Friends,
Thank you for your mail and links.
The names and forms of cantillation marks are, and
have always been, subject to heated debate, which is
why a statistical database would be useful.
A great amount of care has been given to such issues
as pasek/legarmeh, which does not seem to be a big
problem. The meteg/silluk problem has not yet been
satisfactorily resolved, but it concerns only half a
dozen verses in which silluk may be in median verse
position. Otherwise verse-final meteg is silluk.
Musical configuration involves other problems, for
example revia mugrash is considered two signs, geresh
and revia, but we have dealt with this problem.
We can extract cantillation data in Unicode, with only
a few fairly predictable left-to-right problems. This
is all carefully documented along with our best
solutions.
Since we make no claim to know what cantillation marks
are or might mean, we make as few assumptions about
them as possible. We take into account their form and
sequential order, but not much else.
Two hypotheses are supported, syntactic and musical,
which are wildly different, both in the interpretation
of the signs themselves (their number and identity)
and of their hierarchy, rabbinical
conjunctive/disjunctive or superlinear/sublinear.
The goal is simply to create a database of all
cantillation marks, with an application to perform
statistical analysis and pattern recognition, and then
to allow the user to adopt an informed opinion based
on the data, rather than on tradition or anecdotal
evidence.
Once we get a lead programmer interested, we will
prepare a mock-up. It might be a good project for a
graduate student.
Best Regards.
Gabriel
Archive powered by MHonArc 2.6.24.