[DCRM-L] Reconsidering digraphs

[James Larrabee]

I have some difficulty in making the content/form distinction. However, 
I am inclined to think
(certainly I cannot prove) that ae/oe are different in kind from such 
ligatures as st, ct, etc., and
 that this may explain their survival in some languages or contexts. In 
(later?) medieval Latin orthography,
ae and oe were typically spelled "e". The Renaissance humanists restored 
the original orthography
(ae/oe)  in their general campaign to return to what they viewed as the 
classical style of Golden Age
Latinity. On the other hand, perhaps because it was felt to be too sharp 
a break from the one-letter
spelling, or possibly confusing to readers at the time who might think a 
different sound was intended, or
for some other such reason, rather than writing the letters separately 
they were coalesced into
a digraph. Possibly this explains why these diphthongs were singled out 
for special treatment while
others went their merry way. Or perhaps it was a Renaissance theory as 
to how these diphthongs
are pronounced.

It would also explain another difference. The "st, ct. etc." digraphs 
can easily be seen as deriving
from simple connecting strokes in handwriting, passing from one letter 
to the next. The letters aren't
really fused together, just connected. The ae/oe digraphs can't, I 
think, be explained this way, nor
do they look like something arising from handwriting styles (though of 
course they were used in the
humanistic scripts). The "st" when the s is long is just a kern with the 
little gap closed (as with an "ft"--
would anyone really even bother thinking about these as digraphs? Even 
though they are, of course,
one piece of type). Maybe the idea of doing the same when a short s 
precedes a t derived from that
by analogy, then extended to ct. Why? Who knows? Just decoration at this 
point, one would think.

Does anyone with manuscript expertise know if the ae/oe digraphs were 
used in earlier manuscripts?

Printers using the "littera moderna" (black letter) seem not to have 
adopted this "new" orthography, so
had no need of the ae/oe digraphs. Did the usage change at all toward 
the end of the black letter era?
At any rate, they became a permanent feature of the new style or 
"littera antiqua" though perhaps not
adopted everywhere and at all times. What shows, I think, the 
fundamental difference between these and
the "st, ct, etc." ligatures is that when they are used at all in a 
given text or maybe even by a given printer,
they are used consistently. The "st, ct, etc." are used at random (when 
used at all). That seems to indicate
cosmetics rather than orthography; something like swash letters, for 
example.  Also there seems no need
for them in upper case, while AE/OE digraphs are used in upper case as 
in lower--or a sort of "virtual"
ligature formed by moving the two letters noticeably closer together. If 
these facts tend to show that
that the ae/oe usage was a matter of orthography, as I think they do, 
then presumably that will put it under
"content" rather than "form" for our purposes.

The ae/oe digraphs, in my experience, are extremely common in earlier 
(pre-19th cent.) printing, perhaps
well-nigh universal (in Latin, at any rate), if not right up to the 20th 
century. Modern languages mostly don't
have these particular diphthongs. It seems natural for those that do 
(the examples noted) to use the ligatures,
even though it may be partly a matter of historical accident. Printers 
aren't philologists. (National practices
might well have been dictated at a higher level, of course.)

In view of this, I'm inclined to consider the ae/oe ligatures as content 
or orthography, so they should be transcribed
as such. (Likewise, I believe, with "virtual" ligatures in upper case, 
or does that get us into a different discussion?)
Another criterion in the rule revisions, not, I think, yet touched on in 
the present discussion, is that no change
in the existing rules should be made without clear necessity or at least 
a solid reason. In this case, the advantages
of the change seem to outweigh whatever disadvantages may exist. The 
only one appears to be the limitations
of indexing in the electronic context. I don't know how this affects 
others. RLIN has always indexed these digraphs
as if separated, so the difference was invisible from the searching 
point of view. And Juliet has noted that we
agree, or ought to, that the rules should ignore such considerations 
(limitations ought to be addressed where they
exist).

We do admit the limitations of dealing with the shrunken character set 
of modern (post-"handpress era"), however.
Perhaps necessarily, in case of the long s, for instance. But, as Brian 
mentioned, we do have ae/oe ligatures available
and by the same token they are familiar to modern readers, so why not 
use them? And as Bob also pointed out, we also
have upper case letters available, so why not follow the source? 
Definitely another discussion ...

Having gone on about the desirability of changing the rule, I have to 
say that after reading the messages, and going over the
rules again, I am still in the dark as to why the existing rule was 
written as it was. If limitations of machine indexing were
a factor, why the exceptions? If French, Anglo-Saxon, etc., are allowed 
their digraphs, why not Latin, etc.? As a former
member of Brian's group at the DCRB revision conference, I recall that 
the rather brief discussion of this point led to no
further enlightenment. I notice that Appendix B, in addressing 
ligatures, mentions only "ct st etc." along with scharfes-s (really
just a long-s/short-s ligature, at least originally). Was there a 
subliminal message that ae/oe somehow fall into the same
category, though never stated? Were the compilers led astray by their 
own typography? The ct and st examples in App. B
are somehow shoved together in the same manner as ae/oe ligatures. (How 
did they do that?) Real ones don't look like that at all
(presumably they weren't available to the printers of DCRB). A form of 
cheating by typography? (At this point I need one of those
little smily faces.) Or perhaps the rule was a reflection of the shift 
in printing (or rather orthographical?) style of Latin texts, at least
in English-speaking countries (presumably in pursuit of further 
"authenticity") dating perhaps to the later 19th century? And of
course the digraphs went right on being used in English (by some 
printers at least), for Latin-derived words, right
through the same period. If anyone remembers the actual reasons in the 
minds of the original rule compilers, it would be nice
to hear from them.

James Larrabee
Robbins Collection Cataloger
Law Library (Boalt Hall)
University of California, Berkeley


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listserver.lib.byu.edu/pipermail/dcrm-l/attachments/20050223/1d5ec798/attachment.htm