Music and Evolution
Music and evolution An evaluation of the selective pressures on the origins of music and language L.T.T.
Hagemans Blok 3. 6 Bachelor thesis Begeleider: Dr. Harry Smit Universiteit Maastricht Faculteit der Psychologie Biologische afstudeerrichting ABSTRACT It is possible to survive without music. Therefore the origins of mu sic were not immediately considered to be an evolutionary adaptation relieving a specific selective pressure. However, cumulating evidence point to the biological significance of music.
Musicality originates early in development, it can serve as a mood regulator, facilitates group cohesion and cooperation, and it affects pair bonding during sexual selection. Until recently these theories did not provide a complete and precise story of the how and when questions about the evolution of music. This study attends the question of possible selective pressures on the evolution of music. The latest theory called Hmmmmm communication incorporates the former theories and is supported by archaeological evidence dating the origins of music much further back in time.
Hmmmmm communication implies a shared evolutionary ancestor of music and language, which signifies that one is not derivative of the other. During the evolutionary trajectory several important adaptations – like bipedalism and reduced sexual dimorphism resulted in the gradational evolution of Hmmmmm communication that eventually split into the precursors of the language and music we are familiar with today. 2 CONTENTS INTRODUCTION …………………………………………………………………………………………………… 1 THE SCIENCE OF EVOLUTIONARY MUSICOLOGY ……………………………………………. 6 1. 1 Types of evidence …………………………………………………………………………………………… 7 1. 2 Pitfalls of an evolutionary theory of music ………………………………………………………….. 8 2 MUSIC AND LANGUAGE …………………………………………………………………………………. 10 2. 1 Similarities concerning the human species and human society………………………………. 10 2. Similarities concerning the structure of language and music …………………………………. 11 2. 3 Differences ………………………………………………………………………………………………….. 12 2. 4 Independence between music and language ……………………………………………………….. 13 2. 5 Modularity of music and language …………………………………………………………………… 14 3 MUSIC AND EMOTION …………………………………………………………………………………….. 16 3. Mother-infant communication of emotion …………………………………………………………. 16 3. 2 Musical communication of emotion …………………………………………………………………. 17 3. 3 The emotional effects of music ……………………………………………………………………….. 17 4 SEXUAL SELECTION ……………………………………………………………………………………….. 19 4. 1 Mechanisms of sexual selection ………………………………………………………………………. 19 4. Miller’s wrong interpretation: problems with sexual selection of music ………………….. 19 5 GROUP SELECTION …………………………………………………………………………………………. 21 5. 1 Group cohesion …………………………………………………………………………………………….. 21 5. 2 Mood synchronisation and cooperation …………………………………………………………….. 21 5. 3 Social bonding ……………………………………………………………………………………………… 2 6 HMMMMM THEORY………………………………………………………………………………………… 23 6. 1 Indirect selection pressures …………………………………………………………………………….. 23 6. 2 Bipedalism …………………………………………………………………………………………………… 23 6. 3 Singing Neanderthal hypothesis ………………………………………………………………………. 24 DISCUSSION ………………………………………………………………………………………………………. 6 REFERENCES ……………………………………………………………………………………………………… 28 3 INTRODUCTION In reply to the question “what is music? ” people are often inclined to answer that “music is the language of emotion”. Although this phrase has become a catchy cliche, we still know little about why it has a ny meaning to us humans. Intuitively it looks rather comprehensible but when we try to explain music we come into trouble. Why do we see music as a language and how can this non-sentient entity communicate the most profound emotions to us?
It is hard to define music considering the many diverging descriptions for it. Greek philosophers and medieval theorists defined music as tones ordered horizontally as melodies, and vertically as harmonies. But disagreement of music as being only pleasant melodies led future composers to explore darker, harsher, and afterwards atonal melodies. Later on contemporary composers even used noise as a way to make music, and according to them any sound could be seen as music. Musicologist Jean-Jacques Nattiez (1990) states that even within a single society there is no consensus about the border between music and noise.
Therefore, ‘there is no single and intercultural universal concept defining what music might be’. According to another (ethno)musicologist, Bruno Nettl (1983), music is ‘human sound communication outside t he scope of language’. This description covers a lot but then again it fails to define the artistic statement made by composer John Cage in 1952. His composition 4,33 consist of nothing but four and a half minutes of silence signifying that apart from soun d, silence is a key aspect of music. It is even harder to identify the origins of music.
Considering the amount of effort and enjoyment we take in producing and listening to music one has to wonder how on earth all this brouhaha about the arrangement of so unds came to be. The old age and ubiquity of music demonstrate the inbred appreciation people have for music. But where does our predisposition to engage in music come from? When thinking about the origins of music it is tempting to assume an evolutionary explanation for it. After all, everything in nature that exists today must have had some adaptive value in the past otherwise it would not exist at all.
The existence of music is all the more mysterious then, because it does not seem to serve any purpose in our survival or reproduction. This is presumably the reason for the considerable controversy surrounding the questions of the origin, biological significance, and function of music. Steven Pinker (1997) argues that music is nothing more than ‘auditory cheesecake’, an evolutionary spin-off from language. In addition to Pinker, David Huron (2001) questioned the evolutionary origins of 4 music and underlined the post hoc argumentation of evolutionary theorizing.
He also points out the complex genesis of music evolution and agrees with a premusical adaptation but also suggests a co-evolution with culture. Although not very convinced of an evolutionary theory for music, Huron argues that music can facilitate social bonding. In contrast, Charles Darwin (1871) ar gued for music as a biological adaptation in sexual selection. Geoffrey Miller (2000) amplified Darwin’s sexual selection theory emphasizing the fact that music production is highest during adolescence at the time when males are most active in human courtship.
The most promising hypothesis is the ‘Hmmmmm’ theory from cognitive archaeologist Steven Mithen (2005). Hmmmmm stands for Holistic, multi-modal, manipulative, music & mimetic and is thought to be the common evolutionary ancestor of language and music. Although several researchers (e. g. Brown, 2000; Wray, 2002) have set an example for the idea of a precursory proto – or musilanguage, Mithen’s hypothesis provides a more complete picture combining data from multiple disciplines with archaeological evidence.
In order to understand more about the origins of music it is vital to spend a great deal of notice to the evolution of language, but it could also work the other way around. Either way the parallel study of language and music can make a large contributio n to the advance in science in contrast to dealing with them separately. Also, more and more studies are beginning to show the practical benefits of music therapy for mental and physical health. Furthermore, music has positive effects on mood, learning and creativity. Therefore, knowledge of the origins of music can be of great use to science and society.
It is a great mystery that although we could easily survive without music it has such a big impact on our emotions. It has the power to make someone cry or cause shivers down the spine, the ability to cheer up crowds at a concert or soothe a baby in distress. It appears that the faculty of music is one that we industriously endeavour. The question is: what made us this way? Perhaps the evolution of language, emotion or sexual evolution can help to explain the origins of music. However, the theories of music evolution are diverse and on their own they do not provide a clear picture.
The aim of this research is to evaluate the natural and sexual selection theories of music to see if an evolutionary argument can be made for the origins of music and its counterpart language. Therefore the question will be: What are the possible selection pressures for the evolution of music in relation to language? 5 1 THE SCIENCE OF EVOLUTIONARY MUSICOLOGY In providing a basic framework for the study of the biology of music, Wallin (1991) distinguished three branches in this field (fig. 1). The first of which, evolutionary musicology, deals with the origins of music and employs the data of the other two branches as well.
In search of musical origins evolutionary musicologists look for clues provided by evolutionary psychology, archaeology, and animal studies like bird song but especially utterances of primates from chimpanzees, gorillas, bonobos, geladas, gibbons, and vervet monkeys. Apart from that evolutionary musicology makes use of data from neurological studies that consist mainly of fMRI studies for specific brain areas (whether or not shared with language); lesion or pathological studies like aphasia and amusia; and developmental studies with evidence of early innate musical abilities.
Comparative studies can be carried out by making comparisons, of which there are two types. Comparative musicology (also referred to as ethnomusicology) is affected with comparisons between humans from different cultures or ethnical backgrounds. Comparative (or animal) psychology is concerned with comparisons between humans and animals, i. e. nonhuman animals (e. g. song -birds) and humanlike animals (e. g. primates). Fig. 1 Classification of the disciplines within biomusicology by Wallin (1991). This study is focussed on evolutionary musico logy. 6 1. 1 Types of evidence
A clear view of the science of evolutionary musicology is made easier by tak ing a brief look at the types of evidence that are used to discover the origins of music. Genetic evidence. This would be the best type of evidence for an evolutionary account of music, but until now there has not been any solid evidence of a gene for e. g . musical talent or absolute pitch. Both Miller (2000) and Huron (2001) emphasize on the need for genetic evidence for the conclusiveness of the evolutionary origins of music. In the case of language however, the FOXP2 gene suggests some heredity of language (Enard t al. 2002). FOXP2 is a gene that is implicated in the development of language skills, including grammatical competence. It is possible that this gene could be relevant to music as well. Neurological evidence. The presence of specialized brain structures is consistent with innate developmental dispositions producing music abilities in contrast to solely due to learning or conditioning. However, specialized structures are neither a prerequisite nor a sufficiency for music to be an evolutionary adaptation. On the other hand, shared brain structures e. . with language would suggest a shared evolution in the past. Psychological evidence. These can range from developmental data of the early musical abilities of babies to cognitive data about the rules and mechanisms of the systems we call music and language. For a faculty, like music, to classify as a complex human adaptation there are several classic criteria (Miller, 2000). The fact that music satisfies the criterion of an orderly development schedule makes a strong evolutionary argument for its adaptive value. Ethological evidence.
Ethology is the scientific study of animal behavior considered as a branche of zoology. It studies the instinctive natural traits as opposed to those found in laboratory environments. In the case of music, it is concerned with the consistency of musical behaviors with survival and propagation of genes. For example studies about the variety of alarm or mating calls from chaimpanzees to gibbons. Comparative evidence. Comparative psychology also studies animal behavior, but, as opposed to ethology, construes its study as a branch of psychology rather than as one of biology.
Thus, where ethology sees the study of animal behavior in the context of what is known about animal anatomy and physiology, comparative psychology sees the study of animal behavior in the context of what is known about human psychology. Anthropological evidence. This is the evidence that coincides with comparative musicology. For example, different cultures are compared and when a universal feature is found it can be attributed to evolution. Selection pressures have caused a common design 7 feature among cultural diversity. For instance there is not a single culture or cummunity to be found on earth that does not engage in music.
Thus, ubiquity is a sound evolutionary argument for music to have (had) adaptive value. Also, paleoanthropology can provide clues about the ideal circumstances in which music could have arisen. For example bipedalism has had a great influence on music development among others (Mithen, 2005). Archeological evidence. The earliest evidence of music is believed to be a bone flute although there is some debate about its authenticy – of approximately 50,000 years old found in Divje Babe, Slovenia (Ivan Turk, 1995 in Mithen, 2005).
Of course early hominids could have used sticks and stones prior to self-made instrument or simply singing for that matter but these clues are irretracable. Huron (2001) estimates the antiquity of music to be 250,000 years old. However, Mithen (2005) argues that before homonids could actually have begun to sing they had to have the ability to sing. Therfore selection pressures on music are dated back much further to 5 milion years ago were singing was becoming possible through means of other adaptations not directly related to music. Either way, the antiquity of music is another evolutionary argument for its adaptation. . 2 Pitfalls of an evolutionary theory of music When theorizing about the evolution of music one h as to be very cautious. The problem with evolutionary theories is the hindsight bias. It seduces us to post hoc reasoning; that particular adaptations were designed for a particular purpose, while in fact one should rather focus on why all the other designs have ceased to exist. In other words, why has the surviving adaptation escaped the clean-up. There are some problems to tackle, which will become clear by attending to some basic questions.
Is music an evolutionary adaptation? While most evidence is based on the assumption that there had to have been an evolution of music, this does not have to be necessarily so. There are several psychologists who do not think the evidence in support of music as an evolutionary adaptation is strong (e. g. Pinker, 1997; Huron, 2001). However, there is also no evidence that would pose a serious threat for an evolutionary theory of music. Moreover, according to Miller (2000) music satisfies many of the classis criteria for a complex adaptation. How could music have evolved?
Present-day there are numerous possible candidates for the selection pressures of music evolution. These fall into all four categories: natural selection, sexual selection, group selection, and cultural selection. Moreover, Huron (2001) 8 argues that if music is indeed an evolutionary adaptation, then it will probably have a complex genesis. It could be built upon several other adaptations e. g. for language. It might represent several adaptations that are all placed under the same denominator i. e. music. It co uld be involving a complex co -evolution with culture.
Or, it could be that music used to have survival value in the past but is now merely vestigial (e. g. like the human appendix, or tonsils). Thus, there are a lot of hypothesis to choose from. Do language and music have a common precursor? In extension to the previous question, there are distinct patterns of evolution (Hauser & McDermott, 2003). Homoplasies are traits that started as two distinct lineages but ended in an evolutionary convergence (convergent evolution). Homologies are traits that are descendant from a common ancestor. Music is believed to be a homology to language.
However, there are four possible theories for the evolution of music and speech: 1) music evolved from speech, 2) speech evolved from music, 3) both evolved from a common ancestor, or 4) music and language evolved in parallel to each other (Brown, 2000; Mithen, 2005). Are there language and music universals? Since Chomsky (1957), language is said to have a universal basis that can be found in all languages over the world. Chomsky points out to the fact that children learn language very fast and without much effort, which led him to think there is an innate language system.
This system accounts for the fact that grammar is learned automatically and after learning the rules of language one can make infinite combinations of words with finite means. Because of its presumably shared history with language, music is also suggested to have such a universal. However, the notion of a musical grammar is contentious and only one theory (Lerdahl & Jackendoff, 1983) has been put forward until now. Moreover, some studies show that grammar is not reserved to humans only. Starlings are able to recognize recursive grammar (Gentner et al. 2006), which is difficult to reconcile with the notion that recursion (the grammatical process of embedding) is attributed strictly to humans (Hauser et al. 2002). Also, several researchers suggest that language depends much on the use of holistic, that is prefabricate d, expressions (e. g. pig in a poke) instead of grammar (e. g. Wray, 2002). Is music a result of culture? While focusing on biological issues, it is wrong to assume that the effects of culture on music were minimal. It seems implausible that the complexity o f music is due solely to the force of nature.
Therefore, music evolution is bound to have had some cultural influence. 9 2 MUSIC AND LANGUAGE In trying to find out the origins of music and a possible evolutionary history with language it helps to look at t he similarities and differences between music and language. Similarities could serve as evidence for a dependence of music and language to the same underlying mechanisms, while the differences point to a divergence between the two. Of course one has to not e that similarities are no conclusive evidence of a common ancestor while these similarities could be the resulting homoplasies of a convergent evolution.
In order for music to be a ‘language of emotion’ it would be forced to evolve into a system with language-like characteristics, implying a common origin. On the other hand, this line of thought is not very convincing because music and language have just too many important similarities for these to be chance occurrences alone, as pointed out by Brown (2000 ). At the same time Mithen (2005) argues that music is – or eventually became – too different from music to be a spin-off from language (Pinker, 1997). 2. Similarities concerning the human species and human society Universality. Music and language are universal among humans; they are present in all human societies. This can be observed not only on the level of societies or cultures but also on the individual level. Humans have a general capacity to acquire linguistic and musical competence. It is important to notice the unbalance between music and language production. Because language is far more practical for communication it is used more often than music and therefore it is practiced on a daily basis.
This has the effect that the production of music is limited to a relatively small group of people leading to a non-universal appearance. Nevertheless, there are no unmusical people only untrained ones; after all, ‘unmusical’ people are perfectly capable of listening to and distinguishing patterns of sound. If music was taught as intensely as language in the early days at school, we could make music as easily as talking or writing. In other non-western cultures where more people engage in everyday music making, this unbalance exists indeed to a much lesser extent (Blacking, 1973).
Human specificity. Both music and language are specific to the human species only. It is possible to consider bird song or gibbon duets ‘music’, but that would be a human interpretation. Music produced by animals has a simplistic and arbitrary nature and animals are not able to play with words or notes as humans are. 10 Cultural diversity. Among cultures there is a wide variety of music forms and present day more than six thousand languages are spoken in the world (Mithen, 2005). Like languages, cultural differences in music are determined by stylistic, geographical and social boundaries.
The cultural diversity is a result of cultural transmission via generations and between societies, which makes identifying the boundaries difficult. For instance, when did Latin became Italian or Classical music became Romantic? It is also suggested that the cultural diversity is constrained by universal features underlying language and music (e. g. Blacking, 1973). 2. 2 Similarities concerning the structure of language and music Modes of expression. Music and language are the two primary acoustic communicative systems of our species. They both share three different modes of expression: vocally, physically, and in writing.
Vocalizations are recognized in speech and song, body movement is used with sign language (or gestures) and dance, and both systems can be written down. Hierarchical structure. Because of its clear-cut description the following text is a citation of Sloboda’s notion of hierarchical structure: “It is common to consider a human language as comprising three components: phonology – a way of characterising the basic sound units of a language; syntax – the rules governing the way in which sound units are combined; and semantics – the way in which meaning is assigned to sound sequences.
Music seems to break down quite naturally into the same three components” (Sloboda, 1985). Within the phonological level the acoustical elements of both music and language can be divided into two sublevels (Sherer & Zentner, 2001). 1) Segmental features are the acoustic characteristics of the building blocks (segments) of a structure. In music, those building blocks are the individual sounds produced by the singing voice or an instrument, corresponding to phones of the phonemes (a set of phones) in speech. Segmental features are the features that describe the acoustic structure of the individual tone segments (building blocks).
The segmental features are basically the same for music and language, namely: duration, energy, pitch, and – in the case of music – timbre. 2) Suprasegmental features are defined by Sherer and Zentner as ‘the systematic configurational changes in sound sequences over time’, and lie on top of the segmental features. In speech these are for exa mple prosody, intonation and stress, while in music comparable features are melody, tempo and rhythm. Molino (2000) and Mithen (2005) underline the melodic and rhythmic component in speech, which is necessary to perform proper prosody and conversation. 1 Combinatorial systems. As mentioned above, language and music have a structure of hierarchical nature that consists of the small acoustic elements (words or tones) that can be combined into larger phrases (utterances or melodies) and those can be combined aga in to make linguistic or musical discourse (Mithen, 2005). These combinations can lead to recursion: a linguistic or musical phrase is embedded within another phrase of a similar type. With recursion an infinite range of expressions can be generated from a set of finite elements.
While animals are constraint to a finite and small set of utterances in the communicative repertoire, humans can generate infinite numbers of novel sequences that they have never heard before (Hauser et al. , 2002). Thus music and language are both generative combinatorial systems. 2. 3 Differences Symbols. The nature of the discrete units (words or tones) in the hierarchical structures is completely different. Language uses symbols that have a referential meaning based on an arbitrary association between the symbols and the referent (e. g. dog’ refers to a hairy animal that barks but the word does not look or sound like a dog). In music the notes do not refer to anything not even emotions, so there are no symbols in music only emotional expressions. It is the holistic phrase in a piece of music that obtains an emotional load. The only meaning that music can have is that carnival music ‘means’ more to people below the rivers in Holland than for people above. This is similar to the not ion of Wagner’s use of ‘leitmotifs’ in his operas; although it represents something outside the scope of music, the message of the music itself remains without any literal meaning.
However, the formulaic aspects of language (like in a saying for instance) have a greater similarity with music than we tend to think. Just like in music it makes no sense to separate the individual pieces of an expression because then the meaning of the expression is lost e. g. trying to translate the Dutch saying “met je neus in de boter vallen” word by word, does not contribute anything to its understanding. In a similar way separating the opening notes of Beethoven’s Fifth (e. g. by ten seconds) eliminates its meaning completely. Grammar. Because of the rules of grammar, ‘lion k ills man’ means something different than ‘man kills lion’.
Chomsky stated with his ‘poverty of the stimulus’ argument that the acquisition of grammatical rules by children is too incredible considering that the grammar rules have to be abstracted from the utterances they hear. He therefore proposed that there is an innate set of predispositions, a ‘Universal Grammar’, in the human species, which 12 helps in acquiring grammar. Researchers like Lerdahl and Jackendoff (1983) went in search for a musical equivalent of the universal grammar but it soon turned out that the rules of musical style are fundamentally different from the rules of language.
Unlike in ‘lion kills man’ changing the order of the notes in a piece of music does not change its meaning, because there is no meaning to be changed in the first place. Another difference between the rules of music and language is the change through time. Because language has to remain understandable to have a function it has to remain stable and therefore cannot change much (Mithen, 2005). Music on the other hand is changed a lot because deviation and renewal is appreciated in music. We can observe that language indeed changes at a much slower pace than musical styles evolve.
For example the word ‘gay’ meant something different before the 1970’s but the number one on the top 40 hit -list changes every few weeks and a completely new music style can develop within years if not months. Information transmission. Music is a non-referential system of communication and lacks any clear communicative function. There is no information transmitted like in the case of language, but it can move us on a very deep level. Therefore music is said to be purely manipulative and language is said to be referential.
However, language can be manipulative as well, for instance when somebody tells us something to do we are driven to action (Mithen, 2005). 2. 4 Independence between music and language To discover if the similarities and differences between music and language can be reduced to structures in the brain one is assigned to the domain of neuropsychology. From this discipline the lesion studies and studies of pathologies like amusia and aphasia provide a great deal of evidence. Amusia is a deficit in processing or producing music, while apha sia is the deficit in understanding or producing language (Mithen, 2005).
Amusia can be congenital from birth -, or acquired – after brain damage. There are also several types of aphasia. Broca’s aphasia is an intact understanding of language but the ina bility to speak. Wernicke’s aphasia is the reversal of this; while patients are able to speak but their utterances are meaningless and they cannot understand words that are spoken to them. Also differences in degradation can be distinguished; global aphasia comprises the whole loss of language abilities; anomic aphasia on the other hand is restricted to not being able to find the appropriate words when speaking.
A t ypical case of aphasia is that of the Russian composer Shebalin who suffered a severe stroke in the temporal and parietal regions of his left hemisphere. He was not able to 13 understand people or make comprehensible utterances but his musical abilities remained untouched. During his disorder he even finished his Fifth symphony, which was considered a brilliant and artistic piece of music by his contemporaries. Other cases show similar symptoms but there is a lot of variation (Mithen, 2005). Some patients differ in the capacity to recognize environmental sounds, foreign languages and linguistic prosod y.
Again others have different recovery rates for musical, verbal and environmental sound recognition. The existence of a dissociation between recognizing words, music, or environmental sounds suggests that there is not one but rather two distinct modules for processing auditory information: one for speech and one for sound (Peretz & Coltheart, 2003). In the cases of amusia patients are unable to process music but have normal language capabilities. As with aphasia there is a lot of variation in the extent o f the various components of musical abilities that are impaired (Mithen, 2005).
For example after a car accident the French composer Ravel could not write down the music in his head. Other patients lost their tonal knowledge: the implicit rules humans have about the proper structure of music. Tonal knowledge is used when a mistake is recognized in a string of notes for example. The existence of congenital amusia suggests that there is a genetic foundation for the musical capacity. Overall, the findings of impaired musical abilities with intact language abilities argue for neural specificity of a music module (Peretz & Coltheart, 2003). . 5 Modularity of music and language Where does all this evidence lead to? Starting from the antiquity and ubiquity of mus ic and language we can assume that they are both posterity from very precursory adaptations in the distant evolutionary past. Although the similarities and differences between music and language are no hard evidence, they are compatible with an initial sha red evolutionary history following a period of independent evolution. Brown (2000) was one of the first to propose a common ancestor for music and language in his musilanguage model.
If there really were a divergence of language and music this would be ret raceable in differentiation of brain areas for processing of music and language separately. Indeed neurological evidence points to the modularity of music. Double dissociations in cases of amusia without aphasia and aphasia without amusia demonstrate that music and language have little cognitive overlap. The evidence suggests that music and language have a degree of independence from each other but are not entirely independent mental faculties, rather labels for complex sets of processes, while some of 14 hich are shared and some are different (Platel & Peretz 1997). The data provided by pathological patients and brain imaging studies have led to the notion that language and music processing can be explained in reference to the modularity of mind (Fodor, 1983) . According to Peretz and Coltheart (2003) there are different modules for language and music, some of which are shared and others are domain-specific. Therefore, it is wrong to assume a single music module and another one for language but rather several specific modules comprising the faculty of music and language separately (fig. ). The presence of specific evolved music modules is in contradiction with Pinker’s auditory cheesecake hypothesis. Moreover, the psychological data concerning development and e motion support the idea that music has adaptive value. Fig. 2 A modular model of music processing. Each box represents pathways of information flow or communication between process ing components. A neurological anomaly may either damage a processing co mponent (box) or interfere with the flow of information between the boxes.
All components whose domains appear to be specific to music are in green; others are in blue. There are three neurally individuated components in italics – rhythm analysis, meter analysis, and emotion expression analysis – whose specificity to music is currently unknown. They are represented here in blue, but future work may provide evidence for representing them in green (Peretz & Coltheart, 2 003). 15 3 MUSIC AND EMOTION Music has a strong link with emotion and evidence from developmental psychology supports the idea that music has adaptive value for infants.
Mithen (2005) recapitulates that music can successfully communicate intended expressions of emotion and thereby induce emotions in adults but even in newborn babies as well. It appears that music has developmental priority over language even. 3. 1 Mother-infant communication of emotion Compelling evidence is derived from numerous studies in developmental psychology. They primarily hinge on the phenomenon of infant -directed speech (IDS) or ‘motherese’, which is the communication between newborn infants and their mothers or caretakers.
As described by Mithen (2005), IDS is speech with highly exaggerated musical features; high pitched, longer vowels, hyperarticulated, slow, and more repetition. This particular type of communication is directed towards infants and accommodating to their biased interest for IDS. IDS can be considered neither language nor music or perhaps both. Anne Fer nald (1991 in Mithen, 2005) has distinguished four stages of IDS: the first three are concerned with maintaining contact between mother and infant, modulating arousal and emotion, and conveying intentions and feelings.
Only the last stage of IDS is explicitly facilitating language learning. IDS can be seen as a part of the mother -infant interaction, which is described by Colwyn Trevarthen (1999 in Mithen, 2005) as a fine-tuned multi-modal dance of interactional synchrony and a necessary foundation for succe ssful social interaction. This interaction can also include maternal singing which is in fact preferred more by babies than IDS (Trehub & Nakata, 2001). The key features of IDS are that it is universal and that it is used long before infants begin to acquire language (Mithen, 2005).
The adaptive value of IDS is obvious; being able to communicate fundamentals needs such as hunger, distress, frustration, and contentment to its mother benefits the survival of the infant. In the mother-infant interaction music played an important role of enhancing the emotional communication, but exactly how this mechanism evolved is not clear. Mithen (2005) proposes that IDS was derivative of musicality that regulated social relationships and emotional states. But alternatively, according to Ellen Dissanayake (2005) it could also be the other way around; musical capacity enhanced the mother -infant interaction and this adaptation 16 erved as an ancestor to the later development of musilanguage or proto -music in prehistorical hominids. 3. 2 Musical communication of emotion For musical communication to be effective it should be able to reliably transfer emotion from performer to listener. If there were no consistent agreement about which emotion is conveyed in a particular piece o f music then music would not have the function of communicating emotion. It would rather be a randomly interpretative sound source without the possibility for a performer to express his emotional intentions.
Surely such a system would soon extinct because it has no benefits. Many studies have shown that emotions can be reliably communicated between performers and musically untrained listeners (e. g. Juslin, 2001; Sloboda, 2001). Also there is consensus between listeners about the basic emotion that can be e xtracted from a music piece (e. g. Hevner 1936). The successful recognition of emotion is due to the iconic coding of the segmental features mentioned earlier (Sherer & Zentner, 2001). While symbolic coding uses arbitrary symbols as mentioned earlier, iconic codes contain features of the intended object.
For example, pitch could refer to the size of an object as can be heard in Saint -Saens “Carnaval des animaux” where the elephant is represented by the low notes of the double bass. The little birds in the “voliere” are represented by high-pitched tones with fast and whirling melodies that mimic their chaotic fluttering. The high sound energy or volume of a tone could be associated with high activity like aggression or anger while, while soft and mellow tones are often related to low activity like peaceful or sad emotions.
Again there is an evolutionary constraint to be found on the external appearance of music, which is bounded by iconic coding. 3. 3 The emotional effects of music The impelling propensity to engage in music over and over again has to do with the pleasure it gives us. Music can even elicit highly pleasurable experiences of shivers -downthe-spine, chills, or tears. Moreover, these can be attributed to specific musical features as is shown in a study by musicpsychologist John Sloboda (1991). Music seems to have a direct route to our central nervous system and can therefore draw automatic responses from the 17 body.
The neuro -affective foundation of musical appreciation is demonstrated by a study in which subjects that experience intensely pleasurable responses like chills, show brain activity in areas involved in reward, emotion, and arousal (Blood & Zattore, 2001). These are the same areas that respond to other euphoria-inducing stimuli like food, sex, and abusive drugs. This means there is a link between music and survival-related stimuli, which supports the biological significance of music. An alternative explanation for the experience of musical pleasure is proposed by Huron (2001), which he calls non-adaptive pleasure seeking (NAPS) theory.
A typical example is the human taste for sugar and fat. Millions of years ago these were hard to find so any predisposition for preference of sugar and fat would aid to survival. Present -time the resources for sugar and fat are plentiful but the rudimentary reward system for them still exists. This inclines people to eat unnaturally high levels of fats and sugars and causes health problems like obesity and heart disease. Similar like abusive drugs, music could stimulate the pleasure centres of the brain that evolved to reinforce other adaptive behaviors, thereby contributing nothing to enhance survival.
However musical pleasure does not result in neglect or poor survival as apposed to alcohol or heroin abuse for instance, but instead it even has positive effects. The positive effects of music come in various forms. The most obvious of course is that music can induce emotional states (Sherer & Zentner, 2001), which is suggesting that music could serve as a mood regulator. The benefit of having emotions is enabling people to deal with the social world around them (Oatly & Johnson-Laird, 1987 in Mithen, 2005).
Any living being has to make decisions about how to act in his environment, but there is limited time, impartial knowledge, and conflicting goals. Emotions have evolved as a guide to make decisions in difficult situations, which is also described as bounded rationality. The social world is exactly the complex situation were emotions come in handy and not surprisingly more complex emotions are related directly to our social relationships (Mithen, 2005). The ability then to regulate mood and facilitate social and emotional interaction has been of great value for survival (Huron, 2001). 8 4 SEXUAL SELECTION 4. 1 Mechanisms of sexual selection According to Darwin (1871) there are two different types of sexual selection: aggressive rivalry, and mate choice. If music indeed has evolved through sexual selection, the preference for it has obviously been selected by means of mate choice, while there is no aggressive struggle to be won with music. In his effort to extend Darwin’s idea of evolution of music by sexual selection, Miller (2000) elaborates on two key principles of mate choice: indicators and aesthetic traits.
Indicators are features that show the fitness of an individual, for example Miller describes the peacock’s tail as an indicator of good genes. Because the costs of having such a large tail are high – they can easily be spotted by predators and have more difficulty of escaping – only very healthy peacocks can carry it. Aesthetic traits are physical or behavioural features that take advantage of the perceptual bias of those looking for mates. In this case Miller gives an example of birds that for the most part eat red berries, therefore there eyes are sensitive to red and their brain is attracted to red.
Because of this perceptual bias mates will prefer a red coloured bird to a different colour. So, this species would then evolve red feathers as a side effect o f eating red berries. The process of sexual selection can be explained by two important principles. The runaway selection (Fisher, 1930) accounts for the disproportionately large size of a peacock’s tale for example. When the heritable mate preference for large tails will be genetically correlated with the trait itself, a positive feedback loop arises and the tail will become much larger than otherwise expected.
Another mechanism is the handicap principle devised by Zahavi (1997), which states that for a tr ait to become preferable for choosy mates it has to have high costs otherwise it could be easily faked. Therefore the benefits should be relatively low and handicap the individual’s own survival. 4. 2 Miller’s wrong interpretation: problems with sexual selection of music Singing and dancing is suggested to be an indicator of fitness because it takes time and energy that could otherwise have been invested in more beneficial behaviors like sleeping for example (Miller, 2000).
Moreover it takes time to practice, which is unprofitable for the same reasons. Miller believes that in the past music has become a ‘runaway sexual selection of 19 complex acoustic display’ by anthropoid apes and that because it provides no direct survival benefit it must have evolved solely by sexual selection. The problem with this notion is that Miller fails to appreciate the survival benefits of communicating emotions, intentions, information, and facilitating cooperation (Mithen, 2005).
Furthermore, because music-making is largely a gro up activity this contradicts the notion of the selfish gene that is incompatible with a theory of group selection (Dawkins, 1976). If an individual attempts to propagates his selfish genes this would not be facilitated by cooperating with other competing individuals, because their musical display could be sexually selected at the cost of one another. Another problem with the sexual preference of music is the fact that music -making is not predestined for one single sex but men engage in it as well as women do (fig 3).
This contradicts with the sexual dimorphism that is normally present in sexual selection related behaviors (Huron, 2001). If music was a male trait to enhance the sexual selection of his genes then females should have to be no more then a weak reflection of males, otherwise female cost of music-making would be high without any benefit even. Obviously this is not the case and Mithen (2005) suggests that there actually has been a reduction in sexual dimorphism so that females got bigger which explains female participation in music.
This facilitated pair bonding between males and females, which enhanced the survival of their infants because they would be raised by both parents. Fig. 3 Violinist Janine Jansen portraying herself in a rather flirta tious manner, which is not conventional in the present -day classical music world. In the Vivaldi CD -booklet a series of similar pictures are added which serve no musical purpose other than displaying a romantic atmosphere perhaps. Obviously she uses her looks to profile herself as a sexual intelligent musician to increase sales probably. 20 5 GROUP SELECTION
A problem with music as a sexual selected adaptation was the fact that music is made in groups. However, on its own group selection is a plausible argu ment in explaining social bonding and group cooperation. Of course, nowadays passive and individual listening are very common, but this is not the ‘natural’ situation for human music, which is dancing and participating altogether. 5. 1 Group cohesion Of all the cultural diversity in music, whether it is made for fun or professionally, spontaneously or ritually, the one aspect that is the same in all cultures is that music is principally a shared activity.
In relation to sexual selection neurobiologist Bjo rn Merker (2000) argues that this could have arisen from prehistoric males forming a group and making choral music together to attract females that are leaving their group in search for mates. According to Miller (2000) this would help females to compare t he singing and dancing males in the group and choose their mate. However Mithen (2005) rejects this argument based on the fact that this music-making would also attract predators. He believes a more suitable explanation is to build trust among males, and females as well.
This relates to the idea of world historian William McNeill (1995), that groups who depend on one another are inclined to make music together. According to McNeill this is demonstrated in dance and military drill, which creates cohesion within the group. The shared music making leads to so -called ‘boundary loss’: the blurring of personal boundaries and a feeling of merging with the group. 5. 2 Mood synchronisation and cooperation Mithen proposes that the problem of why individuals with se lfish genes favour to cooperate can be approached with the classic work on the prisoner’s dilemma (Robert Axelrod, 1984).
It demonstrates the value of cooperation, but it also shows that it is difficult to achieve because people’s interests are neither completely concordant nor completely conflicting and another’s intentions are difficult to find out. Therefore, Mithen suggests, music-making can show willingness to cooperation while at the same time it has low costs and it is comparable to the first move of the ‘TIT for TAT’ strategy. In this strategy the first 21 move is always to cooperate but in the case of music there is not much to loose if the others defect.
But while the costs of joining in singing and dancing are low and the benefits of cooperation are high, music making can be very easily exploited by free-riders. The principle of boundary loss can provide a simple solution to this problem according to Mithen. The music causes the loss of self-identity in the group and it can also set the mood for cooperation because happy people are more cooperative (Fried & Berkowitz, 1979 in Mithen, 2005). 5. 3 Social bonding An important aspect of why people find music so much fun to do or makes them feel good is social bonding.
To cooperate and consequently overrule the selfish genes poses a bonding problem between members of the group. Social bonding is suggested to be the mechanism of how cooperation could be established. Cognitive scientist William Benzon (2005) suggest this is done by ‘coupling’ where people tune in on each other by synchronizing the sounds they make so that their brain states are coupled together. According to Brown (2000) two design features of music are perfectly suited to form a sense of communion by promoting simultaneous singing and dancing .
Firstly, pitch intervals cause the voices to blend harmoniously when sounding together. This is a possible explanation to how boundary loss arises when one’s own voice cannot be heard anymore among the voices of the other members of a choir. Secondly, te mporal regularity can facilitate motor synchronicity. In addition, Peretz (2001) puts forward that these design features appear to be specific to music and that they are not shared with speech, which requires individuality to be intelligible. 22 6 HMMMMM THEORY
Mithen (2005) devised his Hmmmmm theory in which he proposes that the ancestors of humans and apes developed a form of communication, Hmmmmm communication, 5 million years ago. This communication eventually grew a more and more musical character until in Homo sapiens it split up in language and music. The natural forces that caused the evolution of this Hmmmmm communication can be considered the indirect selection pressures on the evolution of music. 6. 1 Indirect selection pressures The notion of indirect selection pressures regards the gradational nature of evolution.
The human adaptation of the musical faculty did not evolve directly in one piece but instead several other adaptations cleared the path for music to evolve later on. There are several changes in the life of early hominids that led to important adaptations that were significant to music development. Based on archaeological evidence, like fossils of zebras and antelopes, it is concluded that some hominids lived in more open landscapes tha n their forest -dwelling relatives. Consequently they are presumed to live in larger groups because the possibility to escape in trees from predators was lost.
In larger groups the predators can be spotted more easily and there is less chance for an individual to be attacked (Mithen, 2005). Living in larger groups had the consequence of more social tension, but these social groups would have also been very intimate communities sharing knowledge about group members, social relationships, and the environment. The necessity to cope with this complexity involving larger groups could explain the larger brain size proportionally correlated to group size (Dunbar, 1992 p. 27 in Mithen, 2005). 6. 2 Bipedalism These major changes in lifestyle, the change to a more op en environment and living in larger groups, resulted in several adaptations that had huge consequences. The most influential adaptation was the development of bipedalism, which is believed to be a result of decreasing heat stress on the hot planes of the s avannah by walking higher above ground, summarized by the words ‘stand tall and stay cool’ (Wheeler 1988, p. 145 in Mithen, 2005) 23 Bipedalism had a great impact on child bearing.
Because the hips are narrower than with four-legged creatures and due to the increasingly larger brain size, the babies were forced to be born in a helpless not fully developed state. To overcome this problem the child and mother had to grow a strong bond, which enabled the child to develop (learn emotions, language, etc. ) and be able to deal with the complex social world later in life. As discussed earlier this care-giving would have been facilitated by music. Walking on two legs also caused the larynx to drop making more room in the throat to produce more complex holistic utterances.
This built a larger call repertoire of Hmmmmm communication and without the ability to make a wider range of sounds the production of music, singing in this case, would have been impossible. The development of rhythm and dancing could have been a result o f a need to make walking on two legs easier. Because the weight of the body shifts from side to side and changes its centre of gravity, the coordination of talking steps could be facilitated by rhythm. Moreover, walking itself produces a kind of rhythm be cause if one would take irregular steps one would not came very far very fast. . 3 Singing Neanderthal hypothesis According to Mithen Neanderthals were intelligent but their intelligence was domain specific (Mithen, 1996). In contrast to Homo sapiens the y lacked ‘cognitive fluidity’ which enabled humans to make connections between different areas of knowledge like the natural world, physical material, and complex social interaction. For example, knowledge about techniques for tool making could be used to make a necklace for a potential mate. This cognitive fluidity was made possible by the emergence of language, which Mithen believes the Neanderthals lacked.
For them this division between music and language did not occur like with Homo sapiens but they evolved to an even more musical more sophisticated form of complex singing-like Hmmmmm communication (Mithen, 2005). Thus in Homo sapiens music was derived from the remains of the early Hmmmmm communication after language had evolved (fig. 4). Mithen suggest s that the development of language repressed the further development of music in comparison to the sophistication of music in the case of the Neanderthals. In other words Homo neanderthalensis was more musical than Homo sapiens. 24 Fig. The evolution of music according to Mithen (2005) 25 DISCUSSION Overall, we can conclude that music is certainly not a by-product of language as Pinker (1997) suggests. Based on the numerous adaptive values music provides for survival and reproduction we can clearly state that there are indeed evolutionary grounds for the origins of music. The modularity of music denotes a specifically designed adaptation for music (Peretz & Coltheart, 2003). Especially the developmental predispositions found in babies are a compelling argument that music has been useful for establishing a bond between mother and infant (Trehub, 2003).
But what were the reasons why these specific adaptations evolved? In other words what could be their selective pressures? The evolution of music is no meticulously planned design for an adaptation to counter some evolutionary problem, premeditatedly directed by natural selection. It seems that other adaptations made a clear path for music to become favourable and when it came in handy it was used. For instance, Mithen shows that a change of lifestyle resulted in the adaptation of bipedalism. In turn bipedalism had an effect on the need for care-giving facilitation and rhythm for example (Mithen, 2005).
An adaptation related to bipedalism was the reduction of sexual dimorphism that facilitated pair bonding, which could in turn increase the infant’s survival success. All the previous theories suggest that music is an evolutionary adaptation but there is no consensus on how and when it happe ned. The strength of Mithen’s (2005) hypothesis is that he combines all these theories in one comprehensive story and tries to give a specific explanation of how and when music evolved. To achieve this goal he uses a lot of data from ethological and archaeological studies.
Unfortunately because there are no fossil records of prehistorical sounds and due to the scant archaeological data the conclusions are somewhat speculative (Dissanayake, 2005; Benzon, 2005). Nevertheless the Hmmmmm theory is very convincing in providing a complete history of music evolution. Most importantly Mithen tries to consolidate that language and music had a shared ancestor and the idea seems very plausible. However, due to lack of conclusive evidence the question of whether language perhaps did come first or not remains unanswered.
Benzon (2005) argues that it is necessary to find out if ‘the nature of the nervous system is such that the evolutionary trajectory must go through musicking on the way to language’. He thinks this is indeed the case and tries to provide a theoretical model to explain how this occurs. It is important to notice the high flexibility of music. It can have low costs, for example people could sing a simple tune and receive a lot of joy from it. But it can also have 26 high costs; one could go to an expensive concert in which the musician made a great effort to study the notes and spend energy and time on performing.
Either way the benefits are relatively higher. Also music appears to be a ‘facilitator’ of things. It seems that music makes things go a lot easier for instance care-giving, group cohesion and cooperation, and pair bonding. All these behaviors have one thing in common, namely they are all highly involved with emotions. It is evident that music somehow has a direct path to our emotions but the exact mechanisms remain largely undiscovered, partially because emotion itself is difficult to study.
Attempting to answer the question of why music has evolved like it did and where the close connection to emotions originates from can give helpful direction to understanding how these mechanisms work. Mithen’s greatest contribution to the origins of music is his emphasis on archaeological evidence and the vast amount of data that is derived from it. Future studies ar e necessary to further confirm his ideas. Especially combining perspective from multiple disciplines is inevitable to find answers. Therefore the science of language will probably stay in close relationship to music science and expectantly profit more and more from each other.
It is difficult to find universal grammatical features between and within music and language – considering that the only not quite successful attempt was made by Lerdahl and Jackendoff (1983). However it still seems very plausible that there are universals in music and language. Rather than to look at the structure of music perhaps it would be more fruitful to search for a kind of ‘emotional syntax’. Music, typically in the case of Mozart, is filled with musical phrases in the form of question and answer.
The types of music that are more or less bounded by the laws that constitute the musical discourse of question an answering could be analysed in relation to their emotional perception or induction. Hence, the emotional syntax should therefore be able to predict the emotion load of a string of notes. The evolutionary origins of music could provide a base for the study of musical sounds inducing emotions. Probably the study of animals like song -birds could provide more data on the questio n whether sounds sequences are designed by an underlying constraint in nature or that they are produced completely arbitrary without any specific design. 7 REFERENCES Axelrod, R. (1984). The Evolution of Cooperation. New York: Basic Books Benzon, W. (2005). Synch, Song, and Society. A review of The Singing Neanderthals: The Origins of Music, Language, Mind and Body by Steven Mithen. Human Nature Review, 5, 66- 86. Blacking, J. (1973). How Musical Is Man?. Seattle: University of Washington Press. Blood, A. J. , & Zatorre , R. J. (2001). Intensely pleasurable responses to music correlate with activity in brain regions implicated in reward and emotion. Proceedings of the National Academy of Sciences, 98, 11818-11823. Brown, S. (2000).
The “Musilanguage” Model of Music Evolution. In Wallin, N. L. , Merker, B. , & Brown, S. (ed. ), The Origins of Music. (pp. 271-300). Cambridge, MA: MIT Press. Chomsky, N. (1957). Syntactic Structures. The Hague: Mouton. Reprint. Berlin and New York (1985). Darwin, C. (1871). The descent of man, and selection in relation to sex (2 vols. ). London: John Murray Dawkins, R. (1976). The selfish gene. Oxford: Oxford University Press. Dissanayake, E. (2005). A review of The Singing Neanderthals: The Origins of Music, Language, Mind and Body by Steven Mithen. Evolutionary Psychology, 3, 375-380. Enard W. Przeworski, M. , Fisher, S. E. , Lai, S. C. L. , Wiebe, V. , Kitano, T. , Monaco, A. P. , & Paabo, S. (2002). Molecular evolution of FOXP2, a gene involved in speech and language. Nature, 418, 869-872. Fisher, R. A. (1930). The Genetical Theory of Natural Selection. Oxford: Clarendon Press. Fodor, J. (1983). The Modularity of Mind. Cambridge, MA: MIT Press. Gentner, T. Q. , Fenn, K. M. , Margoliash, D. , & Nusbaum, H. C. (2006). Recursive syntactic pattern learning by songbirds. Nature, 440, 1204-1207 Hauser, M. D. & McDermott, J. (2003). The evolution of the music faculty: a comparative perspective.
Nature Neuroscience, 6, 663-668. Hauser, M. D. , Cho