A shorter version of this article can be found in: D. Huron (2001), "Is music an evolutionary adaptation?" Annals of the New York Academy of Sciences, Vol. 930, pp. 43-61.
Addressing the question of music's origins has a long -- though not particularly distinguished -- history. Many cultures have provided colorful stories describing how humans acquired the capacity for music. A few bold or wreckless scholars have ventured to offer biological, psychological, social, cultural or religious accounts of possible origins. Most scholars have wisely steered clear of the issue of music's origins, since clearly the enterprise is patently speculative. At its worst, proposals concerning musical origins are fiction masquerading as scholarship. However, I think there remains some merit in contemplating the question of how music-making got started. Reflecting on such questions can be a potentially informative and perhaps even illuminating exercise. In this lecture, I propose to offer a social account of music's origins that is explicitly linked to one of the most successful theories yet devised: the theory of evolution by natural selection.
Richard Dawkins (1995) reminds us of the importance of natural selection in the following passage:
"All organisms that have ever lived -- every animal and plant, all bacteria and all fungi, every creeping thing ... can look back at their ancestors and make the following proud claim: Not a single one of our ancestors died in infancy. They all reached adulthood, and every single one [successfully reproduced.] Not a single one of our ancestors was felled by an enemy, or by a virus, or by a misjudged footstep on a cliff edge, before bringing at least one child into the world. Thousands of our ancestors' contemporaries failed in all these respects, but not a single solitary one of our ancestors failed in any of them. These statements are blindingly obvious, yet from them much follows: much that is curious and unexpected, much that explains and much that astonishes." (1995; p.2)
The theory of evolution is possibly the most powerful theory yet conceived. It has survived the strongest of challenges, such as the challenge of accounting for altruistic behavior. With the work of Hamilton, Trivers, Wilson, and others, and the extraordinary accomplishments of molecular geneticists, the theory of evolution has gone from strength to strength. If Darwin were alive today, he would be very impressed by just how many phenomena appear to be accounted for by his theoretical legacy.
Evolution is often thought of in purely physiological rather than psychological terms. It is not simply that evolution has shaped immune systems, digestive tracts, and knee caps. Evolution has also shaped our attitudes, dispositions, emotions, perceptions, and cognitive functions. Some of our deepest convictions can be traced to plausible evolutionary origins: we love life, we fear death, and we nurture our children because any group that did not have these dispositions would be at a competitive disadvantage.
In addition to psychological dispositions and attitudes, our cognitive and perceptual capacities are also the products of evolution. Cognitive and perceptual capacities are shaped by and adapted to the world. Clearly, our perceptions of the world are not precise characterizations of how the world really is. But neither are our perceptions arbitrary constructions, for if so, we would soon be dead. In the case of sound, our ways of perceiving and apprehending are first of all conditioned by how sounds behave in the physical world, and by what information sounds encode that might be of value to human survival and procreation.
Similarly, our emotional lives are shaped by evolution. Research by Randolf Nesse has shown that even sadness can serve an essential evolutionary purpose; feeling bad may not be so bad after all. Like pain, feeling bad may be unpleasant, but it may also be biologically useful.
The theory of evolution by natural selection is a distal theory rather than a medial or proximal theory. It is not a theory that explains specific behaviors, such as why you chose to cook ravioli for dinner last night, why you parked in a particular parking spot this morning, or why you decided to learn to play the viola. Evolution proceeds by selecting traits that are adaptive to an organism's environment. For example, evolution did not "originate" or "create" the phenomenon of altruism. Instead, given a certain environment, natural selection favored individuals who exhibited altruistic traits. Evolution does not dictate our behavior: it selects which behaviors are likely to be passed on to subsequent generations -- and it selects only those behaviors that have a genetic basis.
Because so many human behaviors are clearly linked to survival, we might entertain the question of whether musical behaviors also confer some sort of advantage that enhances human survival and procreation. This is a difficult, contentious, and unresolved question. In this lecture, I will not attempt to somehow prove that music is adaptive; rather, my goal here is to convince you that this is a worthwhile question and deserves further thought and discussion.
Many knowledgeable people have concluded that music has no survival value. Indeed, a number of esthetic philosophers have argued that an essential, defining characteristic of the arts is that they serve no practical function. Accordingly, any music that is created for biological (or economic) reasons cannot be considered art. Even among evolutionary psychologists, it has been common to suppose that music is not adaptive. In How the Mind Works, for example, the noted psycholinguist and evolutionary psychologist Steven Pinker has argued that music is a good example of a common human phenomenon that is likely not an evolutionary adaptation.
I think that the evidence one way or another is not particularly convincing. Like others, I am not at all convinced that music is an evolutionary adaptation. However, I think we should investigate matters further before we dismiss the notion out of hand.
Before addressing the question of possible evolutionary origins for music, it is useful to consider some of the dangers associated with forming an evolutionary argument. Not all of the pertinent dangers can be noted here, but let me at least identify six of the more important ones.
Given these possible dangers, why bother attempting to formulate an evolutionary theory of music? Isn't it premature? First, as noted above, my goal here is not to convince you that music is adaptive; my goal is only to convince you that this is a worthwhile question. Understanding the possible origins of music might help inform us about some of the reasons we tend to respond in certain ways. Second, in the spirit of Popper, I will aim to tell an evolutionary story that is able to generate testable hypotheses. Like other evolutionary accounts, my theory will draw on existing knowledge, and so be post hoc in character. As long as this account remains post hoc, Gould and Lewontin's criticisms raise justified and paramount difficulties. However, it is my hope that the theory can be developed to the point where testable hypotheses might be derived.
Before entertaining some possible evolutionary views of music's origins, let us first consider two pertinent complicating points of views. One view is that music is a form of non-adaptive pleasure seeking. A second view is that music is an evolutionary vestige.
Most pleasurable activities, such as eating and sex, have clear links to survival. Such activities ultimately stimulate brain mechanisms that are specifically evolved to reward and encourage adaptive behaviors. Note that once brain mechanisms are in place that permit the experience of pleasure, it may be possible to stimulate those mechanisms in ways that don't confer a survival advantage. We can call these behaviors non-adaptive pleasure-seeking (NAPS). An example of NAPS behavior is found in the human taste for sugars and fats. In pre-modern times, sugars and fats were rare in human diets, but highly nutritious in the amounts available. There are good reasons why human tastes would evolve to reward the ingestion of foods with high fat or sugar content. However, centuries of human ingenuity have succeeded in generating a modern diet that contains unnaturally high levels of fats and sugars -- levels so high as to cause health problems such as diabetes and heart disease. Although such tastes originally conferred an increased chance of survival, in the modern environment, these behaviors have become less adaptive.
Another example of NAPS behavior is found in drug use such as heroin or cocaine. These drugs can directly activate the brain's pleasure centers, simply by injecting or imbibing a substance. Although the channel for pleasure exists for good evolutionary reasons, it may be possible to exploit the channel without any concomitant survival-enhancing result.
As in the case of drugs, it is possible that musical behaviors are forms of non-adaptive pleasure-seeking. That is, music itself may not enhance human survival; music may merely exploit one or more existing pleasure channels that evolved to reinforce some other adaptive behavior(s). We might call this view the "NAPS Theory of Music."
One way to determine whether some pleasure-seeking behavior is adaptive or non-adaptive is to consider how long the behavior has been around. In the long span of evolutionary history, non-adaptive pleasure-seeking behaviors tend to be short-lived. For example, heroin users tend to neglect their health and are known to have high mortality rates. Furthermore, heroin users tend to neglect their offspring -- they make poor parents. Poor health and neglect of offspring are infallible ways of reducing the probability that one's genes will be present in a future gene pool. After many generations, natural selection will tend to mitigate against heroin use. Those individuals who are not disposed (for whatever reason) to use heroin, are much more likely to procreate and so pass along their aversion to the use of such drugs, provided that the aversive behavior is somehow linked to a gene or genes.
The use of alcohol already suggests how NAPS behaviors can transform a gene pool. Although no gene has yet been identified, either for alcohol susceptibility or for alcohol tolerance, the responses of different human populations to alcohol show a suggestive pattern. Large quantities of alcohol became possible only with the advent of agriculture. European and Asian descendents of early agrarian cultures (such as originated in Mesopotamia) manage to deal with alcohol better than descendents of traditional hunter-gatherer societies, such as indigenous peoples in the Americas and in the arctic regions of Europe. Of course there are certain to be non-genetic factors influencing alcohol tolerance and abuse. However, alcohol researchers suspect that genetic factors are at work. Those people who have descended from traditional agricultural societies have a clear statistical advantage in dealing with the non-adaptive consequences of alcohol, and this would be expected if alcohol had been prevalent in these societies for thousands of years.
If music itself has no survival value (and merely exploits an existing pleasure channel) then any disposition towards musical behaviors would tend to worsen one's survival. Spending inordinate amounts of resources (such as time and money) on music might be expected to place music-lovers at an evolutionary disadvantage. In other words, if the NAPS Theory of Music is true, then we might predict that music appreciation would be correlated with marginal existence: as in the case of alcohol, people on "skid row" might be expected to be disproportionately music enthusiasts.
If music is non-adaptive, then the likelihood is that music is a modern invention; otherwise music-lovers would have become extinct some time ago. As we will see, the archaeological evidence indicates that music is very old -- much older than agriculture -- and this great antiquity is inconsistent with music originating as a non-adaptive pleasure-seeking behavior. In short, there is little evidence that musical behaviors have been selected against. All of this suggests that there is little support for the NAPS Theory of Music.
Another view might be that, while music at one time did indeed confer some survival value, it is now merely vestigial. Like the human appendix, at one time this "organ" may have contributed directly to human survival, but now it is largely irrelevant -- an evolutionary leftover. If this view is true, then we would have to ask `What advantage did music once confer?' and `How have things changed so that music is no longer adaptive?'
The adaptive value of some function is often evident in the individual survival costs arising from that function. For example, the larynx of newborn infants is anatomically arranged so that breathing and swallowing can happen at the same time. When the larynx enlarges, our physiological capacity for speech is purchased at the price of the danger of choking. In fact, one measure of the evolutionary advantage of speech is the mortality rates due to choking.
Similarly, an estimate of the evolutionary advantage conferred by music is to measure the amount of time people spend in musical behaviors. In the Atlas mountains of Morocco, full-time Jujuka mountain musicians are supported by the local villagers. That is, there is an entire caste of people whose principal productive activity is music-making. A ready index of the importance of music in such a society may be the ratio of the number of musicians to the number of farmers and herders.
Now let's consider some possible positive answers concerning the evolutionary advantage of music.
Before continuing, we should take a moment to discuss a common, though questionable theory of music's origins. Charles Darwin identified a form of natural selection known as sexual selection. The classic example of sexual selection is the peacock's tail. The function of the peacock's tail is not to promote the survival of the peacock; rather, the function is to promote the survival of the peacock's genes. Sexual selection arises once a particular genetic preference is established by the opposite sex -- in this case, the preference of the peahen for flashy tails. Even if one peahen is not particularly impressed by Las Vegas-style tails, it remains to the female's benefit to mate with the most colorful male if her offspring are more likely to be desired by other females who are fond of colorful tails.
Darwin himself suggested that music might have arisen due to sex selection in mating calls. Like the peacock's tail, the preferences of hominid women could create an escalating competition for ever more elaborate and beautiful melodies. The members of the all-male Vienna Philharmonic notwithstanding, there is nothing to indicate that one sex is more musical than the other, and so there is no evidence of the dimorphism symptomatic of sexual selection. Women may be impressed by men who serenade them outside their balcony windows, but it is questionable whether this says anything about evolutionary origins. After all, unlike female songbirds, female humans are perfectly capable of serenading men.
For the same reason, there is little to support the view that human music-making arose in a manner analogous to the songs of songbirds. In songbird species, only the male bird sings. That is, there is high sexual dimorphism for singing. Once again, in humans, there is no comparable sexual dimorphism.
In presenting a case for the evolutionary origins of music, we can consider five types of evidence:
Genetic evidence. The best evidence of an evolutionary origin would be the identification of genes whose expression leads to the behavior in question. Unfortunately, it is rare for scientists to be able to link particular behaviors to specific genes. Although behavior-linked genes have been discovered in other animals (such as fruit flies), no behavior-linked gene has yet been conclusively established in humans. As in so many other areas, music has attracted a kind of folklore related to heritability. In some cultures, it is common for people to assume or believe that musical talent is partly inherited. More recently, work at the University of California, San Francisco by Baharloo et al. (1998) appears to suggest a genetic component for absolute pitch.
Biochemical evidence. Since genes are expressed in the form of proteins, we would expect to be able to identify proteins that influence musical behaviors. If we cannot find such proteins, then there is little likelihood that music has a genetic basis.
Neurological evidence. The existence of specialized brain structures is neither a sufficient nor a necessary condition for music to be an evolutionary adaptation. Nevertheless, if stable anatomical brain structures exist for music, then this is consistent with music arising from innate development rather than due solely to a generalized learning.
Ethological evidence. Are musical behaviors consistent with survival and the propagation of genes? In order for music to be an evolutionary adaptation, music-related behaviors must somehow increase the likelihood that the musical person's genes will be propagated.
Archaeological evidence. Since complex evolutionary adaptations arise over many thousands of generations, we must ask how widespread music is in biological history? If music originated in the past few thousand years, then it is highly unlikely to be an evolutionary adaptation. Evolution doesn't work that fast.
As noted, there is currently no evidence that links music to any gene. Let's consider the other areas of evidence in more detail.
In 1980, Avram Goldstein published the results of an experiment where the effect of naloxone on musical pleasure was measured. Naloxone is an opiate receptor antagonist -- that is, it is a molecule that attaches to opiate receptors in the brain without activating them. Naloxone is a sort of prophylactic that covers the receptor and prevents it from being activated. In his experiment, Goldstein found that volunteer listeners who had been injected with naloxone reported significantly less musical pleasure than those who had received a saline injection.
Goldstein's experiment does not tell us how music evokes pleasure. But his experiment implies that, however music evokes pleasure, it ultimately causes the release of endorphins that stimulate the brain's opiate receptors. In short, musical pleasure appears to engage the same physiological mechanisms that are used by a wide variety of other pleasure-inducing behaviors.
Most activities that result in pleasure are somehow related to enhanced survival. As we already noted, there are some pleasure-inducing activities that do not appear to have any evolutionary or adaptive value. Is music adaptive pleasure-seeking or is it non-adaptive pleasure-seeking? That is, is music somehow akin to eating (an activity that increases survival)? Or is music like heroin use (an activity of no apparent survival value that simply exploits the biological pleasure-creating machinery designed for other purposes)? I would like to propose that this is one of the most fundamental questions that can be asked about music.
Further evidence related to biochemical concomitants of music was provided by Fukui (1996). Fukui measured the effect of music listening on testosterone production. Evidence supporting a social role can also be found in studies of physiological responses to music. Testosterone is an androgen -- a hormone normally associated with men, but also produced in smaller quantities in women. Testosterone levels are strongly correlated with aggression: high levels of testosterone tend to facilitate aggressive behaviors, similar to the "roid rage" commonly experienced by athletes who use commercial steroids. In addition, testosterone is thought to mediate libido. Low levels of testosterone are associated with lower sexual arousal. (Nelson, 1994; Sherwin, 1988; Wallen & Lovejoy, 1993)
Fukui (1996) carried out a study in which he measured testosterone levels from saliva samples collected from undergraduate-aged participants while they listened to their favorite music. Compared with a control group that listened to no music, the testosterone levels dropped significantly. Moreover, Fukui found no sex-related differences: testosterone levels dropped by a similar proportion in both male and female listeners.
Both Fukui's experiment and Goldstein's experiment provide evidence that music modulates the production of specific proteins in the body. This doesn't prove much, but it does demonstrate that music is not a disembodied abstraction; music engages human physiology at one of the most basic levels. We will have more to say about Fukui's and Goldstein's experiments later.
Let's now move on to consider some of the archaeological facts. The archaeological record shows a continuous record of music-making in human settlements. Wherever you find evidence of human settlement, you find evidence of musical activities.
In 1995, paleontologist Ivan Turk discovered a bone flute while excavating an ancient burial mound in Divje Babe, Slovenia (Anon., 1997). Using electron spin dating, the age of this flute has been determined to lie between 43,000 and 82,000 years old. If the instrument had been made out of wood, it would long ago have disintegrated. So we are fortunate that someone took the time to fashion this particular instrument from the femur of the now extinct European bear.
Of course, finding this flute doesn't mean we've found the earliest musical instrument; this is just the earliest found instrument. It's logical to presume that wooden flutes were fashioned earlier than bone flutes. So it's not inconceivable that wooden flutes existed, say, 100,000 years or more ago.
As musical instruments go, flutes are rather complicated devices. If we look at contemporary hunter-gatherer societies, the most common instruments are rattles, shakers, and drums. For example, prior to the arrival of Europeans, by far the most common instruments in native American cultures were rattles and drums. The same pattern of preferred instruments is evident in African and in Polynesian cultures. If we assume that rattles and drums typically pre-dated the use of flutes, then the ancient music-makers of Slovenia might well have been creating instrumental music somewhat earlier than 100,000 years ago.
But what sort of music-making might have existed prior to the fashioning of musical instruments? It's not unreasonable to assume that singing preceded the making of musical instruments by some length of time. If we suppose that singing predated instrument making by 50% of the intervening time, then music-making might have existed 150,000 years ago -- roughly twice the age of the older estimate for the Divje Babe flute. Even this figure might be a conservative estimate, and the actual origin of music might be twice as old, say around 250,000 years ago.
On the other hand, the Divje Babe flute might truly be an early specimen, and singing might have developed about the same time. Using the most recent estimate for the Divje Babe flute would therefore place the origins of music-making about 50,000 years ago.
In summary, the archaeological record implies that music-making likely originated between 50,000 years ago and a quarter of a million years ago. Although Wurlitzer organs, American Bandstand, and MTV are relatively recent phenomena, music-making in general is really quite old.
The evidence pointing to the great antiquity of music satisfies the most basic requirement for any evolutionary argument. Evolution proceeds at a very slow pace so nearly all adaptations must be extremely old. Music-making satisfies this condition, though we must be careful not to assume that the music of the Pleistocene period bears much resemblance to Brahms or Twisted Sister.
Turning to contemporary anthropology, we can ask, "What does the plethora of existing human cultures tell us about music?" Without taking time to review the evidence, there is one overwhelming conclusion from the modern anthropological record. There is no human culture known in modern times that did not, or does not, engage in recognizably musical activities.
Not only is music-making very old, it is ubiquitous; it is found wherever humans are found. Moreover, I neglected earlier to mention one important fact about the bone flute at Divje Babe: the flute was found in a Neanderthal burial site. The Divje Babe flute isn't even a human artifact. In short, it may be the case that music-making is not just ubiquitous among homo sapiens; music-making may possibly be characteristic of the entire genus homo.
The evidence pointing to the ubiquity of music satisfies another important basic requirement for any evolutionary argument. Relatively few adaptations are not found throughout the entire population of the affected species. For example, if eyelashes confer an evolutionary advantage, then just about everyone should have eyelashes. There are some exceptions to this principle, some of which are very important. For example, humans divide into female and male versions, so there are some genes that are not shared by everyone. Another more subtle example is the gene that codes for sickle-cell -- a gene that protects against malaria, but can also cause anemia.
Ethology is the study of animal behavior; this includes the study of human behavior. When studying a particular animal, ethologists often begin by making an inventory of observed behaviors. What does the animal do, and how often does it do it? Activities that require a great deal of time and large expenditures of energy are understandably considered important. Ethologists assume that behaviors are likely to be optimized. Even behaviors that seem unimportant (such as infant play, or sleeping) often have a serious or critical purpose.
Primates, for example, spend an extraordinary amount of time grooming each other. Ethologists feel obliged to formulate theories that account for the various proportions of resources dedicated by an animal to different activities.
Let's apply the ethological approach to the behaviors we call musical. For the purposes of illustration, let's consider two case descriptions. The first case is that of the Mekranoti Indians of the Brazilian Amazon. And the second case is that of contemporary U.S. society.
The Mekranoti Indians are hunter-gatherers who live in the Amazon rainforest of Brazil. In Mekranoti culture, singing plays a prominent role in daily life. For several months of the year, every morning and evening the women lay banana leaves on the ground where they sit and sing for between one and two hours. The men sing every night starting typically around 4:30 in the morning, but sometimes as early as 1:30 AM. The men sing for roughly two hours each night, and often they will also sing for a half hour or so before sunset.
When singing, the Mekranoti men hold their arms in a sort of cradling position and swing their arms vigorously. The men endeavor to sing in their deepest bass voices, and heavily accent the first beats of a pervasive quadruple meter with glottal stops that make their stomachs convulse in rhythm. Anthropologist Dennis Werner (1984) describes their singing as a "masculine roar." When gathering in the middle of the night, the men are obviously sleepy, and some men will linger in their lean-tos well after the singing has started. These malingerers are often taunted with shouted insults.
Werner reports that "Hounding the men still in their lean-tos [is] one of the favorite diversions of the singers. `Get out of bed! The Kreen Akrore Indians have already attacked and you're still sleeping,' they [shout] as loudly as they [can]. ... Sometimes the harassment [is] personal as the singers [yell] out insults at specific men who rarely [show] up." (pp.245-247)
What is extraordinary about the Mekranoti singing is the amount of time involved -- roughly two hours per day. (Remember, this is a subsistence hunter-gatherer society.) For the evolutionary ethologist, the important question arising from the Mekranoti Indians is why music-making would attract so much of the tribe's resources. We'll return to this question later.
By way of comparison, consider now the prevalence of music in a modern industrialized society like the United States. For the ethologist looking at modern human behaviors, a crude though ready index of the amount of resources we dedicate to a particular activity can be found by measuring economic activity.
There is a widespread misconception that the foremost export sector in the U.S. economy is "high technology." In fact, the preeminent export sector in the U.S. economy is entertainment. Of the various component areas -- films, sports, television, toys, and games -- it is music that ranks foremost.
How big is the music industry? The music industry is bigger than the pharmaceutical industry. People spend more money on music than on prescription drugs. We purchase recordings, go to concerts, buy sheet music, take our children to music lessons, listen to commercial radio, watch film accompanied by music, and encounter Muzak in the local shopping mall. The most active `concert venues' in the world are freeways: a major preoccupation for millions of drivers is listening to music.
Of course financial measures are crude indicators of behavioral significance. The ethological point is simple. In both a hunter-gatherer society and a modern industrial society, we find humans dedicating a notable proportion of resources to music-making and listening. Music may not be more important than sex; but it is arguably more expensive, and it is certainly more time-consuming.
In order to put these behaviors in perspective, suppose you were a Martian anthropologist visiting earth. There are many aspects of human behavior that would have recognizable value. You would see people engaged in growing and preparing food, in raising and educating children, people involved in transportation, health, and governance. But even if Martian anthropologists had ears, I suspect they'd be stumped by music.
If you're still not convinced that music attracts a peculiarly excessive proportion of human resources, consider another comparison. Think of how important food is to human well-being; of how tasty and enjoyable food is and can be. Now how many universities have departments of cuisine or nutrition? Or departments of food sciences, or even departments of home economics? Now consider how many university have departments of music. Why would music figure more prominently than food? To a visiting tourist from Mars, music sticks out; it is a remarkable and bizarre activity that earthlings do.
Of course, we must be careful in drawing any conclusions about adaptations based on observations of modern behaviors. If music-making is an adaptive behavior, then it must have arisen long ago in the environment of evolutionary adaptedness -- namely the Pleistocene period when the vast majority of human evolution occurred.
Just because an animal spends a lot of time on certain activities doesn't mean that the activity represents an evolutionary adaptation. Ethologists must connect the behavior to an explicit evolutionary account. That is, there must exist a plausible explanation of how the behavior would be adaptive.
Before considering such a theory for music, let's examine a non-musical example -- an example that has a richer theoretical literature about its origins. Specifically, let's consider some of the evolutionary arguments that have been advanced to account for the origins of language.
As in the case of music, views concerning the origins of language are necessarily speculative. Nevertheless, we can learn a great deal by considering some of the theories that have been advanced concerning its origin. Until recently, the principal view of language was that it facilitated complex collaborative activities such as coordinating actions during hunting. This account seems unlikely, first, because talking is bad idea when tracking prey, and secondly, because men as a group display inferior language skills compared with women.
A number of anthropological psychologists have suggested that language (and even music) evolved as surrogates for social bonding.
The most empirically grounded of the recent theories of language origins is what might be called the "grooming and gossip hypothesis." Its principal advocate is Robin Dunbar (1997). The theory proposes the following logic.
Animals often live in groups for mutual protection against predators. In general, larger groups are more effective in detecting and warding-off predators than smaller groups. But there are costs associated with maintaining a large group. One cost is that feeding must be much more intensive in a given area and so a larger group must travel greater distances in search of food. A second cost is that as group size increases, threats are more likely to arise from internal conflict within the group rather than from external predators. That is, there is a point where group size effectively minimizes predation, but at the cost of threats from members of the group itself. Nowhere is this more evident than in primates. As a consequence of internal threats, animals within the group begin to form alliances with one another. These alliances reduce the likelihood of conflict due to the threat of group retaliation.
In primates, the principal means by which alliances are formed and bonds maintained is through grooming. Grooming accounts for between 10% and 20% of an individual's daytime activities.
There is good evidence to suggest that the principal purpose of grooming is to form alliances between individuals. First, grooming partners are much more likely to come to the defense of one-another when threatened by another member of the group. Even more important evidence comes from relating the amount of time spent grooming to the size of the group. Different primate species have different typical group sizes. Gorillas, macaques, chimpanzees, bonobos, and so on each tend to form groups that have different average sizes. Primatologists have measured the different amounts of time each species engages in grooming.
A major discovery has been that there is a consistent relationship between group size and the amount of time spent grooming. As the group size increases, the average grooming time also increases. This is an important finding: there is no reason to suppose that animals in larger groups tend to get more dirty than animals in smaller groups, so the increase in grooming is unlikely to be related to cleanliness.
Primatologists widely agree that the increase in grooming time for larger groups arises from the need to form more extensive networks of alliances. In a large group, an individual fares better by having a wider circle of friends, and the way to build primate friendships is through mutual grooming.
Of course alliances can be broken or betrayed. An animal who has been attacked by another animal may well expect a grooming partner to come to their defense. But there are always those individuals who may benefit from your willingness to defend them, but who will not reciprocate by coming to your defense. This is the so-called "free-rider" problem: cunning animals might well exploit those who are foolish enough to groom them.
The free-rider problem means that individual primates ought to be sensitive to the possibility of defection by a grooming partner. Individuals will look for clues about the reliability of those they consider their friends. Indeed, primatologists have described circumstances where a grooming alliance is abandoned by an individual who has witnessed the failure of their partner to come to the defense of a third grooming partner. Untrustworthy animals are not popular grooming partners, and a reputation for reciprocal altruism is important.
In this respect, humans are no different from other primates. As Cosmides and Tooby (1992) have famously shown, human reasoning follows patterns, not of abstract logic, but are optimized for social contracts. Humans have deep-seated notions of justice that follow from betrayals of social alliances: if you collaborate, you deserve to be helped; if you defect, you ought to pay the price.
In the case of humans, the common `group size' has been estimated at roughly 150 people. This is approximately the size of most rural villages in the world. This means that human groups are especially large when compared with other primates. As Dunbar (1997) has pointed out, "If modern humans tried to use grooming as the sole means of reinforcing their social bonds, as other primates do, then the equation for monkeys and apes suggests we would have to devote around 40 per cent of our day in mutual mauling." (p.78).
Dunbar has suggested that language evolved as an alternative to physical grooming. In effect, physical grooming was replaced by "vocal grooming" whose purpose remains the formation and maintenance of friendships or alliances. Such "vocal grooming" has two distinct advantages over physical grooming. First, we can talk to several people simultaneously. This increases the number of people we can bond with at the same time. Second, we can exchange information about people who are physically absent -- that is, we can gossip. Unlike other primates, this means that we can learn about the behavior of others without being limited to direct observation.
Incidentally, Dunbar's theory does not preclude other uses for language. Clearly, language is advantageous in a number of ways. Dunbar's theory simply attempts to account for how language got started in the first place -- it is not necessarily a theory of how language might be adaptive for modern humans.
Nevertheless, Dunbar and his colleagues have conducted a number of studies that illustrate the continuing human penchant for gossip. Even in formal business interactions, only roughly 1/4 of the time is spent negotiating or discussing technical details. The majority of time in business interactions is spent relaying personal information, discussing colleagues, and gossiping about the intentions, betrayals, supports, and reliability of other people -- or establishing our own credibility and worthiness of character.
When did language arise in humans? Estimates vary from as recently as 50,000 years ago to 500,000 years ago. None of the evidence is direct. Archaeologists point to the so-called Upper Paleolithic Revolution, a period when stone artifacts and tools show a marked improvement in quality and range. At this point (50,000 years ago) tools include buttons, needles, awls, and other refined inventions.
Direct evidence for language (such as writing) is only found within the last 10,000 years. In fact, the archaeological evidence for the antiquity of music is stronger than the archaeological evidence for the antiquity of language -- although that doesn't mean that music necessarily preceded the emergence of language.
Note that even language has significant limitations for multiple concurrent social interaction. Dunbar (1997) has noted that "there appears to be a decisive upper limit of about four on the number of individuals who can be involved in a conversation." (p.121). When a fifth or sixth person joins a conversation there is a marked tendency for the group to subdivide into two or more concurrent conversations. It is only in hierarchical situations (such as in a formal lecture) where a single conversation can be maintained in a larger group.
All of this suggests that language is most useful in close interpersonal interactions, such as grooming, gossiping, courting, and conspiring. Note, however, that there are other activities that are of value to members of a social group that involve the entire group (or at least large segments) rather than groups of twos or threes. Chief among these group activities is defense. When under threat, uniform group action is indeed a mighty force, much more powerful than smaller groups of twos and threes.
At this point, we might speculate how music might fit into this account. Let's assume, for the moment, that the hypothesis that language evolved as a surrogate for physical grooming is true, and that language thereby allowed humans to live in larger groups with their attendant complex social relations. We could certainly conceive of a similar function for music. In some ways, music provides several advantages over language. Singing is much louder than speaking, so singing may facilitate group interactions involving more than the four individuals posited as the upper limit for conversation. Although music may not be as effective as language in informing us of the deceptions of others, it does fit within the rubric of surrogate grooming. Recall that in primates, the function of grooming is to provide social bonding opportunities -- not ways of learning about the machinations of individuals who are absent. Therefore, in some ways, music provides a better parallel to physical grooming than is the case for language.
This view of the possible origins for music was essentially proposed by Juan Roederer (1984):
"... the role of music in superstitious or sexual rites, religion, ideological proselytism, and military arousal clearly demonstrates the value of music as a means of establishing behavioral coherency in masses of people. In the distant past this could indeed have had an important survival value, as an increasingly complex human environment demanded coherent, collective actions on the part of groups of human society." (p.356)In light of later work by primatologists such as Dunbar, there appears to be merit in Roederer's hypothesis. Music might have originated as an adaptation for social bonding -- more particularly, as a way of synchronizing the mood of many individuals in a larger group. That is, music helps to prepare the group to act in unison.
Perhaps a helpful image is to imagine the cackling of geese prior to them taking off. How is it that individual geese manage to synchronize their actions so that the entire flock takes flight more-or-less simultaneously? For anyone who has watched geese take-off, there is a clear increase in the volume of cackling ... more and more geese start honking. The general hubbub of honking geese is apt to raise the arousal levels of all geese in the vicinity. This increased arousal (which includes increased heart-rate) would prepare the geese for a significant collective expenditure of energy.
It is this theory of music and social bonding which I believe holds the greatest promise as a plausible evolutionary origin for music. For the remainder of this lecture, I would like to review further phenomena that provide support for this hypothesis. The evidence is going to come from the following five sources:
Consider two mental disorders: Williams Syndrome and Asperger-type Autism. The principal feature of Williams syndrome is mental retardation. Williams syndrome is somewhat unique in that suffers display three additional characteristics. One characteristic is high verbal abilities. Individuals suffering from Williams syndrome take a great interest in words. Their speech is fluent, and peppered with a remarkably sophisticated vocabulary. In fact, when first encountering someone with Williams syndrome the language fluency tends to mask the mental handicap.
In addition to high verbal abilities, Williams syndrome individuals also exhibit high sociability. They are gregarious and sociable. Coupled with the high verbal abilities, this makes Williams syndrome children a delight to work with. Finally, Williams syndrome children exhibit high musicality.
Daniel Levitin and Ursula Bellugi (1997) have described the musical activities of Williams syndrome children at a summer camp in New York state. The children are remarkable. The entire camp is alive with music, string quartets, trios, woodwind groups, and so on. They are `crazy' about music, and relish the social environment of other children with the same social, language, and musical enthusiasms.
Now consider the case of Asperger-type Autism. Autism is characterized by a strong aversion to social interaction. Although most autism is associated with reduced mental functioning, mental retardation is not always evident. There are autistic individuals with normal and above average intelligence as well. Autism is related to an emotional deficit -- notably the failure to develop the so-called secondary or social emotions -- including shame, pride, guilt, love and empathy. For normal children, these secondary emotions typically appear by about the age of four.
Temple Grandin is a high functioning Asperger-type autistic who has become well-known through her writings about her own condition. Concerning love, Grandin talks about her confusion in high school when reading Shakespeare's Romeo and Juliet. I never figured-out what it was all about, says Grandin. In a trip through the Rocky Mountains with Oliver Sacks, Grandin remarked "The mountains are pretty, ... but they don't give me a special feeling, the feeling you seem to enjoy." "You get such joy out of the sunset," she said. "I wish I did, too. I know it's beautiful, but I don't `get' it." (Sacks, p.124). Grandin's experience of music is similar. Although Grandin has perfect pitch, and what she describes as a tenacious and accurate auditory memory, she finds music leaves her cold. She finds the sounds "pretty," but in general, she just doesn't "get" it (p.122). All the fuss about music leaves her mystified.
Grandin's own explanation is that not all of the `emotional circuits' are connected. Sack's interprets the phenomenon as follows: "An autistic person can have violent passions, intensely charged fixations and fascinations, or, like Temple [Grandin], an almost overwhelming tenderness and concern in certain areas. In autism, it is not affect in general that is faulty but affect in relation to complex human experiences, social ones predominantly, but perhaps allied ones -- esthetic, poetic, symbolic, etc. No one, indeed, brings this out more clearly than Temple herself. ... She feels that there is something mechanical about her mind, and she often compares it to a computer ... She feels that there are usually genetic determinants in autism; she suspects that her own father, who was remote, pedantic, and socially inept, had Asperger's -- or, at least, autistic traits -- and that such traits occur with significant frequency in the parents and grandparents of autistic children." (p.123).
The contrast between Asperger-type Autism and Williams Syndrome is striking. On the one hand we have a group of people whose symptoms include high sociability linked with high musicality. On the other hand we have a group of people whose symptoms include low sociability OFTEN linked with low musicality. Together, these mental conditions are consistent with a relationship between sociability and musicality -- and this link is the principal assumption of a group-oriented evolutionary account.
Suppose we asked the following question: What is the most successful piece of music in modern history? Of course the answer to this question depends on how we define success -- and this is far from clear, as esthetic philosophers have shown. Nevertheless, let's use a straightforward criterion: lets assume that the most successful musical work is the one which is most performed and most heard. Using this criterion, you might be surprised by the answer. The most successful musical work was composed by Mildred and Patti Hill in 1893, and revised in the 1930s (Fuld, 1995). The piece in question is, of course, Happy Birthday. Happy Birthday has been translated into innumerable languages and is performed on the order of a million times a day. It remained under copyright protection until the middle of the century. For many people, the singing of Happy Birthday is the only time they sing in public. For other people, the singing of Happy Birthday constitutes the only time they sing.
In some ways, Happy Birthday is the quintessential feminist work. Its composers remain unknown and uncelebrated; the work was created by the collaboration of two women rather than as an egotistic expression of one man. It is a thoroughly domestic work; Happy Birthday is performed in the kitchen or lunch room rather than in the concert hall. No other musical work has evoked so much spontaneous music-making. The work is domestic, amateur, and relationally oriented. Despite its extraordinary success, it remains undervalued as a musical creation.
Happy Birthday plays a role in our evolutionary story because I suspect that for the vast majority of human history, music-making was of this ilk. In Western culture, it is surely the camp songs sung by Girl Scouts or the songs sung by British soccer hooligans that come closest to what might be imagined in Pleistocene homo sapiens. In all of these cases, the music serves an obvious social role and is a critical moment in defining a sense of identity and common purpose.
In light of our evolutionary hypothesis, let's return and reconsider the singing of the Mekranoti Indians. Recall some of the characteristic features -- especially the singing done by the men: the men's singing is done late at night and in the early morning, and their singing is associated with a high degree of machismo. Like most native societies, the greatest danger facing the Mekranoti Indians is the possibility of being attacked by another human group. The best strategic time to attack is in the very early morning while people are asleep. Recall the insult shouted at men who continued to sleep in their lean-tos: "Get out of bed! The Kreen Akrore Indians have already attacked and you're still sleeping."
The implication is obvious. It appears that the nightly singing by the men constitutes a defensive vigil. The singing maintains arousal levels and keeps the men awake.
Of course music-making is also associated with stirring a war-party. North American Indians famously sang and danced prior to initiating an attack on another tribe. One might suppose that engaging in an activity that publicly announces a hostile intention would be counter-productive. War dances might possible warn an enemy of an impending attack. However, the music-making seems to serve a more important role: that of raising arousal and synchronizing individual moods to serve the larger goal of the group.
Apart from arousing individuals, music can also pacify. Recall Fukui's experiment showing that listening to music can reduce testosterone levels. Fukui himself was quick to point out the possible social and evolutionary significance of this finding. In human social groups, lower levels of testosterone are likely to result in less aggression, less conflict, less sexual confrontation or sexual competition, and consequently more group cohesiveness. Where men commonly suffer from testosterone "poisoning," music truly hath charms to soothe the savage breast.
A problem with Fukui's experiment is that he didn't manipulate the type of music heard by his listeners. Listeners simply listened to their favorite music. Depending on his sample of listeners, we might expect whole genres of music that were not represented. We might suppose, for example, that heavy metal, hard rock, or thrash music might well have increased testosterone levels rather than decreasing the levels. Further research is necessary to document the specific hormonal changes associated with different types of musical experiences. However, Fukui's work at least shows that music can have marked effects on hormone levels -- specifically, hormones that relate especially strongly to sociability.
An important question to ask is how precisely music might bring about social bonding. Neurophysiologist Walter Freeman (1995) has proposed a pertinent theory related to the hormone oxytocin.
Oxytocin is most commonly associated with the "let-down" response in new mothers -- that is, the response that enables the flow of breast milk following child-birth. The presence of oxytocin also has dramatic effects on the brain. For example, when a ewe gives birth to a lamb, the olfactory bulb in the ewe's brain is bathed in oxytocin. Following the birth of the new lamb, a ewe will imprint on the smell of the new lamb, but will subsequently fail to recognize the smell of her former offspring. The result is that the ewe will suckle only the new-born lamb.
Neurophysiological research has shown that oxytocin acts as a sort of "eraser" that wipes away previous memories and simultaneously facilitates the storage of new memories. When linked with significant life events, oxytocin is the cement that binds new memories. The amnesic properties of oxytocin are evident in all kinds of learning episodes. However, their strongest effects occur during major limbic activations such as those resulting from trauma or from ecstasy. Pavlov discovered this phenomenon when serious spring flooding affected his lab and nearly drowned his caged dogs. Following their rescue it was discovered that the dogs had to be re-trained from scratch (Pavlov, 1955).
In his book, Societies of Brains, Freeman chronicles a number of circumstances where oxytocin release occurs, and the effects of these releases on neural organization. As we have noted, oxytocin releases are associated with trauma and ecstasy. In addition to child-birth, oxytocin is released in males and females following sexual orgasm. Freeman also suggests that oxytocin is released during trance and while listening to music.
In many cases, the presence of oxytocin is correlated with human and animal bonding circumstances. For example, in the case of sexual orgasm, oxytocin may significantly facilitate pair-bonding in the same way that oxytocin following child-birth facilitates mother-child bonding. Freeman's suggestion that music causes oxytocin to be released has important repercussions for instances of peer-group bonding and social identity. If Freeman is correct, there would be good neurophysiological reasons for lovers to enjoy music while courting, for union members to sing while on the picket line, for religious groups to engage in collective music-making, for colleges to promote alma mater songs, and for warriors to sing and dance prior to fighting.
Thayer and his colleagues have carried out a number of studies concerning how people regulate their moods. One study attempted to determine what people do to try to get out of a bad mood. Of 29 categories of activities, the foremost activity was calling or talking to a friend. The second most frequently reported activity was trying to think positive thoughts -- to give oneself a sort of "pep talk." The third most frequently reported activity -- ahead of a wide variety of behaviors -- was listening to music. Forty-seven percent of respondents reported that they used music to temper or eliminate a bad mood.
Thayer et al. carried out a similar study to determine what people do to raise their alertness or energy level. Listening to music was reported by 41 percent of respondents, following activities such as sleeping, taking a shower, getting some fresh air, and drinking coffee. Finally, in a third study investigating what people do to reduce nervousness, tension or anxiety, listening to music ranked third at 53 percent, following after only calling or talking to someone, and trying to calm down by thinking about a situation.
There are two points to highlight from these studies. The first is that the foremost category of behavior for mood regulation is being with or conversing with a friend. That is to say, our first tendency is to seek mood regulation through social interaction. Moods are contagious, and we rely to some extent on each other to to modulate, reinforce or temper our moods. Although we know that moods are highly influenced by the individual's physiological state -- notably through food, exercise, rest, etc. -- behaviors such as eating, exercise and rest are less frequently used for mood regulation than music.
The second point to highlight is the obvious point that music appears to figure prominently as a method for mood regulation. Although in contemporary society music tends to be experienced in a personalized or individualized listening context, we already know that this context is historically unprecedented. Most music-making in hunter-gatherer societies occurs in a social or group context. Until the invention of the phonograph, the vast majority of music in Western culture was also experienced in social or group contexts. In short, music is not out-of-place in the list of socialized behaviors used for mood regulation.
By way of conclusion, first let me again reiterate that I don't think the evidence in support of music as an evolutionary adaptation is strong. The purpose of this lecture has been to show that there are no obvious or fatal impediments that rule-out a possible evolutionary origin.
We might summarize the basic evidence as follows:
The evidence we have for mood regulation and synchronization is suggestive:
As noted at the beginning of this essay, there is a long history of abuse of genetic claims serving ulterior and often nefarious motives. Even if we assume that musicality has some adaptive function, the repercussions for modern music-making and modern musical enjoyment are likely to be minimal.
Music is now deeply embedded in a cultural/historical context where human musical memories span centuries and the fashion cycle is a significant engine of change. Music is now part of a Lamarkian system where acquired characteristics are transmitted in Dawkinsean "meme-pool" rather than in Mendelian "gene-pool". Like language, the details of musical culture and tastes are largely a product of enculturation.
Nevertheless, it remains worthwhile to attempt to understand where music comes from, and why it has achieved such a ubiquitous presence in human lives. Evolutionary theorizing about music may well remain in the realm of `Just-So' stories. But there is always the possibility of a testable hypothesis emerging, and if so, we'll all wait with interest to see the results.
I would like to extend my thanks to Dr. Kristin Precoda for drawing my attention to the work of Werner regarding the Mekranoti Indians, and to Dr. David Wessel for drawing my attention to Freeman's work concerning oxytocin. This lecture was originally presented at the Department of Music, University of California, Santa Barbara, March 6th, 1998. A shortened version was presented at the Society for Music Perception and Cognition Conference, Evanston, Illinois. August 16, 1999. An edited version of this article can be found in: D. Huron (2001), "Is music an evolutionary adaptation?" Annals of the New York Academy of Sciences, Vol. 930, pp. 43-61.
Neanderthal notes: Did ancient humans play modern scales? Scientific American, Vol. 277, No. 3 (1997, Sept.) pp.28-30.
Aiello, L., & Dunbar, R.I.M. (1993).
Neocortex size, group size and the evolution of language. Current Anthropology, Vol. 34, pp. 184-193.
Baharloo, S., Johnston, P.A., Service, S.K., Gitschier, J. & Freimer, N.B. (1998).
Absolute pitch: An approach for identification of genetic and nongenetic components. American Journal of Human Genetics, Vol. 62, pp. 224-231.
Barkow, J., Cosmides, L. & Tooby, J. (Eds.) (1992).
The Adapted Mind: Evolutionary Psychology and the Generation of Culture. Oxford: Oxford University Press.
Baron-Cohen, S. (Ed.). (1997).
The Maladapted Mind: Classic Readings in Evolutionary Psychopathology. East Sussex, UK: Psychology Press.
Blair, R.J.R. (1995).
A cognitive developmental approach to morality: investigating the psychopath. Cognition: International Journal of Cognitive Science, Vol. 57, pp. 1-29.
Brown, D. (1991).
Human Universals. New York: McGraw-Hill.
Buss, D. (1994).
The Evolution of Desire: Strategies of Human Mating. New York: Basic Books.
Cacioppo, J.T., Petty, R.P. Losch, M.E., & Kim. H.S. (1986).
Electromyographic activity over facial muscle regions can differentiate the valence and intensity of affective reactions. Journal of Personality and Social Psychology, Vol. 50, pp. 260-268.
Darwin, Charles. (1872).
The Expression of Emotion in Man and Animals. London: Murray.
Dawkins, R. (1976).
The Selfish Gene. Oxford: Oxford University Press.
Dawkins, R. (1982).
The Extended Phenotype. San Francisco: W.H. Freeman.
Dawkins, R. (1986).
The Blind Watchmaker. New York: W.W. Norton.
Dawkins, R. (1995).
River Out of Eden. New York: Basic Books.
Dimberg, Ulf. (1987).
Facial reactions and autonomic activity to auditory stimuli with high and low intensity. Psychophysiology, Vol. 24, p. 586.
Dimberg, Ulf. (1989).
Perceived unpleasantness and facial reactions to auditory stimuli. Uppsala, Sweden: Uppsala Psychological Reports, No. 414.
Dissanayake, Ellen. (1988).
What Is Art For? Seattle: University of Washington Press.
Dobzhansky, Th. (1973).
Genetic Diversity and Human Equality. New York: Basic Books.
Dunbar, R. (1997).
Grooming, Gossip and the Evolution of Language. New York: Faber & Faber.
Durham, W. (1991).
Coevolution: Genes, Culture, and Human Diversity. Stanford, CA: Stanford University Press.
Eaves, L., Eysenck, H., & Martin, N. (1989).
Genes, Culture and Personality. London: Academic Press.
Ekman, P. (1973).
Cross-cultural studies of facial expressions. In P. Ekman (editor), Darwin and facial expression. New York: Academic Press, pp. 169-222.
Eysenck, H. (1997).
The future of psychology. In: R.L. Solso (ed.), Mind and Brain Sciences in the 21st Century. Cambridge, MA: MIT Press, pp.271-301.
Freeman, W. J. (1995).
Societies of Brains: A Study in the Neuroscience of Love and Hate. Hillsdale, NJ: Lawrence Erlbaum Associates.
Freeman, W. J. & Grajski, K. A. (1987)
Relation of olfactory EEG to behavior: Factor analysis. Behavioral Neuroscience, Vol. 101, pp. 766-777.
Fuld, J.J. (1995).
The Book of World-famous Music; Classical, Popular and Folk. New York: Dover, 4th edition.
Gaulin, S.J.C. & Robbins, C.J. (1991).
Trivers-Willard effect in contemporary North American society." American Journal of Physical Anthropology, Vol. 85, pp. 61-69.
Goldstein, A. (1980).
Thrills in response to music and other stimuli. Physiological Psychology, Vol. 8, No. 1, pp. 126-129.
Gould, S.J., & Lewontin, R.C. (1979).
The spandrels of San Marco and the Panglossian program: A critique of the adaptationist programme. Proceedings of the Royal Society of London, Vol. 250, pp. 281-288.
Hatfield, E., Cacioppo, J., & Rapson, R. (1994).
Emotional Contagion; Studies in Emotion and Social Interaction. Cambridge: Cambridge University Press.
Levitin, D.L., & Bellugi, U. (1997).
Musical abilities in individuals with Williams Syndrome. Paper presented at the 1997 Society for Music Perception and Cognition, Massachusetts Institute of Technology, Cambridge, Mass.
Lewontin, R.C. (1991).
Biology as Ideology; The Doctrine of DNA. Concord, Ontario: House of Anansi Press.
Miller, W.I. (1997).
The Anatomy of Disgust. Cambridge, MA: Harvard University Press.
What good is feeling bad? The Sciences.
Pavlov, I. P. (1955).
Selected Works. Moscow: Foreign Languages Publishing House. Translated by S. Belsky; edited by J. Gibbons.
Pert, C.B. (1997).
Molecules of Emotion. New York: Scribner.
Pinker, S. (1994).
The Language Instinct. New York: Morrow.
Popper, K. (1935/1959).
Logik der Forschung. Vienna, 1935. Translated as The Logic of Scientific Discovery. New York: Basic Books.
Rozin, P. & Fallon, A. (1987).
A perspective on disgust. Psychological Review, Vol. 94, pp. 23-41.
Rozin, P., Haidt, J. & McCauley, C. (1993).
Disgust. In M. Lewis & J. Haviland (editors), Handbook of Emotions. New York: Guilford, pp. 575-594.
Ridley, M. (1994).
The Red Queen: Sex and the Evolution of Human Nature. New York: Macmillan.
Roederer, J. (1979).
Introduction to the Physics and Psychophysics of Music. New York: Springer Verlag.
Roederer, J. (1982).
Physical and neuropsychological foundations of music. In M. Clynes (Ed.), Music, Mind and Brain. New York and London: Plenum Press.
Roederer, J. (1984).
The search for a survival value of music. Music Perception, Vol. 1, No. 3, pp. 350-356.
Shepard, R.N. (1987).
Evolution of a mesh between principles of the mind and regularities of the world. In J. Dupr\o'e\(aa' (Ed.), The Latest on the Best: Essays on Evolution and Optimality. Cambridge, MA: MIT Press/Bradford Books; pp. 251-275.
Shepard, R.N. (1992).
The perceptual organization of colors: An adaptation to regularities of the terrestrial world? In: Barkow, Cosmides & Tooby (Eds.), The Adapted Mind: Evolutionary Psychology and the Generation of Culture. Oxford: Oxford University Press; pp. 495-532.
Symons, D. (1992).
On the use and misuse of Darwinism in the study of human behavior. In: Barkow, Cosmides & Tooby (Eds.), The Adapted Mind: Evolutionary Psychology and the Generation of Culture. Oxford: Oxford University Press; pp. 137-159.
Tooby, J., & Cosmides, L. (1992).
The psychological foundations of culture. In Barkow, Cosmides & Tooby (Eds.), The Adapted Mind: Evolutionary Psychology and the Generation of Culture. Oxford: Oxford University Press, pp. 19-136.
Trivers, R.L. & Willard, D.E. (1993).
Natural selection of parental ability to vary the sex ratio of offspring. Science, Vol. 179, pp. 90-91.
Wallin, N.L. (1991).
Biomusicology; Neurophysiological, Neuropsychological, an Evolutionary Perspectives on the Origins and Purposes of Music. Stuyvesant, NY: Pendragon Press.
Werner, D. (1984).
Amazon Journey; An anthropologist's year among Brazil's Mekranoti Indians. New York: Simon and Schuster.
The Dancing Chimpanzee: A Study of the Origin of Music in Relation to the Vocalising and Rhythmic Action of Apes. London: Allison & Busby, 1980; revised edition.
Wilson, D. R. (1993).
Evolutionary epidemiology: Darwinian theory in the service of medicine and psychiatry. Acta Biotheoretica, Vol. 41, pp. 205-218.
Wright, R. (1994).
The Moral Animal: The New Science of Evolutionary Psychology. New York: Vintage Books.