A Short Prehistory of Western Music (chapters 1-2)

Background and aims

What is `music' and why is it important?

Music's prehistory

Sources

There are four main sources of information about music's prehistory: (i) archaeological evidence in the form of musical instruments; (ii) archaeological evidence in the form of paintings or sculptures representing musical artefacts or activities; (iii) orally transmitted information preserved in stories, myths, etc. and later written down; (iv) present-day or recent oral cultures whose economy, technology and culture resemble those of societies in prehistorical periods; (v) recent studies of music from biological, neurological and social-anthropological perspectives.

Theories of music's origins

Figure 1: 60,000-year old cave bear femur bone segment with 4 holes (2 complete, 2 partial, one at each broken end). Produces 4 notes: do, re, mi, f	Figure 2 Rough chronology of hominids.

Evolutionist theories

Evolutionist theories of music try to explain its origins in terms of evolutionary adaptation, by which is meant the ability of a species to find effective strategies for survival by means of adapting to their environment. One such theory, presented by Merker (2000), is that music derives from the synchronous chorusing of higher primates. Merker's theory runs roughly as follows.

Chimpanzees in the forests of equatorial Africa live in highly territorial extended family groups. To produce the healthy offspring necessary for the survival of their species, they cannot breed within their own group. Their mating behaviour has therefore adapted so that females from one extended family group go off individually to mate with males from another. Invisible because of dense foliage, a gang of males in the other group draw the approaching female's attention by producing a concerted chorus of vocal sounds. She pays no attention to individual males outside the gang but to those who make the most striking sonic display within the chorus, such behaviour being identified with social acceptance and cohesion, as well as with strength and dominance. You need only think of an insecure young woman's attraction to the loudest lout in a cohort of lager-swilling soccer hooligans, or to the most garish virtuoso in a heavy metal band, to get the general picture.

These observations of behaviour among one type of primates (chimpanzees, not lager-lout or rock-band humanoids) may be important, but it is doubtful whether they explain the origins of the human phenomenon of music. The problem lies in the assumption that all primates from which we descend, or to which humans are similar, regulated mating and the procreation of their species in a similar way for long enough for that behaviour to become genetically hardwired, a process taking millions rather than thousands of years. Failing that interpretation of the theory, the assumption has to be that the primate mating behaviour was somehow passed on automatically by learning from generation to generation and species (primates) to genus (homo erectus) to genus (homo sapiens) without entering the human genome, an unlikely explanation, given the extent of environmental and social change we have had to confront over the last 100,000 years or so.

Dissanayake (2000) puts forward a different theory, arguing that `it is in the evolution of affiliative interactions between mothers and infants -- not male competition and adult courtship -- that we can discover the origins of the competencies and sensitivities that gave rise to human music.' This more palatable theory aligns with observations, presented earlier, about musical `universals', more specifically with those relating to pulse, timbre and mode of vocal articulation (see 3. Musical `universals').

Several other theories stress the importance of what Brown (2000) calls `musilanguage', i.e. that language and music, both sonic and both neurologically intertwined, stem from a common origin, `evolving together as brain size increased during the last two million years in the genus homo' (Falk 2000) . Like Dissanayake's mother-and-infant theory, this explanation also seems quite plausible becuase both Homo sapiens and Neanderthal man had, if our knowledge of the Slovenian bone flute and other early human instruments of music are anything to go by,21 clearly started to treat oral language and music as distinct modes of communication. Although neurologically interrelated, these two sonic systems were used for different functions. This aspect of evolution is important because the separation of music from language can be seen as a trait distinguishing humans from other animals.22

One common objection to the theory of distinction between music and language as a basis for understanding the origins of music as trait of human behaviour seems to argue that if we, as humans, say that birds and whales sing, then we are talking about music, simply because that is how we hear it. The sonic habits of humpback whales provide fuel for this argument. As those great mammals migrate or swim around their breeding grounds, they piece together repeated phrases, singing song after song for up to twenty-four hours at a stretch. Humpback whales have a seven-octave range similar to that covered by the piano keyboard, i.e. a range of fundamental frequencies well within the limits of what humans can hear, and much larger than the restricted range of pitches the human voice can produce. As the months go by, the whales modify their song patterns and most males end up by singing the same new song after a while. Moreover, humpback whale songs contain rhythms and phrases which, strung together, build forms of a length comparable to ballads or symphonic movements. It also seems that their songs contain recurrent formulae which end off different phrases in much the same way as we use rhyme in poetry. One theory about rhymes in whale song is that they help in the breeding season when the males have to remember `what comes next': the more elaborate the whale's song pattern, the more likely it is to rhyme.23

All these traits of whale song come across as typically musical to the human ear. But the `music' of the animal kingdom does not stop there: certain insects produce rhythmic patterns, some even metronomically regulated, which, like those of human music, vary and repeat in longer patterns. More importantly, eleven percent of primate species can produce short strings of notes that, although less musical to our ears than the singing of humpback whales, form a recognisable pattern in time. This behavioural trait, characteristic for most of our own music, is thought to have evolved independently four times within primates.24 All this evidence seems to suggest that music is not exclusive to the human species.

The problem with the objections just raised is that they are anthropocentric, i.e. that they interpret non-human behaviour on the basis of human experience, perception and behaviour. The animals make music standpoint assumes, in other words, that the whales, insects and primates just mentioned hear and react to the sounds they make themselves as we hear and react to them (as music), and that the animals produce those patterns of sound for the same reasons as we make what we hear as comparable patterns of sound in our music. Indeed, Thomas Eisner, Professor of Entomology at Cornell University and an accomplished classical musician, holds that we must `draw a distinction between enjoying animal sounds [as music] and saying that animals make music', even though, after hearing recordings of humpback whales, he admitted: `if a whale calls me up tomorrow and wants to do an evening of sonatas, I would be the first to volunteer' (Milius 2000) . Similarly, although we hear birds as the greatest songsters of the animal kingdom, it is, once again, highly improbable that they make, hear and use their melodies as we make, hear and use our music. Leading US ornithologist Eugene Morton is quite adamant in his refutation of anthropocentrism in relation to birdsong.

`Any analogy to human music is not interesting to me. It doesn't explain anything about how the world is, except how humans want to perceive it. Good on 'em, but I want to understand animals... Birdsong constitutes an avian broadcasting network, letting birds minimise the arduous work of flying about during interactions'.
(Milius 2000) .

If singing can replace the amount of flying around birds would otherwise have to do in order to interact, it is certainly part of a symbolic system. For example, instead of physically repelling every potential invader of its own space, a bird can claim its territory through song; instead of flying round to see if local members of the species are all there before they shut down for the night and that they are all there again in the morning, birds can join in the evening and dawn choruses. Birdsong is in other words a strategy for the survival of individuals within the group, because everyone has to have their place to nest, and for the group as a whole, because they may all need to collect for foraging, migration or for other purposes. Singing is clearly an energy-efficient way for birds to establish these essential relations.

It would in a similar way be absurd to expect whales, who have to cover huge distances in search of food but reconvene for breeding, to keep visual or tactile underwater checks on the whereabouts of each other, as individuals and as family groups, across vast stretches of ocean. In this sense, whale song, by replacing physical and visual contact with sonic communication, also acts symbolically to facilitate the social cohesion necessary for the survival of their species. It is also highly probable that, as with the chimpanzee chorus described above, the various functions of sonic communication in the animal kingdom are linked with what we humans might qualify as pleasure and pain, tension and relaxation, etc., i.e. with what we think of as emotions and which are essential ingredients in the evolutionary process of most sentient beings.25 If such `emotions' are linked to situations in the animal kingdom where what we hear as their `music' is used to signal messages we might understand verbally in terms like danger! or it's OK, we're all here or pick me, I'm best , then it is also probable that the sounds in question are accompanied by patterns of hormone production comparable to those found in humans when stimulated in certain ways by certain sounds in certain situations.26

If there is any truth in the line of reasoning just presented, it would seem that there may be grounds for calling that animal `music' music. After all, the argument would go, what we have described tallies quite well with the seventh of our eight axioms about music (see points 1-6, above), with our observations about ‘concerted simultaneity and collective identity’, and with several other points mentioned under our working definition of music. However, there is at least one human trait that animal `music' does not exhibit: unlike humans, animals do not appear to have two distinct symbolic systems mediated as sound, one system -- language -- more suited, though not wholly dedicated, to the precise denotation of objects and ideas, the other -- music -- more closely, though not entirely, related to the expression of movement, gesture and emotion (see axiom 4).

As stated earlier, language and music, both neurologically intertwined and both using the sense of hearing, seem to stem from a common origin, evolving together as brain size increased during the last two million years of evolution in the genus homo. However, it is first after we humans managed to wipe out our Neanderthal cousins (c. 50,000 years ago) that we start to leave technologically complex objects behind us, the most common of which are musical instruments made of non-degradable materials like bone. The appearance of so many musical instruments so early in our archaeological history indicates that Homo sapiens sapiens must have attached considerable importance to music, while the relative absence of such artifacts in remains of Homo erectus and Homo neanderthalensis suggests that our hominid predecessors either lacked the technological skill to create such instruments or that music (as we know it) played a less important part in their lives.

In short, the separation of sonic representation into two distinct but related spheres of activity -- what we now call language and music -- may have started to evolve in our hominid precursors but seems to have developed most explosively after their demise. Moreover, there is no evidence to suggest that any other species extant in the animal kingdom uses two distinct sonic systems of representation, one to communicate information and a different one to express corporeal or emotional aspects of that information. For example, a blackbird joining in the dawn chorus does not seem to distinguish sonically between, on the one hand, what we would think of as a verbal statement like I'm here and I hear you're all there and, on the other hand, whatever sensations she might have on waking up and on registering that she is chirping away in her own place, that her neighbours are in theirs, that there are males and fellow females in the vicinity, and that she's not the only one of her kind around. We humans seem to be the only species to make such a distinction between those modes of sonic representation. Cross (1999) argues that this distinction between language and music may be the most important thing humans ever did. We will return to this point at the end of the next section which deals with music's importance for another aspect of human development: the process of socialisation.

Music and socialisation

At the age of minus four months most humans start to hear. By the time we enter this world and long before we can focus our eyes on objects at different distances from ourselves, our aural faculties are well developed. Most small humans soon learn to distinguish pleasant from unpleasant sounds and most parents will witness that any tiny human in their household acts like a hyperactive radar of feelings and moods in their environment. You know it's no use telling baby in an irritated voice `Daddy's not angry' because the little human sees straight through such emotional bullshitting and starts to howl.

But baby's hearing is not what most parents notice first about sound and their own addition to the human race. They are more likely to register the little sonic terrorist's capacity to scream, yell, cry and generally dominate the domestic soundscape. Babies are endowed with non-verbal vocal talents totally out of proportion to other aspects of their size, weight and volume: they have inordinate lung power and vocal chords of steel, it seems, capable of producing high decibel and transient values, cutting timbres and irregular phrase lengths, all communicating messages like I 'm uncomfortable or I'm irritated or I'm in pain , or I'm hungry , so do something - now! and which parents have to interpret in terms like change my nappies or produce either breast or bottle for immediate consumption . Maybe these tiny humans have to yell not so much because they can't speak as because they need to dispel whatever state of adult torpor we happen to be in, be it watching TV, chatting, reading or, worst of all, sleeping. Babies seem to know in advance that sharp timbres at high pitch and volume carry extremely well, cutting through whatever constant underlying hum or mumble there may be in the adult world, be it idle conversation, the TV on in the background, fridges, ventilation, etc. Also, irregular rhythms and intonation by definition avoid the sort of repetition that can gradually transform into ambient (background) sound: a baby's yell is always up front, foreground, urgent, of varying periodicity and quite clearly designed to shatter whatever else mother, father, big sister or big brother is doing. That sonic shattering is designed to provoke immediate response. Desires and needs must be fulfilled now .

Now is the operative word in this context. Sonic statements formed as short repetitions of irregularly varying length are also statements of urgency, as well we know from the anaphones of news and documentary jingles -- important, flash, new, the latest update . Babies seem to have no conscious past or notion of future: all is present. The baby's lack of adult temporal perspective in relation to self is of course related to its lack of adult senses of social space, which, in its turn, relates to baby's egocentricity, essential for survival in the initial stages of its life.

In the perspective presented so far it is clear that non-verbal sound is essential to humans. We monitor it constantly from inside the womb until death or deafness do us part from its influence. We use our non-verbal voices to communicate all sorts of messages from the time we are born until we die or turn dumb. Together with the sense of touch, non-verbal sound is one of the most important sources of information and contact with social and natural environments at the most formative stages of any human's development. It is vital to senso-motoric and symbolic learning processes at the preverbal stage of development and central to the formation of any individual's personality. Moreover, everyone has experienced the process by which we gradually learn that we are not the centre of others' constant attention, nor can we be constant objects of immediate consolation or pacification: we have to get used to being just one human subject and social object among many other fellow human subjects. We have to have some sort of working relationship with whatever society and culture we belong to and we cannot live in the vain hope of returning to a state where we are the sonically dominant or foreground figures: we can never regain any imagined or real lost paradise, whatever advertisers may want us to believe.

Different cultures and subcultures establish and develop different patterns and norms for what course the process from baby via child to adult should run. The ultimate goal -- being the ideal male or female adult -- depends on whatever the society in question, on account of its material basis and cultural heritage, sees as desirable, useful and good. Assuming we have all been babies and if, as suggested here, baby's power over the domestic soundscape in the early development of every human is a biological necessity that must be relinquished for that individual to survive among fellow humans in adulthood, then we ought to gain important insights into how any culture works by studying patterns of socialisation that relate directly to non-verbal sound.

Humans can emit a tremendous variety of non-verbal sounds. We have to breathe to stay alive, but we can also cry, shout, yell, call, gurgle, sob, sigh, laugh, giggle, burp, fart, crunch, slurp, gulp, swallow, yawn, groan, moan, growl, cough, splutter, wheeze, sneeze, smack their lips, kiss, hiss, spit, blow our noses, clear our throats, cough up phlegm, etc., while our hearts beat, our tummies rumble, while we walk, run, stroll, jump, hop, scratch our heads etc. We may shudder with fear, tremble with delight, or shiver with cold so that their our chatter. We can also make sounds by dragging, pushing, cutting, hitting, tapping, patting, clapping, caressing, chopping, sawing, hammering, grinding, scratching, scraping, slapping, splashing, smashing etc. Some of these sounds are loud, others soft; some are heavy, others light; some are fast, others slow; some are high-pitched, others not; some are long or ongoing and repetitive, others short and discrete. All these humanly produced sounds are made within a context that is full of sound: in our society we have fridges, freezers, computer drives, traffic, aeroplanes, mains hum, air conditioning and all sorts of other machines; elsewhere we may be able to hear wind in the trees, rain, sea swell, animals, birds, insects, running water, thunder, earthquakes, ice breaking, crisp or slushy snow under our feet, waves breaking on the shore, etc.

Some of these sounds we make ourselves, others we just hear in a wide variety of acoustic settings, including those inside our own heads and bodies. Which sounds are evaluated as pleasant and unpleasant, which ones are deemed to be part of music and which ones not, will all depend on the culture we belong to and what sort of motoric and sonic behaviour prove to be generally compatible with the needs of that community, be it a youth subculture in late capitalism or a nomadic tribe from the stone age. These relationships between non-verbal sound, music and society are of course too complicated to be discussed in detail here. Suffice it to say that although many cultures have no equivalent to what we mean by `music', no human society has ever been without it. All of us have been babies and all of us have had to learn that we cannot for ever remain at the centre of the world around us, acoustically or otherwise. We have to learn to cooperate, to negotiate social space and uses for ourselves in relation to the community we belong to. Music and dance provide socially constructed sonic and kinetic frameworks for that learning process: most of us learn to sing, hum and whistle in accordance with the norms of what our culture regards as music, rather than just yelling, laughing, mumbling, or bashing objects at will in an uncoordinated or unpatterned manner. As we acquire the gift of language we learn to distinguish between humanly organised verbal and non-verbal sound. Moreover, many of us go on to learn how to play an instrument as a way of making sound whose functions are quite different not only to those of spoken language but also to those we make when chopping wood, hammering nails, ironing clothes, doing the washing up, flushing the toilet, taking a shower, walking upstairs, driving a car, eating food, operating machinery, folding a newspaper, closing the door, etc., etc. It would, from the perspective just presented, be absurd to regard music as some sort pleasant but parasitic appendage to human life -- `auditory cheesecake' as one writer put it.27

There are other reasons for understanding music as an essential part of the survival kit for any human society, not as just the cultural icing on the cake of production and economy. These reasons are presented by Cross (1999) and can be summarised in the following simplified terms.

As we have seen, the distinction between two modes of sonic representation (language and music) appears to be a specifically human trait. In fact, the human capacity to process signals from the world around them via different domains of representation (verbal, visual, motoric, emotional, etc.) seems to have been one of our species' great advantages in the evolutionary struggle in that we can sort out abstractions of cause and effect by distinguishing between visual, verbal, sonic and motoric impulses. Those domains of representation are even located in different parts of the brain so that what we hear at a particular time (a sonic event) does not have to represent the same phenomenon as a movement or emotion we may experience at that same time: having to rush up in a panic as the alarm clock goes off does not mean the alarm clock is stressed out, so to speak. Of course, such `domain-specific' signal processing in no way prevents humans from making connections between several simultaneous domain-specific signals if they co-occur on a regular basis. For example, when a loving parent talks in a sing-song voice to a baby while holding and rocking it, the little one receives signals that are at the same time specific to the sonic, motoric and emotional domains of representation. As these combinations of domain-specific signals are repeated, the infant learns to make connections between them so that another, overriding or `embodying' type of representation comes into play. Such combinations of sonic, motoric and emotional signals are sometimes called proto-musical.28

The specific domains relating to (proto-) musical representation, shown in figure 3, are all basically intuitive. They are: [1] physical, i.e. the ballistics and trajectory of a body, including one's own, travelling through space; [2] gross motoric, i.e. the movement of arms, legs, head, etc; [3] fine motoric, i.e. muscular control in moving, for example, fingers, eyes and, most importantly for music and language, vocal chords; [4] linguistic, i.e. prosodic patterning;30 [5] social, in terms of knowing which sounds or movements produce what effect on you and those around you; [6] emotional, in terms of evaluating a situation in response to different body states such as posture, muscular tension or relaxation, adrenalin count, etc.

It should be clear that music is a particularly human type of activity which lets us mix elements from all the domains just mentioned into an integral whole. It is an activity that allows us to represent combinations of signals from its constituent domains in symbolically rather directly interactive terms. It lets us engage in inter-personal activity on many levels, either by making music together or by responding to it. To express ourselves on all these levels, we humans do not need to confront each other with symbolically unmediated emotional, motoric or physical actions: we can use music instead. In other words, music enables relatively risk-free action by providing socio-culturally regulated forms of potentially risky interaction between humans. It also helps our cognitive flexibility, i.e. the ability to mix, switch and correlate across different domains of representation. Viewed from this perspective music may well be, as Cross (1999) concludes, `the most important thing that we humans ever did'.