This research is built on some pretty shaky ground—including some very loaded picking-and-choosing of vocabulary—and because it doesn't have very strong predictive power, hasn't made much of a dent in the community in the last decade since it's been published. Here's a discussion from the blog Language Log that discusses a lot of the methodological problems that show up: https://languagelog.ldc.upenn.edu/nll/?p=4612
Nuts, you beat me to it; I was going to provide that exact same link, together with an excerpt from a comment: "The LWED database reconstructions just aren't good enough to form the basis of a successful effort along those lines."
Aside from the specific details, I feel like papers of this nature would be on much firmer ground if they included controls. In this case for example, that would mean showing that are greater word-correspondences in these eurasian language families than with other language families.
What is interesting is that Chinese is not I this superfamily.
For me, ancient Chinese civilization is so interesting in large part because of how isolated it is. Mesopotamia, Egypt, Persia and even India were in regular communication with each other. Empires such as the Persians and Alexander the Great fought across all these civilizations. The Romans had direct trade with India via the Indian Ocean.
However, Chinese civilization is located far away from those ones. I think there is still debate if Chinese writing rose de novo or was influenced by writing from Sumer.
At the same time, even though it was so distant, China still had a very advanced civilization (and in many ways was more advanced than those of the Middle East and Europe for much of history)
But ancient China wasn't isolated at all. Contact with Persia was extensive. Contact with India was significant. Han China interacted with Greeks left behind by Alexander's conquests: https://en.wikipedia.org/wiki/Dayuan . The Japanese days of the week today are clearly a cultural transmission from the West, mediated through China.
> I think there is still debate if Chinese writing rose de novo or was influenced by writing from Sumer.
I'm not aware of a theory that says Chinese writing was influenced by Sumer. It's generally felt to be its own thing. (Though note that Sumerian is about 2,000 years earlier than our oldest Chinese records, so as a theoretical matter there's no real way to rule it out. There just isn't any evidence of influence.)
> For me, ancient Chinese civilization is so interesting in large part because of how isolated it is.
For the Europeans this might be true, but not for the Indians. There was lot of exchange between Indus and Chinese civilisations. How do you think Buddhism reached China?
Buddhism reaching China was in the 1st century CE.
The Persians ruling over Egypt, the Punjab, and Mesopotamia was 509 years before that.
And a 1000 years before that (1500 BC) the Mittani who had very similars gods and royal names that are very similar to Sanskrit had an Empire in the Middle East.
Also, the Maury Empire had close links with the Seleucid Empire (the war elephants used in many of the Seleucid campaigns came from India)
The Indo-Chinese links as can be seen from history are not nearly as old or as well attested.
If we're comparing European contact to Indo-Chinese contact, your sense of when Europe was integrated with the Middle East/Mediterranean culture seems a bit off. The Persia, the Mediterranean and the Black Sea had a lot of contact prior to the 1st Century but "Europe" as a whole did not. For example, England was not Christianized until the 7th Century, Kievan Rus was the 9th Century, Scandinavia was started in the 10th Century, and the Baltic Crusades lasted until the 16th Century. Cultural exchange in "Europe" absolutely is newer than contact between India and China.
No, I'm saying one comment is saying contact can be seen by spreading of religion, and therefore Chinese and Indian contact were later than European spread. So, looking at the criteria of the above comments, taking them post at their own standards, which is spread of religion, Europe as an entity did not have "contact" in its entirety until a thousand years after China and India.
History and society in western countries is absurdly Euro-centric, it sounds like an oxymoron but it should not be. You must be aware enough to realize what is an absolute truth and what it is just relative from your POV. The other day I had a discussion online mocking the concept that "Aboriginal Australians are so unique because they split from the Europeans first" as if they cannot say exactly the same.
At least in the English language, many, many subjects are less Euro-centric and more Anglo-centric: everything is seen from the vantage of the south-eastern part of a tiny island in the North Atlantic. What's so dark about the Dark Ages? Outside of the north-western boundaries of the Roman empire, not much.
The only reason Chinese is not in this superfamily is that the authors didn't bother to feed their algorithm with any Sino-Tibetan data. If they had done so, it would've been placed somewhere in the "family tree".
>I think there is still debate if Chinese writing rose de novo or was influenced by writing from Sumer.
Hardly. There's no discernable relationship in that respect between China and the Near East. As far as anybody's aware, China's writing is indigeneous.
If you want a real isolated and independent civilisation core, look at central America.
There was travel and commerce between far out regions in that period (1,500-3,500 BCE). The connection between the three old-world starts of writing (Sumerian, Egyptian and Chinese) is, AFAIK, not 100% clarified yet. It's widely accepted that American writing was a separate original invention, it makes sense. I believe these days the accepted understanding is that Egyptian hieroglyphic writing is an independent development that may, or may not, have been inspired by Sumerian writing. Not hard to believe: "you know son, these peoples far far away know magic, they're able to make dents in a table containing a story, and then pass on the tablet to someone else and they can retell the story!" Chinese writing may, or may not, have also come up the same way.
I have a theory that if the East and West are ever going to reconcile their differences, we must find a way to retrace how these two great civilizations separated in the first place. What caused our societies to drift apart? Was it primarily circumstantial factors, like geography and competition for limited resources? Did religion play a role, like the recent East-West schism, where early societies self-segregated according to monotheist and polytheist beliefs? Was it driven by language, where the East and West favored certain modes of expression, eventually manifesting as logographs or alphabets? Did it originate in primordial social contracts, where early famers and shepherds drove nomadic hunter gatherers further to the North and East? Probably the original motive lies buried in the sands of time, but I think these are fascinating questions and would be curious to know what answers future historians might come up with.
Take a look at a globe. Where does east and west start and stop?
When you find yourself in a binary quandary, you are nearly always going to be arguing yourself into complete twaddle. Think: East:West, Black:White, North:South, Man:Woman, "Normal": ... . That way lies madness.
It is trivial to slide into exclusiveness. I try to be inclusive. Back in the day I was the fat child left on the sidelines when picking footie teams. Then I went to a different school that also offered rugby as well as football. That's an odd form of inclusivity but I found my place for a while: I had quite a lot of extra inertia and kinetic energy and discovered how to apply it quite efficiently.
Back to your comment, which I don't think needs down votes but questioning. Where do you perceive East and West colliding? Turkey?
Perhaps. I think a clearer understanding could also give us a better appreciation of our common values and unique differences. There is something we all share in common: a desire for peace, order, stability. And there is something qualitatively different between how Eastern and Western societies are structured to achieve those goals. By studying our historical differences, we can build a better understanding of how our societies are uniquely equipped to handle future challenges.
> I had quite a lot of extra inertia and kinetic energy.
That is an interesting story, thank you for sharing! I agree with you that physical differences are definitely worth celebrating. Fat people have an advantage in many arenas (can survive much longer in famine and are good at physical sports like rugby), just like people who live in high altitudes or chase food to survive. More importantly, humans also have ideological differences and it would seem those differences, real or perceived, are the source of our current tensions.
> Where do you perceive East and West colliding?
Historically, friction has arisen in two places: the Middle East, and the Pacific. The first seems more religiously motivated, which could be a reflection of different cultural values but I think those waters are too muddy to tell. The recent cross-Pacific tension seems like a more direct clash between individualism and collectivism. Studying the environmental factors that gave rise to our differences could help us navigate where each society is better adapted, or how they can complement each other to achieve our common goals.
I also find the Chinese writing system fascinating in that it's the only one that didn't evolve from representing concepts to representing sounds. Maybe there is a connection there?
The same thing happened in Japanese but both countries refrained from taking the next logical step and throw away the ideograms. Much of it has to do with the educated elite wanting to restrict access to knowledge, but it's still curious that the same thing didn't happen in other cultures.
Seriously, it's about cultural continuity. Its not about wanting to restrict access to knowledge; in general the educated elite wishes the illiterate masses would be more educated, not less.
But for example, there is a period in Egyptian history when royal inscriptions start being written exclusively in vernacular Egyptian rather than the (incredibly old) classical form. And that period just happens to be when the Egyptian throne is taken over by Libyans.
I’m not an expert on the matter but, from what I know about Korea and Japan, literacy was a priviledge granted to a small minority. Having a complex script that takes years to learn was a tremendous advantage to them. Maybe it was different in China, and things have certainly changed in modern times, but low literacy has often been beneficial to those in power.
Cultural continuity is questionable, even old texts in European languages often need to be adapted for contemporary readers.
All that said, Chinese ideograms are incredibly cool so maybe that is a good reason to keep them.
> but, from what I know about Korea and Japan, literacy was a priviledge granted to a small minority.
I have heard that a contemporary objection to Hangul was that it would allow anyone to learn to read. I don't know all that much about it either -- in particular, I don't know how much force that objection ended up having compared to other objections -- but it certainly supports your view.
There are two conflicting principles here, which we can observe within the Western European tradition:
- The elites are interested in keeping the masses under control.
- The elites are interested in ensuring that the masses' idea of virtue matches that of the elites.
Education promotes that second goal while arguably undermining the first one. So e.g. we see some Christian traditions that perform religious services in Latin and some that perform them in the local vernacular; we see some traditions that perform services in the vernacular but discourage reading the Bible; and so forth. There is a tension that different traditions resolve in different ways.
But writing systems are mostly about conservatism. To change the writing system, you need an explicit reform effort to succeed. Otherwise it will just keep going the way it is for no other reason than that that's how you write.
> Chinese ideograms
As a side note, Japanese kanji are much closer to being ideograms than Chinese characters are. The first and most important thing you can know about a Chinese character is its pronunciation. The semantics are present to a much greater degree than in most other languages, but they are not dominant. In Japanese, a character doesn't really have a pronunciation by itself, only readings that are specific to particular words.
Cultural continuity isn't just about old texts, but also recent texts in the old style, that people learning to read today want to be able to understand.
> in general the educated elite wishes the illiterate masses would be more educated, not less
Literate masses are more likely to challenge the dominance of the educated elites. In Korea, for example, the yangban pushed back against Hangul primarily for this reason.
The word "ideogram" has fallen out of fashion because it's a very poor description of Chinese characters. A Chinese character doesn't describe an idea: it describes a concrete word, with a definite pronunciation. It's just that the relation between the shape of the character and its sound is much more obscure than most scripts.
An example of an actual ideogram is the exit sign (you know, a green man running out of a door) - it conveys the idea without having a canonical associated word.
The case is not that strong: for example, nobody writes "It's hard to find out how to <exit icon> vim." The icon represents the concept of "a door used for emergency exit", not the word "exit".
I think we're talking about different things. I found an image of the sign you're talking about by searching for "exit sign", but I wouldn't have called it an exit sign and wouldn't really understand what it's supposed to mean.
To me, an exit sign is mounted on the ceiling and just says "EXIT".
Best not let any linguists hear you say "representing concepts to representing sounds". All languages represent sounds as far as I've ever heard from them, and the idea of "representing concepts" is regarded as a remnant of some very broken ideas. :-)
As far as I (a non-linguist, although I've read a bit) know, there are roughly three ways of writing: alphabets, syllabaries, and whatever Chinese is (logosyllabic?).
Alphabets have one grapheme per (roughly) phoneme and have(AFAIK) developed once: Proto-Siniatic, which led to Semitic (Hebrew, Arabic, etc.), Phoenician (which developed into Greek, Latin, etc.) and a few others. Every other alphabet is a direct or indirect descendant. They typically have <50 graphemes.
Chinese is kinda-sorta syllabic, but the language has a great many one syllable words which have a grapheme mapped directly to that use. On the other hand there's things like "coral" (IIRC) which is two syllables and is written with two characters each of which are not used anywhere else. It, its descendants, and any other similar languages if there are any, has some many thousands of graphemes.
Syllabaries have one grapheme per syllable and are the most common form of writing, to the extent that it's pretty clear that they're the normal version of human writing. They typically have a few hundred graphemes.
Mayan and cuneiform are a couple of weird cases. They (AFAIK) are mostly syllabic, but with some logographic-ish parts like Chinese. But the number of graphemes are pretty firmly in the syllabary range.
Tl;dr: Writing is weird and the writing I'm doing now is very much so.
I’m not sure I understand your comment. Granted, I’m not a linguist but there is certainly something unique about the Chinese script.
I don’t know a single word of Chinese, but I followed a link in a sibling post and came across this: 女書 - which my meager knowledge of Japanese allowed me to understand that it means “women’s writing”.
What is the proper name for the ability to read text without knowing how to pronounce it?
This is starting to sound like that debate between Dennett and Chomsky about whether recursion is a universal feature of languages. Linguistics is a strange field.
There are three undisputed independent developments of writing: Sumerian, Mayan, and Chinese. Egyptian may or may not have been influenced by Sumerian; Indus Valley Script may or may not be writing; and Andean quipus (essentially knotted strings) raises interesting questions about what writing even is. All other (known) writing systems developed out of knowledge of these priors.
The development of writing systems seems to be pretty clear, especially because we can use Sequoyah's documented development of Cherokee script as a check. Writing seems to start as logographic: each character starts out as representing a word, often as a stylized representation of that word (for example, draw an eye to represent the word 'eye'). To represent words that don't have easy representation, a rebus system develops: you use homophonic puns (so an eye might also stand for 'I'). This lends itself to a simplification to a syllabic representation, one word for each possible syllable. And usually, it stops there (although note that 'syllable' usually comes out in writing as a consonant-vowel pair, rather than a full syllable).
It's with the Semitic languages that things took an interesting turn. Semitic languages have an interesting feature in that you can get most of the sense of a word even if you drop the vowels. And the writing ended up switching from consonant-vowel pairs to just consonants (what we now call an 'abjad'). Greek opted to adapt this by adding letters for vowels and formed the first 'true alphabet'. Sanskrit opted instead to use reliable markings on consonants to make an 'abugida'.
Chinese never developed their logographic system into a syllabary, although this was probably partially driven by the fact that Chinese itself had most of its words become monosyllabic at the same time. Japanese did continue Chinese script into two full syllabaries.
Korean is the other independent invention of an alphabet: Hangul. Although Hangul is arranged as a succession of blocks representing full syllables, so it can also be viewed as a systemic construction of a syllabary rather than an alphabet.
I mentioned Sequoyah's development of Cherokee script: he knew of writing from the Americans at the time, and even had access to a Bible (although he couldn't read it at all). And his own documentation of his development does expressly indicate that he started trying to build a logography, realized that there were too many symbols that had to be created, and instead simplified it into a syllabary, which was much more conservative in its character count. (And the influence of several character shapes came from that Bible, which is why Cherokee script looks like someone started with Latin script but had no idea had it worked--here is its character set: ᎠᎡᎢᎣᎤᎥᎦᎧᎨᎩᎪᎫᎬᎭᎮᎯᎰᎱᎲᎳᎴᎵᎶᎷᎸᎹᎺᎻᎼᎽᎾᎿᏀᏁᏂᏃᏄᏅᏆᏇᏈᏉᏊᏋᏌᏍᏎᏏᏐᏑᏒᏓᏔᏕᏖᏗᏘᏙᏚᏛᏜᏝᏞᏟᏠᏡᏢᏣᏤᏥᏦᏧᏨᏩᏪᏫᏬᏭᏮᏯᏰᏱᏲᏳᏴᏵᏸᏹᏺᏻᏼᏽ)
There are (at least) a couple other languages where some local genius saw writing and developed a syllabic writing system: Vai (West Africa) and a language of Southeast Asia whose name I can't remember (the invention was fairly recent). Maybe also Yi (spoken in China, but written in an entirely different script).
I am doing a Computer Sceince masters in this field.
I am personally quite sceptical of this study. The methodology seems somewhat arbitrary and I doubt that the signal is there for what they are inferring from the data. I suspect this is a case of finding what you want to find.
If you find this topic interesting, I highly recommend John McWhorter's course, The Story of Human Language. It's available on Audible / The Great Courses. I learned a lot and was fascinated throughout.
I'd also recommend Lyle Campbell's Historical Linguistics: An Introduction for historical linguistics especially. It's a very concise introduction and I believe you could understand most of it without a deep background in linguistics (maybe some understanding of the IPA and phonetics mostly)
The big secret of history is that there was far more mixing between disparate cultures than is currently let on. There are many reasons that this is not widely accepted into the historical cannon, the main one being that some cultures most esteemed figures and events were catalyzed or can be completely attributed to another group of people/culture.
General examples being, some sea faring civilization A far in the past sailed 2000 miles, mixed with some other civilization B, which catalyzed notable events there, which are now attributed to that civilizations B greatness. In schools the children are taught of those events, but evidence supporting a root in civilization A is suppressed or even destroyed.
The beauty of your statement is that it will become more and more provable through fine grained DNA analysis. The history of Neolithic Europe has been determined by recent (<10 years) results of multiple waves (almost) repopulating the old hunter gatherer stocks. There may grow a new school of linguists who deeply incorporate these waves with layers of language.
Interestingly the arrival of horses, chariots and advanced metallurgy in China did not leave much of an impression on the gene pool, at least not detected (detectable?) by current data/methods.
I had a hard time parsing what this actually means in terms of which words are actually connected. I speak Hungarian, which is a Uralic language supposed to have split from Finnish roughly 3000 years ago. I've seen a few instances in a dictionary where the common roots between the two are connected, though it rarely seems connected from casual observation. People often think there are some shared roots between Hungarian and Turkish, due to more recent occupation of Hungary by the Turks, which led to many Turkish loan words, but the connection is mostly superficial.
Hungarian also has a few loan words from Latin mostly from the last couple Hundred years, but I'd be fascinated to learn about any connections between Hungarian and Latin via some shared ancestral roots.
I find this absolutely fascinating, especially in the context of emergent technology. One could speculate: will persistent interactive mass media platforms eliminate or vastly reduce the rate of word replacement? Or will they accelerate it? A competent linguist might be able to give a good opinion but even so it seems debatable. Maybe linguistic change is contingent on the sorts of traumatic cultural divisions that were commonplace before the advent of global culture and technology, or maybe language changes more when more people talk to each other. Is language change accidental and traumatagenic or creative and intentional?
Not really related to this paper, but something that really amuses me is that the two words 'aré?!' ('huh?!') and 'nai' ('is not there') have exactly the same meaning in both Japanese and Bengali. Two languages that you would not expect to be anywhere close to each other.
Ultradeep relationships between languages are to linguistics as perpetual motion machines are to physics. Everybody wants to let you know why their theory works better than everybody else's.
There's a good page on zompist discussing the question of "just how likely is it that all these similar words are nothing but coincidence?"
"ma", "pa", and "da" are all among the first sounds a babbling baby typically learns to make.
(As an aside, in my native language Finnish the most common word for mother, äiti, is notably not a "ma" word. It's apparently a loan from some Proto-Germanic word that has itself disappeared from extant Germanic languages. However, Finnish does have its own ancient "ma" word that means "mother": emä or emo, which is still in use when referred to non-human animal mothers, in poetry, and in a figurative sense in compound words such as emolevy "motherboard" or emäalus "mothership".)
For the benefit of future readers who don't get it, if nothing else: the "salmon problem" (in German Lachsargument) is an old argument that the origin of the Indo-European language family should be in the Baltic region, because the word for "salmon" is found in both the Germanic and Balto-Slavic branches. Later research in the Caucasus found that the reflexes of the word referred to trout, which is found in rivers on the Eurasian steppe. (They're both species in the genus Salmo, so it makes sense that semantic drift would occur in populations moving to a new area.)
