Introduction: Humanism and Its Critics
The language of dehumanization now seems nearly ubiquitous, a dialect of our moral vernacular that is increasingly salient in both popular and academic discourse about interpersonal violence. We find such language, for example, in the testimony of survivors of enslavement, torture, rape, and genocide, who allege that they were treated or regarded as objects, animals, monsters, or the like. It also appears in the accounts that some perpetrators of inhumanity give of their deeds, and in political propaganda that casts some despised or feared group as beyond the pale of humanity.
Talk of victims of inhumanity being regarded as less than human is sometimes supposed to be purely rhetorical. But more often it hints at a form of explanation of why people are motivated to commit inhumanity, particularly when such claims about dehumanization are made in philosophy, feminist theory, social psychology, and genocide studies. The basic idea is that we treat others inhumanely when we fail to see them as human beings, and thus that our treatment of them will tend to be more humane when we (fully) see their humanity. Call this view humanism.
Recently, there has been a critical backlash against humanism and the dehumanization model specifically. Critics of humanism object that the dehumanization model does not fit ‘man's inhumanity to man’ in that people who treat others inhumanely often know and even care that their victims are human. For, it is argued, perpetrators of inhumanity regularly act on moral emotions or for the sadistic pleasure of abusing another human being. Thus, it seems false that torturers, mass rapists, and genocidaires literally see or treat their victims as nonhuman. Critics further point out that recognizing someone's humanity is insufficient for explaining why we treat her humanely because that recognition standardly elicits a range of hostile attitudes that may easily move us to commit inhumanity against her. In light of these objections, humanism may seem psychologically unrealistic, and sympathy for the view may betray a kind of liberal naïveté about inhumanity and an unfounded optimism about humanity—as if our all-too-human inhumanity could be chalked up merely to ignorance of empirical fact or category mistake.
I articulate and evaluate the case against humanism in this essay, with particular attention to its expression in the work of Kate Manne (Reference Manne2016; Reference Manne2018: 133–78; Reference Manne2019; Reference Manne2020: esp. 25–27). Manne's critique merits such close attention because it continues to be highly influential, within and outside philosophy, and because it has generally not been subjected to a depth of scrutiny commensurate with its outsized influence (see, however, Ng Reference Ng2021). I argue that the two most commonly voiced objections to humanist views (and their relatives) are unpersuasive: not only does the evidence marshaled against these views fail to disprove them, it could threaten them only if some questionable assumptions were granted. While my discussion constitutes a partial defense of humanism, it is primarily intended as a plea for greater methodological mindfulness. By providing necessary conceptual ground clearing and routing common lines of attack, I hope to determine what it would take for a humanist project to succeed, thereby paving the way for a full defense of humanism that fulfills its explanatory ambitions.
1. Humanism: What is the Target?
As Manne (Reference Manne2018: 143–46) uses the term, ‘humanism’ refers to a cluster of moral-psychological claims about what explains why people are motivated to treat others humanely or inhumanely. Her characterization of humanism is complex, but two theses constitute the heart of the view. I will call this pair of claims the recognition thesis and the dehumanization thesis, respectively, thereby eschewing Manne's own—somewhat opaque—labels for them (i.e., ‘the moral psychological claim’ and the ‘quasi-contrapositive moral psychological claim’, respectively).
Here, then, is a statement of the core of humanism, in Manne's sense:
humanism =def For every pair of human beings, A and B,
Recognition Thesis: if A sees B as a (fellow) human being, then (ceteris paribus) A is robustly disposed to treat B humanely in virtue of A's recognition of B's humanity; and
Dehumanization Thesis: if A is robustly disposed not to treat B humanely, then (ceteris paribus) A fails to see B as a (fellow) human being and is so disposed in virtue of that failure.
Let us begin by clarifying how, exactly, Manne conceives of the main target of her critique.
First, for the humanist, the concept of a human being is not a purely biological notion that refers to its user's conspecifics—members of the species Homo sapiens. Rather, the concept refers to fellow persons, where a person is characterized by the possession of a range of psychological capacities, including capacities for rationality, autonomy, valuation, emotion, and so on. To see someone as a human being, then, is to attribute some such set of capacities to him; to fail to see him as a human being is to fail to attribute such capacities to him (Manne Reference Manne2018: 141–42). I use the term ‘interpersonal recognition’ to refer to the state of seeing someone as a human being.
Second, while Manne provides no formal definition of what it is for treatment to be humane or inhumane, we can conjecture that treatment of a person is humane if and only if it is motivated by the kind of basic moral consideration appropriate to our dealings with human beings. Likewise, it seems that treatment of a person is inhumane, in her sense, if and only if it amounts either to behaving toward him with extreme hostility, even cruelty, or to treating him with extreme indifference, where these exceed the bounds of ordinary moral consideration.
Third, per Manne (Reference Manne2018: 144), the humanist holds that seeing someone as human strongly disposes the recognizer to treat the recognized party humanely, by activating a motivating state or mechanism such as ‘empathy, sympathy, compassion, or fellow feeling’ toward the latter. Failing to see another as a human being, on the other hand, explains our tendency to treat him inhumanely in virtue of dampening our sympathetic/empathetic capacities vis-à-vis him. Manne argues, however, that the recognition thesis and the dehumanization thesis are both false.
Importantly, Manne claims that humanism, in her sense, is common in philosophy and beyond, and that it can be found in the work of Rae Langton (Reference Langton2009a, Reference Langton2009b), David Livingstone Smith (Reference Smith2011, Reference Smith2016), Nomy Arpaly (Reference Arpaly2003: 75–78), Cora Diamond (Reference Diamond1978), and Raimond Gaita (Reference Gaita2002), among others. But the heterogeneity of this list of theorists—comprising a Kantian, a Humean, two Wittgensteinians, and a pragmatist—invites us to wonder whether their views all fit Manne's official definition of humanism. And a closer look reveals the label to be an awkward fit.
For one, Smith (Reference Smith2021: 231) does not accept the recognition thesis and, thus, does not count as a humanist, in Manne's sense, nor does he construe dehumanization as a failure to ascribe characteristically human psychological capacities. Neither Diamond nor Gaita holds that recognition of someone's humanity strongly disposes the recognizer to treat that person humanely, nor, as far as I am aware, does anyone on Manne's list believe that all or most inhumanity is explained by a failure to recognize the humanity of the victims targeted. If anything, theorists who ascribe explanatory significance to dehumanizing attitudes are apt to accept nearly the converse of the dehumanization thesis instead: that if A fails to see B as a human being, then (ceteris paribus) A is disposed not to treat B humanely. So, we may worry that Manne's critique is directed at a straw man.
Although this interpretive worry has some force, it would be unwise to dismiss Manne's critique out of hand. For despite these infelicities, fully vindicating a view in the general neighborhood of humanism proves to be no easy task. Moreover, in taking the measure of Manne's case against humanism, we stand to learn much about the assumptions that typically frame the debate between humanists (or proponents of similar views) and their critics. Finally, as we will soon see, her critique presents at least a prima facie challenge even to views that do not perfectly match her stated definition of humanism. In particular, her arguments seem to cast doubt on the idea that dehumanization—in the sense of seeing people as other than human—helps explain why people commit inhumanity. In any case, for ease of exposition, I will continue to use the term ‘humanism’ to denote the target of Manne's critique while acknowledging that some of her interlocutors explicitly disavow the label and, indeed, even reject one of the defining theses of the view. Let me now turn to Manne's first objection to humanism.
2. Taking a Dehumanizing—yet Humanizing—View of a Person
The first objection starts with the idea that perpetrators of inhumanity are, in fact, often aware that their victims are human beings. For these perpetrators often seem to be motivated by the sorts of attitudes that it only makes sense to hold toward human beings, including what P. F. Strawson (Reference Strawson2008: 6–7) called the ‘reactive attitudes’ (e.g., resentment and indignation). Holding these attitudes toward others involves attributing characteristically human mental states and capacities to them. Because these perpetrators see their victims as human, they cannot or do not see them as other or less than human. Both the dehumanization thesis and the recognition thesis are therefore false.
Call this the humanization objection. In a forceful statement of it, Manne denies that those who commit atrocities against others generally act from an unawareness of their victims’ humanity. Rather, inhumanity often has a decidedly moralized, indeed interpersonal, dimension that is obscured by taking inhumane agents to represent their targets as nonhuman.
The centerpiece of Manne's argument is an analysis of the motives of Elliot Rodger, who committed the Isla Vista killings in 2014 after vowing to kill the ‘hot blonde sluts’ of the Alpha Phi sorority at the University of California, Santa Barbara. Before the attack, Rodger explained that he was acting so as to avenge himself against the sort of women who, he claimed, had caused him suffering throughout his life by refusing him sex and love. Manne writes that Rodger's resentment appeared to ‘hinge on the women's presumed humanity’, in that he ‘did not deny women's power, independence, or the reality of their minds. Rather, he hated and sought to punish them for evincing these capacities in ways that frustrated him, given his sense of entitlement to their benefit’ (2018: 150, emphasis in the original). In other words, inhumane agents often do not hold dehumanizing attitudes toward their victims: attitudes that involve failures to see them as human beings; hence, dehumanization, in my narrow sense, simply does not exist.
The humanization objection has been further developed in social psychology, among other fields. Paul Bloom (Reference Bloom2017; cf. Bloom Reference Bloom2016: esp. 201–7), for instance, extends Manne's analysis by suggesting that views such as humanism tend to distort the motives of those who sadistically mistreat others; the thrill of treating some people as vermin, for example, ‘lies precisely in the recognition that they are not’—that they are, in fact, human. Harriet Over (Reference Over2021: 6) expands this attack on humanism further still, pointing out that allegedly dehumanized people are ‘often described in ways that only apply to human beings’ (e.g., as criminals or usurpers)—a practice that assumes that the speaker is aware of the difference between (human) outgroup members and nonhuman animals or objects. Furthermore, she claims, while victims of inhumanity are often thought to lack certain mental properties, they are often ascribed characteristically human yet antisocial properties such as cunning, greed, and spite—a point that finds some experimental support in a recent study by Enock and colleagues (Reference Enock, Flavel, Tipper and Over2021). Finally, and along similar lines, Rai, Valdesolo, and Graham (Reference Rai, Valdesolo and Graham2017) conducted a series of five experiments that suggested that the dehumanization of another person facilitates instrumental violence toward him but not morally motivated violence; the latter, they argue, requires seeing him as a human being. (For a similar point in the context of objectification, see Mikkola Reference Mikkola2016: 157; Reference Mikkola and Kronfeldner2021: 331–38).
Critics of humanism might pursue an additional strategy in tandem: to insist that dehumanizing attitudes would be motivationally otiose, playing no significant role in the motivation of inhumane behavior. This anti-humanist strategy appears to gain plausibility once we highlight the virtues of alternatives to humanism, such as Manne's (Reference Manne2018: 150–68) own socially situated view of inhumanity. On Manne's view, people treat others inhumanely when and because we hold negative socially situated stances toward them—seeing them, for example, as enemies, rivals, or thugs. Stances such as these have their home in hierarchical relations of domination among human beings regarded as such. Similarly, humane behavior is explained by the agent's holding positive socially situated stances toward others—example, coming to see them as friends, fellow citizens, and the like. Whether negative or positive, these are humanizing attitudes toward others: attitudes that presuppose that their bearer is aware of their objects’ humanity. Thus, it is unnecessary, and less plausible, to posit dehumanizing attitudes to explain inhumane behavior.
I take the force of the humanization objection to be greatly overstated. Before rebutting it, however, I want to draw out and examine the assumptions that I believe lend the objection some of its superficial credibility. Then, I will canvass some salient evidence for the existence of dehumanizing attitudes on the part of would-be perpetrators of inhumanity.
2.1. Mutual Exclusion?
On the strongest version of the humanization objection, people who see others as human beings cannot also see them as other than human (and vice versa), either as a conceptual or as a psychological matter. Yet the objection relies on some controversial assumptions—for example:
Mutual Exclusion–Seeing-as: If A sees B as human, then A does not see B as nonhuman.
This thesis is wildly implausible, however, as it conflicts with platitudes about the logical grammar of seeing-as claims. For one, seeing x as F does not normally preclude seeing x simultaneously as G, not even when F-ness and G-ness cannot belong to the same thing at the same time. Parents can see their fully grown offspring both as children and as adults, for instance—even though it is (I take it) impossible for a person to be a child and an adult all at once. True, for some F and G, not all x can be seen as both F and G—think of Joseph Jastrow's duck-rabbit. But we need some further reason to think that people cannot be seen as human and nonhuman.
Mutual Exclusion–Seeing-as may seem plausible given a particular view of the link between seeing-as and belief. If A sees B as nonhuman, it might be thought, A does not believe that B is human or believes that B is nonhuman. And perpetrators of inhumanity do typically believe that their victims are human. If these claims are true, Mutual Exclusion–Seeing-as seems to be on solid ground. But the load-bearing assumption does not hold up to scrutiny, either. A surgeon operating on her patient may see him in that setting as a mere collection of organs and tissue, rather than as a minded individual, all the while retaining her belief that the patient is human.
Critics of humanism bent on pressing a strong version of the humanization objection might rely not on Mutual Exclusion–Seeing-as but, instead, on the following, related claim:
Mutual Exclusion–Belief: If A believes that B is human, then A does not believe that B is nonhuman.
As Smith (Reference Smith2020: 145–49; 2021: 233–34) points out, this thesis is credible only if it is impossible for people to hold contradictory beliefs simultaneously, like the belief that someone is human and the belief that she is not. But it is a familiar fact of life that people sometimes hold conflicting beliefs. Consider a chain-smoker who recognizes that his habit will be fatal but retains the conviction—felt strongly in optimistic moods—that he will be one of the lucky ones to survive. There is no psychological impossibility here. As a general matter, then, the schema instantiated by Mutual Exclusion–Belief appears to be false. Once again, the burden of proof is on the critic of humanism to show why it would be otherwise with the belief that someone is human and the belief that she is other than human.
Finally, we can imagine a variant of Mutual Exclusion–Belief with apparent credibility:
Mutual Exclusion–Belief*: If A ascribes any characteristically human psychological capacities or states to B, then A does not fail to see B as human.
This claim implies, by contraposition, that if A fails to see B as human, then A does not ascribe any characteristically human psychological capacities or states to B. Less abstractly, taking a dehumanizing attitude toward someone consists in a complete failure to attribute human psychological capacities or states to her. So, just how plausible is Mutual Exclusion–Belief*?
Well, if the original Mutual Exclusion–Belief is false, it is hard to see why the modified version would fare much better. If it is possible for a parent to believe that his fully grown offspring is an adult while simultaneously believing, implicitly, that she is a child, then there appear to be no grounds for denying the possibility of him believing that she has psychological capacities or states characteristic of adults as well as those that are characteristic of children. Beyond that, though, Mutual Exclusion–Belief* is undergirded by two further controversial assumptions, each of which would be denied by some of the targets of Manne's critique.
First, there is the claim that dehumanization is always all-or-nothing rather than potentially partial (cf. Vaes, Paladino, and Haslam Reference Vaes, Paladino and Haslam2021; Kronfeldner Reference Kronfeldner2018: 26; and Reference Kronfeldner and Kronfeldner2021: 7; for criticism of such views, see Smith Reference Smith2021: 152–53). Second, there is the claim that dehumanization consists of denying characteristically human psychological capacities (cf. Smith [Reference Smith2021: 153–57], who holds that seeing someone as human consists of thinking of her as belonging to one's own natural kind; Phillips [Reference Phillips2015: 8–10], for whom taking a person to be human is a noncognitive assertion of equal social/political status; and Kronfeldner [Reference Kronfeldner2018], who argues that ‘human’ has two senses, one that picks out a biological group [Homo sapiens] and another that picks out a social group [bearers of moral standing]). In any case, my point is that if we are willing to accept this second assumption, the first seems highly dubious. If dehumanization is a matter of denying that someone is human in psychological terms, then it would seem odd to insist that the denial cannot be piecemeal—the effacement of certain specifically human psychological capacities or states, but not others, from the dehumanizer's view.
In any case, going forward, I assume that failing to recognize a person's characteristically human psychological capacities or states constitutes at least one way of dehumanizing him. I do so for dialectical purposes: because Manne conceives of dehumanization in these terms, as do some scholars of dehumanization, it would be important for the debate if it turned out that some agents of inhumanity did dehumanize their (would-be) victims, in this sense. I also make a similar assumption—mutatis mutandis—about seeing someone as a human being: that at least one way of so regarding him is seeing him as human in his psychological capacities or states. Even granting these two assumptions, however, the mutual exclusion theses considered above are implausible.
There is no logical or conceptual obstacle, then, to seeing a person as a human being and as other than human, and this combination of mental states seems psychologically possible. Of course, this point by itself does not show that dehumanization exists. For even if perpetrators of inhumanity can see their victims as nonhuman, it may still be the case that they never or rarely do so (anymore), particularly when motivated by Strawsonian reactive attitudes—such as resentment or indignation—or, indeed, by a bare desire to make a person suffer for its own sake. Thus, we might accept a weaker version of the humanization objection: that even if it is in principle possible for perpetrators of inhumanity to see their victims as other than human, they simply do not. To defuse this version of the objection, we still need to show that some inhumane agents hold dehumanizing attitudes toward their victims, seeing them (also) as other than human.
2.2. Dehumanizing Attitudes: Some Evidence
To begin to evaluate the weaker (hence, more plausible) version of the humanization objection, let us revisit the example of Elliot Rodger. Remember that Manne (Reference Manne2018: 150) holds that Rodger fully saw women as human beings with characteristically human capacities, albeit (probably) underneath a ‘more or less thin veneer of false consciousness’; he had no dehumanizing attitudes toward them, she insists. Is she correct? I believe not, or not obviously. Manne's interpretation conflicts with Rodger's own stated views about women, as detailed (at sometimes stupefying length) in his memoir, in which he habitually speaks of women in terms that are explicitly, even animalizingly autonomy-denying. Here is a particularly telling and representative passage:
Women are flawed creatures, and my mistreatment at their hands had made me realize this sad truth. There is something very twisted and wrong with the way their brains are wired. They think like beasts, and in truth, they are beasts. Women are incapable of having morals or thinking rationally. They are completely controlled by their depraved emotions and vile sexual impulses. (Rodger Reference Rodger2014: 136; see also 117)
The most straightforward interpretation of this passage is that it expresses Rodger's view of women as animals lacking autonomy and rationality—two hallmarks of the human. There is a defeasible presumption in favor of this reading, and it is not undermined by the fact that Rodger saw women as human beings toward whom certain reactive attitudes were, in principle, apt. For in light of the arguments just advanced, seeing a person as human and seeing her as other than human (in this case, as an animal) are not mutually exclusive; thus, Rodger may have just seen women under both guises simultaneously. In other words, without the backing of one of the mutual exclusion theses, Manne's interpretation seems one-sided, and, indeed, unsupported.
Manne's discussion contains two responses to counterexamples such as these. First, dehumanizing language is, she sometimes claims, standardly used to ‘intimidate, insult, demean, [and] belittle’ rather than to make factual assertions about its target's psychological capacities or metaphysical status; that is, such talk does not constitute solid evidence that the speaker harbors dehumanizing attitudes (Manne Reference Manne2018: 163–64; cf. Smith Reference Smith2021: 235). In reply to this objection, however, the humanist can point to clear-cut cases in which dehumanizing language does, in fact, function to assert such factual claims. And such cases are not hard to come by.
As Johannes Steizinger points out (Reference Steizinger2018: 147–48), some claims made by Adolf Hitler seem to be unambiguous denials that Jewish people are human, not just empty metaphors or insults. In Mein Kampf, Hitler (Reference Hitler and Murphy1939: 233–35) theorizes that what distinguishes human beings from animals is that we alone have ‘the idealistic spirit’—‘the willingness of the individual to make sacrifices for the community and his fellow-men’—and that Jewish people lack this property, for in them ‘the readiness for sacrifice does not extend beyond the simple instinct of individual preservation’. (Indeed, as Thomas Brudholm and Johannes Lang [Reference Brudholm, Lang and Kronfeldner2021: 347] emphasize, such a dehumanizing stance is even compatible with moralized hatred toward the target group.) We find comparable rhetoric in the testimony of other genocidaires. For instance, as Hatzfeld (Reference Hatzfeld2005: 47) reports, some Hutu soldiers who took part in the Rwandan genocide characterized Tutsis similarly: ‘[w]e no longer saw a human being when we turned up a Tutsi in the swamps’, one of the genocidaires confessed, ‘I mean a person like us, sharing similar thoughts and feelings’; ‘[w]e no longer considered the Tutsis as humans or even as creatures of God’, claimed another.
Furthermore, claims that express dehumanizing attitudes were historically made about the supposedly truncated psychological capacities of women and enslaved people. For example, Black slaves in the antebellum United States were regularly seen by their White enslavers as lacking in specifically human capacities and as more akin, psychologically, to nonhuman animals: they were regarded as bereft of human-level intelligence, autonomy, and self-control. (For elaboration, see, e.g., Elkins [Reference Elkins1959: 82] and the references in Machery [Reference Machery and Kronfeldner2021: 152–53].) And women have traditionally been regarded as so radically deficient in their intellectual and volitional capacities as to be incapable of the kind of self-government required for participating in political life—a view whose most infamous expression, perhaps, appears in Aristotle's (Reference Aristotle and Reeve1998: 23) Politics but that we can also detect in the rhetoric of contemporary misogynists like Rodger.
Second, Manne (Reference Manne2018: 162) sometimes responds to such counterexamples by insisting that people who appear to hold dehumanizing attitudes are in the grip of wishful thinking instead, in that they are indulging a fantasy about others’ minds rather than misrecognizing them. Her point is that perpetrators of inhumanity want to believe that their victims are other than human—that, say, they have simpler or perhaps alien kinds of minds—but that they do not actually see their victims in this way; deep down, the perpetrators do fully recognize their humanity.
Manne's suggestion strikes me as unduly optimistic, particularly as applied to historically pervasive attitudes toward enslaved people and women. Beyond that, however, it misses the fact that the motives operative in wishful thinking sometimes obscure our perception of the minds of others by leading us to attend less consistently or less astutely to their minds. Again, the fact that inhumane agents recognize their victims’ human mind, on some level or to some degree, does not rule out the possibility that they also misrecognize it. They may even be motivated to blot out their sense of their victims’ mind in response to their recognition of its humanness.
My counterexamples cast doubt on Manne's socially situated view, too. A great deal of that view's plausibility rests on the idea that the socially situated stances that motivate inhumane treatment never involve seeing the would-be victims as other than human—an idea that has been shown to be false. A White person in the antebellum American South seeing a Black person as a slave qualifies as a socially situated stance if anything does, embedded as it is in relations of interpersonal domination. Yet, enslaved people were seen as endowed with only a truncated capacity for experience, thought, and volition compared to White people. Much the same could be said, mutatis mutandis, about seeing someone as a woman, in some historical contexts.
Thus, there is some initial support for the idea that people do sometimes hold dehumanizing attitudes toward others—attitudes that constitute a failure to regard them as (fully) human. But this historical evidence is also bolstered by research on dehumanization in social psychology, where the topic has received considerably more attention than in philosophy (Manne's discussion does not engage with this work).
For instance, Jacques-Philippe Leyens (Reference Leyens, Paladino, Rodriguez-Torres, Vaes, Demoulin, Rodriguez-Perez and Gaunt2000, Reference Leyens, Cortes, Demoulin, Dovidio, Fiske, Gaunt, Paladino, Rodriguez-Perez, Rodriguez-Torres and Vaes2003, Reference Leyens, Demoulin, Vaes, Gaunt and Paladino2007) and colleagues have found evidence in a series of studies that people are less disposed to attribute what they call ‘secondary emotions’ to outgroup members than to ingroup members, where these are more sophisticated emotions (e.g., nostalgia, admiration, fulfillment) that belong to human beings but arguably not to animals. Thus, people seem disposed to subject outgroups to what these authors call ‘infrahumanization’: taking others to be less human than one's own group. In a similar vein, Nick Haslam (Reference Haslam2006) has discovered evidence of two distinct forms of dehumanization: what he calls ‘animalistic dehumanization’ (denying a person such uniquely human qualities as civility, rationality, and moral sensibility) and what he calls ‘mechanistic dehumanization’ (denying a person such characteristically human qualities as agency, personal depth, and emotional responsiveness). (For a critique, see Enock et al. [Reference Enock, Flavel, Tipper and Over2021], for instance.) And in an especially disturbing raft of studies, Nour Kteily and colleagues (Reference Kteily, Bruneau, Waytz and Cotterill2015) have found that—inter alia—Americans commonly hold explicitly dehumanizing attitudes toward some ethnic or religious groups (e.g., Arabs and Muslims), seeing them as endowed with psychological abilities that make them more akin to nonhuman animals than to modern human beings (see Kteily et al. Reference Kteily, Bruneau, Waytz and Cotterill2015 and Kteily and Bruneau Reference Kteily and Bruneau2017). Lasana Harris and Susan Fiske (Reference Harris and Fiske2006) have even found that in neurologically typical observers, members of certain outgroups (e.g., homeless people, drug addicts) fail to activate the medial prefrontal cortex, which underpins our capacity for social cognition. Data such as these are not uncontroversial, to be sure. Still, the evidence considered so far suggests that some people are apt not to ascribe characteristically human mental states, capacities, and personality traits to members of certain groups, thereby failing to see them as human (to that extent), and that this occurs through increasingly well-studied psychological processes.
I conclude that, as currently formulated, the humanization objection is unconvincing. Some perpetrators of inhumanity do, in fact, hold dehumanizing attitudes toward their victims, although they plausibly also see their victims (to some extent) as human. Yet, even if this is the case, we may wonder whether dehumanizing attitudes ever motivate those who hold them to commit inhumanity. Why think that dehumanizing attitudes are ever motivationally significant? To answer this question, we must understand the motivating force of interpersonal recognition. And this topic brings us to a second, rather more challenging objection to humanism.
3. The Motivational Significance of Interpersonal Recognition
The humanization objection is frequently presented alongside, and less often distinguished from, a distinct yet related line of attack against humanism. If the humanization objection alleges that perpetrators of inhumanity typically do not hold dehumanizing attitudes toward their victims (or that such attitudes are motivationally otiose), then the motivational insignificance objection alleges that humane behavior is not explained by the agent's recognition of the patient's humanity. In its most influential form, the complaint is that seeing someone as a human being is insufficient for explaining why we treat him humanely or perhaps of the wrong shape entirely.
Here again Manne leads the charge. For Manne (Reference Manne2018: 148), the problem is that ‘the characteristic human capacities that you share [with the person whom you recognize as a fellow human being] don't just make her relatable; they make her potentially dangerous and threatening in ways only a human being can be’. Only a human being can be ‘an intelligible rival, enemy, usurper, insubordinate, betrayer’, for example, just as only a human being could possibly dominate, ‘coerce, manipulate, humiliate, or shame’ us (147). Thus, we should expect recognition of someone's humanity to include some awareness of the palpably interpersonal threat posed by his human mind. Even if recognizing someone's humanity activates benevolent dispositions toward him (e.g., sympathy, empathy), then, that recognition can sometimes trigger hostile dispositions toward him as well, such as the kind of distinctively interpersonal malice, resentment, contempt, and hatred that we reserve for fellow human beings. Worse still, the latter may well be stronger than the former. It is doubtful, then, that interpersonal recognition generally leads to humane treatment of the one so recognized. If anything, the opposite may be closer to the truth. The recognition thesis is false: humanism fails to explain humane behavior.
The motivational insignificance objection can be read in (at least) two different ways. On the first reading, recognition of someone's humanity is insufficient for humane motivation, given that such recognition also triggers motives for inhumane behavior in many contexts. The second reading is more radical still: that recognition of someone's humanity is incapable of explaining why the recognizer is motivated to treat that person humanely, when he is. On this reading, recognition of someone's humanity is a motivationally inert state of mind, unable to figure as anything other than an enabling condition in the motivation of humane behavior. And in either case, if the alternatives to a humanist project are plausible, and we can explain humane and inhumane behavior while adverting only to attitudes permeated with interpersonal recognition (e.g., socially situated stances), then the explanatory credentials of humanism are in danger. To assess the force of the motivational insignificance objection, I examine each reading in turn.
3.1. Insufficiency?
If the motivational insignificance objection is that interpersonal recognition is insufficient for humane motivation (in virtue of also triggering hostile motives), then its strength is limited. True, it does threaten the recognition thesis, as Manne construes it—again, the claim that if A sees B as a (fellow) human being, then (ceteris paribus) A is robustly disposed to treat B humanely in virtue of A's recognition of B's humanity. But read as stating that seeing someone as human to any degree activates a robust disposition to treat her humanely, this thesis is plainly dubious, and I know of no humanist (or scholar of dehumanization) who accepts it in this form.
To resist the motivational insignificance objection, the humanist can embrace the following alternative formulation of the recognition thesis that retains the spirit of the original:
Strong Recognition Thesis: If A fully (or clearly) sees B as a (fellow) human being, then A is robustly disposed to treat B humanely (at least partly) in virtue of A's recognition of B's humanity.
It is far more difficult to challenge the strong recognition thesis by appeal to the sorts of considerations adduced by Manne. To dispute it, it is not enough to point to a person who meets only the minimal conditions for seeing someone else as human in that he recognizes the other party's humanity to some extent, yet fails to treat the target humanely (or who has no strong motive for so treating him). Instead, one the critic must show either that (1) there is a person who fully/clearly sees someone else as human but is not strongly disposed to treat him humanely or that (2) if such a person is strongly disposed to treat the one so recognized humanely, it is not due to fully/clearly seeing the recognized party's humanity. Thus, cast as a point about the motivational insufficiency of recognition, the motivational insignificance objection does not cut against this thesis without the aid of supplementary premises (that seem open to doubt, in any case)—such as that agents of inhumanity always fully/clearly see their victims as human beings.
As I have argued, there is a solid case for thinking that some agents of inhumanity do not fully/clearly see their victims as human beings, at least while mistreating them; if so, they cannot constitute counterexamples to the strong recognition thesis. For example, by all accounts, a typical enslaver in the antebellum U.S. South regarded his victims as human in some key respects but not in others, or else saw them simultaneously as human and as other than human. If the enslaver befriends one of the enslaved people in question, the transition is best characterized as his coming to see the enslaved person's human mind more clearly or more fully—say, the fact that the enslaved person has an inner life that makes him an intelligible partner in human friendship.
Additionally, the humanist can embrace a more modest version of the recognition thesis, on which seeing someone as human only activates a disposition to treat her humanely:
Modest Recognition Thesis: If A sees B as a (fellow) human being to any degree, then A is disposed to some degree to treat B humanely (at least partly) in virtue of A's recognition of B's humanity.
For this claim to be true, recognition of someone's humanity need only activate some disposition to treat him humanely, which may be blocked by potentially stronger countervailing motives or by general motivational disorders. Per the thesis, then, the disposition so activated by the recognition of someone's humanity need not always or even normally be effective in moving us to act. That does not imply, however, that the disposition activated—or recognition of someone's humanity—would be motivationally insignificant, failing to help explain why the recognizer is moved to humane treatment. While weaker than its predecessor, the modest recognition thesis is still quite interesting, and the humanist need not assert any stronger thesis just to count as a humanist after all. Interpreted in this way, humanism is not obviously threatened by the motivational significance objection.
Remember, too, that Manne grants, for the sake of argument, that recognition of someone's humanity tends to make us empathize with him. If that point is admitted, it is harder to maintain that (fully/clearly) recognizing someone's humanity triggers no disposition to treat him humanely—not even one that is blocked by an even stronger motive. As I will understand it, empathizing is the mental act of taking up or sharing another subject's perspective in an emotionally charged manner—normally, by vicariously feeling what he is feeling or by feeling an emotion that is somehow congruent to his. To empathize with someone, in my sense, is to be prone to feel (the emotion that is congruent to) what he is feeling because he is feeling it, and particularly when he is perceptually available to us.
In the sense that I have in mind, then, empathy generates some affective pressure to share his attitudes—sympathy at his suffering, shame at his contempt, pride at his esteem, perhaps. We can resist the pressure, of course, by means of—inter alia—reflection: by judging that his suffering is deserved, his contempt is baseless, or his esteem insulting, say. But we normally feel some pressure to give his attitudes some weight with respect to the question of how we should feel or what we should think. Empathizing pushes us in the general direction of humane motivation toward its object, although its success or failure depends on the strength of countervailing motives and on the extent of our vulnerability to akrasia, accidie, and the like. If that is the case, the motivational insignificance objection has even less force against the strong recognition thesis and the modest recognition thesis than may be thought.
Perhaps critics could reply that the affective pressure generated by recognition of someone's humanity is weak. There is some evidence from social psychology, however, that a disposition to empathize with a person is, in fact, correlated with treating him with basic consideration. (For evidence of the link between empathy and altruism, for example, see the studies canvassed in Batson [Reference Batson, Forsyth and Hoyt2011].) Accordingly, it is unlikely that empathizing with someone, or empathizing with him fully, produces so little motivating force that it never or rarely succeeds in moving a person to humane behavior, even if it must compete with other motives or with the limits of the agent's rationality. Read as a claim about interpersonal recognition's insufficiency for humane behavior, then, the motivational insignificance objection does not jeopardize the most credible versions of the humanist view, and the onus is on the critics of humanism to show that empathy provides an especially weak motive for humane behavior.
3.2. Inertness?
The motivational insignificance objection can also be read as the stronger claim that seeing someone as a human being is motivationally inert, incapable of explaining why anyone is motivated to act humanely. The critic of humanism may therefore deny that interpersonal recognition is tightly linked to empathetically identifying with him, or with any particular set of motives. She might even claim that recognition is a cognitive or representational state with mind-to-world direction of fit, which restricts it to serving only as an enabling condition for humane motivation instead of as a source of the relevant motives. I will argue that, so construed, the objection tends to rest on an overly intellectualized conception of interpersonal recognition. Much more argumentation is needed to endanger humanism than has been provided so far.
Although she intends for her statement of humanism to be ecumenical, a closer look at Manne's (Reference Manne2018: 142) discussion reveals that she often speaks of interpersonal recognition in intellectualist terms—as a way of ‘thinking’ about people that ‘is generally supposed to comprise (inter alia) thinking of them as having, or at least as having had, the potential to’ exercise a wide range of characteristically human capacities. The suggestion seems to be that interpersonal recognition is a form of belief or judgment about its object's psychological properties. That is why, I submit, it seems natural to Manne to think that the humanist needs to posit some motivating state that is separate from, but triggered by, our seeing someone as human. (For a similar criticism of Manne's view on this score, see Crary [Reference Crary and Kronfeldner2021: 162].)
It will seem mysterious how seeing someone as a human being could play any substantial role in the motivation of humane behavior (or any behavior) if we assume that this state of mind consists in believing that she falls into a biological category—that of (fellow) member of the species Homo sapiens. Why should believing that someone is a member of one's biological species trigger such responses as sympathy or empathy—indeed, any particular motives? And although Manne's conception of a (fellow) human being is richer, it invites a parallel worry: if seeing someone as human consists in believing that he is a subject of certain psychological properties, well, why should that move us to treat him humanely (or in any particular way), either? More generally, though, seeing someone as a human being will seem motivationally inert so long as it is conceived as a belief about him that is of the same form as any other, distinguished only by its content—by the fact that it purports to be about a person and not, say, about a stone or a tree.
But is the humanist, in fact, saddled with the implication that interpersonal recognition is motivationally inert? It is common to conceive of interpersonal recognition along the lines of what we might call the classification model: the view that seeing someone as a human being is the purely intellectual act of classifying an entity as falling under a particular (biological, psychological, metaphysical) category. (For elaboration and critique of this picture, see Tarasenko-Struc [Reference Tarasenko-Struc and Kaspar2020].) It is unclear, admittedly, why a mental act of that kind would be motivationally engaged by its very nature. Insofar as we are unable to imagine a credible alternative to this picture, the implication of motivational inertness will seem inescapable.
We might instead embrace a Kantian humanist position on which seeing someone as a person consists of regarding him as a bearer of moral standing, as on Axel Honneth's (Reference Honneth2001) view. For Honneth (Reference Honneth2001: 125), interpersonal recognition is a kind of ‘evaluative perception’ that is constitutively linked to showing concern and respect toward the one recognized, while dehumanization consists in the occlusion or corruption of our capacity to regard others in this way.
But we also find another, less explicitly moralized alternative to the classification model in Wittgenstein's later philosophy. Wittgenstein appears to deny that seeing someone as human amounts to or rests on the belief that she falls into a particular biological category or that she possesses certain psychological properties. Rather, his view is that interpersonal recognition consists in what he evocatively calls ‘an attitude towards a soul’, a web of person-specific forms of affect and concern toward her in light of her perspective (Wittgenstein Reference Wittgenstein and Anscombe1953: 178e).
In what follows, I want to articulate the view frequently attributed to Wittgenstein, and arguably latent in his later philosophical work, and to set aside the interpretive question of whether this reading of him is faithful to the text. This view finds its fullest expression in the work of David Cockburn (Reference Cockburn1990), Alice Crary (Reference Crary and Crary2007), Cora Diamond (Reference Diamond1978), Raimond Gaita (Reference Gaita2002, Reference Gaita2004), Peter Winch (Reference Winch1981), and, surprisingly, the early John Rawls (Reference Rawls1958) and later Axel Honneth (Reference Honneth and Jay2008). The Wittgensteinian thought, I take it, is that seeing someone as a human being is fundamentally a matter of emotionally identifying with her as such—by being susceptible to such sentiments as sympathy and, perhaps, Strawsonian reactive attitudes (e.g., love, gratitude, resentment) toward her. On this view, interpersonal recognition is not, at the most basic level, intellectual or doxastic. It is not a mental act of dispassionate categorization. Rather, it is the overall, ground-floor orientation of the emotions and will vis-à-vis the recognized party's perspective. Seeing someone as human has built-in motivational content, then, implicating a bundle of interlocking affective dispositions and other forms of noninstrumental concern for her mental states. Call this view the engagement model of interpersonal recognition.
The engagement model easily dispels the mystery of how seeing someone as human could be motivationally engaged by its nature. On this view, seeing someone as human is constitutively linked to a range of dispositions for person-specific modes of affect and concern vis-à-vis her perspective. And some of these modes of affect and concern—sympathy, for one—constitute motives for treating the recognized party humanely. The engagement model therefore fits well with the modest recognition thesis: roughly, the claim that seeing someone's humanity to any degree disposes us—to some degree—to treat her humanely. If the engagement model is tenable, a humanistic view backed by it defuses this strand of the motivational insignificance objection.
Of course, the fact that the engagement model would solve the problem does not show that it is defensible. Fully defending it is beyond the scope of this essay. Still, a proper defense of this position might proceed by first noting that there is a profound mismatch between the more familiar classification model and our experiences of encountering other minded human beings.
Consider, for example, what it is like to experience being looked at by another person. Suppose that you realize that you are the object of another person's gaze. Then, it will normally be very difficult for you to be utterly indifferent to that fact, registering it as just so much affectively neutral data relevant to predicting her behavior toward you. And this is so even if she is a total stranger to you and you have no particular designs on that person (or she on you). Indeed, should you go on to meet her gaze, you will have a particularly intense experience of her perspective. Why should the experience of being looked at by someone feel so charged?
One intuitive explanation is that the experience of being looked at is tinged with a felt susceptibility, on your part, to that person's perspective, and particularly her attitudes toward you. In this respect, encountering people just feels significantly different, from the inside, than encountering other entities such as objects and even nonhuman animals. The experience of another person's perspective feels potentially emotionally loaded in a way that, say, the experience of objects in your perceptual field—or of the perspectives of animals—simply does not.
It is not that another person's gaze always moves us to feel some particular sentiment or other. Sometimes we are able to meet the other's gaze with clinical detachment. Still, when a person looks at us, her attitudes toward us are experienced as immediately relevant to the question of how we are to feel about ourselves—even if we reject her attitudes as inapt, thereby settling that question for ourselves, and feel nothing. I submit that our ordinary experience of being looked at is inflected with a network of dispositions to affect and concern—what Wittgenstein called ‘an attitude toward a soul’. Moreover, this element of affective engagement does not present itself as a mere projection of our sentiments onto the perspectives of others; rather, it strikes us as woven into our awareness of them as minded. The phenomenology of encountering others suggests, then, that seeing someone as human ineluctably contains a kernel of emotional identification with her.
Thus, there is some phenomenological evidence in favor of the engagement model. That evidence is not conclusive, and its strength is a matter for debate. But it should give us pause that the classification model appears so at odds with our experiences of other people's perspectives—a discrepancy that at least imperils the view's unearned status as the default position on the topic. And if the engagement model is defensible, then the motivational insignificance objection is easily defanged, and the humanist is well positioned to establish the modest recognition thesis. People who see others as human beings would thereby be disposed to treat them humanely.
The critics still have a point. This modest humanist view does not exclude the possibility that in seeing someone as human we may also harbor motives for treating her inhumanely. In emotionally identifying with her as a human being, we may feel antipathy toward her as well as sympathy or empathy, and this may give us motives for both humane and inhumane behavior toward her. But, again, that possibility presents a problem only for a humanist view on which seeing a person's humanity unleashes motives that normally suffice for humane behavior toward her. The psychological facts are more complex than that. Yet the humanist need not deny them.
Read as the stronger claim that interpersonal recognition is motivationally inert, the motivational insignificance objection is persuasive only if the humanist is committed to the classification model, which I have argued is not necessarily so. As against this received view, I have contended that the engagement model constitutes a coherent alternative that better harmonizes with the phenomenology of encountering other people. We have been given no compelling reason to believe that interpersonal recognition lacks motivational significance entirely, then. The motivational insignificance objection thus endangers humanism far less than it may seem.
4. Conclusion: Taking a Step Forward
I conclude that the two most prominent objections to humanism (and adjacent views) do not succeed. The humanization objection fails because once we clarify the assumptions underpinning the strongest version of it, we see that they are questionable and readily rejected. We also discover a host of counterexamples that suggest that some agents of inhumanity really do fail to regard their victims as human beings, which enables us to rebut even the weaker and more plausible version of the objection with ease. It is an open question how widespread dehumanizing attitudes are among agents of inhumanity. Still, if my argument is correct, then dehumanization is neither nonexistent nor a marginal phenomenon. The most pertinent question, then, is not whether people ever hold dehumanizing attitudes toward those whom they badly mistreat, but how prevalent these attitudes are and whether they best explain different strains of inhumane behavior.
The motivational insignificance objection, on the other hand, is inconclusive, at best. Read as an allegation that interpersonal recognition is insufficient for humane motivation, it is true but uninteresting, as it does not threaten either the most plausible or the most influential versions of humanism. However, read as the claim that seeing someone as a human being is motivationally inert, the objection presents a far more powerful challenge. To address it, I have recommended jettisoning the overly intellectualized conception of interpersonal recognition frequently presupposed both by the critics of humanism and by humanists themselves—the view that seeing someone as a human being consists, most basically, in holding a belief about him that is of the same form as any other belief about the world. In place of that conception, I have tried to articulate promising alternatives gleaned from the Kantian and Wittgensteinian traditions.
Let me close with three broader methodological points. First, critics of humanism tend to begin with the observation that sadism or morally motivated violence directed at human victims is based on some awareness of their humanity. The critics then infer from that fact that no inhumane behavior can be even partly explained by a failure to recognize the victims as human or that no humane behavior can be even partly explained by interpersonal recognition. The preceding discussion shows, however, that it is premature to draw such conclusions about humanism and that the evidence for it has generally been underestimated by its opponents.
Second, humanism can be an interesting moral-psychological thesis even if it does not entail the implausibly strong claim that seeing someone as a human being is normally sufficient for us to be motivated to treat him humanely. Likewise, humanists are best understood as claiming that the failure to see someone as a human being disposes us to treat him inhumanely (perhaps by suspending a disposition for treating him humanely)—not that most or all instances of inhumanity are best explained by the agents’ failure to recognize their victims as human. Again, it is up for debate just how much of the data are explained in terms of dehumanization.
Finally, while some effort has been devoted to clarifying what it is to see (or fail to see) someone as a human being, far less attention has been paid to the question of how closely interpersonal recognition is tied to the motives for humane/inhumane behavior. In particular, very little has been said about which motives, if any, are built into interpersonal recognition and, more broadly, whether seeing another person's humanity amounts to a dispassionate mental act of categorization or something saturated with affect. Questions concerning the motivational import of interpersonal recognition (and failures of recognition), I submit, deserve a central place in discussions of what moves people to commit inhumanity against their fellow human beings.
