Gervais & Fessler (G&F) have laid out a novel analysis of social emotions, which is sure to stimulate fruitful discussion and future research. Nevertheless, I believe that their model suffers from some conceptual ambiguity, which, if clarified, would give the analysis greater predictive power and fidelity.

The authors posit that low-value individuals are likely to become targets of contempt because they fail to elicit respect. This characterization neglects that value may be high or low along orthogonal dimensions and that these dimensions interact to predict specific emotions. The Stereotype Content Model (SCM) (Fiske et al. Reference Fiske, Cuddy, Glick and Xu2002; Reference Fiske, Cuddy and Glick2007) organizes beliefs about social and cultural groups along two fundamental dimensions of social cognition: warmth and competence. Warmth is attributed to groups that are generally seen as cooperative rather than competitive, whereas attributions of competence are reserved for high-status relative to low-status groups. Crossing the warmth and competence dimensions yields four broad classes of stereotypes and predicts specific corresponding emotions. Groups that are seen as both cooperative and high-status – and therefore warm and competent (e.g., Christians, middle class) – elicit emotions like pride and admiration. Groups that are seen as competitive (in the sense that they are free-riders) and low-status stimulate disgust and scorn (e.g., drug addicts, welfare recipients). The two “off-diagonal” classes provoke ambivalent emotions: cooperative but low-status groups (e.g., elderly, disabled) are seen as warm but incompetent and therefore trigger pity. Competitive but admittedly high-status groups (e.g., rich people, Asians, and other model minorities) activate emotions like envy.

In the SCM framework, groups may be imbued with relational value (based on their cooperation), agentic value (based on their status), neither, or both. Each dimension is associated with distinct suites of behaviors (Cuddy et al. Reference Cuddy, Fiske and Glick2007). Relational value predicts active behaviors. Cooperative groups receive active assistance (e.g., prosocial behavior, charity), whereas competitive groups receive active harm (e.g., harassment, political scapegoating). Agentic value, on the other hand, predicts passive behaviors. High-status groups receive passive facilitation (e.g., realpolitik cooperation) and low-status groups receive passive harm (e.g., neglect, limited access to education, housing). Each class of social groups receives a combination of these active and passive behaviors (e.g., elderly people receive charity, but are also neglected). Rather than treat them as orthogonal sources of value, G&F discuss “value” and “efficacy” as substitutes for one another.

Collapsing across warmth and competence as distinct value dimensions diminishes the predictive power of G&F's model. Furthermore, it considerably expands the swath of people who are likely to become targets of contempt. In the contempt-as-sentiment framework, all social targets with the exception of high-warmth, high-competence targets are likely targets of contempt in either its cool or hot form because they are devalued on at least one dimension.

What determines whether a sentiment – in this case contempt – runs hot or cold? And when it runs hot, what determines whether contempt prompts approach or avoidance behaviors? G&F reference many likely moderators; the hot form of contempt may be invigorated by frequency of contact, stability of the social hierarchy, and network size, among many other factors. The authors ultimately conclude that the emotions, phenomenology, and behavior that result from the contempt sentiment will depend on “any variable that influences the perceived costs and benefits of social relationships” (sect. 6.1, para. 7).

Contrast this formulation with the systematic principles approach taken by the SCM, which begins by identifying the functional relations – competitive, cooperative – between parties to make predictions about downstream emotions and behaviors. Active, “hot” intergroup emotions are amplified when there is a zero-sum relationship between two parties' goals (Fiske & Ruscher Reference Fiske, Ruscher, Mackie and Hamilton1993; Struch & Schwartz Reference Struch and Schwartz1989). Even in the absence of overt competition, the mere perception that a group poses a resource or value threat engenders negative emotions, intergroup conflict, and outgroup derogation (Deutsch Reference Deutsch, Deutsch, Coleman and Marcus2006; Johnson & Johnson Reference Johnson and Johnson1989; Mackie et al. Reference Mackie, Devos and Smith2000; Sherif et al. Reference Sherif, Harvey, White, Hood and Sherif1954/1961). However, as described above, the assessment of a group as friend or foe necessarily intersects with the assessment of their ability to enact their intentions (i.e., their status), good or ill.

Because the sentiment framework relies on a single index of value, it clusters anger, an approach emotion, together with disgust, an avoidance emotion, to produce the “hot” phenomenology associated with contempt. However, using both dimensions from the SCM allows us to differentiate when the hostile emotions route is likely to lead to anger and attack (i.e., in response to low-warmth, high-competence targets; Glick Reference Glick, Newman and Erber2002) versus disgust and exclusion (i.e., in response to low-warmth, low-competence targets). In line with these predictions, we find that participants report they would be most likely to harm low-warmth, high-competence targets (Cikara & Fiske Reference Cikara and Fiske2011), but most likely to say it is acceptable to sacrifice low-warmth, low-competence targets (Cikara et al. Reference Cikara, Farnsworth, Harris and Fiske2010).

Another strength of examining the orthogonal effects of competitiveness and status, as opposed to a unitary value, is that it allows for predictions about how emotional responses towards groups will change as attributions of warmth and competence change. For example, we find that participants are most likely to exhibit pleasure in response to competitive, high-status (relative to other) targets' misfortunes (for a review, see Cikara & Fiske Reference Cikara and Fiske2013). However, providing participants with counter-stereotypic information that decreases these targets' status or increases perceptions of their cooperation significantly reduces participants' Schadenfreude (Cikara & Fiske Reference Cikara and Fiske2012), indicating that it is these dimensions, as opposed to anything intrinsic to the groups themselves, that drives counter-empathic responses.

G&F's broader contribution to the social emotions literature is undeniable. The Attitude-Scenario-Emotion (ASE) model of sentiments integrates seemingly irreconcilable views of contempt and other social emotions. Formally integrating distinct dimensions of social value would deepen the impact of this approach by facilitating even more specific predictions regarding when and why contempt arises, and how it manifests.