Even some ›pure‹ instrumental music – especially some music from the Romantic era – can be heard as containing a persona who is expressing emotions. My own view is that expressing emotion in music in the full Romantic sense should be thought of as in essentials very much like the expression of emotion in ordinary life: it is primarily something that a composer or a persona in the music does or achieves, rather than primarily something detected or experienced by listeners. Finally, I turn to the question of whether and how music can arouse emotions in its listeners. In Book III of The Republic , Plato argued that the musical mode known as the »Lydian« mode should be banned from the education of future governors of the state on the grounds that it makes people lascivious and lazy, whereas the Dorian mode should be encouraged because it makes people brave and virtuous. There is now ample evidence that Plato was right to think that music affects the emotions of its listeners. There are several ways in which it does this. As Peter Kivy has remarked, listeners often get pleasure from the beauty and clever craftsmanship of a well-constructed piece of music. Leonard Meyer has shown how having certain emotions is a mode of understanding certain music. Thus when listening to a piece in sonata form, we might feel anxiety at the delayed return of the tonic, bewilderment when the keys modulate further and further from the tonic and relief when finally the tonic returns. Another way in which music arouses emotions is by getting us to respond sympathetically to emotions expressed in the music by the composer, or his surrogate in the music. Finally, there is good evidence that music arouses emotions and moods in a more direct bodily way as well, influencing the autonomic system and the motor activity of listeners. These various mechanisms of emotional arousal often function simultaneously so as to produce powerful, complex, ambiguous emotional states.

Frank Kelleter uses ›neo-naturalism‹ as a cover term for approaches belonging to cognitive science, evolutionary biology, and neurophysiology, as well as for the efforts of Empirische Literaturwissenschaft (empirical study of literature) – in other words, just about everything that treats the methods and insights of the empirical sciences as a source of anthropological knowledge. This diverse mixture is rounded off with a number of quasi-religious and quasi-philosophical statements by public thinkers, specifically Edward O. Wilson and Steven Weinberg. In this way, Kelleter constructs a compact scientistic ideology that asserts total authority and competes with the authority asserted by the culturalist paradigm. The question of usefulness on a more down-to-earth level is thereby reduced to secondary importance from the beginning. Nonetheless, Kelleter does take the time to demonstrate in detail why the positions he criticizes cannot be supported. Now, it is not particularly difficult to find points to criticize in such a jumble of positions, and I am able to agree with Kelleter entirely on a number of issues. Kelleter's criticism, however, is exclusively destructive. In his portrayal, the things discovered by the neo-naturalists have either long since been known, or are trivial or uninteresting. For the space of a paragraph (and in his abstract), he does acknowledge that neo-naturalist tendencies could help to offset obscurantism and excessive cultural relativism. But nothing is likely to come of that if neo-naturalism is really as flawed as Kelleter portrays it to be. The corrections I provide, of some misleading details in Kelleter's depiction, do not need to be summarized here. In general, a ›not … but …‹ strategy predominates, suggesting that it is necessary to decide between two mutually exclusive claims: not biological conditions but culture. What is lost here is the fact that the very interaction between the two factors, as well as the cultural uses, modifications, and regulations of ›natural‹ givens, could be of particular value and relevance. Accordingly, one is struck again and again by the fact that Kelleter does not consider the lines of enquiry pursued by the positions he reviews on their own terms, but instead declares them to be of no interest or asks questions of them that they were never meant to answer. In material terms, it is regrettable that his coverage of recent biological positions is confined to the case of sociobiology, not that of Evolutionary Psychology. Providing a summary of several approaches based on the empirical sciences would have been eminently sensible here. Evolutionary Psychology is no longer fixated on analogies and homologies with the animal kingdom, but seeks in particular to explain the peculiarities of human cognition on an evolutionary basis. This, moreover, means that it is able to provide the cognitive sciences, which at times place rather too much emphasis on concepts, with an important supplemental form of empirically based cross bearing. Similarly, functional aspects can be added to enrich neurophysiology, and a historical dimension could be added to empirical psychology. Kelleter's objections, though, are of such a fundamental nature that it is of no further importance how accurate or up to date the material details are. He adopts an asymmetric scholarly dualism in which any challenge to hermeneutic scholarship by the empirical sciences is rejected. The distinction formulated in his concept of ›two natures‹ does no more than give him the opportunity to put the first aside. This places him firmly in the tradition of healthy human reason refined by education, whose thought begins with an intuition of life as a whole and therefore perceives scientistic tendencies as a threat to human wholeness. Committing a naturalistic fallacy, Kelleter even suspects that the attempt of cognitive science to study the reading processes of the general public is an attack on his own reading practices. To modernize his dualistic position and make it possible to accuse neo-naturalism of making a fundamental logical error, Kelleter introduces Bennett and Hacker as expert witnesses. They have set out a fundamental criticism of popular neuroscience, but have also made dogma of their own position as an a priori one and thereby immunized themselves against criticism coming from the empirical sciences. Criticism that leads down a one-way street in this manner is not acceptable and fails, at the latest, when the attempt is made to extend the status of unassailability to traditional hermeneutics. The aversion to scientistic and biologistic approaches in cultural studies is probably due primarily to the fear that a position of scientism might replace the hermeneutic one based on everyday language as a source of orientation in the world. Occasionally, philosophizing natural scientists continue to stir up these fears by declaring such hopes, which are really no more than science fiction. This, though, is a case of mistaken (self-)impressions. In practical life, hermeneutic knowledge based on everyday language will retain its primacy for as long as people are born, live, love, and die. The findings of the empirical sciences can only ever provide partial illuminations and corrections for our otherwise ›wholistic‹ real-world orientation (and reading). It does not in any way need to be protected by walls and made resistant to criticism; instead, it should be conceived of as eager for knowledge and open to correction. Someone who takes concepts such as reflection, responsibility, and the like as a source of orientation should be the last to condemn the systematic study of the dependencies and consequences of our behaviour. Understanding the first nature is an utterly essential part of the second.