20 April 2012

Simplex and complex case-marking

Never heard that terminology before?  I'm not surprised -- I made it up.  Why?  Because there's a phenomenon in grammatical case marking that seems completely obvious and observable to me, but to date I've never seen any literature describing it.

Unfortunately, because it is not a widely recognised phenomenon, it is often disregarded as "an error".

What is this phenomenon?  It's that in some languages, case marking only occurs when there is a single "head noun" in the noun phrase -- introduce "and" or "or" and the marking is lost.

The classic example (and this is where most of you will cry "that's wrong") is in English pronouns.  "Me and him went to the cinema." "Him and me don't talk anymore."  Yes, there are some people who will say "He and I" for both those cases, but then a lot of them tend to hypercorrect and say things like "between you and I", which puts the nominative form where the oblique form would be expected in the classical model.  I'm sure there are some people who consistently mark cases in their pronouns within complex noun phrases; but it is observable fact that a great many native speakers of English do not.

And so I propose making a distinction in case marked languages that only mark case in simple, one head-noun noun phrases, which I call "simplex case marking languages", and those that maintain case marking in longer noun phrases, which I call "complex case marking languages".

Enough about English for the moment, let's look at the Romance language family.  Most Romance languages retain complex case marking in pronouns, even though most have lost noun marking.

So Spanish, for example, will use "yo y tú" as subject if strictly required, and "a mí y a ti" as object. This gives us things like yo y tú no tenemos nada en común -- literally "I and you have nothing in common".  But French doesn't do this.  French is an example of a simplex case marking language.  And yet it tries its hardest not to be.  The closest equivalent of the Spanish sentence above would be toi et moi, nous n'avons rien en commun (I think) -- literally "you and me, we have nothing in common".  So the complex subject is relegated to a tag prefixing the main clause, and a simplex pronoun that is properly case marked.  As for as I know, though, you'll never find a case marked complex noun phrase in French.

And this even carries through in the absence of a verb.  eg
¿Quién viene mañana? - Who's coming tomorrow?
Yo.
Whereas the French answer would either be:
Moi, je viens
or more likely simply
Moi.

And in English, we have a similar choice: I am (with the verb) or just me.

Even among those who still support "you and I" style English, there are very few (I hope!) who would try to enforce a distinction between not I and not me.  Modern English really always comes down to not me.  French sticks with moi (me), whereas Spanish will use yo (I) or a mí (me) according to case.
¿Quién lo tiene? - Qui l'a? - Who has it?
yo no - Moi, non. - Not me.

This leaves me wondering: is there an intrinsic link between simplex case marking and the lack of case marking in verbless utterances?  Or do the similarities in pattern between French and English simply further contact effects caused by the English and French history of mutual invasion?

Not quite the same... or is it?
A similar phenonomenon can be seen with Scottish Gaelic. The Celtic languages have an interesting feature -- prepositions have merged with the following pronoun and altered over time, so eadar (between) + sinn (us) = eadarainn.  But this only holds for a simplex nounphrase.  I want to say "between us and them", it wouldn't be "*eadarainn is iad", but rather "eadar sinne is iadsan", ie. the prepositional pronouns can only replace a simplex prepositional phrase.

But Gaelic's in a funny situation compared to French and English.  Fused forms aside, the pronouns appear to have less case marking than common nouns.  Gaelic still has a three way distinction of sorts: a "default" (for lack of a better word), a genitive and a dative.  But many people will argue that this is actually archaic and doesn't reflect real usage -- case marking in Gaelic is a massive bone of contention, and it's something that won't be sorted out until a proper corpus is assembled.

But I think one of the stumbling blocks in the current academic discussion is the failure to take into account simplex and complex case marking.  The handling of prepositional pronouns does suggest that the Gaelic speaker's internal model deals with complex and simplex noun phrases separately.  But more than this, William Lamb's PhD corpus, though small, contained enough for him to say fairly confidently that the dative tends only to be marked when the preposition rules a single noun -- no ands or ors, and no genitival components.

But that would leave Gaelic in a funny state.  A genitive can only, by its nature, appear in a complex noun-phrase, so the division of simplex vs complex is just a little to simple to hold the key here....

14 April 2012

Ooooh... it looks like I was wrong....


For years, I've been toeing the line that non-native languages never become like native languages, and that there's two language processing mechanisms in the brain: one for the infant learner, and one for the adult-learner.  This was science's best guess based on the data they had to hand: victims of "selective aphasia".  These people had suffered brain damage that had affected their abilities in languages, but not equally across languages.  The observed pattern was that all native languages would be affected more or less equally, and all non-native languages would be affected more or less equally, but the two would often be differently affected.
Well, via an article in the New York Times, I came across this paper at PLoS ONE that says otherwise.  Apparently they've been able to track the brainwaves of learners and find that there are similarities between the brainwaves of a proficient learner and patterns typical of the native.  This I find really cool.  (I wish I could afford a brainscanner so that I could measure my proficiency in terms of native-like brainwaves!!)

I'm concerned, though, that one thing they say will be overinterpreted:
Interestingly, both before and after the delay the implicitly trained group showed more native-like processing than the explicitly trained group, indicating that type of training also affects the attainment of native-like processing in the brain.
There are many people who advocate immersion from day one, and this would appear to be proof of the efficacy of the approach.  However, we are dealing with a very simple language here -- 13 words and a handful of extra inflectional morphemes, leading to a sum total of 1404 possible sentences in the language.  Even if the grammar's very different from English, there's still a very small number of "decision points" to be considered, so it is much easier to devine the meaning from the context -- this is something that just doesn't match the experience with real language.

They do, of course, admit as much in the body of the paper:
it may be that the results reported here are due to the limited size of the artificial language
Furthermore, the study says:
In the explicit training condition, participants were provided with 13.5 minutes of input of a type similar to that found in traditional grammar-focused classroom settings. Auditorily-presented metalinguistic explanations structured around word categories (e.g., nouns, verbs) were presented along with meaningful Brocanto2 phrases and sentences (which were also auditorily-presented, together with visually-presented corresponding game board configurations).
It's a shame the paper doesn't include the script.  My concern here is with the mention of "metalinguistic explanations".  Explicit instruction does not necessarily mean a lot of "metalinguistic" explanations -- yes, it will always require some, but often what starts out as an explanation overcomplicates things with unnecessary jargon and overly conscious processing.  Finely tuned explicit input can be very clear to the learner even if it looks messy on paper, and poorly tuned explicit input can be nigh-on impenetrable to the learner even if it looks neat on paper.

Which is why I object when they claim that:
A second possibility is that at end of training the explicit group's dependence on explicit, declarative memory-based knowledge resulted in the inhibition of the learning or use of procedural knowledge (see above), thus precluding anterior negativities. On this view, explicit training actually prevents, or at least slows, the development of native-like processing.
It is impossible for them to state that explicit training in particular prevents anything -- but that the training that they provided did.  (Of course, if I was to see their scripts, I'd definitely find fault with them, but that could always be me rationalising away anything that disagrees with me -- classic confirmation bias.  We're all guilty of it at times.)

Anyway, despite their statements about the difference between the explicit and implicit instruction, the  study does state towards the end:
the implication of declarative and procedural memory in this study is consistent with the predictions made by the declarative/procedural model for second language acquisition. This neurocognitive model posits that during L2 learning grammar initially depends largely on declarative memory, but that gradually aspects of grammar are increasingly learned and processed in procedural memory.
I am heartened to see that they haven't out-and-out supported the hardline distinction between "learning" and "acquisition" that Krashen and his ilk propose.

So it looks like I may have been wrong about the neurological differences between native and adult-learned languages, and I might still be proven wrong about the benefits of explicit vs implicit instruction, but this is not the study that proves that one.  If and when one comes along, I'll rethink my position.  Until then, I'll stick with what I currently believe.

10 April 2012

Decoo's wheel turns again....

I'm always referring to it, but everyone with an interest in language should read Wilfried Decoo's lecture "On The Mortality Of Language Learning Methods".  By his reckoning, the latest "new" popular movement in teaching is long overdue.  (The fact that the communicative approach is still king is probably due to its prevalence in TEFL, and the fact that anything else takes longer to train teachers to do.)
And one candidate is starting to emerge.  There are whispers of "social media" echoing round the classroom.
Not that long ago, the watchword was "web", but what became of that?  Certainly, the web gives us more access to more materials, but what we've seen to date is either simply the existing textbook styles rewrapped as webpages with automatic marking of exercises or a rapackaging of 90s style multimedia packages.
Digitised textbooks have their advantages over the paper textbook (no fiddling with the CD player to find the right track, instant feedback on many errors) at the expense of being something you couldn't just sit down and do on the train.
The multimedia websites, on the other hand, offer very little over the old CD-ROMs, but are a lot slower to use due to constant downloading of materials in between questions.  And I can shove a CD-ROM in my laptop on the train, but I won't be able to get a connection to the internet without running up a fairly hefty 3G bill.

But the "social media" thing is starting to kick in.  LiveMocha is a clone of the overpriced toy courses by Rosetta Stone, but they had the idea to get learners correcting each other's work.  This is not an innovation in any real sense, it's just a poor man's version of paying a tutor.  Of course, Rosetta Stone decided to jump on the social bandwagon too, and started setting up a language buddy/exchange site.  Reports have it that the demographics are heavily skewed and it's difficult to find language buddies for many language combinations.
Steve Kaufmann's Lingq has taken the social approach down a different track.  The website provides a shiny interface that doesn't really do much and gets people to insert screeds and screeds of texts with audio in various languages, and the user is expected to learn by induction from massive exposure.  I don't buy the argument.  The interface doesn't do much to guide me and the quality of the material varies wildly, with some people even just uploading lists of words, contrary to the site's main philosophy.  But this is starting to get away from the notion of "social", and more into the realms of "crowdsourcing".   It's only the credits-based correction system that retains any sort of "social" status, but even that's starting to stretch definitions a bit.

Tatoeba is unashamed crowdsource.  It's a very intriguing idea: a bunch of sentences undergoing unstructured translation from one language to another, and another, and another.  In mathematical terms, they describe it as a "graph" rather than a "table", a concept which probably merits a blog post to itself.  I love it, and I find it a great exercise to translate stuff from various languages into Scottish Gaelic, as it gets me thinking about things I might not otherwise, and helps me identify holes in my abilities.  That said, it's difficult to see what you can actually do with all that data.  It has no concept of grammatical syntagms or paradigms, so you can't search for a structure and see examples of it.

But all these things -- web, social or crowdsource -- have one thing in common: they're outside the classroom.  The internet is a solo phenomenon -- why would anyone work the web in the classroom?  You might as well be at home!
Perhaps the biggest problem with computer-based learning is that it's all independent, so we can all make excuses and slack off.  Maybe we'll start writing a blog post instead of studying, and we'll pretend to ourselves we're doing something productive....

But anyway, all that notwithstanding, there are still people who dream about getting social media into the classroom as an article in today's Guardian shows.
The justification for a focus on social media is, as far as I'm concerned, a non-sequitur:
We are late to the party. Children now default to social media in nearly every aspect of their life. They use it to communicate with their friends, play games and watch TV. Our failure to provide language learning resources must partly be due to teachers and parents who either don't appreciate or don't understand the power of social media.
To paraphrase: kids use it outside the class, so we must use it inside the class.  But what of all the other things kids use outside the classroom?  When I was a child, did the teachers use BMX bikes and Action Man to teach us stuff?  No.  They used TV to an extent, but even then it generally had the goal of making up for the classroom teacher's defiencies -- most primary teachers don't know much science, for example, so we watched a weekly science programme.  So TV wasn't used because it was "what the kids use" but because it was the appropriate tool for the job.  Any attempt to crowbar teaching into social media for its own sake is likely to be as successful as any other attempt adults make to "get down with the kids", and it will just look patronising.
At the end of the article, the author gives "Five ways you can start to engage with your pupils on social media", and if the term "engage with" doesn't immediately turn you off, you'll discover that once again, the "method" is not led by pedagogy, but by technology:
1. Create a Facebook page that your class can 'like'. Start posting updates to your timeline, but not in English. Ask your pupils to translate the text using Facebook's in-line Bing translation tool and ask them to gauge its accuracy.
It looks to me like it's a matter of "it's there so we'll use it" -- correction is an occassional exercise in most classrooms, but here it's being promoted as part of the routing.  Because it's there.
(This may actually be counterproductive -- why draw students' attention to the ubiquity of free translation tools if you're trying to demonstrate to them the value of learning a language?)
2. Create a Twitter account. Start tweeting in a foreign language, keeping in mind that you have a 140 character limit, and see if your pupils can strike up a conversation with you. Impose a non-English only reply and retweet rule.
Nothing of note - nothing groundbreaking.  Write stuff.  But short.  OK.  Understood.  But what stage do you get to in school language before you're able to write over 140 characters anyway.  (See also "idiographic languages"...)
3. Create a YouTube account. Ask each of your pupils to record a video blog, or 'vlog', of their hobbies, thoughts or opinions on topical news stories, but speaking only in a foreign language. Those who want to have their video uploaded should send it to you first.
I'm sorry, but I fail to see any teaching goal that is achieved by putting these on public display on YouTube.  In fact, as a student it would horrify me to think that my errors would be on show, but then I'd probably feel browbeaten into agreeing, seeing as everyone else is doing it.
Why is this any better than setting a spoken homework task that only the teacher and/or the class sees?
4. Create a Pinterest account. Take some pictures of prompt cards, post-it notes or even objects with their description in another language and 'pin' them on your boards. You could even look for photos of the country, or infographics about languages in general, to help your pupils understand more about why they should learn it.
I can't imagine this getting much use.  What is the student's goal in using the pinboard?  Why would I go to it?  What on Earth could you put on it that would grab enough attention to make the endeavour worthwhile?
5. Create a blog or Tumblr. Dedicate it entirely to publishing content in the language you teach. Show your pupils why you love the language and inspire them to do the same. Ask them to write something, however small, and post it for the whole world to admire.
All well and good in a liberal arts atmosphere, but like the videos, where's the innovative step between this and written homework?  Besides, as a learner and a teacher, I prefer much more closely directed work.  Why?  Am I a tyrant?  Hell no.  An open task normally leads to either getting stuck in a rut using the same old "safe" language items over and over again, or frequent overreaching by trying to say something that you haven't been taught yet.
As a learner, that frustrates me.  As a teacher, it makes things harder to mark.  What feedback do you give in the latter case?  Do you write five or six lessons worth of teaching?  Do say just mark it as wrong and say you haven't done that yet?

So I don't think social media's a great answer for the teacher, really.  A fantastic resource for the independent learner, yes.  A great way of maintaining a language in the absence of other contact, yes.  But an important component of the modern classroom, no.


Edit 11/04/2012: Tatoeba's actually better than I thought.  The search engine may not be as flexible or as specific as true corpus concordancing software, but it's pretty powerful, as the docs indicate.  There's a "proximity operator" in there, so you can look for co-occurring words even if you don't have the exact phrase.  Nifty.

08 April 2012

3 months or 5 days?

You may be aware of Benny Lewis and his blog Fluent in 3 Months.  If so, you'll probably be aware that he's generally involved in some controversy or other.  If you had to pick someone as his "arch-rival", you'd probably have to go with Steve Kaufmann.

Now I'm not a fan of either, but this morning I was alerted to a new video Kaufmann posted on YouTube which would appear to be a direct challenge to Benny.  Not only is it called "Fluency in Five Days", but he even makes reference to the 3 month thing in his opening.  Take a look if you like:



The whole thing hangs on the meaning of "fluency" -- in technical terms, fluency is the ability to function flowingly and without (much) hesitation.  The other parts of language are "accuracy" and range.  So what he's proposing to do is just to cram as much and build his accuracy, and then immerse himself for a week to get up to speed in actually using it.

Basically it's a direct challenge to Benny's idea that you have to speak from day one (which is the name of Benny's current commercial product).  I have to say that I agree with Kaufmann on this.  I don't agree with his methodologies (or that of his commercial website Lingq), but I definitely agree that there is value to initial study without any immersive usage... as long as you later go into an immersive situation.

I've always said that the value of an immersive environment is limited by what you know going into it -- the more you know, the more you'll pick up.

I look forward to seeing Steve's results.

07 April 2012

Languages in schools...?

It's well-known throughout much of the world that the UK has an exceptionally bad record at teaching languages in schools.  Every few years, someone stands up and vows to change this.  Years ago the Scottish education department declared that everyone going into primary school teaching had to be able to teach a second language.  As I understand it, England followed suit shortly after.  What ever came of it?  Not much, because as language is only compulsory for two years at high school either side of the border, and the vast majority of teacher-training candidates didn't have any language qualifications at all, and enforcing the law would have led to a serious lack of primary school teachers all round.

So instead they started special "[insert language name here] for primary teachers" courses, like this one from the Open University.  It's for complete beginners, and costs only £275 regardless of where you live in the world -- that's equivalent to 5% of the tuition fees for a year of study at an English university, meaning it can't be particularly substantial.  According to the site, it can be done entirely at your own pace, so it doesn't seem like you'll get any live tuition whatsoever.  This really cannot teach you enough to turn you into an effective language teacher.  And yet it must be done, to pay lipservice to an admirable goal that is simply too expensive to meet.

Looking at the resources that are used in primary school languages, I can only despair.  Take, for example, Languagenut, a computer-based resource claiming to meet the "Key Stage 2 Modern Foreign Languages" syllabus -- that's an English thing, but they also say that they're aligning to the Scottish "Curriculum for Excellence".

It's a colourful and appealing package, but where is the substance?  This isn't Languagenut's fault, really.  The Languagenut software is nothing more than a repackaging of the most common techniques in computer-assisted language learning (CALL) along with the syllabus goals of the primary curriculum.  The problem is in the whole culture of CALL and the whole system of primary education.  The idea of "games" that involve very superficial matching exercises is engrained in both cultures, as is the concept of "modularisation" that leads to the isolation and compartmentalisation of language, instead of continuously constructing a single large overview of the language.  Even if this compartmentalisation wasn't a bad thing in and of itself (and I certainly believe that it is), the act of compartmentalisation always leads to one resource blocking every other resource out, because no resource can ever be fully compartmentalised -- there will always be an underlying order to the material, and no two resources will follow exactly the same order.  Dipping in and out is usually very difficult indeed.

"Resources" will never be a replacement for a teacher in the classroom.  What is required is teachers who know the material.

But maybe there is a way to get the competence we need, at both ends.  I don't know about England, but in Scotland there's a surplus of qualified teachers, as the government made a point of ramping up training in anticipation of a mass retirement that's due any year now.  Why don't we have a government scheme to get them abroad?  The demand for teachers of English is always high, and these guys are trained and experienced classroom teachers.  When they come back, they'll be looking to teach.  So why not set up a scheme where teaching grads get the first shout on teaching assistant posts overseas, reciprocated by bringing in foreign teaching grads as language assistants.  The language assistant schemes are atrophying at the moment, and could really do with government intervention to prevent the local authorities closing them down as an unnecessary expense.

Certainly, it will take something far more radical than a few ministerial proclamations to raise the standard of language teaching to anything useful. If in doubt, have a look at the Scottish Standard Grade or English GCSE exam papers to see what's expected of you after four years of learning....

06 April 2012

Oops!

Ooops... this wasn't a finished post -- it was a set of notes for a future post.  Deleted.

It referred to an MA thesis regarding the teaching of Finnish grammar.

http://scripties.let.eldoc.ub.rug.nl/FILES/root/Master/DoorstroomMasters/ToegepasteTaalwetens/2010/Tilma/MA_0243450_C_Tilma.pdf

04 April 2012

Reinventing old (and useful) wheels, and forgetting etymology

If you collect old books, particularly specialist ones, you'll have noticed that many of them have something called a "subscriber list" at the start.  It seems a bit odd to us.  I've subscribed to several publications, and I never expect to get a credit in the magazine itself.  But some book editors even went as far as to write prefaces praising the subscribers for their support.

A subscriber wasn't someone who just bought stuff regularly but someone whose money was directly responsible for getting the book published in the first place.  I recently heard the Gaelic translation of the term: "fo-sgrìobhadair".  Literally "under-writer", which is etymologically the same word as "subscriber".  It's one of those moments when you want to slap yourself for having missed something so obvious for such a long time.

Essentially, the subscribers of old made sure there was enough money to compile eg. a dictionary and do an initial print-run.  This took the risk off what would otherwise have been a very expensive, low turn-over item.  A subscriber was, in the traditional sense, a patron of the arts.
It's an idea that died out with the increasing corporatisation of publishing.  Why would anyone give charitable donation to a commercial publishing operation?
But every good idea will be had several times, and subscription is back with a new name: "crowdfunding".  A lot of people don't really get it -- they want to look at it as an investment, but it's not.  It's people putting up enough money to ensure the product gets made, with the product itself as the only reward (well, actually, many crowdfunding operations have a sliding scale of rewards, but these are often just trinkets like stickers etc).

This crowdfunding model has been used for everything from industrial design and manufacturing to software development.  It's even gone right back to where it started and been used as a way to get funding to set up a new printrun of out-of-print books.

It's an idea that certainly has its benefits for the language world.
One company put up an adventure game project and said that it would be translated to other languages if enough money was put up.  Unfortunately, the platform they were using didn't let you pledge on the grounds of a specific language.  There are several languages that would have got me to pledge my money (I have never come across an adventure game in Catalan, for instance, and I find adventure games quite a good way to practice language -- although that's a matter for another post) but in the end I didn't pledge.
One of the questions that keeps coming up on the internet is "if you love the language, why don't you teach me for free?"  The answer is quite simple: because I have to make a living.  If I spent a lot of time putting together an effective course then gave it away, that would be several years of my working life out the window, with nothing to show for it.  But if people are willing to stump up the dough up front, there's no problem, and that's what the guys at languagehunters.org are doing -- they've raised cash via a Kickstarter campaign to fund them filming a set of videos teaching the Irish language. Now I'm not convinced by their teaching methodology (regular readers already know what I think about the notion of learning like a child, and I'm not a fan of excessive signing), and most of the methodology is really no different from what millions of teachers around the world are already doing, but the idea of funding in this way certainly appeals to me.

Except that the figures don't really add up.  The Kickstarter page says that they'd spent 400 man hours before even opening the pledge drive -- that's already worth more than the £3820 they asked for, so they're still going to have to get a lot more money later down the line.

Wouldn't it be better if they could raise enough money to make it completely free, available for posterity?  Which all ties into what I was asking last time: to what extent can we justify having the individual pay for and/or produce language materials for others (possibly including corporations) to use later down the line?

I now find my asking myself what language projects I could propose where I could sell my time in advance via Kickstarter.... anyone want me to produce a free Quechua course?  I'll need a plane ticket to Peru, somewhere to stay for a few months, a few classes in the language, some time to write the course............