I read a book called The Horse, The Wheel, and Language by David W. Anthony after watching Simon Roper videos with great interest for the last couple of years.
I can definitely recommend that if you're interested in the topic. A pretty detailed exploration of the Kurgan Hypothesis.
Yes, though it makes sense as majority of Europe and Europe-descended cultures are Indo-Europeans, with Indians and Iranians as the other major cultures.
We know much about Proto-Semitic AND Proto-Afro-Asiatic, thankfully, due to the antiquity of recorded Semitic languages in the Near East. Proto-Austronesian seems to be clear too with the 600+ languages we have access to. Unfortunately the languages between the three (Caucasians, language isolates, Tibeto-Burman, Andamanese, Turko-Mongol-Yeniseian Altaic* sprachbund etc) are much more obscure.
Yeniseian (the originator of such words as Tengri, Kha(g)an, etc) being connected to Native American Apache language was pretty amazing and evidence of the importance of paleo-linguistics in finding out informations about our ancestors
> Yeniseian (the originator of such words as Tengri, Kha(g)an, etc) being connected to Native American Apache language was pretty amazing and evidence of the importance of paleo-linguistics in finding out informations about our ancestors
This is a bit controversial, though. I haven’t read through the original papers, but one person who did so informed me that they found this relationship not very convincing at all. Still, it’s fascinating stuff!
> Yeniseian (the originator of such words as Tengri, Kha(g)an, etc) being connected to Native American Apache language was pretty amazing.
Vajda’s theory is still regarded as very tentative by the community, and even Vajda himself suggests in his most recent book on the subject that work towards proving a connection has barely begun. It is definitely not settled science.
The problem is that there is very little material to work with because all non-PIE languages besides Basque have been lost and what little we have of Tyrsenian (and to an even much lesser Tartessian and some others) is really insufficient to make any substantial claims about their origins or shared features.
Yes of course (I would generally expect most European linguists and anthropologists to be more interested in studying European languages due to obvious reasons).
Thanks. I asked because this is a topic that interests me that I don't know a lot about. I wasn't sure whether there was a theory that (almost) everything is PIE or whether there was an Asian component to your comment or something else that was in my unknown unknowns.
Wow. I asked because this is a topic that interests me that I don't know a lot about. I wasn't sure whether there was a theory that (almost) everything is PIE or whether there was an Asian component to the comment or something else that was in my unknown unknowns.
Don't jump down people's throats based on uncharitable readings of what they wrote.
I'm aware that an overwhelming majority of the global population use non IE languages. I just wouldn't expect that most research related to them would be published in English (and even if it was it would probably be much harder to understand for people who only speak English + other IE languages).
This other commenters are just being pedantic and have nothing to add. What you meant was perfectly valid English and there’s no need to explain you meant European languages when discussing European languages.
However I do not believe your comment about the “vast majority” of languages is accurate. I indo-European languages make up about half of all languages on the planet (3.2 billion) and no other language family comes close.
Even in Europe we have Hungarian, Finnish and Estonian as reasonably large non-indoeuropean languages; Maltese as an example of a Semitic EU language, and of course Turkic languages are also right there.
The question is reasonable, it’s often not a good time to watch a video and one might want to know if it’s worth coming back to. The problem is that the tone is abrasive.
Video should be regarded as science journalism. It's made by someone not doing research in topic but educated enough to follow research and reproduce it in manner suitable for the general public. Preferably though they should've cited the used sources.
Good researchers are often bad educators, and vice versa. It doesn't matter so long as the information is good, and the credentials of the speaker are a poor signal for quality of information.
Recently found this video https://youtu.be/bzRxSVK7qIU where Sanskrit was being compared to Lithuanian. To think, that such geographically distant languages are so similar, as Lithuanian is one of the most well preserved languages in Europe. To get a grasp of what PIE sounded, the closest you can get in Europe is by listening to Lithuanian (especially the older dialects of it)
> as Lithuanian is one of the most well preserved languages in Europe
Just to elaborate on the confusing phrasing here... Lithuanian, among PIE languages, is regarded by linguists as having preserved the most archaic features of PIE. So it basically resembles the first branches of PIE (like Sanskrit, but more Proto-Balto-Slavic) more than any other modern PIE language today. So as linguists trace back systematic transformations from other Proto-Balto-Slavic derived languages back to their root words, it resembles Lithuanian more. Btw, I'm not a linguist. I just worked really hard for a 4.0 GPA, even for electives.
> the closest you can get in Europe is by listening to Lithuanian (especially the older dialects of it)
I'm not sure that's true. I mean Baltic languages are generally considered to be one of the last to diverge from PIE but that doesn't mean the sound that similar. They preserve many archaic features but there has also been a lot of innovation over thousands of years (especially considering that modern standard Lithuanian is to some degree a constructed language).
While certain grammatical features like almost the whole case system have been preserved too a higher degree than in other IE languages. Lithuanian has still lost laryngeals, aspirated stops etc. which means it sounds very differently to what proto Indo-European might have sounded.
> To get a grasp of what PIE sounded, the closest you can get in Europe is by listening to Lithuanian
No you can’t. Especially not the “sounded” aspect. Lithuanian is conservative in some matters of lexicon and morphology, but modern Baltic phonology is very different from PIE. The ancestral stages of Lithuanian after PIE lost the PIE laryngeal sounds, then the distinction of three classes of velars and aspirated/non-aspirated stops, and in more recent centuries there has been some palatalization processes and loss of nasal vowels (though the latter are still denoted in writing). The developments in Baltic-Slavic tone after PIE are also an infamous mess.
All in all, when you listen to Lithuanian, you are listening to just any IE language instead of gaining special insights into PIE.
English is already a good progress in that direction, in many (most?) European languages we have gendered nouns for everything, no matter how nonsensical it is. (la baguette / le pain).
And indeed, modern Scandinavian languages have retained gendered nouns.
Though it's gradually being watered down. Eg. in Norwegian Bokmål (there are two official variants of Norwegian, Bokmål is the dominant one) you can now mostly choose whether to use feminine or masculine forms for a noun if a word takes either (but you can't do this for neuter), e.g. en sol (masculine; a sun) vs. ei sol (feminine; a sun) are both valid as long as you're consistent (combining "en sol" (m; a sun) with "sola" (f; the sun) in the same text, mixing genders, is frowned upon, though some of us do that intentionally - "solen" (m; the sun) sounds to stuffy/posh to me, but I prefer "en sol" to "ei sol").
The problem was that the pronouns and declensions didn’t match. Since this was a case of large numbers of adults learning a new language, they just learned enough to get by and didn’t bother with the complicated bits.
English is increasingly using the singular 'they' as a workaround, since alternative options like 've,' 'ze,' 'zhe,' and 'xer' didn't gain traction. This comes with its own set of issues—initially, it was a bit disorienting for me, and even now it often requires a mental pause to parse what the speaker is referring to.
I'd welcome a more authoritarian approach here: globally adopt a new pronoun, go through a transitional period, and then we're all on the same page. Some sci-fi novels have toyed with this concept.
IMO gendered pronouns don't add much value in many languages (with exceptions - poetry etc.). In Slavic languages, it's relatively straightforward to deduce gender from the pronoun, but in Germanic and Romance languages, it's more obfuscated.
I actually love “they” because it waves off every ambiguity at once: not only you don't have to care about the gender of the person, you don't even have to bother whether it is an individual or a collective (this super convenient when discussing about open-source projects).
English is definitely leading in terms of genderlessness, but that advantage is more than offset by its complete detachment between spelling and pronunciation.
It would be nice to have a language with sensible spelling/pronunciation but no gender.
Or you do it the French IIIème République way: you crush all linguistic diversity[0] from your country and standardize it all around the way it's spoken in London.
Also, in English there's so much detachment from spelling to pronunciation than even choosing to write it down like a random[1] dialect would be an improvement: the problem with English isn't so much about “long a” vs “short a”, it's about “through tough thorough thought, though”.
[0]: sure there's still some pronunciation differences between parts of France, but not much. And it has very little consequence on the spelling/pronunciation relationship.
[1]: in practice, it could be the most commonly spoken dialect, or the dialect that has the most pronunciation in common with others).
in practice, it's usually the snob/spoken dialect of the capital city.
See parisians thinking they're better than anyone else, non-french and other-french included.
Im guessing london didn't implement this method because it's useful for them to keep differentiating themselves from the masses now that everyone can have education
And then since pronunciation shifts, you’ll periodically want to have international treaties adjusting spelling, hyphenation, capitalization and other rules, as the German-speakers have done.
Or freeze it once and for all as the French have chosen to do?
I think the pronunciation shifts might be the real issue with English. All languages shift, but I've got the feeling that English shifts a lot more, faster and more dramatically than most languages. And I'm not sure why.
Dutch seems to be quite the opposite: spelling changes a lot. The best way to read old Dutch texts is to sound them out in your head, because the words look totally different, but they sound the same. Hearing medieval Dutch is probably easier than medieval English, but medieval English is much easier to read.
> All languages shift, but I've got the feeling that English shifts a lot more, faster and more dramatically than most languages. And I'm not sure why.
Probably because they go off to colonize very far away places and then develop separately.
I've got the feeling that Afrikaans is more like old fashioned Dutch. It sounds a bit archaic and sloppy to me ears.
I've got the feeling that most of the shifts in English don't come from the colonies, though, but more from mainland Britain. Various city accents rising in popularity and spreading their quirks around the country, like the recent i -> oi shift.
Also, I heard recently that American is actually closer to how English used to be in the 17th-18th century than UK English, which has had more development, including color -> colour and that sort of stuff.
> It would be nice to have a language with sensible spelling/pronunciation
I get the impression that this is impossible in the long term because pronunciation tends to drift, but spelling is more rigid, so over time they mismatch.
Yet there are languages that are much more phonetic than English. In fact, most probably are. But I've been told that Polish is perfectly phonetic without any weird pronunciation surprises. (I don't know Polish, though.)
And spelling is only more rigid in English. Plenty of languages update their spelling over time. Only in English this is useless because there is already barely any connection between spelling and pronunciation, so the break would be far bigger than in other languages.
> But I've been told that Polish is perfectly phonetic without any weird pronunciation surprises.
I don't know about Polish, but I have had conversations where native speakers think that until exceptions are pointed out, because they are unaware.
There's also the issue of accents, where we expect spelling to be universal across the language, but pronunciation in some accents have drifted (ie. it's not a straightforward mapping).
Polish orthography – however bizarre it may look to foreigners at first glance – does map pretty closely to the spoken language, but it still isn’t perfectly phonetic. For example, the reflexive pronoun się is written with a nasal vowel, but whether the vowel is pronounced as a nasal vowel or as a non-nasal vowel differs across regions.
The fact that English preserves the root bindings in so many cases often makes it easier to figure out what a new word means when you encounter it in text. Just writing the sound would abandon this valuable tool.
Anyway, writing the sound just helps the beginning reader. When you read flowing text you don’t actually read the words aloud in your head, letter by letter.
> initially, [they as a singular pronoun] was a bit disorienting for me, and even now it often requires a mental pause to parse what the speaker is referring to.
If it makes you feel better, the use of “you” as both singular and plural pronouns did not become universal until the 60s (thanks to TV amd some school rules); thee and thou remained common in northern regions of England, particularly around Yorkshire, into the 1940s. Yet you use it today without thinking about it.
And I notice the singular “they” in Jane Austin; likely it well precedes her usage as well.
Meh, Finnish doesn’t have noun genders, but due to phonological changes obscuring what was, centuries ago, a regular declension, it now has a large number of different noun classes that have to be memorized as such. These are so laborious for foreigners to learn that a Finnish language workbook like Harjoitus tekee mestarin can spend 300 pages just on drills of noun classes.
Eh, gendered nouns are kinda helpful in spoken language. It's an extra bit of redundancy which can help distinguish similar-sounding words in noisy environs. Same deal with verb conjugations and adjective declensions.
Granted they don't have much of a point in written word, at least not today when writing is easy to preserve digitally without accumulating errors over time. (No smudges making it hard to make out a word)
I really do wonder though how English seemingly uniquely in I-E languages ended up dropping those features, for the most part.
> I really do wonder though how English seemingly uniquely
Not uniquely. Just within the Germanic languages, Afrikaans, an offshoot of Dutch, is without gender, in a way similar to English.
And there are others in the IE family that have lost gender. Persian is a notable example.
English lost its noun gender right after the Norman conquest. The language underwent rapid and extensive change, different populations merged, dialects flattened, and a lot of vocabulary was borrowed from the new French overlords. Many language structures, other than gender, were also lost at that time. Some have argued English almost underwent a sort of creole effect, where the native grammar of the parent languages is lost, and the next generation invent a new heavily regularized grammar to fill the void. That's going a bit far, I think, still there was a great deal of simplification. The lack of gender in Persian and Afrikaans has a somewhat similar history, in terms of social upheaval/displacement/dialect flattening, having an influence on the loss of gender.
I thought this might be a Simon Roper video! I really enjoy his deep, humble and creative views on these topics, even though he still describes himself as only an enthusiast. Speaking of creative, check out his "A London Accent from the 14th to the 21st Century"[1] for a frankly enchanting 18 minutes of viewing.
What's very weird is that it is probably not because the languages are related, but because the same forces that create pronouns, work in a similar way across languages. It's important the sounds be very different, for example, to minimize ambiguity. And they're some of the first abstract words children learn. Something similar is seen with informal terms for "mother" and "father" which have a very strange similarity across languages (e.g. in Chinese, English and Swahili "baba" means father.)
There's something called the "M-T pronoun paradigm", a trace of earlier proto-languages that aren't reconstructible today, possibly from as far back in time as the Stone Age.
The first couple of episodes of the podcast “The History of English” about PIE are fascinating. I’d highly recommend it for anyone interested in a quick intro to PIE (or the rest of the podcast in general, if you’re interested in the history of the English language).
It would be fun to see a video going into depth about the various speculations on how the laryngeals sounded like. I personally think of them as /h/, /x/, /x^w/, but there are so many arguments for the different sounds that people speculate about.
94 comments
[ 2.0 ms ] story [ 181 ms ] threadI can definitely recommend that if you're interested in the topic. A pretty detailed exploration of the Kurgan Hypothesis.
We know much about Proto-Semitic AND Proto-Afro-Asiatic, thankfully, due to the antiquity of recorded Semitic languages in the Near East. Proto-Austronesian seems to be clear too with the 600+ languages we have access to. Unfortunately the languages between the three (Caucasians, language isolates, Tibeto-Burman, Andamanese, Turko-Mongol-Yeniseian Altaic* sprachbund etc) are much more obscure.
Yeniseian (the originator of such words as Tengri, Kha(g)an, etc) being connected to Native American Apache language was pretty amazing and evidence of the importance of paleo-linguistics in finding out informations about our ancestors
This is a bit controversial, though. I haven’t read through the original papers, but one person who did so informed me that they found this relationship not very convincing at all. Still, it’s fascinating stuff!
Vajda’s theory is still regarded as very tentative by the community, and even Vajda himself suggests in his most recent book on the subject that work towards proving a connection has barely begun. It is definitely not settled science.
Don't jump down people's throats based on uncharitable readings of what they wrote.
However I do not believe your comment about the “vast majority” of languages is accurate. I indo-European languages make up about half of all languages on the planet (3.2 billion) and no other language family comes close.
It's so easy to make up stuff regarding language, that sounds reasonable, but has nothing to do with reality.
Video is not a good format for skimming around to get a quick overall impression.
Download the subs.
True but the question asked is answered at 0th second.
The first thing he says is that he’s not a linguist.
See the recent LK-99, Stanford president and Dan Ariely
I really wish we could get over the whole idea that universities and journal have a monopoly on knowledge in the age of the internet.
It really is an antiquated view, literally.
Just to elaborate on the confusing phrasing here... Lithuanian, among PIE languages, is regarded by linguists as having preserved the most archaic features of PIE. So it basically resembles the first branches of PIE (like Sanskrit, but more Proto-Balto-Slavic) more than any other modern PIE language today. So as linguists trace back systematic transformations from other Proto-Balto-Slavic derived languages back to their root words, it resembles Lithuanian more. Btw, I'm not a linguist. I just worked really hard for a 4.0 GPA, even for electives.
I'm not sure that's true. I mean Baltic languages are generally considered to be one of the last to diverge from PIE but that doesn't mean the sound that similar. They preserve many archaic features but there has also been a lot of innovation over thousands of years (especially considering that modern standard Lithuanian is to some degree a constructed language).
While certain grammatical features like almost the whole case system have been preserved too a higher degree than in other IE languages. Lithuanian has still lost laryngeals, aspirated stops etc. which means it sounds very differently to what proto Indo-European might have sounded.
No you can’t. Especially not the “sounded” aspect. Lithuanian is conservative in some matters of lexicon and morphology, but modern Baltic phonology is very different from PIE. The ancestral stages of Lithuanian after PIE lost the PIE laryngeal sounds, then the distinction of three classes of velars and aspirated/non-aspirated stops, and in more recent centuries there has been some palatalization processes and loss of nasal vowels (though the latter are still denoted in writing). The developments in Baltic-Slavic tone after PIE are also an infamous mess.
All in all, when you listen to Lithuanian, you are listening to just any IE language instead of gaining special insights into PIE.
[1]: https://en.wikibooks.org/wiki/Old_Norse/Pronouns
Though it's gradually being watered down. Eg. in Norwegian Bokmål (there are two official variants of Norwegian, Bokmål is the dominant one) you can now mostly choose whether to use feminine or masculine forms for a noun if a word takes either (but you can't do this for neuter), e.g. en sol (masculine; a sun) vs. ei sol (feminine; a sun) are both valid as long as you're consistent (combining "en sol" (m; a sun) with "sola" (f; the sun) in the same text, mixing genders, is frowned upon, though some of us do that intentionally - "solen" (m; the sun) sounds to stuffy/posh to me, but I prefer "en sol" to "ei sol").
I'd welcome a more authoritarian approach here: globally adopt a new pronoun, go through a transitional period, and then we're all on the same page. Some sci-fi novels have toyed with this concept.
IMO gendered pronouns don't add much value in many languages (with exceptions - poetry etc.). In Slavic languages, it's relatively straightforward to deduce gender from the pronoun, but in Germanic and Romance languages, it's more obfuscated.
It would be nice to have a language with sensible spelling/pronunciation but no gender.
Whose dialect are you going to choose to render phonetically? Mine from the West Country, or that of my friends from Newcastle?
Will you standardise on short a in castle (Newcastle upon Tyne, North East England) or long a in castle (Restormel Castle, Cornwall)?
Or should each person just write phonetically so that we have millions of idiolects?
Also, in English there's so much detachment from spelling to pronunciation than even choosing to write it down like a random[1] dialect would be an improvement: the problem with English isn't so much about “long a” vs “short a”, it's about “through tough thorough thought, though”.
[0]: sure there's still some pronunciation differences between parts of France, but not much. And it has very little consequence on the spelling/pronunciation relationship.
[1]: in practice, it could be the most commonly spoken dialect, or the dialect that has the most pronunciation in common with others).
See parisians thinking they're better than anyone else, non-french and other-french included.
Im guessing london didn't implement this method because it's useful for them to keep differentiating themselves from the masses now that everyone can have education
Or freeze it once and for all as the French have chosen to do?
Dutch seems to be quite the opposite: spelling changes a lot. The best way to read old Dutch texts is to sound them out in your head, because the words look totally different, but they sound the same. Hearing medieval Dutch is probably easier than medieval English, but medieval English is much easier to read.
Probably because they go off to colonize very far away places and then develop separately.
Isn't Afrikaans pronunciation shifted Dutch?
I've got the feeling that most of the shifts in English don't come from the colonies, though, but more from mainland Britain. Various city accents rising in popularity and spreading their quirks around the country, like the recent i -> oi shift.
Also, I heard recently that American is actually closer to how English used to be in the 17th-18th century than UK English, which has had more development, including color -> colour and that sort of stuff.
I get the impression that this is impossible in the long term because pronunciation tends to drift, but spelling is more rigid, so over time they mismatch.
And spelling is only more rigid in English. Plenty of languages update their spelling over time. Only in English this is useless because there is already barely any connection between spelling and pronunciation, so the break would be far bigger than in other languages.
I don't know about Polish, but I have had conversations where native speakers think that until exceptions are pointed out, because they are unaware.
There's also the issue of accents, where we expect spelling to be universal across the language, but pronunciation in some accents have drifted (ie. it's not a straightforward mapping).
Anyway, writing the sound just helps the beginning reader. When you read flowing text you don’t actually read the words aloud in your head, letter by letter.
If it makes you feel better, the use of “you” as both singular and plural pronouns did not become universal until the 60s (thanks to TV amd some school rules); thee and thou remained common in northern regions of England, particularly around Yorkshire, into the 1940s. Yet you use it today without thinking about it.
And I notice the singular “they” in Jane Austin; likely it well precedes her usage as well.
It's in Shakespeare.
(If you go back to Chaucer, "girl" is gender neutral for a child.)
There are a lot different classifier systems in language around the world; gender is just a small example.
All these mechanisms are kinds of ECC.
Granted they don't have much of a point in written word, at least not today when writing is easy to preserve digitally without accumulating errors over time. (No smudges making it hard to make out a word)
I really do wonder though how English seemingly uniquely in I-E languages ended up dropping those features, for the most part.
Not uniquely. Just within the Germanic languages, Afrikaans, an offshoot of Dutch, is without gender, in a way similar to English.
And there are others in the IE family that have lost gender. Persian is a notable example.
English lost its noun gender right after the Norman conquest. The language underwent rapid and extensive change, different populations merged, dialects flattened, and a lot of vocabulary was borrowed from the new French overlords. Many language structures, other than gender, were also lost at that time. Some have argued English almost underwent a sort of creole effect, where the native grammar of the parent languages is lost, and the next generation invent a new heavily regularized grammar to fill the void. That's going a bit far, I think, still there was a great deal of simplification. The lack of gender in Persian and Afrikaans has a somewhat similar history, in terms of social upheaval/displacement/dialect flattening, having an influence on the loss of gender.
[1] https://youtu.be/3lXv3Tt4x20?feature=shared
one, eins, ek
two, zwei, do
three, drei, teen
four, vier, chaar
five, funf, paanch
six, sechs, chhah
seven, sieben, saat
eight, acht, aaath
nine, neun, nou
ten, zehn, das
Seems related to me! Source:
https://l-lingo.com/free-lessons/en/learn-german/numbers-1-1...
https://www.mindurhindi.com/basic-words-and-phrases-in-hindi...
1 – yksi 2 – kaksi 3 – kolme 4 – neljä 5 – viisi 6 – kuusi 7 – seitsemän 8 – kahdeksan
But it almost feels as if its related a little. Or its my brain playing tricks.
English Russian Finnish Hungarian
me, menya, mina, en
we, my, me, mi
thou, ty, te, ti
he, on, hän, ő
What's very weird is that it is probably not because the languages are related, but because the same forces that create pronouns, work in a similar way across languages. It's important the sounds be very different, for example, to minimize ambiguity. And they're some of the first abstract words children learn. Something similar is seen with informal terms for "mother" and "father" which have a very strange similarity across languages (e.g. in Chinese, English and Swahili "baba" means father.)
New insights into the origin of the Indo-European languages - https://news.ycombinator.com/item?id=36930321 - July 2023 (148 comments)
This is a good channel