Biology & Philosophy

, Volume 26, Issue 4, pp 583–593

Empathy’s purity, sympathy’s complexities; De Waal, Darwin and Adam Smith


    • LEI, Wageningen University and Research Centre
Open AccessArticle

DOI: 10.1007/s10539-011-9248-4

Cite this article as:
van der Weele, C. Biol Philos (2011) 26: 583. doi:10.1007/s10539-011-9248-4


Frans de Waal’s view that empathy is at the basis of morality directly seems to build on Darwin, who considered sympathy as the crucial instinct. Yet when we look closer, their understanding of the central social instinct differs considerably. De Waal sees our deeply ingrained tendency to sympathize (or rather: empathize) with others as the good side of our morally dualistic nature. For Darwin, sympathizing was not the whole story of the “workings of sympathy”; the (selfish) need to receive sympathy played just as central a role in the complex roads from sympathy to morality. Darwin’s understanding of sympathy stems from Adam Smith, who argued that the presence of morally impure motives should not be a reason for cynicism about morality. I suggest that De Waal’s approach could benefit from a more thorough alignment with the analysis of the workings of sympathy in the work of Darwin and Adam Smith.


MoralitySocial instinctsSympathyEmpathyDe WaalDarwinAdam Smith

Introduction: biology and morality

The discovery of the embodiment of thinking, including moral thinking, is a long and wide-ranging voyage, in which Darwin was the great pioneer. Progress is not always smooth. For example, when in 1975 E.O. Wilson found it time to remove ethics temporarily from the hands of philosophers and give it to biologists instead (Wilson 1975: 562), he met with wide resistance. Philosophers and social scientists considered biology with suspicion anyway, and Wilson’s proposal was not phrased in a way that invited cooperation. Several decades later, the atmosphere has changed greatly. Ethics is not taken from anyone’s hands, but a party of biologists, philosophers and social scientists has gathered around it. Thinking is now widely considered as embodied and therefore full of emotion, intuition, images and large unconscious parts that make our conscious thought look like the tip of an iceberg. The question is no longer whether morality is somehow rooted in our evolutionary past, but how. Jonathan Haidt, the influential advocate of an intuitionist “new synthesis” in moral psychology (Haidt 2001, 2007), points to two books of the 1990s that deserve special mention in explaining the change of atmosphere: Damasio’s Descartes’ error (1994), which showed that emotions are crucial for moral decision-making and De Waal’s Good natured (1996), which showed that “building blocks” for morality can be found in chimpanzees and other primates (Haidt and Kesebir 2010).

Frans de Waal continues to write influential books, with increasingly outspoken views of human nature and the biological origin of morality. He presents these views as being in direct line with Darwin, who famously wrote in The descent of man (2004/1871) that “any animal whatever, endowed with well-marked social instincts” is bound to acquire a sense of morality as soon as it becomes intelligent enough.

In this paper I compare the views of De Waal and Darwin on the origins of morality. Focusing on the central instincts they see at work when it comes to morality,1 sympathy for Darwin, empathy for De Waal, I will argue that there are not only similarities to be seen, but also important differences. De Waal’s emphasis on empathy will appear as a move in a search for the morally good side of human nature, which is opposed to a bad or self-interested side and which he sees as the direct basis of morally good behaviour. For Darwin, the relation between sympathy and morally good behaviour is more complex. The “workings of sympathy” centrally include a self-interested element.

In order to appreciate the philosophical significance of the differences between Darwin and De Waal, it will be helpful to include Adam Smith in this discussion. Darwin derived his view of sympathy from Adam Smith’s Theory of moral sentiments (1976/1759). De Waal, too, refers to this book. I will argue that Darwin’s analysis of sympathy is indeed in line with Smith, while De Waal’s interpretation is common but mistaken. This is of more than historical importance. The theory of moral sentiments contains important arguments for a morality on the basis of mixed motives. I will suggest that De Waal’s approach would benefit from a more thorough alignment with Darwin and Adam Smith.

De Waal: good natured

When Frans de Waal was educated in ethology, that discipline, under the influence of Konrad Lorenz, strongly focussed on aggression. But when De Waal looked at chimpanzees he did not just see aggression, he also saw reconciliation. From that time onward, he has been elaborating a view in which the good has a place in human and animal nature alongside the bad. Important books in the development of his thoughts are Good natured (1996), Primatesand philosophers (2006), in which he explains his thoughts in dialogue with philosophers, and The age of empathy (2009). The books convincingly show countless continuities between human and other animals that are relevant in the search for origins of morality.

Thinking about evolutionary roots of morality is interwoven with thinking about human nature. De Waal wholeheartedly agrees with John Dewey that we are social to the core and that living together in societies is our state of nature. For the gardener who wants to turn wild nature into a moral garden, nature is not just the enemy T.H. Huxley saw in it. Dewey argued that the gardener can work with nature as much as against it. For De Waal, this means that we are moral beings to the core (1996: 2), and he has increasingly come to identify this core with empathy. In The age of empathy, after citing Darwin’s view that animals with well-marked social instincts will develop morality as soon as they become intelligent enough, he wonders what those social instincts are and in the next sentence reformulates this as the question what makes us care about others (2009: 8). The answer, empathy, is the radiant centre of the book, of which the first sentence, “Greed is out, empathy is in” is a clear mission statement.

In Primates and philosophers, De Waal uses images to clarify his view. We are not selfish spheres covered by a thin layer of moral veneer; it’s more adequate to characterize our moral selves as Russian dolls. Upon opening one moral doll, you will find another, and another. The core is “emotional contagion”; it is the fully automatic way in which we are sensitive to the emotional state of others (2006: 26/39). Mirror neurons presumably afford the core mechanism, while the surrounding dolls add capacities which finally yield full grown human empathy.

Empathy is not the whole story of human nature; De Waal wants to complement Lorenz’s view, not to overthrow it. In humans, as in other primates, he sees two diametrically opposed tendencies. He is not always very precise about the poles. The shadow side is sometimes selfishness, sometimes aggression, sometimes competition or greed. What seems central in the big dualism he sketches is the opposition in which good, disinterested empathy, which leads to altruism, is opposed to bad, self-interested aggression, competition, or greed. The dualism sometimes leads to caricatures, for example when he describes bonobo’s as gentle and sexy, chimpanzees as brutal and domineering, and humans as “bipolar apes” with characteristics of both (2009: 203).

To support the dualistic character of human nature, De Waal refers to the first sentence of Adam Smith’s Theory of moral sentiments in all the books considered here (1996:28; 2006:15; 2009: 2). This sentence says that “How selfish soever man may be supposed, there are evidently some principles in his nature, which interest him in the fortune of others, and render their happiness necessary to him, though he derives nothing from it except the pleasure of seeing it”. It says a great deal about the distinctions we need to make, says De Waal, that Smith, who is so well known for his emphasis on self-interest, also wrote about the universal human capacity for sympathy (2006: 15).

Below, after discussing how Darwin and Smith understood sympathy, I will argue that De Waal’s dualistic interpretation of Adam Smith is unwarranted. But let me first indicate why De Waal shifted his preference from sympathy to empathy. Both concepts are about being in emotional contact with the situation of others and they are often used interchangeably. But empathy’s main focus is on feeling, while sympathy also tends to include agreement, concern, some kind of action. This difference is central in De Waal’s shift.

Darwin did not face a choice between sympathy and empathy, because the word empathy did not exist in English until the beginning of the twentieth century.2 De Waal also started out with sympathy. As he explains in Good natured (1996: 41), where he still preferred sympathy, as well as in The age of empathy (2009: 88), where he had come to prefer empathy, empathy is a basically passive process of information gathering. In comparison, sympathy is more action-oriented. Through its pro-active nature, sympathy is more complex, and speaking about his personal experience he notes that his sympathy is “less spontaneous, more subject to calculation, sometimes even selfish (…) Instead of flowing straight from my empathy, my sympathy takes a detour through a rational filter” (2009: 89). In other words, empathy is more automatic, more basic and more purely disinterested, which is why De Waal prefers it to sympathy in his search for the good side of human nature.3 Only empathy really satisfies the requirements of a moral emotion, of which disinterestedness is the core (2006: 20). In short, De Waal’s shift from sympathy to empathy is motivated by his search for an unambiguously good core in animals and humans.

Darwin: the workings of sympathy

In chapter 4 of The descent of man, Darwin describes the development of morality in four steps. First, the social instincts lead an animal to take pleasure in the society of its fellows. Second, as soon as the mental faculties become highly developed, the brains of an animal will have to deal with images of past actions and motives, which invariably results in feelings of dissatisfaction or even misery. Third, as soon as language develops, the common opinion can become a guideline for behaviour. And fourth, there is the formation of habits.

The second of these steps, which mentions feelings of dissatisfaction and misery, is important for understanding Darwin’s view on how sympathy and morality are related. Sympathy is a central social instinct. Darwin characterizes it, with reference to Alexander Bain and Adam Smith, as being based in our strong retentiveness of former states of pain or pleasure. Those states are reactivated when we see hunger, pain or fatigue in other persons, and this impels us to relieve the suffering of others (2004/1871: 129).

But sympathy is not yet morality, since morality requires normative judgment: a moral being is “capable of comparing his past and future actions and motives, and of approving and disapproving of them” (p. 135). In order to explain how sympathy leads to morality, Darwin emphasizes that different instincts can be in conflict.4 When we act, he says, we often follow strong impulses directed at the gratification of our own needs, at the expense of those of others. But afterwards, when our desires are satisfied and have disappeared, we look back and feel miserable. The social instinct and our deep regard for the good opinion of others become more prominent in our thoughts, and we will feel regret, remorse, and shame. On the basis of such brooding, a person will “resolve more or less firmly to act differently for the future; and this is conscience; for conscience looks backwards, and serves as a guide for the future” (p. 138).

Time and again, Darwin emphasizes the importance of the worry about others’ opinions in this brooding, and he stresses that it is inherent to “the workings of sympathy”. It is Mr Bain who has clearly shown, he writes, that “the love of praise and the strong feeling of glory, and the still stronger horror of scorn and infamy, are due to the workings of sympathy” (p. 133).5

In this understanding of sympathy, the need we have of other people’s sympathy, which includes the fear to be deprived of it, joins our tendency to sympathize. Together, they lead to moral actions. Within the mix, the role of the need for sympathy is large. Writing about the development of morality in primeval man, for example, he suspects that, “it is hardly possible to exaggerate the importance during rude times of the love of praise and the dread of blame” (p. 157).

What results is a complex whole: “Ultimately our moral sense or conscience becomes a highly complex sentiment—originating in the social instincts, largely guided by the approbation of our fellow-men, ruled by reason, self-interest, and in later times by deep religious feelings, and confirmed by instruction and habit” (p. 157).

Darwin, in short, does not characterize sympathy itself as morality or as morally good, but focuses on its complex workings and the mechanisms that lead from sympathy to morality, where self-interest has a large role. His understanding of sympathy is derived from Adam Smith, to whom I now turn to see how he defended moral behaviour against requirements of moral purity.

Adam Smith: mixed motives

Adam Smith’s Theory of moral sentiments was published in 1759, exactly a century before Darwin’s Origin of species and 112 years before The Descent of Man. Sympathy is central in Smith’s understanding of morality. In the first sentence of the book he states that, how selfish soever we may be thought to be, we have an interest in others. Through the imagination we conceive of the passions of other persons, or rather, we imagine what we ourselves would feel in the situation of other persons, as he explains in the chapter.

The need we have for the sympathy of others is introduced in the second chapter, of which the first sentence reads:

But whatever may be the cause of sympathy, or however it may be excited, nothing pleases us more than to observe in other men a fellow-feeling with all the emotions of our own breast; nor are we ever so much shocked as by the appearance of the contrary.

Smith’s book can be characterized as a rich study of the search for balance in giving and taking sympathy, and of the corresponding search for the mutual propriety in dealing with the passions. His “laws of sympathy”, as they have been called,6 consist in this two-directionality: on the one hand we like the sympathy of others and fear its absence, on the other hand we like to be able to sympathize with others (Kerkhof 1992). Finding this balance is a moral achievement for which Smith introduces the famous “impartial spectator”, a hypothetical “man within the breast” who helps us to look at the various needs and perspectives and to determine what, in each situation, is the proper balance between our needs and those of others.

Throughout the book, our need for sympathy and praise receives abundant attention. For example, when Smith wonders what all the ambition, “all the toil and bustle of this world”, is about, the answer reads: “to be observed, to be attended to, to be taken notice of with sympathy” (p. 50). Yet, we are not completely dependent on others, at least not inevitably. People do not just want to be loved and praised, they also want to be lovable and praiseworthy, and this desire, Smith adds, is a goal in its own right, it not just derived from the love of praise. Here again it is the impartial spectator who helps: we must become the impartial spectators of our own behaviour in order to strengthen the love of praiseworthiness (p. 114). Doing so allows the wise among us to achieve a certain independence from the opinion of others and develop standards of our own. Wise men too derive pleasure from being praised, but only for what they think is really praiseworthy (p. 117).

People emerge from Smith’s analysis as imperfect beings in an imperfect world, deeply dependent on one another and therefore driven by a mix of motives. This view unites his two great books, The Theory of Moral sentiments and The Wealth of Nations. The idea that these books deal with opposite sides, or different conceptions, of human nature is known as “Das Adam Smith Problem” and characterizes the understanding of many, including Frans De Waal. The consensus among Adam Smith scholars, however, is that the gap does not exist (Tribe 2008). Convincing arguments against it are given by many authors (e.g. Coase 1976; Kerkhof 1992; Vernon Smith 1998).7 My reading of Adam Smith makes me agree with Vernon Smith’s summary: the “propensity to truck, barter and exchange one thing for another” is the core of Smith’s view of human nature in both books. In the one book, what is exchanged is goods, in the other it is sympathy.

Behaviour with moral merit is quite feasible on the basis of mixed motivation, according to Smith, since we should judge actions, not motives and feelings; the latter “are placed by the great Judge of Hearts beyond the limit of every human jurisdiction” (p. 105). He defends that view against Hutcheson’s idea that good behaviour is necessarily based on pure benevolence and that the trace of anything else, the hope of reward for example, completely destroys all notion of merit. Benevolence, says Smith, may perhaps be the sole motive of the Deity, but “so imperfect a creature as man, the support of whose existence requires so many things external to him, must often act from many other motives” (p. 305).

Smith defends this view also in response to Mandeville, who had been arguing that “public good” derives from “private vice”, the dependence on others’ good opinion, which he called vanity. Smith, having so strongly emphasized the love of praise and the dread of blame, agrees that Mandevilles’s view borders on the truth in some respects (p. 313). Yet he vehemently disagrees with Mandeville, in the same atmosphere in which he disagrees with Hutcheson’s requirements of pure motivation. Wherever we fall short of the most ascetic abstinence, Mandeville sees only gross luxury and sensuality, says Smith. He repeats that a motivational mix does not automatically become morally worthless when it also contains the love of praise, and concludes:

It is the great fallacy of Dr. Mandeville’s book to represent every passion as wholly vicious, which is so in any degree and in any direction. It is thus that he treats every thing as vanity which has any reference, either to what are, or to what ought to be the sentiments of others; and it is by means of this sophistry, that he establishes his favourite conclusion, that private vices are public benefits (p. 212–213).

In Smith’s view, pure motivation can hardly ever be expected of human beings, but that does not imply that wisdom and moral merit are illusions.

De Waal, Darwin, Adam Smith: some thoughts on alignment

De Waal emphasizes the continuity and similarity between his approach and the work of Darwin and Adam Smith; his writing suggests that he interprets their understanding of sympathy as more or less identical with his understanding of empathy. So far, this paper has argued that there are in fact important differences. De Waal, Darwin and Smith all think that the capacity to be emotionally affected by the situation of others is an important place of origin for the development of morality. For Darwin and Smith, this origin leads to moral norms and moral goodness through a complex system of mechanisms that includes selfish needs. Moral goodness is located at the level of actions, not of instincts or motivations. De Waal, on the other hand, focuses on the pure core of goodness in human nature. He locates not just the origin of morality but the very identity of the morally good in our capacity to empathize.

Calling attention to that difference is the main aim of this paper. But I also think that De Waal’s approach leads to some problems and could profit from a more thorough alignment with Darwin and Smith. I conclude by offering three lines of thought to support this suggestion.

First, De Waal’s emphasis on the pureness of empathy creates tensions within his own work, in which he also abundantly encounters selfish needs in the mechanisms that lead to moral norms and moral behaviour. Much of his work deals with fairness and reciprocity. As a consequence of his moralization of social instincts, in which only empathy emerges as really good, his judgement of other sources of morality is far less favourable, since they involve large chunks of self-interest. “We’re all for fairness, as long as it serves us”, he says, explaining that he agrees with Hobbes that we’re interested in justice just for the peace’s sake (De Waal 2009; 183–184). He continues by wondering whether he is “uncharacteristically cynical” here. “You’ve heard me explain at length how incredibly empathic, altruistic and cooperative we are, so why when we get to fairness is self-interest all that I see?” The inconsistency is not as great as it may seem, he says, because all behaviour must in the end serve the actors. But in the domain of empathy and sympathy, and only there, evolution has created a psychological mechanism that makes us genuinely care about others. “I fail to see how the same applies to our sense of fairness. Other-orientation seems such a small part of it”.

These remarks illustrate that De Waal is involved in two projects that are not clearly distinguished. One is to understand how moral norms and behaviour evolve. The other is to judge good and bad motives in human nature. The merging of these projects creates confusions. For example, is a moral being a being that makes moral judgements or a being that is morally good? As a consequence of the entanglement of descriptive and normative activity, the search for mechanisms becomes normatively loaded. Mechanisms and outcomes that include self-interest generate De Waals’s moral cynicism.

The moralization of motives and mechanisms leads to a second problem, which is loss of diversity in the search for the sources of morality. I’m not only thinking of the loss of selfish aspects of sympathy, but also of instincts that have disappeared completely. Apart from sympathy, Darwin mentioned various instincts as relevant for morality: love, faithfulness, self-command, loyalty, obedience, courage (2004/1871: 130–133). For De Waal, the (morally inferior) sense of fairness is the main addition to empathy.

It is interesting to look at parallels with the field of social psychology. Contemporary social psychology also associates moral goodness (in the form of pro-social behaviour, helping, cooperation) with empathy, while loyalty, obedience and conformity are mainly connected with morality through warnings of their moral danger.8

In recent years, social psychologist Jonathan Haidt has described as well as criticized the loss of diversity in social and moral psychology. On the basis of anthropological inventories, he argues that five domains of moral intuitions can be distinguished, but that moral psychology has been narrowing its perspective and now focuses on only two: harm/care (the domain of empathy) and fairness/reciprocity. Three domains suffer from neglect: in-group/loyalty, authority/respect and purity/sanctity. The parallels with the losses in biology are evident; De Waal’s framing of morality is clearly in line with the spirit in moral psychology. Haidt and his co-authors (e.g. Haidt and Graham 2007; Haidt and Kesebir 2010) comment that the lost domains are the domains of community-oriented rather than individual-oriented intuitions and that they deserve to be rediscovered and restudied as important domains of moral intuitions.

Finally, though De Waal’s intention is to focus on our good side, a dualistic moralization of human nature inevitably comes with an unloved bad side. When morality is associated with pure motivation, it becomes fairly rare, while the polluting self-interest is rejected and driven into a corner where it may become more angry and dangerous. Understanding at least some forms of self-interest as the need of socially dependent beings for praise and inclusion allows for moral merit on the basis of mixed motivation. It also enables more compassionate and perhaps wiser ways of dealing with self-interest than rejection.

A more accepting approach to self-interest can hardly fail to reveal that it exists in many varieties. According to John Dewey (relevant fragments of his work can be found in Rogers 1997), self-interest is not immutable but takes shape in processes of action and valuation. Many problematic forms of self-interest really come down to a very narrow and short term understanding of what that interest involves. The judgment that someone is selfish, wrote Dewey, means that this person “ought to define himself on the basis of a wider situation, that he ought to be taking into account factors which as a matter of fact he is neglecting. It means inadequacy almost always” (Rogers 1997: 235). Doesn’t this remind us of the impartial spectator who encourages us to take more perspectives into account? The Theory of Moral Sentiments is still an inspiring book to (re-)start the exploration of mechanisms of morality.


I will hardly pay attention to conceptual issues concerning instinct, intuition or morality, though I will point to some confusions. Neither will I discuss selection mechanisms and nature-nurture issues. I don’t believe these issues are unimportant, but they require more than a small detour and I think that the issues I do address can be meaningfully separated from these other discussions.


The concept of empathy originated in German aesthetics, as Einfühlung, at the end of the nineteenth century. For a review of relations between empathy and sympathy and their history, see Eisenberg and Strayer (1987).


De Waal does not consider empathy as perfect from a moral point of view. A limitation that he is very aware of is that empathy is in tension with impartiality, because it works most forcefully with those who are near or at least visible. Batson et al. (1995) have shown that “empathy-induced altruism” can lead to choices with immoral consequences, for example to allocating resources to people who do not need them most urgently. This phenomenon, as the authors note, is everywhere. It is the basis of media-wars that call for international aid: no aid without media-attention. De Waal acknowledges such limitations of empathy. There is simply far too little empathy in the world for “other people”, he writes; “If I were God, I’d work on the reach of empathy (De Waal 2009; 204).


For example, migrating birds have strong parental instincts as well as migratory instincts. If autumn arrives when they have still dependent young, the migratory instinct wins out: it is so powerful that the birds can be seen to desert their tender young, leaving them to perish miserably in their nests, Darwin notes. What an agony or remorse a swallow would feel, “if, from being endowed with great mental activity, she could not prevent the image constantly passing though her mind, of her young ones perishing in the bleak north form cold and hunger” (p. 137). Such miserable-making reflection is precisely what humans cannot prevent.


The section in Bain’s book that Darwin here refers to is titled: “The feeling of being approved, admired, praised by others, is a heightened form of self-gratulation, due to the workings of sympathy” (Bain 1868: 254). The explanation is that the sympathy of others strengthens us in our own sentiments and opinions. In another reference to Bain, Darwin explicitly refers to selfishness as increasing the power of sympathy: “With mankind, selfishness, experience and imitation probably add, as Mr Bain has shewn, to the power of sympathy; for we are led by the hope of receiving good in return to perform acts of sympathetic kindness to others” (p. 130). While this formulation involves a distinction between selfishness and sympathy, it also states that they cooperate in producing the power of sympathy.


By the Dutch author A.J Kerkhof (1992), who in turn refers to Campbell (1971) Adam Smith’s Science of Morals. The title of Kerkhof’s book, De mens is een angstig dier, is the Dutch translation of a quote from Adam Smith: “Man is an anxious animal” .


The apparent absence of benevolence from The Wealth of Nations is often supported by the much-quoted remark “It is not from the benevolence of the butcher, the brewer or the baker that we expect our dinner, but from their regard to their own interest”. In a lucid paper on Adam Smith’s view of man, R.H. Coase stresses, rightly in my view, that this remark is put it in a very different light by the text that immediately precedes it, where Smith stresses the deep dependence of people in civilized societies upon each other’s cooperation and assistance (Coase 1976: 543–44).


Textbooks of social psychology deal with morality (and evolution) in chapters about “pro-social behaviour” (Hewstone et al. 2008; Aronson et al. 2007) or “Helping and Cooperation” (Smith and Mackie 2007). Empathy is proposed as the basic driver of compassion and altruism. Loyalty and obedience on the other hand are not associated with morality except in the form of moral danger. The moral rejection of these traits is certainly partly connected with an important reason for the flourishing of social psychology after the second world war: the troubling question why so many people in Nazi Germany went along with Hitler. Through contributions such as Milgram’s famous obedience experiments (Milgram 2005/1974), and Zimbardo’s Stanford Prison experiment (re-addressed in Zimbardo 2007), loyalty, obedience and conformity became extremely suspect human tendencies, threatening rather than basic for morality. Milgram (2005/1974) himself saw obedience as a form of morality in its own right, however. In what he calls the “agentic shift”, “morality does not disappear, but acquires a radically different focus: the subordinate person feels shame or pride depending on how adequately he has performed the actions called for by the authority” (p. 147). The morality of obedience conflicts with the morality of empathy and justice. Milgram locates the main trouble in most people’s lack of resources to understand the web of conflicting forces that operate in social situations (p. 31) and to control their behaviour. In his foreword he describes this as a tragic element, “for nothing is bleaker than the sight of a person striving yet not fully able to control his behaviour in a situation of consequence to him”.



I am grateful to Henk van den Belt for his suggestions and encouragement. I also thank the reviewers for useful suggestions.

Open Access

This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.

Copyright information

© The Author(s) 2011