Saturday, June 14, 2008

Mozart and Salieri

Can you tell Mozart's music from Salieri's?

I scored a slightly embarrassing 60% on the quiz: the excerpts are a mix of opera arias and piano concerto movements by both composers. Two of the Mozart excerpts I knew already, but for the other questions I had to guess based on my knowledge of the composers. My theory was that the Mozart excerpts were going to be less predictable, with more unexpected harmonic shifts and asymmetrical phrase lengths, and the Salieri excerpts would be more old-fashioned and foursquare-sounding, but the results did not bear this out. (In my defense, one of the Salieri excerpts I missed was written in 1795, four years after Mozart died, and one of the Mozart excerpts I missed was from a rarely-performed early concerto, written when he was in his twenties. So it would make sense that the late Salieri work would sound more like the mature Mozart, and that the early Mozart work would sound more like the conservative Salieri.)

The Reverent Entertainment site contains many quizzes such as these - one which asks you to tell the difference between MIDI performances and performances by virtuosi (yawn), another which asks you to distinguish between abstract art masterworks and fakes made in Paint (pretty easy), and another which asks you to distinguish between prose passages by Dickens and Bulwer-Lytton (surprisingly difficult). Good for an hour's entertainment, but what's this guy's angle? Turns out that he's gathering results for statistical articles, with titles like "Scientific Evaluation of Charles Dickens":
Are the very famous writers different from the obscure ones?

Edward Bulwer-Lytton is the worst writer in history of letters. An annual wretched writing contest was established in his honor. In contrast, Charles Dickens is one of the best writers ever. Can one tell the difference between their prose? To check this I wrote the Great prose or not? quiz. It consists of a dozen of representative literary passages, written either by Bulwer-Lytton or by Dickens, and the takers are to choose the author of each quote.

The distribution of the scores received by over three thousands quiz-takers is shown in Figure 1.The average score is 5.74 or 48% correct. Due to the large number of quiz-takers the standard error of this average is small: 0.035 or 0.3%. . .

On average, a quote from Bulwer-Litton was selected as Dickens (or great prose) by 52% of quiz-takers, while a quote from Dickens was selected as Dickens by only 48%. Does this mean that Bulwer-Lytton is a better writer than Dickens? Probably not. Table 1 shows for every quote the fraction of people who attributed it to Dickens. This fraction varies between the quotes with the lowest being 36% (No. 9) and the highest 74% (No. 12). This suggests that a different selection of quotes could lead to a different average score. For example, if we remove the most Dickensian Bulwer (No. 12) and the most Bulwerian Dickens (No. 10), and recalculate the scores based on 10 remaining questions, - the average score becomes 51%. . .

The results of the quiz show that people can’t appreciate great prose when the name of a great writer is detached from it. The answer to the question, we started with, is: Yes, they have more readers.
Very interesting, but I can't help wondering if this statistician realizes the flaws in his methodology. In each of the quizzes I've looked at, the questions are designed in such a way to make the "genuine article" and the "fake" look as much like each other as possible. Mozart and Salieri are both represented by 15-second clips from extended works, when everyone knows that to take the measure of an extended work requires you to hear the complete piece in context. (The similarity of the two composers' styles is also a problem: a person who's studied eighteenth-century music well enough to tell the two composers apart has probably heard at least some of the Mozart excerpts before, which invalidates the survey.) In the Dickens and Bulwer-Lytton quiz, both writers are represented by long and rather purple descriptive passages in a nineteenth-century style that most readers will find foreign to any prose they are likely to encounter today. And again, anyone with a wide enough experience of Victorian literature to distinguish between two stylists is likely to have read the Dickens novels involved. The Modern Art quiz is the most flawed, because each of the "fakes" in the quiz is obviously designed to look like an authentic Kandinsky, Mondrian or Rothko. To identify the difference between the real and fake paintings, therefore, requires you either to have seen the paintings before or to recognize the difference in image quality between the canvas paintings and the computer-generated forgeries.

Unfortunately, this interesting site appears to be a front for someone with a cynical Agenda, which is revealed by some of the other quiz titles: Faulkner or Babelfish Translation (ouch!), Jackson Pollock painting or Bird Droppings (double ouch!), and by a tutorial on how to create abstract art in five minutes using Microsoft Word. In other hands, quizzes like these could show that even the Salieris of this world are capable of writing with grace and charm in the style of their time; instead, they're being used to uphold someone's nihilistic conviction that art is bunk and that the great masterpieces of our culture are worthless.



Allen H Simon said...

We have a "great man" view of history, a celebrity culture which thinks that the "great" are 100 times better than the "good", whereas in reality the great are oftentimes only 10% better, or only better after many repetitions. The music of JCF Bach, or Michael Haydn, or Salieri, is perfectly entertaining -- it's only after hearing it the tenth time that you realize its lack of depth in comparison with Mozart. Certain composers have celebrity status, but the idea that there's a chasm between them and the second-rank composers is a media-generated mirage.

Osbert Parsley said...

A valuable point, I think. Even Bulwer-Lytton, who has a reputation as a laughably awful, turgid writer, isn't that much different in style from Dickens; the difference is that his writing style crosses a line of bathos that Dickens only occasionally skirts.

Even the Mozart "brand name" only lives up to its reputation for the works written near the end of his life. I remember sight-reading through a book of his early minuets (K. 2), and they were absolutely awful. No surprise, I'm sure - he was about six when he wrote them. But he had to learn by trial and error in order to reach his mature style, just like any other computer.


"But he had to learn by trial and error in order to reach his mature style, just like any other computer." Mozart was a computer?

To Allen's point, I've always thought of this as the capitalist value system at work - the top of the bell curve is in much shorter supply, so supply-and-demand means its value is out of proportion. We certainly see this in the world of performers, where a few "brand name" soloists can command exponentially more than musicians of comparable skill and talent. Still, I think Mozart's got JC Bach by more than 10% (whatever tha means), but maybe that's because he was apparently a computer.

Osbert Parsley said...

Gaaaaaaaack! Of course I meant to type "composer". Although if he was a computer, his incredible facility would make a lot more sense. I'll let the conspiracy theorists take it from here.