Re: Synonyms in Unicode Terminology

Date: Wed Jan 27 1999 - 17:36:12 EST

       Script and alphabet are definitely distinct terms. Rick has
       distinguished them thus:

       RM>- script / alphabet

       A script is a collection of written symbols viewed as a unit.
       An alphabet is the complete collection of such symbols used for
       a particular purpose, such as a single language. Generally,
       "alphabet" implies that it is a collection of units
       representing phonological units, rather than syllabic units.
       The term "syllabary" is preferred for the latter. "Alphabet"
       is often used in normal discourse to mean "script", but
       technically speaking in the standard, they are different. The
       "Latin Alphabet" technically refers ONLY to the subset of the
       "Latin Script" which is used to write the "Latin Language".

       I'm not entirely happy with this, though this may be the case
       in part to not being sure exactly what he means at certain
       points. (More on this below.)

       I think it's important to distinguish script from writing
       system. I have come across other sources that also make this
       distinction and which give definitions that are pretty similar
       to the definitions I've arrived at. Here are the definitions of
       script and writing system that I currently work with:

       Script: A maximal collection of characters used for writing
       languages or for transcribing linguistic data that have graphic
       representations that share common characteristics of
       appearance, that share a common set of typical behaviours, that
       have a common history of development, and that would be
       identified as being related by some community of users.

       By this definition, Latin is a script, Thai is a script, etc.

       Writing system: An implementation of a script for a particular
       purpose, e.g. for writing a particular language. A writing
       system uses some subset of the characters of the script on
       which it is based with most or all of the behaviours typical to
       that script and possibly certain behaviours that are peculiar
       to that particular writing system.

       By this definition, the English alphabet is a writing system,
       and the French alphabet is a different writing system; Standard
       Lao writing is a writing system, and Tai Dam written with Lao
       script is a different writing system; etc.

       In my current thinking, writing system and orthography are
       distinct. Orthographies are writing systems that are used for
       writing languages for purposes of broad communication, as
       opposed to technographies, for example, which are used to
       transcribe linguistic data for research purposes. John
       Mountford ("A Functional Classification", in Daniels & Bright,
       1996, pp. 627 - 32) gives a 5-way typology of writing systems,
       of which I have mentioned 2 types.

       Mountfords typology is a functional typology of writing
       systems. There are also structural typologies of scripts that
       have been proposed. Peter Daniels ("The Study of Writing
       Systems", in Daniels & Bright, 1996, pp. 3 - 17) provides an
       interesting (though not undisputed) 6-way typology of scripts:

       - logosyllabary (e.g. Chinese)
       - abjad (e.g. Arabic)
       - syllabary (e.g. Katakana)
       - abugida (e.g. Ethiopic)
       - alphabet (e.g. Latin)
       - featural system (e.g. Hangul)

       (These terms can also be applied to writing systems.) An
       alphabet, then, is just one structural type of script.

       I the explanation Rick gave, the term "alphabet" is used in a
       way that could be confused with "writing system", though he is
       correct is saying that the characters in an alphabet represent
       phonological units and that the characters in a syllabary
       represent syllabic units. Admitedly, "alphabet" is commonly
       used in a way that equates it with "writing system", as when
       people refer to the "Latin alphabet". I'd suggest, however,
       that for our purposes we need to adopt more careful usage of
       terminology, as would commonly be done in any research
       endeavour. Thus my reservations about Rick's statement.

       Peter Constable
       Non-Roman Script Intiative, SIL

This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:44 EDT