I'm coming up with an idea for a game that simulates the evolution of languages, but to do that and make it the most realistic, I would need to put in the sounds that the IPA says are possible but we don't have a symbol for because they aren't found in any languages that we know of. Is there any good way to represent them?

  • 1
    You claim there are sounds like that. What are they??
    – Lambie
    Dec 5, 2023 at 20:57
  • @Lambie: Perhaps these sounds.
    – dan04
    Dec 5, 2023 at 23:08
  • But those words and their pronunciation do exist as sounds in English. What sounds in that link cannot be represented by IPA? I think they all can....Perhaps you need to learn the phonemes of English? :)
    – Lambie
    Dec 6, 2023 at 4:38
  • @Lambie: I know the phonemes of English. None of them involves a dental trill, which is a possible sound but not in the IPA.
    – Anonymous
    Dec 6, 2023 at 16:16
  • There is an alveolar trill. For me, dental trill is not a great term. As in perro. There is also an alveolar tap as in pero. One r.
    – Lambie
    Dec 6, 2023 at 16:37

2 Answers 2


Generally, you can either use diacritics (the official way), or invent your own symbols (the unofficial way). The IPA doesn't have special symbols for labiodental stops, so when field linguists encountered them, they either applied the "dental" diacritic to the labial symbols (p̪ b̪) or invented new letters (ȹ ȸ). The latter is more useful if they're separate phonemes and you want to keep your transcriptions clean; the former is more recognizable to people who aren't familiar with your work, especially if it's a phonetic detail that only shows up occasionally.


In principle, there aren't any phonemes without corresponding symbols in the IPA. It used to be that the IPA had no symbol for a labial flap, but after it was persuasively argued that such a sound is phonemic in some languages, the symbol ⟨ⱱ⟩ was added.

The point where confusion over symbols vs. phonemes is most likely to arise is in the use of diacritics. The symbol ⟨ʰ⟩ is a diacritic that can be added to anything, and is exploited contrastively in many languages such as Hindi, likewise nasalization which phonemically distinguishes vowels and other sonorants in many languages is indicated with a diacritic ̃. But there are other diacritics as found in [o̜], [ɪ̟], and [ɛ̠] marking 'less round, advanced, retracted' which only mark sub-phonemic detail.

A realistic game simulating the evolution of language can't just rely on abstract phonemes, since language is not learned by magical transmission of abstract phonemes, it is learned on the basis of actual pronunciation coupled with analysis of physical differences into a system of phonemes and a system of phonetic implementation. The progression of earlier /ki/ to later /tʃi/ doesn't happen in a single step, and the first step is a low-level physical one where a velar becomes more advanced before a high front vowel – [ki] → [k̟i], which can progress further on the path to [kʲi] → [ci] → [tɕi] → [tʃi].

Depending on how this game works in terms of learning phonemic representations from physical outputs, you might make it a principle that certain diacritics (those that never define phonemic contrasts) are only used in physical outputs.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.