Several individuals who have studied "Moralism in Objectivism" have wondered about the biological basis of empathy. Some such people may find that this paper offers additional insight into Damian's views on this subject.

If you haven't yet studied the talk, "Moralism in Objectivism," since the following paper is included as an appendix, it's recommended that you first read or listen to that talk. For quick access, click here.

DRAFT-DO NOT CITE, REPRODUCE, OR DISTRIBUTE WITHOUT PERMISSION FROM THE AUTHOR. NOTE THAT THIS PAPER WAS WRITTEN FOR AN ACADEMIC AUDIENCE, WHICH USES THE TERM "ALTRUISM" SOMEWHAT DIFFERENTLY THAN AYN RAND, THE FOUNDER OF OBJECTIVISM, USED IT.

I. Philosophical Questions

Many people accept without question that they should behave altruistically towards each other and that the government should structure society to force people to do so. I, however, will question these commonly held beliefs and evaluate them using evidence from experiments I have conducted to study altruism in non-human primates. The main questions I wish to answer are (1) Should I behave altruistically towards my fellow humans? and (2) Should the government force people to do so? Of course, the scope of these questions is enormous and would take at least a lifetime to answer completely. However, I will focus on how empirical research in one field of study - animal cognition - can contribute to answers to these questions.

To answer (1), the first question I must ask is, What is the value of behaving altruistically towards my fellow humans? Many philosophers (e.g., Kant, 1785/1993) argue that an action is moral only if it is done for its own sake, not if the intention is to make oneself feel good. A few others (e.g, Rand, 1961) argue that, to the contrary, an action's moral value is determined by its contribution to the actor's happiness. While determining the primary standard of value for the morality of human behavior is the fundamental question of ethics (i.e., the answers to all other moral questions depend on the answer to this one), this is outside of the scope of this thesis. Rather, I will assume that one's own happiness is of at least some value to every person and leave other moral concerns aside.

Once I have determined that my altruistic behavior should be evaluated, at least in part, according to its contribution to my happiness - its emotional value - I can begin to evaluate it. I can casually observe that when I help other people achieve ends that they cannot achieve by themselves, I usually feel good. When I do not do so, I often feel bad. A psychologist might even do a comprehensive study to show that altruistic people are, on average, happier than people who aren't [REF?]. Thus, I might be tempted to conclude that I should, in general, if happiness is my ultimate value, behave altruistically.

However, to conclude this before analyzing the matter in more depth would be premature. Suppose, as a thought experiment, that I could press a button and no longer experience pleasure when I help others. This may seem like an undesirable option - why would I want to deprive myself of a source of pleasure? However, in doing so, I would also be depriving myself of a source of displeasure: the guilt and empathic pain I experience when I don't help someone who needs my help. If I pressed this button, I could go about my business, achieving my own goals (leaving aside the question of what goals would be valuable if I did not care about others) without being pulled by the emotional drive to help others. I may not choose to press this button, but I will not rule out the possibility that I or someone in my shoes would.

Of course, this button does not exist literally, but does it exist metaphorically? That is, can I choose to shut off the system that drives me to act altruistically? The answer to this question depends on the fundamental cause of altruism. If it is a social construct, a product of social norms, then it may be destructible. To a large extent, humans are norm-following beings [REF?], but this fact applies to humans neither universally nor deterministically. For example, Oliner and Oliner (1988), who interviewed members the relatively small proportion of non-Jewish Germans who rescued Jews during the Holocaust, found that while approximately 90% of the rescuers did so simply because they were following the altruistic norms their parents had instilled in them as children, approximately 10% of the rescuers did so because they it concorded with their own moral principles, which they derived independently of the norms of both society at large and their parents. It is not clear what the source of this independence is, but it is clear that it is possible for people - or at least some people - to resist social norms as the cognitive and emotional mediators of their actions. If altruism is, in fact, entirely a social construct, it is at least possible that a person could, through independent thought and probably only with much time and effort, decide to reject it and obtain no emotional value, ceteris paribus, from behaving altruistically. However, if the source of altruistic behavior is, to a large extent, hard-wired and genetically determined, then a person's attempt to eradicate the emotional value of altruism would be mostly futile.

Before proceeding down this line of inquiry, let me return to question (2), Should the government force me to behave altruistically? To answer this, I must know how I would act in the absence of laws. If I saw a starving person on the street and there were no laws requiring me to pay taxes that would be redistributed by the government to this person, would I help this person? We might observe the proportion of people's incomes that they donate to charity, note that, in aggregate, they are not enough to save the majority of people starving on our nation's (let alone the world's) streets, and conclude that the only way to ensure that everyone's welfare meets minimum standards is to require people to give what they would not give of their own volition.

However, this conclusion would also be premature. Our behavior in today's society is not necessarily representative of how we would behave were there no laws. In fact, there are good reasons to expect that we would donate much less to charity in today's society than we would were there no laws. First, those of us who are concerned with the general welfare will have to give much less to charity given that we do have a welfare system than if we didn't in order to achieve a given level of welfare. To illustrate, if we wish everyone to have a minimum of 100 units of wealth and the government provides for 75 of these units, then we have only 25 units to make up for by donating to charity, while if we did not have a welfare system, we would have to donate all 100 units. Also, if there is intrinsic value in helping others (i.e., if we value it as an end in itself, not as a means to an end), this intrinsic value could be decreased if the value is made extrinsic by the government. In support of this hypothesis are experiments (reviewed in Deci, Koestner, and Yan, 1999) in which, for example, children who were paid (i.e., were given extrinsic value) to play a game that they already enjoyed (i.e., that had intrinsic value) ended up playing the game less once the extrinsic reward was removed. Similarly, if people naturally enjoy helping others and are then required by law to do so, the subsequent enjoyment may be decreased and thus our current level of giving might be less than what it would be if there were no laws requiring it.

Philosophers such as Hobbes argued that a government must force people to give to each other because, in a "state of nature," we would not do so. Hobbes (1651/1996) argued that outside of society, people's interests are fundamentally opposed. If two people desire the same thing, which they usually do, they become enemies and desire to destroy each other. Thus, outside of society, people are in a perpetual state of war. He claims that "men have no pleasure, (but on the contrary a great deale of griefe [sic]) in keeping company, where there is no power able to over-awe them all" (p. 88). If true, this entails that people take no pleasure in behaving altruistically towards each other. Indeed, he said that the only factors that checks this perpetual war are fear of death and desire for comfortable living - certainly not good will towards others as a value in itself. Since life in a state of nature is "solitary, poore [sic], nasty, brutish, and short" (p. 89), Hobbes concluded that we must have a government to force us to behave altruistically towards each other. Of course, a conclusion is only as sound as the premises from which it is derived, so we must question how sound Hobbes's premises are. What evidence did Hobbes provide to sustain the claim that there can be no altruism in a state of nature? The only empirical observation he provides is of the "brutish" (p. 89) Native Americans, whom we now know to have been, ironically, quite altruistic towards each other [REF?].

Thus, in order to determine whether people would, in general, behave altruistically towards each other in the absence of laws, we must again ask what the source of altruism is. We may now integrate question (2) with question (1): Is the source of altruism the norms and laws of society, or is it our genetic heritage? Knowing the answer to this question will help us answer questions (1) and (2).

II. Clarification and Identification of Key Concepts

Before we can look for the source of altruism, we must know precisely what we are looking for. There are two things we can look for, each of which is appropriate to focus on in different contexts: altruistic behavior and the psychological mechanism that mediates this behavior. Before we can proceed with our search, we must define the former and identify the latter.

A. Clarification of the Behavior

Altruism is defined and used in many different ways. For clarity, it should be broken up into three distinct but related concepts. Sober (1988) and Sober and Wilson (1998) break altruism into two concepts: evolutionary altruism and either vernacular altruism (Sober, 1988) or psychological altruism (Sober and Wilson, 1998) (vernacular altruism and psychological altruism are synonymous). Evolutionary altruism is defined as behavior that increases the fitness of another and decreases the fitness of oneself (Sober and Wilson, 1998; for similar definitions, see Dawkins, 1976; Trivers, 1971; Wilson, 1975). Fitness, in this context, is equivalent to reproductive success. Vernacular or psychological altruism is behavior that is consciously and ultimately intended to increase another's welfare, "regardless of how, or even whether, they think or feel about the action" (Sober and Wilson, 1998, p. 6; for similar definitions, see Eisenberg & Mussen, 1989; Staub, 1978; Cialdini et al., 1981). Though correlated, evolutionary altruism does not necessarily imply vernacular/psychological altruism. For example, a plant in a crowded area could induce its own death are thereby allow its offspring to survive and this would be considered evolutionary altruism. Neither does vernacular/psychological altruism necessarily imply evolutionary altruism. For example, "If I know that you love to play the piano, I may give you a volume of Beethoven sonatas out of the goodness of my heart" (Sober, 1988, p. 76), and this may actually diminish your reproductive success because you may consequently spend more time at the piano than in the bedroom. The problem with leaving the breakdown of altruism at just the distinction between evolutionary and vernacular/psychological forms is that it is not clear what "the goodness of my heart" means. Sober (1988) explicitly chooses not to address this question, but it is essential to do so for the present analysis. The question is, broadly speaking, Is vernacular/psychological altruism ultimately egoistically or non-egoistically motivated? In other words, Do people help each other ultimately because it makes them feel good or ultimately because they care about the welfare of others? There is a long-standing debate on this question (see, e.g., Batson, 1991; Cialdini et al., 1987), but I will not attempt to address it as stated because it is ambiguous. We first need to distinguish between conscious and subconscious motivation. I may say that my ultimate reason for giving my brother a birthday gift is that I care about him, but subconsciously my real motive may be that I am worried that he will be mad at me if I don't. Alternatively, my subconscious ultimate motivation may be that it simply makes me feel good to see him happy.

Thus, I will introduce two new concepts to distinguish the two different types of vernacular/psychological altruism. The first type is affective altruism, which is defined as behavior that is intended to help another as a direct means to the ultimate end of one's own emotional gratification. While the intention of helping another must be conscious, the knowledge that one's ultimate end is one's own emotional gratification need not be conscious. The second type is cognitive altruism, which is defined as behavior that one believes to be intended to help another as a means to the ultimate end of promoting the other's welfare. Here both the intention of helping and the believed ultimate end of promoting the other's welfare must be conscious. Clearly, affective altruism does not imply cognitive altruism. For example, if I help an old lady to cross the street for the explicit purpose of making myself feel good, this is affectively altruistic but not cognitively altruistic. Also, cognitive altruism does not necessarily imply affective altruism. For example, if I help an old lady to cross the street because I think it is right to help other people but doing so doesn't actually make me feel good, this is cognitively altruistic but not affectively altruistic. Of course, it is possible for a behavior to be both affectively and cognitively altruistic.

For example, if I tell myself that I am helping an old lady cross the street because it is right to help other people but I am really doing so because helping her directly makes me happy, this is both affectively and cognitively altruistic.

For the purpose of this thesis, I am concerned primarily with affective altruism. For altruistic behavior that is reflex-like and non-volitional - as is the case with altruism that is evolutionary but not affective or cognitive - no ethical question can be asked, for an ethical question, by its nature, presupposes volition. Between affective altruism and cognitive altruism, the former is a much better candidate for being part of human nature. Kant (1785/1993) argues that we should follow the categorical imperative (which implies that we should behave cognitively altruistically) not because it is in our nature to do so, but in spite of the fact that it is not in our nature to do so. Social norms advocate cognitive altruism and consider affective altruism, at best, neutral and, at worst, disingenuous. Thus, it is more likely that cognitive altruism than affective altruism is a product of social construction rather than human nature, and therefore, it is on affective altruism that I will focus.

The reader may wonder why I have chosen to focus on benevolence rather than altruism. I did so because altruism is defined differently by biologists, psychologists, and philosophers, and since this thesis may be read by members of any of the three groups, to use the term altruism would be ambiguous, confusing, and possibly misleading. Biologists tend to define altruism purely in terms of behavior: behavior that causes a cost to the self and a benefit to another (e.g., Dawkins, 1976; Trivers, 1971; Wilson, 1975). Psychologists tend to add an intentional component: behavior that causes a cost to the self and is intended to cause a benefit to another without external motivation, as to receive material or social rewards (e.g., Eisenberg & Mussen, 1989; Staub, 1978; Cialdini et al., 1981). Philosophers tend to add a motivational conponent: behavior that causes a cost to the self, is intended to cause a benefit to another, and is motivated by concern for the beneficiary and not with the self. [REFS] Some psychologists, such as Batson (1991), also define altruism with this motivational component. The best definition to use to answer the questions I have posed is the second, which specifies the intention but not the motivation. The reason for this is that to determine how I should behave and how the government should regulate behavior, intentions matter and I do not want to limit types of motivation. Intentions matter because whether I should behave a certain way and whether the government should regulate such behavior depend on the behavior's underlying intention. For example, the ethical and political analysis of giving would be entirely different if the behavior in question were accidentally dropping wallets on the street, giving one's money when one is robbed at gunpoint, or voluntarily giving money to charity to pay for food and shelter for homeless people. Motivation also matters for ethical and political questions, but determining the motivation that underlies benevolent behavior is a very complex, controversial topic (e.g., Batson, 1991; Cialdini et al., 1987) that is outside the scope of this thesis. Rather, I will leave motivation completely unrestricted, so as not to exclude intentionally helpful behavior that is egoistically motivated. In fact, with regard to the ethical question I have posed, I am particularly concerned with egoistically motivated helping because I am specifically analyzing the contribution of helping to one's own happiness. To avert the many possible confusions and misinterpretations that could arise if I use the word "altruism," I stipulate the following: Henceforth, when I use the word "benevolence," I will be referring to the intentional, non-motivational definition of altruism. [REFS SHOWING HOW SOME PHILOSOPHERS USE "BENEVOLENCE"] Also to avoid ambiguity, when I refer to the non-intentional, non-motivational type of altruism, I will use the phrase, "evolutionary altruism" (Sober & Wilson, 1998) to indicate that I refer to the biologist's definition of altruism. [CHANGE ALL REFERENCES TO "BIOLOGICAL ALTRUISM" TO "EVOLUTIONARY ALTRUISM"]

B. Identification of the Psychological Mediator

The definition of affective altruism I have just provided includes a behavioral component and a psychological component. Still, this psychological component, the intention to help another, does not identify the fundamental psychological mediator. By fundamental psychological mediator, I mean the process that gives rise to the intention and which cannot itself be psychologically reduced (though it may be reduced neurologically). This mediator, I contend, is empathy. "Empathy" has been used and defined in many different ways; I will try to choose one definition that is both unambiguous and concords with how the term is usually used. First, empathy must be differentiated from its synonym, "sympathy." In a historical review and an analysis of the distinction between these two concepts, Lauren Wisp (1986) defines sympathy as the "heightened awareness of the suffering of another person as something to be alleviated" (p. 318) and empathy as "attempt by one self-aware self to comprehend unjudgmentally the positive and negative experiences of another self" (p. 318). In short, sympathy is a way of relating to others while empathy is a way of knowing others' experiences. Others define empathy less cognitively and more emotionally. For example, Barnett (1987) defines empathy as the "vicarious experiencing of an emotion which is congruent with, but not necessarily identical to, the emotion another person is experiencing" (p. 146). Batson and Oleson (1991) define it similarly as "an other-oriented emotional response congruent with the perceived welfare of another person" (p. 63), the difference being that the response is to the other's perceived welfare rather than emotion, so that, for example, one can feel empathy for a person who is unconscious. Davis (1994) contends that one of the primary reasons empathy is defined so differently by different people is that the term is being used to denote two disparate processes, which can be called cognitive empathy and affective empathy. Wisp's definition is of cognitive empathy while Barnett's and Batson and Oleson's are of affective empathy. Cognitive empathy, as the name implies, requires more cognitive mechanisms than affective empathy. Specifically, it requires a theory of mind, knowledge that others have minds that are distinct from one's own, for which there is little evidence of its existence in human infants and any animals other than humans (summarized in Tomasello, 1997). Affective empathy, on the other hand, requires an affective mechanism by which the emotional state of one person causes another person to experience the same or a similar emotion. For clarity, I will not use the term "empathy" in isolation; rather, I will always precede it by either "cognitive" or "affective."

While cognitive empathy could result in affectively altruistic behavior, it normally would not do so without the presence of affective empathy. For example, if I am told that my friend has been locked in a closet, I can put myself in his shoes to realize that he is probably suffering, but what will motivate me to help him will be my distress at knowing that he is distressed (especially if I hear his screams) and the knowledge that my distress will be relieved or that I will be happy once he is released. I could also be motivated by a sense of justice or a desire for social reward, but since these, along with cognitive empathy, are much more cognitively complex and likely more culturally contingent than affective empathy, affective empathy seems to be a much better candidate for an innate basis of affective altruism. While there are, to be sure, many learned or socially constructed factors that cause affective altruism, for the purposes of this thesis, I am concerned with innate factors and, in particular, the most likely fundamental innate factor - affective empathy. Introspectively, we can easily discern the relationship between affective empathy and affective altruism. When we see someone suffer - especially if it is someone we care about and who shows many affective signs such as visible wounds, blood, screaming, and a pained facial expression - we suffer also and are often inclined to help that person. In a meta-analysis, Underwood and Moore (1982) did not find a significant correlation between affective empathy and affectively altruistic behavior, but note that the correlation is stronger when empathy is induced by an actual or experienced rather than a hypothetical or imagined situation. Eisenberg and Miller (1987) conducted a more comprehensive meta-analysis and found a low to moderate correlation between affective empathy and affective altruism, though they postulate that the actual correlation is much higher but is underestimated in the meta-analysis due to weak or questionable indices of empathy (such as response to hypothetical or imagined situations) and affective altruism.

III. Is Affective Altruism Innate?

Now that affective altruism has been defined and the psychological mediator we are concerned with has been identified, we can proceed to ask whether and in what sense affective altruism and affective empathy are innate. Before answering the question though, we must establish that the question itself is valid. In other words, is the so-called nativist-empiricist debate a meaningful debate at all?

A. A Valid Question?

Certainly some people who call themselves nativists and empiricists believe it is (e.g., Spelke, 1998; Haith, 1998), but others contend that nature and nurture are inseparable and that asking whether something is innate is meaningless or misguided (e.g., Lehrman, 1953; Hinde, 1974). I will argue that the debate can be misguided, but that it is still possible to ask meaningful questions about the extent to which a cognitive process is innate or learned.

Many psychologists, when addressing such issues, begin with the obligatory statement that nature versus nurture is a false dichotomy (e.g., Jacobson, 1974). Nonetheless, although these critics maintain that psychologists should look at the interaction between the organism and the environment, they still go on to implicitly support this dichotomy with synonymous terminology such learned versus innate behavior, closed versus open programs of development, biological constraints and predispositions, and hard-wired versus soft-wired behavior (Johnston, 1987). I will argue that the nature-nurture dichotomy has shortcomings and can be used inappropriately, but that it can be valid, useful, and interesting when answering a particular type of question. The general type of question is: Is the first appearance of a behavior or psychological mechanism a result of interaction with the aspect of the environment that it concerns? I will consider this question specifically in the context of human behavior and psychology, but let us first explore how the study of animal behavior and psychology has set the stage for the nature-nurture debate.

Lorenz (1937) proposed that deprivation experiments be used to determine whether a given behavior was learned or innate. This works well in simple cases in which removal of an environmental stimulus clearly does or does not eliminate a behavior, but future ethologists showed that many cases are not so simple. Marler (1993) used bird song to show that animal behavior is often best characterized by instincts to learn, treating nature and nurture no longer as a dichotomy but an interaction. Unlike Lorenz, Marler showed that innate mechanisms are not necessarily determined to follow a single, stereotyped path. Instead, certain primitives reside in an organism without any environmental stimulation, but these primitives are modified by innate mechanisms by which organisms learn from their environment.

Before Marler showed empirically why an interactionist approach to nature and nurture is necessary, others argued theoretically that Lorenz's distinction is untenable. Lehrman (1953), Hinde (1974), Smith (1999), and others contend that characterizing a behavior as innate explains nothing about the mechanism by which innate mechanisms arise, and thus contributes nothing to our understanding of the behavior. Lehrman and Hinde claim that all psychologists are doing when they call something innate is "passing the buck" along to other developmental disciplines because they are not explaining the proximate causation and mechanism by which a cognitive process arises. It is often true that nativists do not explain how a cognitive process arises, but when answering the type of question described above, just calling something innate or not can still be very interesting and informative.

For example, when attempting to discover the basis for people's first knowledge about objects, Spelke (1998) asks "whether knowledge of things in the external world develops on the basis of encounters with those things" (p. 192). Asking a nature-nurture question in this manner renders so many of the criticisms of the nature-nurture dichotomy, particularly of nativism, irrelevant. By answering whether knowledge of something results from interaction with that something, one does not specify how the initial state arises, but one can still say something interesting. If, for example, psychologists discovered tomorrow that American babies will know the grammar, semantics, and lexicon of the English language whether or not they hear others speak it, people would not say that these psychologists are saying nothing of import just because they don't specify the mechanism by which this knowledge develops from one's genes.

Johnston (1987) suggests that information cannot be encoded in one's genes because genes code only for proteins, which develop only by interacting with the environment. Analogously, computer software code specifies which pixels on a screen will be turned on and off, and this happens only with help from the environment of the computer's hardware, but it would be absurd to claim that the code does not contain information if the resulting pixels form the words of this essay. To solve this problem, we must not look at the genes and the environment in the abstract, but rather define them for a particular purpose. Certainly the genes that somehow code for innate knowledge must be expressed by a plethora of developmental events that rely on interactions with the environment, but the only aspect of the environment in question is, as Spelke claims, "encounters with those things." If the answer to Spelke's question is interesting, it does not matter whether the environment in any abstract sense of the word contributes to the development of knowledge.

Likewise, innate can be defined for this particular context rather than in the abstract. If knowledge of things in the external world develops not on the basis of encounters with those things, then we can call this knowledge innate. The uselessness of a universal definition of environment is obvious from developmental biology's observations that cognitive processes are often influenced in important ways by interactions with the "environment" that have no obvious relation to these processes. An organism's life follows a developmental chain that requires certain events to happen in certain periods, but it is useless to argue that all developments that depend on an earlier event show that these later developments are learned from the earlier event because this does not explain development in any meaningful way. For example, it would be absurd to conclude from a person receiving brain damage from early malnutrition and thus not being able to speak properly that language is learned from food. A proper interactionist approach - however useful or useless - may assert that all behavior results from an interaction between an organism and its environment, but not that all behavior is learned. We avoid this problem by asking "whether knowledge of things in the external world develops on basis of encounters with those things," thus restricting the definition of the environment to encounters with those things about which an organism has knowledge.

Let us also not forget that the question under consideration refers only to the initial state of knowledge. Nativists do not deny that knowledge is shaped by learning from the environment, but they may contend that a person's first knowledge about a particular domain is not a product of interaction with that domain. That initial knowledge is innate in this sense does not imply that it cannot be modified by the environment in the future (Marler, 1998), nor that it is necessarily present at birth (Johnson, 1998), as with the genetically determined physiological changes that occur at puberty, such as the emergence of pubic hair.

While nature versus nurture is in many senses a false dichotomy, it can be useful and interesting in certain circumstances if the terms of debate are clearly defined. Now that we have established that to ask whether something is innate can be a meaningful question, we can proceed to specify the question. Which is an appropriate question, (A) Is affective empathy innate?, or (B) Is affective altruism innate? When we ask whether something is innate, that something is a mechanism, not an output of the mechanism, so question A is appropriate and question B is not. It is valid to ask, Are the psychological mechanisms that cause affective altruism innate?, but, given that we have identified affective empathy as the psychological mediator of affective altruism of interest, this question is equivalent to question A. Nonetheless, we shall see that it is sometimes more useful to analyze behavior than psychology when answering this question.

B. How to Determine Innateness

There are many ways to answer the question as to whether something is innate. I will focus on six general ways of doing so: evolutionary, developmental, neurological, ethnological, ethological, and experimental.

Evolution

The first way is evolutionary. This approach is primarily theoretical, though, to be sure, evolutionary theory is primarily a product of empirical research (see, e.g., Darwin, 1859). Evolutionary theory can be used to hypothesize how likely it is for a trait to evolve. The evolution of traits can be analyzed both at the behavioral level and the psychological level, with some qualifications. If we analyze behavior, we must ask how the psychology that underlies the behavior is restricted by evolution. If we analyze psychology, we must ask how this psychology translates into behavior. To summarize the relationship between evolution and psychology, "one could think of evolution as generating bounds on utility functions for the species. The alchemy of the endocrine system and the emotions can be thought of as a powerful tool in this work" (Skyrms, 1996, p. 42).

In terms of behavior, four primary theories have been proposed to explain evolutionary altruism by evolutionary biology (for further analysis, see Skyrms, 1996 and Ridley, 1996). First, the most common explanation for evolutionary altruism proposed by non-professionals today and by biologists until the latter half of the last century is that organisms give to each other "for the good of the group." This explanation invokes the theory of group selection, which holds that groups of organisms (typically populations or species) give to each other because groups that do so outperform groups that don't in competition for resources that promote reproductive success, such as territory, food, and mates. While it is true that evolutionarily altruistic groups will outperform non-evolutionarily altruistic groups, this is not usually a stable state of affairs, for within a group composed entirely of evolutionarily altruistic individuals, if a mutant or migrant appears who is not evolutionarily altruistic, it will reap the rewards of others' generosity without paying the price that others pay. This individual's reproductive success will be greater than that of the evolutionary altruists, and so its and its descendents' reproductive success will tend to be greater than that of its fellow group members, and so, over time, the non-evolutionarily altruistic strategy will tend to take over. Mathematical models show that except in very special conditions, group selection loses out to individual selection because individual selection is faster (see, e.g., Williams, 1966 for a critique of group selection).

Sober and Wilson (1998) adamantly defend the coherence and existence of group selection. Their primary argument is that evolutionary biologists who oppose group selection define it out of existence by means of "the averaging fallacy." That is, people say that the average fitness of the altruist is greater than that of the non-altruist, and therefore individual selection, not group selection, causes evolutionary altruism to thrive. However, Sober and Wilson argue, within groups the altruists actually fare worse than the non-altruists, and so individual selection cannot be the cause of the survival of the evolutionarily altruistic trait, but the within-group differences are washed out when the groups are averaged. Moreover, they argue that kin altruism and reciprocal altruism (described below) are each sub-types of group selection since kin and non-kin groups are selected and pairs of individuals who interact repeatedly are selected. While this analysis may or may not be valid, at this point the debate becomes semantic (i.e., pertaining to a definition of a word, not the content of a theory), and so, for the purposes of this thesis, it is not worth considering any further.

Implicit in the preceding analysis is a "gene's eye view" of natural selection (Dawkins, 1976). That is, the unit of selection is properly regarded as the gene rather than the organism, the population, or the group. Genes will tend to increase in a gene pool if they promote the propagation of identical copies of themselves. Regarding organisms as the unit of selection ignores the fact that a gene that promotes the propagation of identical genes in other organisms will be selected for even if it does so at a cost to the organism in which it resides as long as it does so with a net benefit to its own propagation.

This view explains, second, Hamilton's (1964) theory of kin selection, which holds that evolutionarily altruistic behavior will be selected for if it increases the benefactor's inclusive fitness, which is the fitness of its own genes plus the product of the beneficiary's genes and the coefficient of relatedness between them, which is equivalent to the fraction of genes that they share. For an evolutionarily altruistic behavior to be positively selected, Hamilton shows, the cost to the benefactor's own fitness must be less than the product of the benefit to the beneficiary's fitness and the coefficient of relatedness between them. Thus, in this context, this theory predicts that no one will sacrifice its life for 1 other, but one should sacrifice its life for more than 2 siblings, more than 4 half-siblings, and more than 8 first cousins.

Third, Robert Trivers (1971) showed that evolutionarily altruistic behavior will be selected for if it increases the likelihood that another will reciprocate. Suppose that organisms play a non-zero sum game. That is, if one organism gives to the other, the recipient will gain more reproductive success than the donor will lose. While in any given instance, it is to an organism's advantage to not sustain a cost in order to give a benefit to another (given that the recipient is not kin), in the long run it is to the organism's advantage to do so if, in so doing, it receives in return. However, this strategy will not work unless the organisms who play these non-zero sum games do not continually give to "cheaters" who take but do not reciprocate. Robert Axelrod (1984) found in a computer simulation that the most evolutionarily stable strategy is what's known as TIT-FOR-TAT, in which an individual gives initially to a given player and from then on does what the other player did on its last turn. Thus, if the first player "cheats," the second player "punishes," and if the first player "repents," then the second player "forgives."

Fourth, organisms can manipulate others into behaving in ways that benefit their own inclusive fitness (summarized in Ridley and Dawkins, 1981). For example, an organism can deceive another into thinking that it is kin so that the second organism will behave evolutionarily altruistically towards it. Or a predator may pretend to be prey to attract another organism to it. More interesting if we are interested in explaining affectively altruistic behavior mediated by affective empathy are cases like the following: a chimp was observed to limp only when in view of his rivals, possibly to elicit affective empathy so that he would not be attacked (de Waal, 1996).

All of these evolutionarily based behaviors - kin altruism, reciprocal altruism, exploitative altruism - may or may not be explained proximately by affective empathy. A mother bird may feed her infant or a vampire bat (Wilkinson, 1984) may reciprocate blood-sharing because they affectively empathizes with the beneficiary or because it is a reflex to regurgitate food they are prodded the right way. Thu

In terms of psychology, the most compelling theory of the evolutionary basis of affective empathy is Buck and Ginsburg's (1991) communicative gene hypothesis. Buck and Ginsburg argue that the ultimate (in the biological rather than philosophical sense) purpose of affective empathy is spontaneous communication, which they define as "nonintentional, nonpropositional, affective communication about feelings and desires or emotions and motives" (156). Through spontaneous communication, one perceives directly what is going through another's mind by way of signals such as pheromones, vocalizations, and facial expressions. If such communication makes the recipient anticipate feeling good by behaving affectively altruistically towards the information provider, he or she will be motivated to do so. Buck and Ginsburg review evidence that much communicative behavior is genetically encoded and selected upon (see Hauser, 1996 for a review). Most importantly, they argue that it is important to treat spontaneous communication as the basis for selection rather than evolutionary altruism, or, specifically, kin altruism or reciprocal altruism. Kin altruism theory would predict that, ceteris paribus, organisms should be more evolutionarily altruistic towards kin than towards others. The communicative gene hypothesis, on the other hand, predicts that organisms should be more evolutionarily altruistic towards those from whom they receive the most spontaneous communication. In most species and in most humans who do not live in developed nations, animals live with and therefore receive the most spontaneous communication from kin. Those who have migrated into their group are often treated as pseudokin and are also recipients of evolutionary altruism. Goodall (1986) suggests that those who are not members of an organism's social group are often treated as members of another species and do not elicit affective empathy when they show signs of suffering (also see Wrangham & Peterson, 1996). If the communicative gene hypothesis is true, it would not be surprising that people sometimes behave affectively altruistically towards non-kin who show signs of suffering and who are not members of an outgroup.

Development

The second way is developmental. Much is known about how affective altruism and affective empathy develop in children (e.g., Eisenberg and Strayer, 1987; Ungerer et al., 1990; Zahn-Waxler and Radke-Yarrow, 1990). However, most of this knowledge will not help us much in answering the question of innateness. The reason is that once an child is born, it is immediately exposed to social input. A baby is not capable of doing anything affectively altruistic until it is at least two or three years of age, at which point it is capable of giving material objects or verbal support to others, and before this point, it has probably been socialized by its parents and others to behave affectively altruistically.

Since the study of the development of affective altruism is thus not very useful, let us turn to affective empathy. Even before a baby learns language, it can still learn what it needs to appear as if it has affective empathy. For example, a baby may learn that whenever its mother smiles, something desirable will happen like feeding or tickling, and may therefore smile also, and whenever its mother frowns, something undesirable will happen like yelling or neglect, and may therefore frown also. Thus, the most effective way to determine whether affective altruism is innate using the developmental approach would be to see whether newborns have affective empathy.

There is a problem, however, with determining whether affective empathy is innate based only on observations of newborns: it could still develop after birth. While "innate" is often equated with "present at birth," this is not a necessary equivalence. Traits that are innate can emerge well after birth. An obvious, non-psychological example is pubic hair, which emerges at puberty but is not a product of environmental influences aside from adequate nourishment. Traits that emerge at birth can also be not innate, such as familiarity with the mother's voice, which is learned while in the womb. The former problem could lead to a false negative, while the latter could lead to a false positive. In investigating whether affective empathy is innate, the latter is not relevant - at least in terms of facial expressions, a fetus has no exposure to them and therefore cannot learn associations in the womb. The former problem could be relevant, but before evaluating whether there is a problem of obtaining a false negative result, let us see whether or not the result is, in fact, negative.

Perhaps the best developmental test of the innateness of affective empathy was done by Martin and Clark (1982) as a refined replication of experiments by Simner (1971) and Sagi and Hoffman (1976). In this experiment, they showed that 18-hour old infants cried in response to recordings of the cries of other newborn infants, even though these infants had had very little exposure to such cries before then. (Simner [1971] and Sagi and Hoffman [1976] obtained similar results with 70 and 34 hour old infants, respectively.) Thus, it is unlikely that their response is a result of a learned association. Moreover, they did not cry in response to recordings of the cries of 11-month-old infants, infant chimpanzees, nor of themselves. This implies that their crying was not simply a response to an aversive stimulus, which is further corroborated by adult ratings of the 11-month old infant cries and infant chimp cries as more aversive than the newborn cries. The only plausible explanation left for these results is that the newborns experience affective empathy towards each other and that it is innate (moreover, it is peer and species specific).

Still, there are two problems left with this interpretation. First, while unlikely, given the paucity of exposure, it is possible that newborns learn in the first few hours of life to associate distress calls of their peers with negative consequences for themselves. Second, Hoffman (1990) argues that children pass through different stages of empathy, and perhaps the type of empathy that infants possess is not sufficient to cause affectively altruistic behavior. According to Hoffman, the four stages of empathic development are: (1) Global empathy. Before the age of 1, a child does not clearly differentiate itself from others, and therefore experiences others' distress as its own distress and when it sees another hurt, acts as though it was hurt. (2) "Egocentric" empathy. Between the ages of 1 and 2, a child can recognize that it is the victim, not him or herself, who is in actual pain or danger, but, without a developed theory of mind, will still not understand that their internal states are distinct from its own. (3) Empathy for another's feelings. Between the ages of 2 and 3, when a child's theory of mind becomes more fully developed, a child can begin to understand that his or her feelings are distinct from the victim's. Still, such empathy can be aroused only by directly perceivable distress cues from the victim. (4) Empathy for another's life condition. After the age of 3, when a child develops a full-blown theory of mind and appreciates the importance of others' past and future existence in addition to present existence, a child can experience empathy in response to indirect information about a person's general level of distress in addition to direct information about a person's immediate level of distress. In his stage-model, Hoffman does not distinguish between affective and cognitive empathy, but it is clear that the psychological processes that differentiate the first stage from the latter three are cognitive (e.g., self-other distinction, theory of mind), not affective. As Hoffman shows, children are more likely to behave affectively altruistically when they have these additional cognitive components. However, the affective component of empathy is still necessary. To illustrate, a person who could easily understand the mental state of someone in distress but did not experience distress in response would likely not help to relieve this person's distress given that social norms do not play a role. Moreover, the affective component is sufficient, though not as powerful. If a person experiences distress in response to distress cues from others and learns that particular behaviors stop these distressing distress cues, he or she could learn through simple operant conditioning to behave affectively altruistically. (Even though this association may be learned, affective altruism would still be considered innate in the sense I am using it here since its primary psychological mediator, affective empathy, is innate.) Thus, the type of empathy that newborns appear to possess is sufficient to elicit affectively altruistic behavior, and so this is not a problem.

Still, the developmental approach does not yield a conclusive answer as to whether affective empathy and affective altruism are innate, and so other approaches to the problem must be pursued.

Neurology

The third way is neurological. Establishing that a psychological process has a neural correlate does not imply that this psychological process is innate. However, if it is established that an area of the brain that is shown to correlate with a psychological process develops in an organism before that organism has undergone experiences that would allow it to learn or acquire that psychological process, then this suggests that this psychological process is innate. This method of inquiry might be useful in determining whether a pre- or non-linguistic organism uses a particular psychological process that is not always clearly expressed behaviorally, for if the organism cannot say or show what it is thinking, the next best way to find out might be to "see" what it is thinking by looking at a brain scan. Of course, before one can infer what another is thinking from a brain scan, one must first establish the connection between the psychological process of interest and a particular brain area by interviewing or observing an organism that can express the psychological process through speech or behavior.

Eslinger (1998) has reviewed a number of lesion studies on the neurological basis of empathy and established that the primary locus of empathy is the frontal cortex. In particular, he has shown that cognitive empathy and affective empathy probably have distinct neural correlates. In a study of adult frontal lesion patients, Eslinger, (1996) found no significant correlation between cognitive and affective measures of empathy. Moreover, Grattan and Eslinger (1989) found that where cognitive empathy and affective empathy are not distinguished, dorsolateral frontal lesions correlate strongly with impairment in both empathy and cognitive flexibility, while orbital frontal lesions correlate strongly with impairment in empathy but not cognitive flexibility. This is significant because cognitive flexibility (i.e., flexible thinking, as measured by, e.g., persevorative errors on the Wisconsin Card Sorting task) is hypothesized to be part of the foundation for cognitive empathy, mediated by role-taking and perspective-taking. Thus, it appears likely that affective empathy has a unique neural correlate in the orbital frontal cortex. To determine whether and to what extent affective empathy is innate, we may investigate whether and to what extent the orbital frontal cortex develops before an organism undergoes experiences that would allow it to learn to be affectively empathic.

Ethnology

The fourth way is ethnological. Ethnology is the science that studies and compares human cultures. If a cognitive process is universal to the human species despite differences in environments, it follows that it is innate. To be innate in the sense used here, the environments must differ in the element that one hypothesizes to be necessary for the cognitive process's development. For example, if affective altruism is encouraged in some cultures and discouraged in other cultures, and yet people are affectively altruistic in all cultures, this suggests that affective altruism may be innate. However, there are a number of problems with this argument, which I discuss below.

Fiske (1991) has surveyed prosocial behavior in cultures throughout the world and found that they are universally prevalent. In particular, he argues that four basic relational behavior types can explain most human interaction and that they are primarily affectively altruistic in nature. The first is Communal Sharing, which is sharing without expecting specific reciprocation. The second is Authority Ranking, which is the reciprocal relationship of deference and protection between dominants and subordinates. The third is Equality Matching, which is behavior that promotes equality in give-and-take relationships. The fourth is Market Pricing, which is the exchange of goods according to their value. Even Market Pricing, he argues, is often motivated by a conception of fairness rather than rational self-interest. Fiske argues that, in general, these behaviors are not exhibited as conscious means to asocial ends. For example, a subordinate does not defer to a dominant explicitly in order to receive protection in return. Rather, such behavior is intrinsically motivated - it is pleasurable in itself, whatever its ultimate (i.e., fundamental) purpose.

Fiske claims that "Children do not need to learn the elementary relational forms, which are "built in by evolution". But the child must learn the culture-specific rules for implementing them" (p. 208). In brief, he claims that the forms are innate but the rules are not. He does not provide an argument to support this strong claim, but we can evaluate it ourselves. Fiske makes a strong argument for the universality of affective altruism; now we must ask what this implies about its source. It is likely that these relational forms are innate in human beings, for if they are, we would expect them to be universal. However, there are other possible explanations. First, one might argue that no human culture is entirely isolated - that we are all part of one macro-culture. Since we are all descended from a common ancestor, it is possible that these relational forms emerged from socially constructed norms before groups migrated and became isolated. Furthermore, no culture (at least of those that we have investigated) is completely isolated, for there is intergroup migration or communication in all cultures to some extent (this is true by necessity, for if we have investigated a culture, we have communicated with it). However, this argument is weak, for this model would not predict that the relational forms would be so stable but the relational rules would be so diverse. Also, the primary examples Fiske uses are of some the most isolated cultures we know of, such as the rural peoples of West Africa. Second, one might argue that similar environments (i.e., that which is common to all of the earth's inhabitable areas) cause similar behavior. However, it is hard to say what about the earth would constrain human behavior in this way - after all, these relational forms concern how people relate not to the earth, but to each other. Third, one might argue that everyone hit upon an obvious best solution to a common problem. All societies face scarce resources and all four of these relational forms allow people to play non-zero sum games to promote their welfare. However, if this were the case, we would expect such behavior to be extrinsically motivated - for people to think of such behavior as merely a means to the end of one's own welfare - but, according to Fiske, such behavior is intrinsically motivated. Thus, in seems highly plausible, though not certain, based on a cross-cultural analysis, that affective altruism is innate.

Ethology

The fifth way is ethological. Ethology is the naturalistic study of animal behavior. Since non-human animals have little, if any, culture (Nishida, 1987), in the sense that humans intentionally teach each other, it is likely that if animals behave affectively altruistically, such behavior is innate in them. Of course, observing evolutionarily altruistic behavior is only half of the task - the other half is establishing that it is affectively altruistic and, given the focus of this thesis, that it is a product of affective empathy. Once we establish that a non-human animal behaves affectively altruistically, we can use this fact in two ways to support the hypothesis that it is innate in humans. First, we can argue that their behavior is homologous to ours. That is, given the phylogenetic closeness between them and us and the fact that, given similar selection pressures (or lack of negative selection pressures) we would expect the genes for affective altruism to be conserved. Second, we can argue that their behavior is homoplasic to ours. If the affective altruism is very similar in function but quite dissimilar in form, it may follow that the affective altruism in us and them is not a product of shared genes but is a product of functionally similar selection pressures. At the very least, we can argue that the innateness of affective altruism in another species establishes the possibility that affective altruism can be innate in any species, including ours.

Frans de Waal (1996) has chronicled a plethora of observations of evolutionarily altruistic behavior in non-human primates. In particular, he provides a number of examples of apparently affectively altruistic acts motivated by affective empathy. For example, he observed a female chimpanzee screaming when another female gave birth, defend the mother the next day, and groom the mother in the weeks to come. de Waal reasons that the chimp probably felt for the mother while undergoing the painful childbirth and later helped her because of her affective empathy. Also, he has observed chimps and even ring-tailed lemurs caressing, consoling, and caring for sick, injured, and upset conspecifics. While suggestive, none of the examples he describes constitute conclusive proof that affective empathy is the underlying motivator for the apparently affectively altruistic behavior of these non-human primates. Indeed, he admits that "many hard-to-convey details in the behavior of chimpanzees make me intuitively agree" (p. 58) that they are motivated to help by affective empathy. He concludes that "although it is hard to know the intentions of animals, we can speculate about them and may one day reach the point of testing one interpretation against another" (p. 46). The purpose of the next way of testing the innateness of affective altruism is to do just that.

Experiment

The sixth way is experimental. The ideal experiment to ascertain whether affective altruism is innate would be to raise a child in the absence of social norms and rules and observe how affectively altruistically the child and subsequent adult behaves. Of course, this is completely unethical and impractical, so we must settle for something less, which is to combine the experimental approach with the ethological approach just described.

In nature, it is difficult to tell what an animal's motive is to behave evolutionarily altruistically towards a conspecific. Often, it could be motivated by affective empathy, or it could be motivated by some other factor, or it could be a simple reflex-like response to stimuli. For example, when a chimp or a macaque licks the wounds of a conspecific, it could be because the benefactor affectively empathizes with the victim; it could be because licking is a reflexive response to seeing wounds; it could be because the benefactor likes the taste of blood (de Waal, 1996). Similar explanations could be used to explain food giving (which, incidentally, is rarely observed in nonhuman primates [de Waal, 1996]). An experiment that eliminates the possibility of factors other than affective empathy, however, could test conclusively whether affective empathy is, indeed the responsible psychological mediator for the apparently affectively altruistic behavior.

[In the beginning of this paper, I presented a couple of normative philosophical questions and traced how we might answer them down through psychology and other biological sciences all the way to the need for experiments. Now I will trace back from the experimental level to the philosophical level by summarizing and integrating what I have presented.]

Notes

1. By this I do not mean to equate hard-wired with genetically determined. For example, something could be hard-wired - i.e., permanently encoded - into the brain due to pre-natal or post-natal environmental factors, and therefore not be genetically determined.

2. I am not using this in the biological sense that Mayr uses it in to distinguish ultimate and proximate levels of explanation. Rather, I am using it in the philosophical sense to mean the most fundamental end. That is, I may give my girlfriend flowers as a means to the end of making her happy, but making her happy is a means to the ultimate end of making me happy, which cannot be further reduced.

3. By direct, I mean that no subordinate ends separate this means from the ultimate end. For example, if I give a homeless person money as a means to the end of receiving gratitude from this person or praise from others, which is a means to the ultimate end of my own emotional gratification, this is indirect, and therefore is not an instance of affective altruism. Rather, if I give a homeless person money and consequently improve his welfare as a direct means to the end of my own emotional gratification, this is an instance of affective altruism.

4. Even though the terms are the same, I do not mean to imply that there is a necessary link between cognitive empathy and cognitive altruism and between affective empathy and affective altruism.

References

Axelrod, R. (1984). The evolution of cooperation. New York: Basic Books.

Batson, C.D. (1991). The altruism question: Toward a social-psychological answer. Hillsdale, NJ: Lawrence Erlbaum Associates.

Buck, R., & Ginsburg, B. (1991). Spontaneous communication and altruism: The communicative gene hypothesis. In M.S. Clark (Ed.), Review of personality and social psychology (Vol. 12) (pp. 149-175). Newbury Park, CA: Sage.

Cialdini, R.B., Baumann, D.J., & Kenrick, D.T. (1981). Insights from sadness: A three-step model of the development of altruism as hedonism. Developmental Review, 1, 207-223.

Cialdini, R.B., Schaller, M., Houlihan, D., Arps, K., Fultz, J., & Beaman, A.L. (1987). Empathy-based helping: Is it selflessly or selfishly motivated? Journal of Personality and Social Psychology, 52, 749-758.

Darwin, C. (1859). On the origin of species by means of natural selection, or the preservation of favoured races in the struggle for life. London: John Murray.

Dawkins, R. (1976). The selfish gene. Oxford: Oxford University Press.

de Waal, F. (1996). Good natured: The origins of right and wrong in humans and other animals. Cambridge, MA: Harvard University Press.

Deci, E.L., Koestner, R., & Ryan, R.M. (1999). A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation. Psychological Bulletin, 125 (6), 627-668.

Dugatkin, L.A. (1997). Cooperation among animals: An evolutionary perspective. New York: Oxford University Press.

Eisenberg, N., & Miller, P.A. (1987). The relation of empathy to prosocial and related behaviors. Psychological Bulletin, 101 (1), 91-119.

Eisenberg, N., & Mussen, P.H. (1989). The roots of prosocial behavior in children. Cambridge: Cambridge University Press.

Eisenberg, N., & Strayer, J. (1987). Empathy and its development. Cambridge: Cambridge University Press.

Eslinger, P.J., Satish, U., & Grattan, L.M. (1996). Alterations in cognitively- and affectively-based empathy after cerebral damage. Journal of the International Neuropsychological Society, 2, 15. New York: Cambridge University Press.

Fiske, A.P. (1991). The cultural relativity of selfish indivisualism: Anthropological evidence that humans are inherently sociable. In M.S. Clark (Ed.), Review of personality and social psychology (Vol. 12) (pp. 176-214). Newbury Park, CA: Sage.

Grattan, L.M., & Eslinger, P.J. (1989). Higher cognition and social behavior: Changes in cognitive flexibility and empathy after cerebral lesions. Neuropsychology, 3, 175-185. Philadelphia, PA : Philadelphia Clinical Neuropsychology Group.

Haith, M. M. (1998). Who put the cog in infant cognition?: Is rich interpretation too costly? Infant Behavior & Development, 21, 167-179.

Hamilton, W. D. (1964). The genetical evolution of social behavior. Journal of Theoretical Biology, 7, 1-52.

Hauser, M.D. (1996). The evolution of communication. Cambridge, MA: MIT Press.

Hinde, R. A. (1974). Biological bases of human social development. New York: McGraw-Hill.

Hobbes, T. (1651/1996). Leviathan. Cambridge: Cambridge University Press.

Hoffman, M.L. (1990). Empathy and justice motivation. Motivation and Emotion, 14 (2), 151-171.

Jacobson, M. (1974). A plenitude of neurons. In G. Gottlieb (Ed.), Aspects of neurogenesis (pp. 151-166). New York: Academic Press.

Johnston, T. D. (1987). The persistence of dichotomies in the study of behavioral development. Developmental Review, 7, 149-182.

Kant, I. (1785/1993). Grounding for the metaphysics of morals. Indianapolis: Hackett.

Lehrman, D. S. (1953). A critique of Konrad Lorenz's theory of instinctive behavior. Quarterly Review of Biology, 28, 337-363.

Lorenz, K. (1937). Über die Bildung des Instinkrbegriffes. Naturwissenschaften, 25. 289-331. Translated in K. Lorenz (1957), The nature of instinct. In C. H. Schiller (Ed.). Instinctive behavior (pp. 129-175). New York: International Universities Press.

Marler, P. (1998). The instinct to learn. In M. H. Johnson (Ed.). Brain Development and Cognition: A Reader (pp. 454-480). Oxford: Blackwell.

Martin, G.B., & Clark, R.D. (1982). Distress crying in neonates: Species and peer specificity. Developmental Psychology, 18, 3-9.

Nishida, T. (1987). Local traditions and cultural transmission. In B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham, and T. T. Struhsaker (Eds.), Primate societies (pp. 462-474). Chicago: Chicago University Press.

Oliner, S.P., & Oliner, P.M. (1988). The altruistic personality: Rescuers of jews in nazi europe. New York: The Free Press.

Rand, A. (1961). The virtue of selfishness. New York: Penguin.

Ridley, M. (1996). The origins of virtue: Human instincts and the evolution of cooperation. New York: Penguin Books.

Ridley, M., & Dawkins, R. (1981). The natural selection of altruism. In J.P. Rushton and R.M. Sorrentino (Eds.), Altruism and helping behavior: Social, personality, and devlopmental perspectives (pp. 19-39). Hillsdale, NJ: Lawrence Erlbaum Associates.

Sagi, A., & Hoffman, M.L. (1976). Empathic distress in the newborn. Developmental Psychology, 12, 175-176.

Simner, M.L. (1971). Newborn's response to the cry of another infant. Developmental Psychology, 5, 136-150.

Skyrms, B. (1996). Evolution of the social contract. Cambridge: Cambridge University Press.

Smith, L. B. (1999). Do infants possess innate knowledge structures?: The Con side. Developmental Science, 2(2).

Sober, E. (1988). What is evolutionary altruism? Special Issue of Journal of Philosophy, 14, 75-100.

Sober, E., & Wilson, D.S. (1998). Unto others: The evolution and psychology of unselfish behavior. Cambridge, MA: Harvard University Press.

Spelke, E. S. (1998). Nativism, empiricism, and the origins of knowledge. Infant Behavior & Development, 21, 181-200.

Staub, E. (1978). Positive social behavior and morality: Social and personal influences (Vol. 1). New York: Academic Press.

Tomasello, M. (1997). Primate cognition. New York : Oxford University Press. Trivers, R. L. (1971). The evolution of reciprocal altruism. Quarterly Review of Biology, 46, 35-57.

Underwood, B., & Moore, B. (1982). Perspective-taking and altruism. Psychological Bulletin, 91 (1), 143-173

Ungerer, J.A., Dolby, R., Waters, B., Barnett, B., Kelk, N., & Lewin, V. (1990). The early development of empathy: Self-regulation and individual differences in the first year. Motivation and Emotion, 14 (2),93-106.

Wilkinson, G.S. (1984). Reciprocal food sharing in the vampire bat. Nature 308: 181-184.

Williams, G.C. (1966). Adaptation and natural selection. Princetion, NJ: Princeton University Press.

Wilson, E.O. (1975). Sociobiology: The new synthesis. Cambridge, MA: Harvard University Press.

Wrangham, R., & Peterson, D. (1996). Demonic males: Apes and the origins of human violence. Boston: Houghton-Mifflin.

Zahn-Waxler, C., & Radke-Yarrow, M. (1990). The origins of empathic concern. Motivation and Emotion, 14 (2), 107-130.