Home About Subscribe Search Member Area

Humanist Discussion Group

< Back to Volume 33

Humanist Archives: Dec. 5, 2019, 6:12 a.m. Humanist 33.458 - failure of another sort

                  Humanist Discussion Group, Vol. 33, No. 458.
            Department of Digital Humanities, King's College London
                   Hosted by King's Digital Lab
                Submit to: humanist@dhhumanist.org

        Date: 2019-12-05 00:33:49+00:00
        From: lachance@chass.utoronto.ca
        Subject: Re: [Humanist] 33.455: failure of another sort

One further note on intentions. There is none expressed explicitly in the
text. But the author is quite clear in an interview.

A scan the WWW and one finds an interview. Note the characterization of
the passage as an "un-translation":


Joe Milazzo :

On page 67, the reader encounters a poem in Arabic script. If this poem
itself is translated elsewhere within The Consequences of My Body, it is
not marked or identified in any way. Likewise, the reader is not privy to
whether these Arabic lines constitute an original composition, or whether
they are the work one of on those predecessor poets introduced later in
the book. What hinges upon the un-translation of the poem on page 67?

Maged Zaher:

So much hinges on this un-translation and its typeset — so much — I will
leave it at that


Apologies for not quoting this in my earlier posting. (My failure of
another sort). I was caught up in the theme of resisting OCR.

I do think that the ubiquity of machine translation raises some
interesting issues for pedagogy and scholarly communication: where, when
and how to translate (or to respect the phenomenon of "making strange").


>                   Humanist Discussion Group, Vol. 33, No. 447.
>             Department of Digital Humanities, King's College London
>                    Hosted by King's Digital Lab
>                        www.dhhumanist.org
>                 Submit to: humanist@dhhumanist.org
>         Date: 2019-11-29 14:29:34+00:00
>         From: Francois Lachance 
>         Subject: Failure of Another Sort
> Willard
> I have come across an interesting and quirky case...
> Maged Zaher _the consequences of my body_
> Failure of another sort - engineered failure for the monolingual English
> reader - you turn the page in Zaher's book and on the verso is this
> section of what looks like Arabic. No translation, no annotation.
> I  turned to an Arab-speaking friend who informs me:
> [quote]
> Typically, one has to understand what one is reading in Arabic and do so
> with a good grasp of the grammar in order to read it correctly. This is
> one reason why the language is classified as one of the most difficult in
> the world.
> Your poet is intentionally trying to make things even more difficult. The
> Arabic script is in mirror image. And, to complicate things further, it is
> all in upper case letters without any spaces separating the words.
> I am assuming the intention here is to have the text resist optical
> recognition software and hence make it accessible only to a living
> breathing Arabic reading human.
> [/quote]
> In this case the machine is to aid communication, connecting with the
> human, by resisting the machine.
> --
> Francois Lachance
> Scholar-at-large
> http://www.chass.utoronto.ca/~lachance
> https://berneval.hcommons.org
> Attachments:
> zaher-consequences.jpg: https://dhhumanist.org/att/81835/att00/
> --[2]------------------------------------------------------------------------
>         Date: 2019-12-02 14:45:16+00:00
>         From: Alec McAllister 
>         Subject: Re: [Humanist] 33.447: failure of another sort: conceal
> to reveal
> I can think of another explanation: plain old software/human error.
> Arabic is written right-to-left, but not all software "knows" this, and
> even
> some software that can handle RtoL text needs to be explicitly "told"
> which bits
> are RtoL: if the options are wrongly set, the software displays Arabic as
> if it
> were plain vanilla left-to-right text.
> A further complication is that Arabic is written cursively: letters join
> together, as in handwriting. Arabic script has no concept of upper and
> lower
> case as such, but each letter can have different forms (typically up to
> four
> different forms, plus certain ligatures, some of them compulsory),
> depending on
> whether the letter is linked to the letter on the right, on the left, to
> both or
> to neither. The software needs to know about this system too, so that it
> can
> substitute the correct forms, taking account of the context.
> These days, much this information is already encoded within the font
> itself, and
> the rest is handled pretty routinely by the operating system. From the
> user's
> point of view, s/he starts typing in Arabic and it all just happens.
> However, older software, especially fonts, may date from the bad old days
> when
> RtoL scripts needed separate RtoL operating systems. Such software lacks
> the
> features that tell modern, bi-directional operating systems how to handle
> Arabic
> letters.
> In extreme cases, Arabic text comes out running from left to right, and
> with
> every letter in its independent (stand-alone) form. In that form, it is
> extremely difficult to decipher, even by fluent Arabic readers.
> This is especially likely to happen *after* the text has been proof-read
> and
> accepted, and is then handed over to a non-Arabic-speaking graphic-design
> person
> who substitutes a different font because it "looks nicer".
> A similar case occurred in the run-up to the London Olympics in 2012:
> http://www.bbc.co.uk/news/uk-england-london-18911599
> le.jpg] Olympics train poster 'gibberish'
> (http://www.bbc.co.uk/news/uk-england-london-18911599)
> A train company is criticised for producing an Olympics security poster
> which
> reads as "gibberish" in Arabic.
> www.bbc.co.uk
> To err is human, but to make a real mess, use a computer.
> Alec McAllister
> Multilingual Computing Co-ordinator (retd)
> University of Leeds

Unsubscribe at: http://dhhumanist.org/Restricted
List posts to: humanist@dhhumanist.org
List info and archives at at: http://dhhumanist.org
Listmember interface at: http://dhhumanist.org/Restricted/
Subscribe at: http://dhhumanist.org/membership_form.php

Editor: Willard McCarty (King's College London, U.K.; Western Sydney University, Australia)
Software designer: Malgosia Askanas (Mind-Crafts)

This site is maintained under a service level agreement by King's Digital Lab.