|
 |
Biggest open mysteries of Human Language | ||
Share with:
| ||
 Language Learning Forum : General discussion |  | |
| chaotic_thought Diglot Senior Member United States Joined 3544 days ago 129 posts - 274 votes  Speaks: English*, German Studies: Dutch, French | Message 9 of 29 29 June 2015 at 3:53pm | IP Logged | ||||||||||||||
|
Does it make any sense to speak of "complexity" with regards to a spoken language? For example, English is commonly cited as 'complex' compared to other Germanic or Romance languages due to spelling difficulties. And Japanese is commonly cited as 'complex' compared to non-character based languages due to the writing system.
As another example, it is indeed possible to form COMPLEX sentences and SIMPLE sentences in a given language. When writing, you can generally get away with more complex sentences than when speaking, because when speaking your audience can't "re-read" your sentence to figure it out. However, what does it mean, if anything, to say that a LANGUAGE (i.e. the spoken language) is 'complex' in and of itself? Surely all native speakers learn to speak their mother tongues with more or less the same ease as compared with other native speakers. 2 persons have voted this message useful | |||||||||||||||
| ScottScheule Diglot Senior Member United States scheule.blogspot.com Joined 5230 days ago 645 posts - 1176 votes Speaks: English*, Spanish Studies: Latin, Hungarian, Biblical Hebrew, Old English, Russian, Swedish, German, Italian, French | Message 10 of 29 29 June 2015 at 8:35pm | IP Logged | ||||||||||||||
|
Sure it does. Take two languages, otherwise identical, only one has ten irregular verbs and the other has two. The first language is more complex in some fashion than the latter.
1 person has voted this message useful | |||||||||||||||
| robarb Nonaglot Senior Member United States languagenpluson Joined 5061 days ago 361 posts - 921 votes Speaks: Portuguese, English*, German, Italian, Spanish, Dutch, Swedish, Esperanto, French Studies: Mandarin, Danish, Russian, Norwegian, Cantonese, Japanese, Korean, Polish, Greek, Latin, Nepali, Modern Hebrew | Message 11 of 29 30 June 2015 at 3:49am | IP Logged | ||||||||||||||
More or less but not exactly. All native speakers eventually learn their mother tongues successfully, but that has to be true by definition because whatever native speakers do is how we define their language. However, young learners of different languages have measurable differences in how quickly they stop producing errors (i.e. utterances that an adult would not produce). And different groups don't produce exactly the same rate of ungrammatical utterances.
The concept makes sense; a spoken language can be conceived of as a set of behaviors or a system of rules, and either way one may define a measure of "complexity" on it. However, languages are all fiendishly complex and we don't understand how they work in the way we understand algebra or programming languages, so there isn't a standard complexity measure that you can use out of the box.
No one knows for sure but I've always favored a rather pedestrian hypothesis: Suppose clicks are just normal phonemes that languages can have, but the processes of phonetic change only generate them through infrequent conjunctions of favorable phonetic conditions. But once you have a click, you can get more (perhaps clicks that become different through coarticulation split into multiple phonemes). Clicks can also be lost, but this is infrequent as well because they aren't readily confusable with non-click phonemes. The result of all that would be one or a few language families whose members share a varied inventory of clicks, which are not observed elsewhere in the world (except some Bantu languages in southern Africa which have gained clicks through contact).
It is of course absurd that language would have been produced all at once by a genetic mutation. But it could be that humans gradually evolved a "readiness" for language via increasing intelligence, and language was invented all at once by one group of early humans that had evolved the readiness. In my view, the more likely scenario is a set of interdependent changes over time: (1) Humans evolve a readiness to use language through increasing intelligence. (2) Through invention and cultural change, early humans' communication becomes more intricate, leading to proto-language. (3) As processes (1) and (2) continue, genetic evolution also takes place to solidify the processes that develop the brain structures that process the communicative behaviors so that children learn it more reliably--either by the Baldwin effect, or just by standard natural selection. (4) Alongside (1), (2), and (3), linguistic structure changes to match the capabilities of our brains. Any nascent features that are unlearnable by the next generation are automatically weeded out. We have absolutely no way to estimate how long this took to go from ape communication not recognizable as language-like, to fully-fledged human language. I would guess it took between 30 and 300,000 years though.
Something like this is probably needed to explain the fact that after so many thousands of years of language change, they haven't all ended up with the same typology as a kind of stable attractor. Still, there are lots of things that could have introduced a bias for language change in recent times to differ from the distant past: -Intentional standardization -We have writing now -Prestige forms are displacing dialects -Easier travel -Speech communities are much larger Of course, many of the oft-mentioned Indo-European changes took place in tribal societies in which these modern factors were weak or absent. That's probably just an artifact of where early Indo-European languages were in the "cycle:" in other words, the typology of early Indo-European languages was such that they predictably changed to become more analytic, on average, over the centuries. This must necessarily not be the case for languages in general, or else all languages would be analytic by now. The only possibility for something of the sort is that the above characteristics of "advanced" societies might bias languages in that direction. Edited by robarb on 30 June 2015 at 3:58am 2 persons have voted this message useful | |||||||||||||||
| ScottScheule Diglot Senior Member United States scheule.blogspot.com Joined 5230 days ago 645 posts - 1176 votes Speaks: English*, Spanish Studies: Latin, Hungarian, Biblical Hebrew, Old English, Russian, Swedish, German, Italian, French | Message 12 of 29 30 June 2015 at 4:31pm | IP Logged | ||||||||||||||
|
I suspect (don't hold me to it) that writing tends to impede languages from moving in more synthetic directions. Words that would have, otherwise, merged are kept apart by the weight of tradition. I notice when I ask people "What's going on?" I pronounce it [tsgoin on]. Half the time I even say it as a declarative, without the final pitch rise, even though I intend it as a question.
If there were no writing, I imagine children who learned their language from me might very well think my "ts'goin" is a single word, with [ts] being an interrogative prefix or something. Or they might think the entire phrase is a single word: "tsgo-in-'on." But because children learn writing as well, typically, they'll continue to recognize the parts of the phrase as individual words. Contractions are an exception, where words merge in speech AND writing, but they're scarce, comparatively, and even so, they always have that apostrophe there, reminding you that the word can be, if needs be, pulled back apart into pieces and treated analytically. But illiterate proto-Germans had no such reminder, and it seems natural that eventually any knowledge that the parts of the word were originally separate would be lost. (Of course there are lots of languages that contract without apostrophes, so this doesn't apply there). It seems to me the journey from synthetic language to analytic doesn't have the same problem. When the Romans moved from "Romae" (meaning in Rome) to "in Roma" (meaning the same thing), this was easier than moving in the reverse direction, because "Roma" was already a form that existed in the language, as was "in." So no new form had to be created--rather different already existent forms had to be put together in new arrangements. Making new arrangements of existing words is what using language is about, so that's comparatively easy. But if the reverse were the case--say the form "Romae" didn't exist, "in Roma" was standard, and I'm a Greek pedagogue marking a scroll, and one of my students wrote down "Romae" I'd object "There's no such word, cut that out." But of course I'm just speculating. Perhaps a Greek pedagogue would find it just as objectionable if a student started writing "in Roma" instead of "Romae." Regardless, this all seems eminently testable. Do languages with writing move in different directions at different rates, compared to purely verbal languages? Edited by ScottScheule on 30 June 2015 at 4:33pm 2 persons have voted this message useful | |||||||||||||||
| robarb Nonaglot Senior Member United States languagenpluson Joined 5061 days ago 361 posts - 921 votes Speaks: Portuguese, English*, German, Italian, Spanish, Dutch, Swedish, Esperanto, French Studies: Mandarin, Danish, Russian, Norwegian, Cantonese, Japanese, Korean, Polish, Greek, Latin, Nepali, Modern Hebrew | Message 13 of 29 01 July 2015 at 1:23am | IP Logged | ||||||||||||||
I don't know that we can get that much reliable data on changes over time in unwritten languages across multiple families. You could infer changes from reconstructed proto-languages, but that's dicey. Language features might differ in how likely they are to be reconstructible, resulting in a biased sample of what the older unwritten languages had. It's a cool and plausible speculation though! Edited by robarb on 01 July 2015 at 1:23am 3 persons have voted this message useful | |||||||||||||||
 Iversen Super Polyglot Moderator Denmark berejst.dk Joined 6705 days ago 9078 posts - 16473 votes Speaks: Danish*, French, English, German, Italian, Spanish, Portuguese, Dutch, Swedish, Esperanto, Romanian, Catalan Studies: Afrikaans, Greek, Norwegian, Russian, Serbian, Icelandic, Latin, Irish, Lowland Scots, Indonesian, Polish, Croatian Personal Language Map | Message 14 of 29 01 July 2015 at 10:44am | IP Logged | ||||||||||||||
|
A complicated morphology can be regular, and a minimalistic morphology can be full of irregularities. Even the fearsome Latin verbal morphology is reasonably regular - bordering on agglutinative-like - in its main construction principles. For instance the -ba- of the imperfect is common to the active and the passive forms throughout all three persons and two numbers. And apparent irregularities like the "-ui" of the 2. class verbs (monui) versus the -vi- of the 1. class vers (amavi) are still understandable as phonological differences which can be dealt with in a systematic way - even on the fly when you are speaking or listening.
This raises the question of the balance between construction versus atomistic learning of forms. If you know you just can insert -ba- to produce the imperfect of a verb then it isn't really different from using an adverb or an auxiliary verb to achieve the same effect. Actually it may be easier to grasp than a system where some vague idiomatic/stylistic reasons favor the use of one word at the expense of others. But alas, regular systems get eroded, and the result is that all those beautiful symmetries end up as heaps of irregular forms which have to be learnt as isolated items. OK, native speakers will to a large extent skip the construction phase and jump directly to the final complex units, but for language learners who haven't yet learnt all forms of most verbs as individual units a regular system with a lot of forms is still not too difficult to manage - you just have to look beyond the myriads of forms. Another example: the past tense and conditional of Polish. For historical reasons the past tense in the Slavic languages is based on a participle with an l (which in Polish mostly becomes ł, pronounced almost as English w), and this form agrees with the subject in gender and number. But far ago the ancient Polish speakers apparently missed the personal endings so much that they decided to add them after the -l-morpheme (although with some freedom in moving them to other words in a sentence). And in the conditional they followed the same pattern, only this time they also included a -bi- which originally was an inpendent word (actually another old verbal form which lost its independence). But then they added a few intricacies of their own. In the singular the -l-thing has masculine, feminine and neutrum, but in the plural it is personal masculine against all the rest. And the result is a set of fiendishly complicated inflection tables, which furthermore have to be used with caution because of phonologically based changes. Did I forget to mention the prefixes and infixes that transform imperfective verbs into perfective and vice-versa? But stop: instead of counting individual combinations of all these fixes and morphemes and soundchanges and god knows what you should see them as elements that can be combined according to relatively simple rules. If there wasn't some kind of logic behind the kaleidoskopic vagaries then even native children couldn't learn their first language. Somehow they must be able to grasp a system behind the madness. Edited by Iversen on 01 July 2015 at 12:00pm 1 person has voted this message useful | |||||||||||||||
| ScottScheule Diglot Senior Member United States scheule.blogspot.com Joined 5230 days ago 645 posts - 1176 votes Speaks: English*, Spanish Studies: Latin, Hungarian, Biblical Hebrew, Old English, Russian, Swedish, German, Italian, French | Message 15 of 29 01 July 2015 at 5:25pm | IP Logged | ||||||||||||||
Yeah, I was thinking more of looking at languages in the short term. How do unwritten languages seem to moving now? But as you say, the data just might not be there. Maybe in 20 years we can compare languages to our current data. 1 person has voted this message useful | |||||||||||||||
| ScottScheule Diglot Senior Member United States scheule.blogspot.com Joined 5230 days ago 645 posts - 1176 votes Speaks: English*, Spanish Studies: Latin, Hungarian, Biblical Hebrew, Old English, Russian, Swedish, German, Italian, French | Message 16 of 29 01 July 2015 at 5:30pm | IP Logged | ||||||||||||||
I imagine there's an upper bound on the complexity of a language that can reasonably be learned by a community. If there are two many irregularities, the human brain will probably just not be able to take it. It would be like doing advanced math without a calculator. That being said, some languages are still more complex than others, although none are (obviously) too complex to learn. Edited by ScottScheule on 01 July 2015 at 5:32pm 1 person has voted this message useful |
|  | 
Printable version
|
| Copyright 2024 FX Micheloud - All rights reserved No part of this website may be copied by any means without my written authorization. |
