5.0155 Responses: E-Accents (2/104)

Elaine Brennan & Allen Renear (EDITORS@BROWNVM.BITNET)
Tue, 18 Jun 91 11:04:48 EDT

Humanist Discussion Group, Vol. 5, No. 0155. Tuesday, 18 Jun 1991.

(1) Date: Thu, 13 Jun 91 19:27:53 CDT (27 lines)
From: "C. M. Sperberg-McQueen" <U35395@UICVM>
Subject: Re: 5.0142 Qs: E-Accents

(2) Date: Mon, 17 Jun 91 10:11 BST (77 lines)
Subject: Re: accents (Re: 5.0142)

(1) --------------------------------------------------------------------
Date: Thu, 13 Jun 91 19:27:53 CDT
From: "C. M. Sperberg-McQueen" <U35395@UICVM>
Subject: Re: 5.0142 Qs: E-Accents

Two queries from Humanist 5.0142:

>A colleague in Slavic, new to computers, asks if there is a standard
>character set for Czech or Croatian-- specifically, where are the
>characters with hacek placed?

Try ISO 8859 part 2 and ISO 6937. The latter is widely used in teletype
and similar services, though not in computing. It uses floating
diacritics, which makes it hard to implement on some devices.

>Is there a recognised convention for showing accents in email?
>(so for example, e acute might be represented as e followed by a forward

Not really, though a number of people have developed ad hoc conventions
and there are a few proposals which have actually been thought through.
Three have recently been discussed extensively on the ISO10646 list at
JHUVM, to which you should refer if you are really interested in the

-C. M. Sperberg-McQueen
(2) --------------------------------------------------------------85----
Date: Mon, 17 Jun 91 10:11 BST
Subject: Re: accents (Re: 5.0142)

I'm not sure if an accent convention exists for European languages
(presumably one must) but one is certainly developing for the transcription
and e-sending of Sanskrit (and hence most S Asian languages). I enclose
a recent posting from the INDOLOGY list, which may be interesting because it
deals with the potential problem of using inverted commas:


This message has the dual purpose of a) seeing whether I have
correctly recorded the addresses of those who wish to participate in
the discussions on Sanskrit syntax, and b) making a proposal about
standardizing the way we send Sanskrit by e-mail.

Sending diacrtically marked romanized Sanskrit by e-mail is
difficult, and sending devanagari is (as far as I know) impossible.
May I suggest that for the purposes of these discussions we adopt
the set of conventions being used on the Indological Forum

These conventions are in fact the commands used to print Devanagari
in the fonts designed by Frans Velthuis for TeX typesetting
software. I find them simple and intuitive.

Velthuis's conventions are as follows:

1) LONG VOWELS ARE DOUBLED. (This also happens to be phonetically

The vowels of Sanskrit, then, are a, aa, i, ii, u ,uu etc.


Thus the retroflex class of vowels and consonants is written:
.r .t .th .d .dh .n .s

Visarga is written: .h

The palatals that take diacritics are:
^n and "s

The guttural nasal is written: "n


Since quotation mark (") has been appropriated as a diacritic, one
can use doubled open and close quote marks to indicate quotation:
``This is within quotation marks.''

Since all of us can distinguish Sanskrit from English and can
probably make a good guess at distinguishing between words that are
being used and words that are being mentioned, I propose that we
minimize the use of quotation marks. That is, instead of writing:

The word ``"sabda'' means word.

one could just as well write:

The word "sabda means word.


Like any new set of conventions, these may seem a bit awkward at
first, but I have found it easy to adapt.

If any of you have principled (or even unprincipled) preferences for
some other set of conventions, please state them fo us to consider.

Richard Hayes <cxev@musica.mcgill.ca>
Faculty of Religious Studies, McGill University"

Marcus Banks