Easy to Consume, Difficult to Produce
Music is something that people produce, and it is also something that people consume.
But there is a fundamental asymmetry between the production and consumption of music, which is that anyone can consume music, but not everyone can produce music, or, to be more precise, most people can produce music, but the music they produce would be of such poor quality that other people would not choose to consume it.
Most of the time, the production of music can be divided into two distinct stages or components: composition and performance.
Both of these are difficult, ie, composition is difficult, and performance is difficult.
The difficulty of performance is typically overcome by the relatively straightforward but arduous strategy of engaging in years of practice.
There is no known strategy to overcome the difficulty of composition: some people compose music, and some people never compose music. And even people who compose music may only ever compose one item that is considered worth listening to by most other people.
Variation In Agreement
We might wonder if the difficulties of composition and performance are truly different things.
One relevant observation is that people agree more about which performances are the best, and they agree less about which compositions are the best.
This distinction suggests that two very distinct "sub-modules" in the brain are responsible for evaluating these two aspects of musical production.
At the same time, our brains must combine the evaluation of these two different aspects in a manner that is closer to a multiplication (of scores) than an addition: it is not sufficient to listen to a favourite song played badly followed by someone else's favourite song played really well – what we really want is to hear our own favourite song played really well.
The Difficulty and Competitiveness of Music
Because music can only be evaluated subjectively, the concept of "difficulty" in relation to Music needs to be defined relative to the criteria that listeners set when listening to music, if we are to avoid claims that "anyone can produce music".
In practice, the production of music is highly competitive, and most people will only bother to listen to music that is as good as the best music that they have access to.
There are some situations where people will make some allowance for the circumstances, for example when listening to live music down at the local pub.
There are also situations where people will politely listen to music that they don't really want to listen to, but those situations are all about the politeness, and are relatively infrequent.
Nevertheless, most of the time, for most listeners, music is only "good enough" if it is also "as good as possible".
Two Lesser Difficulties: Loudness, and Multi-Part Music
The difficulties of composition and performance are the major difficulties relating to music, and if you can compose and perform to a level where a significant number of people choose to listen to your music, then you will be regarded as a fairly brilliant musician.
There are two additional aspects of music that can be considered to make the production of music more "difficult" to some degree.
The first of these lesser difficulties is that of loudness. If we like music, then we usually like it even more if it is loud. Prior to modern technologies for amplification, making music loud would actually have required some degree of effort, either from the individual performer performing their instrument or singing more loudly, or by coordinating the performance by a group of performers.
A second lesser difficulty is the desirability of multi-part performance, ie multiple singers, multiple instruments, or both of those at once. As with the difficulty of loudness, achieving multi-part performance requires either extra effort from the individual performer, or coordinated performance of a group of performers.
Technology
There are many technological solutions that have been developed to help with the composition, performance, dissemination and playback of music.
Most of these solutions are fairly recent, especially compared to the time-scale of human evolution, and the existence of these technologies somewhat obscures the intrinsic difficulties of producing music.
For example:
- The existence of recording technologies means that the very best compositions and performances can be created, perfectly reproduced and then distributed to almost anyone in the world. Which means that everyone has the option to only ever listen to the highest quality music, and no one ever needs to listen to something "second-rate" if they don't want to.
- Technologies like AutoTune can convert fairly good performances by the best available performers into consistently near-perfect performances.
- Making the music louder is as easy as turning the knob on your amplifier. This is so easy that we employ public servants whose full-time job it is to deal with complaints from people about their neighbours who play music too loudly.
- Technology also helps with multi-part performance – if you can't actually sing and play at the same time, or your group is not quite good enough to play all their own parts consistently and brilliantly and simultaneously, then multi-track recorders can be used to record different tracks separately, fix up or replace any defective recordings, and then merge the individual parts to produce the final high quality result.
In earlier times, when these technologies did not exist, the intrinsic difficulties of making "good" music would have more strongly constrained how often people could listen to music, under what circumstances they could do so, and how good any of that music would be.
Music and Language
Music is somewhat similar to Language, even though, at the same time, it's not actually a language.
One major difference between Music and Language is that Language has an obvious purpose, which is to communicate, whereas Music appears to have no obvious purpose beyond the purpose of being musical.
Given the apparent relationship between Music and Language, and the relatively obvious function of Language, a common approach to the Mystery of Music is to ask questions about what is the precise relationship between Language – the thing that we know what it is – and Music – the thing that we don't know what it is.
Another difference between Music and Language, which may help to throw light on this relationship, is the Difficulty of Music.
In some senses Language can be considered to be "difficult". For example, it is very difficult for non-human animals to achieve even a very limited competency in the production or consumption of human language. And if one attempts to create a machine to produce and a consume human language, ie by writing software, that turns out to be severely non-trivial.
But at the level of the average human producer of language, language is not difficult. Almost everyone learns to be a competent producer of language. And even though full language acquisition takes a while, no special "motivation" or conscious persistence is required to get there.
The non-difficulty of human language production is related to the standards set for human language consumption. As consumers of language, we do not demand the "best possible" composition or performance of language. Rather, we require language production to be "good enough" for its purpose, which is to communicate.
I will advance the hypothesis (in more detail below), that if the requirements that we apply, as consumers, to Music, were similarly applied to Language, then Language would cease to be usable for its primary function, ie as a system of communication, and it wouldn't be Language any more.
The Sexual Selection Hypothesis
One hypothesis advanced about the purpose of music relates to sexual selection, specifically that human males use their musical prowess as a means of impressing and attracting female mates.
This was originally suggested by Charles Darwin, and has more recently been advanced by evolutionary psychologist Geoffrey Miller, who gives the example of Jimi Hendrix as a famous musician who had sex with hundreds of females (chapter "Evolution of human music through sexual selection" in the book "Origins of Music", edited by Nils Wallin et al).
Sexual selection does seem to provide a promising explanation for the observed difficult of music.
The weakness of the Sexual Selection hypothesis is that most known instances of sexual selection involve requirements, where the selected partner (usually the male) must satisfy the requirement being selected for if the other partner (ie the female) is to choose them as mate.
For example, it's not just that a peacock might have more sex if it has blue feathers and a fancy tail, it's that the peacock won't have any sex at all if it's not the right colour or doesn't have the right sort of tail presented in the correct manner to the female.
Musical prowess may be one way for a human male to attract a female, but it is not the only way, and it is certainly not necessary. Without necessity, the selective pressure is relatively weak, and the hypothesis is less convincing.
To put it another way, if we are to believe that sexual selection explains music, we must believe that millions of fans of Rock and Roll have to listen to the music, so that just a few hundred star performers can attract a few dozen or (if extra good) a few hundred mates. Which is a lot of work by a lot of people for relatively little benefit.
The Group Hypothesis
Another theory about the purpose of music has to do with group performance.
Of the four difficulties of music I have identified, at least three of them can be overcome, to a greater or lesser degree, by means of group performance:
- The performance quality of a group is typically better than that of an individual, because any errors are averaged out. So a group of moderately in-tune singers can match the singing quality of one really good individual singer.
- A group of performers is obviously louder than an individual performer.
- A group can more easily perform multi-part music.
Given these effects that group performance has on the quality of the music, people do make the effort to get together in groups and make music together, more so in circumstances where modern musical technologies are not available.
One version of the Group hypothesis is that music is beneficial simply because it encourages people to get into groups and engage in the coordination required to do a group performance. The assumption is that it's a good thing just for people to get together and do stuff together, even if the "stuff" they're doing is otherwise pointless.
The weakness of this hypothesis is that the only things the group is doing while performing music is that they're performing the music and they're listening to the music.
Also, the effects of music on mood, which might contribute to a feeling of "togetherness", tend to cease fairly quickly once the music stops.
If there was some benefit to people participating in group activities (which there is), then we might expect the evolution of a simple desire to be in groups, and there is no reason for such a desire to be confounded with the desire to engage in a specific activity which is otherwise pointless.
The Disabled Language Hypothesis
I will now advance a hypothesis of my own, about the relationship between Language and Music.
My claim is that Music is like Language, and has indeed evolved from Language, but now serves a purpose distinct from Language, and this purpose does not involve communication. Furthermore, Music has evolved features so that it can't function as a communication system in the way that Language does. In effect Music is Language that has been disabled.
Under this hypothesis, it is likely that some of the "difficulties" of Music are the results of constraints which have evolved as part of the criteria for musicality.
In particular:
- Language has syntax, and any competent producer of a language can produce an unlimited number of language utterances which are valid according to the syntax of that language. Music has its own "syntax", but it is not a syntax like that of a spoken language, and it does not allow original and "valid" utterances to be easily and freely produced. If original and valid utterances cannot be freely produced, then Music is not usable as a communication system.
- To function as a communication system, Language has to be something that anyone can produce, to a level of quality which is satisfactory for any other person to "consume". With Music, only highly practised performers produce music good enough for most people to listen to. You can't use Music to communicate with other people if other people don't want to listen to you.
The Mental State Hypothesis: Music as a Signal
The Disabled Language hypothesis tells us that Music is not a system of communication, but it doesn't tell us what Music is.
If Music is not a system of communication, then all the content of music, complex though it may be, has no intrinsic meaning.
This leads to the hypothesis that Music is an arbitrary signal, and that the purpose of the human response to Music is entirely contained in the effect that Music, as a signal, has on the listener.
There is evidence that a primary effect of music is its effect on mood, and I would go further and state that Music induces a specific mental state which is peculiar to Music.
The exact nature of this mental state is somewhat elusive (even though it's something that we all experience), but the following aspects can be identified:
- A tendency to partially disconnect from immediate reality
- A tendency to more fully experience emotions associated with thoughts unrelated to immediate reality (eg "daydreams")
To relate this hypothesis back to Language, we can suppose that some component of the processing of acquiring a first language, in the first months of life, involves a similar mental state, which occurs in response to language-like sounds. This mental state serves a useful purpose in the language acquisition process, but once a certain stage of language acquisition is reached, that mental state is no longer useful, and no longer occurs.
This effect on mental state is part of a human "language acquisition instinct" which evolved at some early time in the evolutionary history of the human species.
Then, at some later time in the evolution of the human species, this same mental state proved beneficial, as part of dealing with life in general, even apart from the process of learning language. As a result, Music evolved as a pseudo-language which continued to evoke this altered mental state, throughout a person's lifetime,
Over time, Music evolved characteristics that prevented it from being confused as a learnable language.
Apart from this requirement to prevent Music from being processed as an actual language, there are a couple of additional reasons why the difficulty of music might be beneficial, which I present here as additional hypotheses:
- The Limitation Hypothesis
- The Group Safety Hypothesis
The Limitation Hypothesis
Given that a person can choose to listen to music at any time, the possibility arises that people might choose to listen to music all the time, so that they are always in the altered mental state induced by music.
The mental state induced by Music involves partial disconnection from immediate reality, and it would not be good to be in this state all the time.
The Limitation hypothesis is that some of the features of Music exist to limit how much time anyone can spend listening to music, and the difficulties of music are one way that such limitation can happen.
The Group Safety Hypothesis
Music induces an altered mental state, which includes a desire to partially disconnect from immediate reality.
This altered mental state does leave the individual music listener vulnerable to attack, for example from other hostile individuals.
One way to mitigate this risk would be to only listen to music when in a large group of other (presumably friendly) people.
So one reason why music has aspects which encourage or require group performance is that it encourages the listener to only listen to music when in the relative safety of a large group, and not in a more vulnerable situation by themselves.
This hypothesis is a bit different to other "Group" hypotheses, because the assumption is not that Music encourages group performance because group activity is generally a good thing, but more specifically that Music encourages group performance because being in a group while listening to music is a good thing.