The Temptation to Define
One asks the question:
This question invites us, indeed tempts us, to give an answer.
Most words in most languages can be defined in terms of other words, and typically those definitions don't have to be terribly long.
So, surely, if "music" is a word, then we must be able to give a definition for it.
This optimistic assumption has led to a veritable industry of people giving all sorts of definitions of music – see https://en.wikipedia.org/wiki/Definition_of_music for a list of some of them.
To give a "popular" example (which seems to be popular based on how short it is), as credited to the composer Edgard Varèse:
I will consider this example below, in my section on the Non-Refutability of Bad Definitions of Music.
How do we know what "music" is?
Most of us know, in some sense, what "music" is.
(I qualify this with "most", because there is a percentage of people who, it turns out, don't really "know" what music is, and who only know of music on a second-hand basis, as something that other people experience and talk about.)
Every person who knows about music knows about it because of two things:
- A set of known examples of music.
- That person's own subjective experience of listening to a subset of that set of known examples.
This is how individual people "know" what music is.
That is, we only know about music subjectively.
Our objective scientific ignorance about music
One might think that there is a scientific understanding of music which goes beyond the personal subjective experience of listening to a specific set of examples.
But, actually, there isn't.
It is worth spelling out exactly how much we don't know about music:
- We don't know of any objective criterion to distinguish what is music from what is not music.
- We don't know of any process or algorithm to reliably generate new good examples of music.
- We don't know what processes occur when someone listens to and enjoys music.
- We don't know what biological function music serves, if any.
- We don't fully understand what determines differences in different people's preferences for music (although we know that cultural exposure plays a significant role).
It may seem unbelievable that everyone, scientist or not, can be so ignorant, on so many levels, about the basic nature of something that most of us are very familiar with.
But that is how it is.
The only correct definition of "music" is one that gives an actual list of examples, and then tells the person reading the definition to listen to those examples.
This kind of definition is nothing like the kind of definition that people expect when you say you are going to "define" something, and it is not the kind of definition you can put into a dictionary. (You could if you really wanted to, but in practice noone does.)
The Non-Refutability of Bad Definitions of Music
One problem with the subjectivity of our knowledge of music is that it is quite impossible to objectively refute any proposed verbal definition.
For example, someone might claim that music is "organized sound".
To refute this definition, all I have to do is the following:
- Create a sound recording consisting of some sounds which might be considered to be "organized" in some fashion.
- Listen to that recording.
- Observe that I do not perceive that recording to be musical based on my own subjective response to that recording.
Subjectively, I have refuted the definition.
But how can I prove to someone else that I have found a counter-example?
An advocate for the "organized sound" definition could just claim that my counter-example actually is music.
And there is no way I can objectively refute that claim.
The most I can hope to do is find some largish group of people who agree, subjectively, that my counter-example is non-musical, and take this as proof that it is a valid counter-example.
But even then my refutation will fail, because it can always be claimed that "what is music to one ear is not music to another".
Moving forward ...
Where does this leave us?
The only way to make progress is, I believe, to shift the burden of proof.
As individuals, interested in studying and researching the mystery of music, we need to stand up for the right to reject all definitions that can be subjectively refuted.
If a definition fails to describe music in a manner that corresponds to our personal subjective experience of things that are music and things that are not music, then we should reject that definition.
And the best definition is ... no definition at all.
Which means, in practice, given my own personal subjective experience of music, and given my own understanding of all the attempted "definitions" of music that I have ever read, every verbal definition of music that has ever been given is actually wrong.
Whatever it is that music is, none of those definitions comes close to actually telling me what music is.
Also, just for the record, I can't give you a definition of "music" in so many words. The most I can do, as explained above, is give you a list of examples, and then tell you to listen to them. Even then I have to assume that there is enough overlap between my musical taste and your musical taste that you will then "know", from listening to my list of examples, what I mean when I use the word "music".
I may have lots of crazy speculative ideas about what music is (read the rest of this blog to see what those ideas are), but I am not in a position to give you a verbal definition of "music" any more than anyone else is.