Slashdot Mirror
Grammar Traces Language Roots
mlewan writes "Researchers use grammar to trace relations between Papuan languages. What is interesting is not that much that they use grammar features to do this, but that they seem to have given up using vocabulary as a help."
One can always, eventually, make out structure from random noise. At what point do you stop blindly searching for the sake of it?
Furthermore, whilst vocabulary within a language can diversify yet remain part of the same language, vocabulary can easily bleed between diverse languages. Consider that the word "television" might exist in Chinese (I've no idea whether it does or not). That doesn't mean that Chinese was in any way derived from Latin (or Greek, I dunno). Words can migrate with ease. You certainly wouldn't expect Chinese grammar to suddenly mimic ours though.
Argh.
The vocabulary varies within the three major countries that speak it (America, England and Australia)
Eh?
Religion for nerds. Stuff that really matters
It has long been known that Japanese and Korean have had similarities in grammar, but both have been classified as language isolates as a result of not being able to find strong vocabulary links as nice as Indo-European languages. Some consider the two languages to be a part of the larger Altaic language group. Maybe this new method of investigation will turn up more useful results than the vocab link which is increasingly becoming a dead end.
UK: I haven't got a nose. US: I don't have a nose.
UK: Microsoft are delaying Longhorn. US: Microsoft is delaying Longhorn.
Also, grammar certainly does change quite a bit even in the course of a thousand years. E.g., "With this ring I thee wed" is a remnant of when English used Subject-Object-Verb ordering (like German) instead of Subject-Verb-Object, whereas most of the so-called "strong irregular verbs" in English can be traced back to proto Indo-European (~7000 BC). English has also lost almost all of its declinations for case, except for pronouns.
Nevertheless, this new technique does sound like a promising tool for historical linguistics.
"Technically, 'he ain't got it' is just as wrong as 'he doesn't has it', considering that 'ain't' means 'am not'. The only difference between your examples is that one error (he ain't) is widely accepted in every-day speech and one (doesn't has) isn't."
How about something like: "Aren't I supposed to go to school?"
This is perfectly standard English, though "aren't" is a contraction of "are not". So, the sentence, when changed into a statement rather than a question, says: "I are not supposed to go to school".
Strange, at first glance. However, contractions don't have to do precisely what one might say they're supposed to do. The evolution of language need not be absolutely precise.
Still, language does precisely what it's supposed to do--convey meaning. So, meaning is use, and language evolves, regardless of how prescriptivists would have it.
You don't mention whether you evaluated the research itself, or merely the report of the research. I get the sense from our post that you are commenting on the data that was presented by the journalist, and not the data that was presented by the researchers.
You demonstrate that the two features listed in the news article, when applied to English and German, don't demonstrate a relationship that we know is there. It is possible, however, that the remaining 123 features that the article did not list would correctly show the close relationship between Engish and German.
The fact that there is "one, and only one" in the past does not mean that this must always be the case.
A good reason to be skeptical. But I don't think that an academic such as yourself should be utterly dismissive without reading the actual literature and evaluating whether the researchers answered the criticisms of prior research.
I speak German, Swiss-German, Dutch, Afrikaans, French, English and some Spanish and Turkish. One thing that really amazed me about Turkish is that, despite being seperated for over 1000 years, a Turk can still make himself understood throughout central asia from Azerbaijan to Kazakhstan. The languages have changed very little from proto-Turkic. Whats more, once you learn the grammatical system on which Turkish is based, you immediatley notice the exact same or at least very similar features throughout the Ural-Altaic language group, from Finnish, Hungarian, through to Turkish and Mongolian: The way that these languages almost uniformally have no concept of grammatical gender (no word for he or she), the way that these languages universally use the concept of adding prepositions as suffixes onto the end of words instead of being seperate as is generally the case in Indo-European languages, the very large case system also added as suffixes to the ends of words, and the concept of vowel harmony, where, in the beginning of a word which has back vowel such as a,o and u, or front vowels such as ä, ü or ö, will force the rest of the word to also change their vowels to fit in with the pattern.
It is amazing that this structure of these languages has remained so solid such that Hungarian and Finnish, which have no common words, have a very similar grammatical structure after having being seperated for almost 3500 years.
This is absolutely not the case with Indo-European languages where a modern English person can usually not understand their own language from 1200 years ago, much less German or Dutch which were both very closely related to Old English at the time. Granted Old English changed very much with the viking invasions when it mixed with Old Norse and then once again when it mixed with old French after the Norman invasion, such that the structure of a modern English sentence resembles Scandinavian more than it does German, but its vocabulary resembles German/Dutch and French.
In summary, I think that language is a reflection of both society and environment. People will make up new words to fit changing circumstances, and language structure will change when different languages meet. Simply trying to match grammatical patterns will work well on some language groups such as Ural altaic, but not so well on others, such as Indo European where vocabulary patterns are better matched (try matching English's almost complete lack of grammatical cases with Czech's 7 cases). Pattern matching on languages should try to take not only historical environmental situations into account, but also language group mixing, language evolution patterns if possible, and integrate those with vocab and grammatical patterns.
For a really good question, one should ask oneself how on earth old languages evolved in the first place, since they were alomst uniformly far more complex grammatically than those we speak to day.