[DCRB-L] [WG2] Contractions & expansions

[James Larrabee]

--Boundary_(ID_+7BD37YbtsIzjtehEY+GFg)
Content-type: text/plain; charset=iso-8859-1
Content-transfer-encoding: 8BIT

On considering our present treatment of early ms.-type contractions, I
have to wonder exactly what is the rationale behind some of it.

If we don't transcribe "don't" as "don'[o]t" or "do[ ]n'[o]t" or maybe
"don[o]'t", why do we transcribe "q" with a circumflex/tilde over it as
"q[uam]" (including the tilde)? If we don't transcribe "Me." or "Me" or
even "M." as "M[aitr]e." (a French lawyer's title, quite common on title
pages) why do we transcribe "dñs" as "d[omi]ñ[u]s"? I think anyone
dealing with earlier materials, from the incunable period until well
into the 16th century, whether as a "user" or as a cataloger, must find
this treatment 1) extremely cumbersome to construct; 2) unreadable; 3)
ugly; and 4) possibly not doing a good job in representing the source in
the way it represents itself. In addition, it may be impossible for a
cataloger insufficiently familiar with the language. In many cases, a
table is not going to do the job.

The purpose of transcription is representation (according to our
conventional, well-motivated rules), not translation. For the purposes
of transcription, why should so many interpolations, or even any, be
necessary (not even just an option, or something to throw into a 246)?
In many early-period transcriptions, the interpolations outweigh the
text or come close.

I would submit that this situation arose as a result of approaches to
quasi-facsimile transcription by catalogers. In typewriting days,
catalogers could and sometimes did write in the squiggles and strokes to
approximate the actual appearance of the original, which seems pretty
close to quasi-facsimile transcription using more sophisticated
methods--custom-produced types. So it was felt that the contraction or
abbreviation signs had to be represented. When they could not,
interpolations were supplied not for the purpose of explanation or
translation for the unskilled user's sake, but simply to indicate that
something was there that could not be transcribed. In practice, there
could undoubtedly have been some unconscious confusion or conflation
with bracketed corrections or items of the "[i.e. ...]" sort. Fine
distinctions between transcription and translation did not necessarily
occupy cataloger's minds. But I suppose they must occupy ours.

Do we have a sound reason, in non-facsimile transcription, to reproduce
marks of contraction which mostly do not exist in modern typography?
They are not part of orthography but are more analogous, it seems to me,
to punctuation. They are not a fixed part of the spelling of a word,
like modern accent marks, but were applied as and when the typographer
thought fit, within their generally recognized but quite fluid usages.
Would it be appropriate to ignore them in the 245 when it seems
necessary, to include them when it seems possible, even easy to do so
(the tilde/circumflex is the one obvious example that comes to mind,
since it exists), and in either case to reserve amplifications or
translations (sans "[]") for 240/246 &c. fields, which seems their
proper place?

If not, why not?

It's true that we transcribe the apostrophe in contractions (as
elsewhere), but that is because it exists and it is easy and standard to
do so at present. Our transcription, representing as it does a sort of
conventional but reasonable halfway house between a facsimile or
quasi-facsimile transcription on the one hand, and a full modernization
on the other (but certainly leaning strongly to an accommodation with
modern style in many ways, capitalization, punctuation, roman only,
etc.), ought to accommodate this.

Possibly another way of putting it would be to say that, given there is
no very precisely determinable line between quasi-facsimile
transcription and our simplified but historically informed
transcription, it is to some extent arbitrary where we draw it (within
the parameters so well presented in the position paper). Thus, the
practical advantages of the system I am proposing ought to be sufficient
to justify abandoning the universal expansion approach. In a word, if
there is no clear and compelling reason to do something (however
arbitrary) about these contraction conventions, standardizing to a more
modern usage should be preferred. Or one might say that when, on the one
hand, the textual appearance is so far from present typographic
possibilities, and on the other hand the attempt to represent this text
ends up so far from the actual appearance anyway, we are not succeeding
either in quasi-facsimile transcription or in simplified transcription.

In short, I would suggest ignoring, for transcription purposes, such
signs of contraction as cannot be reproduced, say, on our current
keyboards, reproducing those that can be (the tilde-circumflex is the
only one that comes to mind), and reserving such expansions as seem
necessary or appropriate for other title fields.

The preservation of the circumflex has the additional historical
justification in the fact that it survived the ms. era much longer than
any of the other signs; past the end of the 16th century, certainly. One
must deal with it fifty times for every instance or two of the others.

A point I might add: as a "user" along with being a cataloger (and
obviously, we are major catalog users in our very activity as
catalogers), I need to work from printouts of my rough draft when
consulting printed catalogs some of which are half way across the campus
at the Bancroft, and I cannot trudge through a rainy East Bay morning
with a 1478 imprint tucked under my arm (nor at any other time). I find
it important to make a close comparison between my transcriptions, not
only of the title page, but of colophons and other areas of the book
(not all of which I may leave in the final record), and those in the
incunable catalogs. It's a lot easier to do so if I have typed just
what's in the source, with the tildes at any rate in place; and ease
will also make for accuracy. And I think the same must also be true at
least to some extent for many users of the catalog, otherwise why do we
bother with transcription at all? Goff, for example, doesn't.

James Larrabee
Robbins Collection Cataloger
Law Library
University of California, Berkeley
Berkeley, CA 94720
(510) 642-1114
larrabee@law.berkeley.edu



--Boundary_(ID_+7BD37YbtsIzjtehEY+GFg)
Content-type: text/html; charset=us-ascii
Content-transfer-encoding: 7BIT

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
On considering our present treatment of early ms.-type contractions, I
have to wonder exactly what is the rationale behind some of it.
<p>If we don't transcribe "don't" as "don'[o]t" or "do[ ]n'[o]t" or maybe
"don[o]'t", why do we transcribe "q" with a circumflex/tilde over it as
"q[uam]" (including the tilde)? If we don't transcribe "Me." or "Me" or
even "M." as "M[aitr]e." (a French lawyer's title, quite common on title
pages) why do we transcribe "d&ntilde;s" as "d[omi]&ntilde;[u]s"? I think
anyone dealing with earlier materials, from the incunable period until
well into the 16th century, whether as a "user" or as a cataloger, must
find this treatment 1) extremely cumbersome to construct; 2) unreadable;
3) ugly; and 4) possibly not doing a good job in representing the source
in the way it represents itself. In addition, it may be impossible for
a cataloger insufficiently familiar with the language. In many cases, a
table is not going to do the job.
<p>The purpose of transcription is representation (according to our conventional,
well-motivated rules), not translation. For the purposes of transcription,
why should so many interpolations, or even any, be necessary (not even
just an option, or something to throw into a 246)? In many early-period
transcriptions, the interpolations outweigh the text or come close.
<p>I would submit that this situation arose as a result of approaches to
quasi-facsimile transcription by catalogers. In typewriting days, catalogers
could and sometimes did write in the squiggles and strokes to approximate
the actual appearance of the original, which seems pretty close to quasi-facsimile
transcription using more sophisticated methods--custom-produced types.
So it was felt that the contraction or abbreviation signs had to be represented.
When they could not, interpolations were supplied not for the purpose of
explanation or translation for the unskilled user's sake, but simply to
indicate that <i>something was there that could not be transcribed. </i>In
practice, there could undoubtedly have been some unconscious confusion
or conflation with bracketed corrections or items of the "[i.e. ...]" sort.
Fine distinctions between transcription and translation did not necessarily
occupy cataloger's minds. But I suppose they must occupy ours.
<p>Do we have a sound reason, in non-facsimile transcription, to reproduce
marks of contraction which mostly do not exist in modern typography? They
are not part of orthography but are more analogous, it seems to me, to
punctuation. They are not a fixed part of the spelling of a word, like
modern accent marks, but were applied as and when the typographer thought
fit, within their generally recognized but quite fluid usages. Would it
be appropriate to ignore them in the 245 when it seems necessary, to include
them when it seems possible, even easy to do so (the tilde/circumflex is
the one obvious example that comes to mind, since it exists), and in either
case to reserve amplifications or translations (sans "[]") for 240/246
&amp;c. fields, which seems
<i>their</i> proper place?
<p>If not, why not?
<p>It's true that we transcribe the apostrophe in contractions (as elsewhere),
but that is because it exists and it is easy and standard to do so at present.
Our transcription, representing as it does a sort of conventional but reasonable
halfway house between a facsimile or quasi-facsimile transcription on the
one hand, and a full modernization on the other (but certainly leaning
strongly to an accommodation with modern style in many ways, capitalization,
punctuation, roman only, etc.), ought to accommodate this.
<p>Possibly another way of putting it would be to say that, given there
is no very precisely determinable line between quasi-facsimile transcription
and our simplified but historically informed transcription, it is to some
extent arbitrary where we draw it (within the parameters so well presented
in the position paper). Thus, the practical advantages of the system I
am proposing ought to be sufficient to justify abandoning the universal
expansion approach. In a word, if there is no clear and compelling reason
to do <i>something</i> (however arbitrary) about these contraction conventions,
standardizing to a more modern usage should be preferred. Or one might
say that when, on the one hand, the textual appearance is so far from present
typographic possibilities, and on the other hand the attempt to represent
this text ends up so far from the actual appearance anyway, we are not
succeeding either in quasi-facsimile transcription or in simplified transcription.
<p>In short, I would suggest ignoring, for transcription purposes, such
signs of contraction as cannot be reproduced, say, on our current keyboards,
reproducing those that can be (the tilde-circumflex is the only one that
comes to mind), and reserving such expansions as seem necessary or appropriate
for other title fields.
<p>The preservation of the circumflex has the additional historical justification
in the fact that it survived the ms. era much longer than any of the other
signs; past the end of the 16th century, certainly. One must deal with
it fifty times for every instance or two of the others.
<p>A point I might add: as a "user" along with being a cataloger (and obviously,
we are major catalog users in our very activity as catalogers), I need
to work from printouts of my rough draft when consulting printed catalogs
some of which are half way across the campus at the Bancroft, and I cannot
trudge through a rainy East Bay morning with a 1478 imprint tucked under
my arm (nor at any other time). I find it important to make a close comparison
between my transcriptions, not only of the title page, but of colophons
and other areas of the book (not all of which I may leave in the final
record), and those in the incunable catalogs. It's a lot easier to do so
if I have typed just what's in the source, with the tildes at any rate
in place; and ease will also make for accuracy. And I think the same must
also be true at least to some extent for many users of the catalog, otherwise
why do we bother with transcription at all? Goff, for example, doesn't.
<p>James Larrabee
<br>Robbins Collection Cataloger
<br>Law Library
<br>University of California, Berkeley
<br>Berkeley, CA 94720
<br>(510) 642-1114
<br>larrabee@law.berkeley.edu
<br>&nbsp;
<br>&nbsp;</html>

--Boundary_(ID_+7BD37YbtsIzjtehEY+GFg)--