What I mean is the /j/ sound (as in English "yellow" and French hyène) (hyène means hyena) is the most common semivowel around the world. This phoneme is highly conserved in Indo-European languages (this language family include Italic languages like French and Italian, Germanic languages like English and Dutch, Slavic languages like Russian and Polish, etc.).

Also, the /i/ sound (as in English "hippie") (the word "hippie" is also used in French) in one of the most, if not the most common vowel around the world.

Both phonemes are almost universal, as the /m/ consonant (as in English "mother", and French mère) (mère means "mother").

I know many languages that lack the /w/ semivowel like Albanian, Armenian, Greek, and Russian (the /w/ semivowel is found in both English as in "world", and French as in "oiseau") ("oiseau" means "bird").

In the universe I am creating, there is a species named ogres (their scientific name is Homo obesus, so they are still humans, just not Homo sapiens) (Homo obesus means "obese human"). In the most spoken language used by ogres, there are five basic vowels:

  1. /a/ (as in French arbre) (arbre means "tree")
  2. the sound corresponding to the a in English "fall" (and the o in French ordinateur) (ordinateur means computer);
  3. /u/ (as in English "cool", and French ouvrier) (ouvrier means worker);
  4. /y/ (as in French univers) (univers naturally means "universe");
  5. and the sound corresponding to French eu (as in French euphémisme) (euphémisme naturally means "euphemism").

There are no diphthongs in the most spoken language used by ogres, but the are long vowels, and two nasal vowels: the sound corresponding to the an in French antagoniste (antagoniste naturally means "antagonist"), and the sound corresponding to the on in French mouton (mouton means both "sheep" and "mutton").

Also, the most spoken language used by ogres has both the /w/ semivowel, and the semivowel corresponding to the u in French fruit (fruit naturally means "fruit").

So, I wonder why would a language created by mammals from the Homo genus lack the /j/ semivowel, and even the /i/ vowel.

  • 6
    $\begingroup$ This seems like a very open ended question, especially given the how diverse human languages are. There's probably an uncountable number of reasons why a human language would lack /j/ and /i/. On this site we strongly discourage brainstorming and idea generation questions. Do you have an idea in mind? Why don't you try asking for a reality check on that rather than asking for any explanation? $\endgroup$
    – sphennings
    Feb 24 at 19:59
  • 2
    $\begingroup$ Māori, the Polynesian language spoken in New Zealand, is a real-world language which does not have the semivowel /j/. but does have the semivowel /w/. (For example, the word wahine "woman", which has been adopted in some varieties of English.) $\endgroup$
    – AlexP
    Feb 25 at 5:13
  • 2
    $\begingroup$ Does there need to be a specific reason? My language has 9 vowel sounds. Arabic, in turn, has only 3 (if long and short versions are not distinguished) . They just happen to come from different language families. $\endgroup$
    – crizzis
    Feb 25 at 15:02
  • $\begingroup$ Same reason English lacks the g sound in Dutch and Afrikaans. Jou sleg bliksem! $\endgroup$ Feb 26 at 14:22
  • 3
    $\begingroup$ Se also conlang.stackexchange.com $\endgroup$
    – TRiG
    Feb 26 at 15:58

4 Answers 4


TL;DR: There do indeed exist real-world human languages which do not have the semivowel /j/, the vowel /i/, or both. The very well-known Ancient Greek did not have a phoneme /j/. Māori, the Polynesian language spoken in New Zealand, is a real-world language which does not have the semivowel /j/, but does have the semivowel /w/.

There is nothing to explain, because the requirements are fulfilled by real human languages spoken by real humans.


Since there may be some readers who are not fully familiar with the International Phonetic Alphabet:

Symbol Sound
/j/ The sound of the letter y in the English words yes and you.
/w/ The sound of the letter w in the English words we and well.
/i/ The sound of the letter y in the English word happy. Very slightly different from the sound of the letter i in the English word kit, which is usually transcribed /kɪt/. (/ɪ/ is a little bit less close, or, if you want, a little bit lowered, and a little bit less front than /i/.)
/u/ No exact English equivalent. Like oo in goose /ɡuːs/, but short instead of long. Slightly different from oo in foot /fʊt/. (/ʊ/ is a little bit less close, or, if you want, a little bit lowered, and a little bit less back than /u/.)
/a/ No exact English equivalent. Like a in father /ˈfɑːðə(ɹ)/, but short instead of long. Similar to the sound of o in mother /ˈmʌðə(ɹ)/. (Note that this usage of the IPA symbol /ʌ/ is specific to English dictionaries; elsewhere, the sound would be transcribed /ɐ/ or maybe /ɜ/, whereas /ʌ/ transcribes an open-mid back unrounded vowel, which is clearly different.)
/ai/ Only very slightly different from the sound of the letter i in the English words kind, pine and tile, or the letter y in the English word my.
/au/ Only very slightly different from the sound of ou in the English words mouth and sound.
  1. "Standard" (= Attic) Ancient Greek does not have /j/ and /w/. (But it does have diphthongs such as /ai/ or /au/; hoewever, those are not normally analyzed as a sequence of vowel + /j/ or /w/, and they do not behave as such.)

    And Attic Greek is not some obscure language...

    • Among modern European languages, Swedish does not have /w/. (But it does have /j/.)

    • Māori, a Polynesian language spoken in New Zealand, does not have /j/. (But it does have /w/. For example, the word wahine "woman", which has been adopted in some varieties of English.)

    • And the Tongan language does not have any semivowels at all.

  2. Classical Arabic does not have /e/ and /o/.

  3. (Late) Proto-Indo-European did not have the phonemes /i/ and /u/ (and it also almost didn't have /a/). But it did have /j/ and /w/, and in some contexts those were syllabicized as [i] and [u]. The standard analysis of PIE phonology presents only four vocalic phonemes, /e/, /eː/, /o/ and /oː/.

    Which shows how tricky this all is. The question asks about phonemes (because it uses /slashes/). In PIE, there is no phoneme /i/, but there is a phone [i] surfacing as the syllabic pronunciation of /j/. (And in the same way, [u] surfaces as the syllabic pronunciation of /w/.)

  4. Broadly speaking, languages which have only a very few distinct vocalic phonemes have at least /a/, /i/ and /u/. But Wikipedia gives examples of languages (Adyghe and Sepik) with a (so-called "vertical", because of how the vowels are represented on the standard vowel triangle) three-central-vowels system of /ɨ/ (central-close), /ə/ (central-mid), /a/ (central-open); and even languages with a two-central-wovel system (Arrernte, Circassian, and Ndu), namely /ə/ and /a/.

  5. In recent years we have all heard about Abkhazia and South Ossetia, over which Georgia (the country, not the American province) fought and lost a war with Russia in 2008.

    And guess what, the Abkhaz language has only two vowels, variously given as /ɨ/ (similar to Russian ы, Turkish ı or Romanian î) and /a/; or /ə/ (schwa, very common, similar to the sound of the letter a in the English word about, or Romanian ă, or Bulgarian ъ, or German e in words such as Auge or Tinte, or the sound of French final e muet in Opera French and Poetry French) and /ɑ/ (similar to the sound made by the letter "a" in the English word palm, but short instead of long).


  • The "sound corresponding to French eu" is either /ø/ (as is peu /pø/), or /œ/ (as in jeune /ʒœn/). (The difference is that /œ/ is more open than /ø/; similar to the difference between é /e/ and è /ɛ/.)

  • "[The phoneme /j/ is] highly conserved in Indo-European languages":

    No. For example, PIE /j/ became either /h/ or /z/ (maybe /dz/) in Ancient Greek; as a well-known example, Latin jugum /ˈjugum/ (French joug /ʒu/, Italian giogo /ˈdʒoɡo/, Romanian jug /ʒuɡ/) corresponds to Greek ζυγός (zygos). (From Proto-Indo-European *yugóm, which gave yoke in English and иго (igo) in Russian.)

    We have to look no further than our beloved Romance languages:

    /ʒ/ is the sound of the letter s in the English words measure and pleasure; /dʒ/ is the sound of the letter j in the English words jet and job.

    • Latin jocus /ˈjokus/ > French jeu /ʒø/, Italian gioco /ˈdʒɔko/, Romanian joc /ʒok/, Spanish juego /ˈxweɡo/.

    • Latin jaceo /ˈjakeoː/ (infinitive jacêre) > French gis /ʒi/ (infinitive gésir /ʒezir/), Italian giaccio /ˈdʒattʃo/ (infinitive giacere /dʒaˈtʃere/), Romanian zac /zak/ (infinitive zăcea /zəˈtʃe̯a/. (Spanish does preserve a sound sort-of like /j/: yazco /ˈʝaθko/, infinitive yacer /ʝaˈθeɾ/.)

    • Latin juro /ˈjuːroː/ (infinitive jurâre /juːˈraːre/) > French jure /ʒyr/ (infinitive jurer /ʒyre/), Italian giuro /ˈdʒuro/ (infinitive giurare /dʒuˈrare/), Romanian jur /ʒur/ (infinitive jura /ʒuˈra/), Spanish juro /ˈxuɾo/ (infinitive jurar /xuˈɾaɾ/).

  • 5
    $\begingroup$ Great Answer, very comprehensive. $\endgroup$
    – UVphoton
    Feb 24 at 21:39

PIE is Everywhere

OP's question references "most common" phonemes - it's important to know that basically all of the most popular languages on Earth descend from the same language family (the major exceptions being Mandarin and Arabic).

Proto Indo European (PIE) is the reconstructed / hypothetical root language that Germanic (English, German), Romance (French, Spanish, etc) Indo-Aryan (Hindi, Urdo, etc) and several other languages families descend from. Turns out domesticating the horse gives your culture a HUGE "first mover" advantage. (pun intended)

Something like 4.5 Billion people speak PIE derived languages... which is about half the planet.

So as long as you say your language is not related to PIE, you can probably get away with doing a lot of things that look "weird" to Westerners, since Western languages all have the same root.

And that's before you account for any anatomical differences for your Orge people.

  • 1
    $\begingroup$ What's the pun? $\endgroup$ Feb 25 at 20:47
  • $\begingroup$ @AzorAhai-him- I really hope it's not an inversion this classic xkcd. $\endgroup$
    – No Name
    Feb 27 at 0:20
  • $\begingroup$ I'd call Cantonese and Japanese major exceptions to that rule as well. $\endgroup$
    – Hearth
    Feb 27 at 2:33
  • 1
    $\begingroup$ @AzorAhai-him- domesticating the horse gives you a "first mover advantage" -- first mover advantage is a business term describing getting to market before your competition. The first person to domesticate the horse... was literally the first person to ever MOVE that way. They were the first people to experience the mobility of galloping, the first to build chariots and carts, and their culture spread across 2 continents over a couple hundred years. $\endgroup$
    – codeMonkey
    Feb 28 at 15:26
  • 1
    $\begingroup$ @Vikki - I went to the Wikipedia page for "most spoken languages" and ordered them by "L2" (total Speakers) -- 8 of the top 10 were Indo-European, and the other two were Arabic and Mandarin. Top 10 is an arbitrary cutoff, but it seemed a reasonable one. $\endgroup$
    – codeMonkey
    Feb 28 at 15:37

I think you could possibly look into the anatomical differences of the homo sapiens and your homo obesus. My understanding is that consonants are often associated with the stopping of air flow, while the vowel sounds are more associated with the shape of the tongue, mouth and lips. Going through the “ah” “ee” “ey” “oh” and “ooh”, you have pretty distinct shapes and position of the mouth, the "ah" more with the mouth open, the "ee" more with the teeth together and like a smile etc. The website https://www.visiblebody.com/blog/something-to-talk-about-anatomy-of-speech-sounds also points out the position of the tongue with respect to vowel sounds. I suppose speech therapists when helping children also try to help them with things like tongue position when trying to say some sounds.

That perhaps is more of a justification for why there could be the ability to produce some sounds and not others.

On the linguistic side as to why some languages tend to use some types of sounds that are produceable and not other sounds that are just as producible, I suspect you could blame it on dialects and how dialects shift. (Not sure if dialect is the best word to describe what I mean.)

Even with English there is quite a bit of drift in vowel sounds. There is something called the "Great Vowel Shift" where between the 1400s and 1700s english speakers changed how they pronounced vowels in very significant way. Even in the last hundred years linguists have been seeing a lot of vowel shifting in the U.S.

I guess another example might be related to how your homo obesus might hear sounds, or how their brain is wired. For example, I think Mandarin, is a tonal language, and some western people can pick up the tonal shifts, and others have great difficulty with it. One common explanation that being exposed to a tonal language when you are a baby helps build the neurological connections to better detect the changes in pitch.

  • 3
    $\begingroup$ Just a note: the phonetic phenomena corresponding with Mandarin tones are also used in English. The difference is that in English they carry syntactical information, and we speak of intonation, whereas in Mandarin they carry lexical information, and we speak of tones. For example, when we say who? as a question, we make just about the same phonetic gesture as in Mandarin hú 狐 "fox"; but in English the phonetic gesture carries the syntactical meaning "this is a question", whereas in Mandarin it differentiates the word from, for example, hù 怙 "to rely (on something)". $\endgroup$
    – AlexP
    Feb 24 at 22:27
  • 1
    $\begingroup$ "Why some languages tend to use some types of sounds that are produceable and not other sounds that are just as producible": Simply put, because each and every language uses only a small number of phonemes. Yes, many many phonemes are "producible", and many of them are even easily producible. For example, all English speakers can easily learn how to make the front rounded vowels represented by ö and ü in German (or eu and u in French); but English already has more than enough vowels, thank you, and it does not need any more. $\endgroup$
    – AlexP
    Feb 24 at 22:33


The human mouth and associated anatomy (esophagus, larynx, diaphragm, etc.) is capable of an infinite variety of phonemic combinations. The choice to use one over the other is ultimately one of preference. One might think it comes down to using sounds that use less energy (lowest calorie use), but they would be wrong. Many cultures value overt and energetic expression. Consider the Maori Haka dance, which is actually taught in school. An outsider would think they were about to be tied to a stake and broiled for dinner, but they are actually honoring and congratulating you. Watch New Zealand win a soccer match, you’ll understand. Then there are sounds that are almost universal yet never make it into a sentence. Stick your tongue out and blow. We call that a “raspberry.” All of humanity knows what it means, but no one has ever given that sound it’s own symbol or spelling. Why? Because humans.

So the true answer is anything you want. But in world, an ogre mouth is just set up wrong. That answer will sell a lot better than any philological thesis on human/ogre culture dynamics.

  • 5
    $\begingroup$ Actually, by accident we do have a symbol for the sound we make when "blowing a raspberry"; it's a linguolabial trill, [r̼] (usually a devoiced linguolabial trill, [r̼̊]). But you are right, it is not known to be used in any spoken language. (We have a symbol for it only because we have a general diacritic for linguolabial articulation ◌̼, and we can apply it to a trill [r] or [ʙ].) $\endgroup$
    – AlexP
    Feb 24 at 23:03

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .