Usually, constructed languages of all kind (naturalistic or not) draw their sounds from existing natural languages.

Are there conlangs with constructed sounds, i.e., sounds that do not occur in natural languages (or are at least very rare in natural languages)?

  • 2
    @ChrisF If you have an answer, please post it below. Comments do not have the feature necessary to vet or edit whatever someone might say here, so we do not use comments for answers or partial answers. Commented Feb 9, 2018 at 15:20
  • Vernor Vinge's story “Conquest by Default” features aliens who can voluntarily close their noses, allowing a stop ‘p̃’ and a fricative ‘ṽ’. (When it was first published, those characters were not readily available, so they were replaced with ‘%’ and ‘#’ iirc. I hope some reprint will restore the proper letters!) Commented Apr 26, 2018 at 4:00

6 Answers 6


Click consonants are rare in natlangs, but rather popular in conlangs, though despite this, given the existence of things like !Xóõ reality is stranger than most fiction in this specific regard. Language-game/conlang(?) turned actual spoken register Damin had several phonemes not seen in (other?) natural languages, such as an ingressive glotally interrupted unvoiced lateral fricative /ɬ↓ʔ/, an egressively released bilabial click /ʘ↑/ and stuttered bilabial trill /ʙ\ʙ/.

Searching for the various hypothetical combinations as well as ext-IPA sounds on google turns up a fair few instances of conlangs using sounds such as percussives, implosive fricatives, velopharyngeal and nareal fricatives, etc.

Outside of these humanly-possible, but non-attested outside of disordered speech phonemes, some languages made up for non-humans include sounds that are impossible for humans to make, such as this one spoken by a race of giants, including a special type of whistle-like sounds, or this one made for creatures with bird-like anatomy, differentiating between 4 different types of contraction of muscles in the throat overlaid with 5 different pitch contours.

  • 1
    There is a hypothesis that Proto-Human (a hypothetical first language from which all other human languages are monogenetically related) may have been a click language, which would actually make click languages more "natural" than non-click languages. Commented Feb 19, 2018 at 10:25

There's also the brilliant masterpiece kay(f)bop(t), which features

  • the dextral lateral click (a click made on the right side of the mouth)

  • the sinistral lateral click (a click made on the left side of the mouth)

  • the manual stop (a clap, which may only occur in morphemes pertaining to penguins)

  • the faciomanual click (a facepalm, which may only occur in function words)

(note that these aren't actually official names but come from the Conlang Critic video on kay(f)bop(t))

These are already stretching it, so I don't think kay(f)bop(t)'s phonemic hats could be counted as "sounds" per se.

(And yes, this is a joke language, but I couldn't resist a chance to mention it.)

  • 2
    It is probably challenging to distinguish dextral and sinistral clicks acoutically.
    – Sir Cornflakes
    Commented Feb 7, 2018 at 14:11
  • 3
    Yeah, I'd call the last two the bimanual percussive and the faciomanual percussive, respectively.
    – bb94
    Commented Feb 7, 2018 at 18:50

In Ithkuil, a geminate /h/ can be produced as a bidental fricative, a sound which is only attested in a single dialect of a single natural language.


The geminated version of Ithkuil h is pronounced in either of two ways: (1) as a “bi-dental” fricative, in that the jaw is completely closed and the upper and lower teeth are in near-contact along their entire length; the resulting sound is somewhat similar in timbre to both a voiceless interdental fricative (as in English thin) as well as the English f-sound, however there is absolutely no contact by the tongue with the teeth or gums when pronouncing this sound; no IPA equivalent; or (2) as the voiceless pharyngeal fricative found in Arabic (spelled ح)and in various Northwest and Northeast Caucasian languages. This second allophone should not be employed if the resulting pharyngealization distorts the timbre of the adjacent vowels to the extent that their place of articulation changes (e.g., the vowel û being made to sound like ô).

  • Didn't know that! To that list we can add Queranarran, which also has the bidental fricative. I'm not sure I'd consider it a "constructed" sound or not. I mean, it's a real sound and everything...
    – elemtilas
    Commented Apr 28, 2018 at 22:42

I recall some years ago, someone submitted a greeting for Conlangery that was a bunch of insect sounds. According to the conlanger it is meaningful, though I don’t vet greetings too thoroughly.

Also, in my earliest conlang, Yeltax, there is a harmonic tone, with two simultaneous pitches. It’s impossible for humans to produce in normal speech, as the aliens who speak Yeltax have a syrinx, which makes two base frequencies possible. (Note: I made that before I learned any phonetics, and now I think I’d want to think about the acoustics here to distinguish this from the harmonics we use for vowels.)

In general, though, most conlangers don’t make up too many sounds. Many of the gaps in the IPA chart are physically impossible, and it’s questionable that we’ve missed a major articulator. I think most of the true inventions will be for non-human languages.

Leaving behind sound, though, it could be said that Rikchik has a robust alien phonetics that was invented from scratch. Since the aliens have many tentacles for signing, and a very non humanoid appearance, their signs can’t be related to any of the features of signs in human sign languages. Once again, invented phonetics is probably going to be for aliens.


Old Entish is a possibility, though not really by direct attestation. The Professor says of Old Entish that it is "...slow, sonorous, agglomerated, repetitive, indeed long-winded; formed of a multiplicity of vowel-shades and distinctions of tone and quantity which even the loremasters of the Eldar had not attempted to represent in writing" (LotR Appendix F)

It is after all spoken by Ents and Huorns (sentient tree-like beings). Lots of deep, rolling hoomhomming and baruurundillanding, possibly with subsonic undertones.


I am currently working on a conlang making heavy use of:

-Interdentalized sounds, like what @J. Siebeneichler pointed out

-Fully Interdental sounds, where you literally have to bite your tongue

-A Rounded Schwa

-and a plosive that isn't quite a pharyngeal plosive but sounds like it as it requires the speaker to press the back of their tongue directly against the back of their throat.

Given, this language isn't really meant for humans, so many of it's other oddities are downright unusable to humans.

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.