Towards a Theory of Emotional Communication

In their paper, "Towards a Theory of Emotional Communication," Anne Bartsch and Susanne Hübner outline a model of emotional communication where emotional communication is conceptualized as a process of mutual influence between the emotions of communication partners. To elaborate this general notion further, four working definitions of emotional communication are introduced, each of which is based on a different theory of emotions. In the second part of the paper, an integrative framework is proposed that reconciles the four working definitions and their underlying theories of emotion. According to this framework, emotional communication comprises three interrelated levels of complexity: 1) innate stimulus-response-patterns, 2) associative schemata, and 3) symbolic meaning. Finally, Bartsch and Hübner discuss how emotional communication can be described in terms of general communication theory, and conclude that the three complexity levels are heterogeneous with regard to definitional issues in general communication theory. Hence, emotional communication cannot be subsumed under a single theory of communication. Taken separately, however, each complexity level of emotional communication can be related meaningfully to approaches in general communication theory. Anne Bartsch and Susanne Hübner, "Towards a Theory of Emotional Communication" page 2 of 8 CLCWeb: Comparative Literature and Culture 7.4 (2005): Anne BARTSCH and Susanne HÜBNER Towards a Theory of Emotional Communication Over the past two decades, emotions have become a rapidly growing field of research and theorizing in disciplines as different as cognitive, social, and developmental psychology, anthropology, linguistics, neuroscience, philosophy, and others. One of the central themes that emerged from the joint efforts of emotion research is that emotions are not just private inner experiences, but inherently social and communicative phenomena. In this paper, we explore the following: If emotions are inherently social and communicative, is it also true that communication is inherently emotional? To date, communication has mainly been conceptualized as a process of exchange, mutual influence, co-orientation, normative control, etc., of cognitive information processing. But what if we think of communication as a process of exchange, mutual influence, co-orientation, and normative control of emotions? We do not propose, however, that this view with focus on emotion would exclude the cognitive approach; rather, we explore the role of emotions and their specific contribution to the process of communication. Further, we outline a model of emotional communication where the objective of the model is to integrate recent developments in emotion research into a communication-theoretical framework (see Bartsch and Hübner). What exactly do we mean by "emotional communication"? The basic idea of the theoretical framework we propose is that people do not only communicate in order to exchange information, but also in order to exchange emotions. In other words, emotional communication is conceptualized as a process of mutual influence between the emotions of communication partners. To elaborate this general concept of emotional communication further, we propose four working definitions, each based on a different theory of emotion. The first working definition is based on neuroscience models, the second on appraisal theories, the third is based on the prototype approach, and the fourth on social constructivist theories of emotion. The four emotion theories mentioned are chosen because each provides us with an explicit and detailed account of the processes that give rise to emotions. For, if we know the processes that give rise to emotions, we may be able to explain how these processes are influenced by similar processes in other people. Neuroscience models of emotion assume that emotions are caused by specialized brain systems (see, e.g., LeDoux; Panksepp; Damasio). Basically, emotional brain systems have two functions: First, they analyze the emotional meaning of stimuli. This is accomplished by associative paring of stimulus features with innate emotion elicitors. Interestingly, emotion expressions of others seem to count among these innate emotion elicitors. Emotional brain systems are involved in the processing of expression of others critically. Second, emotional brain systems control a variety of emotional responses such as hormone release, activation of the autonomic nervous system, vocal, facial, and motor expression, allocation of cognitive resources to the situation that elicits emotion, etc. Taken together, these functional properties of emotional brain systems lead to the following conclusion: If the same brain systems that give rise to emotions are also involved in the expression of emotions and in the processing of expressions of others, emotion expression will then lead to an activation of similar brain systems in communication partners. Thus, on the basis of neuroscience models of emotion, emotional communication can be defined as a process of reciprocal activation of emotional brain systems. Appraisal theories of emotion describe the processes that give rise to emotions at a quite different level of analysis. According to appraisal theories emotions are elicited by cognitive appraisals (see, e.g., Lazarus; Scherer; Smith and Ellswort; Ortony, Clore, and Collins; Frijda). Appraisal criteria relevant to emotion include novelty, valence, goal conductiveness, certainty vs. uncertainty of outcomes, self vs. other's agency, coping potential, and normative evaluation. Each emotion is caused by a specific pattern of appraisals. Emotion expression is, therefore, assumed to carry detailed information about cognitive appraisals. Some appraisal theorists even believe that emotion expressions are composed of several muscle movements, each of which signals a different appraisal component. For example, according to Klaus R. Scherer, the appraisal information conveyed by emotion expression can be used by communication partners in two ways: The appraisals of others can be used to resolve ambiguities in one's own Anne Bartsch and Susanne Hübner, "Towards a Theory of Emotional Communication" page 3 of 8 CLCWeb: Comparative Literature and Culture 7.4 (2005): appraisal of the situation and appraisals can be used to infer the motives and the likely behavior of others (which are often relevant to one's own appraisal of the situation). To summarize, emotion expression carries appraisal information that is likely to affect the appraisals of others, and consequently does affect their emotions. Thus, on the basis of appraisal theories, emotional communication can be defined as a process of information exchange about cognitive appraisals. The prototype approach was developed originally as a theory of natural language concepts and it was applied to emotions by Shaver and his colleagues (see Shaver, Schwartz, Kirson, O'Connor; Fischer, Shaver, Carnochan). The prototype approach assumes that knowledge about emotions is represented in the form of nonverbal emotion scripts. Emotional scripts include knowledge about typical eliciting situations, typical reactions, and self-control procedures. These emotion prototypes have two closely related functions: They structure the personal experience of emotions and they are used to understand the emotions of others. This double function of emotional scripts in emotion elicitation and emotion perception leads to the following conclusion: If the same knowledge structures are activated when experiencing an emotion and when observing emotions in others, the vicarious experience of an emotion must then have essential features that correspond to the personal experience of an emotion. The person who experiences an emotion vicariously may be aware that the emotion actually "belongs" to someone else, but nevertheless she/he will experience an emotion. Thus, on the basis of the prototype approach, emotional communication can be defined as a process of reciprocal activation of emotional scripts. Next, social constructivist theories of emotion also assume that emotion concepts play a causal role in emotion elicitation (see, e.g., Averill; Harré; Shweder; Lutz; on [radical] constructivism, see Riegler ). Social constructivists, however, believe that emotional knowledge is represented in a purely symbolic manner. According to this view, the meaning of emotion words is constituted by a set of rules that specify the kinds of persons, situations, and actions to which the emotion word applies. The rules that govern the use of emotion words and other symbolic expressions are thought to be equivalent to the social norms concerning emotions. In turn, this means that if a person finds herself/himself in a situation to which a specific emotion word applies, this person has both the moral right and the moral obligation to experience the emotion and to behave accordingly. Therefore, emotions can be considered as social roles (see, e.g., Averill). As with social roles in general, the individual is not free to self-ascribe emotions at will. The self-ascription of emotions must not only rely on culturally determined emotion rules and in order to become a fact of socially shared reality, emotional roles must be authorized and reciprocated by complementary role behavior of others. The implications of this theory for the concept of emotional communication are straight forward: If emotions do exist only as a fact of socially shared reality, the existence of an emotion in the heads of other people is then as constitutive to the emotion as its existence in the head of the person having the emotion. Thus, on the basis of social constructivist theories of emotion, emotional communication can be defined as a process of symbolic negotiation of emotions. To summarize, the four emotion theories introduced above clearly support the assumption that the emotions of communication partners influence each other. This is quite remarkable given that these theories describe the processes that give rise to emotions in quite different terms. Appraisal theories consider the emotion eliciting process as a form of cognition, whereas neuroscience models claim that emotion and cognition are distinct forms of information processing. The prototype approach assumes that knowledge about emotions is represented in the form of nonverbal prototypes, whereas social constructivist theories claim that emotion concepts are symbolic representations. Neuroscience models describe emotions as bodily phenomena, whereas social constructivist theories consider them as purely mental constructs, etc. On the one hand, there are four good reasons to believe that the emotions of communication partners influence each other. On the other hand, there are multiple reasons to doubt whether the four emotion theories mean the same thing when they speak of emotions. If the concepts of emotion implied in these theories cannot be reconciled, it would be misleading to apply the same term "emotional communication" to all of the working definitions derived there from. We are convinced, however, that this is not the case. For, in spite of the fact that there are substantial controversies between the proponents of strong verAnne Bartsch and Susanne Hübner, "Towards a Theory of Emotional Communication" page 4 of 8 CLCWeb: Comparative Literature and Culture 7.4 (2005): sions of the four emotion theories, there are also moderate and intermediate positions trying to reconcile them. It is beyond the scope of this study to discuss in detail why we believe that the strong versions are wrong and why we propose hat the conciliatory versions are right (on this, see Bartsch and Hübner). Basically, the answer depends on the position one takes with respect to the Zajonc-Lazarus debate about the primacy of affect over cognition, or vice versa, the imagery debate, the nature-nurture debate, and the mind-body problem. As this impressive list of unresolved controversies shows clearly enough, our position, favoring a unified theory of emotion would be open to criticism in multiple ways. If we suppose, however, that the emotion theories considered describe important aspects of the same phenomenon, then a comprehensive theory of emotional communication would have to take into account the entire range of processes specified by the four working definitions. One important step towards an integrated theory of emotional communication is the assumption that different levels of cognitive complexity are involved in emotion elicitation. Such hierarchies of complexity levels are proposed by scholars across the whole field of emotion theory and can therefore be seen as a common denominator. The three levels of cognitive complexity that are mentioned most often are innate stimulus-response patterns, associative schemata, and symbolic concepts. If we assume that at least three levels of cognitive complexity are involved in emotion, the next question we need to ask is how are the different levels interacting? Most theories are not very specific with regard to this question and it is often assumed that these different levels work in parallel and contribute to the final product of emotion in an additive manner. In our opinion, the most elaborate version of an emotional processing hierarchy is proposed by Kurt Fischer, Philipp R. Shaver, and Peter Carnochan in their paper published in 1990, "How Emotions Develop and How They Organize Development." According to Fischer, Shaver, and Carnochan, the different levels of cognitive complexity emerge in a fixed sequence during child development. The most basic level consists of innate stimulus-response patterns. This level is present at birth, whereas the levels of associative schemata and symbolic concepts emerge at about four months and two years, respectively. A central assumption of the model is that higher levels of cognitive complexity such as associative schemata and symbolic concepts are composed of simple cognitive skills on the more basic levels. The emergence of superordinate skills leads to a reorganization and differentiation of the basic skills. Nevertheless, the skills on the subordinate levels continue to function independently of the newly acquired level of cognitive complexity. Thus the subordinate levels can work without the superordinate levels, but not vice versa. How, then, can the four working definitions be located within this hierarchy of cognitive skills? At least three of the emotion theories considered have clear affinities to one of the complexity levels. Neuroscience models are primarily about innate stimulus-response patterns, the prototype approach applies to associative schemata, and social constructivist theories deal with symbolic emotion concepts. The only approaches that cannot be easily identified with any of these levels are appraisal theories where the process of appraisal is basically the same for all three levels of cognitive complexity (see Lazarus; Leventhal and Scherer; Smith and Kirby). The situation is appraised with respect to the same set of appraisal criteria and the same patterns of appraisal lead to the same emotional reactions. Thus, the working definition based on appraisal theories would apply equally to all three levels of emotional communication. Taken together, the argumentation considered here leads to a model of emotional communication that comprises three interrelated levels of complexity. Level one consists of the reciprocal activation of emotional brain systems, level two consists of the reciprocal activation of emotional scripts, and level three consists of the symbolic negotiation of emotions. In addition, all three levels can be characterized as a process of information exchange about cognitive appraisals. Further, as a consequence of the interrelated architecture of cognitive complexity levels, our model posits that the emotional exchange processes specified by the four working definitions contribute to a unified process of communication. As already mentioned, the model of Fischer, Shaver and Carnochan assumes that processing skills on the superordinate levels are composed of simple processing skills on the subordinate levels. Thus, the superordinate levels of emotional communication involve automatically communication on the more basic levels. This means, for instance, that emotional communication on a symbolic level will not work if it does not activate nonverbal emotion Anne Bartsch and Susanne Hübner, "Towards a Theory of Emotional Communication" page 5 of 8 CLCWeb: Comparative Literature and Culture 7.4 (2005): scripts, which, in turn, depends on the activation of emotional brain systems. However, despite its dependence on basic levels, each of the higher levels of emotional communication does add new constraints and new degrees of freedom to communication processes on its subordinate levels. Thus, it can be hypothesized that an important function of the higher levels is to regulate emotional communication on the more basic levels. Another implication of the interrelated organization of cognitive complexity levels is that the model allows for communication processes that cross level boundaries. In other words, the level of cognitive complexity that is used to express an emotion can differ from the level of cognitive complexity that is used to interpret the expression. This is possible because the cognitive structures that underlie the four levels are closely intertwined. As a consequence, typical aspects of emotion expressions on one level often coincide with typical aspects of emotion expressions on other levels. Such coincidences make it possible that emotion expressions are reframed by communication partners on a different level of cognitive complexity. For example, innate emotion expressions can be interpreted as a symbolic act of role assertion if the expression is associated with a conventionalized social meaning in a given culture. On the other hand, symbolic expressions can trigger innate response tendencies if they mimic typical stimulus features of innate emotion elicitors. This seems to be the case, for instance, with music. Studies of music perception find that the emotional impact of music is predicted by the same acoustic features that characterize the vocal expression of emotions (see, e.g., Juslin and Sloboda). To summarize, our model of emotional communication comprises three levels of complexity. Level one consists in the reciprocal activation of emotional brain systems. Level two consists in the reciprocal activation of emotional scripts. Level three consists in the symbolic negotiation of emotions. The three levels are interrelated in such a way that the higher levels of emotional communication automatically involve communication processes on the more basic levels. We propose the hypothesis that an important function of the superordinate levels is to regulate emotional communication on the subordinate levels. Such a model allows for communication processes that cross level boundaries, because typical aspects of emotion expressions on one level often coincide with typical aspects of emotion expression on other levels. However, one question remains: In how far is it justified to consider the emotional exchange processes described by the model as a form of communication? A wide range of general communication theories are based on concepts referring to symbolic forms of exchanging messages. To mention some examples, symbolic communication theories include: Rational Argumentation Theory (Cox and Willard), Symbolic Convergence Theory (Bormann; Bormann, Cragan, Shields), Narrative Paradigm Theory (Fischer), and in the domain of social psychology Symbolic Interactionism (Mead). Of course, each theory is based on specific concepts like rational arguments, fantasy themes, stories, and meanings, but all theories have in common that they describe communication as a symbolic message transfer between sender and receiver (symbolic, in so far as all theories mentioned include the application of symbol systems like language, descriptive gestures, reference objects, etc.). What does this mean for our model of emotional communication? The highest complexity level of the model -level three -clearly meets the criteria proposed by symbolic communication theories, such as symbol systems, conventionalized meanings, etc. To give you an example, Symbolic Interactionism in the tradition of George Mead can be applied in a meaningful way to the symbolic negotiation of emotions on level three. The theoretical starting point of Symbolic Interactionism is that no kind of meaning exists outside of the interaction between people. In other words, meaning is considered as a shared response to each term, which in the course of interaction has achieved the status of a significant symbol and the meaning of which is shared by a community. With respect to emotional communication, this means, that not the emotional expression or the understanding is negotiated in the communication process but the meaning of emotion within the social context. To summarize, the mainstream of general communication theories based on criteria such as symbolic message exchange and rationality can be applied to level three of our model. Levels one and two, however, refer to non-symbolic forms of message exchange. The number of communication theories that deal with such non-symbolic forms of communication is significantly smaller. However, there is a series of approaches that describe nonverbal behavior and its function in Anne Bartsch and Susanne Hübner, "Towards a Theory of Emotional Communication" page 6 of 8 CLCWeb: Comparative Literature and Culture 7.4 (2005): communication process such as Proxemics, Kinesics, Paralanguage, or Vocalics, and similar theories such as Interpersonal Deception Theory (Buller and Burgoon), Expectancy Violation Theory (Burgoon), Marital Communication Theory (Fitzpatrick), pragmatic communication theories in the tradition of Paul Watzlawick, etc. Generally speaking, these theories can be subsumed under the term of nonverbal interaction theories. To summarize the common theoretical statements, all of these theories are based on the assumption that an adequate understanding of message transfer in social contexts requires a consideration of nonverbal cues as an information channel to convey knowledge about each other. Nonverbal interaction theories can be applied to the second level of our model, which is the level of nonverbal emotion scripts. On the basis of emotional knowledge which is organized in script-like structures with a beginning and an end, communication partners develop expectations about the course of interaction. Moreover, the chronological structure of scripts enables the participants to shape intentions to regulate this interaction. It is important to note that criteria such as reciprocal expectations and intentions, which play a central role in nonverbal interaction theories, do not necessarily mean self-conscious expectations and intentions. Rather, expectations and intentions are based on implicit characteristics of interaction such as the temporal unfolding of interaction scripts. Finally, let us turn to the neurological level. Only a few approaches try to explain the biological foundations of human communication. Among these theories are: The Autopoietic Systems Theory of Humberto R. Maturana and Francisco J. Varela, the Communibiology Approach of Michal J. Beatty and James C. McCroskey, and the Theory of Spontaneous Communication of Ross Buck. The theory of Buck is the most explicit with regard to the influence of genes on the reciprocal activation of emotional brain systems. Similar to all evolutionary approaches, Buck assumes that social communication has an evolved basis in innate displays and pre-attunements, which are determined genetically and serve the co-orientation of attention. According to the communicative gene hypothesis of Ross Buck and Benson E. Ginsburg, not only the emotional display is innate but also the skill to receive emotional signals. Since, from an evolutionary perspective, innate expression skills appear reasonable, only if there exist also innate receiving skills. This leads to the conclusion that there are not only selfish genes, but also social genes that enable communicative co-orientation. To conclude, there is no single communication theory that covers the entire bandwidth of emotional communication processes specified by the four working definitions. The widespread concept of communication as a symbolic message transfer can be applied only to level three. The description of emotional communication at the more basic levels requires a concept of communication which includes innate co-orientation processes and reciprocal formation of expectations based on emotional scripts. That means that the phenomenon of emotional communication is quite heterogeneous with regard to definitional issues in general communication theory. Hence it cannot be treated within a single theory of communication. Rather, there are interesting parallels in the controversies about emotion and the controversies about communication. These parallels make emotional communication an especially promising subject for communication theory and research. Note: The above paper is an updated version of Anne Bartsch and Susanne Hübner, "Die Macht der Emotionen. Zur aufmerksamlitenden Funktion von Emotionen in medialen Dipositiven" in SPIEL: Siegener Periodicum zur Internationalen Empirischen Literaturwissenschaft 22.1 (2003): 38-48.