I’d like to reinforce a point that Adam made earlier (and I implied in my response) about the records being discussed in this thread.


The only category of undifferentiated descriptions allowed in RDA is for persons (see 8.11) and PCC has disallowed even that. RDA does not recognize a “generic” or “undifferentiated” category of expression description, and for good reason. Undifferentiated descriptions cannot work in the entity-relationship database structure that FRBR/RDA is based on (at least in my opinion), and toward which we are moving (at least I hope so). I understand the practical desire for such a category, but it needs to be understood that it is completely extra-RDA (and given that PCC has disallowed the category for persons even though there were very practical reasons for that, too, I’m not sure I understand the urge to create a new category of undifferentiated entity descriptions for expressions).


So, because RDA doesn’t recognize any other category, RDA’s instructions at 6.27.3 for creating an authorized access point for an expression assumes that the authorized access point is for a specific expression. 6.27.3 says to begin with the authorized access point for the work and then to add one or more elements (content type, date, language, or other distinguishing characteristic) as needed to identify the expression. No priority is given for which element to add, but as many may be added as needed to differentiate the expression entity being described from other expression entities related to the same work. So, if (for example) addition of the language alone is sufficient to differentiate the expression described from all other expressions, that’s all that’s needed (though further additions are not forbidden). The fact that you haven’t added anything more than the language does not mean that the description is for a “generic” expression (which, again, RDA doesn’t recognize). That is, you can’t recognize an authority record for a “generic” expression vs. a “specific” expression simply by noticing that only language has been added to the authorized access point.


I created a record for the Spanish expression of J.G. Ballard’s novel Unlimited dream company this morning (no2014131000). There is only one Spanish expression, a translation by Enrique Pezzoni and Marcial Souto first published in 1990. I created the authorized access point by adding “Spanish” after the authorized access point for the work. I did not add anything more because that addition under 6.27.3 was sufficient to distinguish this expression from all other expressions of this work. This description (NAR no2014131000) was not intended to be a “generic” expression record/authorized access point. It very definitely applies to this expression and this only (which happens at the moment to be the only Spanish expression). It does, appropriately, contain things like the 046 field for the date of expression, even though similarly to the records that initiated this discussion, the only thing added to the authorized access point was the language.


This is also my practice when I encounter an existing authority record for an expression when (a) only language has been added to the authorized access point form for the work and (b) I discover that it is the only expression in that language. I record elements in the record pertinent to the expression (e.g., date of expression, language, other characteristic, relationship to translator or editor, etc.). If there is only one expression in the language the question of whether the original record represented a “generic” or “specific” expression is moot—it is “specific” (to say nothing of the earlier point about RDA not recognizing the former category anyway). If on the other hand I discover that there are in fact other expressions in the same language, I will add a new record to the file for the expression(s) I need.


LC has now removed the dates from the records in question, apparently intending to convert them back into “generic” descriptions. LC can of course appropriately revise or remove changes that they themselves made earlier, but given that the records apparently represented the only expressions in the languages in question, I’d maintain that this action didn’t actually convert the record “back into a generic” record. The record already stands for a specific expression, the only one in the language described.


Making authority descriptions as specific as possible by adding RDA elements (and incidentally, for the case of expressions, clarifying that a record represents a specific expression) is not straightening deck chairs, in my opinion. It is absolutely crucial to the future usefulness of the database and the ability to integrate our current records into the future ER environment (which I believe is possible, but only if we prepare the data).




Robert L. Maxwell
Ancient Languages and Special Collections Cataloger
6728 Harold B. Lee Library
Brigham Young University
Provo, UT 84602

"We should set an example for all the world, rather than confine ourselves to the course which has been heretofore pursued"--Eliza R. Snow, 1842.


From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On Behalf Of Frank, Paul
Sent: Thursday, October 02, 2014 1:12 PM
To: [log in to unmask]
Subject: Re: Date of expression in DLC authority records?




The dates in these expression records were added to the 046 field early in 2013, before the May 2013 update to LC-PCC PS 6.27.3.


Mark, Bob, and Adam—your analysis was correct, but we will err on the side of caution and do as Stephen suggests.


So, after discussion with colleagues in the Policy and Standards Division, the dates in the 046 fields of these expression records have been removed, so that the authorized access points now represent “generic” expressions of the English translations.  


There is still a need for the issues and recommendations raised by the PCC Task Group on Expressions to be formalized and incorporated officially into PCC documentation. I will raise this issue with the PCC Policy Committee.  


Paul Frank

Acting Coordinator, NACO and SACO Programs

Cooperative Programs Section

Cooperative and Instructional Programs Division

Library of Congress

101 Independence Ave., SE

Washington, DC 20540-4230


[log in to unmask]





From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On Behalf Of Matthew C. Haugen
Sent: Thursday, October 02, 2014 2:54 PM
[log in to unmask]
Subject: Re: [PCCLIST] Date of expression in DLC authority records?


As the chair of the PCC Access Points for Expressions Task Group, I think Stephen has characterized our recommendations correctly. Though the Policy Committee accepted the report in early 2013, our recommendations were rather preliminary, and nearly two years later, I'm still at a loss for more specific solutions.


At the time, we were wary of defining a "super-expression" category to enshrine collective or generic expression AAPs, but acknowledging that the co-existence of undifferentiated and differentiated work/expression NARs is somewhat inevitable, and not wishing to require or forbid either, we recommended that 046/3xx fields NOT be separately encoded in work/expression NARs that do in fact represent multiple expressions or could potentially do so in the future; more simply put, that separate 046/3xx elements generally not be recorded unless the same data is also an element in the access point. For elements used as part of a work/expression access point, we recommended encoding that information in separate fields as well, following general best practices for use of controlled vocabularies, standardized dates, language codes, etc.


In the case Adam describes regarding a single expression, a date in 046 but not in the access point can reflect a single expression at the time it was created; this would preclude its being used as a generic record, however, in the event that another expression comes along, which may or may not be the intent.


I would assume, that, as is the practice for undifferentiated personal name records, any action taken to differentiate an expression currently represented on a generic record is best done by creating a new record for that expression, rather than reusing an existing expression record by changing its level of differentiation.  So I agree with Stephen's reluctance to add a date to the AAP to resolve ambiguity or conflict; I don't think this was the intent of the task group.






On Thu, Oct 2, 2014 at 9:42 AM, Stephen Hearn <[log in to unmask]> wrote:

Bob's case against using dates in generic (or "generic") expression authorities makes sense, too, and agrees with the recommendation of the PCC Access Points for Expressions Task Group. There's still an ambiguity, though--when the title looks generic and the 046 contains a specific date, which data element determines the AAP's intended scope? Does the AAP look generic? Page 9 of the task group report comments, "Usually, then, these special fields [e.g., 046] would only apply when those same elements are also included as additions to the access point."  So maybe these are cases where the 046 was applied in error and the generic intent of the record is more evident from the AAP.


In practical terms, given that the authorities have been posted for some time and may have been used to control bib AAPs, I'd be reluctant to resolve the ambiguity in the authorities by adding a date (per the task group recommendation) to the AAP, since the unspecified AAP form may have been used generically to control other expressions.  It would be safer to remove the 046, let the existing authority stand as generic, and create a new authority with more thorough specificity if needed.






On Wed, Oct 1, 2014 at 1:56 PM, Robert Maxwell <[log in to unmask]> wrote:

The presence of a recorded date of expression means the record is no longer a “generic” record covering all English language expressions of a given work. The record applies only to the expression that was created in that year (or first published in that year). Different expressions are created in different years, so by recording a specific year you are pinning the record down to a specific expression.


Bizarrely, there is no evidence for the 046 date 1927 in n  85022980, but it does appear from an OCLC search that an English language expression of Rølvaag’s I de dage was indeed published in 1927, so this record would correspond to that particular expression, not another (perhaps there is only one English language expression of this work, but if there are more than one, n  85022980 can only apply to the expression first created in 1927).


Randomly looking at a few of the other examples you give, most of them don’t seem to have any evidence for the 046 field data either.


In my opinion if any expression-related attribute aside from language (377) is recorded in an authority record, that takes the record out of the realm of “generic” language expression and it becomes a record for a specific expression, the one that corresponds to whatever attributes have been recorded (in these cases, date of first creation of the expression). So my response to Mark’s question would be “No, these records should not be treated as describing a generic expression, they describe a specific expression.” But I have no idea what the LC catalogers were thinking when they enhanced these records with 046 fields.




Robert L. Maxwell
Ancient Languages and Special Collections Cataloger
6728 Harold B. Lee Library
Brigham Young University
Provo, UT 84602

"We should set an example for all the world, rather than confine ourselves to the course which has been heretofore pursued"--Eliza R. Snow, 1842.


From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On Behalf Of Mark K. Ehlert
Sent: Wednesday, October 01, 2014 9:59 AM
[log in to unmask]
Subject: Date of expression in DLC authority records?


I've come across several authority records exclusively sourced and edited by DLC (per the 040 field) for English translations that include 046 dates that specify the date of expression.  The heading itself is generic following LC-PCC PS 6.27.3--Name. Title. English--but the record is made specific with that added date.  Many, but not all, of the records were last edited in February-March 2013, which point to the Phase 2 RDA Authorities project.  However, none of the 1xx/4xx headings use the $f, which I presume would be the source for the automated population of the 046 field.  Several 046s also don't match the source dates in the 670s; manual intervention is implied there.


Random examples by LCCN:
n  90636367
no 97019257
n  80148532
n 2012079161
n 2013001274
n 2013037659
n 2013062190

So a question of practice for those of us who aren't LC: Should these records be treated as generic expression?  Or specific expression?  I presume the 046s aren't meant to act as the start date(!) for a collection of English expressions covered by a single record.  I have one translation record in front of me I'd like to edit, but don't know how to go about enhancing it, if at all (n  85022980, last touched in September 2013).


Mark K. Ehlert                 Minitex
Coordinator                    University of Minnesota
Digitization, Cataloging &     15 Andersen Library
  Metadata Education (DCME)    222 21st Avenue South
612-624-0805            Minneapolis, MN 55455-0439

  "Experience is by industry achieved // And perfected by
the swift course of time." -- Shakespeare, "Two Gentlemen

of Verona," Act I, scene iii




Stephen Hearn, Metadata Strategist

Data Management & Access, University Libraries

University of Minnesota

160 Wilson Library

309 19th Avenue South

Minneapolis, MN 55455

ORCID:  0000-0002-3590-1242



Matthew C. Haugen
Rare Book Cataloger
102 Butler Library
Columbia University Libraries
E-mail: [log in to unmask]
Phone: 212-851-2451