Re: [lojban] pedagogy and lojban

[Edward Cherlin <lojban-out@lojban.org>]

On Wednesday 31 July 2002 02:19 am, G. Dyke wrote:
> Hi
>
> I'm currently teaching Lojban to a Frenchman who knows very little
> English (and has no interest in learning it - English, not Lojban).
> I'd set out with the idea of translating the lessons, but I don't
> quite like the way in which they are presented (in addition to
> which, I find them a bit long for the purposes of teaching one
> person).
>
> I was wondering whether anyone had any opinions about the way to
> introduce various concepts. 

I had some TESL (Teaching English as a Second Language) training in 
the Peace Corps, including a taste of the Defense Language 
Institute's "wavelet" teaching system. It takes the concept of 
structure drills a step further. Instead of trying to teach whole 
branches of grammar at once, it divides the task into a succession of 
small steps, each of which is drilled to the point of mastery before 
proceeding further. 

A structure drill uses a sentence form with variable parts. The 
instructor gives some cue, and the student utters a sentence in the 
same form but with some part changed. For example, 

I: "You are a student."
S: "I am a student."
I: "I am the teacher."
S: "You are the teacher.

In the example above, the teacher varies the person of the pronoun 
and the description. Then the student makes a transformation of the 
pronoun and copula.

A sequence of wavelet drills on this form could start with third 
person pronouns, then do first and second person singular, then mix 
all three, then do each case of the plural, then mix all of the 
forms. In Lojban there are more pronouns but no problem of verb 
agreement, so the sequence would be rather different.

I started to make a list of Lojban language features suitable for an 
introductory set of drills, but I haven't gotten far enough to 
present anything.

Is anyone interested in pursuing this?

> (My ambition is that the textbook be
> baselined from a translation of my French textbook zo'o). There is
> particularly the problem that introducing small things like no
> "la,doi" in names etc is a pain. Do I add a "read more here" at the
> end of each chapter or return to each subject at a later
> time(achapter dealing with morphology for instance). So here is an
> outline of what I have and how I think it should proceed.
>
> - pronunciation
> - three categories of lojban words
> - names (practice in phonological adaptation)

Don't try to teach all of pronunciation or adaptation of names at 
once. Start with the easy cases, so that the student has the 
experience of success to begin with, and introduce more difficult 
cases after the student has a framework to locate them in.

> - gismu as an application of brivla
> - bridi structure (put the {cu} in immediately)
> - introducing things that can act like sumti :
> -- mi
> -- do
> -- ti
> -- ta
> -- tu
> -- zo'e
> -- ko
> - getting rid of unnecessary zo'e
> - explain elision of cu
> - add {la cmen.} to sumti
>
>
> - explain le as extracting x1 place of selbri
> - complete cu explanation along with ku
> - introduce se te etc as means of extracting other places of selbri
> - use them in normal bridi
> - introduce fa fe fi
>
> - attitudinals
> - discursives
>
> - events as means to fill in the sturcture of sutra
>
> - numbers
> - lo, finish explanation of la and le
>
> - ...
> [Yet to be thought out]
> - ...
>
> ma se jinvi do
>
> mu'omi'e greg

I would have created a completely different order, at a much finer 
grain. I don't have time to lay out even as much as I have done in 
this vein today, but I'm willing to talk about it in pieces over time.

I would start with observatives and attitudinals, and proceed rapidly 
to variations on a simple sentence, such as

mi tavla do [la lojban] [la lojban]

starting with two or three pronouns, and proceeding to just a few 
other languages and topics of conversation, simple tenses, and the 
use of simple attitudinals in sentences. Then change sumti, and run 
through the process at a slightly higher level, including some of the 
elementary place structure manipulations. The obvious next candidate 
is "klama". :)

At some point, after several more sumti have been introduced, one can 
add selected simple little pieces of negation, abstraction, tanru, 
quantification, subordinate clauses, conjunctions, and the rest of 
the apparatus. Every selma'o will require its own sequence of drills, 
first at a very elementary level, and later in more complex 
utterances.

After running through the simplest cases of all of the fundamentals, 
come back around and do each category again. Run through a portion of 
the previous level very rapidly, and then again with a few of the 
variations and complications added, and so on round and round to 
fluency.

At the highest level, there could be drills along the lines of 
negated compound sentences or quantifications, and on to all the rest 
of the full logical toolkit taken out (if anyone wants to) to the 
Rube Goldberg level. :)

A glance through the Refgrammar shows the possibility of hundreds, 
perhaps thousands, of steps in such a sequence. Finding them is not a 
problem. Putting them into a usable order that moves along fast 
enough without skipping over something essential is a real challenge.

Of course, drill isn't the only thing needed. Drill can make the use 
of structure automatic in a formal setting, but then the student 
needs guided conversations where those structures can actually be 
used. Games are a good way to handle that structuring.

I have to stop now, but I will be happy to take any of these topics 
up again another time.
-- 
Edward Cherlin
Generalist
"A knot! Oh, do let me help to undo it!"
--Alice in Wonderland