Phonology of numbers: BSL TEENs

Thu May 23 22:56:43 UTC 2002

I think I dropped a ball while juggling ideas in my last message, Adam...

I had written:

What seems to me to be going on in each of these cases is that
un-symmetric signs are allowed if there is a morphological motivation
that is more powerful than the phonological pressure to respect
Symmetry. This seems to be the case quite systematically in two-handed
number systems which compound FIVE on the non-dom. hand with some
other number on the dominant hand, combining them with some movement to
derive numbers beyond 10. This is exactly the case for the BSL numerals
that
you talk about, Adam.

And you commented:

> My problem with the simultaneous compound analysis for BSL signs like
> SIXTEEN is that the FIVE on the non-dominant hand combines with SIX
> on the dominant hand (an I handshape or an A with an extended thumb
> bent at the middle joint). The SIX handshape can occur alone to mean
> 'SIXTEEN' or together with the non-dominant FIVE handshape to mean
> 'SIXTEEN'. It's thus not clear how the FIVE handshape contributes any
> meaning to this sign (i.e., it's not clear to me that FIVE handshape
> means 'five' in this sign).

I'm pretty sure I was wrong in claiming that the BSL numeral system
works like LSF-type systems, where once FIVE is reached, ONE and so on
on the dominant hand are then compounded with FIVE on the non-dominant
hand. If I remember right, BSL 1-5 fall under what could be called a
cumulative IMRPT extension series (starting with extended I(ndex) for
ONE and progressively adding each subsequent finger up to FIVE). Then,
for SIX to NINE the language has *two alternative* cumulative extension
series, either TIMR or PRMI. (I've always wondered how these two systems
are distributed across the BSL signing community...) As far as I know,
the non-dominant hand doesn't come into play until TEN -- is this right?
And TEN, if I remember correctly from the BSL Dictionary, is two /5/
hands moving briefly forward in a supinated position. So, I *hope* I
have the basic facts right now, because whether what I'm about to say is
useful, or a waste of time, hangs on this. So I'll take the plunge, in
hopes of being corrected if I'm hopelessly wrong in my assumptions about
the facts.

On the assumption that TEN in BSL is formed with two /5/ handshapes, it
could (possibly) be analysed morphologically as being composed of two
FIVE signs simultaneously compounded. Another analysis would treat the
sign as monomorphemic but being lexically specified for two hands, to
which a phonological /5/ handshape specification spreads as in other
symmetric two-handed signs. In this case, the original bimorphemic
structure can be assumed to have "bleached" away. If this analysis is
correct, then it would not be unexpected to find simplification to a
one-handed version in some contexts (via Padden and Perlmutter's Weak
Drop or its converse Strong Drop). I think it is plausible that this is
what is happening in SIXTEEN-NINETEEN in BSL. If you simultaneously
compound SIX with the two handed TEN, apart from growing an extra arm
and hand, the only way to deal with the resulting mismatch between the
phonological specification and the available articulators is to delete
one hand's worth of phonological information. SInce the only way to do
this without jettisoning essential information is to get rid of one hand
in TEN, this is the most likely route to take. The lexical /5/ handshape
of TEN then associates only to the non-dominant hand, recoverable
nonetheless as TEN, while /I/ or /A.../, /Iv/ or /L/ etc., depending on
the dialect, associate to the dominant hand. What you are dealing with,
in sum, is just a simultaneous compound of the units numeral morpheme
plus TEN, which has been phonologically reduced via Strong Drop. With
the up-down alternating movement, you have an extra piece of information
that these are -TEEN series numbers. The process I have described so far
seems plausible as a either a synchronic or (in part) a diachronic
explanation of the form of these compound numerals. In fact, it is
equally plausible in purely synchronic terms that the -TEEN morpheme is
composed of the /5/ handshape on the non-dominant hand *plus* the
specification(s) -- however represented by a given model -- for the
alternating movement. (A hypothesis about a possible origin for this
movement just below...)

Coming back to the question of Symmetry violations, BSL seems to allow
this combination of different handshapes plus alternating movement due
to a morphological motivation for the extra complexity, but Auslan
appears from what you say to be more finicky, along the lines discussed
in my previous message. Just one question to clarify things for me: in
the Auslan variants, do you still have a repeated up-down movement in
the one-handed signs?

A final word on the origins of the alternating movement itself. I talked
about the (n)-TEEN signs as simultaneous compounds of a unit plus TEN. I
can imagine this coming from an earlier stage in which the compound was
sequential, with the unit sign followed by the two-handed TEN. In this
case, the non-dominant hand would (typically) have to rise from resting
position to the flexed forearm position typical for a number. This is
one part of the story. The other has to do with what I *believe to be*
the movement associated with TEN in BSL. In SL of the Netherlands
(northern dialect at least), I know that the two supinated /5/ hands
move briefly forward and downward (in anatomical terms, the flexed
forearms extend slightly). I seem to recall this as being illustrated in
the BSL dictionary: as I said before, my memory may well be wrong and
what I have to say here hinges on this and would collapse as a result.
*If* this is indeed how the sign is formed, then in the (n)-TEEN signs,
the dominant forearm would move slightly downward and outward from the
units sign at the beginning of the (sequential) compound to the TEN at
the end. The result, taken together with the non-dominant hand rising
into position for the sign TEN, is the dominant and non-dominant hands
moving in opposite and complementary directions. There are only two
steps from this sequential compound to the overall form of the -TEEN
signs as you have described them. First, the handshape of the units sign
on the dominant hand spreads onto the final position, dislodging the /5/
as it were; second, the simultaneous movement of the two hands in
complementary directions is reinterpreted as alternating movement, with
a return movement on each hand inserted to fill in the cycle.

This hypothesis leads me to wonder exactly how ELEVEN through FIFTEEN
are formed. Are they all one-handed? I seem to remember (at least one
variant of) ELEVEN being formed exactly like ASL TEN, in other words,
with an /A'/ handshape plus a repeated twisting forearm movement. How
are the others formed?

..
..
..

I was hoping to post something this week about exceptional handshapes
and related movements in number systems, but I have been attending to
other matters as well, so I'll leave this to next week sometime. Suffice
it to say for the moment that though certain rare handshapes are
"forced" into a language's inventory by virtue of forming part of a
given numerical (or other type of) series, other phenomena that at first
glance appear unusual may in fact be unexpected clues to underlying
properties of sign language phonologies that are otherwise difficult to
observe. More on this next week...

Chris

p.s. Please let me know if I am in fact mistaken or not in the
assumptions I made about BSL number signs! It would help me rethink my
hypotheses if necessary.

==================================================
Christopher Miller
Department of Linguistics
University of Manitoba
Winnipeg MB
R3T 2N2
Canada

+1 204 474-8343 (office)
+1 204 474-7671 (fax)
+1 204 951-5002 (mobile phone)

millerc at ms.umanitoba.ca
==================================================