Hacker News new | ask | show | jobs
by mnhn1 1665 days ago
> people never musicalize conversational speech

I found this assertion confusing. Speech in say, English (just to narrow it down), is musical inherently. The pitch and rhythm of a speaking voice, especially how they change while an individual is speaking, is meaningful meta-information about what is being said, and how the speaker feels about it, or what they mean by what they say.

We _absolutely_ use the music of everyday speech to create emphasis, for example, and in many other ways. Actors use pitch and rhythm when speaking to convey emotions. It is often the musical content of speech that makes or breaks the performance. Mismatching speech melodies with the dialog sounds all wrong.

I dunno, 10 years or so ago I wrote a whole MA thesis on this exact subject. It's long enough ago (and I've not stayed in the field I was in) that I've long forgotten many of the sources, but there is plenty of stuff out there that deals with speech and music.

To veer off into pure opinion: I definitely think music gains some of its emotional impact by virtue of its relationship to speech, given that we can interpret so much from the music of speech itself, and if that kind of metadata is presented _independently_ of natural speech, there's often something pleasing or interesting about about that. We also are super good at listening for other meaningful sounds though, like things that might kill us.

My two cents is that music, like other art and things like sports or games, leverages senses, instincts, and skills that evolved initially for other purposes, and uses them recreationally, playfully. To varying degrees, humans seem to like stimulating and playing with their senses in different ways.

I'm not convinced that the attempt to explain it the way the author does is worthwhile. Parts of it ring true and parts of it (like the role of discerning truth and the claim that people don't musicalize their speech) I think are off track, and maybe also constrained by a far too limited perspective of what music is in the first place.

1 comments

You're right, but you may want to consider that highly monotone communicators may not actually notice most of the non-literal signals that get passed in typical social situations.

Models be built from something, something related to the modeler's interpretation of their own interface into reality.

That said, it's nice that such a model was made; it's a nice reference / jumping off point. Someone more sensitive to their percepts and the nuances of life would be hard pressed to formalize any model at all; they'd be hard pressed to unfocus from the complexities and responsibilities of social life to do the abstract work of modeling.

Sure. There's also the converse situation where people who speak in tonal languages might not develop associations with melodic patterns in speech and any strong meta-meaning, since in those languages, pitch is actually carrying top-level meaning.

People who have congenital amusia, iirc, also tend to struggle with understanding tone languages. It seems true that if you aren't able to distinguish pitch well, you aren't good at encoding/decoding messages that are present through the medium of pitch differences.

I didn't mean to suggest above that this is a general thing for all humans. Like every other recreational stimulate-the-senses activity, some people don't actually care for it/aren't affected by it anyway.

I see. Could you say more about meta-meaning?
I'm using that to group all the things that are communicated through the musicality of the voice in addition to the words themselves. Like if a person is happy, upset, or surprised, in a way you can pick up from the tone of their voice.

Here's a small example, about how the minor third interval between two notes which is perceived as "sad" in Western music, is also present in English speech that sounds "sad" to the listener: https://www.researchgate.net/publication/44642274_The_Minor_...