Re: brevigraphs: abbreviation or typography?
M. J. Driscoll <mjd <at> HUM.KU.DK>
2005-12-02 13:52:14 GMT
Dear all,
For years I've been collecting materials (and boring people to death) on the
subject of abbreviations and their expansions, and in particular how best to
mark them up. So anyone with any sense should stop reading now.
The short answer to the question is yes, brevigraphs are abbreviations, one of
four basic types, seen from the point of view of the means through which the
abbreviation is achieved, the other three being: suspension, where only the
first letter or letters of a word are written, generally followed (and
frequently also preceded) by a point or with a superscript stroke; contraction,
where the first and last letters are written, normally also with a superscript
stroke, or, less commonly, a point or points; and superscript letters, where a
superscript vowel will normally represent that vowel preceded by r or v, a
superscript consonant that consonant preceded by a.
The most common sign of abbreviation is the superscript stroke or bar, which
can indicate the suppression of one or more nasal consonants (m or n), and is
also used as a more general mark of abbreviation in suspensions and
contractions. Although in appearance there is no discernible difference between
the two signs, in terms of their use they are quite distinct. From the point of
view of their function, therefore, abbreviation signs may be said to fall into
two categories, those which indicate that something has been omitted, without
suggesting what that something may be, and those which always refer to a
particular combination of graphemes, regardless of the lexical item in which
they occur. There is obviously a correlation between the two systems:
suspensions and contractions by necessity make use of a general mark of
abbreviation, while superscript letters and tittles (some of which actually
derive from letters) have a specific graphemic reference. The brevigraphs are
of both types, since some, such as the inverted c representing con, have a
specific graphemic reference, while others, for example the nomina sacra, have
more in common with suspensions and contractions.
It is customary in scholarly editions to expand abbreviations, that is to
supply the letters which have been omitted; the letters so supplied are often
typographical distinct from the others in order to indicate exactly what is in
the original and what has been supplied. The TEI Guidelines provide a means of
encoding abbreviations and their expansions through the use of the <abbr> and
<expan> tags. It is up to the encoder to decide whether the abbreviation or its
expansion is to be the base form; in P4 the other form could be given as an
attribute value, while in P5 both can be provided, wrapped within <choice>
tags.
The problem is that the two elements, <abbr> and <expan>, do not really mirror
each other, at least not completely (there are historical reasons for this
which I won't go into here). Underlying this problem is the fundamental
question of what, exactly, "the abbreviation" is. Is it the mark, sign or
letter that indicates that something has been suppressed, or is it the entire
word? And similarly, is "the expansion" only the letters which have been
suppressed and have therefore been supplied by the transcriber, or is it,
again, the whole word? Given the distinction between abbreviations with a
lexical reference (suspensions, contractions and some of the brevigraphs) and
those with a graphemic reference (superscript letters and signs and the
remainder of the brevigraphs), the most reasonable answer would appear to be
that it is both. It strikes one as counter-intuitive to treat abbreviations
with a lexical reference on anything other than the whole-word level. In
English and other languages "p." is a common abbreviation for the word "page"
(or the equivalent), but the dot can in no way be said to "stand for" the
letters "age": rather it is the first letter, followed by a point, which
represents the whole word. Moreover, this initial letter can be doubled to
indicate the plural, e.g. "pp.", for "pages". It seems nothing short of
perverse to maintain that in such a case, the first p somehow "really is
there", whereas the second and attendant dot are not, but are rather an
abbreviation representing the suppressed "ages". At the same time, however, the
superscript 9-shaped sign does only stand for "us", regardless of the lexical
item in which it occurs, and it would be somewhat forced to treat it any other
way, especially in cases where there is more than one abbreviation within a
single word; treating such abbreviations on a whole-word basis would blur the
connection between the abbreviation sign and its expansion. One obvious
solution would be to distinguish between the two types of abbreviations (and
expansions), using a type attribute on <abbr> (and <expan>), with the values,
say, "lex" or "graph". Another would be to tag abbreviations and expansions
solely on a whole-word basis and use some other means, the <supplied> tag for
example, to indicate which letters have been supplied. A third possibility
would be to rethink the entire system, which is what I have been trying to do
for the last several years, thus far without any great success.
Should anyone still be reading, and be interested in reading more, I've written
an article on abbreviation practices in some modern European languages,
available at: http://www.staff.hum.ku.dk/mjd/thoughts.html
all the best,
Matthew
M. J. Driscoll
Arnamagnaean Institute
Copenhagen University
RSS Feed