Wild gorillas compose songs that they hum during meals | Hacker News Reader

Wild gorillas compose songs that they hum during meals

newscientist.com

163 pointsTurukawa37 comments

I've always hypothesized that humming while eatting (or saying "Mmmmm") is the perfect pitch to vibrate the hummer's tounge, which makes the flavors experienced more pronounced.

It would explain why we say "Mmmmm" not something higher pitched. Plus you can feel your tounge moving when you say "Mmmmm".

Any thoughts?

"m" is a consonant. What does it mean to assign it a pitch?

"m" is a nasal sonorant. It's assigned a pitch the same way nasal and oral vowels are- based on the positions of its major formants.

Or, less technically, when someone successfully hums a non-monotone tune- you have just heard what it means for them to assign a series of differing pitches to "m".

"m" is a voiced, nasal consonant. It can be pitched as much as any vowel.

In this context, it means nothing. As you point out, GP is conflating pitch with pitchless phonemes, for example, the phrase "we say "Mmmmm" not something higher pitched" doesn't make any sense.

However, I suppose GP meant to write "we tend to say "Mmmm" with a specific, relatively low pitch", a statement with which I don't particularly agree. I would rather say it's done at an average conversational pitch, or even at a higher pitch since it often has a pitch accent conveying surprise. Which I believe is the main purpose of saying "mmm": communication, not vibrational tongue massage. Consider that when you say "mmm", you voluntarily choose a pitch intonation according to what you're trying to convey, ie. how pleasantly surprised you are.

You can do Fourie transform to any sound, and if you see that certain frequency and it's modulations have a prevalence over all others, then you can talk about this sound's pitch.

https://en.wikipedia.org/wiki/Pitch_detection_algorithm

https://en.wikipedia.org/wiki/Fundamental_frequency

Just one: I don't feel my tongue moving when I "Mmmmmm".

Not every culture says "Mmm".

I actually thought they did! I'd love to hear about what cultures dont.

Chinese say 好吃！ (hao chi, literally "good eatin'!")

Thai say aroy ("tasty!") aroy mak ("very tasty!") or perhaps, if memory serves, gin dee ("good eatin'")!

'Mmm' is a very western sound I think. Probably not that historically established either, I wouldn't be surprised if it was a post broadcast media development. Remember how uptight the Victorians were. Perhaps it's French?

Maybe it's the easiest noise to make with a closed mouth?

Can anybody really hear anything in those samples? They seem random noise.

I couldn't hear anything.

More samples would be cool.

The word "happy" is in the title but has no support in the article.

Are we talking about music that was inspired by or inspiring the happy mood? If it's the latter, I would say it's true since eating is one of the better experiences you can have, especially as a simple animal.

Ok, we replaced "happy songs" with "songs".

HN readers are like sandpaper for titles.

The Ku Pteleopsis Hylodendron Humming sounded like C D F D D to me. Very deep, possibly octave below left-most C on a piano, but for those saying it's just noise I'd say there's some signal there.

The Bal Gilbertiodendron Dewevrei Singing clip I can't pick anything out. Bit too quiet to tell.

Programmers also do that. This is somehow linked to survival in the long run.

> The other was singing – a series of short, differently pitched notes that sounds a little like someone humming a random melody (listen to clip below).

> “They don’t sing the same song over and over,” says Luef. “It seems like they are composing their little food songs.”

Right. A "random melody". Also known as "random noise".

Is it so implausible that these creatures who have displayed startling emotional depth (Koko and her pet cat, anyone?) might also display musicality?

Why are we so quick to dismiss the notion that anything not human might do something creative, like hum a happy little tune to itself?

Oh, no. That's just "random noise".

The grandparent might be specifically talking about the sound clips from the article. They don't sound melodic, let alone musical (at least to my human ears).

I believe it's more of a semantic thing. A melody is inherently non-random. "Random melody" is nearly oxymoronic.

I think it's fair to call a non-repeating sequence of randomly selected, sufficiently tonal notes, on a familiar scale, in the right frequency range, a melody. So I don't think it's fair to say a melody is inherently non-random. I'd say that there only needs to be a sequence of sounds identifiable as individual notes, with a sufficiently dominant pitch as to sound tonal.

In other words, random within a non-random subspace of the universe of possibilities. Like hitting random notes on a piano, without having a piano to start with.

As someone who has composed aleatoric music and had many discussions with people who have devoted their lives to generative music, I have to say it's debatable whether what you described can strictly be called a "melody" unless you take the broadest dictionary definition ('a pleasing sequence of notes'), though I do think you can call it "melodic" as in 'melody-ish'. In other words, I think there's an essential difference between melody as conceived prior to the advent of generative music, which implies a decision by the composer that 'this is the melody', and a 'random melody' where that decision is made in a different manner, i.e. "whatever this process produces is 'the melody'". In the case of a randomly generated score, it gets a bit confusing because it's not clear if the composer used a generative process to determine 'the melody' or if it can meaningfully be called 'random'. Ultimately we just don't have words for such things. Perhaps conveniently, generative music still plays a very minor role in music culture outside of i.e. some of Brian Eno's recent work and the enduring influence of John Cage.

Aren't you then saying that melody-ness is being conferred by the composer? That would make it extremely difficult to confirm that something is a melody or not by observation alone (and thus likely impossible when determining whether other apes can come up with melodies).

That said, I think the article is not using "random" in the literal sense but in the sense of "arbitrary", as in "novel" (rather than being simply a reiteration of a previously observed tune).

"Is it so implausible" isn't really the bar we're trying to reach, though.

> these creatures who have displayed startling emotional depth (Koko and her pet cat, anyone?)

I'm not convinced that Koko is much more than a long-term, well-publicized Clever Hans. Can you explain what emotional depth you think is evidenced in this case? If possible, please do so without relying on the interpretation of Koko's main caregiver, who does not seem to abide by scientific principles.

The position of a 'skeptic' that you adopt is frequently adopted for things related to animal emotion, empathy, intelligence etc. It is often couched in a higher 'we have higher standards for a scientific method'. But it is somewhat misguided.

I find these inferential questions no more fraught than the question that a fellow human, for example, you are capable of intelligence, empathy, emotion. All we can do is make external observations and not get into your psyche and feel what you feel. If the basis of assigning emotions etc to a second person is the fact that I am aware of my own emotions, then it is structurally very limited. It is incapable of making inferences beyond the human kind, and even then it is quite suspect.

I am fine as long as the same standards are used for animals and as is used for humans, without resorting to arguments mirroring 'I know I feel therefore I know you do' or 'I know you feel because you told me so'.

If Bob says he has a message for you from Alice, and the message he delivers is exactly in line with Bob's incentives, are you skeptical of Bob? Even if you knew in advance Alice was sending you a message via Bob, does the fact that the message and Bob's incentives are in alignment not cause you any skepticism?

I addressed Koko specifically because her case was brought up and is especially famous. I believe that a press release about the death of a celebrity who visited the gorilla once, 13 years before[1] is either a) cynical, exploitative fund-raising or b) misguided interpretation, relying on the interpreter's own emotions to decide those of the animal being interpreted. These declarations are made by the caretaker who has a long, intimate relationship with the animal.

My skepticism is aimed not at the idea that non-human animals are intelligent, empathetic, or emotional, but rather that single human interpreters, working for decades with a single animal subject, are reliable scientific indicators of any possible conclusion. A decent Slate article here[2] goes beyond Koko. I am arguing that the evidence most commonly presented is weak, not the hypothesis.

I found this article about parrots and PTSD[3] fascinating as some of the examples are almost certainly Clever Hans effect and others are in fact best explained via a close pair-bonding humans recognize as love.

1: http://www.koko.org/koko-tribute-robin-williams

2: http://www.slate.com/articles/health_and_science/science/201...

3: http://www.nytimes.com/2016/01/31/magazine/what-does-a-parro...

Just to make sure, I was not critical of _your_ skepticism (as in personally), it is necessary for science. But was pointing out that emotion in a 2nd being is inordinately difficult to make any claims about. We give human emotion / empathy a free(er) pass compared to the standards we use for other animals. I think a Turing test equivalent is the most one can do in these cases.

For example you and I may agree that something is color yellow but still be seeing something completely different. We just agree on the name because that's how they have been labeled to us when we were learning the colors by name. This is just for simple concepts like colors, for complex things such as emotion intelligence it gets much more fraught.

No, it's not implausible, it's just not supported by the evidence gathered. No one is dismissing the possibility, I am dismissing the idea that this observation tells us anything substantively new.

Not really. A melody is any sequence of tones that one could perceive as being musically significant or interesting. You could generate a random melody by randomly playing notes from a pentatonic scale.

Random noise is musically tone-less and without structure, closer to e.g. Metal Machine Music or such.