These comments are in reference to Section I.4, “Series that exist in more
than one language expression,” and to the response from SCS included there.
With all respect for the work of the Task Force, I share the reservations
expressed by SCS.

My specific concern is with bilingual editions of the works of a particular
author. Here it might be more accurate to talk about “Series that _contain_
more than one language expression”  -- “Series that exist in more than one
language expression” can be understood to refer to translations of series,
which is not necessarily the same thing.

I’m talking about such things as an edition of the works of Saint Augustine
in English and Latin. As I understand it, such a works edition falls under
the RDA definition of a series, because there is a title common to all the
parts, as well as individual titles for each part. Further, it constitutes
a work in its own right, while each of the individual parts is also a work
on its own.

The presence of original text and translation in a particular resource is a
criterion that often helps our users select the resource they need.

Under AACR2, we had many records along these lines (using a made-up

100 0_ Augustine, Saint, Bishop of Hippo.

245 10 City of God

490 1_ Works of Saint Augustine ; v. 1

800 0_ Augustine, Saint, Bishop of Hippo. Works. English & Latin. 1951 ; v.

I realize that the use of two languages in the authorized access point for
the series is counter to the spirit of RDA. But if we change our practice,
what do we have to offer to our users that is as useful as the heading
above? It still seems to me that this is an efficient way to signal to the
user that “here is the edition that meets your needs, that contains the two
languages you are looking for.”

I have been wondering about possible solutions to the problem the Task
Force has been grappling with. Here is a thought:  In such situations, add
a qualifier to the heading such as: (Bilingual version, English and Latin).

Could this be justified as an instance of “Other Distinguishing
Characteristic of the Work” under RDA 6.6? Placed in such a qualifier, this
information might fit within the parameters of RDA, while retaining the
utility of our past practice.

Following the argument presented by the Task Force (and leaving aside the
question of replacing the “1951” with some other qualifier), it appears
that the recommended practice here would be to have two 800 access points
in the bib record:

800 0_ Augustine, Saint, Bishop of Hippo. Works. English. 1951 ; v. 1.

800 0_ Augustine, Saint, Bishop of Hippo. Works. Latin. 1951 ; v. 1.

This puts us in the position of having two access points that both point to
the same entity, namely the work entitled “Works of Saint Augustine” – so
we no longer have a one-to-one correspondence between entity and access

It also means that users browsing a list of titles under the author name
Augustine won’t be able to tell which of the works editions listed there
are bilingual, and which are monolingual. Instead they would see something
like: (I’ll fall back on dates for differentiation, for the sake of the

Works. 1948 ; v. 1

Works. 1948 ; v. 2 (for some other edition containing only the Latin text)


Works. English. 1951 ; v. 1 (for our bilingual English-Latin edition)

Works. English. 1951 ; v. 2

Works. French. 1960 ; v. 1 (for some other edition containing French
translation only – or it might be a bilingual French-Latin edition; we
wouldn’t be able to tell)

Works. Latin. 1951 ; v. 1 (for our bilingual English-Latin edition)

Works. Latin. 1951 ; v. 2

We need an RDA practice that is as efficient as the older practice in
meeting the needs of users.
Charles Croissant
Senior Catalog Librarian
Pius XII Memorial Library
Saint Louis University
St. Louis, MO 63108