I'm trying to pinpoint a linguistic concept that may or may not exist.

Let's say you have a complete set of "units" (i.e., sounds/letters/moras) in a language. This can be many things, depending on the language. If the language has an alphabet, this set of units may be the alphabet itself. If there isn't any 1-to-1 mapping between sounds and the alphabet, this set of units may be a set of sounds (i.e., in English, just the letter "a" has multiple readings). It can also be a complete set of sounds (e.g., moras, expressed with Hiragana in Japanese).

You randomly select a reasonable amount of such units and attach them together, forming an artificial "word". Obviously, this artificial "word" may or may not be a real word in that specific language. I believe that the probability of such an artificial word constituting a real word is different across language so language families. If this probability is relatively high, it would mean that a large percentage of "garbled", artificial sounds produced in that language would actually constitute real words.

In that case, the language in question would have a large degree of <the term I am looking for>, e.g., some sort of saturation, perhaps "phonetic saturation"?

I would appreciate any opinions or tips!

  • 1
    I am not aware of a specific term for the rate of usage of the allowed space of words in a given language. Commented May 22, 2023 at 8:00
  • In programming sometimes we call this semantic density.
    – ngn
    Commented May 22, 2023 at 19:29
  • @ngn Thanks, looks like semantic density is a good starting point!
    – vivasra
    Commented May 23, 2023 at 3:39

1 Answer 1


It seems that the concept you are looking for is discussed in The sound pattern of English pp. 416 ff. where an algorithm is posited to compute a scalar "distance from lexicon". The point of that computation is to distinguish existing words like [brɪk] from possible non-extant words [blɪk] (they apparently did not know of the existence of that art supply store when they wrote this) vs impossible [bnɪk] and highly-impossible [bnzk]. Their system is based on a phonetic theory of "similarity", where [b] and [p] are more similar than [b] and [ʃ] in computable ways.

  • Thank you, I will definitely look into it.
    – vivasra
    Commented May 23, 2023 at 3:38

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.