Listen to part of a lecture in a linguistics class.
All right, so far we have been looking at some of the core areas of linguistics, like syntax, phonology, semantics, and these are things that we can study by looking at one language at a time, how sounds, and words, and sentences work in a given language. But the branch of historical linguistics, involves the comparison of several different languages, or the comparison of different stages of a single language.
Now, if you are comparing different languages, and you notice that they have a lot in common. Maybe they have similar sounds and words that correspond to one another that have the same meaning and that sound similar.
Let's use a real-world example. In the 18th century, scholars who have studied the ancient languages, Sanskrit, Latin and Greek, noticed that these three languages had many similarities. And there might be several reasons why languages such as these had so much in common. Maybe it happened by chance, maybe one language was heavily influenced by borrowed words from the other. Or maybe, maybe the languages developed from the same source language long ago, that is, maybe they are genetically related, that was what happened with Sanskrit, Latin and Greek. These languages had so many similarities that it was concluded that they must have all come from the same source. And talk about important discoveries in linguistics, this was certainly one of them.
The scholars referred to that source language as Proto-Indo-European. Proto-Indo-European is a reconstructed language. Meaning, it is what linguists concluded a parent language of Sanskrit, Latin and Greek would have to be like. And Proto-Indo-European branched out into other languages, which evolved into others. So in the end, many languages spoken all over the world today can trace their ancestry back to one language, Proto-Indo-European, which was spoken several thousand years ago.
Now, one way of representing the evolution of languages, showing the way languages are related to each other, is with the family tree model. Like a family tree that you might use to trace back through generations of ancestors, only it's showing a family of genetically related languages instead of people. A tree model for a language family starts with one language, which we call a mother language, for example, Proto-Indo-European. The mother language, is the line on the top of this diagram, over time, it branches off into new daughter languages, which branch into daughter languages of their own, and languages that have the same source, the same mother, are called sisters. They share a lot of characteristics, and this went on until we are looking at a big upside down tree languages like this. It is incomplete of course, just to give you an idea. So that's the family tree model, basically.
Now, the tree model is a convenient way of representing the development of a language family and of showing how closely related two or more languages are. But it is obviously very simplified. Having a whole language represented by just one branch on a tree doesn't really do justice to all the variations within that language. You know, Spanish that spoken in Spain isn't exactly the same as Spanish that is spoken in Mexico, for example.
Another issue is that languages evolve very gradually, but the tree model makes it look like they evolve over night, like there was a distinct moment in time when a mother language clearly broke off into daughter languages. But it seems to me it probably wasn't quite like that.