Re: Unicode Mongolian

From: Timothy Partridge (timpart@perdix.demon.co.uk)
Date: Mon Mar 20 2000 - 14:48:59 EST


Peter Constable said:

> Hans:
> >"If the Mongolian script is adopted into horizontal text,
> >its lines are rotated another 90 degrees counterclockwise
> >so that the letters join left to right, and the columns
> >are transcribed to the equivalent lines (first column
> >becomes first line, and so on). IF SUCH TEXT IS VIEWED
> >SIDEWAYS,THE USUAL MONGOLIAN COLUMN ORD ER APPEARS
> >REVERSED, BUT THIS ORIENTATION CAN BE WORKABLE FOR SHORT
> >STRETCHES OF TEXT (emphasis added by Hans). Note that
> >there are no bidirectional effects in such a layout,
> >because all text is horizontal left to right."
>
> >Apart from an obvious mistake: - "first column becomes
> >first line, and so on" is wrong, rather last column
> >becomes first line, and so on is true, - what this
> >means in other words is the following: Unicode
> >necessitates a change of orthograp hy standards,
> >making the native reader reverse his habitual reading
> >order...
>
> I believe you may be misunderstanding the intent. Unicode
> doesn't require any changes to the orthography. The text you
> quoted is discussing what to do if a person wants to mix
> languages in a paragraph of text, Mongolian with something
> written LTR. The author is suggesting that it can be acceptable
> to present the Mongolian LTR. Apparently, the author seems to
> think this is accepted practice. This is provided in the
> Unicode book as an implementation guideline, but is not
> something that Unicode requires.
>
> If you believe that this implementation for mixing Mongolian
> with LTR text is not the preferred way to do it (I must admit,
> it's not what I would have expected), then I suggest you write
> to the Unicode list to say so. The text can be changed in the
> next version, and probably would be if there are clear
> precedents to suggest that the other way is used or even
> preferred.

> >It would have been easy to make Mongolian an horizontal
> >RTL script, (i.e. rotati ng 90 degrees clockwise). Then
> >viewing the text sideways would result in normal
> >Mongolian reading order! This way has been used by
> >scholars during the last 200 years, when Mongolian
> >studies grew up.
>
> It sounds like you're aware of a pretty good precedent for
> using the other rotation (making the Mongolian RTL).

Just over a year ago I made some comments about the draft bidi algorithm to
which Mark Davis responded.

Tim Partridge said:
> > ... in Table 3-5, the category AL lists Arabic, Thana and Syriac scripts.
> > Should it also include Mongolian? According to page 539 of "The World's
> > Writing Systems" [1], Mongolian is historically derived from a right-to-left
> > script, but rotated. (90 degrees anti-clockwise.) Of more practical
> > importance to implementers is the fact that European number strings are
> > printed rotated 90 degrees clockwise in Mongolian text. (For examples see
> > page 91 of "Writing Systems of the World" [2], or page 206 of "The Languages
> > of the World" [3].) Mongolian could therefore be considered to be a
> > right-to-left script printed vertically in the common Arabic style for mixed
> > direction texts. Unfortunately none of my books give information as to how
> > the traditional Mongolian digits are treated (AN or EN).

> > References
> > [1] The World's Writing Systems, Editors Daniels and Bright, OUP,
> > ISBN 0 19 507993 0
> >
> > [2] Writing Systems of the World, Akira Nakanishi, Tuttle,
> > ISBN 0 8048 1654 9
> >
> > [3] The Languages of the World (New Edition 1995), Kenneth Katzner,
> > Routledge, ISBN 0 415 11809 3

Mark Davis replied:
> I will talk to the Mongolian experts about this.

I don't know what the result of Mark's investigations was.

As further evidence of Mongolian being in an Arabic style vertical format
(LTR texts rotated clockwise, RTL anticlockwise.) see page 346 of The Blackwell
Encyclopedia of Writing Systems, Florian Coulmas, Blackwell Publishers Ltd,
ISBN 0 631 21481 X. This shows a 1933 monolingual dictionary of Cyrillic and
Mongolian equivalents. If this is rotated anti-clockwise the Cyrillic does
read left to right but the lines go bottom to top of the page. It's not just
the Mongolian that reverses its normal order.

The original Uyghur script was RTL, but later the page turned anticlockwise.
Mongolian retained this orientation and the order of the vertical lines on
the page being left to right, unlike the neighbouring Chinese.

   Tim

-- 
Tim Partridge. 
Any opinions expressed are mine only and not those of my employer



This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:21:00 EDT