UniLang

What I'm hearing about the Leaving Certificate from Irish students may be biased by their hate for the school system, but I've noticed many of them say that the way to pass the Irish exam is to simply memorise the answers beforehand. So maybe not a good way to actually measure competency in the language. Maybe TEG tests could work better in that respect.

Either way, you would (try to) measure the competency among those who take the test, not the average in the general population, so not sure what the results would tell you anyway.

kevin wrote:Either way, you would (try to) measure the competency among those who take the test, not the average in the general population, so not sure what the results would tell you anyway.

What you'd really need is a comprehensive language survey. It can be done--it has been elsewhere--but it's expensive and difficult and I'm not sure what the real practical use of obtaining this kind of data would be.

dEhiN wrote:

md0 wrote:I find it pretty exciting from a theoretical point of view too. I am now reading some research on Berlin Russian-German bilinguals and their age-of-onset-related effects are more nuanced than the traditional views of either "L1=L2 but distractions" or "L1 always =/=L2 because biology".

Could you please explain what "L1=L2 but distractions" and "L1 always =/= L2 because biology"? What are the distractions being referred to? Is it supposed to mean L1 influence on L2? If so, then why not say "L1=L2 but influence"? And what does biology have to do with the reason L1 always doesn't equal L2? Wouldn't it just be L1 always doesn't equal L2 (i.e., someone always using L1 means they won't be good or versed in L2)?

What I wanted to mention very briefly is that there are two extreme positions in the field of language acquisition: one that argues that competency in an L2 would have been (and can be) equal to competency in one's L1 if this was happening in a vacuum, but real life outcomes (which show that L2 skills are lower than L1) can be explained by environmental factors (e.g. because adults have simply more responsibilities and things that require their attention in their day-to-day lives). That kind of (non-linguistic) distractions. The other extreme position is that there's a clock in the brain that's counting down to the moment access to the L1 learning mechanism is cut-off (critical period hypothesis) and any language learnt after that relies on different, non-language specific mechanisms and/or* compensation mechanism that try to map features of the L2 to similar-enough features of the L1 and they all come together to explain why L2 competence is not comparable to L1.

I don't want to imply that the truth is in the middle though. The evidence is more favourable to the second position, but not all the way.

* a debate within the debate!

linguoboy wrote:[*]The same folks have told me that learners have thanked RTÉ for running programmes narrated by speakers with non-traditional Irish like Manchán Magan (notorious for his stunt documentary series No Béarla) because they find it easier to understand than native traditional Irish. I've had L2-speakers complain to me before about RTÉ running interviews with "an old farmer with three teeth" than no one can understand rather than finding more intelligible speakers.

This reminds me of the following Try video: https://www.youtube.com/watch?v=rJ4SMU6knFA

At about 3:51, these Irish English speakers* listen to "an old farmer with three teeth", which is hilarious! Out of the 3 sets of "tryers", 2 are couples - Marie and Richard, and James and Séan; I find it interesting though to hear their Irish English accents as for me, especially James and Séan sound quite different. Ryan (the guy on the right of the 3rd set) has a thick Irish English accent, which I sometimes find hard to understand. Despite this, it's actually Marie who seems to understand the most of the "old farmer".

*As far as I know, all those on this channel grew up only/mostly speaking English and learned Irish later on in school.

md0 wrote:What I wanted to mention very briefly is that there are two extreme positions in the field of language acquisition: one that argues that competency in an L2 would have been (and can be) equal to competency in one's L1 if this was happening in a vacuum, but real life outcomes (which show that L2 skills are lower than L1) can be explained by environmental factors (e.g. because adults have simply more responsibilities and things that require their attention in their day-to-day lives). That kind of (non-linguistic) distractions. The other extreme position is that there's a clock in the brain that's counting down to the moment access to the L1 learning mechanism is cut-off (critical period hypothesis) and any language learnt after that relies on different, non-language specific mechanisms and/or* compensation mechanism that try to map features of the L2 to similar-enough features of the L1 and they all come together to explain why L2 competence is not comparable to L1.

Thanks for explaining that! Do you know if there have been any studies done on speakers of two L1 languages who, for environmental reasons, basically end up almost exclusively using only one of the L1 languages for a number of years. I wonder if these speakers' second L1 language effectively whittles away to the equivalent of an L2 language? Or, if, because of this 'critical period', during which they presumably learned both L1 languages, even non-use of one of the languages doesn't affect it in any significant way?

"'Top o' the morning' is Irish 'shrimp on the barbie'."

LOL. Nailed it.

dEhiN wrote:Thanks for explaining that! Do you know if there have been any studies done on speakers of two L1 languages who, for environmental reasons, basically end up almost exclusively using only one of the L1 languages for a number of years. I wonder if these speakers' second L1 language effectively whittles away to the equivalent of an L2 language? Or, if, because of this 'critical period', during which they presumably learned both L1 languages, even non-use of one of the languages doesn't affect it in any significant way?

Yes, that's a very common situation in heritage linguistics, concerning mostly "home languages" vs societal language in second-/third-generation households (where early bilingualism is likely, as opposed to first-generation). The phenomenon is called "language attrition" and whether the decline of competence in the heritage language is identifiable as different than the lower competence of an L2 speaker or if the end result is the same is such a cool topic and as all cool topics are, it's under debate

If I wasn't under immense time pressure this week, I would have tried to dig up some studies. Unfortunately, what I worked on in the field of heritage linguistics involves primary school-age children, so we aren't describing attrition yet and I don't have those citations handy. Look for work by Maria (Masha) Polinsky's work though, she's the main figure in heritage linguistics.

I think I can regardless of any critical period, you can lose a language (or rather, show a decline in competence).
Less confidently, I think there's at least some evidence to support that language attrition of an L1 is qualifiedly different than incomplete L2 mastery.

linguoboy wrote:They're aren't, and I'm not sure how there even could be given that it's such an inherently subjective definition. Perhaps you could try to get an idea by looking at Leaving Certificate scores, but that's a weak proxy at most since not everyone tests well, the exams don't cover all aspects of the language, and doing well on a language exam is no indication that you use the language in daily life. At the end of the day, all official Irish statistics on language use are self-reported and thus subject to all manner of cognitive biases.

Does the typical graduate of an Irish-medium program outside the Gaeltacht speak what might be considered "good non-traditional Irish"? If so that might work as a lower estimate.

This is going to be as subjective and variable as any other measure of mutual intelligibility. It's going to depend a lot on which dialects, which speakers, what your exposure to them has been in and out of the school system, etc. So I'm afraid all I have to offer is anecdotes (mostly gleaned from conversations with Irish-speakers in learners' fora and IRL):

Thanks! That was illuminating. I guess it would be better to ask how much exposure these L2 speakers typically get to traditional Irish. For example, will listening to popular Irish-medium TV, radio, and podcast shows expose you to traditional Irish?

Yasna wrote:Does the typical graduate of an Irish-medium program outside the Gaeltacht speak what might be considered "good non-traditional Irish"?

You mean like the number of school-leavers who have attended a Gaelscoil?

Yasna wrote:I guess it would be better to ask how much exposure these L2 speakers typically get to traditional Irish. For example, will listening to popular Irish-medium TV, radio, and podcast shows expose you to traditional Irish?

It's all going to depend on what programmes and podcasts they listen to. There's so much content available in non-traditional Irish these days that it's very easy to not have to ever listen to traditional dialect speech if you'd rather avoid it (much like those dialect-speakers who avoid certain dialects).

linguoboy wrote:You mean like the number of school-leavers who have attended a Gaelscoil?

Yeah, I was trying to express it more inclusively to also include Irish-medium university programs, but looking now there doesn't appear to be any.

It's all going to depend on what programmes and podcasts they listen to. There's so much content available in non-traditional Irish these days that it's very easy to not have to ever listen to traditional dialect speech if you'd rather avoid it (much like those dialect-speakers who avoid certain dialects).

Do traditional Irish and non-traditional Irish essentially exist in separate ecosystems? For example, would it be unusual for a traditional Irish speaker to appear on a podcast done in non-traditional Irish?

Yasna wrote:

It's all going to depend on what programmes and podcasts they listen to. There's so much content available in non-traditional Irish these days that it's very easy to not have to ever listen to traditional dialect speech if you'd rather avoid it (much like those dialect-speakers who avoid certain dialects).

Do traditional Irish and non-traditional Irish essentially exist in separate ecosystems? For example, would it be unusual for a traditional Irish speaker to appear on a podcast done in non-traditional Irish?

Hopefully kevin or someone can answer this. I've heard little or nothing in terms of crossover, but maybe I'm just not listening to the right stuff.

I don't really listen much to podcasts or other media in Irish, and if I do take the time, I'll probably only bother if I know they will have good Irish, so my impressions are probably biased, too.

It's probably fair to say that regarding radio stations there is a relatively sharp divide between Raidió na Gaeltachta for traditional Irish and other stations for so called Urban Irish, the non-traditional version of the language. There was that one study that compared the speech of moderators of both types of stations and found out that Urban Irish moderators, even though they are fluent in their version of the language, use significantly less complex sentences on average and make lots of grammatical errors (like leaving out initial mutations in I think it was something like between a third and a half of the cases).

I feel TV is a bit different because you find both forms of the language on TG4, though still maybe less so in a single programme.

On the other hand, there must be some kind of influence between both forms of the language (maybe through school). One of the most obvious examples even for someone who doesn't speak the language might be the non-traditional English [ɹ] which just used to be part of a non-native accent, but is now a feature of the native Irish of younger speakers.

And finally a word on Gaelscoileanna: Some use the word "Gaelscoilis", as if it were a separate language, to refer to the stereotypical bad non-traditional Irish. Of course, that's not the whole reality. I have (online) met Gaelscoil students with great Irish and others who have basically "English with Irish words". It depends a lot on the school, the teachers and of course also the student. But I think it's safe to say that just knowing that you attended a Gaelscoil doesn't give me confidence that you have good, even if non-traditional Irish.

I'll finally leave a video here for you:

https://www.youtube.com/watch?v=3kw-of3UBgg

kevin wrote:I'll finally leave a video here for you:

https://www.youtube.com/watch?v=3kw-of3UBgg

Very interesting and illuminating video! I question though his statement from about 6:29 to 6:41, where he says something to the effect that this "Gaelscoilis" isn't being used by students outside the immersion schools because the hybrid language doesn't meet the needs of real-world communication. I'm sure there's an aspect of that - especially if, as I understand him, this hybrid language is essentially Irish vocabulary strung together using English grammar.

But, I would think that the biggest reason students of the immersion schools don't speak Irish (non-traditional or otherwise) outside of school is due to lack of impetus. If the students are native English speakers who grew up in English speaking households and who basically exist in an English environment, apart from the immersion schools, there's no impetus for them to practice any form of Irish outside of school.

It also seems to me that the hybrid language spoken in the schools probably arose out of the externally imposed structure of the schools: that the students have to speak Irish. This creates a forced environment which explains why the students' brains would focus on learning Irish from a semantic level, but not worry so much about grammar rules and the like. They already have a grammar framework from within which to insert this new Irish vocabulary - English. If, in the process, they pick up some Irish grammar rules and remember it enough to use it while speaking, so be it. But if they don't remember at all, or remember wrongly, it doesn't matter because the recipients (i.e., other students) will understand them anyway.

dEhiN wrote:It also seems to me that the hybrid language spoken in the schools probably arose out of the externally imposed structure of the schools: that the students have to speak Irish. This creates a forced environment which explains why the students' brains would focus on learning Irish from a semantic level, but not worry so much about grammar rules and the like. They already have a grammar framework from within which to insert this new Irish vocabulary - English. If, in the process, they pick up some Irish grammar rules and remember it enough to use it while speaking, so be it. But if they don't remember at all, or remember wrongly, it doesn't matter because the recipients (i.e., other students) will understand them anyway.

It reminds me of the lightly-pidginised German I heard while living in a dormitory in Germany that was mostly for Goethe Institute students. Although they were quite internationally diverse, most had some knowledge of English so they would frequently fall back on it (although in general they were motivated to learn a standard form of German, living as they were in a largely German-speaking environment). But in the creativity shown, maybe it's more like Camfranglais or one of the other mixed languages that arises in countries where colonial languages are still overwhelmingly used in education, at least at the higher levels.

----

So, I think it may have finally clicked what I find so odd about Indonesian.

For years I've been carrying around this assumed set of linguistic universals, a couple of which can be roughly summarised as:

1. Function words are shorter than content words and often show phonetic reduction.
2. The more frequent particular content words are, the shorter they tend to be.

I'm sure these seem very Anglocentric or at least Eurocentric, but they've been tested against a range of languages, starting with Korean, which I learned in college. Korean is an agglutinative language and, if not quite reaching the level of elabouration famously found in Turkish, conjugated verbs can still get quite lengthy. However, the basic roots are quite short--generally a single syllable for core native verbs. "Come" and "go", for instance, are 오- /o-/ and 가- /ka-/. The copula is 이- /i-/ (and maybe even drop out entirely, leaving behind only verbal inflections seemingly attached directly to the predicate noun). Case affixes are generally monosyllabic too and disproportionately feature the neutral vowel 으 (/u/, generally [ɯ]), also used to break up clusters in borrowed words. [Compare Turkish, where "come" is gel-, "go" is gid-, the copula is also i- (and liable to drop), and the case markers are of the form -V, -CV, -CVC, and -VCV.]

None of this, however, is true of Indonesian. "Come", "do", and the copula are, respectively, datang, pergi, and ada (often appearing in the trisyllabic form adalah). None of the most basic verbs are monosyllabic and many contain one or two heavy syllables. For instance, those with the syllable structure CVCVCor CVCCVC include: duduk sit, hidup live, makan eat, masuk entre, pindah move, terbang fly, tidur sleep, turun go down, etc. Case clitics don't exist as such, given the analytic nature of the grammar, but a similar role is filled by prepositions and (apart from the most basic of all, locative di and allative ke) they're hardly less complex, e.g. dengan with, untuk for, oleh by, seperti like.

The result is very disorienting. I'm used to looking at sentences in language I hardly even know and being able to intuit which are the function words or lighter verbs and which are the content words. But this doesn't work at all with Indonesian. I look at a basic sentence like:

Mereka tidak datang karena kematian kakeknya.

and everything seems to have roughly equal weight. Who would guess, for instance, that mereka here is a personal pronoun, tidak is a negator, and karena is a preposition? Compare a version of the same sentence in Welsh:

Ddaethon nhw ddim o achos marw eu tad-cu.

Despite the lack of obvious cognates, I wager y'all would have a better chance of recognising these in nhw, ddim, and o achos, respectively.

kevin wrote:I don't really listen much to podcasts or other media in Irish, and if I do take the time, I'll probably only bother if I know they will have good Irish, so my impressions are probably biased, too.
[...]
And finally a word on Gaelscoileanna: Some use the word "Gaelscoilis", as if it were a separate language, to refer to the stereotypical bad non-traditional Irish. Of course, that's not the whole reality. I have (online) met Gaelscoil students with great Irish and others who have basically "English with Irish words". It depends a lot on the school, the teachers and of course also the student. But I think it's safe to say that just knowing that you attended a Gaelscoil doesn't give me confidence that you have good, even if non-traditional Irish.

Fascinating, thanks!

linguoboy wrote:So, I think it may have finally clicked what I find so odd about Indonesian.

For years I've been carrying around this assumed set of linguistic universals, a couple of which can be roughly summarised as:

1. Function words are shorter than content words and often show phonetic reduction.
2. The more frequent particular content words are, the shorter they tend to be.

I never thought of this, but I think implicit assumption wise, the first one is also a lens I use when trying to analyze sentences in languages I don't know. I'm not sure I've ever really considered the second one as true, though I could see how this would make sense.

linguoboy wrote:The result is very disorienting. I'm used to looking at sentences in language I hardly even know and being able to intuit which are the function words or lighter verbs and which are the content words. But this doesn't work at all with Indonesian. I look at a basic sentence like:

Mereka tidak datang karena kematian kakeknya.

and everything seems to have roughly equal weight. Who would guess, for instance, that mereka here is a personal pronoun, tidak is a negator, and karena is a preposition? Compare a version of the same sentence in Welsh:

Ddaethon nhw ddim o achos marw eu tad-cu.

Despite the lack of obvious cognates, I wager y'all would have a better chance of recognising these in nhw, ddim, and o achos, respectively.

To be honest, I wouldn't have been able to guess for Welsh! (Well, I probably would've ventured that nhw is a pronoun, but not the other two). However, I would've figured that ddaethon was most likely a content word.

linguoboy wrote:For years I've been carrying around this assumed set of linguistic universals, a couple of which can be roughly summarised as:

1. Function words are shorter than content words and often show phonetic reduction.
2. The more frequent particular content words are, the shorter they tend to be.

I'm sure these seem very Anglocentric or at least Eurocentric

Or maybe (roughly) Indo-Eurocentric.

There's this (Estonian):
Oleme kõikide nende aastate jooksul teinud temaga tihedat koostööd.
We have worked closely with him for all these years.
Any clues there? (I'm genuinely curious, since once you know what each word means it's harder to see it from the perspective of someone who doesn't. Although, Linguoboy, you've studied some Finnish so that might be "cheating". There are cognates.)

Of course my sentence is a bit contrived to avoid very short words; it's a natural sentence, but I chose it over your original sentence for that reason. In the sentence that you used as an example, there would be the negator ei which does follow your "universals".

Tema vanaisa surma tõttu ei tulnud nemad.

So there's also:

Vend ei saanud tulla õe töö tõttu.
The brother wasn't able to come because of the sister's work.

Both of these have ei as a negator. On the other hand, in Estonian other sentences have less "recognizable" negators such as mitte, polnud or polekski. And although pronouns can be short like ma, they can also appear in longer forms like minaga, which is just as long as Indonesian mereka (but conveys a bit more information since it means "with me").

Linguaphile wrote:There's this (Estonian):
Oleme kõikide nende aastate jooksul teinud temaga tihedat koostööd.
We have worked closely with him for all these years.
Any clues there? (I'm genuinely curious, since once you know what each word means it's harder to see it from the perspective of someone who doesn't. Although, Linguoboy, you've studied some Finnish so that might be "cheating". There are cognates.)

Yeah, "oleme" immediately leaps out at me as cognate to Finnish olemme. Besides that, I think nende and temaga both have the look of function words, being less phonetically complex than the words around them.

Linguaphile wrote:On the other hand, in Estonian other sentences have less "recognizable" negators such as mitte, polnud or polekski. And although pronouns can be short like ma, they can also appear in longer forms like minaga, which is just as long as Indonesian mereka (but conveys a bit more information since it means "with me").

Yeah, my experience with agglutinative languages like Estonian or Turkish is that although certain function words (like pronouns) might be longer due to inflexion, you soon start to recognise patterns in which suffixes get repeated. In the Uralic languages in particular, agreement aids this quite a bit. When you come across a string like "...jotka olen jalossa punaisessa miehessä huomannut...", it's not hard to figure out that something more than mere coincidence is at work.

So a brief followup: I'm beginning to realise that one of the reasons Indonesian seems so exceptional to me is that I've been learning a rather formal variety of Indonesian. In informal registers, contractions and reductions are more common and the appearance of sentences tends to fall more in line with what I'm accustomed to. For instance, a sentence like:

Saya tidak mau menjadi bintang. "I do not want to become a star."

might appear in a more informal register as:

Ku nggak mau jadi bintang. "I don't wanna be a star."

Although content words are also affected (e.g. the loss of the men- prefix on menjadi "become"), the changes especially seem to affect auxiliaries (e.g. sudah > dah, tidak > (ng)gak), prepositions (e.g. daripada > dari), and other function words. There are also familiar phonetic reductions, like /a/ > /ə/ in unstressed syllables due to Javanese influence.

I realized I haven't been reading enough news articles in Chinese. It's a strange issue for me to have since in the past I overused news articles with languages like German and Japanese. The problem is that mainland news turns me off due to the censorship and propaganda, and Taiwanese news turns me off because of the script (traditional), variety (Taiwanese Mandarin), and the relative unimportance of Taiwan (outside of the political status issue). I've decided I'm just going to bite the bullet and read more from both sources.

Yasna wrote:I realized I haven't been reading enough news articles in Chinese. It's a strange issue for me to have since in the past I overused news articles with languages like German and Japanese. The problem is that mainland news turns me off due to the censorship and propaganda, and Taiwanese news turns me off because of the script (traditional), variety (Taiwanese Mandarin), and the relative unimportance of Taiwan (outside of the political status issue). I've decided I'm just going to bite the bullet and read more from both sources.

Could you find news sites that aren't specifically Chinese but have Chinese versions of the articles? I can't find the site now, but I've come across European news sites that will display the same article in multiple (European) languages. Perhaps something similar exists for Chinese?