[Basic] Learning languages

[johndallman]

This is interesting: Wikipedia's article on the "Common European Framework of Reference for Languages" an EU classification scheme for language proficiency levels. The time expected for learning to various levels is congruent with GURPS rules, so much so that I wonder if this stuff was an input into the 4e language rules.

[Critical]

That, and the GURPS rules, are a little dubious just because a speaker of Language A will not always learn Language B & C at the same rate. It is harder for a native English speaker, for instance, to learn Mongolian than to learn Spanish.

But, yeah, the GURPS system gets the job done well enough.

[Not]

Are the difficulties reciprocal? Can a speaker of language A learn quickly language B, while a speaker of B learns A with difficulty?

[Polydamas]

I don't think so: the classic example is Spanish and Portugese. In general a speaker of an innovative language may have the core features of a less innovative relative (as old words, fixed expressions, etc.), while the speaker of a less innovative relative may not have access to the new features of the other language.

[trans]

Originally Posted by johndallman

This is interesting: Wikipedia's article on the "Common European Framework of Reference for Languages" an EU classification scheme for language proficiency levels. The time expected for learning to various levels is congruent with GURPS rules, so much so that I wonder if this stuff was an input into the 4e language rules.

Language teachers in Germany use this scheme all the time. A couple of months ago my wife asked me to pick up some materials that were at the B1 level, and it was clear what she was talking about. So to that extent it does work.

Originally Posted by Not

Are the difficulties reciprocal? Can a speaker of language A learn quickly language B, while a speaker of B learns A with difficulty?

I've got quite a lot I can say about this, so this will get long-winded.

The expectation would be that it would be easier for the speaker of a standard language to learn a patois or creole, but I'm not sure that's really true.

Learning individual aspects of languages, as opposed to all-around command of the language, is in fact easier in one direction than the other.

Take an English speaker learning Russian - "oh, that must be really hard, with that weird alphabet". But most people could learn Cyrillic just fine in a week, together with a few other things along the way. All the same, Russian speakers learning the Latin alphabet have it marginally easier, if only because they have been more or less exposed to Latin letters already.

But when it comes to spelling or reading aloud, English speakers have it much easier, because Russian is almost always what-you-see-is-what-you-get. The exceptions are few and easy to remember.

Now we come to basic grammar, understanding and being understood. Here English speakers have a harder time, because in Russian you have to know the noun cases, their endings, and other grammatical forms -- otherwise you won't know who is doing what to who (not whom), or whether man bites dog or dog bites man. In English you can throw words out there, and mostly people will understand you. The meaning is often determined by the word order, but that usually isn't much of a problem.

When it comes to writing poetry, however, Russian's flexible word order and diverse word endings make rhyming a snap. Meter is relatively easy, too, and can be very expressive. Even those few exceptions to phonemic orthography mentioned above can be used to good effect - Pushkin did some amazing stuff with that.

Well, Pushkin wasn't learning Russian as a second language, but I think an English-speaking third-year student of Russian (non-immersion, about eight hours a week) could write more convincing poetry than the other way around.

Now German. Again, spelling aside I think it is marginally easier for German speakers to speak, read and write simple English than the other way around, largely due to prior exposure. Even at a more sophisticated level English is probably easier - traditional, highfalutin' German often delights in compound sentences with multiple parenthetical sub-sentences, with the main verb -- when it is not forgotten completely -- at the end coming.

See Mark Twain's The Awful German Language.

But paradoxically, when it comes to legalese, German is much easier. Yes, the infamous Amtsdeutsch is intimidating, but it really is an extension of "normal" German, rather than something from another planet like English legalese. In fact, the normal citizen is expected to understand legalese (though not necessarily reproduce it). The original texts of the relevant laws and regulations are part of the training for most jobs.

Contrast that with English-speaking countries, where legalese is the domain of specialists.

Btw, it seems obvious in GURPS to treat languages like Area Knowledge in that characters will know more about "stuff" connected with their skills: characters with Teaching skill will tend to know educational jargon, lawyers will know legal jargon, etc.

[trans]

That example of the verb in German at the end coming wasn't even the main verb. Here it is again:

Traditional, highfalutin' German often in compound sentences with multiple parenthetical sub-sentences, with the main verb -- when it is not forgotten completely -- at the end coming delights.

[vicky_molokh]

Originally Posted by trans

Learning individual aspects of languages, as opposed to all-around command of the language, is in fact easier in one direction than the other.

Take an English speaker learning Russian - "oh, that must be really hard, with that weird alphabet". But most people could learn Cyrillic just fine in a week, together with a few other things along the way. All the same, Russian speakers learning the Latin alphabet have it marginally easier, if only because they have been more or less exposed to Latin letters already.

But when it comes to spelling or reading aloud, English speakers have it much easier, because Russian is almost always what-you-see-is-what-you-get. The exceptions are few and easy to remember.

I have to totally disagree about what-you-see-is-what-you-get. It's an illusion that one easily falls into; Russian just likes to sneak up on you and shower you with surprises. Silent consonants in the middle of the word, ros/roz/ras/raz choices and swapping, arbitrary doubling and non-doubling of consonants ('agressive', anyone?), e/i and a/o ambiguities with Over9000 rules and 100500 exceptions, the list goes on. Oh, you can make yourself understood after a couple of weeks of study, but reading out loud will sound ridiculous even if you learn the phonetic inventory perfectly, because of all those mismatches.

I'd say that reading even French and German out loud is easier (fewer mismatches); yes, I know writing in either is difficult. And then there are languages with a near-1:1 mapping between spoken and written form - Japanese Kana (only), Ukrainian, Spanish, AFAIK Icelandic, Esperanto, Tatar (Kazan) etc. (I have no idea how closely Belarusian matches its sounds and letters - it seems to be halfway between Russian and Ukrainian it terms of writing complexity, but I can't speak the language at all.)

Originally Posted by trans

Now we come to basic grammar, understanding and being understood. Here English speakers have a harder time, because in Russian you have to know the noun cases, their endings, and other grammatical forms -- otherwise you won't know who is doing what to who (not whom), or whether man bites dog or dog bites man. In English you can throw words out there, and mostly people will understand you. The meaning is often determined by the word order, but that usually isn't much of a problem.

The problem with Russian, Ukrainian, Belorusian etc. cases isn't even their sheer number (although one could claim that there are 9 instead of the officially recognised 6 if one looks deep enough). It's how varied and arbitrary the case endings (suffices?) are, and the occasional root-changing that goes on in them. Conversely, Tatar (Kazan) has six cases too, but they're made just gluing one out of four possible suffices to the word, and you can figure out which one based on the last letter of the word.
Likewise, Japanese seems to have numerous but reasonably simple case-like formation with its -ni/-to/-de etc. endings (I'm not sure it's proper to call them cases).

BTW, the very fact that different languages have different case and pre-/postnoun agreement is an interesting complication. So one speaks _ English or French, speaks Russian-like, speaks using Ukrainian [language], speaks on Russo-Ukrish [pidgin] etc.

Originally Posted by trans

When it comes to writing poetry, however, Russian's flexible word order and diverse word endings make rhyming a snap. Meter is relatively easy, too, and can be very expressive. Even those few exceptions to phonemic orthography mentioned above can be used to good effect - Pushkin did some amazing stuff with that.

Speaking of poetry, I was surprised to find out that English poetry relies on accented-nonaccented alternation of syllables. I never had that pointed out to me in school.

Originally Posted by trans

Now German. Again, spelling aside I think it is marginally easier for German speakers to speak, read and write simple English than the other way around, largely due to prior exposure. Even at a more sophisticated level English is probably easier - traditional, highfalutin' German often delights in compound sentences with multiple parenthetical sub-sentences, with the main verb -- when it is not forgotten completely -- at the end coming.

Lately I'm finding that verb-at-the-end is actually more common than I imagined earlier. Japanese, Tatar, German, surely there are many more. Originally I thought it's an eastern thing.

[malloyd]

Originally Posted by trans

The expectation would be that it would be easier for the speaker of a standard language to learn a patois or creole, but I'm not sure that's really true.

It's probably not, creoles really are different languages than their ancestors.

Some languages are easier to learn that other is one of those things that comes up all the time in anecdotes but persistently fails to show up in studies that measure anything beyond the most basic levels - that it may show up at those most basic level is probably because you can manage some communication with no grammar at all, and it probably is easier to memorize and understandably pronounce a few hundred vocabulary words when they use the same sounds as or are actual cognates of words you already know.

I like to think of this as all human languages are of about the same total complexity, but you can move some of the complexity between various subsystems - e.g. if you leave out cases, that makes that bit of grammar that much easier to learn, but adds a burden to learning prepositions or fixed word orders or whatever the language actually uses instead to clarify them, if your language only uses a few sounds, then it'll need longer and harder to tell apart words that one that has more possible syllables to work with to convey the same information and so on.

[Flyndaran]

I would qualify that as most languages are of the same complexity. Some really are quite simple. There's one that lacks words for numbers, colors, or ability to nest clauses within sentences.

[vicky_molokh]

Originally Posted by malloyd

I like to think of this as all human languages are of about the same total complexity, but you can move some of the complexity between various subsystems - e.g. if you leave out cases, that makes that bit of grammar that much easier to learn, but adds a burden to learning prepositions or fixed word orders or whatever the language actually uses instead to clarify them, if your language only uses a few sounds, then it'll need longer and harder to tell apart words that one that has more possible syllables to work with to convey the same information and so on.

But there's more than one way to organise a subsystem. For example, take cases and plurals. English has simple-ish plurals and simple-ish cases. German, Ukrainian and Russian have moderately complex plurals on their own, but they get ridiculously complicated once you combine plurals with cases (I might be misremembering about German, though) - the problem is that the ending indicating singular/plural and the one indicating case intertwine in highly arbitrary and illogical ways (depending on sex of the noun, its group and some incalculable amount of exception). And finally let's look at Tatar, where all plurals are -lær/-lar/-nær/-nar (and you know precisely which to use based on the last letters in the root), and cases are likewise all done by 4 endings each; and you just lump the plural-ending (or lack thereof) and the case-ending, and there you go.

Basically, word-flexing case systems are a dead end as far as lingtech goes - they lack any redeeming features in the form in which they exist.