177 comments

[ 3.3 ms ] story [ 183 ms ] thread
The word is ‘huh’.
Isn't it "juj" in Spanish? :)
We don't say that in spain, the closest would be "eh?" or "ah?", maybe in latin american they use something I'm not aware of.
I’d edit the title, current one seems clickbait. Maybe:

The word “huh” exists in all human languages, according to research

In the past I've edited titles, carefully crafting them to give a better indication of the content, only to have people complain that the title has been changed, and subsequently to have the mods change it back.

So I've given up even trying to put any thought into the title ... if the article seems like it would, should, or could be interesting to the HN community, I post it.

I've been trained not to put any work into it.

same. kind of absurd that when you try to un-editorialize some clickbaity bs into a tldr; it gets mod-reverted to the original vague crap.

this modding style has definitely discouraged me from submitting articles here with good content but unfortunate headlines.

And if you read the article and look at the list of examples, it's more the variants of "what?" in different languages.

"Huh" and variants exist additionally in quite a few as well.

They must have missed Swedish. We don't have 'huh', we have 'va'.
I was thinking the same, I haven't heard 'huh' in Sweden, and 'va' doesn't sound much like it.

I also read in the source: "Therefore we collected data from recordings of naturally occurring informal conversations in a sample of 10 languages from 5 continents, varying fundamentally in terms of phonology, word structure, and grammar (languages 1–10 in Figure 1)". That sounds pretty far from "all languages" as in the title.

That's really weird. So, all languages do not have word for say 'water' ? or say 'father' ?
(comment deleted)
Look at the table, non of the words are literally "huh", but they convey the same meaning. The title is confusing, I had the same thought as you at first.
Yeah the table has an Icelandic example: "Ha". Though I only lived in Sweden for a year so I can't remember but Norwegian has the word "Hæ" which certainly has the same meaning as the English "huh"
Vasaru?

I don't think the point is that it would sound like "huh" in all languages, but rather that there is a word used like "huh" in all languages. "Va" definitely fits, as you even said "we have 'va'", proving the point of the article.

I'd expect many more words to exist in all languages. It would seem that there is a core set of words that are absolutely fundamental to any sort of spoken communication (yes/no, for example), and would be very surprised if this word "huh" is the only one so far found to exist in all languages.
The article says "sound the same", but then the words they found don't sound the same across the listed languages, so it's confusing.
If you read more carefully you'll realize that the hypothesis is that "huh" is universal in both meaning and pronunciation.
Isn't Mandarin a typical example without yes and no? You answer affirmatively but repeating the verb and negatively by negating the verb.

(it's not really that simple, but it illustrates the point)

Irish Gaelic is similar, affirming or negating the verb.
In some situations, yes, but you can say yes and no. 是 (shì, roughly translates to "is" in most situations) and 不是 (bú shì, no is). It's definitely not as cut and dry as the English yes and no, and it gives Chinese kids hell when they're learning English (I taught there for 6 years).
Definitely and to add to that, to my understanding (better than basic, far from fluent), 是 is more of a fallback verb following the same pattern than it is a translation of "yes".
I thought that mama/papa/nana/dada or some close equivalent existed in all languages, as it is the first basic sound that babies can say.
I was thinking “mom “ and “papa”.

If “huh” is the standard for a word then also laughing words like “haha” are also universal.

The physical sound of "haha" might be the same everywhere (to some degree, people laugh differently obviously) but different languages write "haha" different. Spanish: "jaja" or "jeje", Japanese: "wwwww"
In simplified Chinese: 哈哈哈!
That's starting to get into paralinguistic territory. Sneeze sounds, for example, are learned to a degree (think some people sneezing very quietly or very loudly), and can be relearned even though they are involuntary in their timing, unlike "real" words.
(comment deleted)
I would say "haha" exists in all languages but what do I know?
That's not a word though, that's an emotive sound, like crying. Sure, "haha" is a word you can use on the internet but it's the English interpretation of the sound laughter makes.
Most languages probably have an "interpretation of the sound laughter makes" and across the languages I know it's a fricative-vowel syllable repeated several times. That's about the level of similarity shown for 'huh?' across languages in the study.
'Huh?' is significant because it forms a communicative purpose, it forms a request from listener to speaker to recall the previous spoken phrase. The study is showing that all languages have a sound for that purpose.
It’s pretty clear that there are some grunts/vocalizations that are shared amongst all humans (and possibly primates) these would yes/agreement, no/disagreement, and question mark/surprise those aren’t language dependent.

I’m not sure if it’s actually convergent evolution like the research paper implies my bet would that these predate humans by quite a bit, and some languages have a stronger onomatopoeic relationship between the base grunts and their question word.

Even in English ‘What’ has the basic huh in it, wha/ha it wouldn’t surprise me if the T was added to make it more distinct or allow for more tonal information be passed along such as surprise, displeasure, disappointment and even anger.

My non-linguistically educated opinion is that huh and and other grunts like a-ha (not the band) are probably much older than humans and they are universal just as female copulatory sounds are in humans and other primates (there is more to girls moaning in bed than just pleasure).

https://en.m.wikipedia.org/wiki/Female_copulatory_vocalizati...

What is "co" [tso] in my language and people here are making the same arguments with it
Well depending on the tonal formalization of it at least how I would pronounce it would probably be very similar to huh just a non-guttural version of it that may have evolved because the language itself is non-guttural.

The question really is if a non-English or any second language speaker of your language understand “huh” as in not the word but the grunt, my bet would be yes.

Not sure I consider "huh" to be an actual word. If so, I'm pretty sure blowing raspberry also exists in all human languages, and some non-human.
I have never seen that word huh written or pronounced in Spanish, French o Italian.

... But know it exists. I suppose more than one reader has pronounced aloud, in his native language, so it exits.

The title is misleading. It's not about the specific word "Huh" but about the fact that every language has a word that takes its function - namely to trigger a speaker to repeat their previous sentence.
And at least the one they claim for German ("häh?") is a general expression of "I don't understand", but it doesn't mean "can you repeat that?", it means "I don't understand", which you'd say if you don't understand e.g. an explanation. If you didn't understand somebody because you weren't listening, they are mumbling or there was a noise, you'd say "Wie bitte?" or something similar, not "häh?"

I suppose if you widen the concept of "is the same" enough, you'll find all sorts of words that exist in all languages. Sure, they may sound different, they may not really convey the same meaning ... but they exist!

Same in Hebrew and Arabic and quite a few other languages, there is no onomatopoeic word for huh but the vocalization of huh/ah exists.

Some part of me wonders when huh and haha were even added to some languages it wouldn’t surprise me if a lot of basic primate/human vocalizations in written forms are fairly new and only appeared as more complex fictional literature and heck even comic books/illustrated literature became a thing.

The word in French is generally written "hein"

https://fr.m.wikipedia.org/wiki/Hein_(interjection)

Hmm, I think that would be “hein ?” then, without the question mark “hein” can have a lot of different contextual meanings that are quite different from “huh”.
Can you give some examples? I’m only conversational in French at best so maybe I haven’t encountered other uses yet but I’m struggling to think of a sentence with “hein” where “huh” wouldn’t fit in the English translation.

Would you say it’s more like the Canadian “eh” that can be tacked on to just about anything?

I think in most cases it will indeed be used in an interrogative tone, and written followed by a question mark to reflect that.

It can also be used to mean "isn't it", to ask for agreement, approval (very informal).

Example: il fait beau, hein? (The weather is good, isn't it?).

The proper way to say "isn't it" is "n'est ce pas", but it's very rarely used in practice, and English people speaking French are often caricaturized ending all their sentences with "n'est ce pas", which is not very far from reality in my experience ;)

It's not the kind of thing you will find in formal discussions, but it's a very common interjection in spoken french. I cannot tell you the rules, it's just something people use to complete their sentence or communicate something about the context.

(all the examples below are from real chat discussions)

Exemple 1: "Non mais sérieusement, c'est pas très malin, hein..."

Translates to: "Seriously, that's not really smart..."

"hein" here presses on the fact that the speaker is a bit impatient or annoyed.

Example 2: "Je suis pas forcément anti Facebook hein!"

Translates to: "I'm not necessarily against Facebook!"

Here it is used to communicate that there may have been some misunderstanding in the context of the discussion, and making clear that even if I said stuff negative toward facebook, I'm in fact not anti-facebook.

Example 3: "J'ai aucune expérience avec Haskell hein"

Translates to: "Just so you know, I have no experience with Haskell"

The "just so you know" is communicated through the use of "hein" at the end.

Example 4: "C'est de ma soeur, hein, pas de ma part"

Translates to: "To be clear it's from my sister, not me"

Here too, the "to be clear" is communicated through the use of "hein".

Be aware: the interpretation is purely subjective and depends on the context, and mostly the tone, so it's not that clearly communicated via text! Also, my french is actually swiss-french.

Hope that helps, and tell me if "huh" would communicate the same meanings :)

Another one I'm thinking about now:

Person 1: "C'est pas correct, laisse moi te montrer comment ça marche. Clique sur ce bouton, et ensuite sur celui-ci. Tu vois?" Person 2: "Heeeeeeein!" (very long one that kind of sound like a sin wave from high pitch to lower pitch)

Translates to:

Person 1: "That's not correct, let me show you how that works. Click on this button, then that one. All good?"

Person 2: "Ooooooh, I see!"

Thanks! This is super helpful! This is the kind of “slang” (not really sure if that’s the right word) that’s really hard to learn from books and even from watching Netflix in French. I’ll have to try to pay attention to “hein” on TV shows I watch. I wonder if my brain has just been skipping right over the word.

Thanks again!

You're welcome :)

In TV shows people often speak French in a more dramatic/theater-like way, it's likely that they remove some details like that. I face the same issue as I'm learning German, the way people speak in German movies is a sanitized version that sounds more like drama (theater) than real dialogs.

If you listen to Twitch stream in French, or even some Youtube videos or some podcasts, or people on live TV (basically anything that isn't edited to cleanup the audio), I expect that you will find a lot of those interjections!

Twitch streams! Thats such a good idea and I definitely, will thanks :)
In my city's Spanish it's "eh?" or "hm?"
The "hmm?" is the "word" that's shared by all human languages, not "huh".
Spanish equivalent I think it would be more like "hmm".

"Eh" in Spanish exists but is more to call for attention, kind of like "ahoy" http://udep.edu.pe/castellanoactual/e-eh-y-he-2/

In my region we use "¿eh?" more like an informal abbreviation for "¿qué?", so I guess it's similar to huh.
And meanwhile, the difference between "hmm" and "huh" is pretty much whether or not your mouth is open when you make the sound. Seems plausible that even these might be related.
EDIT 2: TIL that the flame-war detector triggers a review by the mods, so this comment is irrelevant, and wrong. I was going to remove it and leave the thread tidier than it was, but I can't, probably because it has replies. So it will remain ... please ignore it and carry on. As you were.

===================================== Original:

Huh ... more discussion than upvotes, I suspect this will trigger the "flame-war" penalty and disappear from the front page, possibly within minutes (10:42 BST)

Shame, I thought this was an interesting discussion. If you agree and haven't done so already, your upvoted is needed to keep this on the front page. If you disagree, let it get buried.

EDIT: as I say in a reply below, I'm just interested in the discussion. If you don't want to give me the karma that's fine, just find some comments of mine and downvote them in balance.

If the flame-war detector is triggered, the mods get alerted and decide whether they're actually dealing with a flame war or not. No need to beg for upvotes.
Is that really true? I've not seen them say so, and it's never come up when I've discussed it with them. I'll have to ask ... thanks for the information/suggestion.

If people want to keep the item but don't want to give me the votes, feel free to downvote one of my comments. I don't care about the karma, I wanted to see the discussion develop and evolve.

You're right, I hadn't seen that. I'll remove my comment.

Thanks for the information ... today I learned.

IIUC the penalty is applied automatically but they get a warning [1] and then they check if the penalty is correctly applied and may revert it.

[1] I guess by email, but it may be a SMS or a big red blinking light in the headquarter.

Also, perhaps they get the warning when a submission pass a lower threshold and the penalty is automatically applied with higher threshold.

All languages tend to use shorter words for things that are more common.

The more commonly the thing comes up in conversation, the shorter the word tends to be.

Given how vast the pronunciations are (even in the examples they chose), I think they are creating an illusory correlation.

french chien, german hund, italian cano, and portuguese cao are all dog. They all have vastly different pronunciations, but they all trace their origins to a single proto-indo-european word. And in english, we use that same word to mean hunt, because that's what dogs do for us.

so the question is not whether words are pronounced the same, but whether they can be traced back through a lineage. where the sound does offer clues when taken as part of a larger tapestry. Another comment here says that the french huh is written (and presumably pronounced) hein. Well isn't that interesting? huh-hunt and hein-chien? (french uses irregular phonetic spelling because it used to be pronounced differently than today) Can we find a pattern of similar changes that would lead us to believe that the historical ancestor word of huh-hien was in use by an early tribe of people before their descendents separated deviated, physically and linguistically?

so, that's in essence a good piece of what linguists study; as in all areas of expertise, non-experts have difficulty contributing as their anecdotal experience has frequently already been catalogued and is part of the hypothesis.

The English cognate of chien, etc. is hound. Hunt is apparently unrelated. (Source: wiktionary)
That's the point though, there is no evidence linking these words to anything related.

The icelandic 'ha' and spanish 'ke' sound nothing alike and have no known relation.

The authors are trying to suggest that because everyone has these short responses that they must be related.. which only makes sense if you don't stop to think about how languages are formed.

Interestingly, dog is unrelated to that root proto-indoeuropean word.
I can’t remember where I read it but apparently the etymology of the word dog is not known. Old English used hund (or something similar) and sometime between then and now ‘dog’ turned up. Of course we could still say hound and people would most likely understand you meant dog.
Surprised no one has mentioned gynantonix
(comment deleted)
The researchers discuss the question whether “huh” is a word or just a grunt in the linked paper. (https://journals.plos.org/plosone/article/file?id=10.1371/jo...):

> Huh? exhibits linguistic conventions that speakers need to learn in order to use the form properly. A learner of Spanish has to know that repair is initiated with the mid front unrounded vowel ‘‘e8’’, a learner of Cha’palaa has to know that the form is more like ‘‘aQ’’ with falling intonation, and a learner of Dutch has to know that a glottal fricative at onset is common: ‘‘h38’’. Its acquisition follows a normal trajectory, at least in American English-speaking children [37]. Second language learners’ reports confirm that the precise form of this interjection has to be learnt, and that intuitions are not necessarily a reliable guide in this process [38].

> Perhaps there is a continuum from non-linguistic vocalisations like sneezing and crying to prototypical conventional lexical items like bless you and pain [39]. Our evidence suggests that huh? is more on the word side of that continuum. Based on the fact that huh? is integrated in multiple linguistic subsystems and conventionalised in language-specific ways we conclude that huh? a lexical word.

It's definitely not a word in all languages, probably not in most of them.

It's an utterance like "oooh" and "hmm" and "haha" which are pretty much universal as well.

Can the people who're downvoting the above comment mention whether it's factually incorrect ?
It certainly has more than a whiff of "HN commenter solves question experts have been discussing for some time" to it.
I'm not so sure about this... I speak 3 languages and it does exist in only two of them. The closest you can get to "huh" is probably "eh" in the last one, though that is still fairly close. Mind you, those are all European languages so they probably have influenced one-another significantly over the centuries (even the Slavic language of the bunch). I guess "huh" might have been transferred over to other parts of the world during the time of the colonial empires, making it more common. But all languages? I Doubt it. I happened to be chatting with an old friend who is originally from Lebanon and she said that's not the case in Arabic for instance(with a grain of salt since she also says she's anything but fluent in Arabic).
I speak two, and in one of those, the meaning is completely different. In my native language, the nasal "huh" is actually and acknowledgement as in "yes, I will do it" or "ok, go ahead". Our version of the quizzical "huh" is actually a nasal "aa" sound (as in cat) with an upward inflection.
I think the chart showed transliterations similar to "aa". It was not claimed that the exact sound or spelling of "huh" exists in every language.
The table in the article considers "eh" (like in Spanish) and "a" (like in Russian) as equivalent to huh in a given language.
Then clearly "huh" doesn’t exist in all languages, proving GP point.
Per the linked article in OP:

>They recorded bits of informal conversation from 31 dialects across 5 continents and suggested that the word ‘huh’ (and its variants) is possibly a universal repair initiator that exists in all languages, performs the same function and sounds roughly the same across languages.

It is apparent that they were aware that it isn't pronounced the same in all languages (because of course it isn't), but that it's a possible universal repair initiator that sounds similar. The title of the article is linkbait, I agree, but the article explains better.

But isn’t this the case for more than just “huh” ? For example, expressing pain (eg “ouch”) exists in all languages as well, although they all have to be translated as well.
Yes, and the article addresses this. It says that pain sounds are closer to grunts on the word-grunt spectrum, and they argue that “huh” is more word-like ok that spectrum.
>expressing pain (eg “ouch”) exists in all languages as well

I'm pretty sure it's not.

This is like saying 'yes' exists in all languages because there's an equivalent in every language... It's... a bit clikbaity
Agreed.

On the topic, sí seems to be yes, but backwards.

(comment deleted)
If I am not mistaken, there is no "yes" in classic Latin.
You are correct. In fact old English didn’t have it as well. "Yes" comes from "gese" which means "So be it".
According to wiktionary entry on Yea.

From Middle English ye, ȝea, ya, ȝa, from Old English ġēa, iā (“yea, yes”), from Proto-Germanic ja (“yes, thus, so”), from Proto-Indo-European yē (“already”)

Wiktionary about yes says

From Middle English yes, yis, from Old English ġēse, ġīse, ġȳse, ġīese (“yes, of course, so be it”), equivalent to ġēa (“yes", "so”) + sī(e) (“may it be”), from Proto-Indo-European yē (“already”). Compare yea.

So "yes", seeming so simple, was once a compound word.

I was taught "ita vero" for yes and "minime vero" for no during Latin lessons at school.
Latin had a few ways to affirm or negate. Both "ita" (it is) and the more forceful "ita vero" (it is true) could be used for yes, but you could also repeat the verb from the question with modification to affirm. "Have you eaten?" -> "I have eaten."

For negation you could do the same verb-repeating, but add "non" to negate.

Irish doesn't have words for "yes" and "no" -- answers must echo the verb: "Are you okay?" "Am." "Did you tell him to come over?" "Didn't tell."
Unsurprisingly, it's similar in Scottish Gaelic and Welsh. Mandarin and Cantonese (and probably other Sinitic languages) don't either, with much the same arrangement as the Celtic languages.
In Welsh there's (Nag-)Ydw and (Nag-)Oes, and Dim.
No Shi? I had no idea about Mandarin.
There are many yes-like words, but not exactly "yes". There's "it is", "it's correct", and "it's good" that may be used depending on context to mean yes.
Enter 嗯(polyphonic):

ng1: indeed, so what? ng2: huh? ng3: what? ng4: yeah, yes. ng~: nooo.

Amazing! I wonder about the universality of head nods (for yes) and head shakes (for no), and whether languages that lack a "yes" or "no" might still use nods.
Not universal, a few exceptions[1] :

> There are several exceptions: in Greece, Iran, Turkey, Bulgaria,[2] Albania, and Sicily a single nod of the head up (not down) indicates a "no".

Though for Bulgaria at least it seems it's shifting with many people who lived abroad at some point(myself included) and have picked up the correct/more universal form. Which makes it worse: Now no one really knows whether you mean yes or no...

[1] https://en.wikipedia.org/wiki/Nod_(gesture)

There's also the head bobble, common in south India, which looks a lot like an ongoing head shake but generally means something like ongoing agreement.

https://en.wikipedia.org/wiki/Head_bobble

Referring to it as "ongoing agreement" reminds me of something I think of as the "confessional no." When addicts talk about their addiction, there's often a rhythmic shaking of the head from side to side like a continual "no," but that speeds up and slows down with some relationship to the grammar of the sentences used.

You can probably make yourself do it by saying out loud: "When I was young, I thought I understood everything." It's like an ongoing disagreement.

What's being argued here is like saying "french" and "francais" are completely unrelated words because they sound different, despite being somewhat similar in sound and having a clear shared origin and meaning.

You can go pretty much anywhere on earth and say "huh? uh? ummm?" in response to something that you don't understand and people will get that you don't understand. Say another word, like "Great", and they might not know that you're affirming whatever they said.

And "yes" doesn't exist in all languages. Many languages require the speaker to repeat back part of the statement or the verb to affirm something.

What they mean is that all languages have a similar sounding, very short morpheme to signal repair. That's quite the language universal.
Different languages have different ways of recording/formalizing what are fundamentally the same onomatopoeic inspirations, though.

In English, frogs croak; in Japanese, they kerokero. It’s clear that the same sound is being portrayed there, just broken up differently due to English having an alphabet where Japanese has an abugida.

Neither language’s encoding from the source “sound of a frog croaking” to a word is lossless, because humans can’t make that exact sound with their vocal cords; and so none of our systems of talking about sounds, be they written languages, spoken languages, or formalisms like IPA, can contain a good representation of the required movements. But both languages above get close enough that we can recognize the sound being referred to from its per-language encoding when we hear it spoken, and from there, know what the word is, even without speaking the relevant language.

The idea here is that there is a thing humans can communicate by making the sound of their language’s equivalent word for “huh?”, which speakers of all other human languages can equally recognize as having the same meaning as their own language’s word “huh?”, without understanding anything about the language being spoken. That communicability is what effectively makes the words “the same” word.

It’s not really that the word (written symbol; exact spoken phonemes) for “huh?” is the same between languages. It’s the abstract sound pattern that’s the same. Just like a frog’s croak, as an abstract sound pattern, is the same as any other frog’s croak, and unlike anything that’s not a frog’s croak. The difference is that all humans can make a “huh?” noise; and so we do get perfect mutual intelligibility out of it. We don’t manage to represent the “sound that ‘huh?’ makes” well in human languages; but when we say it, we still say it the “right” (mutually-intelligible) way nevertheless.

The argument is that the word for "huh" in all languages can be traced to some common ancestor language. "eh" and "huh" are very similar phonetically, and could readily evolve from one to the other or from some common ancestor that's neither).
As a teen in the 80s I went to corner stores which in my city were run by Lebanese people (immigrants of the 1973 civil war). I collected the coins from video games.

As a fan of languages I found the Arabic language fascinating. One thing I noticed was the "eh" sound used for a huh or what.

The older couple would argue frequently so I dare say some of the words may have been swear words.

...all languages that we checked.
>> They recorded bits of informal conversation from 31 dialects across 5 continents and suggested that the word ‘huh’ (and its variants) is possibly a universal repair initiator that exists in all languages, performs the same function and sounds roughly the same across languages.

Can we compile a list of more than 31 languages where this interjection sounds nothing like "huh"?

I'll start with Greek:

  - Αποφασίσαμε να μετακομίσουμε στην Αλάσκα (We decided to move to Alaska)
  - Ε; (/Ɂe/)
Here's an ipa chart where you can click the symbols to get the sounds:

https://www.internationalphoneticalphabet.org/ipa-sounds/ipa...

French: hein \ɛ̃\",
Or "euh"
Not in France, as far as I know - or I'd be curious to know in which region it is used like this.

"Euh" is typically the equivalent of "erm"; it signals hesitation [1]. It can also be uttered when a person is surprised.

[1] https://en.wiktionary.org/wiki/euh

I am French and I took the definition of "huh" from the article:

> Huh is a very specific repair strategy that does not target some part of the statement but rather the statement as a whole.

... and I misread it :). I missed the word repair.

You are absolutely correct with the definition of "euh" - I usually use it when I have doubts or do not agree in a soft way (it then sounds more like heeuuuu...pfff).

Parisian native here, studied linguistics and specifically French/English in college, it's been many years. In my experience there is no equivalent to the use of "huh" in a single word in French so this article title made me a bit skeptical. Couldn't find any corroboration in the linked article. Only hinging that there "rough" equivalents to "huh" in all languages.

As you two suggested, "huh" is tricky to adapt to as a native French speaker learning English, especially American English, because "huh" is not equivalent to any single easily muttered statements in French as far as I know. I would convey the meaning of "huh" in French with: - "ah ouais" (express that I didn't realize something) - "bah dis donc" (express soft surprise or shock) - "mouais" (express soft skepticism or disapproval)

Even those don't quite feel like the encompass the uses cases for "huh".

One thing that stuck out to me in the article is that they only mention one meaning of "huh". As you have pointed out, there are multiple meanings of "huh". You have "huh?" meaning "I don't understand, please repeat and clarify". There is "huh???" for "express soft skepticism or disapproval". And "huh!" for "express soft surprise or shock". Or "huh." for "express that I didn't realize something". Even "huh..." for "I need to think about this more".
(comment deleted)
I don't speak or read Greek, but I do want to note that this counterexample does not prove there is no "huh"-like repair initiator in Greek.

In American English, there is "Eh?" "Huh?" and "Eh?" are similar in function but sound different.

The "h" sound (/ɦ/, I think) does not exist in Greek, but I agree that "Eh" sounds closer to the Greek exclamation than "huh".

On the other hand, the article seems to say that it's specifically "huh" that's universal. Anyway, I'm just curious to see how many counter-examples we can find.

Funnily enough, Ancient Greek did have a leading h- sound. It’s since disappeared, along with tonal accents and some vowel sounds.

I’m not sure the h itself is really necessary for the main argument, but possibly it got dropped somewhere along the way? That’s pure speculation though, I haven’t seen any evidence for that in print.