This is an excellent article! However, I find the definition of phonemes very unsatisfying: > Phonemes are not exactly the same thing as sounds, and they’re not exactly the same thing as letters. They are kind of like connecting categories that map letters to sounds … Phonemes are a way of saying, If you see a letter “r” is this situation, use Sound A for it. If you see “r” in this other situation, use a different So…
(Not saying you are wrong, I'm trying to clear up my misconceptions) I didn't realize that phonemes were perceptually distinct units of sound, in the same way graphemes are visually distinct units. (yes there is the whole "some speakers recognize different sounds in their brain as the same" wrinkle). What would the term be for units of sound which are perceptually distinct regardless of listener's perspective? E.g. […
Yes, exactly! This is a great way to describe it. (I think I might steal it for the future, even.)
> What would the term be for units of sound which are perceptually distinct regardless of listener's perspective?
The closest is ‘phone’, as mentioned by umanwizard, but I’m not sure there is any one word for this, because the concept itself is poorly-defined, and you can make two sounds arbitrarily close. Can you tell apart [a æ]? What about [a̝ æ̞]? How close can they get before you can’t distinguish them by ear? Before a speaker of another language can’t distinguish them by ear? Before you can’t distinguish them even with a spectrogram? And the language you speak really is important — I’m fine distinguishing [p pʰ], I think, but I know people who can’t even hear a difference. Linguists talk as if there were specific sounds, but really there’s only ‘fuzzy’ sets of sounds which people perceive as being more or less similar.