I find it confusing as well, but as Karen points out, that's the way the
FRBR model does things. It seems to be driven by the need for the work to be
such an abstract thing that it is prior to words. However, it does seem to
me that the meaning of the language of a particular expression is not
complete without reference to the original language.
One of the FRAD drafts
actually did propose original language as an attribute of the work ("The
language in which the work was first expressed"), but that axed so it seems
to have been a very conscious decision on the part of the creators of FRBR.
The idea does seem to have generated some controversy. From ALA's feedback
on this draft:
" At least one task force member was a bit uneasy with this attribute,
noting that, although the attribute has a certain utility, the "work" entity
is abstract in FRBR and is not associated with any particular language (e.g.
"Ancient Greek" is the language of the first expression of the Iliad, but
not the language of the work, which encompasses what all of the expressions
have in common). Others thought that an original language attribute was
appropriate for "work" (for textual works, anyway), that all expressions of
a work do have the same "original language" even if the language of the
expressions themselves can differ, and that the attribute is necessary for
determining whether or not the expression represents a translation. It was
suggested that the attribute would not be appropriate for a "superwork"
entity, were one to be defined."
From: Beacom, Matthew <[log in to unmask]>
Thank you, Karen,
It has been awhile since I refreshed my memory with actually reading FRBR.
Language is an attribute of the FRBR expression and not the FRBR work
entity. I must still have a dominate pre-FRBR concept of work in my mind! I
need another 5 years in the re-education camp.
From: Code for Libraries [mailto:[log in to unmask]] On Behalf Of
Sent: Monday, December 13, 2010 10:51 AM
To: [log in to unmask]
Subject: Re: [CODE4LIB] Announcing OLAC's prototype FRBR-inspired moving
image discovery interface
Quoting "Beacom, Matthew" <[log in to unmask]>:
Sometimes I feel like we should all have the FRBR diagram tattoo'd on our
arms so we can consult it any time anywhere. :-)
> With as complex a thing as a film--so many "authors", images, music,
> dialog, acting, sets, costume, etc., etc., etc., applying the FRBR
> model is tough, and your implementation is quite sensible. However, I
> had a small question about one thing you said about FRBR not allowing
> language at the work level. That doesn't seem right to me.
> How could the language of a thing that is primarily or even partially
> a work made of language--like a novel or a motion picture with spoken
> dialogue would not necessarily be considered at the work level and not
> at some other level.
Matthew, I can't answer how it is possible but I can tell you that it is a
fact: language is an attribute of Expression, not of Work. That's kind of
the key meaning of frbr:Expression -- it is the Expression of the Work, and
the Work doesn't exist until Expressed. So Work is a very abstract concept
in FRBR. (Which is why more than one attempted implementation of FRBR that I
have seen combines Work and Expression attributes in some way.)
Not only that, but Kelley's model uses something that I consider to be
missing from FRBR: the concept of a "original Expression." For FRBR (and
thus for RDA) all expressions are in a sense equal; there is no privileged
first or original expression. Yet there is evidence that this is a useful
concept in the minds of users. Some recent user studies  around FRBR
showed that this is a concept that users come up with spontaneously. Also, I
can't think of any field of study where knowing what the original expression
of a work was wouldn't be important.
> Because of the way we treat translations--not just in FRBR--as what
> FRBR calls expressions not as new works, a translation from the
> original language to another would be considered an FRBR expression.
> Could you explain this a bit more?
The FRBR relationship "translation of" is an Expression-to-Expression
relationship. (See my personal "cheat sheet" of RDA/FRBR relationships ).
> Thank you.
> -----Original Message-----
>> This also allowed us to get around some of the areas of more orthodox
>> FRBR modeling that we found unhelpful. For example, FRBR doesn't
>> allow language at the Work level, but we think it is important to
>> record the original language of a moving image at the top level.
[log in to unmask] http://kcoyle.net