The concept of stereotype has been employed in different disciplinary and theoretical contexts to cover a wide range of phenomena and serve manifold purposes, from psychoanalysis to Marxism to Derridean poststructuralism and postcolonialism. As noted in the Introduction, the approach to ethnic stereotypes adopted in this monograph derives from the field of social cognition. This may elicit skepticism in some quarters given the fraught relationship between biblical studies and anything that smacks of “psychology.” Gerd Theissen’s observation from 1983 that “every exegete has learned that psychological exegesis is poor exegesis”Footnote 1 has hardly been overhauled by time. The knee-jerk rejection of many New Testament scholars to psychology is, on one level, understandable and justified, because many early attempts to bring psychology and biblical studies into conversation were based on the “depth psychologies” of figures like Sigmund Freud and Carl Jung. In such approaches, the unconscious plays an important role, and the focus is on individual pathologies. Applying depth psychologies to biblical literature involved the exegete in psychoanalyzing the text’s author, which is, at best, a highly speculative endeavor.Footnote 2
The methodological problems associated with this type of research do not apply to cognitive approaches, which are concerned with how the human mind operates, not with the psyche of any individual person. Applying research in this area to the letters of Paul does not require us to speculate about the state of mind of the historical Paul but only to accept the basic premise that Paul was subject to the same mental processes as any other human being. That Paul lived in a very different time is of limited relevance; the way the human brain functions has changed over time, but within this evolutionary process the two millennia that separate us from Paul are but a drop in the bucket.Footnote 3 While stereotyped groups and the content of stereotypes change over the course of time and differ widely between cultures, there is no reason to think that the cognitive processes involved were substantially different in antiquity.
Social cognition is not a methodological panacea or the only way to approach the topic of this book, but I hope to demonstrate in this chapter that it usefully alerts us to possible ways in which ethnic stereotypes are reflected in early Christian literature and provides us with analytical language and concepts to try to explain these textual moments. This does not mean that every social cognitive study can be uncritically applied to the interpretation of early Christian texts. In what follows, I will focus on general insights related to basic cognitive processes and well-established research findings that are relatively uncontroversial.
The chapter will first turn to definitional questions related to stereotypes in general and ethnic stereotypes in particular (Section 2.1) and subsequently explore the manifold ways in which they exert influence on how human beings perceive, evaluate, and remember other people and social situations (Section 2.2). I conclude by discussing some preliminary implications for the study of Pauline and other early Christian literature.
2.1 Defining Stereotypes
Despite the pejorative connotations of the word “stereotype” in popular usage, from a cognitive point of view, a stereotype is in several important ways simply a form of categorization. We categorize people and objects all the time. This allows us to make sense of and interact with the world around us, which would otherwise be chaotic to the point of being incomprehensible. Categories allow us to classify disparate things under one header (e.g., as books, even though no two volumes are exactly the same). Categories thereby simplify the data available to our senses and facilitate our perception, as well as understanding and memory of the outside world.
Categorization is never, however, merely a matter of sorting things out. Categories also cue expectations. On the basis of the shared characteristics of the members of a category, we can draw inferences about the likely characteristics of a person or object. In this way, we can move beyond our immediate observations. Upon categorizing an object as a book, for instance, we deduce based on previous experience and prior knowledge that we will be able to open this object, that we will find letters or pictures inside, that we can write in it, that it is flammable, and so on. We can be reasonably sure about all of this without having to verify it. Moreover, once the object has been categorized as a book, we are able to form an idea of the ends to which and contexts in which it is likely to be used. For instance, in the case of a scholarly book, we will assume that it will be used for research purposes by a student or professor in a university library or in an office.
Social stereotypes function in much the same way. When someone is categorized as a lawyer, for instance, a set of expectancies is triggered about the appearance, personality traits, conduct, and activities of this person. By means of these expectancies, stereotypes help us to process the otherwise overwhelming complexity of the social world, allow us to predict what a person will be like, and help us determine how we should act in relation to them. Stereotypes are therefore an essential aspect of human interaction. They are ubiquitous and by no means limited to high-prejudice people. They are also not limited to specific situations (e.g., situations in which one group feels superior to another), nor are they necessarily deficient in the sense that they contain completely inaccurate information. From a social cognition perspective, “stereotypes are typically viewed as cognitive structures that contain our knowledge, beliefs, and expectations about a social group.”Footnote 4 They are “pictures in our head,” as Walter Lippmann put it in his classic work Public Opinion (1922).Footnote 5 These pictures or cognitive structures can be given verbal expression in oral or written communication.
The word “our” in “our knowledge, beliefs, and expectations about a social group” bears emphasis; stereotypes are culturally shared and refer to knowledge, beliefs, and expectations that are held by members of a (sub)culture, not just by select individuals. This implies a level of superficiality: stereotypes contain information that is known by a broad swath of people, not just by specialists. Stereotypes are also superficial in the sense that they are indiscriminately applied to a large group of people in a generalizing manner. They are inaccurate in this sense, but that does not mean that they are necessarily completely wrong. Shared knowledge, beliefs, and expectations about members of social groups can be quite accurate, but can also be misguided, and of course what is generally true of the group can be a deeply inaccurate assessment of individual members of that group.
For the purposes of the study of antiquity, the culturally shared nature of stereotypes means that the existence of a stereotype must be established on the basis of multiple sources. In order to qualify as a stereotype, it has to be attested by several, ideally independent texts. Unanimity is not, however, required. There is no question that in present-day contexts, stereotypes are operative that a substantial part of the population does not believe to be true. As will be explored in more detail below, this does not mean that these stereotypes have ceased to exert influence, not even on the people that dispute their accuracy.
This has to do in part with the fact that people do not actively choose to develop such expectancies. Our brains create stereotypes on the basis of personal experience and of what we are told by others, including parents, friends, media, and other socializing agents. The contents of stereotypes are cultural products. Stereotypes are accordingly culture-specific and naturally subject to change; how individuals and groups are stereotyped (i.e., how they are grouped and which traits are assigned to them) varies over time, even as the process of stereotyping itself remains generally consistent, as noted above.
Stereotypes are not necessarily negative in content. It is true that we commonly rate outgroups more negatively than ingroups, but this does not mean that stereotypes about outgroups are uniformly negative.Footnote 6 Groups are often given opposite scores in the categories of competence and warmth. In many Western cultures, old people, disabled people, and Black people are stereotypically seen as incompetent yet warm, whereas Jews, Asians, and professionals are regarded as competent yet lacking warmth. A more limited number of groups are usually viewed as both competent and warm (e.g., middle-class white families) or incompetent and lacking warmth (e.g., drug addicts).Footnote 7 We can observe similar patterns in antiquity, for instance, with regard to “skill and intelligence,” on the one hand, and “spirit,” on the other hand. According to Aristotle, “The peoples of cold countries generally, and particularly those of Europe, are full of spirit, but deficient in skill and intelligence.” The opposite score is given to the inhabitants of Asia: “The peoples of Asia are endowed with skill and intelligence, but are deficient in spirit.” To no one’s surprise, Aristotle believed that “the Greek race … unites the qualities of both sets of peoples. It possesses both spirit and intelligence.”Footnote 8 The Greeks are uniquely marked by positive traits, whereas the assessment of other groups is mixed.
Ethnic Stereotypes
Based on the preceding, ethnic stereotypes can now be defined as a subset of stereotypes that refer to culturally shared knowledge, beliefs, and expectations about members of an ethnic group. But what counts as an “ethnic group”? Here too, it is useful to approach the matter from a cognitive perspective. Rather than thinking of ethnic groups as objectively observable, substantial things-in-the-world, it is arguably more accurate to see them as “collective cultural representations, as widely shared ways of seeing, thinking, parsing social experience, and interpreting the social world.”Footnote 9 This perspective highlights the importance of perception and categorization. Ethnic groups do not exist independently in the world but are created and maintained by people’s categorizations of the world. This is not to say that ethnic groups are figments of the imagination; it is rather a way of explaining why such groups can be very real even if membership is not grounded in any objectively observable facts. If people think of themselves or others as belonging to an ethnic group and that assessment is sufficiently widely shared, the group is a relevant social category. Since people’s categorizations of the world may differ and develop over time, the boundaries and existence of ethnic groups are malleable and negotiable rather than facts of nature.
The criteria for identification of an ethnic group are likewise contingent.Footnote 10 The Harvard Encyclopedia of American Ethnic Groups lists no fewer than fourteen indicia of ethnic identity, including common geographic origins; migratory status; race; language or dialect; religious faith or faiths; ties that transcend kinship, neighborhood, and community boundaries; shared traditions, values, symbols; literature, folklore and music; food preferences.Footnote 11 The editors state that all of the groups included in their encyclopedia are “characterized by some of [these] features, although in combinations that vary considerably.”Footnote 12 In this and similar approaches to ethnicity, there is no single feature that functions as a sine qua non for identification of a group as ethnic.Footnote 13 While this accurately reflects the wide variety of features that can be invoked in conceptualizations of ethnic groups, such a broad conception is of limited analytical use. As Rogers Brubaker has noted, “it may be that ‘ethnicity’ is simply a convenient – though in certain respects misleading – rubric under which to group phenomena that, on the one hand, are highly disparate, and, on the other, have a great deal in common with phenomena that are not ordinarily subsumed under the rubric of ethnicity.”Footnote 14
Some scholars have proposed a more restrictive definition. In his influential work on ethnic identity in Greek antiquity, Jonathan Hall argues for a “somewhat narrow definition of ethnicity” that limits ethnic groups to social entities with “a putative subscription to a myth of common descent and kinship, an association with a specific territory and a sense of shared history.”Footnote 15 Such an approach is more useful for heuristic purposes, but it comes with the trade-off that it does not always accurately reflect the manifold ways in which participants have constructed and identified ethnic identities across different contexts.
Even Hall’s more restrictive definition of ethnicity can be difficult to apply to ancient texts, which are rarely clear and explicit about the grounds on which group identity is ascribed. This frequently renders it difficult to determine if a group is understood to possess the characteristics that Hall has identified as central to ethnicity. Possibly ethnic groups are not always identified with γένος, ἔθνος, genus, or similar terms, and even if they are, the use of these words is not necessarily very helpful since their semantic range is so capacious that they cannot form the basis for any overall taxonomy that is generally applicable, beyond a single text or author.Footnote 16
To illustrate the difficulty in operationalizing Hall’s criteria, consider the claim made in Titus 1:12 that “Cretans are always liars.” Are “ethnic” Cretans (in Hall’s sense) in view, in which case some people on the island are not necessarily liars, because their genealogical roots lie elsewhere? Or is the reference to people who inhabit the island, to the possible effect that Cretans who moved elsewhere may have shed their mendacious ways? Or does the author simply assume that all inhabitants of Crete are genealogically Cretan and that no Cretan has ever left the island’s territory? It is perhaps also possible that “Cretans” must be construed as a cultural identity: people who live a “Cretan” lifestyle. To raise these issues is to ask questions that were evidently not of any particular concern to the author.
If the label “Cretan” presents us with difficulties, the interpretive challenges are at least similarly daunting in the case of other categories, such as Ioudaioi (Jews/Judeans), an identification that could involve genealogy, geography, and culture as well as religion and cult. Roman identity could likewise be based around genealogy and geography (the inhabitants of the city of Rome and/or its empire), denote a predominantly cultural identity, or serve as a political identifier (a Roman citizen). Greek identity, again, could revolve not (only) around genealogy or geography but also around participation in Hellenic culture and the use of the Greek language.Footnote 17 Further complicating the matter is that most of these categories overlap and intersect in various ways. It is, for instance, often difficult to separate religion from culture, and both are closely related to shared history and genealogy.
Even if we can determine with a reasonable degree of certainty, in some instances, on which grounds ancient authors identify a group or person as Jewish, Roman, Greek, Cretan, etc., it cannot be assumed that the same perspective applies in other instances. For some ancient authors, Greek identity was a matter of shared descent, but others associated it with the region of mainland Greece, and yet others saw Greek identity as something obtainable by paideia alone.Footnote 18 Such variety is at times present even within the work of a single author. Jerome, as noted in the Introduction, referred to the Galatians both as a people (gens) and as the inhabitants of a province or region, and there is also considerable slippage in his work between “Galatians” as a referent to the general population and to the Christ-believers among them.Footnote 19
Excluding usages that do not meet the criteria of Hall’s more restrictive definition of ethnicity is not only impracticable, it would be detrimental to the analysis, as it would impose distinctions that ancient texts and audiences did not (consistently) make. Such an approach would amount to cordoning off a great deal of evidence that, in terms of ancient perceptions of the groups in question, is highly relevant and would obscure from view significant discursive shifts such as those just noted with respect to Jerome.
Ethnic group is, in short, a complex category for three, partially interrelated, reasons: (1) Ethnic groups are not objectively observable “things in the world” but a function of people’s classificatory activities, which differ and develop over time, rendering ethnic (group) identity unstable; (2) these classificatory activities can be based around a wide variety of variously constellated criteria; (3) it is frequently difficult to know which criteria inform ancient authors’ classification of a group, and hence difficult to determine to what extent they think of these groups in ways that would fit various modern definitions of “ethnic.” For these reasons, blanket pronouncements about which people do and do not constitute an ethnic group are often too simplistic. In this study, I will instead discuss on a case-by-case basis if and how the groups in question are construed as ethnic (or not) in different contexts. As we will see, all of the groups selected for case studies were construed in some contexts and some respects in ways that match even Hall’s restrictive definition of ethnicity.
I should clarify that I do not wish to suggest that these case studies are the only or best examples of ethnicity in Paul’s letters or that Pauline authors had a consistent notion of ethnicity that is most effectively elucidated by looking at the selected groups. Rather, I suggest that the groups discussed in the study can be generally categorized under the rubric of ethnic group. Although a different categorization would not materially affect the exegetical argument of the individual chapters that follow, the category of ethnic group draws attention to important commonalities between these groups as well as to the relation between this set of Pauline texts and claims made about ethnic and racial groups in later eras.
The difference in size of these groups is not particularly relevant to this discussion. The case studies collected in this monograph feature groups that operate on different levels of scale, but this does not necessarily mean that one is more “ethnic” than the other, for ethnic identity can operate on multiple levels, and multiple ethnic identities can coexist alongside each other.Footnote 20 Like today, multiple ethnic identities were arguably the rule rather than the exception in antiquity. Irad Malkin discusses the fascinating example of a citizen of Syracuse who, depending on context, could be identified as “Syracusan,” “Siceliot,” “Dorian,” or “Greek.”Footnote 21 Paul himself may well have thought of his identity variously as “Jewish,” “Israelite,” “Hebrew,” “Tarsian,” “Cilician,” “Roman,” and perhaps also “Greek.”Footnote 22 Paul’s example effectively illustrates that Roman citizenship did not require the rejection of other identity markers. Those who were Roman citizens did not necessarily regard “Roman” as their primary ethnic identity. Indeed, Acts claims that Paul was a Roman citizen but has him identify himself not as “a Roman” but as “a Jew, a Tarsian from Cilicia, a citizen of an important city” (21:39).Footnote 23
The reality of multiple identities does not rule out ethnic stereotyping. People face stereotypes based on some aspect of their identity that is selected as relevant by the stereotyping party. Whether the stereotyped party agrees with the relevance or accuracy of this identification is not important.Footnote 24 This dynamic also explains why stereotypes exist about the inhabitants of multiethnic cities, like today’s Paris or ancient Corinth. Demographic diversity did not stop ancient authors from thinking about the populations of multiethnic communities in generalizing and essentializing ways. Nor is this form of cognitive dissonance restricted to antiquity. Indeed, there is no such thing as a purely “autochthonous” population; local identities are always the result of intermixing and outside influences, but this does not preclude essentialist rhetoric.
Finally, a word about the relationship between ethnicity and race. The definition of both of these terms is contested and therefore, necessarily, so is the relationship between them. Whereas some, especially in the US context, regard the two as functionally synonymous, others argue for distinct definitions. The difference is sometimes construed in terms of culture (ethnicity) versus biology (race) or insider perspective (ethnicity) versus outsider designation (race). Based on the preceding discussion, it will be apparent that such clear-cut distinctions are difficult to sustain. Ethnicity is frequently used with reference to outgroups and on most definitions encompasses aspects of biology or at least genealogy. Another way of distinguishing the two terms is to take race as a modern, post-Darwinian “scientific” category, with the corollary that it would be anachronistic to use the term in the study of antiquity. Ethnicity, however, is likewise a modern category, and using modern concepts in analyzing and describing ancient evidence is not necessarily problematic, as long as they are clearly defined. Indeed, rendering the past intelligible in contemporary terms is arguably at the heart of the historian’s task. It could, moreover, be argued that avoiding the language of race in the study of antiquity obscures significant similarities between ancient and modern discourses, rendering it needlessly difficult to recognize the throughlines that connect the present with the past. I will make no attempt to resolve these and other related definitional questions within the confines of this study.Footnote 25 For present purposes, I will use ethnic group as the default term, reserving the language of race primarily for the reception-historical sections of each chapter, which generally focus on the nineteenth century, a period when race was a key concept in the academy as well as society at large. I have also retained the word “race” when it appears in translations and interpretations of ancient sources. By doing so, I hope to draw attention to how closely connected race and ethnicity are in many contexts – more specifically, that regardless of any conceptual distinctions that we may wish to impose, in terms of their historical development, notions of race and ethnicity are thoroughly intertwined and overlap and intersect in interesting and important ways that still await more thorough analysis.
2.2 Stereotypes at Work
Having established a general conceptualization of ethnic stereotypes, the present section will explore in more detail when and how they impact social interactions. Stereotypes exert much more influence than people generally recognize. This is partly a function of pejorative associations of the word “stereotype,” which makes people hesitant to admit to their use. More importantly, however, people underestimate the use of stereotypes because it can take place largely outside of awareness. When we encounter a person or the symbolic equivalent of their presence (like a description), we instantly categorize them in terms of social role, gender, ethnicity, and so on, and this triggers stereotypes associated with those categories. Since this takes place instantly, we may not even notice it, a process sometimes referred to as implicit stereotyping or implicit bias; people are categorized and this triggers a set of expectancies even if perceivers are unaware of this.Footnote 26
If stereotypes are cued outside of awareness, this raises the question of whether we are doomed to engage in stereotyping. In order to get at an answer to this question, it is important to recognize that stereotyping involves several cognitive steps. The first step is identification of the attributes of a person, followed by association of those attributes with a social category (in this case, an ethnic group). The third step is the mental activation of the characteristics that are linked to this category, that is, the cuing of the stereotype.Footnote 27 Finally, in order for stereotyping to occur, the stereotype must be used, that is, it must impact our impressions, affects, judgments, or actions.
The first three steps (identification, categorization, and stereotype activation) take place largely automatically. As Gordon Moskowitz notes, summarizing the outcomes of several earlier studies:
As an automatic process, stereotype activation would occur without the … observer intending it, without the observer being constrained by cognitive load and limits to processing resources, without his/her being aware of spreading activation, and regardless of whether he/she is an egalitarian person. Either a high- or low-prejudice observer would have the stereotype triggered by virtue of the fact that each type of person knows the stereotype. Personal beliefs that lead to either acceptance or rejection of the stereotype would play no part in the triggering of the stereotype. All that would matter is that the observer has learned the association between the group label and the stereotype that exists in the culture.Footnote 28
Stereotype activation is thus largely automatic, but does the stereotype, once activated, necessarily get used, that is, does it impact our impressions, affects, judgments, or actions? To answer this question, it is useful to think of stereotypes as hypotheses: A stereotype hypothesizes that a person who has been categorized shares the perceived characteristics of other members of the category. It is possible for people to carefully test this hypothesis and reject it if it seems wanting. If so, even though the stereotype has been activated (step 3 above), it is not put to use (step 4) or its use is corrected. It requires considerable effort, however, to keep a stereotype from being put into use.Footnote 29
To a large extent, this has to do with our remarkable ability to identify confirming data points. Once an hypothesis is in place, humans are liable to develop tunnel vision: We easily notice what supports the hypothesis and do not register contradicting data as well. This is sometimes referred to as confirmation bias or “myside” bias and has been described as “ubiquitous” and as a “built-in feature of the mind” that functions “independently of subjects’ intelligence, cognitive ability, or motivation to avoid it.”Footnote 30 Because of a preference for easy processing, we filter out whatever data does not fit our stereotype (ethnic or otherwise) and highlight what confirms it. Our perception is skewed as a result. As David Hamilton and Terrence Rose have demonstrated, the “illusory correlation” that a stereotype posits between membership of a social category and certain traits leads us to believe that there is more evidence supporting the stereotype than has been observed. We tend to believe that stereotype-consistent behaviors occur more often than they actually do, and we tend to underestimate the prevalence of stereotype-inconsistent behaviors.Footnote 31 Applying this to Paul raises the possibility that he would have more readily recognized behavior and statements that confirmed stereotypical expectations he may have had about Gentiles, Galatians, Corinthians, etc., than any information that was inconsistent with the stereotype.
A lot of behavior is not unequivocally consistent or inconsistent with stereotypes but can be interpreted in various ways. In such ambiguous cases, the availability of a stereotype can guide our interpretation of the evidence. This has been strikingly demonstrated in a famous experiment by Birt L. Duncan, who found that white people observing a white person shoving a Black person labeled this as a violent act only 17 percent of the time, but when the roles were reversed (a Black person shoving a white person) this was seen as violent 75 percent of the time.Footnote 32 Another classic study, by John Darley and Paget Gross, offered further evidence of the way in which stereotypes impact perception.Footnote 33 In this study, participants were asked to judge the academic ability of a little girl. Some participants were informed that she lived in an affluent neighborhood, while others were told that she came from a poor neighborhood. When asked to judge her academic ability without receiving any further pertinent information, participants’ ratings were not biased. However, when participants were shown a video of the girl taking a test on which she performed ambiguously (answering both easy and difficult questions sometimes right and sometimes wrong), participants did use the stereotypical information derived from their knowledge of her socioeconomic status to rate her academic ability. She was judged to have above average academic abilities by those who thought she came from an affluent neighborhood and below average abilities by those who believed she lived in a poor neighborhood. What this experiment demonstrates is that even among people who were unwilling to judge others on the basis of stereotypes (“the rich are smart, the poor are not”), stereotypes still heavily influenced their perception and assessment. Moreover, these and other experiments show that if evidence can be interpreted as supporting a stereotype, that is exactly how most people will interpret it.Footnote 34
In addition to very strong evidence for the influence of stereotypes on the perception and assessment of people and behaviors, a number of studies suggests that our recall of past encounters and situations can be influenced by stereotypical notions as well. This would mean that even if our initial perception was not significantly skewed and information was non-stereotypically encoded in memory, our retrieval of that information is subject to biasing mechanisms. In much the same way that we tend to perceive stereotype-consistent information better, some studies suggest that we recall it more easily than stereotype-inconsistent information.Footnote 35 The relevance of these insights to the study of Paul’s letters is evident: Memory and recall played an important role in the composition of Paul’s letters as this process involved the apostle’s memory of assemblies that he had left some time ago and, in some cases, the memory of coworkers who informed him (after a journey that took days or weeks) about what had taken place since Paul’s departure.
It will be clear by now that avoiding the influence of stereotypes is no small feat. Once the stereotype has been triggered (and again, this can happen automatically) we tend to interpret available evidence in such a way that it confirms the stereotype. As a result, our conscious beliefs about stereotypes and any egalitarian values that we may hold matter less than we might like to think. If we are familiar with a stereotype, it will be triggered and we are likely to find or construe evidence to support it. We may believe that our assessment is based solely on the behavior that we have observed and is therefore “unbiased,” but our perception of this evidence has been skewed under influence of the stereotype. In the study by Darley and Gross, participants believed to have based their assessment on the girl’s performance on the test; and in Duncan’s study, white observers believed to have labeled the behavior of a Black person as aggressive based on the action that they witnessed.Footnote 36 Yet these studies show that in both cases a stereotype guided their assessments. This suggests that if a person’s assessment of the actions of a group or individual aligns closely with stereotypical views, it is reasonable to hypothesize that this overlap may be at least partly explained as the result of confirmation bias. For example, if someone believes that a Black individual is lazy, there is a decent chance that this assessment is not based solely or even at all on the behavior of the individual in question but on the pernicious stereotype of Black laziness. This result should be taken into consideration in the study of Paul’s letters and other ancient texts; if the depiction of the addressees or other characters closely aligns with stereotypical views concerning the ethnic group to which they are assigned, it is reasonable to hypothesize that this depiction is based in whole or in part on a stereotype.
While stereotyping may be difficult to avoid, the situation is not entirely hopeless. Many people today want to correct for bias because of egalitarian ideals. In such cases, it is necessary that people are aware of biasing influence, can accurately assess how it has skewed their impressions, and have the cognitive resources (sufficient time, focus, ability, and volition) to correct their views. Research shows that people who have chronic low-prejudice beliefs or have undergone substantial training may be able to inhibit the impact of stereotypes in this way.Footnote 37 But unless a concerted effort is made to avoid it, stereotypes exert considerable influence on our perceptions, as well as assessments and memory. As noted in the Introduction, there is no substantial ancient evidence of objections to ethnic stereotypes, let alone of concerted efforts to inhibit their influence.
I note, finally, that just as there are factors that may inhibit stereotype usage, there are factors that encourage their use and impact. People are even more likely than normal to engage in stereotyping if our mental capacity has been reduced by time pressure or emotions like anger. Stereotyping, especially negative stereotyping, also increases when we have had a recent experience of failure or, more generally, when we are feeling low and wish to reaffirm our self-worth. In such instances, we sometimes try to feel better about ourselves by derogating others. This mechanism is especially effective if the blow to the self-esteem came from someone who can be stereotyped in such a way that the credibility of the negative evaluation can be questioned.Footnote 38 These research results too may be relevant to the interpretation of ancient literature, including the letters of Paul, whose authority and credibility appear to have been regularly put into question, and who seems rather reproachful in Galatians and elsewhere.
Contact Hypothesis
An obvious counterargument to what has thus far been said about the ubiquitous impact of ethnic stereotypes and, in particular, the applicability of these insights to the study of Paul, is that Paul was not generally writing to strangers. Common sense has it that regardless of the initial impact of stereotypes, they will quickly diminish in influence once people get to know each other and are confronted with stereotype-inconsistent evidence. It would indeed seem reasonable to assume that people are more likely to think in stereotypical ways about strangers than about those they are acquainted with. In the latter case, the influence of stereotypes should be limited by the abundance of concrete, individuating data that has become available. While some early evidence suggested that this is indeed the case, subsequent studies have shown the matter to be more complex.Footnote 39 In one study, participants read about John, who was either presented as an accountant or as a construction worker.Footnote 40 The latter are stereotypically viewed as more aggressive, but once the participants got to know John a bit by hearing about the gentle manner in which he responded to someone who taunted him about his marriage and ridiculed his wife, they accurately described John as unaggressive, regardless of whether they thought he was an accountant or a construction worker. However, when asked who was more likely to get into a fight at a bar, make rude comments to women, or punch someone, participants still thought John would more readily engage in this behavior if he was presented to them as a construction worker, than if they thought that he was an accountant. This outcome suggests that it is not the case that stereotypes cease to influence our perception and assessment past the initial moment of contact or exposure. As Ziva Kunda concludes: “Our stereotypes may shape our reactions to stereotyped individuals even when these individuals are familiar to us.”Footnote 41
Research on the so-called contact hypothesis further bears out this impression. The contact hypothesis posits that if people come into contact with a stereotyped group, the stereotypes they hold about them will disappear. But to once again quote Ziva Kunda:
Extensive research on the contact hypothesis gives little reason for optimism. For contact to reduce prejudice, the members of the two groups must have equal status, an opportunity to get to know each other, exposure to evidence that disconfirms the stereotypes, shared goals, and active cooperation. And even then, contact is not always effective at reducing prejudice. Negative stereotypes often resist change even in the face of intense manipulations involving cooperation with members of the stereotyped group over long periods of time.Footnote 42
Moreover, even if stereotypes undergo change, the process takes place according to “the anchoring and adjustment heuristic,” which refers to the phenomenon that people’s perception is often based to a considerable extent on initial impressions (the “anchor”), which are routinely impacted by stereotypical expectations. When faced with individuating information that does not fit the stereotype, the perception of the person in question is “only adjusted until [participants] reached a plausible conclusion, not necessarily an accurate conclusion.”Footnote 43
In short, the simple fact that Pauline authors were familiar with many of the people they addressed does not mean that stereotypes did not exert influence on their perception and evaluation of them. Also relevant in this connection is the simple fact that these letters evidence by their very existence a lack of contact. Paul was physically separated from his communities and did not have direct access to them; he relied on letters and word of mouth for information about his friends and foes. Paul did not directly observe any stereotype-inconsistent data when he composed his letters.
Stereotype-Inconsistent Information
Even if unambiguous stereotype-inconsistent information is directly available, we often fail to register it adequately. It was previously noted that we tend to perceive and interpret behavior in such a way that it confirms to the stereotype (confirmation bias). We have also already noted that we underestimate the prevalence of stereotype-inconsistent information pp. 30–31. This ties in with the more general phenomenon that it is relatively difficult for people to perceive differences among members of a group. People are liable to exaggerate the similarities within a group as well as the differences between groups. As a result, once we assign a person to a social category (and we do this very quickly), it becomes harder for us to perceive any differences between that individual and other members of the group.
But what happens when, despite all this, we do notice unambiguously stereotype-inconsistent information? A good deal of empirical evidence shows that people are much more ready to explain away a considerable amount of stereotype-inconsistent information than to question or substantially alter the stereotype. For instance, in cases where an individual’s actions are clearly stereotype-inconsistent, we tend to interpret this as flawed evidence or as an exception to the rule rather than as evidence that the stereotype about the group is incorrect. Sometimes this involves the creation of subtypes. Individuals who exhibit stereotype-inconsistent behavior are placed in a separate group, which allows perceivers to retain stereotypes about the broader group.Footnote 44
Another mechanism used to retain stereotypes in the face of contradictory evidence is to attribute it to the situation or circumstances rather than to the personality of the person exhibiting said behavior. And if it is seen as a personality trait, perceivers sometimes distinguish between the inner and outer person and label the inconsistent behavior as reflective of the outer person, rather than of the true, inner person of the individual exhibiting that behavior.Footnote 45
The stable nature of stereotypes and the cognitive effort we invest in maintaining them may suggest that stereotypes will always remain the same. That is fortunately not the case. Stereotypes are subject to change, and over the last century or so, many ethnic stereotypes have become less pejorative. Yet this is a process that can be measured in decades rather than days or weeks.Footnote 46 And importantly in the present connection, these changes are the result of a conscious effort to avoid stereotyping, which is a modern concern without parallel in antiquity.
Conclusion
Over the course of this book, starting with Chapter 3 on Paul’s depiction of Gentiles, we will put to interpretive use insights derived from research in social cognition as discussed in the preceding pages. At this point, it may be helpful to highlight some preliminary implications for the study of early Christian literature, in addition to some of the quick asides already made above. Of first-order importance is that stereotyping is a ubiquitous phenomenon that is by no means restricted to high-prejudice people but affects all human beings. It is so central to the way human beings process the outside world that we can safely posit that Paul and other early Christ-followers were influenced by stereotypes as well. Secondly, while common sense may suggest that stereotypes are only at work in contexts where there is an information deficit (i.e., where people do not know much about the person or situation they are facing), empirical evidence shows that stereotypes can play a role even when people are familiar with each other. For the purposes of the present study, this means that we cannot conclude that stereotypes did not impact the interactions between, for example, Paul and the Corinthians, simply because Paul had previously visited Corinth.
Another important implication has to do with the considerable degree to which stereotyping takes place automatically and proceeds outside of awareness, to the effect that it often impacts the perception and judgment of low-prejudice people as well.Footnote 47 It follows that the question of whether early Christian authors espoused egalitarian ideals does not settle the question of the impact of stereotypes. Even if Galatians 3:28’s famous dictum “there is neither Jew nor Greek” and similar statements are taken as manifestos of social equality, it does not follow that this somehow sheltered early Christians from the influence of ethnic stereotypes.Footnote 48
The unconscious impact of stereotypes suggests that we should not restrict our inquiry to moments in which stereotypical notions are expressly stated but should also consider evidence that may indirectly betray their influence. Duncan’s experiment and many other studies showed stereotypes to be clearly at work despite the belief of the participants that they based their assessment strictly on the behavior they witnessed. This suggests that a stereotype-consistent assessment of an individual may very well be reflective more of the stereotype than of the individual in question. For our purposes, this means that if a Pauline text describes an ethnic group (or its individual members) in ways that closely align with a culturally available stereotype, it is plausible that this depiction was influenced by that stereotype.
Another implication to be noted here is that we should not be surprised to find inconsistencies in the portrayal of ethnic groups in ancient authors. Stereotypes have proven able to withstand a considerable amount of stereotype-inconsistent data. It follows that we cannot conclude that, for example, the stereotype that people from Pontus are uncivilized and backward has been discarded if we find evidence of Pontics who are learned and sophisticated. Or to give another example: While Galatians may sometimes be depicted as constant in their resolve, this does not necessarily indicate that the stereotype of Galatian fickleness has ceased to be operative. Methodologically, this means that in reconstructing ancient ethnic stereotypes, we must give priority to statements that express stereotypes in direct ways rather than to incidental descriptions of their conduct. If the latter are inconsistent with the stereotype, this does not necessarily mean that the stereotype is no longer current.
A final methodological point concerns the unicity of stereotypical character traits. Contemporary stereotypes of ethnic groups exhibit a good deal of overlap. A penchant for criminality, for instance, is stereotypically attributed to various ethnic groups in many cultures. It is obviously not the case that, for example, the stereotype that Roma are criminals does not exist because other ethnic groups are viewed similarly.Footnote 49 Likewise, in the study of ancient literature, it cannot be assumed that, for example, Corinthians were not stereotypically considered sexually licentious on the grounds that this charge was leveled at other groups as well. It simply means that it was too attractive a charge to limit to one group. What must be established is that a trait was widely seen as characteristic of the group. To insist that in order for a stereotype to be identified it must be unique to that ethnic group is to place the evidentiary bar impossibly high and does not do justice to how stereotypes operate.
Open access