LINGUIST List 2.264

Sunday, 2 June 1991

Qs: Register, Turkish, Diacritics

Editor for this issue: <>


  1. , Query: "Register" as a technical term
  2. Mark Seidenberg, Turkish query
  3. John Baima, ISO10646 and dicritical marks

Message 1: Query: "Register" as a technical term

Date: Fri, 31 May 91 16:36:00 EDT
From: <GL250007Venus.YorkU.CA>
Subject: Query: "Register" as a technical term
I would appreciate hearing what "register" means as a technical
term. I have a sense of the term in Halliday's use, and wonder how
this meshes with other uses.

You may send replies to me directly or to the list. My email address

Bill Greaves
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 2: Turkish query

Date: Fri, 31 May 91 15:04:06 PDT
From: Mark Seidenberg <>
Subject: Turkish query
Could someone please point me to (a) a computer-readable Turkish lexicon
(a list of words and their pronunciations), and/or (b) work on morphological
parsing in Turkish?

Thanks in advance.

Mark Seidenberg
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 3: ISO10646 and dicritical marks

Date: Sat, 25 May 91 18:08:35 CDT
From: John Baima <>
Subject: ISO10646 and dicritical marks
ISO10646 is an attempt by ISO to devise a global character coding
standard. It seems to me that the fate of ISO10646 will have a
significant impact on linguists in the years to come.

I have been one of the people arguing for the inclusion of floating
diacritical marks. However, several in the ISO world strongly object
to them. One person who has frequently spoken against floating
diacritical marks is Mr. Johan van Wingen. He is a representative to
ISO from the Netherlands and is on several of the ISO committees. In
the following message, Mr. van Wingen is responding to one of my
messages to the ISO10646 list (quoted with ">"). I can respond, but I
would like to get the responses of the people of LINGUIST and then
post a summary (if anyone wants to respond directly, the ISO10646 list

Thanks for any responses,

John Baima
Date: Fri, 24 May 91 17:51:00 CET
From: "J. W. van Wingen" <BUTPAAHLERUL2.BITNET>
Subject: RE: Re: ECMA enhancement proposal for ISO/IEC DIS 10646
Sender: Multi-byte Code Issues <ISO10646JHUVM.BITNET>

Dear Colleagues
As I told you before, I have got a tremendous backlog in replying to
messages in this list, which deserve one. Here is the first.

> While a couple of people have responded to this message, I would like
> to add one more point. Minority languages need floating diacritical
> marks. There are about 6,000 languages in use today. Within the
> lifetime of ISO10646, many minority languages will enter the computer
> age. Most of these languages will require diacritical marks because a
> majority of them are tone languages. Other languages will need
> diacritical marks for other reasons. Most minority languages will be
> written in the national language script with as few additional
> characters as possible, with the addition of diacritical marks.

Here some issues are confused.
1. Minority languages are so special that they must be written with
2. Because they are so small, no fixed letter/diacritic combination
 is available in the general repertoire.
3. Thus for these languages an ad hoc solution is necessary: the
 Floating Diacritic that combines at will a letter with a diacritic
 (this has the advantage that when a minority language is repressed
 by the regime, no modification of the ISO repertoire needs to be
 proposed to remove the letter).
4. Because minority languages are written with FD, ISO 10646 should
 contain this concept.

The essence of the argument is that it pertains to a spelling problem,
not a coding problem, and thus is strictly spoken outside the scope of
this discussion list.

The first question is: why should languages be written using special
characters or diacritics. Finnish and Hungarian are (distantly) related
languages, but with the first the difference between long and short
vowels is indicated by doubling the letter (aa/a), with the second by
by putting an acute over the long one. Which solution is the better?
The only answer can be given by experimental research.

If for a language that never has been written before, a writing system
is developed, one can discover a certain pattern. First, several
missionaries produce a system, each different from the other. Then a
linguist publishes a grammar or a lexicon, unifying the spelling, and
using an Extension of the International Phonetical Alphabet, totalling
to at least 60 small and capital letters. This system is then taught
at schools; a few books, and some monthly journals are printed. Then,
a New Government raises the language to Official status, and appoints
a committee to simplify the spelling, because the cost of ordering
special typewriters and printers is becoming too high.

> I don't pretend to tell people what they should or should not do, but I
> think that it needs to be made clear that a vote against diacritical
> marks is a vote against thousands of minority groups.

This is putting the fault at the wrong door. It assumes that the way of
spelling must be in that way and cannot be in another. In fact, it is
the fault of the linguists who designed the spelling while ignoring
problems of technical reality, requiring corrections at a later stage
from the government. It occurred even with the spelling for Dutch, by
de Vries en te Winkel, highly respected scholars. To understand the
rules, knowledge of Ancient Gothic was needed. Only after a painful
process, a simplified system was established by law, in 1954.

> But again, when
> ECMA, SHARE, European Governments, and others vote (officially or
> unofficially) against diacritical marks, they are voting against
> minorities. It isn't just us Americans (including the "California
> Dictate") who are concerned about minorities, right?

I have the greatest respect for the Soviet linguists who designed the
spelling for Kaukasian languages in the 1930ties by avoiding diacritics
almost completely. They realized that writing all the 75 phonemes of
Abkhazian each by a separate letter would create an unmanageable system.

We should be concerned about minorities, especially where they have been
the victim of a linguist. One of the worst jokes these people produced
was to designate the dot over the i as a diacritic. The people of Turkey
have to pay dearly for it, up to this day.

I have a suggestion. Let we introduce a law, that every linguist who
wants to introduce a new character, not yet existing, has to pay an
amount of $ 100 000 as a contribution to the cost caused to the
information industry as a result of his invention.

Best regards, Johan van Wingen

[End Linguist List, Vol. 2, No. 0264]
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue