Children tend to look like their parents, and the same is true of languages.  Everyone knows, for example, that French, Italian and Spanish look a lot like Latin.

What is less well appreciated is that you can follow this logic back in time, for a long, long way.  Linguists have known for more than 200 years that there's a lot of similarity between Latin, Greek, German, and surprisingly...Sanskrit.  Languages geographically in-between, such as Farsi, are also related.  You can see it not only in similarity of words, but also grammar.

Words that are essentially the same between languages are called cognate.  The Deus of a Latin prayer is cognate with the Zeus of Greek.  Father in English, Vater in German, pater in Latin, patēr in Greek, pitar in Sanskrit are all the same word.  They're cognate.

Together they form a mega-family of languages that stretch from India to Europe, and therefore are known as the Indo-European family.

You can build an ancestral family tree for Indo-European by looking at how much each language has in common with the others, and making the reasonable assumption that anything two languages have in common must originate from their common ancestor.  Inevitably this must take you back to a single original ancestor, which is called Proto-Indo-European, or PIE for short.

Here's a family tree, which I've linked to from the work of Jack Lynch at Rutgers:

This diagram is the best I've yet scene, because it cuts out the hundred or so extra, highly obscure, and utterly dead languages in the family.  You probably don't care about Early Proto-Tocharian.

People have reconstructed PIE by taking that which is common across the entirely family, and tracing the evolution of sounds and grammar backwards to take an educated guess at the original language.

PIE is thought to have originated somewhere in southern Russia or the Caucasus, probably in about 4,000BC, give or take a millennium.  The most popular theory is that the original speakers were a culture called the Kurgans.  Whoever the original speakers were, they migrated in waves across Europe and down through Asia.

It interests me that almost everywhere it went, PIE and its children dominated.  The natural assumption is conquest did the job--consider for example how Spanish and Portuguese came to be the standards in South America--but even in places where the arrival was peaceful, people mostly adopted the PIE structure.  The biggest failure probably is that language Tocharian I mentioned before.  It was an intrusion into China, but withered in the face of Chinese, the world's oldest known extant language.

There seems to be something about PIE that sits well with the human brain.  It appears to be a very good language for thinking about mathematics and physics.  If you exclude all the progress made by PIE speakers, which means everything achieved by Greek, Germanic, English, Sanskrit, Hindi, Latin, French, Spanish and Italian speakers, then there's not a lot left.  That might perhaps be an issue of cultural domination rather than linguistic advantage, but even so it's remarkable.

PIE was highly inflected, and that might be part of its strength.  You can say an awful lot in a few words.  Linguists originally thought PIE must have been somewhat like Sanskrit, because grammatically it's one of the simpler members of the family, but they now know it's the other way round.  Sanskrit is one of the most advanced members because it's simpler.  PIE was grammatically complex.

The earliest recorded PIE language is in fact the proto-Greek of Linear B, decoded by Michael Ventris.

We can tell a lot about their life from the language.  For example, PIE has a word for horse.  But there's no word for wheel.  (How do we know that?  Because every PIE language has a cognate for the Latin equus, but the word for wheel is different everywhere.)

Several people have had a go at writing something in PIE.  The script must obviously be modern since this is long before writing was invented, but it's fascinating to look at anyway.  The most famous thing written in PIE is Schleicher's Fable.  It's been updated several times since he wrote it in 1868(!), and every time someone updates they go out of their way to make the script more confusing with more silly accents, so here's the original, in PIE and then in English:

Avis akvāsas ka
Avis, jasmin varnā na ā ast, dadarka akvams, tam, vāgham garum vaghantam, tam, bhāram magham, tam, manum āku bharantam. Avis akvabhjams ā vavakat: kard aghnutai mai vidanti manum akvams agantam. Akvāsas ā vavakant: krudhi avai, kard aghnutai vividvant-svas: manus patis varnām avisāms karnauti svabhjam gharmam vastram avibhjams ka varnā na asti. Tat kukruvants avis agram ā bhugat.

The Sheep and the Horses

A sheep that had no wool saw horses, one of them pulling a heavy wagon, one carrying a big load, and one carrying a man quickly. The sheep said to the horses: "My heart pains me, seeing a man driving horses". The horses said: "Listen, sheep, our hearts pain us when we see this: a man, the master, makes the wool of the sheep into a warm garment for himself. And the sheep has no wool". Having heard this, the sheep fled into the plain.


Amalia T. said...

I would love to know how different Linear B Proto-Greek (spoken I mean) and Classical Greek were, and also to modern Greek. I would love to see that evolution laid out.

Gary Corby said...

Linear B to Classical must be hugely different. There's something like a 1,200 year gap. Think of the difference between Chaucer and modern English!

Classical to modern is not so huge. Classical Greek evolved into koine, which is the trading language the Bible was written in, which in turn became modern Greek.

Geoff Carter said...

Good article, such an important area of study.
Men with horses might well spread a language over such large distances.

Amalia T. said...

It's still understandable. And Classical Latin and Medieval Church Latin are understandable to one another--at least as written. That's a pretty large time gap, there.

Gary Corby said...

That's a very good point about Chaucer being understandable. So, maybe they could talk to each other, as long as they spoke slowly?

The oldest known written Linear B is dated to circa 1400BC. That's a 900 year gap to the very beginning of the classical period. Tacking on a few hundred years, on the assumption it took some time before people wrote it, I figure 1200 years. 900+ for sure.

Incredibly, 1400BC is still 2,600 years after PIE is believed to have begun as a language.

Amalia T. said...

I suppose it kind of depends on how late into Linear B the speaker is coming from, too-- if it is someone from the YOUNGEST known Linear B sample, speaking to say, the Oldest known Classical Greek writings, that would change things, assuming that the language evolved steadily over time toward Classical Greek, wouldn't it? The later the speaker, the more likely they would be able to understand.

It's kind of mind blowing to think that PIE was 2600 years old before Linear B appeared on the scene. I have to wonder what caused pieces of PIE to break off into different dialects-- just geographical isolation and migration? It seems odd that there would not have been SOME level of contact maintained, or that each new wave of migration wouldn't have drawn the evolving language back toward its roots.

Maybe I'm thinking about it wrong. After all, France and Spain and Italy are all on top of one another, certainly had contact, and still split apart into three different languages. I guess there is not really any stopping the evolution of language.

I recently read a book where an author sent a modern day English speaker back in time to 800 or 900 AD where the character encountered Saxons and Norsemen, and understood them perfectly. I remember thinking it was pretty outlandish, but that's the same 1200 year gap we're talking about with Linear B and Classical Greek.

Gary Corby said...

Geographic distance, for sure. It only needs a tiny bit. Try putting a Cornishman and a Scotsman side by side.

I'm afraid I don't believe for a moment that a modern day English speaker could understand a Saxon or a Norseman without enormous difficulty. You'd need, at the very least, a strong working knowledge of Middle English grammar. But knowing Old High German would be even better. Also, I assume you know about the Great Vowel Shift that began in the 1300s.

It occurs to me Amalia that a lot of your characters must have been late Linear B speakers.

Amalia T. said...

That was my response too-- you should have heard me going on about it to el husband, who was equally flabbergasted by that stretch of the imagination. It totally blew all my suspension of disbelief.

But yes. My Greek Heroes in Helen's book and Pirithous' are all Late Linear B Speakers -- hence my interest!

Susanna Fraser said...

I could buy a time traveler understanding Saxons and Norsemen or whoever IF it was fantasy time travel--i.e. the character stepped through a stone circle or put on an antique ring or whatever, and BAM! they're in 800 AD. It's magic that took them there, so I'll accept that magic lets them speak and understand the language.

But if it's science fiction time travel--i.e. the story asks me to believe some future scientist has invented a time machine--I want to see the traveler studying the language, not to mention getting their shots for smallpox and the like and making sure they're not carrying any disease strains that might prove lethal to the unexposed past population, before they go.

Gary Corby said...

Yep, that sums it up very well.

Karin said...

How about the Arabic languages? Didn't find them on the map, I think, but they have been rather useful in the development of natural sciences too. Would they fall in the extant category as well, together with Chinese?

Gary Corby said...

Hi Karin, Arabic isn't a member of Proto-Indo-European. It's Semitic. Thanks for the idea; I'll make that the next post...

dipylon said...

Gary, your chart is missing Armenian, which is an important branch. Also, Kurgan was the name of a certain kind of burial, not of a people.

dipylon said...

Another tidbit: the Hittite word for water was "waatar", so if any of you rode the Time Machine and were stranded in Hattusa, you would not go thirsty.

Gary Corby said...

The chart cuts out a whole pile of languages! Apologies to the missing. I picked a summary rather than comprehensive, to show just how widely the family goes. A comprehensive chart of Indo-European would be very, very big.

mk54 said...

I've noticed that everywhere there are clusters of the R1B1 haplotype and Celtic languages there are also modern day sources of tin and copper. Celts were a Bronze Age phenomenon...and Bronze is made of copper and tin, tin is relatively rare. A straightforward possibility is that the Celts were prospectors, miners and metalurgists....well armed and capable of producing wealth whereever they could find the raw materials. Also, probably not a coincidence that the Germanic offshoot of the R1b1 Celts moved to where bog iron could be found...after all it was the Iron Age by the time time happened.