The Benefits of Music

Remember "The Mozart Effect"? The 1997 book by Don Campbell, "The Mozart Effect: Tapping the Power of Music to Heal the Body, Strengthen the Mind, and Unlock the Creative Spirit", discusses the theory that listening to Mozart (especially the piano concertos for some odd reason) may temporarily increase one's IQ and produce many other beneficial effects on mental function. Campbell recommends playing specially selected classical music to infants, in the expectation that it will benefit their mental development. These theories have been largely disproven. I always wondered, why Mozart? Why not Bach? Or Beethoven? Or Debussy, for that matter? The answer is the very mundane one that a French researcher just happened to pick Mozart for his research.

Research continues however and some recent studies seem to show that while music lessons don't develop transferable skills, playing video games does. Wow, really? A number of researchers have been looking into claims made that music develops all sorts of transferable mental skills and found that while there may be correlation (kids who take music lessons also happen to have other mental skills), there is no case to be made for causation. Here is a blog post that gives a pretty good summary.

Much of the literature makes the mistake of inferring causation from correlation, and fails to control for confounding variables. Glenn Schellenberg, a psychologist at the University of Toronto who studies transfer from music specifically, has new, not yet published work showing that the association between music lessons and cognition disappears when demographics and personality are held constant. In other words, the apparent benefits from music lessons have more to do with which kids take music lessons than they do with the lessons themselves. In addition, studies of adult professional musicians show no cognitive benefits over comparable professional non-musicians.

This causation/correlation problem is the elephant in the room in so many statistical studies. Statistics are really incapable of distinguishing the two, which is why it seems that every conclusion about, for example, what diet is good for us, is overturned a decade or two later when different correlations are discovered. My sense is that unless you have an idea about why such and such is happening, you really don't know much.

When I think about it, while it is certainly tempting to say that everyone should take music lessons as there are real, lasting benefits from doing so, this is probably just not true. Some people get a lot out of music lessons, while others get very little. I used to say that the discipline that you need in order to study music successfully is a positive benefit, but the truth is that if you really lack discipline, you will continue to lack discipline when you study music and not get very far. If you have the potential to learn a disciplined approach, then music may offer the opportunity to develop it. But that's about it.

But of course music offers profound benefits, if only to those who have the ears for it! I think the aesthetic benefits are primary, as they should be for any art form. The beauty and sublimity of music is a good and an end in itself. You either get it or you don't. Studying music as a child may well open that door for you. Or not. It is certainly worth a try. But all these attempts to show that music has or has not the ability to make you smarter or a better person or something are really rather wrong-headed, aren't they? We don't expect the practice of doing water-colors or writing poetry to improve us in these ways.

The ongoing pressure to demonstrate scientifically the value of various traditional aesthetic pursuits such as music is an artifact of the attempt to put everything to the scientific test. Understandable, but often what Gilbert Ryle called a "category error". Statistics and science have very little application to the aesthetic appreciation of music. We should just accept that.

Now let's listen to some Debussy. Here is "Nuages" (Clouds) from Nocturnes for orchestra: