Authors Replies

Replies to Flanagan, Seok, Cudd, and Oh

Stephen T. Asma and Rami Gabriel

Columbia College Chicago, Illinois, USA

Asma, Stephen T. and Rami Gabriel. 2020. "Replies to Flanagan, Seok, Cudd, and Oh." Journal of Philosophy of Emotion 2, no. 2: 38-52.

Abstract: In our response to the comments from Owen Flanagan, Bongrae Seok, Ann E. Cudd, and Jea Sophia Oh, we address the role of normativity in the study of emotions, the nature of the dialectical model we put forward, and the place for rationality and ethics in our project. Some misunderstandings are clarified, and we accept helpful correctives from our gracious interlocutors.

Key Words: normativity, dialectics, ethics, shame, guilt

We thank our interlocutors for their thoughtful analyses of our book. In this response, we address each set of comments separately, though we have adduced the main themes of the critiques to the following concerns: the role of normativity in explaining emotions, the role of practical reason and rationality in emotions, the dialectical model, a taxonomy of shame and guilt, whether the developmental stages of emotions are isolable, anthropocentrism, and an eco-feminist lens on characterizations of the mind.


Flanagan’s critiques are substantive and helpful. In general, our disagreements seem to fall along two major fault lines: the issue of emotional exaptation and the issue of emotional correctness. Our project tends to stress the homologous similarities of affective states across the mammalian clade, and leads us to focus more on primary and secondary levels of the emotions; although the last third of our book is deeply concerned with tertiary,

[page 38]

higher-level, human emotions as well. Flanagan in contrast seems to be suggesting more discontinuity between tertiary emotions and their animal affective roots. He suggests that many human emotions are exapted, in the sense that their functions have changed dramatically in the context of culture and cognitive entanglement. We do not disagree, and find this tension more a matter of emphasis.

Our view is that core elements remain in many exapted emotions: “fear of losing my job” does indeed have phenomenological and physiological continuities with fear of snakes, albeit minor. Just as the pentadactyl hand digit plan is a homology for all tetrapods, such that the human hand, the bat wing, and the whale flipper all have the same basic forelimb parts, so too a mouse that’s afraid of a cat, a human being who is afraid that he’s about to be fired, and a baboon being afraid of a leopard are all experiencing some core similarities. Context and cognitive appraisal will certainly change the particular threats and responses, but the energetic urge to defend oneself or flee will be present, as will the amygdala activation, the production of stress hormones, and the increase in blood pressure and heart rate. The hippocampus and prefrontal cortex (and executive functions generally) will determine if threats are pressing, real, or actionable, but the negative affect (feeling states) and the behavioral responses will be continuous across clades.

Considering the aptness or fittingness of the emotions is an important pragmatic project precisely because, as we argue in the latter part of the book, cultural contexts domesticate the experience and expression of moral emotions. We admit that the question of the propriety of emotional expression and the relation of propriety to ‘correct ways’ of behaving within given cultural contexts is a worthwhile area of study. Our issue is that when this debate about norms becomes the main way to analyze emotions, deeming them correct or incorrect vis-à-vis contingent ethico-historical scenes, we risk overlooking the pre-cultural norm context that is the cradle of mammalian emotions.

As Flanagan states, we may be making too many assumptions about the phenomenological continuity of emotional states across species, and maybe even across time for the same human. Here, we might be guilty as charged. It may indeed be an article of faith or a promissory note that non-human animals experience similar phenomenal feeling states as humans. The privacy of other minds--particularly non-linguistic minds--makes this difficult to settle. Thinkers like Frans de Waal and Jaak Panksepp have decided to err on the side of caution and to attribute feelings, like suffering, to animals because the ethical stakes are so high. But De Waal (2018) also felt comfortable anthropomorphizing most human feeling states to animals. What is the quality and quantity of evidence that should lead us to attribute fear, anger, even love to non-human animals?

Nevertheless, the continuity of core affective neural systems, and possibly sentient states therein, makes strong constructionism unconvincing for us. Our main critique of constructionists is that they try to have it both ways: arguing that the emotions are minimally constrained judgements (when they want to celebrate the diversity and non-nativist aspects of emotions, and to dismiss Panksepp’s neuroscientific work), but also switching to arguing that emotions are innately constrained, automatic processes (like perception) when they need to grudgingly acknowledge the biological (fast cognition) aspects of emotions. That inconsistency is problematic, but may not speak against other forms of constructionism (e.g., the Sapir-Whorf hypothesis through Robert Solomon and Catherine Lutz).

[page 39]

Additionally, Lisa F. Barrett and other strong constructionists end up denying emotions to nonhuman animals and babies, and we find this both empirically dubious and evolutionarily naïve. Propositional bias rears its head again when we claim that only a speaking mind can give us warrant for an inner subjective state. Why cannot the moving body (behavior) be (under controlled and scientific constraints) a reliable window into inner subjective states? Indeed, the advantage of human self-report through language to describe inner feelings states may also obfuscate our understanding of the continuity of emotional states across time in the same subject. Describing and reporting my own inner states can be how the medium (speech and conceptualization) alters the message of affective and behavioral experience. We recognize that this is a bigger problem than we can solve here.

We rest on surer footing when we warrant affective continuity on evolutionary grounds. Evolutionary thinking about pain/pleasure as adaptive mechanisms--together with the general conservative tendency of variation and selection--suggests that the same motivating mechanisms for vertebrates to move toward resources and away from threats, would be conserved and remain as deep homologies (just like Hox genes are deeply conserved albeit heavily modified by epigenetics and so on). On this view, fear moves the animal away from threat because it feels terrible. Take away the feeling state and you may be able to even more directly slave the perceptual state to the motor action, but that is not how mammals seem to work. We are open to the possibility that such a direct path (perception to motion, without an affective mediator) characterizes the phylogeny of insects, and probably will characterize the phylogeny of artificial intelligence, but doubtful that evolution would bestow all the behavioral and chemical processes of fear to the mammalian clade, but only the felt sense of fear to a tiny branch of great apes.

Another of Flanagan’s major concerns is the question of emotional correctness. Our dismissal of the pursuit of emotional correctness is taken by Flanagan as an abandonment of the ethical project of norming the norms. He is worried that we are denying the ethical project of asking when an emotion (e.g., fear, pity, compassion, rage) is correct for an individual or society, and when it is misleading or in need of reeducation? But the disagreement is not as dramatic as it seems. Our focus there is not to discount norming our norms, but to move us away from the dominant strategy in analytic approaches of isolating truth-conditions for emotive language. Our reading of the majority of philosophy of emotion (and even good philosophers of intentionality, like Ruth Millikan) is that the inquiry frequently devolves into thought experiments and test-cases for assessing when the emotion is “true.” Our critique is that this is a strange worry to have (influenced by indicative language verificationism) when emotions are mostly imperative events or states, not indicative events or states. Simply put, we don’t want emotional studies to reduce to epistemology.

We are emphasizing the causative rather than truth-function aspects of the emotions (and stressing the biology) to counterbalance those philosophers and psychologists who tend to see emotions primarily as a subset of cognitive judgement. We are trying to load as much of the appraisal work of emotions to the body, rather than the linguistic representational system so common to conscious reflective human appraisal. That doesn’t mean we reject all forms of such conceptual appraisal; we just think the load-bearing structure for most emotional experience is the body. And even if we can assess the incorrectness of the emotion for the adaptive goals of your family or society--via metacognitive reflexive representation--we’re still not going to “fix” it or make it more “apt” unless we employ embodied forms of behavioral change (somatic habituations).

[page 40]

Flanagan’s more nuanced understanding or concern about “correctness” (norming norms via social and/or virtue assessment) is not in the same tension with our project as the analytic search for indicative truth functions. In fact, our later chapters make it clear that virtue ethics is precisely the sort of approach that can and should correct unhelpful social and personal emotions.


We thank Seok for his thoughtful analysis which devolves into three major issues: (1) the cognitive ontology of emotions, (2) the developmental story of shame and guilt, (3) how to characterize the relation between layers of the mind. Seok writes, “Even though Asma and Gabriel are wrong in assuming the unity of emotion at the fundamental level of affect, they can still argue for the pervasiveness of diversely manifested affect in the human mind, society and civilization.” Relying on work by developmental psychologist Jean Decety, he argues that empathy is better understood as three different types namely, cognitive, affective sharing, and prosocial motivation. He writes, “they serve different functions, utilize different processes or mechanisms and are affected differently by brain damage.”

An intriguing question this brings up is whether or not emotions are natural kinds. We understand natural kinds to be biological entities that have a reliable causal story, or unified histories. Griffiths (2002) argues that emotions are not natural kinds, whereas Khalidi (2013) provides a helpful interpretation of how to conceive of natural kinds in psychology through tracing their causal histories. The controversy surrounding basic emotions and the ascent of the affective sciences seems pertinent here (Leys 2017).

While we do not think these arguments are definitive, we suggest the beginning of an answer is to be found in the phylogenetic causal histories of homologous neural systems and the similarity of sentient/motivational functions of basic affective systems. Panksepp’s description of the seven basic emotional systems in our opinion offers a descriptive, causal story of the history of units of neural function which we would like to call a kind. The reason is that the pertinent emotional states have hallmark neurochemical signatures and serve functions that have reliable, replicable causes. While the given emotional system undergoes a significant set of transformations across ontogenetic development, the neurochemical system and the broad function are maintained.

We think that what unites an emotion across a range of expressions is precisely the sentient experience of its motivational function in the service of allostasis. While emotional contagion and cognitive empathy use different mechanisms, like perceptual imitation in the former and hand-wringing cogitation in the latter, they serve related functions. Empathetic processes connect us to each other’s feeling states, this can even occur across species. We are convinced by De Waal’s Russian doll model of empathy and Panksepp’s affective neuroscience that there is a set of mechanisms related by function and experiential motivational state that can be meaningfully dubbed as types of empathy. The reason these expressions differ is partly due to the interactive relation between primary drives, secondary learned associations and tertiary linguistically and cognitively structured emotions. We are intrigued by Seok’s suggestion that empathy may have gone through two stages of co-option.

This brings us to Seok’s other major critique, which his discussion of shame and guilt may be considered an illustration, the relation between levels. He writes, “A distinctive stage (i.e., a layer) in an evolutionary process does not have to be continuous with and transformative of its previous stage(s) as Asma and Gabriel believe. It

[page 41]

can be a stage of a discrete and additive process of co-option.” He refers to co-opting as an additive process rather than a decoupling. He continues, “Co-option is a conservative tinkering process where new functions are added to the existing functions, yet the existing functions are not completely replaced by the new ones.”

In a footnote, Seok suggests we rely on taxic rather than serial homology. In fact we think both types of homology occur in nature and rule neither out. In general, we have sympathies with the co-option view, but will take this opportunity to clarify our use of a dialectical model for the relation between affect at different levels of the mindbrain.

A co-option model in contrast to a dialectical model is helpful, as we describe in our book through a portrayal of the brain as an evolutionary kluge (Marcus 2008), but it fails to capture how the separate evolution, or discrete level of a tertiary type of shame for example, has a profound effect on the original type derived from primary drives. A dialectical model allows that separate trajectories have holistic effects on the thesis and antithesis resulting in a synthesis. The organism is a unity, so the given co-option or dialectical developments must feedback on earlier elements of the emotion. This could be through associational processes or cultural scripts that condition the expression and use of the given emotion. We are using the term dialectical in its broadest form to try to invoke the multi-level causal picture of the brain. It may not be the most apt terminology, but it has been used in biology to signify a picture of complicated causality wherein biological expression is due to the synthesis of forces and countervailing forces (read Levins and Lewontin 1985). We do not deny additive processes, in fact the brain is indeed replete with redundant processes, but we are in earnest in our search for a language or metaphor that will offer clear characterization of a constant interactivity between evolutionary segments of the brain.

As a way of illustrating our approach, we provide here a dialectical story of shame. Shame originates as a protective emotion but it can become social anxiety, neurosis, or a normative check on social behavior within a hierarchy of filial piety. Not only that, shame can also fire ressentiment, as Nietzsche ([1895] 1995) argues in The Antichrist. These are related dialectically because the proximal cause is the sentient state. Whereas the ultimate cause is a cultural story of the social norms to which the proximal sentient state redounded, we suggest this interaction then feeds back into the developmental cultural learning experienced by the individual and subsequently his social group. Granted there are cases of the cultural version of shame taking on a discrete, unique expression in rare cases of cults or small irregular communities. But in these cases, the cultural form affect the individuals by the same means, i.e., as a proximal emotional motivation

At the primary level, shame, having a protective tenor may stem from some form of FEAR in its freezing behaviors which can be shaped by secondary level conditioning that links it to a set of causes, for example, I feel scared of shame when expressing personal views in front of parental authority figures, or ashamed/fearful about pubescent bodily indicators in social situations with my cohort. At the tertiary level, these conditioned behaviors can be formed or rationalized into intellectual systems like filial piety or humility relative to one’s position in the hierarchy.

Seok makes much of the distinction between continuity and transformation in psychological processes. What makes a psychological process a homolog? For us, it has to do with the biological nature of the behavior. But structure to function mapping is not simple. Are there neurochemical signatures? Are there bodily signals, or

[page 42]

specific perceptual inputs that lead to the behavior? How about the functions of the different behaviors? Two behaviors could have different neurochemical signatures but fulfill the same function, i.e., a many to one psycho-neural relation; “individuation according to structure and function do not always coincide, in that sense that the structure-to-function mapping can be many-to-many rather than one-to-one or even many-to-one (as in multiple realization)" (Khalidi 2017, 196). Furthermore, structure and function may actually be tracking different types of causal patterns or temporal and spatial trajectories; cognitive psychology leans on taxonomies on the basis of causal history whereas neuroscience is not suited to this type of individuation (Khalidi 2017, 196). We find these issues raised by Seok fascinating and will address them more extensively in future work.


We are pleased Cudd found our book The Emotional Mind to be stimulating and that there are many points of agreement between us. Cudd’s main critique is that our book “mischaracterizes the role that rationality and philosophical theories of rationality play in understanding and shaping our experience.” Specifically, she thinks we “confuse descriptive and normative uses of rationality in prudential, moral, and social theorizing.” In what follows, we will address Cudd’s main criticism, and also her more peripheral objections that “the book is insufficiently critical about the emotional origins of the theories it explores, leading to its uncritical acceptance of androcentric ideas and it all but ignoring the great and growing problems of group-based inequality.”

Cudd brings our main disagreement into focus by claiming that our discussion of emotionally based social norms is not sufficient to replace or avoid the important ethical work of rational assessment of said lived normative systems. Since Flanagan also has a similar objection, we are left to admit that we could have been and should have been clearer on this issue. We argue that a purely rational actor theory of human behavior (e.g., Homo economicus) does not sufficiently explain the proximate motivational states of agents, nor the irrational results of experiments like the ultimatum game; but our affective approach succeeds in precisely these lacunae. For example, we argue that emotions like revenge motivate human behavior, sometimes adaptively (albeit sub-optimally) and sometimes deleteriously. We may burn every bridge in our social support system and compromise our own survival, all for the sake of vengeance. The rational choice theory will not help us understand such common human behavior. To this point Cudd objects, “Perhaps not; but [rationality] will help us see why revenge is better resisted from either a prudential or a moral point of view.”

From this and other objections, both Cudd and Flanagan seem to think that our biological view is denying the normative project (of moral philosophy, etc.), but we don’t see that implication. It is possible that philosophers are especially sensitized to the traditional fact/value split and therefore see us as making some sort of naturalistic fallacy. We think our project is in part a work of descriptive realism that has captured real human value formation, motivation, and action. That is not in tension with the superordinate question of whether an emotional disposition should be indulged, repressed, or redirected in a given culture; that work can be done even better once the mind’s emotional nature is better understood (our project). Subsequent work will be done explicitly by moral philosophers (who will have the usual lack of impact on large scale populations), and implicitly by everyday umwelten lived-cultures (e.g., family, religion, politics, etc.).

[page 43]

This is why Cudd’s worry is not entirely relevant to our project. She says, “Without robust rationality and critical moral thinking (also evolutionary adaptations) about our emotion-based urges and biases, we cannot hope for a better future.” Our book is not really about making a better future, but of course we don’t want to make a worse future either. Generally, we leave the moral evaluation of bio-values for moral philosophers, religious leaders, politicians, and so on. So much work in the humanities has become advocacy and activism that it is becoming the dominant lens for evaluating all research. We cannot defend our work for its lack of activism, when we were not explicitly directed to such goals. Cudd says that our book ignores “the great and growing problems of group-based inequality.” Our response is to point out that our book also ignores the problems of gambling in baseball, the question of water on Venus, and the relative merits of George Harrison’s solo albums. All those issues, and of course group-based inequality, are worthy topics, but our book is not about them, has no pretense to explain them, and forwards nothing in those arenas. It seems disingenuous to critique a book on the grounds that it’s not about something else.

Cudd also suggests that our arguments in chapter eight seem inconsistent, to the extent that we criticize rationalist explanations of human behavior but then adopt adaptational explanation in our emotional causal strategy. Regarding the ultimatum game, Cudd correctly characterizes our rejection of rational choice theory because it fails to predict the frequently employed strategy of rejecting an offer that is low, whereas our emotional approach explains this easily. Cudd objects, however, that we then go on to say why the “emotional” response is “actually rational” (sic) by describing the adaptive value of pre-commitment displays of anger and other emotions. We are intrigued by this criticism but think it may be relying on a confusion about “rational” versus “adaptive.” We think systematic irrational behaviors cannot be understood as a rational calculation in the ultimatum game agent (as a conscious cost-benefit analysis), but of course we think the punishing strategy or refusal of money is rationally understandable to biologists as an adaptation in many kinds of social exchange scenarios. Our view not only captures the experiment results as accurately as a rationalist utilitarian theory that takes the long view of debt and payment (in an idealized rational reconstruction), but our theory also captures the felt sense and proximate motivation of the agents themselves.

Cudd misinterprets our view (and we take some blame for this) as denying any value to a rational choice description. Instead, we are arguing that an alternative explanation of dramatic social interaction--namely our affective affordance approach--is equally consistent with (or better than) rational behavioral predictions and also captures the psychology of the agents better than the usual approaches. Those usual approaches either project rational judgment into the actors themselves, or remain agnostic about proximate mechanisms. We feel justified going beyond agnosticism to connect motivated social exchanges to the specific feeling states articulated in our book (e.g., Panksepp’s seven affective systems).

Proximal and ultimate explanations of behavior are accurate if they pick out the real proximal and ultimate causal forces, otherwise they are intriguing redescriptions of phenomena. As a redescription of the motivations behind our behavior, the rational actor theory is tolerable since it helps us make certain kinds of predictions by focusing on the intentionality of goal directed agents, but rational agent theory is underdetermined by the facts of human cooperation. Human cooperation will also admit of a complex emotional/affordance theory, but with some advantages over the rational model—for example, apes make similar social contracts and alliances with conspecifics, but without conscious rational cost-benefit analysis. Our affective theory shows how cooperative

[page 44]

and occasional anti-social behaviors can be adaptive without enlisting powers of rational deliberation in the agents.

The “ultimatum game” was first developed in the 1980s and has become an often repeated and well-known case of irrational economics. In the ultimatum game two players are asked to divide $10. Player one proposes a specific division of the money, keeping amount $X and offering amount $Y to player two. Player two then accepts or rejects the offer. If player two rejects the offer made by player one, both players lose all of the $10. According to a purely rational agent theory, player two should always accept the offer from player one, because even a small payment is better than zero (the result of rejecting the offer). But player two frequently rejects a low offer, resulting in no money for both parties. Emotions explain this easily, where rationality (deliberative calculation based on utilitarian rules) does not. But anger or outrage in the face of some insultingly low offer is not just a maladaptive tantrum. Anger can be a precommitment mechanism that demands more fair distributions in the future. My anger is both a frustration of the endogenous SEEKING system (proximate cause), but also a signal/threat to conspecifics that I require a substantial share of resources (adaptive ultimate cause). Thinking about emotions like love, loyalty, and anger as commitment mechanisms makes sense of many human behaviors. It predicts many behaviors, like altruistic cooperation, indignant “bridge-burning,” and so on, better than rational actor theory.

Not only do emotional explanations solve anomalies of the rational agent model, but they also explain many so-called maximal utility behaviors as well as or better than purely cognitive deliberation. Consider a rudimentary cooperative action of our ancestors, like huddling behavior. We might make a rational decision that physical closeness and contiguity will keep a group of us warm through a cold night. But we might also feel, with our bodies directly, that huddling together increases warmth and so the attraction of our bodies is an affordance process, not a deliberation. Your body is an action-oriented-representation (AOR) for my body, and vice-versa. Our perception of each other comes loaded with imperative attraction in this case. Huddling together is driven by the homeostatic interests of our thermoregulation systems, and the way such regulation feels to us. Both competing proximate causal scenarios—conscious deliberation and affective affordance—are compatible with and facilitate an ultimate causal explanation (huddling is adaptive evolved behavior). But just as there is no conscious rational decision making in natural selection, so too there may be little or no conscious decision making in the individual reciprocal warmth of huddlers. Of course, as social organizations scale-up from huddled hunter-gatherers to tribes, chiefdoms, and even states, the ability to weigh costs and benefits in the virtual reality of a representational mind takes on greater influence. Our advanced cognitive abilities can process increasingly complex social data from our tribes, and this improves our cooperative capacities. Here, human reason takes over, and at this stage of human development we agree with Cudd’s assertion of its importance.

Another confusion emerges in Cudd’s pro-reason approach. And this same confusion occurs somewhat in Flanagan as well, so it’s best to clarify our view. Cudd reads our critique of Kantian and Utilitarian ethics as a dismissal of normative ethics generally, reading us more reductionistically than is warranted. Actually, while we admit to a primarily descriptive approach, we also throw our support behind the specific normative tradition of virtue ethics. We think that a virtue ethics frameworks (Aristotle and Confucius) and the sentimentalists (from Hume and Smith to Feminist Care Ethics) are much better at balancing the descriptive/normative tension that emerges from our emotions-first view of the mind. The objections to Kant and to Utilitarians are not objections

[page 45]

to ethics per se, but to a variety of overly rational idealism that predominate debates in modern ethical theory. Yes, Kant did consider the complexities of human anthropology, and even reminded us that, “Out of the crooked timber of humanity no straight thing was ever made” (Kant [1784] 1963, 6th Thesis). But, according to the Groundwork for a Metaphysic of Morals, his view of morality demands that a person act from the “good will” only, which is a sense of duty or obligation motivated by a principle of reason rather than any motivation polluted by a desire for a consequence or by an emotional feeling. Admittedly, Kant hints at a more nuanced understanding of the interaction between empirical factors and maxims in his late lectures in the Anthropology From a Pragmatic Point of View.

From our emotions-first view of the mind (wherein mind is always tethered to affective interests), Kantian disinterest looks like a pipe-dream. Our view is not that Kant and Mill, for example, are wrong because they are normative and we do not trust normative approaches; we think the virtue ethics of normative approaches are strong and compatible with our view of the mind. Rather, we think deontology and hedonic-calculus approaches are descriptively wrong about how the mind works, descriptively wrong about the motives of real agents (usually mischaracterized as egoistic selfishness), and prescriptively wrong about how we should live (radical nemocentric egalitarianism).

Yes, rationality has a role to play in human ethics and economics (as Cudd demands), and of course art and science and all the other human domains, but practical reason (phronesis) or skillful coping (read Dreyfus 2014) is the key, not idealized syllogistic, theoretical reason. We agree with Cudd that the biases discovered in recent behavioral economics should be incorporated into new economics such that science is improved, and we see ourselves doing the same by incorporating affect into the development and functioning of mind. Cudd thinks we are caught in some inconsistency because we both critique the hyper rationalism prevalent now and try to advance science and rational understanding of the mind. But Cudd’s critique assumes a univocal sense of “rational” and we recognize that there are modes and levels of rationality, in which practical reason (including thinking with images and judging through affordances) would be a much older kind of mind than theoretical reason (which requires the evolution of concepts, symbol systems, and language). Whereas the reason employed to evaluate a good scientific model is theoretical reason.

More importantly, we are trying to articulate a large range of emotional learning and communication that exists in social learning below the level of explicit rational cognitive teaching. We think there is an emotional niche that is broader and older (phylogenetically and ontogenetically) than rational teaching/learning. We argue that affect plays a significant role in cultural evolution and we want to see how far up we can scale those processes.

In order for social learning to be adaptive—contribute to survival and reproduction—the learning has to be discriminating (not completely random). It is better to learn from a good hunter, or tool maker, or musician or whatever, than simply copying one’s siblings, for example. Copying expert individuals is “success biased transmission” in cultural evolution circles, and certain non-rational patterns (for the agent) feed into that success (e.g., conformity bias, prestige bias, content bias, unbiased transmission or drift, and so on). Content bias for example refers to the way that affectively negative-coded information is remembered more and transmitted more than positive or neutral information. So affect is a crucial but understudied aspect of non-rational discriminations or judgements. In this case, affect is in the guise of a negative valence, but we point out that

[page 46]

positive affect is a crucial albeit understudied ingredient in prestige bias, or the attraction of charisma in habit formation and cultural transmission.

Folding these non-rational forms of affective cultural learning/transmission into the story of the human mind is not discounting reason altogether, but just reducing its importance a bit. Moreover, in pulling non-rational threads together into a fuller scientific account, we have not violated our own project and succumbed to reason after all. Rather we have given a fuller accounting of adaptive success: a biological logic that contains both rational and non-rational mechanisms of mind. In short, we claim modern ethical theory does not accord emotions a positive role in practical deliberations.

Additionally, Cudd levels a series of critiques that strike us as off base: implying, for example, that our work has unexamined values (like androcentrism). This style of criticism is difficult to counter since the number and tenor of values and centrisms that our book may be consistent with is almost innumerable. Cudd seems to imply that our book is consistent with a conservative viewpoint or male-centric viewpoint. We disagree, and hasten to point out that our book is very consistent with and can serve as a robust defense of liberal values and care-based feminist views of nature since our book explicitly argues for the reality of animal pain and sophisticated animal emotions.

Cudd implies that our description of the nuclear family is derived from a “1950s white middle-class perspective.” But this is incorrect. Our description, cited clearly in the text, is derived from some well-established social history and anthropological literature (e.g., Wiessner 1983; Johnson and Johnson 1975; Johnson and Earle 2000, etc.), most of which is based upon the study of non-Western indigenous cultures. Additionally, our discussions of nuclear family structure is informed by established research in social evolution, including: Chapais 2008, Palombit 1999, Fortunato and Archetti 2010, Gavrilets 2012, Marlowe and Berbesque 2009, and so on.

Cudd and Seok separately raise interesting and helpful critiques of our treatment of shame. Cudd for example suggests that our view of shame could be improved by a feminist approach like that of Sandra Bartky, who claims that, “Shame is a prevailing emotion for women that maintains male dominance. Shame in women is not about morality (as guilt is) but about social submission—keeping women in their place” (Cudd response). We thank Cudd for alerting us to her work. Bartky (and Cudd) may well be correct about shame as a tool of female oppression. Philosophers and psychologists disagree about the difference between shame and guilt. For example, June Price Tangney and Ronda L. Dearing, authors of Shame and Guilt (2002), suggest that shame is a more damaging emotion because it focuses on the failure of the person themself. Shame ascribes very negative emotional evaluation to the character or the “self” that is doing the offending action. When I feel shame, I feel that I am bad. Guilt, in their view, focuses not on the person so much as the sin or offending behavior. In their view, shame is older and more primitive, and it demonizes the person, whereas guilt is a feeling that judges the actor’s behavior, rather than the actor. The authors amass evidence from clinical and developmental psychology in order to conclude, “the pattern is pretty clear-cut: guilt is good; shame is bad” (Tangney and Dearing 2002, 136).

We suspect that Tangney and Dearing are correct in claiming that shame is older than guilt. A major reason for agreeing is that shame seems much more dominant in the ancient ethics of the Greeks and Romans, as well as

[page 47]

the Warring States period of Chinese ethics (e.g., Confucius). The way we would characterize this is that shame is public and requires the disapproval of other people, so it thrives in “face-cultures,” like Ancient Rome and contemporary China. The Judeo-Christian tradition, by contrast, stressed a deeply private sense of contrition between you and God directly. Even if no one knew of your sin and your penitent heart, God still knew. Guilt has no escape. Guilt is an inner state of pain, whether or not the community passes judgment. For this reason, some have argued, including Nietzsche and Freud, that guilt is a much more damaging emotion, as it tilts toward perseveration and neurosis very easily.

We can see our own veiled value system (Western modernity) better if we contrast it with the one that preceded Christianity. For the pagans, honor and pride were valued, but for the Christians it is meekness and humility; for the pagans it was public shame, for Christians, private guilt; for pagans there was a celebration of hierarchy, with superior and inferior people, but for Christians there is egalitarianism; and for pagans there was more emphasis on justice, while for Christians there is an emphasis on mercy (turning the other cheek).

We welcome Cudd’s critique that we should have acknowledged the feminist criticism of attachment theory. We stand duly chastened. Still, the majority of objections to attachment theory were directed at John Bowlby’s work, in particular for his equating attachment with “maternal instinct,” which was a notion criticized by some feminists. Since Bowlby’s and Ainsworth’s work in the 1970s, however, many aspects of attachment theory have been studied, corroborated, and developed beyond the “maternal instinct” model (read Hrdy 2009; Young and Wang 2004). Our work draws on this more recent work, including the neuroscience literature that ignores gender and sex as bonding, and focuses instead on the robust neuro-chemistry of early infant-caregiver interaction (i.e., the oxytocin and endogenous opioid increases for babies, mothers, fathers, and alloparents).


In response to Oh’s comments, we clarify that CARE/care is a relational ethical term that can be extended through rational agential processes. Indeed, our commitments within normative schemes are precisely extensions of CARE systems beyond immediate kin to non-kin and fictive kin, including domesticated animals, sports teams, alumni organizations, bands, hometown pride, etc. These are real ethical commitments, and they seem to rely on or at least correlate with hormones including endorphins, enkephalin, prolactin, vasopressin, and oxytocin. They cannot be reduced to these chemicals, which are necessary but not sufficient, but there is no reason our knowledge about these neurological elements should be left out of the story.

Oh rightly points out that the deep evolutionary roots for care are not exclusive to primates. We agree, and feel comfortable extending a care system provisionally to all mammals. Empirical work by Jaak Panksepp (1998), Frans De Waal (2018), and especially McGraw and Young (2010) reveal care behaviors, neurochemistry, and affective states across a wide range of mammals. Things get less clear when we move to other vertebrates and non-mammalian clades. For example, birds seem to display levels of care for offspring that border on mammalian dedication, yet they are phylogenetically descended in the reptile clade, not the synapsid (where every other mammal falls). We look forward to the empirical work that will clarify the neurochemistry substrates of avian care, and whether care systems are homologous or analogous from the evolutionary perspective.

[page 48]

How far can the human care system be extended? Oh asks this important question and suggests that it can become global. The commentator says, “I suggest a planetary expansion of family toward an interconnected eco-system for making a cosmopolitan harmony with the ‘eco-family’ via ‘eco-caring’ activity.” Suffice it to say that we are less sanguine about that level of expansion. From our point of view care and empathy may have idealistic goals, but they are fundamentally biological processes. The feeling of care is triggered by a perception or internal awareness and soon swells, flooding the brain and body with subjective feelings, behaviors, and oxytocin and endogenous opioids. Care is like sprint racing. It takes time, duration, energy, systemic warm-up and cool-down, practice and a strange mixture of pleasure and pain (attraction and repulsion). Like exercise, it is not the kind of thing one can do all the time or extend indefinitely. We will literally wear down the care system in short order, if we ramp-up the care system every time we see someone in need. The nightly news would render us exhausted and depleted. The limbic system is not designed to handle the kind of constant stimulation that cosmic care proponents expect of it.

If care is indeed a limited resource, then it cannot stretch indefinitely to cover the massive domain of strangers, nonhuman animals, and the planet. Of course, when we see the suffering of strangers in the street or on television, our heartstrings (innate social sentiments) vibrate naturally. We can have contagion-like feelings of sympathy when we see other beings suffering, and that is an adaptive response; but that is a long way from the kinds of active preferential devotions that we marshal for members of our respective families. Real family members donate organs to us, bring soup when we are sick, watch our kids in an emergency, open professional doors for us, rearrange their schedules and lives for us, protect us, and fight for us, and we return all this hard work. Our tribes of kith and kin are “affective communities,” and this unique emotional connection creates altruism, generosity, and selfless loyalty. There’s an upper limit to our tribal emotional expansion. For these reasons we are dubious about “eco-care” for an “eco-family.” Having said that, we do think that rationally-based effective altruism can and should reach out to protect the planet and our animal cousins. We are also intrigued by recent projects to mine cross-cultural traditions, blending affective roots and adaptive ideologies to meet the growing dangers of the anthropocene (read Ivakhiv 2018).

Another interesting issue raised by Oh is the question of animal culture. In our book we confine most of our discussions to primate cultures, but it remains unclear as to which other animal species have culture. This is an especially interesting question because we suggest that in addition to ecological niches and cognitive niches, evolutionary psychologists should consider the affective niches of animal groups. Culture does not only transmit tools and ideas, but also affectively-laden behavioral norms. We are open to expanding the recognition of animal cultures beyond primates, as long as we do so guided by the data. For example, excellent research showing the emotional state-changes of playing rats reveals clear patterns of stimulus-response, brain wave changes, preferential contexts, and behavioral tendencies (Burgdorf et al. 2020), but we should be wary of calling rat play “cultural” unless we also simultaneously observe transmission of behavioral conventions, like adjustable habits, teaching, and learning.

Turning now to the helpful comments about cross-cultural comparative work on Asian spiritual traditions. Oh argues that we are too fast and loose with the Asian religions discussed late in the book. In our defense, we are trying in those sections of the book to show how affective roots are present in common religious experiences. Obviously fear and anger can be seen in Abrahamic views of God as the angry father, or in Hindu representations of Kali or Shiva. But we are also curious in that final chapter about the emotional core of the

[page 49]

common religious experience of the “unconditioned” or the absolute. Yes, the Chinese notion of Tian, the Hindu notion of Atman/Brahman, and the Buddhist notion of sunyata (Nothing) are quite different, and we concede this gladly. However, they share a common conceptual, and we think also affective, notion of boundlessness: of that which is limitless, and expansive. At any rate, all these religious emotions are concessions to human finitude, and the overwhelming nature of something bigger or something not constrained inside a traditional ego. These are the aspects that led us to find commonality in these diverse traditions.

We are grateful to Oh for providing very helpful nuance and clarity on the Buddhist concept of sunyata, the Hindu Atman/Brahman, and the Chinese Tianming. These are indeed different in substantial ways, as the commentator points out, and our overemphasis on common “unconditioned” phenomenology may have distorted them. In particular we are intrigued to see the comparison of our anti-dualist use of “emotional mind” and the Chinese 心 (xin/shim) (heart-mind). Contrary to Slingerland (2018), who argues that xin is not a case of Chinese holism but yet another case of universal dualism, we agree with Oh’s reading of the heart-mind as a precursor to more recent embodied mind theories. This is outside our direct expertise but we watch on for further developments in the debate.

[page 50]


Burgdorf, J. S., S. M. Brudzynski, J. R. Moskal. 2020. "Using Rat Ultrasonic Vocalization to Study the Neurobiology of Emotion: From Basic Science to the Development of Novel Therapeutics for Affective Disorders." Current Opinion in Neurobiology 60: 192-200.

Chapais, Bernard. 2008. Primeval Kinship: How Pair-Bonding Gave Birth to Human Society. Cambridge, MA: Harvard University Press.

De Waal, Frans. 2018. Mama’s Last Hug. New York, NY: W.W. Norton and Company.

Dreyfus, H. L. 2014. Skillful Coping: Essays on the Phenomenology of Everyday Perception and Action. Edited by M. A. Wrathall. New York, NY: Oxford University Press.

Fortunato, Laura, and Marco Archetti. 2010. “Evolution of Monogamous Marriage by Maximization of Inclusive Fitness.” Journal of Evolutionary Biology 23, no. 1: 149–156.

Gavrilets, Sergey. 2012. “Human Origins and the Transition from Promiscuity to Pair-Bonding.” Proceeding of the National Academy of Sciences 109, no. 25: 9923–9928.

Griffiths, Paul E. 2004. “Is Emotion a Natural Kind?” In Thinking About Feeling: Contemporary Philosophers on Emotions, edited by Robert C. Solomon, 233–249. New York, NY: Oxford University Press.

Hrdy, Sara. 2009. Mothers and Others: The Evolutionary Origins of Mutual Understanding. Cambridge, MA: Harvard University Press.

Ivakhiv, A. 2018. Shadowing the Anthropocene: Eco-Realism for Turbulent Times. Goleta, CA: Punctum Books.

Johnson, Orna R., and Allen Johnson. 1975. “Male/Female Relations and the Organization of Work in a Machiguenga Community.” American Ethnologist 2, no. 4: 634–648.

Johnson, A. W., and T. Earle. 2000. The Evolution of Human Societies: From Foraging Group to Agrarian State. 2nd ed. Stanford, CA: Stanford University Press.

Kant, I. [1784] 1963. “Idea for a Universal History from a Cosmopolitan Point of View.” In On History. Translated by Lewis White Beck. Indianapolis, IN: The Bobbs-Merrill Company.

Khalidi, Muhammad Ali. 2013. Natural Categories and Human Kinds: Classification in the Natural and Social Sciences. Cambridge, UK: Cambridge University Press.

[page 52]

______. 2017. “Crosscutting Psycho-Neural Taxonomies: The Case of Episodic Memory.” Philosophical Explorations 20, no. 2: 191-208.

Levins, Richard, and Richard C. Lewontin. 1985. The Dialectical Biologist. Cambridge, MA: Harvard University Press.

Leys, R. 2017. The Ascent of Affect. Chicago, IL: University of Chicago Press.

Nietzsche, F. (1895) 2000. The Antichrist. Translated by Anthony Mario Ludovici. Buffalo, NY: Prometheus Books.

Marcus, Gary. 2008. Kluge: The Haphazard Construction of the Human Mind. Boston, MA: Houghton Mifflin Co.

Marlowe, Frank W., and Julia C. Berbesque. 2009. “Tubers as Fallback Foods and Their Impact on Hadza Hunter-Gatherers.” American Journal of Physical Anthropology 140, no. 4: 751–758.

McGraw, L. A., and L. J. Young. 2010. "The Prairie Vole: An Emerging Model Organism for Understanding the Social Brain." Trends in Neurosciences 33, no. 2: 103–109.

Palombit, R. A. 1999. “Infanticide and the Evolution of Pair Bonds in Nonhuman Primates.” Evolutionary Anthropology 7, no. 4: 117-129.<117::AID-EVAN2>3.0.CO;2-O.

Panksepp, J. 1998. Affective Neuroscience: The Foundations of Human and Animal Emotions. New York, NY: Oxford University Press.

Slingerland, E. 2018. Mind and Body in Early China: Beyond Orientalism and the Myth of Holism. New York, NY: Oxford University Press.

Tangney, J. P., and R. L. Dearing. 2002. Shame and Guilt. New York, NY: The Guilford Press.

Wiessner, P. 1983. “Style and Social Information in Kalahari San Projectile Points.” American Antiquity 48: 253–276.

Young, L. J. and Z. Wang. 2004. “The Neurobiology of Pair Bonding.” Nature Neuroscience 7, no. 10: 1048–1054.

Stephen T. Asma and Rami Gabriel © 2021.

Author emails: and