Author Reply: The “Social” Is Not Merely Another Level of Reality

At the end of his commentary, Richard Shweder calls for more cross-cultural data to inform our theories of emotion: “One does look forward … to tasting (or at least reading) more on-line … empirical reports from diverse cultural settings. I for one will wait for the tasting; and only then draw strong conclusions about the nature and distribution of the ‘emotions.’” (Shweder, 2014, p. 323). I was struck by this call for “tasting,” because this is exactly how I started my career as a cultural psychologist interested in emotions.

For the first 15 years of my research career, I “tasted” emotional experiences in different cultures (Mesquita, 1993, 2001; Mesquita & Karasawa, 2002), and I catalogued the many ways in which these experiences were different, based on both my own and other people’s research (Mesquita & Frijda, 1992; Mesquita, Frijda, & Scherer, 1997). The work showed that it is certainly possible to describe cultural differences in emotions in terms of their components: In questionnaire research we found cross-cultural variations in terms of the appraisal, action readiness, behavior, and long-lasting cognitive consequences that constitute certain emotions.

The different lower level patterns across cultures gave us little reason to assume that emotions were essences (see Zachar, 2014). But I have come to take nonessentialism one step further, abandoning the idea of emotions as states, and focusing instead on emotions as processes that develop in context. Granted, it is possible to describe the emotions of a person at one point in time and out-of-context, just as it is possible for a biologist to describe the anatomy or the state of a plant without considering its environment, but we miss some indispensible information. No biologist would describe the leaves of the plant as brown, without relating this to the fact that the soil is water-deficient. In contrast, many emotion psychologists seem satisfied describing the phenomenon of interest out of context (see, Mesquita, 2010): The focus of much of the field’s theorizing is on emotions as static entities, rather than on emoting as a contextualized process.

When I started my cross-cultural research, I asked large groups of participants from each culture to freely describe emotional episodes (to interviewers with the same cultural backgrounds). The goal was to supplement the existing appraisal and action readiness questionnaires with culture-specific dimensions (see Mesquita, 2001). However, it gave me the unanticipated advantage of looking at hundreds of emotion narratives in Dutch, Turkish-Dutch, and Surinamese-Dutch immigrants, European Americans, Mexican Americans, and Japanese. Appraisal (or componential) theory could be meaningfully imposed on the data, and yielded interesting cultural differences in the constituting components of emotions.

However, these interviews told me a story of their own, which moved well beyond the variations in appraisal and action readiness. First, the interviews were not so much stories about inner feelings, not even stories of individuals’ perceptions of the world, but to a large extent, they were the stories of individuals relating to their social contexts. For the most part, they reflected the shifting position of the individual in a dyadic relationship, or in a larger social context; emotions in many cases mirrored either shifts in power or status (see Clay-Warner, 2014), or shifts with regard to other social dimensions such as closeness or warmth (Wiggins, 1979). The second story that these interviews told us was that emotions develop over time, and in connection with events in the environment.

I was trained, to use Moors’ phrasing (2014), as a second flavor appraisal theorist (having completed my training with Nico Frijda). From this vantage point, it is easy to ignore the story that the narratives told us. We did everything to examine the data only from within the restrictions of our theory, but what the stories really told us was something outside the scope of our appraisal theory: emotions are part of social interactions and events, and they develop over time.

For example, we asked North American and Japanese participants to describe an event in which someone else—in this case, someone not intimate—had offended them. In the typical American scenario for offense by a nonintimate other, the event was part of a buildup. But even if we restricted ourselves to the definitive blow, participants’ appraisals and action readiness were spun out over a longer period of time. Participants reported how they tried to gain approval from others for their own view that the offender was wrong or even a bad person. This approval then fanned their righteous indignation, and others’ confirmation also helped to redeem their self-esteem, and justified their aggression.

The matching Japanese scenario read very differently: Many Japanese participants wondered what they themselves could have done to prevent the offending situation, or how they had given reason for it; they tried to understand the situation from the point of view of the offender, and they typically had not shared their feelings with other people. Therefore, they did not gain social approval for their anger, and they ended up relativizing how much they had been harmed. Their actions differed as well: The most common response was “doing nothing.”

Data like these are complex, in part because they force us to abandon the sort of essentialism on which our science is based (Barrett, 2013). As a consequence, psychologists are not fond of these data, as I have found out the hard way (Mesquita et al., 2007 [unpublished manuscript]). Reviewers have wanted to know if the American and the Japanese appraisal and action readiness modes respectively, are attributes of “the same thing”: Do Americans and Japanese experience the same anger? The question is the remnant of a deep-seated belief in essentialism, the belief that there is a thing deep down—the emotion of anger—and that we are looking for its features. Therefore, if we find different features in different cultures, the question becomes: did they really pertain to the same thing?

Our sociodynamic model of emotions (Mesquita & Boiger, 2014) suggests that emotions are not essences deep down, but rather emerge from interactions with the social world. Different from Agnes Moors’ suggestion (2014, p. 304), our model neither “eschews talk of mental processes” nor “de-emphasizes the role of appraisal.” Rather it places mental processes, appraisal included, in the context of the social interactions in which they occur. Placing emotions in their social context is like describing the air and the soil of a plant: It is not merely a different level of description, but rather it adds information that renders the quality of the emotion understandable and predictable. It makes it possible to not just “taste” emotions in different cultures, but to understand just how different tastes were achieved in the course of individuals’ interactions with their social worlds.