Ancient history is fascinating- trying to trace the origins of civilization to answer where do we come from. I just think that we need to concentrate our efforts more on the subject of the original first language, what I like to call the "mother tongue".

If you really think about it, by finding traces of the mother tongue we could effectively answer the question of where and when did civilization really start. Now I must point out I am not a linguist but the thought of tracing words through different civilizations is certainly intriguing.

For example, take the word father in English, vater in German and padre in Spanish. It indicates of an ancient original root.

It is widely accepted that the predesecors of the European languages are mainly Greek and Latin. Furthermore, Greek and Latin is closely related with Sanskrit- the so called proto (first) European languages. The Indo-European language group is one of the biggest however there is also the Proto-Semitic (Jewish, Arabic). Now imagine if we can compare the two and find the root of the original mother tongue and its origins.

I've got to rush off, but you might try looking under "linguistics" and "anthropology" in Google. Yes, we anthropologists have done this and have mapped out language families. They show the migration of people across the world.

Did you know that we can, by analysis, figure out how long ago a language family group split from another family group?

Fascinating stuff. You might also read some of the Noam Chomsky books on linguistics.

The development of Indo-European isn't quite as linear as you suggest. Proto-indo european split into several major groups, which then further developed and as distances between tribes increased to give the various languages we have today. The language groups include: Celtic, Germanic, Romance (this includes latin), Slavic, Baltic, Albanian, Greek, Armenian, Iranian & Indo-Aryan. These are the surviving groups today, there's a couple obscure extinct groups as well. As an example, slaving langueas (eg, Polish and Russian) are most closely related to Baltic, and then to the Iranian & Indo-Aryan groups, while being rather distantly related to Latin. This is not very surprising, the precursors of the Romans were in Europe for a very long time (over a Millenium into B.C... can't say how much further beyond that with certainty). On the other hand, the Slavs only arrived in Europe a few centuries after the beginning of the current era.

So Indo-European languages certainly didn't all come from Greek and Latin. If we're talking about English, then the grammar is more Germanic, but it definitely also had massive Romance influence. The point is the development of Indo-European is more like a branching tree with a significant amount of interconnection further along the line.

But you're right, it is interesting to trace the way the original language branched out. For example, a curious feature of Indo-European is the development of the Centum and Satem languages. This would have been the first major split in the Indo-European family. Centum includes Romance, Greek, Celtic and Germanic... while Satem includes Slavic, Baltic, Indo-Aryan, Iranians and Albanian.




There are several interesting distinctions between these two major groups but one striking thing that should be noted is the Satem shift of palato-velar stops to fricatives (and affricatives, but all articulated at the front of the mouth). Where as in Centum the palato-velar stops were merged with plain velars.

Best to illustrate this with an example :-/. The original indo-european word for 100 is '*k^mtom". In the Avestan language (closely related to vedic, a form of sanskrit), the sound the palato velar '*k^' shifted to an 's', so in Avestan, 100 is 'satem'. Where as in latin the '*k^' become a plain 'k' hence it's word for 100 being 'centum'. This begins to explain how a word as different as a 'hundred' in English and 'sto' in Russian can be so closely related (the 'h' in English's 'hundred' is actually a later consonantal shift, described by Grimm's law).

Languages tend to do curious things too Let's use a slavin example, btw Slavic is a nice language group to search for changes because it diverged less than a millennium ago... so it's languages are a lot more similar than say Germanic. Proto-slavic for example had 'softened' consonants (easiest way to explain that is a consonant with an 'ee' sound at the end, like in english been). This is still present in modern Russian, while it has been abandoned in most other slavinc languages, it was even further exagarated in Polish with further softening where the softened consonant became palatalised. For example, the soft 'd' sound became palatalised into a soft version of the english 'j'. Other examples include soft 't' into soft 'ch' etc. The problem was that slavic already had a 'ch' sound (similar to the english one), to distinguish it from the new soft 'ch' a new set of consonant shifts occured making sounds like 'ch' 'j' and 'sh' a lot stronger. In fact, the only other language(s) that distinguishes soft and hard versions of these sounds is Chinese (here the distinction is even more pronounced). A much simpler but similar shift also accured in Greek, as it had no 'sh' sound at all, as a results, its 's' sound became ever so slightly palatalised to fill in this gap... that is why the Greek 's' when spoken by a real Greek sounds neither like an 's' or an 'sh' but something in between. Although curiously a native Greek speaker will swear that when they say 's' it sounds exactly the same as the 's' pronounced in English.

Like I said, there's also a lot of exchanges between languages... so there isn't just branching. For example the origin of Japanese remains as ellusive as ever. On theory see it being brought to Japan (superseeding the original native tongue of the earliest japanese natives) through Korea, and even earlier from south asia. This would explain why its grammar is somewhat similar to the Altaic language group (the settlers would have picked it up traveling near the Korean peninsula in the north), and why many of its words and early customs seem to be derived from the inhabitants of south asia. This is rather interesting because it would imply mixing not just within language families, but between them!



Anyway, I'm sure you get the idea



All these linguistic changes are quite interesting to trace, and can tell you a lot about the origins of, migrations and even early social development (an example often quoted here is that Proto-Balto-Slavic borrowed it's word meaning 'God' from proto-Indo-Iranian... which has led some to imply an exchange of concepts not just word names).


There is a problem tracing things all the way back to your proposed mother tongue however. All the changes and shifts and what not can be traced all the way to the original languages for each language family. There are many language families, for example: Indo-European, Hamito-Semitic, Ural-Altaic, Mon-Khmer, etc.

Over a century ago a British guy proficient in linguistics was staying in India and noticed the similarities in grammar between Indian and European languages. He also became intrigued by the vast grammatical differences he noted in semitic. This is more or less where modern linguistics was born, because until then people had only been interested in tracing European languages.


It should be noted the grammars of language families are vastly different. Let's look at some examples:

Indo-European languages conjugate their verbs by changing the endings. For example: drive, driving, drove etc. All of these are conjugations of "to drive". However, in semitic languages (such as Hebrew), conjugation works completely differently. Here you take a root verb such as SGR (which roughly means 'to close') and you conjugate it by changing the vowels that go in between the consonants. So for example: SoGeR - I close, SaGaRta - you closed, SiGoR - we will close.
Chinese is another contrast yet again, there is absolutely no conjugation in mandarin... instead it relies heavily on 'empty' words (like englih 'of', words which have no meaning on their own but indicate how the elements of a sentence relate to each other). Mandarin has no tense, no number, etc... so there is no way to distinguish 'rock' and 'rocks' to achieve the latter you have to say something like 'many rock'. There is no 'I write' and 'I wrote', but you can say something like 'I write yesterday'.

Khmer is interesting because there are no personal pronouns like in european or semitic languages. There is no simple 'you' or 'they', they all depend on who says the pronoun, and who its directed at. On the other hand Khmer has a lot of terms which might seem very broad and general but are in fact quite precise if their context is taken into account.

Japanese again has a grammar quite alien to european languages. There is a lot of detail involved, but when one looks at a Japanese sentence you will notice that the order in which the subject, objet and verb are placed is quite strange. In English you would say "I eat food" so subject then verb then object. But in Japanese the order will end up being subject, object, verb. Meaning the details of the subject and objet are specified first, and finally the action connecting them is specified last. A sentence structures that says: "I food eat" is quite different. This is an oversimplification but you get the idea.

In Japanese you don't use tone of voice as you do in european language. For example you don't raise your voice at the end of a questioning sentence, instead you add '-ka' to indicate the question. Japanese is bitonal and its intonation is quite dull compared to something at the opposite extreme like Italian. Chinese takes it a step further, it doesn't just use tone of voice to indicate how we make a statement... the tone of voice actually indicates which particular word you're saying. So we can take two chinese words which would sound exactly the same to a european speaker, but they will have a completely different meaning in chinese because you say raise the voice when saying one, and drop your voice when saying the other.



The point is the differences are vast. So vast in fact that I thought linguistics did not see any connection between families at all. Now if my info is up to date one can not trace beyond the language family barrier. Byrd, I'm wondering if you can either confirm or deny that. And if you can trace further, then I'd be very curious as to how its done

Oh, and I've probbably made some errors in what I've written so feel free to correct (though that migh be a bit of an overwhelming task)

[Edited on 26-5-2004 by hetman]

Great post, hetman. You got a vote for WATS.

Something I find interesting is that latin used to have more forms that it has now. The classical version we read basicly has five: nominativus, genitivus, dativus, accusativus, ablativus. You can identify two other cases in some very rare cases: the vocativus (only in singular male words of the -us group) and the locativus (only in islands and cities). These two forms are left over for original indo-european languages.

Great posts here people. Nice to see such debate on linguistics. What I am trying to point out tough is the connection ever so subtle between the proto languages of the aforementioned groups in your posts.

Another example I stumbled across is certain similarities between ancient greek and chinese. Now on the surface these two languages have nothing in common- check out these words though and you can draw your own conclusions. (Note thanks to my chinese friend I was able to draw comparisons).

Greek Chinese
hygea (health) yu (to heal)
goulia (sip of water) guliao (a little water)
yios (son) yi (son, descendent)
iskios, skiazomai (shadow, to be afraid) xia (to scare, shadow)
lera (monkey) lei ren (monkey)

Now, I understand that there are vast differences between various language families, what I am trying to pose though is that there has to be a connection between all the proto languages, if we find that we might be able to establish even the origin of the mother tongue.

Semitic, Akkadian, Ugaritic, Hebrew, Arabica. Is called AfroAsiatic group, also known as the Hamito-Semitic language family are on the most older known languages.

There is a common link between all the proto families, it all lies within the root of the word as well as the perceived meaning.

I didn't realize how long this thread was so I have to go right now, but later I will be posting a few links between European and Native American languages. (I think it specifically compared Slavic, German, English and Navajo... but i could be mistaken). There were striking similarities and the effect of it was basically that Africans and Europeans, rather than Asians, settled the Americas. It also points out that a Tower of Babbel story immediately follows the deluge story in certain south american traditions. It all adds up to either a very strong trans-atlantic link, or perhaps a common source of the people from those regions.

Great work vagapond, that is what I am trying to establish- the common link between our proto lanaguages.

The thing is that when it comes to the way language changes over time, the similarity in the way words sound is the first thing to go! This is not because only individual words are changed, in which case perhaps it would be possible that some words remained the same. It's also because shifts in entire sets of sounds occur (like I showed above). That means that even withing a single language family, individual languages use vocabulary that is usually unrecognisable to speakers of another language in the same family.

In other words, if you're trying to find links between Chinese and Greek, isolated cases of similarity between the pronounciation of individual sounds is not going to give you any indication of relationship.

If you wanted to find links between families, perhaps you'd be looking for similarities in grammar or something. But similarities in grammar doesn't necesarily show common ancestry... could be just a common way in which the human brain works. It's tricky stuff this

Well that is precisely the point it is tricky but think of the possibilities. I have located my linguistics prof, we are meeting for a catch up coffee (lol he is also my client) so I will ask for some guidance on this. Give me till tomorrow morning aussie time (it is afternoon now) and I will get back to you.

I think you missed my point. I was saying that any similarities in the pronounciation of individual words between languages in entirely different families can have one of two possible origins. First, pure chance. Second, cultural exchange.

Keep us posted about the meeting with the lecturer though. That should be interesting

Fascinating thread

the Hungarian Language is interesting for its relation to the Sumerian language ( of all things!) they share many common roots and legend(s) of a sumerian migration to what is now Eastern Europe. language is a great tool to trace our origins.

I just had a most fascinating discussion with my ex linguistics prof. He was adamant that there is a very strong possibility that there is indeed a proto lanaguage whose characteristics have been spread through different families over time. What he pointed out was, the only way to find the original tongue is to analyze roots of the essential words such as water or fire. Anyway he promised to email some of his own observations so that should be very interesting.

You may have a point about the mother language coming from the new world, due to the fact that ancient cities are being unearth in south america older than the oldest cities in the middle east. I think we are going to discover more of this in the coming years. We also have to remember that most of the primitive languages were spoken rather than written.

[Edited on 28-5-2004 by marg6043]

Very nice links, I have the feeling that we will never actually find the mother language.

Most of the language we asociated to ancient man is because it has a written for of it.

if no one's mentioned it yet, trying googling for "nostratic," which is one of the names people looking into the possibility of a single global language; you should find plenty to keep you busy.

gvret: seconding what hetman said about the sound/meaning relation being the first to go, the primary problem with the example you're using -- greek vs chinese -- is that you need to be VERY careful when you're assuming the words were pronounced like that...the pronunciation of chinese has changed rather dramatically over the past thousand years, as has the pronunciation of greek.

if there is some universal proto-language there may in fact be a link between greek and chinese, but if there was a link the kind of data you'd want to have showing the link would be something like this:

for some basic concept -- like, say, fire -- you'd start by finding the oldest known way the greeks pronounced it and the oldest known way the chinese pronounced it, let's call them Old Greek Fire and Old Chinese Fire. these probably wouldn't be that similar, but because the way that languages change their pronunciation over time tends to follow pretty regular patterns -- read up on historical linguistics for some example patterns and the evidence for them (i'd recomend something but i only know syntax and semantics stuff) -- you could reason your way from Old Greek Fire and Old Chinese Fire to what we'll call Proto-Old Greek Fire and Proto-Old Chinese Fire, which would correspond to very reasonable guesses as to what the Old Greek Fire and Old Chinese Fire were pronounced like a thousand or two years before.

Note that we're leaving the realm of actual documentation when we do this step, though: we said Old Greek Fire and Old Chinese Fire were the oldest forms we had evidence for, so what Proto-Old Greek Fire and Proto-Old Chinese Fire are is essentially our best guesses, but they could be somewhat or totally wrong for all we know.

Nevertheless, let's say we get to the Proto-Old Greek Fire and Proto-Old Chinese Fire stage and the pronunciation looks more similar. We might then have a hint that these two words are related, and if there's a plausible way for them both to have descended from a common root word then we might have some evidence for a common ancestral language.

For an example of how this would work, say, for example, we figure out that Proto-Old Greek Fire and Proto-Old Chinese Fire are "kit" and "gim", respectively (remember these are both totally made up -- the greek one should look like "pyr-" or something, anyways -- and in our thought experiment are supposed to be just guesses). Then we could guess that something like the following happended:

some ancient language had "ki?" (ie, ki and something we're not sure about) as the fire word, and then in greek somehow that "?" became a "t" (or it could be a "t" and didn't change -- we just don't know) and in chinese the k became a g (which sometimes happens in languages) and the "?" became an "m".

That reasoning gives us an example as to how we'd expect greek and chinese to have diverged from some common root language, and so we'd predict that if we found other candidate words from the root language then they'd diverge similarly, ie if "kal" meant horse in the root language then we'd expect it to become "kal" in greek and "gal" in chinese.

a good researcher would test the "k stays k in greek and goes to g in chinese" theory against all the relevant words known from the root language, and if it was a consistent pattern then, and only then, would there be strong evidence for a common ancestor; otherwise, you've maybe got interesting information, but you can't conclude anything from it with any degree of certainty.

i've skipped a few possible problems that might come up -- what if your oldest example from greek isn't as old as your oldest example from chinese, for example...how do you correct for that? -- and exaggerated the amount of information you'd have about potential words in the root language: a lot of the time, you'd have something like "(vowel?)k(vowel)t" as your candidate word, making it REALLY hard to draw firm conclusions.

This is why looking for grammatical similarities can be more helpful -- grammar changes less over time (though it still does change -- old english had declined nouns but we pretty much got rid of those) -- but we have less understanding of why grammars change with time and how they change, so it's hard to do the kind of guesswork that you can do with sound changes over time.

i should wrap this up, i guess, but i'd say this: it's probably very likely that there are one or a few original languages that most or all of the currently spoken languages derived from, but if we have any chance at all of getting enough evidence to prove or even strongly argue for such a claim, the divergence can't have happened more than 20k - 25k years ago, if that -- it's simply too hard to figure out stuff about (unwritten) languages that old, and very near impossible for things older than that. if the divergence happened more than that long ago, we'll probably never know with any real certainty.

there's a nasty problem with any theory of a single original language, and that's the fact that if you put two or more human children (young, like <= 2-4 years) together, they'll make an entirely new language of their own; do a google for something like "linguistics pidgins creoles" and you'll see what I mean. since that's the case, it's entirely possible for their to be very many "original" languages in the world, and there's less theoretical need to trace everything back to some root language...entirely new languages can pop up in a generation or two in the right circumstances...meaning that even if MOST of the world's languages can be traced to some root language there's probably a lot of languages out there that aren't at all related to the mother language.

sorry about the length...but it's an interesting topic and an interest of mine