The Role of Reduced Humanity in Producing Linguistic Discrimination

Abstract

This article addresses the role of perceived (reduced) humanity and group membership of others in producing linguistic discrimination. Study 1 assessed the effects of these factors on a subtle measure of linguistic discrimination, namely, linguistic abstraction. Study 2 considered the explicit level of verbal abuse. Results highlighted that target’s reduced humanity led to enhanced linguistic discrimination toward the target, while group membership moderated this effect in specific conditions. Overall, the evidence of this set of studies sheds light on the role of humanity and its interplay with social categorization on discrimination outcomes.

Keywords

reduced humanity linguistic discrimination linguistic abstraction verbal insults

Social psychologists have dealt with the origins of social discrimination and prejudice since the very beginning of their science (e.g., Allport, 1954). A large number of empirical studies have consistently shown that dichotomous ingroup/outgroup categorization is a fundamental factor in eliciting intergroup bias (Tajfel & Turner, 1979). By contrast, less is known on the role of humanity as an antecedent of social discrimination. Recently, a considerable amount of research has focused on the phenomenon of dehumanization, that is, the tendency to deny others’ full humanity considering them as less than human (for reviews, see Haslam, Loughnan, Kashima, & Bain, 2008; Vaes, Leyens, Paladino, & Miranda, 2012). Dehumanization, besides some exceptions (Costello & Hodson, 2010, 2011; Loughnan, Haslam, & Kashima, 2009; Pereira, Vala, & Leyens, 2009; Vaes, Paladino, Castelli, Leyens, & Giovannazzi, 2003; Vaes, Paladino, & Leyens, 2004, 2006; Vaes, Paladino, & Puvia, 2011; Wohl, Hornsey, & Bennett, 2012), has predominantly been considered as an outcome variable, rather than a factor affecting intergroup relations. However, as several scholars underlined (e.g., Opotow, 1990; Staub, 1989), the denial of others’ humanity often served as a legitimizing principle of aggression and extermination of entire social groups. In this vein, Tajfel (1982) argued that “the ‘marking off’ from the rest of population” (p. 486) of Jews in Nazi’s Germany was rather “a marking off from their definition as being human, and all that this entails” (p. 486, emphasis added), being “the extreme of the process of deprivation of individuality, of the complete definition of human beings through categories to which they belong” (p. 487, italics in the original), thus stressing an interplay between humanity and social categorizations in accounting for such forms of discrimination. Fortunately, considering others as less human than ourselves does not always produce overt and dramatic consequences, but it may lead to widespread forms of social discrimination such as those conveyed by language, that is, the primary means through which discrimination is communicated and maintained (Maass, 1999).

In the light of the above considerations, in this article we addressed the consequences of being described with reduced humanity. We also addressed the role of group membership and its interplay with (reduced) humanity. Specifically, we examined whether portraying individuals as characterized by reduced humanity leads to biased representations of them in language use. Study 1 assessed the effects of these factors at the implicit level of linguistic discrimination in terms of linguistic abstraction (Maass, 1999; Semin & Fiedler, 1988). Study 2 focused on the explicit level of verbal abuse (Semin & Rubini, 1990).

Denial of Others’ Humanity

Leyens et al. (2000) highlighted an implicit bias through which people attribute less human essence to outgroup members than to ingroupers as a function of ingroup/outgroup categorization. This tendency, labeled infrahumanization, has been mainly studied in terms of attribution of uniquely human (UH) secondary emotions (for a review, see Vaes et al., 2012). Haslam (2006) contended that dehumanization results from both denial of UH characteristics (e.g., secondary emotions, Leyens et al., 2000; human traits, Haslam et al., 2008; prosocial human values, Struch & Schwartz, 1989) and of human nature (HN; that is, what is intrinsically human and central to humans). The former process leads to animalization of human beings, whereas the latter implies mechanization/objectification (Haslam, 2006).

Another interesting perspective on the issue of dehumanization was highlighted by Struch and Schwartz (1989) who considered lack of prosocial human values (i.e., a UH feature distinguishing humans from animals) as an indicator of outgroup dehumanization. The authors showed that perceiving outgroup members as lacking in concern and compassion for others mediated the effects of intergroup conflict on endorsement of aggression toward the outgroup.

Effects of Dehumanization in Intergroup Relations

Vaes et al. (2003) highlighted that ingroup members expressing themselves in terms of (positive and negative) secondary emotions stimulated prosocial behavior to a higher extent than did outgroup members displaying secondary emotions. By contrast, ingroupers and outgroupers expressing primary (not UH) emotions elicited prosocial behavior to the same extent. Convergent evidence was obtained by Vaes et al. (2004) who showed that participants took the perspective of an ingroup target more easily than that of an outgrouper when ingroupers described their feelings in terms of secondary emotions. Similarly, Vaes et al. (2006) highlighted that priming UH emotions activated the human concept to a higher extent when secondary emotions were attributed to the ingroup than to the outgroup. Vaes et al. (2011) also found that objectified women were associated with less human concepts by both males and females. Loughnan et al. (2009)—based on Haslam’s (2006) model—manipulated humanness of a fictitious group and examined whether attribute-based dehumanization led to metaphor-based dehumanization and vice versa. Results showed that participants explicitly learned to dehumanize the groups and inferred the corresponding type of attribute- or metaphor-based perception.

One of the few contributions that explicitly acknowledged the role of dehumanization in the explanation of prejudice and discrimination in intergroup contexts is the work on the interspecies model of prejudice (Costello & Hodson, 2010, 2011; Hodson, MacInnis, & Costello, 2013) according to which perceiving a greater human–animal divide so that humans are regarded as different and superior to animals leads to higher prejudice toward outgroupers (e.g., immigrants or Black children). Similarly, Wohl et al. (2012) showed that intergroup forgiveness diminished when outgroupers expressed UH emotions in their apologies.

Taken together, the reviewed contributions suggest that people hold the belief that outgroupers are characterized by a reduced human essence (cf. Leyens et al., 2000; see also Yzerbyt, Rocher, & Schadron, 1997), and for this reason, once they confront the evidence that outgroupers display human features, they react to this by discriminating against them in various ways. One exception to these findings is the work by Pereira et al. (2009) who showed that people disagreed less strongly with the intention to admit Turks (i.e., outgroupers) into the European Union when they were associated with secondary than with primary emotions.

To sum up, the available evidence is rather ambiguous with reference to the interplay between humanity and group membership in affecting intergroup discrimination. In this vein, the display of secondary emotions is either taken as evidence of outgroup’s humanity (e.g., Pereira et al., 2009) or questioned and rejected as pretense against outgroup’s perceived lack of humanity (e.g., Vaes et al., 2006; Wohl et al., 2012). One could contend that, in previous studies, participants could easily reject the evidence of outgroupers’ humanity because it was based only on the display of secondary emotions. Indeed, the evidence at their disposal was rather weak and, very likely, not persuasive. Moreover, previous studies did not experimentally address the interplay between others’ humanity and group membership by manipulating both factors. In this vein, a clear-cut demonstration of the interplay of these factors is still needed.

Discrimination Through Language

One of the most pervasive and effective means of discrimination is language use (Maass, 1999). A large amount of research has provided consistent evidence on linguistic intergroup bias (LIB; for reviews, see Maass, 1999; Rubini, Menegatti, & Moscatelli, in press) as the tendency to describe socially desirable behavior of ingroupers and socially undesirable behavior of outgroupers, in relatively more abstract terms (cf. Semin & Fiedler, 1988) than socially undesirable behavior of ingroup members or socially desirable behavior of outgroup members. The implication of this biased language use is the higher generalizability across different situations and time of ingroup positive behavior and outgroup negative behavior. Recently, Rubini, Moscatelli, and Palmonari (2007) detected linguistic intergroup discrimination in the context of minimal groups (see also Moscatelli, Albarello, Prati, & Rubini, 2014; Moscatelli, Albarello, & Rubini, 2008; Moscatelli & Rubini, 2011; Rubini, Moscatelli, Albarello, & Palmonari, 2007) by analyzing the free language employed to describe ingroup and outgroup minimal behaviors such as intergroup allocations through matrices. They highlighted linguistic discrimination consisting in people’s tendency to describe ingroup’s behavior with positive terms at a higher level of abstraction, while using negative terms at a lower level of abstraction. The opposite pattern was found for descriptions of outgrouper’s behavior. Since communicators do not exert intentional control on the use of language abstraction, the LIB is conceived as an implicit measure of discrimination (Franco & Maass, 1999; Maass, 1999).

Another linguistic, but overt, means of discrimination and derogation relies on verbal insults, which—as contended by Semin and Rubini (1990)—deny what is near and dear to a person. Abusive language represents an effort to make someone despicable (Van Oudenhoven et al., 2008), and it varies to a great extent depending on the content that epithets convey, for instance, denial of some core features of the person (e.g., intelligence, physical ability) or exacerbation of other aspects of human behavior (e.g., sexual behavior). Very few empirical studies have addressed verbal insults as a means of discrimination. Semin and Rubini (1990) examined the cultural construction of the concept of person in different contexts through verbal insults. Their work was based on E. R. Leach’s (1964) contention that such epithets pertain to dirty words and animal abuse, with the former category referring to sex and excretion and the latter to animals. This language depicts the taboo domain referring to disapproved and stigmatized behavior linked to the sphere of intimacy. Once such behavior occurs outside intimacy, it is perceived as extremely inappropriate and offensive. In this vein, the person performing such behavior might be considered as lacking one of the fundamental properties of human beings, that is, the ability to control one’s impulses and behavior. Semin and Rubini (1990) showed that in individualistic contexts, the denial of the concept of person was conveyed by insults denying psychological/physical properties of the addressee and insults related to the taboo domain (i.e., insults related to sexual organs, sexual behavior, excrement, and animality). The classification of insults adopted by Semin and Rubini (1990) largely overlaps with Van Oudenhoven et al.’s (2008) work in terms of verbal abuse in different cultures, which stressed the dichotomy between lack of intelligence and ethical/moral taboo (e.g., sexuality, animality, excrement). Similarly, Haslam, Loughnan, and Sun’s (2011) studies on metaphor-based dehumanization highlighted two kinds of offensive animal metaphor: those that are degrading—that call into question humanity—and those that are disgusting and related to moral depravity. In view of all this, verbal insults can be considered a means that may well capture the consequences of (reduced) humanity of the addressee.

Overview of the Studies

Two studies experimentally addressed the role of reduced humanity and group membership of a target in producing linguistic discrimination. In order to provide participants with persuasive evidence of target’s (reduced) humanity, we created a paradigm whereby two humanity dimensions—already investigated in dehumanization studies (i.e., display of secondary vs. primary emotions, Leyens et al., 2000; having concern vs. little concern for other human beings, Struch & Schwartz, 1989)¹—were manipulated, instead of only one as in previous studies (cf. Pereira et al., 2009; Vaes et al., 2003).

In Study 1, linguistic discrimination was assessed at the implicit structural level of linguistic abstraction (Semin & Fiedler, 1988) as may be found in natural language. Study 2 focused on the target’s overt devaluation through an explicit measure of discrimination, namely, verbal insults (Semin & Rubini, 1990).

Study 1

Study 1 tested the role of humanity versus reduced humanity of a person and his or her group membership on linguistic discrimination in terms of the abstraction and valence of predicates (Semin & Fiedler, 1988) used in the free descriptions of the target.

Since exclusion from the human group is conceived as a factor leading to moral disengagement (Bandura, 2002) toward others’ well-being and exclusion from one’s moral community (Opotow, 1990; Staub, 1989), we assumed that target’s reduced humanity would lead to higher linguistic discrimination than did target’s humanity. In line with the previous findings on linguistic discrimination in intergroup settings (e.g., Moscatelli & Rubini, 2011), we expected that the target would be described with positive terms at a higher level of abstraction in the humanity than in the reduced humanity condition (Hypothesis 1a). By contrast, the target would be described with negative terms at a lower level of abstraction in the humanity than in the reduced humanity condition (Hypothesis 1b).

The main focus of the contribution was the interplay between humanity and group membership. In this respect, there is evidence that salience of a superinclusive category (as humanity is) reduces intergroup differentiation and discrimination (cf. Gaertner & Dovidio, 2000). By contrast, research on the ingroup projection model (for a review, see Wenzel, Mummendey, & Waldzus, 2007) highlights that when group members identify with their ingroup and simultaneously with a superordinate (positively evaluated) category, they show an ingroup projection bias which enhances intergroup differentiation rather than hindering it. In this vein, the infrahumanization studies (Vaes et al., 2003; Vaes et al., 2004, 2006; Vaes et al., 2011; Wohl et al., 2012) provide coherent evidence that the outgroup is treated worse even when it is humanized (i.e., when expressing secondary vs. primary emotions). Following the ingroup projection model, we thus hypothesized a moderation effect of target’s group membership in the humanity condition: We expected that the human ingrouper would be described with positive terms at a higher level of abstraction and negative terms at a lower level of abstraction than would the human outgrouper (Hypothesis 2). In the reduced humanity condition, no differences in the level of linguistic discrimination were expected in terms of group membership, due to belongingness to a negatively evaluated common category.

Method

Participants and design

Two hundred eighty-six undergraduate students (M_age = 22.87, SD = 3.91; females = 50.0%) participated in the study on a voluntary basis. All participants were Northern or Southern Italians. Participants were randomly assigned to a 2 (target’s humanity: humanity, reduced humanity) × 2 (target’s group membership: ingroup, outgroup) between-participants factorial design.

Preliminary studies

Two preliminary studies were conducted to test the effectiveness of the humanity manipulation used in the main studies. In Preliminary Study 1 (N = 80), four different descriptions of a target (see the online appendix for full details of “Method” and “Results”) resulting from the combination of the two humanity versus reduced humanity dimensions were examined in order to choose the combinations that best discriminated between high and low humanity of the target. Preliminary Study 1 also tested whether both dimensions were considered UH features and whether participants’ judgments of the target were influenced by both humanity dimensions to the same extent. Preliminary Study 2 (N = 160) controlled evaluation of the two humanity facets (see the online appendix for detailed “Method” and “Results”).

Procedure

Participants received a paper-and-pencil questionnaire containing a description of a male or a female target, whose gender was paired with that of the respondent in order to avoid making this categorical dimension salient. At the end of the experimental session, participants were debriefed and informed about the aims of the research. The target’s humanity was manipulated by assigning participants the humanity versus reduced humanity description of the target chosen on the basis of Preliminary Study 1 evidence (see the online appendix). The target’s group membership (ingroup vs. outgroup) was manipulated by matching versus mismatching target’s and respondent’s regional belongingness (i.e., Northern/Southern Italian). An example of the target’s descriptions is the following (see the online appendix for all target’s descriptions): “Francesco was born and lives in a town in the North of Italy. He is a person that has concern for other human beings. In his relationships with others, Francesco displays empathy and hope, dejection and melancholy” (i.e., humanity condition, Northern Italian male).

Measuring linguistic discrimination

After being exposed to target’s profile, participants were asked to describe him or her in their own words. Linguistic productions were coded by two independent coders, blind to the experimental conditions. Following Semin and Fiedler’s (1988) linguistic category model, they distinguished between descriptive action verbs (DAVs), interpretive action verbs (IAVs), state verbs (SVs), and adjectives (ADJs). The semantic valence of predicates (positive vs. negative) was also coded. Intercoder agreement was high (for predicate type: Cohen’s κ = .86; for predicate valence: Cohen’s κ = .90). Disagreement was solved by discussion.

Manipulation checks

Participants rated the extent to which Francesco/a was “a representative member of the human group” and “a typical human being” (α = .63).² They also rated the extent to which having concern (vs. little concern) for other human beings was a positive characteristic in order to control for the biased evaluation of this human aspect that emerged in Preliminary Study 2 (see the online appendix). All judgments were expressed on 7-point Likert-type scales (1 = not at all; 7 = very much).

Results and Discussion

Manipulation checks

The 2 (target’s humanity: humanity, reduced humanity) × 2 (target’s group membership: ingroup, outgroup) ANOVA on target’s human representativeness revealed only a significant effect of target’s humanity (M_humanity = 4.70, SD = 1.18; M_{reduced humanity} = 3.98, SD = 1.49), F(1, 282) = 20.67, p < .001, η² = .068. Also the 2 × 2 ANOVA on the item assessing positivity of having concern (vs. little concern) for others showed only a significant effect of target’s humanity, F(1, 282) = 22.71, p < .001, η² = .075, revealing that having concern (M = 4.64, SD = 1.46) was evaluated as a positive characteristic to a higher extent than was having little concern (M = 3.82, SD = 1.45). This measure was entered as a covariate in further analyses.

Linguistic abstraction

Positive and negative abstraction scores were computed for each participant by assigning weights of 1, 2, 3, and 4 to DAVs, IAVs, SVs, and ADJs, respectively. The summed weights were then divided by the total number of terms used. The two abstraction indexes ranged from 1 to 4. A 2 (target’s humanity) × 2 (target’s group membership) × 2 (linguistic abstraction: positive, negative) ANCOVA with repeated measures on the last factor was conducted entering positivity of having (little) concern for other human beings as a covariate. Importantly, the effect of the covariate was not significant, F(1, 280) = 1.13, p = .288, η² = .004. The analysis highlighted a significant effect of linguistic abstraction, F(1, 280) = 4.02, p = .046, η² = .014, and significant two-way interactions of Linguistic abstraction × Target’s humanity, F(1, 280) = 337.87, p < .001, η² = .547, and Linguistic abstraction × Target’s group membership, F(1, 280) = 16.11, p < .001, η² = .054, which were qualified by the three-way interaction of Linguistic abstraction × Target’s humanity × Target’s group membership, F(1, 280) = 21.26, p < .001, η² = .071.

To decompose such interaction, 2 (target’s humanity) × 2 (target’s group membership) ANCOVAs were carried out on each abstraction index separately (see Table 1 for means and standard errors of positive and negative abstraction indexes).

Table 1.

Positive and Negative Abstraction as a Function of Target’s Humanity and Target’s Group Membership.

		Target’s group membership
Linguistic abstraction	Target’s humanity	Ingroup	Outgroup
Positive abstraction	Humanity	3.45 (0.10)	2.94 (0.10)
Positive abstraction	Reduced humanity	1.32 (0.10)	1.42 (0.10)
Negative abstraction	Humanity	1.73 (0.11)	2.53 (0.11)
Negative abstraction	Reduced humanity	3.23 (0.11)	3.23 (0.11)

Note. Standard errors are presented in parentheses.

As for positive abstraction, the 2 × 2 ANCOVA revealed a non-significant effect of the covariate, F(1, 280) = 0.25, p = .616, η² = .001. In line with Hypothesis 1a, target’s humanity overwhelmingly affected the level of abstraction in target’s descriptions (M_humanity = 3.20, SE = 0.07; M_{reduced humanity} = 1.37, SE = 0.07), F(1, 280) = 322.28, p < .001, η² = .535. That is, the target was described using positive terms at a higher level of abstraction in the humanity than in the reduced humanity condition. The effect of target’s group membership (M_ingroup = 2.39, SE = 0.07; M_outgroup = 2.18, SE = 0.07), F(1, 280) = 4.68, p = .031, η² = .016, and also the two-way interaction of Target’s humanity × Target’s group membership, F(1, 280) = 9.68, p = .002, η² = .033, were significant. Pairwise comparisons (Bonferroni) revealed that the ingrouper was described with positive terms at a higher level of abstraction in the humanity than in the reduced humanity condition (p < .001). The same result emerged for the outgrouper (p < .001). In line with Hypothesis 2, in the humanity condition, target’s group membership led to more abstract positive descriptions of the ingrouper than of the outgrouper (p < .001). In the reduced humanity condition, target’s group membership did not affect the abstraction of positive terms (p = .504).

As for negative abstraction, the 2 × 2 ANCOVA yielded a non-significant effect of the covariate, F(1, 280) = 0.84, p = .361, η² = .003. As expected (Hypothesis 1b), target’s humanity effect (M_humanity = 2.13, SE = 0.08; M_{reduced humanity} = 3.23, SE = 0.08), F(1, 280) = 92.89, p < .001, η² = .249, was significant. That is, the target with reduced humanity was described with negative terms at a higher level of abstraction than the human one. The analysis produced also a significant effect of target’s group membership (M_ingroup = 2.48, SE = 0.08; M_outgroup = 2.88, SE = 0.08), F(1, 280) = 13.44, p < .001, η² = .046, and a Target’s humanity × Target’s group membership interaction, F(1, 280) = 13.35, p < .001, η² = .046. As expected, pairwise comparisons (Bonferroni) highlighted that the ingrouper was described with negative terms at a higher level of abstraction in the reduced than in the humanity condition (p < .001). The same effect emerged for the outgrouper (p < .001). In line with Hypothesis 2, in the humanity condition, the ingrouper was described with less abstract negative terms than the outgrouper (p < .001). In the reduced humanity condition, target’s group membership effect was not significant (p = .992).

This study provided very strong evidence on the role of reduced humanity as an antecedent of social discrimination. Very interestingly, the effect size of the interaction between humanity and linguistic abstraction was huge and much greater than that of the interaction between group membership and linguistic abstraction, on which the LIB (Maass, 1999) is based. It must be remembered that in the Italian context, the Southern/Northern distinction is generally very marked (Maass, Ceccarelli, & Rudin, 1996) and traditionally gives rise to reciprocal stereotypes and prejudice between members of the two groups. This implies that humanity, to a higher extent than group membership, is an overarching means of definition with an independent effect on social discrimination. In the same way, it protects outgroupers from derogation when they are portrayed as full human beings. In contrast, when ingroupers are depicted with reduced humanity, they are not saved from linguistic discrimination and are described as if they were black sheep (Marques & Yzerbyt, 1988). This finding reflects people’s tendency to linguistically derogate ingroupers who are perceived as deviant, marginal representatives of the ingroup. In the same way, it might be likely that participants refused to think that their group included persons not representative of the human group.

Taken together, the novel evidence we gathered reflects a “linguistic human bias,” that is, the tendency to describe the human target with positive terms at a higher level of abstraction and negative terms at a lower level of abstraction than the target with reduced humanity. We believe that this finding proves the effectiveness of our manipulation in showing the strength of humanity as a very influencing factor of social perception, and demonstrates the importance of using “spontaneous” yet subtle measures of social differentiation.

Study 2

If Study 1 assessed linguistic discrimination of others characterized by reduced humanity at the implicit level of language abstraction, Study 2 went a step further by analyzing how reduced humanity of individuals can result in overt offensive language.

Participants were exposed to the same target descriptions adopted in Study 1. To gather spontaneous production of epithets toward the target, they were provided with a scenario in which the protagonist performed a behavior (i.e., crossing the road without looking right and left) that might, in principle, provoke both relatively protective utterances such as warnings—expressing concern for the danger that the target person may encounter—and derogative insults—expressing devaluation toward the addressee. Very plausibly, warnings can imply prosocial motivations and moral concern (Opotow, 1990) toward other human beings, especially when they are perceived to be in danger. In contrast, abusive language can underlie antisocial motivation and be used to express verbal aggression (Van Oudenhoven et al., 2008) or derogation of individuals who are not protected by moral concern (Bandura, 2002) because they reveal reduced humanity.

Terms were classified according to criteria of conceptual similitude and etymological meaning, following the classification developed by Semin and Rubini (1990) on the basis of E. R. Leach’s (1964) pivotal work. The coding was further adapted in light of the characteristics of the scenario and of Van Oudenhoven et al.’s (2008) classification distinguishing between insults related to lack of human properties (e.g., intelligence) and insults referring to taboo (e.g., sexuality, animality, excrement).

Given the characteristics of the scenario, it was hypothesized that participants would address the human target with protective warnings to a greater extent than the target with reduced humanity. Conversely, we expected that the latter target would lead to the formulation of derogative insults toward him or her to a greater extent than would the human target (Hypothesis 1). In addition, in line with the findings of Study 1, we predicted that target’s group membership would affect the formulation of epithets only in the humanity condition (Hypothesis 2): A greater amount of protective warnings would be directed to the target belonging to the ingroup than to the outgroup. Conversely, the outgrouper would be addressed with a greater amount of offensive insults than would the ingrouper. In the reduced humanity condition, group membership would not affect epithets addressed toward the target.

Method

Participants and design

Three hundred fifty-two undergraduate students (M_age = 22.34, SD = 3.66; females = 50.6%) from several Northern and Southern Italian universities (n_North = 176; n_South = 176) participated in the study on a voluntary basis. All participants were Northern or Southern Italians. Again, gender of the target and the participant was matched. A 2 (target’s humanity: humanity, reduced humanity) × 2 (target’s group membership: ingroup, outgroup) between-participants factorial design was used.

Procedure and materials

Scenario for verbal epithets

After reading target’s description, participants were presented with the following sketch of a situation:

Imagine yourself in this context: You are driving along the road you usually take to go home. Francesco [Francesca] is walking along the road when suddenly he [she] crosses over without looking left and right. You have to brake sharply to avoid running him [her] over. You lean out of the window and start talking to him [her]. What would you say?

Participants were invited to write down freely and honestly in the provided empty lines all the epithets that they would address to the target in such a stimulus situation. At the end of the experimental session, they were debriefed and informed about research aims.

Coding verbal epithets

The epithets used by participants were coded by two independent judges who were blind to the aims of the study. The coding system included three main categories (see Table 2): (a) warnings, a type of epithets revealing concern for the target; (b) insults denying physical/intellectual properties of the person (E. R. Leach, 1964; Semin & Rubini, 1990; Van Oudenhoven et al., 2008); and (c) insults related to taboo, utterances related to sexuality, animality, or excrement (Semin & Rubini, 1990; Van Oudenhoven et al., 2008).³ In coding the verbal epithets into different categories, the judges were almost unanimous. Intercoder agreement was high (Cohen’s κ = .85). Disagreement was solved by discussion.

Table 2.

Descriptions and Examples of the Types of Epithets Included in the Coding System.

Type of epithet	Examples
Warnings	Breaking of certain norms of civil behavior (e.g., Si guarda prima di attraversare!—Look before you cross!); lack of attention and mental awareness (e.g., Sveglia!—Wake up!); warnings implying possible consequences (e.g., A momenti ti tiro sotto!—I nearly ran you over!); commands and imperatives to change behavior (e.g., Ritornate a casa!—Go home!).
Insults denying physical/intellectual properties of the person	Insults denying psychological properties (e.g., Idiota!—Idiot!; Matto/a!—Mental case!); insults denying physical properties (e.g., Sei cieca!—You are blind!; Non hai gli occhi!—You do not have eyes!).
Insults related to taboo	Insults exaggerating sexual properties and behavior (e.g., Testa di cazzo!—Dickhead!; Fottiti!—Fuck you!); insults referred to immoral sexual behavior (e.g., Puttana!—Whore!; Bagascia!—Slag!); insults referring to subhuman or animal characteristics (e.g., Vacca!—Cow!; Cagna!—Bitch!); epithets referring to excrement (e.g., Faccia di merda!—Shitface!).

Manipulation checks

As in Study 1, participants completed two items on target’s human representativeness (α = .62). Positivity of having concern (vs. little concern) for other human beings was controlled. For both measures, the response scale ranged from 1 (not at all) to 7 (very much).

Results and Discussion

Manipulation checks

The 2 (target’s humanity: humanity, reduced humanity) × 2 (target’s group membership: ingroup, outgroup) ANOVA on target’s human representativeness revealed significant effects of target’s humanity (M_humanity = 4.64, SD = 1.37; M_{reduced humanity} = 3.51, SD = 1.47), F(1, 348) = 57.36, p < .001, η² = .141, and target’s group membership (M_ingroup = 4.34, SD = 1.58; M_outgroup = 3.84, SD = 1.43), F(1, 348) = 11.48, p = .001, η² = .032.⁴

The 2 × 2 ANOVA on the item assessing positivity of having concern (vs. little concern) for other human beings revealed only a significant effect of target’s humanity (M_humanity = 6.14, SD = 1.14; M_{reduced humanity} = 2.29, SD = 1.44), F(1, 348) = 781.01, p < .001, η² = .692. This item was entered as a covariate in the main analyses.

Verbal epithets

Participants produced 787 insults and warnings. Frequencies of use of epithets (see Figure 1) were converted into percentages for each category of epithet (i.e., warnings, insults denying physical/intellectual properties of the person, insults related to taboo), which were then subjected to an arcsine transformation to normalize data and to remove the intercell dependencies (see Menegatti & Rubini, 2012). Since gender of the target could, in principle, affect formulation of epithets (Haslam et al., 2011; Van Oudenhoven et al., 2008), these proportions were submitted to a 2 (target’s humanity) × 2 (target’s group membership) × 2 (gender of the target/participant: male, female) × 3 (type of epithet: warnings, insults denying physical/intellectual properties of the person, insults related to taboo) ANCOVA with repeated measures on the last factor and positivity of having (little) concern for others as a covariate. As shown in this analysis, the effect of the covariate was not significant, F(1, 343) = 1.54, p = .215, η² = .004. The analysis revealed a significant effect of type of epithet, F(2, 342) = 10.66, p < .001, η² = .059. The interactions of Type of epithet × Target’s humanity, F(2, 342) = 8.90, p < .001, η² = .049, and Type of epithet × Target’s group membership, F(2, 342) = 27.40, p < .001, η² = .138, were significant. Also a two-way interaction of Type of epithet × Gender of the target/participant, F(2, 342) = 17.77, p < .001, η² = .094, emerged. Such interaction is not discussed here, since it is not relevant to the hypotheses.⁵ Importantly, the three-way interaction of Type of epithet × Target’s humanity × Target’s group membership, F(2, 342) = 4.92, p = .008, η² = .028, was significant. To better understand whether the results support the hypotheses, 2 (target’s humanity) × 2 (target’s group membership) ANCOVAs were conducted separately on each type of epithet. Such analyses revealed different patterns of results as a function of type of epithet (see Table 3 for means and standard errors of the different types of epithet).

Figure 1.

Frequencies of use of the different epithets as a function of target’s humanity.

Table 3.

Warnings, Insults Denying Physical/Intellectual Properties of the Person, and Insults Related to Taboo as a Function of Target’s Humanity and Target’s Group Membership.

		Target’s group membership
Type of insult	Target’s humanity	Ingroup	Outgroup
Warnings	Humanity	0.56 (0.03)	0.43 (0.03)
Warnings	Reduced humanity	0.42 (0.03)	0.20 (0.03)
Insults denying physical/intellectual properties	Humanity	0.16 (0.04)	0.29 (0.03)
Insults denying physical/intellectual properties	Reduced humanity	0.25 (0.04)	0.27 (0.04)
Insults related to taboo	Humanity	0.08 (0.03)	0.11 (0.03)
Insults related to taboo	Reduced humanity	0.13 (0.03)	0.26 (0.03)

Note. Standard errors are presented in parentheses

As for warnings, the analysis yielded a non-significant effect of the covariate, F(1, 347) = 0.12, p = .730, η² = .000. In line with Hypothesis 1, target’s humanity (M_humanity = 0.49, SE = 0.03; M_{reduced humanity} = 0.31, SD = 0.03) significantly affected the formulation of warnings toward the addressee, F(1, 347) = 16.17, p < .001, η² = .045. The analysis also produced a significant effect of target’s group membership (M_ingroup = 0.49, SD = 0.02; M_outgroup = 0.31, SD = 0.02), F(1, 347) = 48.70, p < .001, η² = .123. The two-way interaction of Target’s humanity × Target’s group membership was close to significance, F(1, 347) = 3.78, p = .053, η² = .011. However, pairwise comparisons (Bonferroni) highlighted two parallel simple effects of target’s humanity in the ingroup (p = .010) and in the outgroup (p < .001) conditions, and of target’s group membership in the humanity condition (p < .001)—as expected (Hypothesis 2)—but also in the reduced humanity one (p < .001).

The 2 × 2 ANCOVA on arcsine-transformed proportions of insults denying physical/intellectual properties revealed a non-significant effect of the covariate, F(1, 347) = 2.56, p = .110, η² = .007. Unexpectedly, this analysis showed no difference due to target’s humanity (M_humanity = 0.22, SE = 0.03; M_{reduced humanity} = 0.26, SE = 0.03), F(1, 347) = 0.57, p = .451, η² = .002. Nevertheless, mean scores were in the expected direction. By contrast, the effect of target’s group membership was significant (M_ingroup = 0.20, SD = 0.02; M_outgroup = 0.27, SD = 0.02), F(1, 347) = 6.23, p = .013, η² = .018. An almost significant two-way interaction of Target’s humanity × Target’s group membership, F(1, 347) = 3.83, p = .057, η² = .011, also emerged. Unexpectedly, pairwise comparisons (Bonferroni) highlighted no significant difference due to target’s humanity both in the ingroup (p = .120) and in the outgroup (p = .728) conditions. In line with Hypothesis 2, the human ingrouper received fewer insults denying physical/intellectual properties than the human outgrouper (p = .001), whereas the same effect was not found in the reduced humanity condition (p = .679).

The 2 × 2 ANCOVA on arcsine-transformed proportions of insults related to taboo revealed a non-significant effect of the covariate, F(1, 347) = 0.03, p = .873, η² = .000. As expected (Hypothesis 1), target’s humanity significantly affected the use of these epithets addressed toward the target (M_humanity = 0.10, SE = 0.02; M_{reduced humanity} = 0.19, SE = 0.03), F(1, 347) = 5.16, p = .024, η² = .015. Also the effect of target’s group membership (M_ingroup = 0.10, SE = 0.02; M_outgroup = 0.19, SE = 0.03), F(1, 347) = 12.64, p < .001, η² = .035, and the two-way interaction, F(1, 347) = 4.90, p = .028, η² = .014, were significant. Pairwise comparisons (Bonferroni) highlighted that only the outgrouper with reduced humanity received a higher amount of insults related to taboo (p = .003). No difference was found for the ingrouper (p = .359). Unexpectedly, target’s group membership did not have a significant effect in the humanity condition (p = .337) but only in the reduced humanity one (p < .001).

To sum up, the results of this study showed that warnings were addressed more to the human target than to the less human one. In contrast, insults related to taboo were mainly directed to the target with reduced humanity. These findings fully support Hypothesis 1. Insults denying physical/intellectual properties were not differently addressed to the two targets. Very likely, the scenario at stake (i.e., crossing the road without looking right and left) might have led participants to formulate this type of epithet rather than target’s (reduced) humanity per se.

With regard to the interplay between target’s humanity and group membership (Hypothesis 2), the pattern of results varied across the different types of epithets. As expected, warnings were addressed to the ingrouper to a higher extent in the humanity condition, and—unexpectedly—also in the reduced humanity one. Probably, due to the protective (Opotow, 1990) function of warnings, participants did not refrain from showing ingroup bias even when the ingrouper was depicted with reduced humanity. This finding can be related to the greater effect size of group membership in comparison with that of humanity. This result was consistent across the three types of epithets considered. By contrast, regarding insults denying physical/intellectual properties, participants differentiated between the ingrouper and the outgrouper only in the humanity condition, thus fully confirming Hypothesis 2. Regarding insults related to taboo, unexpectedly, participants refrained from discriminating the human target as a function of his or her group membership. Probably, this evidence might be explained by participants’ tendency to avoid addressing the human target with very derogative epithets (as revealed by the low mean scores of these insults). Similarly, taboo insults might be regarded as a very offensive means of derogation compared to those denying physical/intellectual properties of a person. This contention is very close to Haslam et al.’s (2011) argument that animal metaphors related to repulsion and depravity are extremely offensive. By contrast, in the reduced humanity condition a different, yet unexpected, pattern was found: Reduced humanity allowed participants to actively derogate the outgrouper to a greater extent than the ingrouper, who once again was protected by his or her group membership.

Taken together, such multifaceted evidence is indicative of the conditions under which overt derogation of others can be expressed. In particular, when the outgrouper was portrayed with reduced humanity, participants seemed to disengage from any form of moral concern (Bandura, 2002) and feel legitimized to further deny his or her humanity through very offensive, taboo-related insults.

General Discussion

In the literature on social discrimination, dehumanization has mainly been treated as a dependent variable (cf. Haslam et al., 2008; Leyens et al., 2000; Vaes et al., 2012), and the role of reduced humanity of others has been considered in only few contributions (Costello & Hodson, 2010, 2011; Loughnan et al., 2009; Pereira et al., 2009; Vaes et al., 2003; Vaes et al., 2004, 2006; Vaes et al., 2011; Wohl et al., 2012). This set of studies was designed to fill this gap by experimentally manipulating humanity and group membership of a target and assess the effects of these factors on linguistic discrimination.

Based on the argument that humanity is the most important good people distribute to others (Walzer, 1983), the target’s humanity was varied by combining two UH dimensions (i.e., secondary emotions, Leyens et al., 2000; prosocial human values, Struch & Schwartz, 1989). Group membership was operationalized in terms of Northern/Southern regional belongingness. Unlike previous studies where only a single human dimension was at stake (Pereira et al., 2009; Vaes et al., 2003; Vaes et al., 2004, 2006; Wohl et al., 2012), the target’s humanity was portrayed in a more complex fashion by combining two UH dimensions. Such manipulation successfully affected the language participants used to describe/address the target.

Overall, the findings of the two studies revealed that humanity—that is, membership of the most superinclusive group one could possibly belong to (cf. Turner, Hogg, Oakes, Reicher, & Wetherell, 1987)—led participants to describe or address the human target in a more favorable way than the less human one. By contrast, exclusion from the human ingroup led to increased linguistic derogation. This evidence is particularly noteworthy because it differs from previous findings (e.g., Vaes et al., 2003; Vaes et al., 2004, 2006; Wohl et al., 2012). Very likely, the fact that participants were exposed to a target who, independently of his or her group membership, was portrayed as a human being in a more complex fashion than in previous research, challenged the biased representation of outgroupers as characterized by a lower human essence (cf. Leyens et al., 2000) than ingroupers. This led to relatively unbiased descriptions of outgroup members. It is plausible that thinking about an outgrouper depicted in terms of multiple human features triggers a counterstereotypical mindset (cf. Hutter & Crisp, 2005; Prati, Vasiljevic, Crisp, & Rubini, in press), which leads to reduction in linguistic bias. This finding was particularly evident in Study 1, where the effect of humanity was much stronger than that of group membership.

In contrast, we showed that when the target was described as lacking multiple UH dimensions, participants derogated him or her to a higher extent than the human one. Again, this finding is not consistent with previous studies (e.g., Vaes et al., 2003; Vaes et al., 2006) since it shows that when ingroupers do not possess multiple human features they are treated in a biased fashion as much as outgroupers are. Besides the aforementioned evidence, in this set of studies we also highlighted the conditions under which people can still differentiate between the ingroup and the outgroup when common superinclusive belongingness is at stake, in line with evidence on the ingroup projection tendency (Wenzel et al., 2007).

Language Abstraction

In Study 1, a linguistic human bias emerged based on the evidence that the human target was described more favorably than the target with reduced humanity. These results effectively extended the linguistic intergroup discrimination phenomenon (Maass, 1999; Menegatti & Rubini, 2013; Moscatelli & Rubini, 2011; Rubini & Menegatti, 2008, 2014) by showing that this behavior can be rooted not only in group membership but also in (reduced) humanity. Moreover, by tailoring language abstraction and valence of terms, participants communicated the two facets of linguistic discrimination (Moscatelli & Rubini, 2011; Prati, Menegatti, & Rubini, 2014; Rubini et al., in press), that is, ingroup favoritism—expressed through positive descriptions at a higher level of abstraction—and outgroup derogation (cf. Brewer, 1999)—expressed through negative descriptions at a higher level of abstraction. These facets of discrimination emerged as not only a function of group membership (in the humanity condition), but also as depending on target’s humanity.

Verbal Abuse

Interestingly, the language of insults made it possible to detect different dimensions along which discrimination appears. On the one hand, warnings—a more protective way of addressing the target—very plausibly revealed participants’ concern for the well-being of the human ingrouper or outgrouper and his or her inclusion in their moral (human) community (Opotow, 1990). On the other hand, the target’s reduced humanity led respondents to go beyond moral constraints (Bandura, 2002) and overtly derogate the addressee under specific conditions. It should be stressed that an interesting interplay between target’s humanity and group membership emerged and that this varied as a function of the type of epithet.

Besides this, the findings on verbal insults are very interesting since they suggest that the denial of some human features conveyed by the (reduced) humanity manipulation encouraged participants to deny the target further human characteristics through specific insults, as in the case of the less human outgrouper. The evidence of Study 2 might also be enlightening with respect to the literature on the two fundamental dimensions of social judgment, that is, morality and competence (C. W. Leach, Ellemers, & Barreto, 2007; see also Fiske, Cuddy, Glick, & Xu, 2002): One can argue that insults denying physical/intellectual properties might be conceived as an overt means of discrimination based on the denial of competence and abilities, whereas insults related to taboo might be more closely related to a lack of control over one’s own behavior and (private) morality. Insults referring to excrement or abnormal/exacerbated sexual behavior might be linked to a lack of moral foundation in terms of purity and sanctity, which are indeed considered high moral values (Graham et al., 2011).

The implications of our evidence on verbal insults as a means to derogate others might also be very interesting if related to Haslam’s (2006) model of dehumanization in terms of denial of UH versus HN characteristics: One can argue that denial of intellectual/physical abilities might be linked to objectification, leading to an automata like perception of others, whereas denial of morality might be more closely related to animalization.

The Role of Others’ Humanity in Social Perception

The evidence from these studies showed the crucial role of reduced humanity in driving social discrimination. In this vein, our findings stressed the importance of the most inclusive level of self-categorization in Turner et al.’s (1987) model—the human one—as a criterion of self/other definition that can deeply affect social judgment.

Indeed, the role of human identity in intergroup relations has remained almost neglected at the empirical level (for an exception, see Albarello & Rubini, 2012). From a different point of view, Leyens et al. (2000)—following Yzerbyt et al.’s (1997) contention that groups are perceived as possessing different inherent essences—showed that individuals attribute UH features to a lesser extent to outgroupers. In their line of research (as in many others; for example, Haslam et al., 2008) dehumanization was mainly conceived as an outcome of social categorization.

In this contribution, taking a different stance, we manipulated the target’s humanity as an antecedent to social discrimination. Our findings highlighted that when the target was characterized with reduced humanity, there was room for linguistically discriminating him or her, irrespective of group membership. This was true at the implicit level of linguistic abstraction, as well as at the overt level of verbal insults. Overall, the findings of this set of studies suggest that we should give proper consideration to the interplay of the ingroup/outgroup level of categorization and the human one (Turner et al., 1987) in order to better understand subtle and overt forms of discrimination.

Taken together, the results of this line of research show the protective role of humanity in preventing group members and especially outgroupers from being discriminated against. We believe that this evidence is due to the multidimensional manipulation of humanity we used, which allowed participants to consider outgroup members under a new “human light.” However, group membership moderated the effect of humanity, especially when verbal insults were considered. It should be mentioned that the behavioral scenario in Study 2 probably drove participants’ attention to the target’s group membership more than in Study 1. Moreover, unlike language abstraction—conceived as a subtle means of discrimination—verbal insults are more consciously controllable and constitute an overt means of offensive discrimination. This might have led participants to use them in a biased fashion to protect their ingroup fellows whenever they could (cf. Tajfel & Turner, 1979).

We believe that the phenomenon of linguistic discrimination is due to many contextual factors such as the ones we have reviewed—namely the interplay between target’s humanity and group membership, the nature of the linguistic devices at participants’ disposal (whether controllable or not), and the characteristics of the experimental scenario that may reproduce features of real life to a varying extent.

Limitations and Future Directions

Future studies should directly compare the effect of target’s humanity to that of target’s group membership in order to test whether humanity overrides group membership. Besides the two humanity facets employed to manipulate others’ humanity, further dimensions should be considered, for instance, HN traits (Haslam et al., 2008), the combination of UH and HN characteristics, or self-enhancement egoistic human values (Schwartz, 1992)—which might be more closely linked to competence (Fiske et al., 2002) than prosocial human values. By combining two UH dimensions to manipulate target’s humanity, our work lends support to a holistic, multifaceted concept of humanity of a person: In other words, humanity can be conceived as a multidimensional concept which relies on several different and yet integrated characteristics that together contribute to defining what a human being is. Further research should tackle more thoroughly this issue.

Our evidence also suggested that the various typologies of insult might differ in terms of severity as well as in terms of the dehumanized perception of the addressee that they refer to. Such issues should be considered more deeply in future work.

Besides the above considerations, the evidence of this set of studies is very interesting since it shows for the first time that the way we portray others as endowed with full or diminished humanity works as a factor legitimating different forms of discrimination. At the implicit level of linguistic abstraction, reduced humanity led to a linguistic human bias mainly expressed in terms of abstract derogative language. At the overt level of verbal abuse, it removed moral constraints that usually inhibit people from addressing others with very offensive insults related to lack of human properties, to taboo, or to moral depravity of the addressee.

Footnotes

Acknowledgements

We are grateful to Michael Wenzel and the two anonymous reviewers for their helpful comments in the editorial processing of the article. We are also thankful to Gun Semin, Silvia Moscatelli, and Michela Menegatti for their comments on an earlier draft of the article. We also thank Martina Aiello and Francesca Chirico for their assistance in data collection and coding.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research reported in this article was supported by funds (n. 71-2010-1587) from the Alma Mater Studiorum–University of Bologna and by FIRB-Futuro in Ricerca 2012 funds (n. RBFR128CR6_004) from Ministry for Education, University, and Research (MIUR).

Supplemental Material

The online supplemental material is available at .

Notes

References

Albarello

Rubini

(2012). Reducing dehumanisation outcomes towards Blacks: The role of multiple categorisation and human identity. European Journal of Social Psychology, 42, 875-882. doi:10.1002/ejsp.1902

Allport

(1954). The nature of prejudice. Reading, MA: Addison-Wesley.

Bandura

(2002). Selective moral disengagement in the exercise of moral agency. Journal of Moral Education, 31, 101-118. doi:10.1080/0305724022014322

Brewer

M. B.

(1999). The psychology of prejudice: Ingroup love or outgroup hate? Journal of Social Issues, 55, 429-444. doi:10.1111/0022-4537.00126

Costello

Hodson

(2010). Exploring the roots of dehumanization: The role of animal-human similarity in promoting immigrant dehumanization. Group Processes & Intergroup Relations, 13, 3-22. doi:10.1177/1368430209347725

Costello

Hodson

(2011). Social dominance-based threat reactions to immigrants in need of assistance. European Journal of Social Psychology, 41, 220-231. doi:10.1002/ejsp.769

Fiske

S. T.

Cuddy

A. C.

Glick

(2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition. Journal of Personality and Social Psychology, 82, 878-902. doi:10.1037/0022-3514.82.6.878

Franco

F. M.

Maass

(1999). Intentional control over prejudice: When the choice of the measure matters. European Journal of Social Psychology, 29, 469-477. doi:10.1002/(SICI)1099-0992(199906)29:4<469::AID-EJSP938>3.0.CO;2-S

Gaertner

S. L.

Dovidio

J. F.

(2000). Reducing intergroup bias: The common ingroup identity model. Philadelphia, PA: Psychology Press.

10.

Graham

Nosek

B. A.

Haidt

Iyer

Koleva

Ditto

P. H.

(2011). Mapping the moral domain. Journal of Personality and Social Psychology, 101, 366-385. doi:10.1037/a0021847

11.

Haslam

(2006). Dehumanization: An integrative review. Personality and Social Psychology Review, 10, 252-264. doi:10.1207/s15327957pspr1003

12.

Haslam

Loughnan

Kashima

Bain

(2008). Attributing and denying humanness to others. In Stroebe

Hewstone

(Eds.), European review of social psychology (Vol. 19, pp. 55-85). London, England: Psychology Press. doi:10.1080/10463280801981645

13.

Haslam

Loughnan

Sun

(2011). Beastly: What makes animal metaphors offensive? Journal of Language and Social Psychology, 30, 311-325. doi:10.1177/0261927X11407168

14.

Hodson

MacInnis

C. C.

Costello

(2013). (Over)valuing “humanness” as an aggravator of intergroup prejudices and discrimination. In Bain

Vaes

Leyens

J. Ph.

(Eds.), Humanness and dehumanisation (pp. 86-110). London, England: Psychology Press.

15.

Hutter

R. R. C.

Crisp

R. J.

(2005). The composition of category conjunctions. Personality and Social Psychology Bulletin, 31, 647-657. doi:10.1177/0146167204271575

16.

Leach

C. W.

Ellemers

Barreto

(2007). Group virtue: The importance of morality (vs. competence and sociability) in the positive evaluation of in-groups. Journal of Personality and Social Psychology, 93, 234-249. doi:10.1037/0022-3514.93.2.234

17.

Leach

E. R.

(1964). Anthropological aspects of language: Animal categories and verbal abuse. In Lenneberg

E. H.

(Ed.), New directions in the study of language (pp. 23-63). Cambridge, MA: MIT Press.

18.

Leyens

J. Ph.

Paladino

M. P.

Rodríguez

R. T.

Vaes

Demoulin

Rodríguez

A. P.

Gaunt

(2000). The emotional side of prejudice: The role of secondary emotions. Personality and Social Psychology Review, 4, 186-197. doi:10.1207/S15327957PSPR0402

19.

Loughnan

Haslam

Kashima

(2009). Understanding the relationship between attribute-based and metaphor-based dehumanization. Group Processes & Intergroup Relations, 12, 747-762. doi:10.1177/1368430209347726

20.

Maass

(1999). Linguistic intergroup bias: Stereotypes-perpetuation through language. In Zanna

M. P.

(Ed.), Advances in experimental social psychology (Vol. 31, pp. 79-121). San Diego, CA: Academic Press. doi:10.1016/S0065-2601(08)60272-5

21.

Maass

Ceccarelli

Rudin

(1996). Ingroup bias: Evidence for in-group protective motivation. Journal of Personality and Social Psychology, 71, 512-526. doi:10.1037/0022-3514.71.3.512

22.

Marques

J. M.

Yzerbyt

V. Y.

(1988). The black sheep effect: Judgmental extremity towards ingroup members in inter- and intra-group situations. European Journal of Social Psychology, 18, 287-292. doi:10.1002/ejsp.2420180308

23.

Menegatti

Rubini

(2012). From the individual to the group: The enhancement of linguistic bias. European Journal of Social Psychology, 42, 36-40. doi:10.1002/ejsp.856

24.

Menegatti

Rubini

(2013). Convincing similar and dissimilar others: The power of language abstraction in political communication. Personality and Social Psychology Bulletin, 39, 596-607. doi:10.1177/0146167213479404

25.

Moscatelli

Albarello

Prati

Rubini

(2014). Badly off or better off than them? The impact of relative deprivation and relative gratification on intergroup discrimination. Journal of Personality and Social Psychology, 107, 248-264. doi:10/1037/a0036704

26.

Moscatelli

Albarello

Rubini

(2008). Linguistic discrimination in minimal groups: The impact of status differentials. Journal of Language and Social Psychology, 27, 140-154. doi:10.1177/0261927X07313652

27.

Moscatelli

Rubini

(2011). The impact of group entitativity on linguistic discrimination: Ingroup favouritism and outgroup derogation in the explanation of negative outcome allocations. Social Psychology, 42, 292-299. doi:10.1027/1864-9335/a000071

28.

Opotow

(1990). Deterring moral exclusion. Journal of Social Issues, 46, 173-182. doi:10.1037/0022-3514.93.2.234

29.

Pereira

Vala

Leyens

J. Ph.

(2009). From infra-humanization to discrimination: The mediation of symbolic threat needs egalitarian norms. Journal of Experimental Social Psychology, 45, 336-344. doi:10.1016/j.jesp.2008.10.010

30.

Prati

Menegatti

Rubini

(2014). Reducing linguistic outgroup derogation: A matter of multiple categorization and intergroup contact. Manuscript submitted for publication.

31.

Prati

Vasiljevic

Crisp

R. J.

Rubini

(in press). Some extended psychological benefits of challenging social stereotypes: Decreased dehumanization and a reduced reliance on heuristic thinking. Group Processes & Intergroup Relations.

32.

Rubini

Menegatti

(2008). Linguistic bias in personnel selection. Journal of Language and Social Psychology, 27, 168-181. doi:10.1177/0261927X07313653

33.

Rubini

Menegatti

(2014). Hindering women’s careers in academia: Gender linguistic bias in personnel selection. Journal of Language and Social Psychology, 33, 632-650. doi:10.1177/0261927X14542436

34.

Rubini

Menegatti

Moscatelli

(in press). Strategic use of language abstraction in achieving symbolic and practical goals: Evidence from intergroup, relational, organizational and political domains. European Review of Social Psychology.

35.

Rubini

Moscatelli

Albarello

Palmonari

(2007). Group power as a determinant of interdependence and intergroup discrimination. European Journal of Social Psychology, 37, 1203-1221. doi:10.1002/ejsp.391

36.

Rubini

Moscatelli

Palmonari

(2007). Increasing group entitativity: Linguistic intergroup discrimination in the minimal groups. Group Processes & Intergroup Relations, 10, 280-296. doi:10.1177/1368430207075156

37.

Schwartz

S. H.

(1992). Universals in the content and structure of values: Theoretical advances and empirical tests in 20 countries. In Zanna

(Ed.), Advances in experimental social psychology (Vol. 25, pp. 1-65). New York, NY: Academic Press. doi:10.1016/S0065-2601(08)60281-6

38.

Semin

G. R.

Fiedler

(1988). The cognitive functioning of linguistic categories in describing persons: Social cognition and language. Journal of Personality and Social Psychology, 54, 558-568. doi:10.1037/0022-3514.54.4.558

39.

Semin

G. R.

Rubini

(1990). Unfolding the concept of person by verbal abuse. European Journal of Social Psychology, 20, 463-474. doi:10.1002/ejsp.2420200602

40.

Staub

(1989). The roots of evil: Psychological and cultural origins of genocide and other group violence. New York, NY: Cambridge University Press.

41.

Struch

Schwartz

S. H.

(1989). Intergroup aggression: Its predictors and distinctness from in-group bias. Journal of Personality and Social Psychology, 56, 364-373. doi:10.1037/0022-3514.56.3.364

42.

Tajfel

(1982). Instrumentality, identity, and social comparisons. In Tajfel

(Ed.), Social identity and intergroup relations (pp. 483-507). Cambridge, UK: Cambridge University Press.

43.

Tajfel

Turner

J. C.

(1979). An integrative theory of social conflicts. In Austin

W. G.

Worchel

(Eds.), The social psychology of intergroup relations (pp. 33-47). Monterey, CA: Brooks-Cole.

44.

Turner

J. C.

Hogg

M. A.

Oakes

P. J.

Reicher

S. D.

Wetherell

M. S.

(1987). Rediscovering the social group: A self-categorization theory. New York, NY: Basil Blackwell.

45.

Vaes

Leyens

J. Ph.

Paladino

M. P.

Miranda

M. P.

(2012). We are human, they are not: Driving forces behind outgroup dehumanisation and the humanisation of the ingroup. In Stroebe

Hewstone

(Eds.), European review of social psychology (Vol. 23, pp. 64-126). London, England: Psychology Press. doi:10.1080/10463283.2012.665250

46.

Vaes

Paladino

M. P.

Castelli

Leyens

J. Ph.

Giovannazzi

(2003). On the behavioral consequences of infra-humanization: The implicit role of uniquely human emotions in intergroup relations. Journal of Personality and Social Psychology, 85, 1016-1034. doi:10.1037/0022-3514.85.6.1016

47.

Vaes

Paladino

M. P.

Leyens

J. Ph.

(2004). Perspective taking in an intergroup context and the use of uniquely human emotions: Drawing an E on your forehead. Revue Internationale de Psychologie Sociale, 17, 5-26.

48.

Vaes

Paladino

M. P.

Leyens

J. Ph.

(2006). Priming uniquely human emotions and the in-group (but not the out-group) activates humanity concepts. European Journal of Social Psychology, 36, 169-181. doi:10.1002/ejsp.279

49.

Vaes

Paladino

M. P.

Puvia

(2011). Are sexualized women complete human beings? Why men and women dehumanize sexually objectified women. European Journal of Social Psychology, 41, 774-785. doi:10.1002/ejsp.824

50.

Van Oudenhoven

J. P.

de Raad

Askevis-Leherpeaux

Boski

Brunborg

G. S.

Carmona

. . . Woods

(2008). Terms of abuse as expression and reinforcement of cultures. International Journal of Intercultural Relations, 32, 174-185. doi:10.1016/j.ijintrel.2008.02.001

51.

Walzer

(1983). Spheres of justice: A defense of pluralism and equality. New York, NY: Basic Books.

52.

Wenzel

Mummendey

Waldzus

(2007). Superordinate identities and intergroup conflict: The ingroup projection model. In Stroebe

Hewstone

(Eds.), European review of social psychology (Vol. 18, pp. 331-372). London, England: Psychology Press. doi:10.1080/10463280701728302

53.

Wohl

M. J. A.

Hornsey

M. J.

Bennett

S. H.

(2012). Why group apologies succeed and fail: Intergroup forgiveness and the role of primary and secondary emotions. Journal of Personality and Social Psychology, 102, 306-322. doi:10.1037/a0024838

54.

Yzerbyt

V. Y.

Rocher

Schadron

(1997). Stereotypes as explanations: A subjective essentialist view of group perception. In Spears

Ellemers

Haslam

S. A.

(Eds.), The psychology of stereotyping and group life (pp. 20-50). Oxford, UK: Blackwell.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.19 MB