Fwd: digraphs and sorting

[James Kari]

---------- Forwarded message ----------
From: Robert Hsu <hsu at hawaii.edu>
Date: Mon, Jul 30, 2012 at 6:31 PM
Subject: Re: digraphs and sorting
To: James Kari <jmkari at alaska.edu>
Cc: gmholton at alaska.edu, "S. Hargus" <sharon at u.washington.edu>


Hi Jim--  Thanks for forwarding this thread from Gary.



Gary:  I don't know of any studies, and in any case general studies may

not be too relevant since in a particular language maintenance situation

this is probably a decision for the speakers, like orthography design.



   Subjectively I think I prefer digraphs and trigraphs to be alphabetized

as sequences of single characters like th in English--fewer primary letters

to remember the order of.  That would reduce the need for a mnemonic aid

like the footers in the Koyukon dictionary.  Also, from a tree-traversing

point of view, if there are fewer distinct units--fewer possible branches--

searches can be whittled down faster, though the depth would increase a bit.

Still, these are only theoretical speculations and subjective feelings.



   A similar situation occurs when certain segments or distinctions like

voicing or dialectal variation are merged in alphabetization, e.g.,  gh/g,

k/ch, b/m  in Koyukon, or  nh/n, s/z  in Deg Xinag.  Other distinctions are

also commonly merged, like long and short vowels.  Merging also reduces

the number of primary letters, possibly speeding look-up.  It has the

further benefit, when disinctions with low functional load or which are

otherwise easily confused are merged, of simplifying look-up when the user

is not sure of the values (length, voicing, etc.) in the word.



   Apart from ease of use, another issue is whether the phonemic inventory

need be enshrined in the organization of a dictionary designed for language

maintenance, as opposed to a reference dictionary.  (In fact, for the

former, must the orthography itself be phonemically entirely faithful?

The fully phonemic transcription can always be placed in a separate
band.)




   This doesn't answer your original question about digraphs.  But with

on-line dictionaries being practical now, we can have the luxury of

presenting a dictionary in more than one form without worrying about cutting

down more trees.  The one example of this I know is the Marshallese On-line

Dictionary, which is presented in both the original phoneme-by-phoneme order

and "unified" order which merges plain and modified letters:



    www.trussel2.com/MOD



You could do the same with the different treatments of digraphs, and gauge

user preference.  Although looking up a word through an electronic interface

is a different experience from using a book, browsing the electronic

versions should give a feel for the consequences of the different ordering

systems, something impossible to capture in the abstract.



Best wishes,  Bob




On Wed, Jul 25, 2012 at 8:29 AM, James Kari <jmkari at alaska.edu> wrote:

> Hi Bob,
> how are you? this is a subject you know something about.
> Lots of activity here. I attach an abstract for Ath. conf next month in
> Bellingham. program is at uaf.edu.anlc/alc
> best regards
> Jim
>
> ---------- Forwarded message ----------
> From: Mike Morgan <mwmbombay at gmail.com>
> Date: Wed, Jul 25, 2012 at 7:22 AM
> Subject: Re: digraphs and sorting
> To: ATHAPBASCKAN-L at listserv.linguistlist.org
>
>
> I would imagine that it would depend on the existing level of literacy
> in a majority language, and what orthographic conventions are used by
> that majority language community. If most members of a community are
> already used to the idea that "th" is a separate sound but is
> alphabetized after "tg" and before "ti", then presumably following
> this practice in creating an orthography would make things easier.
>
> Though, of course, making things easier is NOT ALWAYS the primary
> goal, and sometimes OPPOSING any existing majority language practices
> is preferred as it increases the degree of distinctiveness of the
> minority language.
>
> as I said though, although I think I have seen studies on this
> somewhere, I have no idea where or when they might have been...
>
>
> PS technology can also enter into the mix: Welsh has a long history of
> treating "ch", "dd" and "ll" as separate digraphic characters, and
> alphabetizing them accordingly (so "ch" follows "cz", etc). With the
> advent of word processing, but before special software which
> alphabetized according to welsh rules, things started to change, and
> many people followed English practices ("ch" after "cg"). Now that
> such software is ubiquitous, it is (mostly) back to tradition...
>
> mwm
>
> On Wed, Jul 25, 2012 at 8:40 PM, Gary Holton <gmholton at alaska.edu> wrote:
> > Hello,
> >
> > I know this list doesn't get a lot of traffic, so apologies in advance
> > for spamming you with this query. For years I've accepted without
> > question the orthodoxy which sorts dictionary entries by digraph
> > rather than by single characters. This makes obvious sense, since
> > digraphs such as th or even trigraphs such as tł' are single phonemes
> > and hence shouldn't be relegated to secondary status within a
> > dictionary. On the other hand, we also know that many languages do
> > just fine treating digraphs as separate characters for the purposes of
> > dictionary sorting (e.g., English has no "th" section; Malay has no
> > "ng" section). So, my question is, does anyone know of any usability
> > studies -- or even just subjective account --  comparing the relative
> > advantages of each approach within a language maintenance situation?
> >
> > Gary Holton
>
>
>
> --
> mwm || *U* C > || mike || мика  || माईक || マイク || மாய்க் (aka Dr Michael W
> Morgan)
> sign language linguist / linguistic typologist
> academic adviser to "Nepal Sign Language Training and Research Centre"
> project
> NDFN, Kathmandu, Nepal
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/athapbasckan-l/attachments/20120730/0d588e70/attachment.htm>