Re: [lojban] Amusing stuff

[Ivan A Derzhanski <iad@MATH.BAS.BG>]

Robin wrote:
> Ivan A Derzhanski wrote:
> > [...] comparing sizes of published dictionaries is not a sound
> > way of comparing the sizes of vocabularies, although it is perhaps
> > useful as a guideline.  There are many other factors, both
> > linguistic (the productivity and predictability of derivation
> > and compounding) and extralinguistic (the diligence of the
> > lexicographers, the resources available to them).
> >
> Yes - it all depends on what you count as lexis.  As I said, Turkish
> has a very small "dictionary" vocabulary, but someone once wrote
> a book explaining all the forms of one verb (sevmek) - it had over
> 1400 entries

The bulk of these don't count, being inflected forms of the same
word(s).  Only the different voices (reflexive _-in_ / reciprocal
_-is~_, causative _-dir_/_-t_ (which can be doubled or tripled),
passive _-il_ and the various combinations of these -- a total of
24 possibilities) can and often do develop unpredictable meanings,
so a dictionary should register them at least when they do.

> (this was originally his PhD thesis, which shows that
> even having a PhD in linguistics may not mean very much!).

Was it a PhD in linguistics or in Turkish philology?

> > I've never approved of the statement `English has more words than any
> > other language', because I'm not convinced that the people who say so
> > have done an exhaustive search, or even considered some likely rivals
> > such as Hindustani.
> 
> Not sure on this one. The people at the OED have done pretty exhaustive
> research, but then they haven't looked at Hindustani.  Similarly with
> Ottoman Turkish - one view is that since educated speakers were (at
> least) triglossic, one should also include the entire vocabularies of
> Arabic and Persian.

Yes, that's the sort of thing I had in mind.  Assuming that English
does have more words than any other language, why might that be?
It's not because it has particularly productive word-building
mechanisms; on the contrary, it's very restricted in that regard
as Indo-European languages go.  Its inherited lexicon can't be very
different in size from that of Dutch or German.  Remains then the fact
that it has ingested large portions of the vocabularies of other lgs,
such as French, that it has been in contact with.  Which is not common,
but not unique either.

> My students tend to produce consistent errors, like "thougth" and
> "ougth" or suffer from first-language interference, while many
> native-speaker friends seem completely at sea.

Your students have the advantage of having had their first exposure
to the notion of reading and writing through Turkish, which means
that one of the first things they've learnt (though not necessarily
explicitly) is that speech consists of sounds and writing is a more
or less straightforward matter of representing each sound by its
corresponding letter.  They're in possession of an intuitive idea
of phonology, which is a very handy thing to have when they meet
languages in which the sound:letter correspondence is not quite so
straightforward.  By contrast, English native speakers are often
completely confused about what speech sounds are and what it means
to write them down.

> > (5) It has in fact been suggested that irregular inflected forms serve
> > a purpose (they're a sort of mental shortcuts, since they are looked up
> > as they are, so no time is spent on their morphological analysis), [...]
> 
> Is there any empirical evidence to support this theory?  It's
> interesting, but historical (and occasionally phonological)
> explanations seem more productive.

I wish I could say more.  I don't even remember who I heard it from,
let alone the details.  I may still have a handout from that talk,
if there was a handout, but even that I'm not sure about.

--Ivan



------------------------------------------------------------------------
Law.com is the preeminent online destination for legal professionals.
Visit Law.com for exclusive content from American Lawyer Media, online
CLE Seminars, Practice Centers and Career Listings.
http://click.egroups.com/1/5803/4/_/17627/_/963125890/
------------------------------------------------------------------------

To unsubscribe, send mail to lojban-unsubscribe@onelist.com