Psychology

For 150 years, a basic belief in the universality of emotion has been a cornerstone of modern psychology; indeed, in a recent survey of current researchers in the field, 88 percent of affective scientists endorsed the statement “there is compelling evidence for universals in any aspect of emotion” (Ekman, Reference Ekman2016, p. 32). However, this statement belies the fierce theoretical disagreements about emotion that occur in psychology, and the devil is in the details. While most psychologists probably agree that the fundamental capacity for emotion is a human universal, there have been intense conflicts regarding more specific issues, most notably whether individual emotions themselves universally occur in human populations. Those on the pro-side of the debate generally maintain that we can point to a set of specific human emotions that are meaningfully universal – that is, they consistently produce observable manifestations – which are thought to have developed as adaptive responses to frequently reoccurring selection pressures in the evolutionary history of the species.

This universalist position is most readily advocated by a family of basic or discrete emotion theories (Shiota, Reference Shiota2024). These theories are anchored in the work of Darwin (Reference Darwin1872), who famously examined facial expressions corresponding to discrete emotional states: “I have endeavoured to show in considerable detail,” he writes, “that all the chief expressions exhibited by man are the same throughout the world” (p. 361). In the mid twentieth century, Darwin’s analysis was influentially revived by Tomkins (Reference Tomkins1962), who largely inaugurated the modern research tradition on discrete, universal emotions (see also Plutchik, Reference Plutchik1962). Tomkins argued that there were nine universal emotions, reflected in innate affect programs – subcortical structures that, when activated, reliably trigger a pattern of motivational, behavioral, and physiological responses, including displays of the stereotypical facial patterns investigated by Darwin. Indeed, the presumed universality of emotional facial expression was central to Tomkins’s theory, so his students began to seek empirical evidence concerning the cross-cultural facial recognition of emotion (Ekman, Reference Ekman1971; Izard, Reference Izard1971); these researchers (most notably Ekman) would go on to develop the so-called Basic Emotion Theory (BET), the central theoretical framework that posits the natural existence of a small number of certain categorically discrete, universal human emotions, such as happiness, fear, anger, sadness, disgust, and surprise (Cordaro, Reference Cordaro, Al-Shawaf and Shackelford2024; Ekman, Reference Ekman1992, Reference Ekman, Dalgleish and Power1999; Ekman & Cordaro, Reference Ekman and Cordaro2011; Levenson, Reference Levenson1999, Reference Levenson2011; Shiota, Reference Shiota2024).

As we will see further in Section 3, BET has evolved considerably over the past half-century, but advocates generally agree that basic emotions are biologically universal evolutionary adaptations that are “physiologically, neurologically, functionally, and behaviorally distinct from one another” (Cordaro, Reference Cordaro, Al-Shawaf and Shackelford2024, p. 5). Historically, much BET research has focused on the facial expression of emotion; decades of cross-cultural analysis has led theorists to conclude that each basic emotion triggers a stereotypical pattern of universal facial movements that can be recognized by people across the world at a much higher rate than chance alone (Elfenbein & Ambady, Reference Elfenbein and Ambady2002; Ekman, Reference Ekman1993; Ekman et al., Reference Ekman, Friesen and O’Sullivan1987; Izard, Reference Izard1994; Witkower, Rule, & Tracy, Reference Witkower, Rule and Tracy2023). Evidence suggests that other forms of affective social communication may also be universally recognized, such as bodily expression (Witkower et al., Reference Witkower, Hill, Koster and Tracy2021) and nonverbal vocalizations (Juslin & Laukka, Reference Juslin and Laukka2003; Laukka & Elfenbein, Reference Laukka and Elfenbein2021; Sauter et al., Reference Sauter, Eisner, Ekman and Scott2010, Reference Sauter, Eisner, Ekman and Scott2015). Basic emotion theorists equally argue that discrete emotions characteristically inspire certain response patterns in the body’s autonomic nervous system (ANS); particular basic emotions are thought to generate specific ANS responses that can be to some extent distinguished from one another (Ekman, Levenson, & Friesen, Reference Ekman, Levenson and Friesen1983; Kreibig, Reference Kreibig2010; Levenson, Reference Levenson, Ekman, Heider and Friesen1992, Reference Levenson, Davidson, Scherer and Goldsmith2003; Levenson, Ekman, & Friesen, Reference Levenson, Ekman and Friesen1990; Levenson et al., Reference Levenson, Ekman, Heider and Friesen1992; but see Levenson, Reference Levenson2014 for methodological issues). In general, most BET theorists have historically maintained that the central emotions of human life have some sort of discrete, observable universal signatures that are biologically grounded in the body.

As mentioned, basic and discrete theories of emotion maintain that particular emotions are universal biological adaptations, so evolutionary psychology (Buss, Reference Buss2015; Cosmides & Tooby, Reference Cosmides and Tooby2013; Tooby, Reference Tooby2020) provides complementary evidence for the universality of emotion. (Though, to be clear, many nonuniversalist theories of emotion equally ground their models in evolutionary theory, so it is important not to simply conflate an evolutionary approach with a universalist, discrete emotion approach.) Put simply, researchers working in this mode view emotions as universal adaptations of our species’ mental architecture, which emerged in the long course of mammalian development as responses to increase fitness in light of certain ubiquitous situations (Cosmides & Tooby, Reference Cosmides, Tooby, Lewis and Haviland-Jones2000; Keltner, Haidt, & Shiota, Reference Keltner, Haidt, Shiota, Schaller, Simpson and Kenrick2006; Nesse, Reference Nesse1990; Nesse & Ellsworth, Reference Nesse and Ellsworth2009; Plutchik, Reference Plutchik2001, Reference Plutchik2003; Tooby & Cosmides, Reference Tooby and Cosmides1990a, Reference Tooby, Cosmides, Lewis, Haviland-Jones and Barrett2008, Reference Tooby, Cosmides and Buss2015; Tracy, Reference Tracy2014; see also J. H. Turner, Reference Turner1996, Reference Turner2021). Many evolutionary approaches to emotion (and psychology more broadly) locate discrete emotions within human nature – that is, “the evolved, reliably developing, species-typical computational and neural architecture of the human mind and brain” (Cosmides & Tooby, Reference Cosmides, Tooby, Lewis and Haviland-Jones2000, p. 91; see Tooby & Cosmides, Reference Tooby and Cosmides1990b) – and thus universality consequently follows, as further suggested by studies that analyze analogous emotion-like expression in our evolutionary ancestors (Kret et al., Reference Kret, Prochazkova, Sterck and Clay2020; Parr et al., Reference Parr, Waller, Vick and Bard2007; Vick et al., Reference Vick, Waller, Parr, Smith Pasqualini and Bard2007). A number of discrete emotions have been analyzed in terms of their universal adaptive functions, such as fear (LeDoux, Reference LeDoux2012); jealousy (Buss, Reference Buss2018), anger (Sell, Tooby, & Cosmides, Reference Sell, Tooby and Cosmides2009), pride (Cheng, Tracy, & Henrich, Reference Cheng, Tracy and Henrich2010), and compassion (Goetz, Keltner, & Simon-Thomas, Reference Goetz, Keltner and Simon-Thomas2010).

Beyond evolutionary psychology, some researchers in the field of cross-cultural psychology aim to detect not only variants in mental functioning across different populations but also points of regularity and consistency (Keith, Reference Keith and Keith2019; Sinha, Reference Sinha2002). This work has contributed to the development of basic emotion theories by conducting research pointing to the broad universality of emotion in different cultural contexts, most notably in the realm of facial expression and recognition (Hwang & Matsumoto, Reference Hwang, Matsumoto, Mandal and Awasthi2015, Reference Hwang, Matsumoto, Carducci and Nave2020; Manokara et al., Reference Manokara, Đurić, Fischer and Sauter2021; Matsumoto, Reference Matsumoto1990, Reference Matsumoto1992; Matsumoto & Ekman, Reference Matsumoto and Ekman1989; Matsumoto & Willingham, Reference Matsumoto and Willingham2006).

Other psychological approaches to emotion complement BET’s findings on universality; indeed, some influential scholars like Izard (Reference Izard1977, Reference Izard1991, Reference Izard2007, Reference Izard2009, Reference Izard2011) do not consider themselves BET theorists per se, but still advocate for discrete, universal emotions. Finally, it must be noted that basic and discrete emotion theories are not the only psychological models of emotion that consider the existence of particular universal emotions. We may also look, for instance, at the category of appraisal theories. Appraisal theory generally maintains that emotions emerge from a computational cognitive process in which humans subjectively evaluate the meaning of stimuli they encounter via a series of categorical criteria (Ellsworth, Reference Ellsworth2024); along with basic/discrete theories, it is one of the four primary theoretical models of emotion in the contemporary affective sciences. (We will encounter the final two, psychological construction and social construction, in the next section.) While appraisal theory, broadly speaking, is thus an intellectual competitor of BET, certain models nonetheless posit a similarly universal approach to particular emotions: Moors calls these “flavor 1” (Reference Moors2014, p. 304) or “biological” (Reference Moors2022, p. 175) appraisal theories. These models, she writes, split emotional episodes “into a limited number of subsets, corresponding to the specific emotions figuring in natural language (e.g. anger, fear, sadness),” and are thus “compatible with affect program theories” (Reference Moors2014, pp. 303, 304). Appraisal theories that might be thought to fall under this umbrella include Arnold (Reference Arnold1960), Lazarus (Reference Lazarus1991), Oatley & Johnson-Laird (Reference Oatley and Johnson-Laird2011), and Roseman (Reference Roseman1984, Reference Roseman2011, Reference Roseman2013).

Neuroscience

Affective neuroscience seeks to elucidate how emotions function at the neural level (Aromy & Vuilleumier, Reference Armony and Vuilleumier2013). For many researchers, a fundamental premise is that the human emotional brain evolved from earlier animal brain systems – with “partly separate neural circuits for different emotion related responses [including] autonomic output, freezing, fixed action patterns, and unconditioned approach or withdrawal” (Rolls, Reference Rolls, Watanabe, Hofman and Shimizu2017, p. 252) – and thus much work on human emotionality is anchored in a universalist perspective, which maintains that “emotions are ubiquitous across species and evolved by natural selection” (Adolphs & Anderson, Reference Adolphs and Anderson2018, p. 308; see also Adolphs, Reference Adolphs2017).

In the 1990s, a number of pioneering neuroscientists began to study the basic neural components of emotion systems, finding structures in the brain that give rise to affective feelings in humans (Johnston & Olson, Reference Johnston and Olson2015). “At least for some emotions,” Ledoux argued, “the evidence for an innate, biological organization is quite strong” (Reference LeDoux1996, p. 121); he famously reconstructed the neural pathways of fear responses in the rodent brain, which pointed to human functioning (see also Ledoux, Reference LeDoux2000). In his study of patients with brain injury, Damasio argued that the limbic system (particularly the amygdala) was vital to the operation of “primary” emotions, which he describes as “innate [and] preorganized” (Reference Damasio1994, p. 133; see also Damasio Reference Damasio1995, Reference Damasio1999, Reference Damasio2003); for him, emotions form a category of action programmes, or “sets of innate, programmed physiological actions aimed at addressing the detected [neural] changes and thereby maintaining or restoring homeostatic balance” (Damasio & Carvalho, Reference Damasio and Carvalho2013, p. 144). Though disagreeing with Damasio’s emphasis on the role of bodily feedback on emotional experience, Rolls (Reference Rolls1999) similarly examined the “neural bases of emotion,” taking a primarily evolutionary perspective; he argued specifically that “developments in primates in the structure and connection of neural systems involved in emotion such as the amygdala and orbitofrontal cortex [are] particularly important for understanding emotion in humans” (p. 75). Somewhat differently, Panksepp (Reference Panksepp1998, Reference Panksepp2005, Reference Panksepp2011, Reference Panksepp, Zachar and Ellis2012; Montag & Panksepp, Reference Montag and Panksepp2016; Panksepp & Bivven, Reference Panksepp and Biven2012; see also Montag & David, Reference Montag and Davis2020) identified the basic emotional circuits of mammalian brains, arguing that we can detect neural mechanisms for affective systems related to SEEKING, RAGE, FEAR, LUST, CARE, PANIC, and PLAY; these “underlying circuits for primary-process emotions were evolutionarily programmed/prewired” (Panksepp & Watt, Reference Panksepp and Watt2011, p. 390). This work, like that of Lane, Reiman, Ahern, Schwartz, and Davidson (Reference Lane, Reiman, Ahern, Schwartz and Davidson1997), assumes that neural correlates for basic emotional functioning can be discovered.

In human subjects, this first wave of affective neuroscience relied on things like lesion, electrical stimulation, and imaging studies to analyze the neural underpinnings of emotions; researchers generally attempted to associate particular areas of the brain with particular discrete emotions. Two early meta-analyses found at least partial support of basic emotion theory; Phan, Wager, Taylor, and Liberzon (Reference Phan, Wager, Taylor and Liberzon2002) associated particular individual emotions with specific locations within the brain, while Murphy, Nimmo-Smith, and Lawrence (Reference Murphy, Nimmo-Smith and Lawrence2003) identified “considerable support … for the affect program accounts of emotion,” noting that while emotions may not be “represented by entirely distinct neural circuits, it seems reasonable to conclude that the underlying neural systems are separate in part” (p. 227). The meta-analysis of Vytal and Hamann (Reference Vytal and Hamann2010) offered stronger evidence: not only “each of the basic emotion states examined (anger, fear, sadness, anger, and disgust) was consistently associated across studies with characteristic patterns of region brain activity” but also “each basic emotion was reliably distinguished or differentiated from the other emotions on the basis of its characteristic pattern of brain activation” (p. 2879).

This emphasis on patterns is crucial, because when subsequent experiments failed to show a one-to-one correspondence between brain regions and discrete emotions – for example, when the amygdala turned out not to be the simple “fear center” of the brain – researchers in the last decade or so began to use more sophisticated methods of imaging (such as multivariate pattern classification) to record the broad patterns of cross-region neural activity that seem to correspond to individual emotions (Kragel & LaBar, Reference Kragel and LaBar2013, Reference Kragel and LaBar2014). Such studies seemed to find “mappings between neural activation patterns and categorically distinct emotional experiences” (Kragel & LaBar, Reference Kragel and LaBar2015, p. 1447) and suggest that “information encoded in both neural ensembles and whole-brain activation patterns can be utilized to predict affective dimensions and discrete emotions with high levels of specificity” (Kragel & LaBar, Reference Kragel and LaBar2016, p. 453). For example, Saarimäki et al. (Reference Saarimäki2016) used multivariate analysis to show that “all 6 basic emotions have distinguishable but spatially distributed neural signatures in the human brain,” signatures that “generalize across different emotion-eliciting conditions and also across individuals” (p. 2564). Such experimental results suggest that “different emotions are represented in the brain in a distinguishable manner, yet in partly overlapping regions” (Saarimäki et al., Reference Saarimäki, Ejtehadian and Glerean2018, p. 477). Accordingly, current thinking suggests that the “discreteness” of basic emotions is “best understood as widespread, system-level patterned activity, rather than selective regional or systemic engagement during specific emotions” (Nummenmaa & Saarimäki, Reference Nummenmaa and Saarimäki2019, p. 7), and the most recent work looks for whole-brain functional (as opposed to physical) connectivity patterns in the experience of basic emotions (Saarimäki et al., Reference Saarimäki, Glerean and Smirnov2022).

Sociology

“Traditionally,” it was said not long ago, “emotion is a topic more central to psychology than to sociology” – but foundational sociologists like Marx, Weber, and Durkheim gave considerable attention to the operation of affective forces, and since the 1970s the sociology of emotion has truly emerged as a dedicated subfield of research (Smith-Lovin & Winkielman, Reference Smith-Lovin and Winkielman2010, p. 327; for overviews see; Ariza, Reference Ariza2021; Bericat, Reference Bericat2016; Lively, Reference Lively2024; Stets, Reference Stets2010; Stets, Reference Stets2012; Stets & Turner, Reference Stets and Turner2006, Reference Stets, Turner, Lewis, Haviland-Jones and Barrett2008, Reference Stets and Turner2014; Turner & Stets, Reference Turner and Stets2005). In many ways, debates about emotion in sociology ran parallel with those we have already explored in psychology (Smith-Lovin & Winkielman, Reference Smith-Lovin and Winkielman2010). Almost immediately, sociologists began to question whether emotions should be understood as biologically based, quasi-universal phenomena – the positivist, naturalizing, or organismic view – or whether they were more meaningfully shaped by social and cultural forces – the constructionist view, which we will see more of in Section 2 (Hochschild, Reference Hochschild1983a; Kemper, Reference Kemper1981). In an early contribution to the field, Kemper (Reference Kemper1981) neatly delineated some initial terms of this opposition; in terms of universality, the positivists most vitally emphasized “the importance of the biological and physiological substrate in the determination of specific emotions” (p. 336). Consistent with other fields, in the most basic sense these contrasting viewpoints continue to account for “the two main trends of sociology of emotions up to the present day” (Longo, Reference Longo2020, p. 42).

Those outside the discipline may be initially surprised to find the enthusiasm for the universalist outlook, but this in part reflects the larger legacy of much early sociology, which often “looked for what is generic and universal in human behaviors and patterns of socio organization” (J. H. Turner, Reference Turner2021, p. 4). Given sociology’s historical interest in making “systematic and universalistic” claims about “the nature of man and society,” it thus makes sense that much work in the field has maintained “that there is a common human nature [and] that emotions, sentiments, feelings, and passions do not vary over time” – it is “only their representation, the forms of their expression, and their philosophical or doctrinal rationalization” that are contextually dependent (Romania, Reference Romania, Cerulo and Scribano2022, pp. 106, 107–108). The universalist position of sociology is founded on the same general premises as basic/discrete theories in psychology – that a small set of “primary” emotions are an evolutionarily shaped, biological constant of human nature – and researchers in this mode argue that “a complete theory of emotion must ultimately deal with the fact that emotion is biologically rooted … regardless of the degree of social conversion, construction, or management” (Kemper, Reference Kemper and Kemper1990, p. 21). Turner and Stets (Reference Turner and Stets2005), for example, begin their sociology of emotion by announcing that “although there are cultural differences in how emotions are expressed and interpreted, it is now clear that some emotions are universal [and] generated from evolved body systems”; “most scholars would agree,” they suggest, that certain basic emotions are “primary or biologically based,” and we can thus safely “conclude that happiness, fear, anger, and sadness are universal among humans, with a few other emotions as potential candidates for inclusion in the list of primary emotions” (pp. 11, 13). For such positivist scholars, there is accordingly “nothing antisociological in finding that physiology plays a differentiating and crucial role in the emotions,” and this opinion is to a degree common: even sociologists who recommend skepticism toward the “lure” of the neurosciences acknowledge that, in general, constructivist work has “not managed to upset a naturalizing view of emotion” in sociological thought (Kemper, Reference Kemper1981, p. 342; Kleres, Reference Kleres, Hopkins, Kleres, Flam and Kuzmics2009, pp. 14, 13). Even beyond basic affective sentiments, we can find a universalizing orientation toward more socially elaborated emotion; consider, for example, Jacobsen’s (Reference Jacobsen2019) recent collection Emotions, Everyday Life and Sociology, a volume whose chapters on discrete emotions make statements like “courage is universally valued” (Marvasti, Reference Marvasti2019, p. 71), “embarrassment as an emotional experience is universal” (Jacobsen & Kristiansen, Reference Jacobsen and Kristiansen2019, p. 105), and “envy is a universal social problem” (Clanton, Reference Clanton2019, p. 150).

But aside from the issue of biological rootedness, sociologists often show universalizing sympathies in their accounts of how emotions function socially – indeed, the very search for “a general sociological theory of emotion” implies the possibility of universal principles of process and function (J. H. Turner, Reference Turner1999, p. 134), and universalism thus underpins the work of those who believe “the task of the sociology of emotions [most rightly concerns] the interconnection between social structure … and certain physiologically specific emotions” (Longo, Reference Longo2020, p. 42). J. H. Turner (Reference Turner2002), for example, in his theory of interpersonal behavior argues that “transactional needs drive the flow of interaction in certain universal directions, despite the widely varying contexts of encounters” (p. 28); emotions play a central role in how these encounters unfold, and they are thus thought to have universal functions that align with their biological universality. Another example is Barbalet (Reference Barbalet1998), who argues that “emotion terms can be developed in and applied to the analysis of social structure”: thus “rationality, class structure, social action, social conformity, basic rights, and social change [can be] considered through discussion of a particular emotion or set of emotions which both characteristically pertains to each of them and elucidates the processes to which each is subject” (p. 1). Perhaps most notably, Kemper (Reference Kemper1978, Reference Kemper1987, Reference Kemper and Kemper1990, Reference Kemper2006; Kemper & Collins, Reference Kemper and Collins1990) very much honors the physiological grounding of basic emotions, but argues that they are made salient in social encounters through linkage to the two central dimensions of personal interaction: power and status. Most importantly for our current purposes, a “fundamental assumption” of his theory is that “the power-status antecedents of specific emotions apply universally across the spectrum of social and demographic categories” – because “were the primary emotions to vary in their relational precursors, considerable social ambiguity would result,” a fact that fundamentally problematizes how “emotion might, in an evolutionary sense applying to all humans, have emerged” (Reference Kemper2006, pp. 109–110). Thamm (Reference Thamm2004), building on this work, finds that “the structure of human groups and emotions are universal,” and that “there is a direct link between specific universal social substructures and specific universal emotions” (pp. 189–190).

Anthropology

“Affect has never been a focus of anthropological research,” Epstein (Reference Epstein1992, p. 2) wrote some time ago, and Stodulka (Reference Stodulka and Storch2017, p. 12) more recently observed that “emotions are rarely a primary theoretical focus of ethnographies and predominantly remain implicit subject matter”; this is reflected, for example, by the fact that there is no section devoted to emotion in the 102 chapters of 21st Century Anthropology: A Reference Handbook (Birx, Reference Birx2010). In anthropology, emotions have, to some degree, been undertheorized in the broader sense, with the focus usually fixed on giving accounts of what they do in a given society (Beatty, Reference Beatty2014).

But this does not mean that anthropologists have remained quiet about the subject. Researchers increasingly acknowledge that emotions crucially shape the anthropologist’s experiences in the field (Behar, Reference Behar1996; Davies & Spencer, Reference Davies and Spencer2010; Lo Bosco, Reference Lo Bosco2021; Spencer & Davies, Reference Spencer and Davies2010; Stodulka, Selim, & Mattes, Reference Stodulka, Selim and Mattes2018; Stodulka, Dinkelaker, & Thajib, Reference Stodulka, Dinkelaker and Thajib2019), and beyond this, there has more generally been foundational work on anthropological emotion in the last 50 years. (It is important to recognize, though, that the heyday of modern anthropological research on emotion was in the last decades of the last century: noting in 2013 the “apparent thinness of current emotion research,” Beatty observed that “an overview of highlights in the anthropology of emotion would show that the major contributions have mostly been made some time ago” [p. 415].) Anthropological accounts of emotion have shown both explicit and implicit interest in the issue of affective universality – perhaps unsurprisingly, given that human universality is a larger concern of anthropology, in the sense that the discipline has historically been at least partly concerned with “the study of human nature in light of human variation” (Shore, Reference Shore2000, p. 81; see Roughley, Reference Roughley2000). In terms of emotion, the field has seen a division between universalist and constructivist positions that broadly reflects what we have seen elsewhere (Leavitt, Reference Leavitt1996; Svašek, Reference Svašek, Milton and Svašek2005).

Although much anthropology of emotion has taken a constructivist approach, there is a core universalist sympathy that runs through the history of the discipline. Following Boas (Reference Boas1910), modern anthropology was historically premised on the so-called psychic unity of mankind, the notion that there is “an essential similarity of mental endowment” in all peoples, and that “diverse [cross-cultural] phenomena are based on similar psychic processes” (pp. 372, 384; see Shore, Reference Shore1996; Beatty, Reference Beatty2019). But, as we will consider in the next section, commitment to this position does not mean that the discipline became fundamentally invested in enumerating transcultural human universals: Boas’s students primarily utilized our species’ common psychic potentiality to establish a cultural relativism that investigated the unique features of different population groups, and this tendency has endured in some of the most important subsequent anthropological works on emotion, especially (we will see) works that developed cultural relativism into a genuine constructivist position.

Despite this, however, there have always been anthropologists who adopted a universalist approach to the basic question of emotion. H. Geertz (Reference Geertz1959), for example, in her study of Javanese socialization processes, argues that “the range and quality of emotional experience is potentially the same for all human beings”; concepts like “anxiety and hostility, insofar as they are operationally defined in terms of scientific theory, refer to basically human – that is, universal – emotions” (p. 225). In his analysis of symbols, V. Turner (Reference Turner1967) similarly refers to “referents of a grossly physiological character, relating to general human experience of an emotional kind” (p. 54). Though Myers (Reference Myers1979) acknowledged that “socialization selects, elaborates, and emphasizes certain qualitative aspects of emotion,” his treatment of Pintupi affective vocabulary explicitly affirmed H. Geertz’s earlier position (353).

The 1980s saw the most concentrated and explicit expansion of the constructivist position in anthropology, but affective universalism still endured among some scholars. In a famous review of C. Geertz (Reference Geertz1980), Leach (Reference Leach1981) took aim at the entire field designated “cultural anthropology” – exclaiming that “I can make no sense of a line of thought which claims that ‘passions’ are culturally defined,” and denounced that assumption “that human individuals are products of their culture rather than of their genetic predisposition” (p. 32). Shweder and Levine’s (Reference Shweder and LeVine1984) Culture Theory: Essays on Mind, Self, and Emotion – a classic collection on the anthropology of emotion – contains several essays that support emotional universality; Spiro (Reference Spiro1984), for example, suggests that “human feelings and the ways in which they work are determined not so much by the characteristics of particularistic culture patterns but by the transcultural characteristics of a generic human mind” (p. 334), while Levy (Reference Levy1984) argues that “there is no reason to think that some dimensions of the emotional feeling do not vary cross-culturally” and that “the central tendencies named by various emotional terms are probably universal” (pp. 223, 229). R. Rosaldo (Reference Rosaldo and Bruner1984), in the same year, published a celebrated, autobiographical essay that chronicled how he came to understand the Illongot practice of headhunting only after the tragic death of his wife, a fellow anthropologist; though not a strict universalist, there is some degree of regularity implied in the statement that his own parallel “personal experience” enabled him to “apprehend the force of emotions in another culture” (p. 188).

Some anthropologists continued to anchor their subsequent work in biology, against the fashion of the day; Gerber (Reference Gerber, White and Kirkpatrick1985), for instance, assumes that “emotions arise at least in part from a biological substrate … which limits the range of emotional variation across cultures” (p. 121), while Epstein (Reference Epstein1992), drawing on Tomkins and Ekman, laments that “cultural analysis ignores biological factors” (p. 11). Lyon (Reference Lyon1995), going further, argues that “the cultural constructionist approach to emotion has real limitations,” because “emotion is more than a domain of cultural conception, more than mere construction, and thus cannot be treated merely as parallel to constructions such as self and person” (p. 247): It is vital, she argues, to honor the role of the biological body in emotional experience. And more generally, the broad importance of cross-cultural universals in anthropology was buttressed by the work of Brown (Reference Brown1991, Reference Brown and Roughley2000, Reference Brown, Wilson and Keil2001); facial expressions of emotions (among many other things) are considered in his analysis to be meaningfully universal.

Finally, the presumed universalism of emotion has played a specific methodological role in the theorization of anthropological fieldwork. I refer to the workings of empathy, because it has historically been thought that a framework of shared universal experience allows researchers to empathetically comprehend cultures that are drastically different from their own (Leavitt, Reference Leavitt1996). While constructivists have attacked this position as ethnocentric, it still has currency; Lindholm, for example, suggests that the “whole disciplinary practice [of anthropology] is based on feelings of empathy” (Reference Lindholm, Conerly and Edgerton2007, p. 31).

Philosophy

The philosophy of emotion is a robust research field (for overviews, see Brady, Reference Brady2024; Campeggiani, Reference Campeggiani2023; Cohen & Stern, Reference Cohen and Stern2017; Deigh, Reference Deigh2013; Deonna & Teroni, Reference Deonna and Teroni2012; Goldie, Reference Goldie2010; Griffiths, Reference Griffiths, Kingston, Banerjee, McKee, Chien and Vassiliou2017; Moors, Reference Moors2022; Scarantino, Reference Scarantino2016; Scarantino & de Sousa, Reference Scarantino and de Sousa2018; Solomon, Reference Solomon2004; Tappolet, Reference Tappolet2023). But compared to those in other disciplines, philosophers have devoted somewhat less attention to explicitly arguing for or against the universality of emotion. In 2002, Solomon – one of the most prominent philosophers of emotion in the last half-century – remarked that philosophers instead “tend to argue over whether an emotion is a feeling or a set of beliefs or judgments or evaluations … [so the questions] of whether the feelings of emotion are learned or innate and how they are based on physiological if not also neurological processes hardly get raised” (pp. 119–120). More recently, Tappolet (Reference Tappolet2023) observes that most philosophers of emotion are centrally concerned with “develop[ing] a theory that aims at spelling out what emotions consist in, so as to be able to distinguish emotions from other states” (pp. 11–12) – so although they have shown a robust interest in the multitude of ways that emotions might be considered, the universality question per se has not been the subject of the same rigorous debate that has been found in other fields. What has been fiercely debated is the related question of whether or not emotions are a natural kind (Charland, Reference Charland2001, Reference Charland2002; Cochraine, Reference Cochrane2019; De Sousa, Reference De Sousa1987; Griffiths, Reference Griffiths1997, Reference Griffiths and Kaszniak2001, Reference Griffiths2004a, Reference Griffiths and Solomon2004b; Prinz, Reference Printz2004; Rorty, Reference Rorty and Rorty1980; Scarantino, Reference Scarantino2009, Reference Scarantino, Zachar and Ellis2012a, Reference Scarantino2012b; Taylor, Reference Taylor2020) – that is, whether they form a class of entities about which scientific generalizations can be made. Many philosophers, Solomon (Reference Solomon, Marks and Ames1995) noted earlier, seem to work with the assumption that “there is some natural repertoire of emotions that constitutes ‘human nature,’ some set of ‘basic’ emotions” (p. 258) – a fundamentally universalist premise – but there is disagreement about whether all emotions are universal, or whether all things that get called emotion belong to the same scientifically valid category.

But despite this, Goldie (Reference Goldie2002) observes that there is a long philosophical tradition of believing that a certain small set of “basic emotions [are] common to all humans”: it is present in the work of the Stoics, Descartes, Spinoza, and Hobbes, to name just a few (p. 87). As Griffiths (Reference Griffiths, Kingston, Banerjee, McKee, Chien and Vassiliou2017) notes, more contemporary work on the emotions is commonly “pursued by adherents of philosophical naturalism” – that is, “the view that philosophy deals in knowledge of the natural world no different in principle from that revealed by the sciences” (p. 107); Brady (Reference Brady2024) suggests that this “naturalistic or empirically based philosophy … takes its lead from empirical science, seeking to make theoretical sense of empirical data [and thus] closely resembles psychological theorizing” (p. 216). This work tends to show a preference for the basic emotion paradigm in the mode of Tomkins and Ekman; this is perhaps because, Tappolet (Reference Tappolet2023) argues, basic emotion theory can be seen as generally compatible with what have been identified as the primary philosophical approaches to emotion (motivation theories, evaluation theories, and feeling theories; see Scarantino, Reference Scarantino2016; Scarantino & de Sousa, Reference Scarantino and de Sousa2018).

In this sense, many philosophers align themselves (to some degree) with what might be thought of as universalist psychological theories of emotion. Charland (Reference Charland1995), for instance, approvingly cites Ekman (and other empirical researchers) to conclude that “affect is an independent perceptual system that exhibits significant modular factors” (p. 295). Griffiths (Reference Griffiths1997) does not believe that the category of emotion is a natural kind, but he still acknowledges the existence of certain discrete affect programs, and argues that “the imperialist claims of social constructionists to explain the whole domain of emotional phenomena are unjustified” (p. 9). Like Griffiths, Goldie (Reference Goldie2002) endorses the notion of affect program responses (pp. 105–106). DeLancey (Reference DeLancey2002) grounds his philosophy of mind and artificial intelligence in the workings of affect programs and basic emotion theory; what’s more, he explicitly sets out to “confront social constructionism,” arguing that “it is not … a viable alternative to the affect program theory or any other naturalistic theory of the emotions” (p. 70). Also anchoring his work in affect programs, Prinz (Reference Printz2004) argues that “basic emotions are innate emotions that are not derived from other emotions” (p. 88); in his theory, all emotions are embodied appraisals, with basic emotions giving rise to complex emotions (via blending or cognitive elaboration). Clark (Reference Clark2010) argues for basic emotions, and believes that they and higher cognitive emotions “have deep roots in our biology” (p. 90). Finally, Cochraine (Reference Cochrane2019) emphasizes “type-specific response patterns” of emotion (p. 86), and takes a dimensional approach to argue that certain “basic features of [emotions] are culturally universal” (p. 209). The universalist premise of basic emotion theory, then, has been productively used in philosophical work.

History

In past centuries, historians routinely adopted a universalist approach to emotion, considering affectivity in terms of an unchanging human nature (see Boddice, Reference Boddice2024; Reddy, Reference Reddy2009). There is, however, no meaningful contemporary strand of historical research that is anchored in a universal approach to emotion: as we will see in the next section, history is perhaps the most unified discipline when it comes to the universalism/constructivism division, and opposition to universality has been said to be a foundational premise of the historical study of emotion.

Linguistics

Psychologists have shown regular attention to the connection between emotion and language (see Gallois et al., Reference Gallois, Vanman, Kalokerinos and Greenaway2021), but this relationship is of particular concern to those working in the language sciences. This is with good reason: “all speaking and writing,” Wilce (Reference Wilce2009) notes, is “inherently emotional to a greater or lesser extent,” in the sense that “nearly every dimension of every language at least potentially encodes emotion” (p. 3; see also Pritzker, Fenigsen, & Wilce, Reference Pritzker, Fenigsen and Wilce2020; Schiewer, Altarriba, & Ng, 2022–2023). Similar to what we’ve seen in other fields, in linguistics, a “debate has been raging between researchers who defend a more universalist perspective on emotions and others who feel that emotions should be investigated using a more relativist perspective, with a focus on the differences across languages and cultures” (Dewaele, Reference Dewaele2010, pp. 17–18). The universalist perspective on emotion emerges from a larger disciplinary investment in universality: since the second half of the twentieth century, Bickel (Reference Bickel, Enfield, Kockelman and Sidnell2014) observes, “linguistics has sent a strong message that despite all diversity, languages are built on a single universal grammar” (p. 102), and “a long-standing and dominant view” has maintained that “language is the product of an innate, universal, domain-specific and encapsulated module” (Majid, Reference Majid2012, p. 432; see, for example, Degli Esposti, Altmann, & Pachet, Reference Degli Esposti, Altmann and Pachet2016; Greenberg, Reference Greenberg1975). Indeed, the concept of universality has long been central to modern linguistics, and researchers have variously found patterns of universality in natural language components such as semantics, connotative or affective meaning, phonology, grammar, and lexicon, among others (Hupka, Lenton, & Hutchison, Reference Hupka, Lenton and Hutchison1999, p. 247).

Studies on the linguistic universality of emotion have generally attempted to demonstrate patterns of cross-cultural regularity in the semantic and categorical meanings of emotion words. Some representative examples can give a sense of this research. Earlier work by Herrmann and Raybeck (Reference Hermann and Raybeck1981), for example, used multidimensional scaling to investigate the meaning of emotion concepts across six cultures; they found “cross-cultural agreement [was] quite high” (p. 194). Romney, Moore, and Rusch (Reference Romney, Moore and Rusch1997) found that “English-speaking and Japanese-speaking subjects share a single model of the semantic structure of emotion terms”; culture-specific models based on English and Japanese norms “account[ed] for relatively little of the total variance” (p. 5489). Church, Katigbak, Reyes, and Jensen (Reference Church, Katigbak, Reyes and Jensen1998) compared the lexical and conceptual organization of emotion terms in English and Filipino, finding “better support for the cross-cultural comparability of emotion concepts and experience than for a strong social constructivist view” (p. 63). Moore, Romney, Hsia, and Rusch (Reference Moore, Romney, Hsia and Rusch1999) similarly found that “Chinese-, English-, and Japanese-speaking subjects assign basically similar meanings to 15 common emotion terms”; they argue that “the differences among the languages are genuine and statistically significant but small,” meaning that the findings are “consistent with previous research traditions that posit cultural and semantic universals” (p. 529). In his “semantic description of emotion predicates in Amharic,” Amberber (Reference Amberber, Harkins and Wierzbicka2001) begins by “assum[ing] a distinct cognitive domain which specialises in the emotions, and that the basic architecture of this cognitive domain is innate and universal” (p. 35); within this framework, his study “shows that emotional universals are borne out by the Amharic data” (p. 65). In a wider study, Hupka, Lenton, and Hutchinson (Reference Hupka, Lenton and Hutchison1999) found remarkable consistencies in the labeling of emotion categories across a survey of the world’s languages; these findings, they conclude, “strongly suggest that the lexical regularity, if not based on social and phenomenological experiences common to the human species, surely is founded on innate principles of human language” (p. 260). And in the twenty-first century, scholars from several disciplines created a psycholinguistic instrument (called “GRID”) to compare the cross-cultural meaning of emotion words (Fontaine, Scherer, & Soriano, Reference Fontaine, Scherer and Soriano2013); researchers using this tool have found “evidence for substantial convergence of emotion concepts across cultures,” leading to the conclusion that “cognitive representations of lexical meaning (i.e. semantic concepts) of emotions such as joy, anger, or sadness are remarkably similar within and across languages and cultures” (Loderer et al., Reference Loderer, Gentsch and Duffy2020, pp. 1483, 1480).

Beyond semantic meaning, researchers have looked for cross-cultural consistency in other domains. For example, a valence analysis of millions of words in 10 languages led Dodds et al. (Reference Dodds, Clark and Desu2015) to conclude that “human language present[s] an emotional spectrum with a universal, self-similar positive bias” in usage frequency (p. 2394). This work extends even to the sounds of words: Auracher, Albers, Zhai, Gareeva, and Stavniychuk (Reference Auracher, Albers, Zhai, Gareeva and Stavniychuk2010) found that “participants tend to use similar consonants independent of the mother tongue to encode specific emotional states” (p. 3) – cross-linguistically, nasal (m, n) sounds are linked to sad feelings, while plosive (p, b, t, and d) sounds are linked to happy feelings – while most recently, Shakuf et al. (Reference Shakuf, Ben-David and Wegner2022) have shown evidence for emotional universality in prosody (i.e., the rhythm, stress, and intonation of speech), meaning that speakers can accurately assess certain kinds of emotional content (such as anger, fear, happiness, and sadness) even in languages they do not understand.

Finally, metaphors hold particular importance to the language of emotion, as research suggests that their usage is “especially frequent when the topic is emotion, and their frequency increases with emotional intensity” (Crawford, Reference Crawford2009, p. 130; see Ortony & Fainsilber, Reference Ortony, Fainsilber and Wilks1989). Matsunaka, Chen, and Shinohara (Reference Matsunaka, Chen and Shinohara2023), in fact, note that “the issue of universality and cross-linguistic variability of emotion metaphors is one of the most controversial topics in cognitive linguistics” (n.p.). From a universalist perspective, cognitive linguists in the conceptual metaphor theory tradition (Lakoff & Johnson, Reference Lakoff and Johnson1980) argue that certain primary metaphors reflect simple experiential patterns that “map fundamental perceptual concepts onto equally fundamental but not directly perceptual ones”; because “humans everywhere share the basic patterns of perception and experience that are reflected in primary metaphors, these patterns ought to show up in languages around the world,” and many primary metaphors (such as BIG IS IMPORTANT) are in fact “widespread across languages that are not related genetically, areally, or culturally” (Grady, Reference Grady, Geeraerts and Cuyckens2007, pp. 192, 194; see also Grady, Reference Grady1997).

In terms of emotion, linguists have influentially argued that “systems of emotion metaphors arise from the physiology of emotions” (Lakoff, Reference Lakoff2014, p. 4). Lakoff (Reference Lakoff1987) draws on the work of Ekman to suggest that basic human emotions are “basic-level concepts [with] basic-level primary, and centrality” that manifest in embodied metaphors reflecting physiology (p. 14). In the classic example, the fact that the experience of anger entails both the increase of skin temperature and blood pressure and the decrease of visual perception and fine motor control is said to lead to the cross-language ubiquity of metaphors such as boiling mad, he exploded, blind with rage, etc. (Lakoff, Reference Lakoff2014, p. 4; see Lakoff & Kövecses, Reference Lakoff, Kövecses, Holland and Quinn1987; Wilkowski et al., Reference Wilkowski, Meier, Robinson, Carter and Feltman2009). Emanatian (Reference Emanatian1995), for example, suggests that “the two semantic domains of eating and heat may be favored cross-culturally as vehicles for conceptualizing lust and sex” (p. 164), while Crawford (Reference Crawford2009) reviews empirical evidence for the ubiquity of concepts such as GOOD as UP, GOOD is HIGH in Pitch, GOOD is BIG (or MORE), GOOD is BRIGHT, and GOOD is GETTING CLOSER, concluding that “abstract concepts [like emotion] are conceptualized in terms of more concrete, embodied physical domains” (p. 138).

Literary and Cultural Studies

Although emotion has long been understood as central to the analysis and appreciation of literature, there has not been a tremendous amount of modern scholarship that specifically addresses the universality of emotion. To some extent, it was simply assumed for centuries that the mark of a great book was its ability to universally appeal to readers, usually at an emotional level (see Long, Reference Long1909). In the last 50 years, however, there has been a dramatic shift in how literary critics think about universality; as we will see in the next section, it is a taboo concept for the majority of twenty-first-century literary theorists, most of whom find the idea of universality to be (at best) naïve and (more regularly) a politically retrograde tool of ideological oppression.

While universality has largely fallen out of fashion in English departments, there are some theorists who have attempted to revive an approach to literary criticism based on the assumption of a common human nature; the “literary humanism” of Mousley (Reference Mousley2007, Reference Mousley2011) is one such example, which (in terms of emotion) notes that one “problem with anti-universalism is that it has discredited the language of human feeling and engagement which we might want to use to describe the experience of literature” (Reference Mousley2007, p. 8). Human nature is even more central to scholars working in the subfield of evolutionary literary theory (J. Carroll, Reference Carroll and Richter2018; J. Carroll, Clasen, & Jonsson, Reference Jonsson and Shackelford2020; Jonsson, Reference Jonsson and Shackelford2020, Reference Jonsson2021), which sometimes touches upon emotion; J. Carroll (Reference Carroll2022a), for example, “uses an evolutionary model of human nature to organize emotion terms within eight categories, [which] serve as the framework for understanding emotions in individuals (authors, characters, and readers), literary genres, and literary periods” (p. 86).

But there are other scholars who more specifically devote themselves to analyzing the universality of emotion; among these, P. Hogan (Reference Hogan2003, Reference Hogan2008a, Reference Hogan2008b, Reference Hogan2011a, Reference Hogan2011b, Reference Hogan2013, Reference Hogan2018, Reference Hogan2022) is by far the most prominent and influential. Having studied an enormous range of works from unrelated, pre-colonial literary traditions, Hogan first argues for the existence of literary universals, primarily realized in cross-cultural patterns of recurrence in various elements of literature. But even more importantly, his expertise in affective science causes him to posit that “the universality of human emotion systems (such as attachment and fear) provide the groundwork for an explanation of story universals”: that is, “different emotion systems define different types of goals,” and “on the basis of the cross-cultural research … we can isolate and describe universal story patterns in terms of agents pursuing goals, goals that are themselves a function of our specific emotion systems” (Reference Hogan2022, p. 335). Hogan thus sees the universality of emotion as central to the operation of literature more broadly, and he has traced the intersection of literary and emotional universals across many groundbreaking books and articles. (The connection between basic literary forms and basic emotional experiences was also explored by a variety of ancient theorists, such as Aristotle and those in the Indian rasa tradition; see N. Carroll, Reference Carroll2022; L. P. Hogan, Reference Hogan and Hogan2011.)

Besides Hogan, a handful of other literary critics have similarly adopted a universalist perspective. We encountered above Oatley’s general theory of discrete emotion, but he (a fiction writer himself) has also developed a foundational body of research pertaining to literature and emotion (Reference Oatley2006, Reference Oatley2011, Reference Oatley2012, Reference Oatley2016); most recently, he has developed a structural account of how poetry evokes emotion, arguing that poems prompt various mental simulations that generate “basic emotions [like] happiness, sadness, anger, and anxiety” (Johnson-Laird & Oatley, Reference Johnson-Laird and Oatley2022, p. 1). Oatley is a cognitive theorist, and those in the broader subfield of cognitive literary studies sometimes equally touch upon emotion from an implicitly universalist position (see Zunshine, Reference Zunshine2015). More specifically, literary theorists interested in embodied cognition and simulation – particularly the effect of the mirror neuron network – often take an universalist perspective; Wojciehowski and Gallese (Reference Wojciehowski and Gallese2022), pioneers of this approach, argue that “the activity of reading fictional narratives activates the same neural circuits that we use in everyday life – circuits that underpin all of our own actions, emotional and sensory experiences” (p. 62; see also Gallese & Wojciehowski, Reference Gallese and Wojciehowski2018; Wojciehowski & Gallese, Reference Wojciehowski and Gallese2011).

In recent years, literary scholars have continued to intermittently ground their work using a universalist approach to emotion, either indirectly or directly. Asher (Reference Asher2017), for example, focuses on the ethical implications of literary emotions, but speaks in passing of “universal emotional patterns” and “universal emotional currents” (pp. 159, 163). Extending the work of Hogan, Singh (Reference Singh2021) suggests that “a universal narrative structure (the sympathetic plot) recurs because it induces pleasure” (p. 186). Zeng (Reference Zeng2022) argues that “universality in literature means that a piece of literary work is able to reflect universal emotions,” applying this standard to advocate for the emotional universality of Emily Brontë’s poem “Remembrance” (p. 313). And my own work of literary criticism (Irish, Reference Irish2018, Reference Irish2020, Reference Irish2023) often adopts the premise that certain prototypical features of emotion are functionally universal, even if the specific ways that emotions manifest are shaped by cultural and historical circumstances.

Beyond literary studies, work on the dynamics of art more generally will often adopt a universalist perspective, as in the attempt of Menninghaus et al. (Reference Menninghaus, Wagner and Wassiliwizky2019), to define the structural features of aesthetic emotions. Furthermore, the literary treatment of emotion complexly overlaps with so-called Affect Theory – a branch of poststructuralist cultural theory that is, in practice, often developed and employed by literary scholars (see Ahern, Reference Ahern2024; Gregg & Seigworth, Reference Gregg and Seigworth2010). Affect theorists sometimes engage universalist emotion paradigms – Sedgwick and Frank (Reference Sedgwick and Frank1995), for example, were famously influenced by Tomkins – and their emphasis on the human body is meant as a corrective to their critical tradition’s hyperfocus on language. Nonetheless, the poststructuralist foundation of their work fundamentally ensures that they are not biologically determinist in a way that would be familiar to universalist scientists – and it is also generally the case that their interaction with affective science is superficial and idiosyncratic, as individual experiments and theories are used as a platform for philosophical musings, rather than as a source of empirical data (see Hogan, Reference Hogan2018).

Psychology

While the modern constructivist tradition in psychology is usually said to start with the work of James (Reference James1884, Reference James1890), historical accounts of anti-universalism often point to the landmark study of Schachter and Singer (Reference Schachter and Singer1962), which suggested that people will apply different emotional labels to the same physiological experience depending on the context in which that experience occurs. In response to the rise of BET, more psychologists in subsequent decades would tend to the contextual features of emotion, most notably in the development of social constructivist approaches (see Mesquita & Parkinson, Reference Mesquita and Parkinson2024; Oatley, Reference Oatley, Lewis and Haviland1993). Reflecting (as we will see) a broader trend in the social sciences, constructivist psychologist denied that a basic set of emotions are cross-cultural universal, stressing instead that affective experience is primarily shaped by the specific social parameters of an individual’s developmental and cultural context; most notable is the work of Averill (Reference 58Averill, Plutchik and Kellerman1980, Reference Averill, Ekman and Davidson1994, Reference Averill2012), whose approach privileges not “a biological level of analysis,” but rather sees emotions as “socially constituted syndromes, the meaning and function of which are determined primarily by the social system(s) of which they are a part” (Reference Averill1982, p. 19; see also Ratner, Reference Ratner1989). Others questioned the categorical validity of the BET model, arguing that there is little evidence to suggest that a small set of English-specific emotion terms accurately reflects a universal quality of affective experience, and that such a method of classification is not actually scientifically valuable (Ortony & Turner, Reference Ortony and Turner1990; Russell, Reference Russell1991; see also Ortony, Reference Ortony2022). Indeed, the categorical tendency of BET was explicitly challenged by the fresh development of dimensional models of affect, which assessed emotions not in terms of individuated universal signatures, but rather saw them as characterized by descriptive dimensional features such as degree of valence and degree of arousal; rather than positing sharply discrete emotions, such approaches usually maintain that individual emotion words plot affective experience within a continuous, gradient space with fuzzy boundaries (a circumplex), reflecting various measures of the core dimensions (Barrett, Reference Barrett1998; Barrett & Russell, Reference Barrett and Russell1999; Feldman, Reference Feldman1995a, Reference Feldman1995b; Plutchik & Conte, Reference Plutchik and Conte1997; Posner, Russell, & Peterson, Reference Posner, Russell and Peterson2005; Russell, Reference Russell1979, Reference Russell1980; Russell & Barrett, Reference Russell and Barrett1999). Barrett and Russell’s work on the dimensional aspects of affect led to the development of psychological constructivist approaches to emotion, the most prominent intellectual competitor to basic/discrete emotion theories today.

Though psychological construction can be approached in different ways (Barrett, Reference Barrett2013), the most important (by far) emerges from the work of Barrett and colleagues, whose evolving model has been called conceptual act theory (Barrett, Reference Barrett2006b; Barrett, Wilson-Mendenhall, & Barsalou, Reference Barrett, Wilson-Mendenhall, Barsalou, Barrett and Russell2015; Lindquist & Barrett, Reference Lindquist and Barrett2008), the theory of constructed emotion (Barrett, Reference Barrett2017a, Reference Barrett2017b), and, most recently, the constructed mind approach (Barrett & Lida, Reference Barrett and Lida2024). While it is impossible to do justice to the enormous amount of literature this research program has generated, its current outlook basically argues that the human mind is a constantly simulating entity that uses predictive computational principles to construct emotional meaning from the sensory signals it receives from the body and environment (Barrett & Lida, Reference Barrett and Lida2024; see also Lindquist, Reference Lindquist2013; Russell, Reference Russell2009; and the citations in this paragraph). For our purposes, the key point is that what folk psychology calls “emotion” has no inherent biological meaning in this model; discrete “emotions” as typically understood do not actually exist in nature, and thus are in no way universal, but rather reflect a “population of events whose physical features will be highly variable, and whose functional features will be variable” (Barrett & Lida, Reference Barrett and Lida2024, p. 365; see Barrett, Reference Barrett2006a, Reference Barrett2012, Reference Barrett2022; Lindquist et al., Reference Lindquist, Siegel, Quigley and Barrett2013).

Building on the skepticism of earlier scholars (J. M. Carroll & Russell, Reference Carroll and Russell1996; Fridlund, Reference Fridlund1994; Russell, Reference Russell1994; Russell & Fehr, Reference Russell and Fehr1987), constructivists take particular aim at BET’s claims that facial expressions of emotion are universally produced and recognized; they argue that these conclusions are based on a faulty methodology that does not reflect the actual conditions in which emotions are expressed. In contrast, their experiments and research point to positive evidence that context and language labels vitally facilitate the perception of emotion, that different cultures perceive emotions differently, that the same emotional expression can signal a multiplicity of emotions, and that the so-called basic emotions do not actually generate predicted facial expressions in a reliable way (Aviezer, Hassin, Ryan, et al., Reference Aviezer, Hassin and Ryan2008; Aviezer et al., Reference Aviezer, Hassin, Bentin, Trope, Ambady and Skowronski2008; Barrett, Mesquita, & Gendron, Reference Barrett, Mesquita and Gendron2011; Barrett et al., Reference Barrett, Adolphs, Marsella, Martinez and Pollak2019; Brooks et al., Reference Brooks, Shablack and Gendron2017; Crivelli et al., Reference Crivelli, Jarillo, Russell and Fernández-Dols2016; Durán & Fernández-Dols, Reference Durán and Fernández-Dols2021; Durán, Reisenzein, & Fernández-Dols, Reference Durán, Reisenzein, Fernández-Dols, Fernández-Dols and Russell2017; Fang et al., 2022; Gendron, Reference Gendron2017; Gendron et al., Reference Gendron, Roberson, Vyver and Barrett2014; Gendron, Crivelli, & Barrett, Reference Gendron, Crivelli and Barrett2018; Hess & Hareli, Reference Hess and Hareli2019; Israelashvili, Hassin, & Aviezer, Reference Israelashvili, Hassin and Aviezer2019; Jack et al., Reference Jack, Garrod, Yu, Caldara and Schyns2012; Lee & Anderson, Reference Lee and Anderson2016; Lindquist et al., Reference Lindquist, Barrett, Bliss-Moreau and Russell2006; Nelson & Russell, Reference Nelson and Russell2013). Constructivists similarly argue that empirical findings do not support the claim that discrete emotions have a specific physiological signature, given the robust evidence that different instances of the supposedly same emotion can trigger different types of bodily responses (Barrett, Ochsner, & Gross, Reference Barrett, Ochsner, Gross and Bargh2007; Chentsova-Dutton et al., Reference Chentsova-Dutton, Gold, Gomes and Ryder2020; Hoemann et al., Reference Hoemann, Khan and Feldman2020; Lang, Reference Lang2014; Mauss & Robinson, Reference Mauss and Robinson2009; Mendes, Reference Mendes2016; Mendes & Park, Reference Mendes and Park2014; Siegel et al., Reference Siegel, Sands and Van den Noortgate2018; Tsai et al., Reference Tsai, Chentsova-Dutton, Freire-Bebeau and Przymus2002).

Beyond psychological constructivism, other psychologists similarly resist a thoroughly universalist approach to emotion. Some, for instance, point to the fact that mainstream psychology draws many of its conclusions – including its universalist conclusions – by research emerging from a WEIRD (white, educated, industrial, rich, and democratic) context, a demographic that hardly represents the world’s population (Arnett, Reference Arnett2008; Henrich, Heine, & Norenzayan, Reference Henrich, Heine and Norenzayan2010). While the field of cross-cultural psychology, we saw in the last section, is (in part) alert to similarities in different social populations – including how universal principles may lay behind apparent cultural variation – the field of cultural psychology, writes one of its founders, emphasizes “psychological diversity, rather than psychological uniformity” (Shweder, Reference Shweder, Kitayama and Cohen2007, p. 827): It is “the study of the way cultural traditions and social practices regulate, express, transform, and permute the human psyche, resulting less in psychic unity for humankind than in ethnic divergences in mind, self, and emotion” (Shweder, Reference Shweder, Stigler, Shweder and Herdt1990, p. 1). Different cultures, it is argued, have different kinds of psychologies (Markus & Kitayama, Reference Markus and Kitayama1991), and this has obvious implications for emotions, which emerge not as universal phenomenon but as contingent cultural constructions (Boiger et al., Reference Boiger, Ceulemans and De Leersnyder2018; De Leersnyder, Mesquita, & Boiger, Reference De Leersnyder, Mesquita, Boiger, Gelfand, Chiu and Hong2021; Mesquita, Boiger, & De Leersnyder, Reference Mesquita, Boiger and De Leersnyder2016, Reference Mesquita, Boiger and De Leersnyder2017). It has been said that emotion (as it is usually understood) may not even be a useful theoretical construct to consider mental states across human populations generally (Shweder, Reference Shweder, Manstead, Frijda and Fischer2004).

Cultural psychologists thus argue that “emotions … are neither natural nor universal” (Mesquita, Reference Mesquita2022), and their research explicitly considers that the meaning of certain emotions (like shame or humiliated fury) are not culturally universal (Kirchner et al., Reference Kirchner, Boiger and Uchida2018; Kollareth, Fernández-Dols, & Russell, Reference Kollareth, Fernández-Dols and Russell2018). Work on emotion in cultural psychology thus often falls under the broad umbrella of social constructivist approaches, as in Mesquita and Boiger’s sociodynamic model – which, seeing them as “largely functional to the sociocultural environment in which they occur,” argues that emotions are “dynamic systems that emerge from the interactions and relationships in which they take place” (Mesquita & Boiger, Reference Mesquita and Boiger2014, p. 298; see Boiger & Mesquita, Reference Boiger and Mesquita2012a, Reference Boiger and Mesquita2012b, Reference Boiger, Mesquita, Barrett and Russell2015). In its emphasis on population variance, cultural psychology also shares terrain with the more specific subfield of indigenous psychologies (Ciofalo, Reference Ciofalo2019; Kim, Yang, & Hwang, Reference Kim, Yang and Hwang2006; Paredes-Canilao et al., Reference Paredes-Canilao, Babaran-Diaz, Florendo, Salinas-Ramos and Parker2015). Tending to indigenous perspectives entails a “movement from investigation of psychological universals to study of culture as a psychological system [with] a bottom-up approach that builds theoretical views based on local phenomena” (Keith, Reference Keith and Keith2019, p. 10); this rejection of universalism is seen in work on emotion, as in studies of how affect is particularly understood in the indigenous thought systems of West Sumatra (Heider, Reference Heider2011), China (Sundararajan, Reference Sundararajan2015), India (Bilimoria & Wenta, Reference Bilimoria and Wenta2015), and Tanzania (Hoemann, Reference Hoemann, Gendron and Crittenden2024). Finally, we saw in the last section how evolutionary psychology has been used to underpin universalist approaches, but there have recently emerged more explicitly constructivist approaches to the evolution of emotion (Bliss-Moreau, Williams, & Karaskiewicz, Reference Bliss-Moreau, Williams, Karaskiewicz, Shackelford and Weekes-Shackelford2021; Lindquist et al., Reference Lindquist, Jackson, Leshin, Satpute and Gendron2022); these models, which reflect the fact that evolutionary thinking has always been a component of psychological construction, view discrete emotion categories as concepts that have emerged from a process of cultural evolution (see Mesoudi, Reference Mesoudi2016).

Constructivist theorists may acknowledge limited universality in certain particular aspects of emotion – that is, the basic psychological “ingredients” that give rise to emotion may be a typical part of human mental architecture, and there may be a general regularity in how people categorize the co-occurring response patterns of certain emotion episodes – but the more salient point is that they see the experience of emotion as “primarily a subjective interpretation constructed within the mind[,] rather than reflecting the objectively measurable effects of a dedicated neural program’s activation” (Shiota, Camras, & Adolphs, Reference Shiota, Camras and Adolphs2023, p. 431). As such, their position is usually thought of as an anti-universalist approach to emotion that stands in contrast to the universalism of basic and discrete theories of emotion.

Neuroscience

Though much work in affective neuroscience, we saw in the last section, is premised on emotional universalism, this is not absolute, as constructivist approaches to the brain have also generated important research. An early example of constructivist neuroscience is the work of Brothers (Reference Brothers1997), who emphasized the social embeddedness of human brains; taking issue with the outlook of LeDoux and Damasio, she argued that the “failures” of neuroscientists to “find emotion in hardware of the brain … prove[s] that the isolated-mind concept of emotion must be discarded” (p. 111). In the twenty-first century, however, opposition to the universalist perspective of affective neuroscience has come primarily from proponents of the constructivist theory of emotion championed by Barrett; this outlook challenges the conclusions that neural “signatures” can be discovered for discrete emotions, arguing instead that “emotion categories are unlikely to have distinct and innate physical correlates within the brain that are replicable across different contexts” (Clark-Polner et al., Reference Clark-Polner, Wager, Satpute and Barrett2016, p. 153).

Early in the development of psychological constructivist theory, Barrett et al. (Reference Barrett, Lindquist and Bliss-Moreau2007) argued that there is a “tremendous amount of evidence that is inconsistent” with the idea that certain discrete emotions are “biologically basic and derive from architecturally and chemically distinct circuits that are hard coded into the human brain at birth” (p. 297); in their view, “the evolutionary legacy to the newborn is not a set of modular emotion circuits that are hardwired into the subcortical features of the mammalian brain but may be, instead, a set of mechanisms that compute core affect and allow affective learning, as well as those that allow conceptual learning and categorization” (p. 305). For such scholars, neuroimaging evidence suggests that “many of the brain regions consistently activated during emotional experiences and perceptions show consistent activation in meta-analyses of other mental phenomena,” meaning that “brain regions are implementing basic psychological operations that are not specific to any emotion per se, or even to the category ‘emotion’” (Lindquist & Barrett, Reference Barrett2012, p. 535). In 2012, a much-cited meta-analysis from the same research team “failed to locate a specific brain basis for discrete emotion categories,” finding instead that “common brain activations exist across emotion categories” and “the bulk of the empirical evidence is more consistent with the hypothesis that emotions emerge from the interplay of more basic psychological operations” (Lindquist et al., Reference Barrett2012, pp. 141, 142; see also Wilson-Mendenhall, Barrett, & Barsalou, Reference Wilson-Mendenhall, Barrett and Barsalou2013). Others agreed that there is “strong evidence against the existence of super-specialized macro-anatomical structures for representations of single emotion categories (e.g. amygdala for fear)” (Guillory & Bujarski, Reference Guillory and Bujarski2014, p. 1887) and that “emotion categories are not contained within any one region or system, but are represented as configurations across multiple brain networks” (Wager et al., Reference Wager, Kang and Johnson2015, p. 1); a growing number of studies from the Barrett research team and others generally reflect that neural activity for emotion is distributed across the whole brain and overlaps with other cognitive processes (Huang et al., Reference Huang, Jastorff and Van den Stock2018; Raz et al., Reference Raz, Touroutoglou and Wilson-Mendenhall2016; Touroutoglou et al., Reference Touroutoglou, Lindquist, Dickerson and Barrett2015; Zhou, Reference Zhou, Zhao and Qi2021).

Other research supporting a constructivist view suggests the importance of contextual classification and categorization at the neural level (Brooks et al., Reference Brooks, Shablack and Gendron2017; Brooks et al., Reference Brooks, Chikazoe, Sadato and Freeman2019) and that using the predetermined “labels” of folk emotion already distorts our interpretation of neural evidence (Azari et al., Reference Azari, Westlin and Satpute2020). Most recently, discrete emotion theories have been challenged by pointing to the process of neural degeneracy, the fact that “instances of an emotion (e.g. fear) are created by multiple spatiotemporal patterns in varying populations of neurons” (Barrett, Reference Barrett2017b, p. 3); the fact that “multiple functional neural network patterns can result in instances of the same emotion category” is said to confound the notion that discrete emotions have particular neural signatures (Doyle et al., Reference Doyle, Lane and Brooks2022, p. 996).

Though constructivists acknowledge that studies in nonhuman animals have found dedicated neural circuits that control behaviors such as escaping, freezing, and fighting, they suggest that “there are a number of arguments for why a neural circuit for a behavior cannot be considered a neural circuit for an emotion per se” – for example, that an animal might variously flee, freeze, or fighting during an ostensible “fear” state seems to complicate the notion that a single neural mechanism controls what we understand as the emotion of fear (Touroutoglou et al., Reference Touroutoglou, Lindquist, Dickerson and Barrett2015, p. 1257). They also object to how discrete emotion theorists interpret the findings of techniques like multivoxel pattern analysis, arguing that such research still indicates that “when it comes to the observed pattern for any single emotion category, variation is the norm”; that the physical correlates of emotion are “highly variable within an emotion category from instance to instance” is taken to mean that there is no neural signature of discrete emotion, as many universalist theorists have historically predicted (Barrett & Satpute, Reference Barrett and Satpute2019, p. 12; see Clark-Polner, Johnson, & Barrett, Reference Clark-Polner, Johnson and Barrett2017).

Sociology

In the previous section, we saw how sociologists of emotion can have great sympathy for the universalist outlook – but this does not mean, in turn, that the positivist approach is self-evidently dominant in the field. Longo (Reference Longo2020), in fact, argues that it is best practice for sociologists to explicitly deemphasize any presumed innate universality of emotion, and others seem to agree. “In sociology,” it was said not too long ago, “we have been reluctant to overcome our historic tendency to associate all biology with reductionism” (Franks, Reference Franks2010, p. 2) – and even more recently, “the idea that emotions are primarily biological phenomena” has been called an “obstacle to the sociological study of emotions” (Harris, Reference Harris2015, p. 3) – so it makes sense that the positivist emphasis on the physiological foundations of emotion has not fully taken hold. Turner and Stets (Reference Turner and Stets2005) even suggest that the sociology of emotion has been conventionally framed in constructivist terms: “for most sociologists, emotions are socially constructed in the sense that what people feel is conditioned by socialization into culture and by participation in social structures” (p. 2). Indeed, as Lively (Reference Lively2024) has recently put it, “the bulk of sociological scholarship on emotion,” irrespective of specific focus, tends to ask questions like “how do historically and culturally specific norms influence the experience and expression of emotion … and what role do emotions play in reifying and disrupting cultural and social constraints” – inquiries obviously premised on the fact that certain vital aspects of emotion are shaped by culturally and historically contingent social forces (p. 289). There is thus a rich tradition of constructivism in sociology, which runs in parallel to the positivist scholarship reviewed in Section 1.

The constructivist position fundamentally suggests that “human emotionality is shaped by the ensemble of social relationships that bind human beings to one another” (Denzin, 1990, p. 90). Emotions are “aspects of a specific culture and social context which determine when, and in which social circumstances, they may be properly manifested and how they should be managed”; they are seen as “socially defined in the sense that it is the social that sets their meaning and normative standards that regulate their expression and management, [while] the biological component is reduced to the organic mechanism which trigger emotions” (Longo, Reference Longo2020, pp. 4, 43). This outlook is seen primarily in what Bericat (Reference Bericat2016) describes as cultural theories, which “see emotion not as mere biological responses but as social feelings” (p. 499). With this approach, constructivist sociologists emphasize processes such as emotional labeling and strategies of emotional management – that is, ways in which the expression and experience of emotion are shaped by contextual forces.

The constructivist tradition of modern emotion sociology is most usually said to inaugurate with Hochschild (Reference Hochschild1979), whose work on emotion management came to emphasize “secondary acts performed upon the ongoing nonreflective stream of primary experience” – that is, “how social factors affect what people think and do about what they feel (i.e. acts of assessment and management)” (p. 552; see also Hochschild, Reference Hochschild1983b). At the same time, Shott (Reference Shott1979) began to focus on both “the socialization of emotion, which generates variability in affective experience across cultures [and] the construction of emotions by the actor, a process greatly influenced by situational definitions and social norms” (p. 1318). A few years later, Thoits (Reference Thoits1985) argued that “only through language do we know what we feel, and implicitly, why” – that is, when “a cultural label is placed on a conjunction of situational cues physiological sensations, and bodily gestures” (pp. 232–233) – and she later affirmed that “along with most sociologists, I assume that emotions are not simply innate, biophysiological phenomena,” because “considerable variability exists historically and cross-culturally in the situational causes, experience, meaning, display, and regulation of emotions, including basic ones” (Reference Thoits1989, p. 319). Perhaps most stridently, Gordon’s work on emotional labeling (Reference Gordon and Kemper1990) maintains that “emotion should be viewed as an ‘open system’ in which the entire combination of elements is socially constructed, rather than a closed system in which society merely activates or stimulates the fixed connections” (p. 153; see Gordon, Reference Gordon, Rosenberg and Turner1981).

And there is, of course, plenty of more recent scholarship that takes an anti-universalist approach. Longo (Reference Longo2020), for example, acknowledges that “approaches which consider emotions as universal, transcultural and homogeneous, based as they are on some essentialist conception of human action, are endowed with a cogent persuasiveness” – but he begins his study of emotions, society, and the self by arguing that “the very possibility of a sociology of emotions lies not in homologies but in differences,” in the sense that “if emotions were experienced and managed in the same way across historical times, cultures and the different strata of the same society, any attempt at a sociology of emotions would be frustrating” (p. 1).

Anthropology

Constructivism, we have already seen, has played a vital role in how anthropologists have historically theorized emotion. Constructivist accounts have emphasized several categories of evidence: the “celebrated counter examples” (Beatty, Reference Beatty2014, p. 546) of cultures that seemingly don’t experience certain basic emotion profiles, such as sadness for Tahitians (Levy, Reference Levy1973) or anger for the Utku Iniut (Briggs, Reference Briggs1970); cultures that experience emotions that are seemingly foreign and untranslatable to the Western tradition, such as the Ilongot liget (M. Z. Rosaldo, Reference Rosaldo1980), the Balinese lek (Geertz, Reference Geertz1973), or the Japanese amae (Lebra, Reference Lebra1976); or cultures that seem to understand the fundamental evaluative framework of emotion to reside beyond the Western sense of interiority, such as the Ifaluk, who see it in terms of situational events (Lutz, Reference Lutz1982), or Samoans, who see it in terms of social goodness (Gerber, Reference Gerber1975).

As noted in the previous section, the Boasian emphasis on the psychic unity of humankind did not lead early twentieth-century anthropologists to seek patterns of cultural uniformity; instead, its rejection of cultural evolution models (and the hierarchization of cultures that they underpinned) resulted in a relativistic interest in the particularities of culture and personality, as evidenced in the work of Mead, Benedict, and Bateson (Bateson & Mead, Reference Bateson and Mead1942; Benedict, Reference Benedict1935; Mead, Reference Mead1928, Reference Mead1930, Reference Mead1935; see Beatty, Reference Beatty2019; Stodulka, Reference Stodulka and Storch2017). The anti-universalist bent is suggested by Mead’s study of “three primitive societies,” which led her “to conclude that human nature is almost unbelievably malleable, responding accurately and contrastingly to contrasting cultural conditions” (Reference Mead1935, p. 280). Other early scholarship, such as that of Labarre (Reference Labarre1947) and Birdwhistell (Reference Birdwhistell1970), challenged the notion that nonverbal emotion cues like gestures and facial expressions were actually as universal as the Darwinian camp claimed. In the middle of the century, relativistic anthropology was further invigorated by the research of C. Geertz, with vast implications for the understanding of emotion; across a variety of studies, he argued that “not only ideas, but emotions too are cultural artifacts in man” (Reference Geertz and Scher1962, p. 735), that “the passions are as cultural as [social institutions]” (Reference Geertz1980, p. 124), and that “our ideas, our values, our acts, even our emotions, are, like our nervous system itself, cultural products” (Reference Geertz and Platt1966, p. 114). Given his emphasis on emotion as a cultural concept, the attack on universality was explicit: to attribute meaning to “universal properties of the human mind,” he argued, “is to pretend a science that does not exist and imagine a reality that cannot be found” (Reference Geertz1973, p. 20).

The work of C. Geertz paved the way for the heyday of constructivist anthropological accounts in the 1970s, 1980s, and 1990s, most notably advanced by Lutz (Reference Lutz1982, Reference Lutz1986, Reference Lutz1988; Abu-Lughod & Lutz, Reference Abu-Lughod, Lutz, Lutz and Abu-Lughod1990; Lutz & White, Reference Lutz and White1986). This work, she reflected some decades later, “relativized, historicized, and contextualized both the emotions or emotionality and the Euro-American psychological science that asserted itself as at the helm of understanding emotion” (Lutz, Reference Lutz2017, p. 182). As this suggests, the approach is particularly premised on the correction of an alleged Eurocentric essentialism in previous anthropology, which sought emotions “in the supposedly more permanent structures of human existence – in spleens, souls, genes, human nature, and individual psychology, rather than in history, culture, ideology, and temporary human purposes”; accordingly, to look critically “at the Euroamerican construction of emotion is to unmask the ways in which that schema unconsciously serves as a normative device for judging the mental health of culturally different peoples” (Lutz, Reference Lutz1986, pp. 287, 288). Similarly, it was noted in the last section that some anthropologists appealed to the presumed universality of empathy in their work on emotion, but constructivist accounts of this period often explicitly denounced this practice; Lynch (Reference Lynch and Lynch1990), for example, advises that we must “reject empathy as a naive and ethnocentric practice, a form of Western imperialism over the emotions of the Other” (p. 17; see also C. Geertz, Reference Geertz1975).

In “going beyond [a] psychobiological framework to include concerns with emotion’s social relational, communicative, and cultural aspects” (Lutz & White, Reference Lutz and White1986, p. 405), this research particularly emphasizes the role of language in the construction of emotion; many of these studies “approach emotion through language and understand language as inescapably and fundamentally social,” meaning that “emotion can be said to be created in, rather than shaped by, speech in the sense that it is postulated as an entity in language where its meaning to social actors is also elaborated” (Abu-Lughod, & Lutz, pp. 10, 12). More specifically, Lutz’s study of the Ifaluk people “treat[s] emotion as an ideological practice rather than as a thing to discovered or an essence to be distilled,” in order to “demonstrate how emotional meaning is fundamentally structured by particular cultural systems and particular social and material environments” (Reference Lutz1988, pp. 4, 5). Similar research includes that of Abu-Lughod (Reference Abu-Lughod1986), which emphasizes “the cultural construction of the sentiments” (p. 206); Grima (Reference Grima1992), which argues most fundamentally that “emotion is culturally constructed rather than universal” (p. 6); and especially M. Rosaldo, which resists the anthropological tendency “to view affective life more as a ‘sign’ that points to social rule than as itself a sphere of meaning that [is] public and socially significant”(Reference Geertz1980, p. 35), and affirms that “affects, whatever their similarities, are no more similar than the societies in which we live” (1984, p. 145).

Although emotion is somewhat less of an anthropological focus in the twenty-first century, many scholars today still adopt a fundamentally constructivist outlook. Illouz and Wilf (Reference Illouz, Wilf, Hopkins, Kleres, Flam and Kuzmics2009), for example, in their more recent study of love, affirm that “the anthropology of emotion has successfully refuted [the] view – hitherto dominant – that emotions are physiological responses to situations, which in turn make them universal invariants of human action”; they thus align themselves with the long research tradition (following C. Geertz) that has “showed repeatedly that emotions are shaped, through and through, by cultural meanings, and in particular by cultural conceptions of the person” (p. 124).

Philosophy

As stated in Section 1, many contemporary philosophers employ notions of affect programs or biologically “basic” emotions in their analysis. This does not mean, however, that there are no dissenters. Harré (Reference Harré and Harré1986), in a collection devoted to the social construction of emotion, argued that “the overwhelming evidence of cultural diversity and cognitive differentiation in the emotions of mankind has become so obvious that a new consensus is developing around the idea of social construction” – a consensus that upturns the “universalistic implications” of earlier research (p. 3; see also Harrré, Reference Harré and Roughley2000). In the same volume, Armon-Jones (Reference Armon-Jones and Harré1986b) explicitly lays out “the thesis of constructionism”: “emotions are characterized by attitudes such as beliefs, judgements and desires, the contents of which are not natural, but are determined by the systems of cultural belief, value, and moral value of particular communities” (p. 33; see also Armon-Jones, Reference Armon-Jones and Harré1986a). Finally, at one point in his career, Solomon (Reference Solomon1984) explicitly opposed the understanding that emotions “can be presumed a priori (and falsely) to be more or less the same in all human beings”: he rejects the “familiar but fallacious” view that “emotions can therefore be taken to be more or less universal human phenomena,” leading him to the thesis that “emotions are to be construed as cultural acquisitions, determined by the circumstances and concepts of a particular culture as well as, or rather much more than, by the functions of biology and, more specifically, neurology” (pp. 239–240). Solomon seems to have adopted a more hybrid position later in his career, but continued to maintain that philosophers must not a priori assume emotional universality (see Solomon, Reference Solomon, Marks and Ames1995).

While emotional constructivism per se seems less popular with twenty-first-century philosophers, contemporary thinkers do usually grant that social and cultural context is an important part of emotional experience. Finally, it must again be noted that many philosophers argue that neither the folk category emotion nor specific emotion categories (i.e. anger) have natural boundaries – but this does not mean, as we saw in Section 1, that they reject the possibility that certain emotional responses (i.e. affect program fear) are biologically natural and thus universal.

History

In the last forty years, the history of emotion has emerged as a vibrant research cluster (for overviews see; Barclay, Reference Barclay2020, Reference Barclay2021b; Boddice, Reference Boddice, Handley, McWilliam and Noakes2018, Reference Boddice2020b, Reference Boddice2024; Knatz, Reference Knatz, Geroulanos and Sapiro2023; Matt & Stearns, Reference Matt and Stearns2013; Plamper, Reference Plamper2015; Rosenwein & Cristiani, Reference Rosenwein and Cristiani2017; Schnell, Reference Schnell2021). “Historians of emotion,” it has been recently said, “can, with confidence, settle debates about universalism” (Boddice, Reference Boddice2019, p. Reference Boddice1994), and the boldness of this statement reflects the fact that historians are essentially unified in their outlook on the question of emotional universality.

The “key premise of the history of emotions is that emotion varies across time and place and so has a history that can be explored by scholars” – so unsurprisingly, this naturally positions virtually all historians to take an anti-universalist stance in their research (Barclay, Reference Barclay2021b, p. 457). Historians occasionally suggest that their work is premised on an essentially shared understanding of emotion – Eustace (Reference Eustace2008), for example, writes that “we are able to analyze eighteenth-century emotion today only because of a shared physiology of feeling that stretches over the centuries” (p. 12) – but this is generally not the focus of historical research (nor is it the focus of Eustace). Three decades ago, Danziger (Reference Danziger and Leary1990) noted that many historians of the day had “overlook[ed] the possibility that the very objects of psychological discourse, and not just opinions about them, have changed radically in the course of history” (p. 336); as the contemporary field of emotion history began to develop in the intervening years, scholars took this caution to heart. “To put it simply,” Dodman (Reference Dodman, Barclay, Crozier-De Rosa and Stearns2021) has recently argued, “unchanging, basic emotions hard-wired by evolutionary processes go against the very idea of a history of emotions” (p. 17). It is thus with good reason, he continues, that “most historians stick to tried and tested social constructivist positions”; though many “historians of emotion have been very conscious of work in psychology, biology and neuroscience,” the emphasis is almost entirely on the scientific work that “provides models for imagining the body as plastic and historical, as much as universal” (Barclay, Reference Barclay2021a, p. 114). Historians of emotion thus overwhelmingly agree with Rosenwein (Reference Rosenwein2001) that “emotions themselves are extremely plastic” and that “it is very hard to maintain, except at an abstract level, that emotions are everywhere the same” (p. 231). “In this field,” Knatz (Reference Knatz, Geroulanos and Sapiro2023) thus observes, “past emotions … are thought of as qualitatively different, ontologically divergent over time” – and there is little room for notions of universality (p. 275). Accordingly, in history of emotion work, constructivist models of emotions are routinely held up for praise (Boddice, Reference Boddice2020a, Reference Boddice2020b; Bound Alberti, Reference Bound Alberti2018; Eustace et al., Reference Eustace, Lean and Livingston2012; Reddy, Reference Reddy2020; Rosenwein, Reference Rosenwein2021; Tepora, Reference Tepora2020), while the basic and discrete models of “Neo-Darwinists, hard-line geneticists, behaviourists [and] biological determinists” are subjected to fierce critique (Boddice, Reference Boddice, Handley, McWilliam and Noakes2018, p. 46).

“Recent theoreticians of the history of emotion,” Tepora (Reference Tepora2020) notes, have “vigorously defended the malleability of emotions over time and across cultures by rejecting the universalism of basic emotions,” and a sampling of statements from contemporary scholars confirms this assessment (p. 98). Rosenwein, for example, elsewhere notes that “universalist and presentist views of the emotions [are] problematic enough” that emotion history is a necessary field (Reference Rosenwein2010, p. 10), and it is “very unlikely that emotions are invariable” (Reference Rosenwein2016, p. 2). Plamper (Reference Plamper2015) states that “what is universal [in emotion] amounts to a molehill when compared to the mountain of data on cultural difference” – and besides, “the universal is often uninteresting” (p. 33). Warning against “the temptation toward the universality of emotional phenomena,” Boddice (Reference Boddice2017) suggests that “to say that emotions change over time is incompatible with claims that there is something fundamentally transcendent or basic about (some) emotions”; accordingly, he affirms that “the history of emotions implicitly challenges basic-emotions models [and] should explicitly do so” (pp. 11, 12). Bound Alberti (Reference Bound Alberti2018) writes that “Ekman’s reductionist biological model has been criticized by more nuanced theorists who demonstrate that rather than being universal, emotions are developed within complex power relations, and through the lenses of disciplinary classifications that are themselves historically specific” (p. 243). Reddy (Reference Reddy2020) believes that the “weaknesses” of BET are “in fact so obvious that one must ask why this theory has had such a long life” (p. 168). And Dixon (Reference Dixon2020) denounces the “essentialism, ethnocentrism and anachronism” that arises when one considers “unchanging, universal emotion[s]” (pp. 11, 3). Indeed, it has recently been said that the entire field of emotion history “sees itself as a reaction to universalistic and biological conceptions of emotions” (Schnell, Reference Schnell2021, p. 25).

Linguistics

In the previous section, we saw how a universalist approach to language dominated the field of linguistics in the second half of the twentieth century – but this was not always the case. In 1969, it was actually said that “the prevailing doctrine of American linguistics and anthropologists has, in this century, been that of extreme linguistic relativity,” meaning that “each language is semantically arbitrary relative to every other language” – according to this outlook, “the search for semantic universals is fruitless in principle” (Berlin & Kay, Reference Berlin and Kay1969, pp. 1–2). And there has been a more recent general challenge to universalism, with at least some researchers claiming that “the idea of a single underlying linguistic system different only in surface realization seems increasingly unlikely” (Majid, Reference Majid2012, p. 432). Evans and Levinson (Reference Evans and Levinson2009), for example, argued that “languages differ so fundamentally from one another at every level of description (sound, grammar, lexicon, meaning) that it is very hard to find any single structural property they share”; accordingly, there are “vanishingly few universals of language,” and linguistic universality is just a myth that has endured despite “a massive accumulation of counter evidence” (pp. 429, 430).

Most famously developed by Sapir (Reference Sapir1921) and Whorf (Reference Whorf1956), theories of linguistic relativity maintain that the properties of language shape human thought, to some degree or another. Extreme versions of the so-called Sapir-Whorf hypothesis seem unlikely – people are self-evidently able to conceptualize and understand concepts that are not linguistically present in their native language – but more flexible notions of linguistic relativity, which still reflect “just how profoundly the variation in languages can influence forms of cognition” (Downey & Gillett, Reference Downey and Gillett2023, p. 694), give rise to the notion of emotional linguistic relativity, the possibility that “the way we feel may depend, at least in part, upon the language[s] we speak” (Ponsonnet, Reference Ponsonnet and Schiewer2022, p. 1033). Consistent with psychological constructivist theories, such a perspective argues that “the way we [linguistically] conceptualize emotions is likely to influence the way we experience them” (p. 1054). This outlook is shared by many linguistic anthropologists, who embrace cultural particularism and believe that the language of emotion is “inseparable from culture, and as likely to shape as be shaped by mind” (Wilce, Reference Wilce2014, p. 79; see also Besnier, Reference Besnier1990). Bamberg (Reference Bamberg, Niemeier and Dirven1997a), for example, began by “view[ing] emotions from the starting point of language” – that is, “how language forms reflect or construct what is commonly taken as an emotion” (p. 209) – to develop a “linguistic-constructionist” approach that inverts the “traditional, realist picture of the relationship between emotions, cognitions and language” (Reference Bamberg1997b, pp. 310, 335).

Critiques of linguistic emotional universality also take other forms. The very field of historical linguistics, for example, reflects the fact that “people’s words for emotions have been and will be alive throughout their histories, reflecting changing ideas of what emotions are, how they relate to each other, how they relate to other concepts, and how they affect us” (Tissari, Reference Tissari2017, p. 93). In terms of metaphor, we have seen how conceptual metaphor theory posits the ubiquity of certain primary emotional metaphors based on universal physiology; however, evidence suggests that different languages (such as English and Hebrew) vary in the way and extent to which somatic metaphors are employed (Kidron & Kuzar, Reference Kidron and Kuzar2002). Even seemingly stable metaphors are not always cross-culturally consistent; for example, a recent study of Mlabri (a language of Thailand and Laos) shows emotional mappings of DOWN as desirable and UP as undesirable, confounding the “commonly reported HAPPY IS UP metaphor [that] is said to link to universal bodily correlates of emotion” (Wnuk & Ito, Reference Wnuk and Ito2021, p. 195). These metaphors are “grounded in the bodily experience of positive low-arousal states,” suggesting that in their metaphoric usage “cultures draw on the available sensorimotor correlates of emotion in distinct ways.” And it has been argued that the cognitive linguist positing of primary metaphors based on fundamental experience and physiology is an “ahistorical method [that] obscures the possible role of cultural traditions as a source of emotion concepts” (Geeraerts, Reference Geeraerts2006, p. 227). Beyond differences in metaphor, the social construction of emotion language may be particularly indicated by multilingual speakers, in the sense that “in any pair of languages, specific emotion concepts may overlap completely, partially, or not at all”; those adept in multiple languages thus report tension and difficulty in negotiating certain emotion concepts across languages, leading to the hypothesis that “bilinguals’ concepts may, in some cases, be distinct from those of monolingual speakers” (Pavlenko, Reference Pavlenko2008, pp. 152, 147; see also Pavlenko, Reference Pavlenko2005, Reference Pavlenko2006). Above all, many skeptical linguists argue that emotion theorists often “seem unaware of absolutising their own language (with its built-in culture and concepts), i.e. of making universal claims on the basis of language-specific categories” (Dewaele, Reference Dewaele2010, p. 18).

Finally, psychologists in the constructivist tradition have also conducted complementary research on language – unsurprisingly, given that language seems to be “especially important to the acquisition and use of [the] conceptual knowledge” that underpins the theory of constructed emotion (Doyle & Lindquist, Reference Doyle and Lindquist2018, p. 62). Gendron, Lindquist, Barsalou, and Barrett (Reference Barrett2012) argue that “emotion words provide an important (although often unrecognized) context in emotion perception” (p. 321). Doyle and Lindquist (Reference Doyle and Lindquist2018) find that “emotion words support the acquisition of new conceptual knowledge that then biase[s] subsequent perceptual memory for emotional facial actions” (p. 72). Satpute and Lindquist (Reference Satpute and Lindquist2021) found that neural processes associated with language are also involved with emotion, leading to their constructivist conclusion that “language shapes emotional experience” (p. 216). And most recently, Ogren and Sandhofer’s (Reference Ogren and Sandhofer2022) study of emotion labeling in children suggested “language may in fact be important for emotion understanding and therefore that linguistic differences across cultures may lead to differences in emotion perception cross-culturally” (p. 175).

Literary and Cultural Studies

As noted in the previous section, modern scholars have largely overturned the long-held interpretive belief in the existence of literary universals. “In many areas of literary studies today,” Gallese and Wojciehowski (Reference Gallese and Wojciehowski2018) note, “and for the past several decades, the idea of universals has been heavily challenged … [t]here are persistent and ongoing concerns that any claims to universality will be ensconced as social or cultural norms that delegitimize or stigmatize persons or groups whose experiences do not conform to those norms” (n.p.). We may consider the case of Shakespeare, who has long been considered the quintessential example of a universal poet. In recent years, researchers have demonstrated that the apparent global appeal of Shakespeare owes no small part to the coercive dynamics of imperialism, as the playwright was purposefully utilized in the British colonial education system to Anglicize native inhabitants of colonized lands. Shakespeare’s reputation as “universal” thus reflects how a conscious political program disingenuously enshrined a white, male, European viewpoint as culturally normative (to the determent of native perspectives), leading many literary scholars today to see Shakespeare’s supposed universality as little more than a racist myth historically deployed as “an instrument of colonial hegemony” (Weissbourd, Reference Weissbourd2023, p. 207).

In general, universality has extremely low currency in modern literary studies, so scholars devote little explicit time advancing a nonuniversalist theory of emotion: a constructivist position is rather assumed. A typical account can be found in Scott (Reference Scott2019), who introduces his analysis of emotion in postcolonial literature by affirming that “I am not suggesting, of course, that emotion can be easily universalized … quite the opposite: I view emotion not as the product of some ahistorical psychic essence, but as a ‘felt response’ to specific social, political, and economic forces” (p. 27). The “of course” indicates the extent to which a universalist perspective on emotion would be unusual in contemporary research on literature; typical sympathies are much more apt to align with Gross’s (Reference Gross2010) sense that “emotional encounters that we might attribute to our biologically ground intuitions must always be considered in a larger political context where the status of social institutions are at stake” (pp. 47–48). There, is, however, a small amount of work that directly argues against the universality of emotion; in their treatment of neuroscience and literature, for example, both Armstrong (Reference Armstrong2020) and Comer and Taggart (Reference Comer and Taggart2021) ground their understanding of emotion in the work of Barrett and generally adopt a constructivist perspective.

Because of its poststructuralist genealogy, cultural theory is heavily suspicious of biological determinism and universalizing tendencies. Ahmed (Reference Ahmed2004), arguably the most prominent and influential cultural theorist of affect, explicitly announces “a critique of the assumption that emotions are innate or biological (p. 17). Leys (Reference Leys2017, Reference Leys2021) forcefully criticizes affect theorists for their regular use of universalizing scientists like Ekman – but, as noted in the previous section, even affect theorists who use such science for philosophical speculation would probably not consider themselves emotional universalists in the same way that a psychologist or neuroscientist might.

Assessing the Debate

Before proceeding, however, it is important to reflect briefly on the central contention between universalists and anti-universalists, now that both sides of the debate have been presented. One consistent barrier to research, it must be said, is that partisans of each position will too often display a reluctance to engage the other with a general good faith, leading to a polarized entrenchment where distorted claims about theoretical opponents are tossed back and forth, while important unfavorable evidence is conveniently ignored. Boddice (Reference Boddice2024), for example, has recently written that “the methods of the universalists have been found wanting on numberless occasions,” yet “the universalist school has not substantially engaged with any of this, but rather carried on regardless” (p. xi). While this is certainly true in particular cases, I believe it is somewhat misleading as a general statement, and (more importantly) an identical charge can be levied at anti-universalists: many constructivists refuse to engage with the ongoing developments of their theoretical adversaries, choosing instead to overwhelmingly focus on how the supposed refutation of a single research program (that of Ekman) on a single aspect of emotion (expression recognition) is enough to unequivocally and absolutely discredit all theories of basic or discrete emotion, despite the amassed evidence for patterns of regularity and consistency in different emotion domains across cultures. Indeed, many constructivists devote their time to attacking what Shiota (Reference Shiota2024) calls the “cartoon” version of BET, an oversimplified and essentially invented model of discrete emotion that is not actually maintained by any current researchers in the field (p. 313).

In this sense, basic and discrete emotion theories in 2024 have evolved to be significantly more complex and nuanced than the more rigid, traditional classic BET of the late twentieth century – that is, the BET that is most often held up for critique. Constructivist theorists, we have seen, have particularly attacked BET by homing on flaws in early emotional recognition studies – but, as Keltner, Sauter, et al. (Reference Keltner, Tracy, Sauter and Cowen2019) more recently note, the “field has moved beyond relying exclusively on forced choice labeling of expressions, and progress is being made in understanding how social perceivers infer intentions, motives, action tendencies, and relational properties of signaler and perceiver in brief expressions of emotion” (p. 135). To this end, Keltner and Cordaro (Reference Keltner, Cordaro, Fernández-Dols and Russell2017) ground their research in basic emotion theory, but argue that there is evidence for discrete multimodal expression patterns in far more emotions than the “Big 6” of traditional BET; Cordaro et al. (Reference Cordaro, Sun and Keltner2018), for example, identified “cross-cultural core patterns of expressive behavior [for] 22 emotions [and] a gradient of universality for the 22 emotions” (p. 75; see also Cordaro et al., Reference Cordaro, Sun and Kamble2020; Cowen & Keltner, Reference Cowen and Keltner2020). In terms of the brain, Celeghin, Diano, Bagnis, Viola, and Tamietto (Reference Celeghin, Diano, Bagnis, Viola and Tamietto2017) argue that “evidence in favor of the neurobiological underpinnings of basic emotions outweighs dismissive approaches,” when we realize that a theory of discrete emotions doesn’t strictly demand (as is sometimes claimed) a traditional one-to-one localization between anatomical structures and emotion signatures: “moving the focus of neuroscientific research from individual brain regions to networks, and from the simplistic region based one-to-one localizations to more sophisticated network-based one-to-many relationships between neural structure and function seems to prefigure a more modern and neurobiologically plausible approach to the study of basic emotions,” and will indeed prove that BET is “still tenable and heuristically seminal” (pp. 1, 8; see also Scarantino, Reference Scarantino2012c). A number of researchers seem to agree, emphasizing that basic emotions demonstrate a “discreteness … best understood as widespread, system-level patterned activity, rather than selective regional or systemic engagement during specific emotions” – that is, “neurophysiological data support the view of multiple discrete emotion systems that are organized in a distributed fashion across the brain, with no clear one-to-one mapping between emotions and brain regions” (Nummenmaa & Saarimäki, Reference Nummenmaa and Saarimäki2019, pp. 7, 3; see also Saarimäki et al., Reference Saarimäki, Ejtehadian and Glerean2018).

Thus, basic and discrete emotion theory is still a thriving and viable research program, which has not (despite what is sometimes said) been overturned by the rise of modern constructivist theories. Scarantino (Reference Scarantino, D’arms and Jacobson2014, Reference Scarantino2018; see also Scarantino & Griffiths, Reference Scarantino and Griffiths2011), for example, has worked on a “new BET” that reflects constructionist critiques (Reference Scarantino, Barrett and Russell2015, p. 334). He argues that “basic emotions are evolved programs that coordinate more basic ingredients such as facial muscle responses, autonomic blood flow, subjective experiences, respiratory and vocal change, motor patterns, thoughts, memories, images”; what “unites the ingredients of a basic emotion together is an evolved and specialized basic emotion program,” which “was selected to coordinate organismic resources to deal successfully with fundamental life tasks such as avoiding dangers, removing obstacles, coping losses” (p. 335). In the new BET, “what is universal when it comes to basic emotions are first and foremost the evolved programs that run them” – but variability in manifestation of basic emotion is to be expected, because these programs need to be context-dependent and flexible to address natural variation in the fundamental life tasks they are designed to address (p. 336). Accordingly, the theory crucially posits that there will be a variety of functional variants in basic emotion manifestation, that is, different ways that the ingredients of emotion will specifically present while still “preserving the task-oriented nature of the response” (p. 340). Reflecting this flexibility, basic emotions in this model thus become “input-output open affect programs … that allow for learning to affect both what activates the program (input) and what responses the program brings about (output)” (Reference Scarantino2018, p. 78). Denying first that basic emotions require a one-to-one correspondence between particular emotions and particular responses, and then that particular instances of the same basic emotion require a coordinated response, Scarantino suggests that a newly conceived BET can address the perennial objections of constructivists. What’s more, this new BET importantly sees basic emotions as not aligning with folk psychological emotion categories, but rather as reflecting “theoretically motivated subcategories such as unconditioned basic fear, conditioned basic fear, body-boundary violation basic disgust, core ingestive basic disgust, defensive basic anger, and so forth” (Reference Scarantino, Barrett and Russell2015, p. 363); the emphasis is on motive states, and flexibility of response (Reference Scarantino2018).

As already suggested, another modern expansion of basic/discrete emotion theory can be seen in the recent research of Cowen, Keltner, and colleagues. Their foundational work found that the evaluation of 2,185 short video clips resulted in “27+ linearly separable dimensions of reported emotion experience” (Cowen & Keltner, Reference Cowen and Keltner2018, p. 275), dimensions that are “represented within a semantic space best captured by categorical [emotion] labels” – but the “boundaries between categories of emotion are fuzzy rather than discrete,” meaning that “many categories of emotional experience share smooth gradients with other semantically distinct categories, forming smooth transitions between particular varieties of reported emotional experience” (Cowen & Keltner, Reference Cowen and Keltner2017, pp. E7900, E7903). Functional magnetic resonance imaging (fMRI) analysis equally showed evidence for these more than twenty-seven distinct emotion categories, finding that “neural representations of diverse emotional experiences during video viewing are high-dimensional, categorical, and distributed across transmodal brain regions”; furthermore, “ratings of individual emotions could accurately be predicted from activity patterns in many brain regions, revealing that distributed brain networks contributed in distinct ways to the representation of individual emotions in a highly consistent manner across subjects” (Horikawa et al., Reference Horikawa, Cowen, Keltner and Kamitani2020, pp. 14, 13). Subsequent studies found evidence for twenty-four distinct kinds of emotional experience in vocal bursts (Cowen, Elfenbein, et al., Reference Cowen, Elfenbein, Laukka and Keltner2019), and twenty-eight emotions evident in naturalistic expression (Cowen & Keltner, Reference Cowen and Keltner2020; see also Cowen et al., Reference Cowen, Keltner and Schroff2021).

Such “high-dimensional taxonomy of emotion” (Cowen, Sauter, et al., Reference Cowen, Sauter, Tracy and Keltner2019, p. 85) led to the development of semantic space theory, a computational approach suggesting “upwards of 25 distinct kinds of emotions, each with their own patterned profile of associated responses” (Cowen & Keltner, Reference Cowen and Keltner2021, p. 132). Contrary to the position of constructivists, this model holds that distinct categorical labels better account for emotional experience than a small number of dimensions like arousal and valence – indeed, in this high-dimensional approach, “dimensions of semantic spaces of emotion most typically correspond to individual emotions” (Keltner, Brooks, & Cowen, Reference Keltner, Brooks and Cowen2023, p. 243) – but at the same time, it distances itself from traditional categorical methods of BET, arguing that the fuzziness of boundaries renders individual emotions distinct but not discrete. Still, as we have seen, Keltner, Sauter, et al. (Reference Keltner, Tracy, Sauter and Cowen2019) and Keltner, Tracy, et al. (Reference Keltner, Tracy, Sauter and Cowen2019) nonetheless argue for the viability of a modern, revised basic emotion theory, and his semantic space model can in some ways be seen as an example of it, in the sense that it uses new analytical techniques to conclude that “21 emotions meet Ekman’s criteria for ‘basic’ emotions, being associated with distinct antecedents, experiences, expressions, and neurophysiological correlates” (Keltner, Brooks, & Cowen, Reference Keltner, Brooks and Cowen2023, p. 245). A recent study (Floman et al., Reference Floman, Brackett and LaPalme2023) also found that participants reliably matched core relational themes (Lazarus, Reference Lazarus1991) to twenty-four distinct emotions, providing evidence for semantic space theory and the foundational, universal nature of its emotional dimensions.

Defining Universality

The universality question, then, is far from settled. This is particularly apparent when we also consider the fact that the very notion of “universality” is itself a more flexible and fluid designation than is often thought. For good reason, the matter of universality is often understood (particularly by anti-universalists) to reflect a rigid, absolute binary. But as Russell and Fernández-Dols (Reference Russell, Fernández-Dols, Russell and Fernández-Dols1997) noted long ago, when speaking of universality we need not be beholden to “the presupposition that we face an either-or choice: either randomness … or full universality” (p. 16) – that is, somewhat surprisingly, there are actually many different kinds of universality, which makes the universality of emotion a more complex and more potentially plausible proposition.

In the context of facial expression research, Russell and Fernández-Dols thus proposed the notion of minimum universality, which “predicts a certain amount of cross-cultural similarity in interpreting facial expressions without postulating an innate emotion signaling system”; the designation minimal, in this formulation, is “meant to emphasize that at least this much universality appears to exist” (Reference Russell, Fernández-Dols, Russell and Fernández-Dols1997, p. 16). Other options also exist. In linguistics, for example, it is perfectly normal to differentiate between “two senses of ‘universal’: the absolute (exceptionless) and the statistical (tendency-based) universal” (Bickel, Reference Bickel, Enfield, Kockelman and Sidnell2014, p. 110). While these concepts have a technically precise usage in the context of language science, Hogan (Reference Hogan2003) has fruitfully employed them outside of linguistics in his exploration of literary (and more generally, human) universals, demonstrating that universality need not only refer to things that occur in all instances without exception (i.e. absolute universals) but can also reflect things that occur in populations significantly more than we would expect by chance alone (i.e. statistical universals). Speaking of emotional expression, for example, Lee and Anderson (Reference Lee and Anderson2016) note that there are some general patterns of cross-cultural consistency – in the sense that “if our expressions were purely higher-order associations, each shaped arbitrarily for social communication, there could not be any recognition of expressions across cultures” – which means that “basic expressions need not be universal in the strong sense but in having maintained statistical stability across the myriad of influences of culture and context they would indicate a common ancestry” (p. 497). Other scholars have similarly made distinctions between different types of human universals, such as Lonner’s (Reference Lonner, Triandis and Lambert1980) seven-level structure or Brown’s (Reference Brown, Wilson and Keil2001) five-plus categories. On a different note, in anthropology the notion of contingent universals refers to the shaping rules and principles that govern a particular cultural context (Harré & Llored, Reference Harré, Llored, Jovanović, Allolio-Näcke and Ratner2018; Shweder, Reference Shweder1991). And in terms of psychological universality, Norenzayan and Heine (Reference Norenzayan and Heine2005) differentiate between four levels, reflecting how various cultures employ cognitive tools: they posit “nonuniversals (different tools), existential universals (same tool but differential functions), functional universals (same tool and same function or use but differential accessibilities), and accessibility universals (same tool, use, and degree of accessibility)” (p. 772). Manokara and Sauter (Reference Manokara, Sauter and Leersnyderforthcoming) have recently used these criteria to consider existing evidence about emotion, finding different degrees of universality at different levels of analysis. In terms of emotion, it seems clear that the general concept of universality must not be seen as an absolute/absent binary, as there are many different ways that emotions may be thought (or not thought) to be meaningfully “universal.”

Another common confusion regarding emotional universality entails the assumption that advocating for universal emotions inherently entails arguing for biological innateness. While it is certainly true that many theorists believe that universal aspects of emotion are “hardwired” into human biology – and many constructivists argue that the lack of discernible biological “signatures” proves that emotions are not universal – it is perfectly possible that some aspects of human emotion might be meaningfully universal but not biologically innate. As Hogan (Reference Hogan2008a) notes, human universals (including universals of emotion) might arise “from regularities in the physical environment, recurring developmental experiences that are not genetically programmed, emergent features of the phenomenology of self-consciousness, convergent results of group dynamics, network patterns, etc.” (p. 203). Similarly, Camras, Fatani, Fraumeni, and Shuster (Reference Camras, Fatani, Fraumeni and Shuster2016) observe that “what are often considered to be universal discrete emotions” may actually reflect “universal environmental circumstances,” in the sense that the apparent “innateness” of emotion may lie “in the universality of environmental ‘control parameters’ that engender the organization of responses as much as in the sets of responses themselves” (pp. 262–263). Clore (Reference Clore, Fox, Lapate, Shackman and Davidson2018) suggests that “there may be more ‘basicness’ and ‘universality’ in … schemas and scripts about emotions than in emotions themselves” (p. 189). Power and Dalgleish (Reference Power and Dalgleish2015) write that the apparent universality of certain “basic” emotions might reflect that “there are a number of common and central appraisal scenarios, distinct from each other, which emerge in human societies and which underlie and shape emotional development,” even if “the existence and development of these appraisal scenarios will differ somewhat across cultures” (p. 66). And Majeed (Reference Majeed2023), taking a discrete approach, advocates for “a new version of faculty psychology that takes emotions to be products of developmental modules: non-innate systems which behave like modules, but form on the basis of various ontogenetic processes” (p. 1439). These possibilities are all at least plausible, so it seems clear that the truth of human emotional universality will not hinge on the proving or disproving of innate, biologically affective signatures.

It is also important to note that investigations into the universality of emotion are naturally complicated by the fact that when scholars consider emotion, they are often examining very different things. Beatty (Reference Beatty2019) reminds us that even the question of what emotions are “depends on what level of analysis you are using, what order of reality you are describing” (p. 18), and this has obvious implications for any assessment of emotional universality. We find this generally acknowledged in an exchange between neuroscientist Jaak Panksepp and psychologist James A. Russell concerning categorical versus dimensional models of affect: Panksepp (Reference Panksepp, Zachar and Ellis2012) notes that “one key point of consilience” might be found in the fact that “Jim’s empirical work proceeds completely at the tertiary-process level of psychological analysis, and mine largely at the primary process level” (pp. 313, 317). Decades ago, Mesquita and Frijda (Reference Mesquita and Frijda1992) recognized that “whether cross-cultural differences or similarities are found depends to an important degree on the level of description of emotional phenomenon” (p. 179); recent cross-cultural analysis of emotional expression suggests that while display rules vary in different cultural contexts, there are certain stable underlying principles, such as “a universal norm for the management of emotional expressions” and a universal “display rule norm for greater expressivity toward ingroups than toward outgroups” (Hwang & Matsumoto, Reference Hwang, Matsumoto, Carducci and Nave2020, p. 260). Furthermore, Manokara and Sauter (Reference Manokara, Sauter and Leersnyderforthcoming) crucially argue that “the question of whether emotions are universal or not must consider different components of emotions separately, and should recognise the possibility that the answers may not be consistent across components” – in other words, “the degree to which cross-cultural similarities and differences are observed may depend on which component of that emotion is being examined,” and thus “universality for different components of emotions can be most usefully understood within a framework that differentiates between different degrees of universality” (p. 2).

What’s more, the “psychological construct of a universal human nature,” Lewis, Al-Shawaf, Conroy-Beam, Asao, and Buss (Reference Lewis, Al-Shawaf, Conroy-Beam, Asao and Buss2017) note, “refers to species-typical psychological mechanisms, not universal manifest behavior” (p. 366), so it is quite obvious that the question of whether all human beings share universal emotion potentials cannot simply be answered by pointing to the obvious fact that all human cultures exhibit different emotional practices (see Mesquita, Reference Mesquita, Mayne and Bonanno2001). This leads Smaldino and Schank (Reference Smaldino and Schank2012) to observe that “the problem with identifying invariant emotional categories is complicated by the fact that people are adaptive systems, whose emotions are highly flexible,” meaning that universal “invariants must therefore be sought in the mechanisms that allow them to produce these feelings, [the] mechanisms of affective responses” (p. 164) – and indeed, it has long been recognized that “constructionists have tended to focus on the management of emotions, in response to emotion norms; and positivists, on their production” (MacKinnon, Reference MacKinnon1994, p. 126). Finally, it may also simply be the case that certain emotions are more universal than others. For example, some scholars argue that core primary emotions – evolutionarily basic, largely hardwired responses to adaptive challenges – mix together to form more culturally elaborated secondary (or even tertiary) emotions (Kemper, Reference Kemper1987; Plutchik, Reference Plutchik1980, Reference Plutchik2003; J. H. Turner, Reference Turner2000, Reference Turner2021). These sentiments still have a biological basis, and may be part of a universal psychic architecture, but they are shaped much more specifically by social forces: Turner and Stets (Reference Turner and Stets2005), for example, suggest that “the most reasonable conclusion is that the expression of primary emotions is hardwired and universal, and that the capacity for first- and second-order emotions is also wired into the human neuroanatomy, with the gestural expression of these emotions determined by socialization into the emotion culture of a society” (pp. 19–20).

Hybrid Models

Thus, there are many ways to think about emotional universality in a more nuanced manner than is sometimes assumed – and, even more importantly, such nuances allow for potential reconciliation with anti-universalist positions. As the first two sections of this Element have indicated, there is undeniable evidence for both consistency and variance in cross-cultural human emotional experience, and for this reason scholars from many disciplines have become increasingly inclined to develop models of emotion that account for both similarities and differences. In the general realm of psychology, for example, some scholars like Colombetti have addressed the issue by using a dynamic systems approach, which views emotions “as complex dynamical patterns of brain and bodily events” (Reference Colombetti2013, p. 46; see also Colombetti, Reference Colombetti2009); the model of Lewis & Liu (Reference Lewis and Liu2011), for example, “integrates the nativist assumption of prespecified neural structures underpinning basic emotions with the constructionist view that emotions are assembled from psychological constituents” (p. 416), while that of Cunningham, Dunfield, and Stillman (Reference Cunningham, Dunfield and Stillman2013) aims at “not only understanding the homogeneity of emotional experience (the similarities among instances of ‘fear’), but also the heterogeneity of emotional experiences (the differences among the ‘fear’ episodes” (p. 353). Most recently, in fact, Wood and Coan (Reference Wood and Coan2023) have explicitly argued that a dynamic systems approach can reconcile the seeming opposition of basic/discrete emotion theories and constructivist theories on the matter of universality: they propose that “at the level of goal-directed behavior, emotions are relatively universal, discrete, and adaptive – one might say evolved – but also necessarily constructed through the body’s interaction with the environment” (p. 443). There is also the network psychometric model of Lange, Dalege, Borsboom, van Kleef, and Fischer (Reference Lange, Dalege, Borsboom, van Kleef and Fischer2020; see also Lange & Zickfeld, Reference Lange and Zickfeld2021), in which “emotions are conceptualized as systems of causally interacting emotion components”; attempting to unite the major strands of modern emotion theory, this approach allows for “(a) identifying distinct emotions (central in affect-program theories), (b) between- and within-person variations of emotions (central in constructionist theories), and (c) causal relationships between emotion components (central in appraisal theories)” (p. 444). Other approaches invite us to reframe our very understanding of seemingly firm concepts like “construction.” Acknowledging the seemingly indisputable evidence that some aspects of emotion are constructed, Parkinson (Reference Parkinson2012) nonetheless moves beyond traditional models of social construction, exploring instead how “a set of processes whereby interpersonal, institutional, and cultural factors make their mark as emotions are put together over time”; crucially, in this account “no absolutist claim that all emotions are fully constituted by these processes is intended” (p. 291).

Beyond psychology, compromise positions are becoming increasingly important in other disciplines. In sociology, Ariza (Reference Ariza2021) writes, “the need has been identified for the sociology of emotions to add complexity to its formulations of perception and experience by incorporating their biological correlates, without neglecting the social and phenomenological aspects” (p. 168); Bericat (Reference Bericat2016) argues that the field needs “a greater degree of interaction between theory and empirical research, structural and cultural approaches and micro- and macro-perspectives” (p. 505). To this end, Demertzis (Reference Demertzis2020) adopts an “intermediate approach of mild constructionism,” reflecting that “basic emotions provide a minimum of affective-cultural universals, a thin foundation whereby an infinite array of situationally formed emotions flourish” (p. 44). Similarly, J. H. Turner (Reference Turner2021) believes that “we do not have to take a firm stand on the essentialist and constructivist arguments for the nature of emotions,” when we acknowledge that “many emotions … are biology based, with this base being expanded by reflexive thinking about emotions, by collective and self-discourse about emotions, and by codification of various beliefs and normative views on emotions that are part of a population’s and various subpopulations’ culture” (p. 177).

In 1996, Leavitt noted that “anthropology is divided between views of the emotions as primarily biological and as primarily sociocultural in nature” – but just as he then proposed “ways around the meaning/feeling dichotomy,” so too have other, more recent anthropologists adopted a more hybrid position (p. 514). Such approaches note that emotions must “be thought of as both biological and cultural, as consisting of both physical feeling and cultural meaning” (Milton, Reference Milton2005, p. 198). Quinn and Mathews (Reference Quinn and Mathews2016), for example, suggest that “the highly variable selves that ethnographers have documented cross-culturally all build upon the universal human self described by neurobiologists”; for them, “the link between cultural selfhood and this neurally-based self is emotional arousal” (p. 359). Stodulka (Reference Stodulka and Storch2017) notes that while “feelings are related to biographical, social, politics, and cultural dimensions,” a “too skeptical Geertzian rhetoric that rejects the universal biological dimension of emotion-related phenomena and withdraws to a constructivist phenomenology obstructs anthropology’s epistemological resourcefulness” (pp. 14, 11). His “integrative anthropology of emotion” thus sees affects as “bio-cultural processes [that] relate physiological arousals and their cognitive appraisals with their surrounding local worlds in terms of a mutually shared cultural rhetoric” (p. 15).

More elaborately, Beatty (Reference Beatty2019) considers “emotion in its different guises and from different perspectives,” in “recognition of the diversity of emotion in form, context, and use”; accordingly, he does “not come down decisively for or against a natural kind view of emotion, or even for or against the coherence of the emotion concept,” but rather takes a “strategic scepticism [that allows him] to open things up, recovering some of the essential detail of person, history, and circumstance that has been lost in reductive approaches” (p. 12). Accordingly, he develops a “narrative approach” that “does not depend on a view of emotions as discrete biological events governed by mechanisms that evolved to deal with threats and opportunities, [but] nor does it embrace an out-and-out cultural relativism” (p. 17). His outlook – which anchors both the apprehension and reporting of emotion within a complex account of emotion’s core narrativity – thus affirms “a multiplex reality [that] salvages the bridge to cross-cultural understanding that relativism would sweep away,” but that “also avoids the empiricist limitations of basic emotions theories that tend to ignore all that concerns the anthropologist – motives, plans, meaning, the social drama” (p. 19). When describing, for example, an anthropological event like a funeral scene, he rightly notes that “pan-human, culturally specific, [and] idiosyncratic” factors are all at play (p. 193); in this, he follows Nussbaum (Reference Nussbaum2001), who notes that “all human emotions … bear traces of a history that is at once commonly human, socially constructed, and idiosyncratic” (p. 177).

In linguistics, we have seen how certain primary metaphors seem to consistently reflect universal cognitive architecture, but Winter and Matlock (Reference Winter, Matlock and Hampe2017) remind that this doesn’t mean that such metaphors are simply acultural; they suggest that “linguistic and cultural reflections of primary metaphors may thus ‘feed back’ into the underlying conceptual structure,” leading to the conclusion that “primary metaphors are both cultural and embodied” (p. 100). For example, the cognitive linguist Kövecses (Reference Kövecses2000a) argues that the evidence of language use points to a “body-based social constructionism,” a view that sees emotions like anger as “both (near) universal and culture-specific,” in the sense that they reflect “universal elements of the body (human physiology) and culture-specific elements of cultural explanation (of different kinds)” (p. 169; see also Kövecses Reference Kövecses2000b). Perovíc and Vuković-Stanatović (Reference Perovíc and Vuković-Stanatović2021) have more recently found that the “level of universality and culture-specificity depends on how generally we define the conceptualization” of the metaphor – their analysis of love metaphor, metonymy, and cultural scripts suggests that “more general and abstract metaphors displayed more universality [i.e. metaphors at the superordinate-level], whereas more cultural specificity was likely to be found in the basic-level metaphors, i.e. narrower metaphors” (p. 45).

Finally, we saw in Section 2 how historians tend to be vehement opponents of universalism, and even those that incorporate the empirical sciences into their approach will often still reject the notion that any aspect of emotion is meaningfully universal. There are, however, some exceptions. Reddy, we saw, attacked basic emotion theory (Reference Reddy2020), but his entry into the field (Reference Reddy1997) was entitled “Against Constructionism”; more notably, his most enduring work, The Navigation of Feeling (Reference Reddy2001), elaborates a “formal theory that establishes emotions as largely (but not entirely) learned,” a model that “leaves plenty of room for cultural variation” but also posits “a core concept of emotions, universally applicable, that allows one to say what suffering is, and why we all deserve to live in freedom” (p. xi). More recently, Dixon (Reference Dixon2020) – who, we saw, also attacked universalizing discourses – notes that one might oppose basic emotion theory in its narrowest conception while still acknowledging that there may be “some physiologically grounded dimensions to … emotion-like experiences in all cultures” (p. 29). But perhaps the most promising compromise model is that of Firth-Godbehere (Reference Firth-Godbehere2021), who uses the mechanics of disgust to exemplify “the way emotions can be understood from both the universalist and constructivist standpoints”: it “could easily be the case,” he writes, that “there’s a basic evolved feeling designed to stop us from poisoning ourselves and picking up parasites,” which “is then adapted, shaped, and manipulated by culture” (p. 259). In this way, it is perfectly possible for historians to be sensitive to the historical and cultural contingencies of emotion without denying that certain functional affective properties may be generally consistent across the species.

The Biocultural View

Beyond such specific disciplinary approaches, scholars are increasingly acknowledging that our analysis of emotion should be more cross-disciplinary, and that we should attempt to bridge theoretical boundaries by drawing on the insights of both traditionally universalist and traditionally anti-universalist disciplines. Indeed, Tappolet (Reference Tappolet2023) has recently concluded that “neither the arguments for biological determinism nor the ones for social constructionism are convincing,” because it is more “plausible that emotions are instead seamless products of the interaction of nature and culture” (pp. 39, 55).

In the study of emotion and beyond, an increasing number of scholars (like, we saw above, Stodulka) are thus embracing some sort of biocultural approach, which J. Carroll et al. (Reference Carroll, Clasen and Jonsson2018) define as an “integrative research program designed to investigate the causal interactions between biological adaptations and cultural constructions”: that is, it posits that “cultural processes are rooted in the biological necessities of the human life cycle,” while “human biological processes are constrained, organized, and developed by culture” (p. 1). One must be slightly wary with terminology, because in some places (particularly in humanist spaces) biocultural implies less of a synthesis between universalist and constructivists positions than one might suspect; for these scholars, J. Carroll (Reference Carroll2022b) notes: “the biological part of the biocultural model is only the domain-general brain that enables culture,” meaning that they use the term biocultural to account for an outlook that is, in practice, almost entirely constructivist (p. 83). But there are plenty of more conciliatory instances. Such versions of a biocultural outlook are premised on culture-gene coevolutionary theory, which posits that “cultural and genetic selection both affect how the mind and brain give rise to behavior” (Chiao & Immordino-Yang, Reference 63Chiao and Immordino-Yang2013, p. 57; see Boyd & Richerson, Reference Boyd and Richerson1985); indeed, decades ago, Nesse (Reference Nesse1990) noted that “debates about whether traits result from nature or nurture” will increasingly “become less simplistic as more people realize that all phenotypes represent the outcome of genotypes interacting with environments” (p. 279).

Accordingly, scholars now focus on “the feedback loops between genetic evolution and cultural practices across the course of human and prehuman evolution” (J. Carroll et al., 2018, p. 3), and biological reductionism is avoided by acknowledging that “cultural evolution affects biological evolution, and that cultural practices can exert pressure on the human genome,” in the sense that “physical, but also material, historical, social and cultural environments in which we live, and even the type of human relations characterizing our life, all influence the expression of our genes” (Wojciehowski & Gallese, Reference Wojciehowski and Gallese2018, 10). Sauter and Russell (Reference Sauter and Russell2024) elaborate:

“In short,” they conclude, “over time, changes in culture can lead to genetic changes; and genetic differences can influence culture.” At the level of individuals, an increased understanding of the brain’s neuroplasticity similarly attests to the viability of a biocultural approach (Rubin, Reference Rubin2009). Reconciling universalist and constructivist approaches, the biocultural outlook thus allows us to imagine a model of the mind in which emotional mechanisms “do not start out as empty shells, devoid of information” – certain “circuits are already there, unfolding over development, influenced but not totally constructed by experience” (Ekman & Cordaro, Reference Ekman and Cordaro2011, p. 367). Or, as Jonsson (Reference Jonsson2021) succinctly puts it, “human emotions are flexible within constraints” (p. 28).

Mason and Capitanio (Reference Mason and Capitanio2012), for example, reconfigure the concept of basic emotions within terms of epigenetic development:

Buck (Reference Buck2014) similarly theorizes a developmental-interactionist outlook, which he terms a “general biosocial approach that views emotions in terms of emergent systems involving an interaction between biological potential and social experience over the course of development” (p. xx). Recent social functionalist theory (Keltner et al., Reference Keltner, Sauter, Tracy, Wetchler and Cowen2022) equally considers “the bidirectional influences between culture, relationship, and emotion” (p. 388); a model that “extend[s] Basic Emotion Theory,” it focuses on how discrete emotions organize around human social relational needs, but still acknowledges that “culture shapes emotion in profound ways, in the beliefs, practices, rituals, ceremonies, and institutions that shape the contexts, appraisals, and forms of conceptualization that imbue emotion with culturally specific meaning” (pp. 389, 395; see also Keltner, Haidt, & Shiota, Reference Keltner, Haidt, Shiota, Schaller, Simpson and Kenrick2006). Similarly, some recent work on emotion in cross-cultural psychology now explicitly frames itself in biocultural terms, with goals of “integrating biological universality with cultural specificity”; this model posits “a biologically based core emotion system that is calibrated, regulated, and elaborated by culture,” and acknowledges that “different domains of emotion are more relatively influenced by biology or culture” (Matsumoto & Hwang, Reference Matsumoto, Hwang, Matsumoto and Hwang2019, pp. 361, 398; Matsumoto & Hwang, Reference Matsumoto and Hwang2012, p. 91). Kamiloğlu, Cong, Sun, and Sauter (Reference Kamiloğlu, Cong, Sun, Sauter, Al-Shawaf and Shackelford2024) vitally note that “evolved mechanisms do not imply strict uniformity,” and thus argue that “evolved psychological mechanisms result in cultural differences instantiated as variations around common themes of human universals” (p. 983). And Watzl (Reference Watzl and Valsiner2019) goes even further, suggesting that “social mechanisms may be so deeply intertwined with other biological mechanism that it makes no sense to ask about their relative contributions” (p. 46). It seems clear that, as Boddice (Reference Boddice2024) has recently put it, “there is no culture-free or value-neutral context to the study of human ‘nature,’ and there is no ‘nurture’ without framed biology” (p. 13).

A particularly valuable biocultural account of the complex entanglement of emotional regularity and emotional variance may be found in the ongoing efforts of researchers working under the umbrella of cultural neuroscience (Chiao, Reference Chiao2009, Reference Chiao2015, Reference Chiao2018; Chiao & Immordino-Yang, Reference 63Chiao and Immordino-Yang2013; Chiao et al., Reference Chiao, Li, Seligman and Turner2016; Han, Reference Han2017; Han et al., Reference Han, Northoff and Vogeley2013), a relatively new, multidisciplinary approach that draws from fields such as cultural psychology, anthropology, and cognitive science. Put simply, cultural neuroscience “examines how cultural and biological mechanisms mutually shape human behavior across phylogenetic, development, and situational timescales” (Chiao, Reference Chiao2015, p. 282; emphasis added); it is thus premised on culture-gene coevolutionary theory. Because of its linkage to cultural psychology, much work in cultural neuroscience is concerned with investigating the “neural substrates of cultural diversity of human cognition and emotion” and how “individuals from different socio-cultural contexts show distinct patterns of brain activity in cognition and behavior” (Han et al., Reference Han, Northoff and Vogeley2013, pp. 337, 338). Yet, at the same time, the field equally recognizes “a hierarchy of neural universals within levels of the nervous system,” including “neural states, ranging from molecules to systems, that are … observable to people of all cultures,” “neural states that have the same function or are physically implemented for the same purpose across situations or cultural contexts,” and “neural states that are implemented with the same degree of ease and frequency across cultural contexts” (Chiao, Reference Chiao2018, p. 35). As such, the final goal of cultural neuroscience is to “unveil both culturally universal and culturally unique neural processes” (Han, Reference Han2017, p. 22), and researchers are indeed “discovering both generalizable and culturally specific mechanisms of the mind, brain, and behavior” (Chiao et al., Reference Chiao, Li, Seligman and Turner2016, p. xix).

Cultural neuroscience shows a particular interest in the working of emotion, and research has already demonstrated how “cultural meaning-making shapes the biological correlates of emotional feelings” (Immordino-Yang & Yang, Reference Immordino-Yang and Yang2017, p. 34) and how the brain’s construction of emotion is “at least partly open to cultural influence” (Immordino-Yang, Yang, & Damasio, Reference Immordino-Yang, Yang and Damasio2014, p. 1). But most exciting is how the field promises to synthesize various insights from the major (competing) models of emotion that we have examined in the previous two sections. According to Chiao (Reference Chiao2015), cultural neuroscience helps us realize that “the four classical theories of emotion (e.g. basic emotion, cognitive appraisal, psychological construction, and social construction) are all, to a certain degree, simultaneously correct,” in the sense that each explores “processes of emotion generation and experience, but each at a different level of analysis, and all of which are important and necessary for adaptive emotional behavior to occur” (p. 281). Cultural neuroscience thus provides an overarching framework for attending to both “the dual processes of genetic and cultural inheritance” – the domain of BET and social constructionism – and “the refinement of cognitive and neural architecture that accompanies genetic and cultural selection of human emotional behavior” – the domain of appraisal theory and psychological constructionism. For this reason, it is an especially promising methodology for moving beyond the simplistic opposition of emotional universality and emotional variance. A vaguely related approach with a somewhat different focus has been called cross-cultural affective neuroscience (CAN): emerging from the work of Panksepp, it studies the “two-way interaction between self and culture,” in terms of how “(1) universally shared subcortical affective systems are initially regulated uniquely in each mother-infant bond and subsequently by family models and cultures and (2) culture, by effecting family models and mothering styles, influences the degree to which subcortical basic affective systems are reinforced or inhibited” (Özkarar-Gradwohl, Reference Özkarar-Gradwohl2019, p. 4).

Underlying Structure

Biocultural approaches thus have great potential to account for both patterns of emotional consistency and emotional variance, in a way that complicates any easy distinction between universalist and anti-universalist perspectives. The existence of variance within a universal framework might also be clarified if we turn to approaches positing that emotional particularity emerges from some sort of underlying universal structure. For example, in the 1980s and 1990s, scholars in different fields took a prototype perspective, which generally posited that cross-culturally ubiquitous prototype models of emotions lie behind cultural particularity (Fehr, Reference Fehr1988, Reference Fehr1994, Reference Fehr and Baldwin2005; Fehr & Russell, Reference Fehr and Russell1984, Reference Fehr and Russell1991; Lakoff & Kövecses, Reference Lakoff, Kövecses, Holland and Quinn1987; Russell & Fehr, Reference Fehr1994; Shaver, Wu, & Schwartz, Reference Shaver, Wu and Schwartz1992; Shaver et al., Reference Shaver, Schwartz, Kirson and O’Connor1987).

Parrott (Reference Parrott2010, Reference Parrott2012) and Frijda and Parrott’s (Reference Frijda and Parrott2011) more recent concept of ur-emotions is a particularly helpful, prototype-like model that identifies a structure underpinning emotional variance. Proposing that we move away from the concept of basic emotions, Parrott introduces the idea of ur-emotion, which “refers to the structure or archetype that underlies an emotion that is evolutionarily related and recognizably shared between cultures and species, but [is] not identical to the occurring emotion itself” (Reference Parrott2010, p. 20). According to this theory, “it is not multicomponential response patterns [of emotion] that are universal … but rather something more abstract”; ur-emotions are thus “the core element in multicomponential response patterns” (Frijda & Parrott, Reference Frijda and Parrott2011, pp. 406, 407), in the sense that they are “an aspect of actual emotional states, but are not themselves actual, occurring emotions” (Parrott, Reference Parrott2010, p. 20). These core elements, more specifically, are “motive states or states of action readiness” that “specify particular motivational, motor, and cognitive response processes” – that is, they reflect certain universally available human response patterns, such as affiliate (“achieving or accepting close interpersonal interaction”), reject (“refusing interaction”), and desire (“achieving positive hedonic outcome”) (Frijda & Parrott, Reference Frijda and Parrott2011, pp. 406, 408, 409). (These tendencies are similar to Lazarus’s [Reference Lazarus1991] notion of core relational themes.) The great value of this approach is that “ur‑emotions can be perceived in similar but nonidentical emotions in different cultures or species”; so, for example, “when comparing the anger of the British to that of the Utku Eskimos or Yanamamo native South Americans, ur-emotion suggests the similarity without suggesting that the vast differences in socialization, evaluation, social norms, cultural centrality, and social consequences have no effect on the emotion’s form, function, or subjective experience” (Parrott, Reference Parrott2010, pp. 20, 18). Parrott elaborates:

The theory of ur-emotions, then, explains how the universal motivational state of submission can variously form the basis of culturally inflected emotional experiences such as shame, awe, admiration, humility, or respect, depending on how the event is appraised and what additional affective components build upon the core. It also explains how the culturally particular sentiment amae – a word in Japanese referring to the pleasurable abandonment of oneself to a caregiver, and an emotion usually said to be nonnative to the affective repertoire of Anglo-Americans – still emerges from the cross-culturally universal position of submission (Frijda & Parrott, Reference Frijda and Parrott2011, p. 411). This, it seems to me, is a quite sensible way to account for the fact that emotions are culturally constructed from universal human potentials.

From the perspective of linguistics, universal structure is also featured in the theory of Natural Semantic Metalanguage (NSM), a model that similarly tends to both emotional universality and cultural particularity (Harkins & Wierzbicka, Reference Harkins and Wierzbicka2001; Wierzbicka, Reference Wierzbicka1986, Reference Wierzbicka1992, Reference Wierzbicka1996, Reference Wierzbicka, Athanasiadou and Tabakowska1998, Reference Wierzbicka1999, Reference Wilkowski, Meier, Robinson, Carter and Feltman2009; for overviews see Durst, Reference Durst2003; Goddard, Reference Goddard, Xu and Taylor2021). Like Ortony, Clore, and Collins’s (Reference Ortony, Clore and Collins1988) attempt to “specify, in as language-neutral a manner as possible, the characteristics of distinct emotions” (p. 9), NSM works by “establishing lists of semantic primitives that exist in all languages and that could allow differentiation between synonyms or their translation equivalents” (Santos & Maia, Reference Santos and Maia2018, p. 5):

Different feelings are conceptualized as sets of clauses within the metalanguage, meaning that “there is an almost infinite set of possibilities for emotions to be lexicalized in different societies, while at the same time, every emotion can be translated and therefore understood by every human being” (Santos & Maia, Reference Santos and Maia2018, p. 5). Most crucially, Wierzbicka argues that English psychological terms like anger and sadness are distinctly not universal, but they nonetheless can be distilled into scripts reflecting the universal meta-language, which allows us “to explore human emotions from a culture-independent perspective”: we can do this, she argues, “when we stop trying to describe other people’s emotions through English terms such as anger, sadness and fear (or even through their ‘scientific’ versions such as ‘affect program anger,’ ‘affect program sadness’ and ‘affect program fear’) and try to understand them instead through the conceptual vocabulary shared by speakers of all languages, that is, through universal human concepts” (Reference Wilkowski, Meier, Robinson, Carter and Feltman2009, pp. 3, 13; see also Wierzbicka, Reference Wierzbicka2014). Although NSM has been subject to criticism (Matthewson, Reference Matthewson2003; Riemer, Reference Riemer2006), the methodology has nonetheless been fruitfully used to examine a variety of emotion concepts (Gladkova, Vanhatalo, & Goddard, Reference Gladkova, Vanhatalo and Goddard2016; Goddard, Reference Goddard, Baider and Cislaru2014, Reference Goddard2015; Ye, Reference Ye, Harkins and Wierzbicka2001).

A final valuable prototype-like approach can be found within a particular strand of appraisal theory. It will be remembered from Section 1 that certain appraisal theories posit that the human appraisal mechanism gives rise to discrete, essentially universal emotions; other appraisal theories, however, split “emotional episodes into a large or even infinite number of subsets, each characterized by a unique situation and hence a unique pattern of appraisal values” (Moors, Reference Moors2014, p. 303). Scherer’s (Reference Scherer1984, Reference Scherer, Scherer, Schorr and Johnstone2001, Reference Scherer2009b) component process model belongs to this second category. This model sits somewhere between the extreme poles of basic emotion theory and constructivism; it proposes that “research move from a discrete emotion approach to an emotion process approach,” but also states that while it is true that “individuals construct their own categorization of experience and choose an emotion label accordingly” (i.e. the psychological construction outlook), it is also “essential to theoretically link this act to what precedes it in the emotion process, allowing at least some degree of prediction” (Scherer & Moors, Reference Scherer and Moors2019, pp. 738, 739).

In Scherer’s theory, humans respond to stimuli in light of several appraisal objectives, evaluating the events via a “subjective assessment … of consequences and implications on a background of personal needs, goals, and values” (Reference Scherer2009b, p. 1309); consequently, emotions are dynamic processes that emerge when such multilevel appraisals give rise to things like physiological response patterns, motivational changes, and categorization/verbal labeling. While Scherer may see universality in the appraisal mechanism (Reference Scherer1997), he distinctly does not see emotional experiences themselves as universal; the component process model “does not assume the existence of a limited set of discrete emotions or affect programmes,” but instead, because there is no limit to the number of ways that events could be subjectively appraised by an organism, “considers the possibility of an infinite number of different types of emotion episode[s]” (Reference Scherer2009b, p. 1316).

But even as it acknowledges the boundless nature of emotion, the component process model still accounts for patterns of emotional regularity, via the concept of modal emotions (Scherer, Reference Scherer, Ekman and Davidson1994, Reference Scherer, Davidson, Scherer and Goldsmith2009a). Scherer notes that in the course of normal human existence, “some combinations [of appraisals]” have tended “to occur more frequently than others as part of an emotional reaction,” probably “due to general conditions of life, constraints of social organisation, and similarity of innate equipment” that the species possesses; there are therefore “some major patterns of adaptation [in] life … that reflect frequently recurring patterns of environmental evaluation results” (Reference Scherer, Ekman and Davidson1994, pp. 25, 28; Reference Scherer2009b, p. 1316). He uses the name modal emotions to account for such prototypical “patterning of expression, autonomic arousal, action tendencies, and feeling states” that arise from the “prototypical pattern[s] of appraisal” elicited by common life conditions in the species (Reference Scherer, Ekman and Davidson1994, p. 30). Given “the prominence and frequency of occurrence of these episodes of highly similar emotional experiences, it is not surprising that they have been labeled with a short verbal expression, mostly a single word, in most languages”; for this reason, the subjective experience designated by the English word frustration (for example) is in effect “universal and ubiquitous,” in the sense that “all organisms, at all stages of ontogenetic development, encounter blocks to need satisfaction or goal achievement at least some of the time” (p. 28).

Crucially, with the theory of modal emotions, “no effort is made to find or define a definite number or homogenous, integral categories or mechanisms that justify an a priori definition of basic or fundamental” – instead, what’s acknowledged is the fact that certain prototypical appraisal patterns give rise to emotional episodes prototypically similar enough to have warranted inclusion in a broad linguistic or conceptual category (p. 30). Rather than positing discrete emotions or affect programs, the model thus postulates “fuzzy sets of modal emotions (as a result of a categorisation of qualia) on the background of an infinite variety of emotional processes and their qualia representations” (Scherer, Reference Scherer2009b, p. 1334). In this, it aligns with the suspicion of linguist Daneš (Reference Daneš and Weigand2004), who approached the universality versus cultural specificity of emotion debate by suggesting that “perhaps it would be [best] to use the metaphor of a field or space of fluctuating fuzzy elemental emotional states … with relatively ‘condensed islands,’ more or less different in various cultures and identified by them by means of particular labels” (p. 31).

It has recently been argued (Boddice, Reference Boddice2024) that all attempts to find universal underpinnings of emotion – such as Parrott’s ur-emotions, Wierzbicka’s NSM, and Scherer’s modal emotions – are inherently flawed, because they are bound to particular-language constructs. I find this position unconvincing, as it is premised on a linguistic relativism that is not, in my mind, supported either by common sense or by existing evidence, which points to countless forms of cross-cultural similarity and ubiquity in emotional language and emotional experience that cannot occur by chance alone. It is perfectly possible, for example, to find language-neutral ways to investigate whether “people of different cultures have similar subjective experiences when encountering equivalent situations” (Manokara & Sauter, Reference Manokara, Sauter and Leersnyderforthcoming, p. 11), and it goes without saying that individuals are not emotionally bound by the linguistic repertoire of their native language. While it is absolutely true that the Ifaluk concept of song and the English concept of anger are not exactly equivalent, it seems odd to me that we should not be able to talk about their structural and functional similarities within their respective cultures, and prototype approaches are one way to do so that does not rely on an uncritical collapsing of the distinction between them.

Concluding Thoughts

It seems likely that we are still far away from settling the “universality question.” It is definitely possible, of course, that new affordances might allow future research to make empirical discoveries that will shed light on the issue; for example, technological advancement and improved research design may allow us, at some point in the future, to conclusively determine whether emotions have certain explicit “signatures” in the body (see Levenson, Reference Levenson2014). But until then, it is likely that researchers from across all disciplines will continue to amass evidence for both points of cross-cultural and cross-temporal affective contact and points of cross-cultural and cross-temporal affective divergence, and it will be the job of researchers to create models of emotion that account for both sets of findings. Given this, it may actually be valuable, at least rhetorically, to strategically exchange the concept of universality for one of ubiquity. As noted, universality is a poisoned term for many – especially humanists – and it may be too difficult to resuscitate for general scholarly usage, even if more technical applications (as we have also seen) entail a much less rigid concept than is often assumed. Speaking instead of emotional ubiquity – that is, emotional forms that recur in many, most, or ostensibly all contexts – may be more palatable to cross-disciplinary audiences. Saucier, Thalmayer, and Bel-Bahar (Reference Saucier, Thalmayer and Bel-Bahar2014) offer a precise definition of observed ubiquity – “empirical occurrence in all members of a representative subset of all cultures,” which in turn “means a high probability of universality” (p. 200). But even used more informally, ubiquity may be a helpful way to get at regularities that occur in the emotional lives of human populations, but one that does not discount the obvious fact of emotional variability.

Whatever the case, it seems that valuable advancements will come when researchers are increasingly inclined to consider whatever perspective on emotional universality is not common in their native discipline or theoretical framework. Ariza (Reference Ariza2021), for example, recently notes:

In other words, I think we will be best served the more that universalist theorists attempt to account for emotional variance in their models, and the more that anti-universalist theorists attempt to account for emotional regularity in their models. Because, most fundamentally, human universality and human difference are bound to one another. Long ago, Huntington and Metcalf (Reference Huntington and Metcalf1979) remarked that “cultural difference works on the universal human emotional material” (p. 43); more recently, Hogan (Reference Hogan2008b) reflects that “there is no such thing as human culture or human cultural difference without human universality,” in the sense that “cultural difference is variation on human universality” (p. 145). Roughley (Reference Roughley2000) similarly wonders if “there aren’t some concepts whose applicability to all humans needs to be assumed in order to even make sense of any version of the thesis of cultural relativism”: if, for example,

Thus, it seems clear that we need to consider universality and difference in tandem.

In terms of emotional universality, I think the most sensible approach to the matter is encapsulated by Semin (Reference Semin2012): while acknowledging that emotions are by definition dynamic and constructed by socio-cultural contexts, he notes that “you can construct only by what your body, brain, and the social-physical conditions afford you to construct” (p. 230). That is, though human functioning (such as emotional experience) “finds different expressions in terms of the variable and evolving social environments they are part of and give rise to,” this functioning is nonetheless still “constrained by relatively invariant ecological, existential, material, and biological conditions” – so, in the broadest sense, “our knowledge is advanced by understanding the balance between variation and systematicity.” To this end, Matsumoto and Wilson (Reference Matsumoto and Wilson2022) have recently argued that “emotions represent an area of study that can be characterized by the simultaneous co-existence of both universality based on biological substrates and cultural differences based on learned constructions” (p. 926), and an increasing number of scholars are realizing this. Indeed, when reflecting on the content of their volume, the editors of The Nature of Emotion: Fundamental Questions (Fox et al., Reference Fox, Lapate, Shackman and Davidson2018) observed that “although the basic model and the constructivist model are often pitted against each other, most of the contributors … seem to acknowledge and, more importantly, make use of, both perspectives” (p. 404). As we move forward, this willingness to not only acknowledge but also attempt to reconcile universalist and anti-universalist positions will make us well positioned for the twenty-first-century study of emotion and beyond. I have surveyed many different disciplines in this Element, and the way that each field approaches such a reconciliation will undoubtedly be different – given that each varies in how it specifically understands matters of theory, evidence, and even “truth.” But I think that all scholars are capable of building, rather than burning, theoretical bridges, and it seems clear that reading and thinking widely, with an aim to integrating various perspectives, will only serve to further elucidate the fascinating contours of human emotional experience.