Sunday, August 21, 2011

Music and Emotion

Philosophers talk about emotions as having objects. If you are angry, you are angry about something or at someone. If you are sad, you are sad about something, the same with happy. They contrast object-directed emotions with moods, which don't have an object. Moods are things like cheerful or gloomy feelings that do not relate to anything in particular. This is a simplification, of course. Here is an article with more detail--see especially section 3 on emotions and intentional objects.

What does this have to do with music? We sometimes talk about music as being sad or happy. There is an anecdote about Schubert that applies. He had played at a salon in Vienna and afterwards someone came up and asked him why his music was always sad. He replied that all music is sad. Perhaps true of Romantic music, but certainly not of Classical or Baroque music. I want to propose that music is neither happy nor sad, nor does it really depict other ordinary emotions. The reason is those objects: music has no object. If there is a piece of music that is slow, languorous, gloomy, we may well talk about it as being 'sad', but it is sadness without an object, therefore a mood, not an emotion. Wagner called the first movement of Beethoven's string quartet in C# minor, op 131, the saddest thing in all music. Here it is:

If it is truly sad, then why do we enjoy listening to this and similar music? I think the answer is that it is only 'sad' by metaphor. This, and all music without any text, is really only 'about' musical beauty. We enjoy beauty in music. Sometimes that beauty comes in slow, languorous forms and sometimes it comes in bright, sparkling forms. Here is the very next movement of the quartet:

Similarly, there is no such thing as 'angry' music: music doesn't make us angry and composers are not angry when they write it. This is why I am always a little leery of attaching much importance to biographical details as a guide to what a piece of music 'means'. Music, again, music without text or lyrics, doesn't mean anything in the usual sense. Perhaps I should say "most music" because there are some interesting exceptions. Sometimes composers have been known to code meaning into a piece through the use of either notes that refer to someone or something, or to construct music with some numerological significance. This is interesting, of course, but it is essentially an extra-musical ploy or reference.

One famous example is Shostakovich's coding of his own initials into several pieces. When speaking the names of the notes in German, E flat is 'es' and B natural is 'h'. Therefore DSCH stands for Dmitri Shostakovich and corresponds to the notes D, E flat, C and B natural. Here is the opening of his Eighth String Quartet, which is permeated with this figure:

Bach did the same at the end of the Art of Fugue with a new subject using the notes BACH. Another example from Bach is the first fugue, in C major, from the Well-Tempered Clavier that uses a subject of exactly 14 notes. B = 2, A = 1, C = 3 and H = 8 for a total of 14. One musicologist claims that the Mass in B minor, also by Bach, uses numerology as well. There are, apparently 2345 measures in the whole piece and the only repeated music, the Dona nobis pacem, is based on a theme that begins with the 2nd, 3rd, 4th and 5th scale degrees. Well, maybe... In any case, these coded references are essentially non-musical and while they may be significant, it is something apart from the fundamental musical structure.

But music can have a dramatic effect on us, making us tap our foot, move our head, breath more heavily, even tingle with goose-bumps. How does it do this? I don't think we really know, though in a neurological lab at McGill University, they are looking at the brain to see if they can find out. I think I prefer to think of it as a mystery: how mere vibrations in the air, impinging on our ears, can have such a profound effect.


Jon Silpayamanant said...

Ah, the glory days of Western Philosophical treatments of musical meaning and aesthetics! Brings back memories (especially about my undergrad thesis).

I think the idea of coding letters (or names or words) in music isn't really a whole lot different than coding those things in language. Basically we're just assigning specific lexical items to a particular sound. That these musical sounds have definite pitches that uniquely define themselves in musical space rather than other phonological qualities isn't really that much different than how tonal languages use lexical pitches to determine meaning.

Th idea of musical meaning (for it's own sake) gets closer to that formalistic approach to analysis that Europe (and eventually America) adopted due to the success of science. Mathematics (through Logical Positivism) and Philosophy (through Analytic Philosophy) were those field's equivalents to Western Serial formalism at its height. But I'm not so sure how much meaning the idea of 'musical meaning' even has.

I think the idea of coding emotions might have some truth to it--it seems like those who are studying the borders of music and language (cf. Brown's 'Musilanguage' hypothesis) have zeroed in on both being forms of communications. They just happen to be more efficient at communicating certain things and those things might not overlap.

I think that's why there are still great aesthetic traditions (outside of the West) that do talk about music transmitting (primarily) emotional meaning. The Indian theory of 'Rasa'; the Arabic aesthetics of 'Tarab'; and hell, even Western music has some roots in the Greek 'Doctrine of Ethos.' I think Western Art music has gotten so far along the formalistic path that it crowded out the richness of some of these older (and in most cases more universally recognized) theories of what music is about.

As a result, just as in lost or dead languages, the community of those who understand what or how the form of communication works gets lost. Indians have no problem accepting that this raga or that raga is supposed to evoke this or that mood or emotion because it still does evoke those things in that population, and with a higher than chance occurrence, same with Arab musicians and what the maqams are supposed to evoke. Control groups (often Americans) perform at chance level.

I guess we shouldn't mistake the means of communication with the message being communicated, right? And maybe, since we've gotten so relatively fragmented as a culture that 'hegemonic' musical meanings just can't exist anymore since there is no common music-speaker community.

Bryan Townsend said...

Hi Jon,

Thanks for making a fascinating comment! I just had a look at your blog and your music-making spans an amazing range of styles. Thanks for the pointer to Steven Brown's theories. Hadn't run into that before.

One thing that you allude to, that I didn't discuss, is the development, within a genre or repertoire, of semantic associations of certain musical gestures. In opera there is the sighing motif of the minor sixth, for example. Another is the descending chromatic bass line that is often a lament. But my feeling is that what the composer is doing, in detail, in the piece, is more important than these associations.

Jon Silpayamanant said...

You're welcome, Bryan! I found your blog through some of the comments you've been making at Greg Sandow's blog--really enjoy reading what you say over there so popped over here!

And you're absolutely right about the semantic specific musical gestures. We could probably include light motifs (ala Wagner) and idea fixe (Berlioz) within this category. And maybe the idea of 'text-painting'--using specific musical contours to represent specific real world referents (e.g. flutes for birdsong; Smetana's usage of melodic movement to represent the Moldau; etc.). Maybe Messiaen's very specific and idiosyncratic musical spelling.

I think that on the whole, though, Western Art music has moved away from the whole Ethos Doctrine idea in its celebration of the individual 'author' (i.e. composer) of musical works and as a result we have various trends toward the usage (if there are any) idiosyncratic musical meanings--I think Messiaen is just one particular extreme case of that.

So there's a split between all these more programmatic musical treatments and so-called 'absolute music' which has become the default 'music language' which, as I mentioned is paralleled in so many other aspects of Western Culture. Serialism (and in some ways, Minimalism) became the logical endpoint of that direction where even the musical meaning is nothing more than a concrete realization of more abstract structures.

If you get a chance, pick up the "Origins of Music" which Brown edited--a fascinating collection of articles by composers, theorists and scientists and is a great introduction to the growing fields of Biomusicology and Evolutionary Musicology. There is a piece by Brown that lays out the foundations of his idea of Musilanguage in it!

Bryan Townsend said...

Hi Jon,

Thanks for the link to The Origins of Music. I will have to read that. Looking at the description I was suddenly struck that this reminds me of the big 18th century argument over the origins of language...

Music with text always can have a narrative, but instrumental music has always posed a problem. Ever since the Renaissance it has made allusions with bird calls and so on. There is even a Baroque piece that purports to depict a kidney operation! And of course, the examples you mention. But I take the view that instrumental music with no text, while it can quote, allude, gesture and so on, cannot actually tell a story. Though Berlioz certainly gave it a shot!