Program for Cooperative Cataloging
Standing Committee on Training
Series Policy Task Force
Interim ReportJanuary 17, 2014
The PCC/SCT Series Policy Task Force, consisting of Ed Jones, Robert Maxwell, George Prager, and Paul Priebe,was tasked in late 2012 with reviewing existing series policies and responding to series policy questions in the light of RDA implementation. The group is associated with, but is not a subgroup of, the PCC/SCT Series Training Task Force and the PCC/SCT Series Review Group.
The Series Policy Task Force sees its work as comprising three activities: (1) to make recommendations on new policy issues that arose when RDA was implemented, (2) to make recommendations as to the disposition of the LC-PCC Policy Statements that currently exist and language in DCM-Z1 about series, and (3) to make recommendations for additions and changes to series-related PCC documentation, including the LC-PCC Policy Statements and DCM Z1. The existingLC-PCC Policy Statements were simply transferred directly from the LCRIs and have not been examined with RDA in mind. As such they are the only remaining part of the LC-PCC PSs that need examination and revision.
The Task Force spent most of 2013 discussing and giving recommendations for policy questions that arose as the Series Training Task Force created its training materials, and as the Review Group fielded questions from catalogers attempting to apply RDA to series authority records. It has not yet begun systematically reviewing the LC-PCC PSs, but has decided it needs to send this interim report so that the PCC policy and training committees can think about some of the policy recommendations. The Task Force recommends that decisions be made on these proposed policy statements, but that they not be published until the existing LC-PCC PSs and series information in DCM-Z1 has been examined in order not to introduce inconsistency in the documentation.
These policy recommendations were circulated to the Series Training Task Force in July 2013 and the Policy TF received feedback. There was not agreement in the Training TF or acceptance of all the recommendations, and we have tried to reflect these cases below. The recommendations are those of the Policy Task Force.
The Policy Task Force intends next to begin review of the existing LC-PCC Policy Statements and DCM-Z1 sections relevant to series (see Appendix II).
Organization of this document:
1
I. FRBR issues
II. Transcription Issues
III. Access Point Issues
IV. Miscellaneous Issues
Appendix I. Proposed Policy Statements
Appendix II. Other Series-Related Policy Statement That Still Need Review
1
I. FRBR issues
I.1.Mixing of FRBR EntitiesWithin Series Authority Records
Discussion
Some commentators have argued that series authority records represent a melding of attributes of several FRBR Group 1 entities—works, expressions, manifestations, and even items (in the form of local treatment elements)—and indeed that such melding was unavoidable. We believe that this occurs no more on series authority records than on any other kind of authority record. For example, records for non-series works and expressions routinely contain variant access points for forms that occur only in a subset of manifestations (often only one). To this extent, while a certain “mixing” is probably inevitable, the general goal of authority records that correspond to discrete FRBR Group 1 entities is not only achievable but desirable. The following discussion addresses this question in terms of the affected elements in the MARC 21 Format for Authority Data.
These are the fields that occur in series authority records and are not found in other kinds of authority records:
643. Series place and publisher/issuing body. It has been argued, as evidence that we mix FRBR entities on series authority records, that this is manifestation information. This field is similar to the 670, which also contains manifestation-level information. 670 occurs on all authority records for every kind of entity, including persons, corporate bodies, families, works, expressions. No one argues that the presence of 670 fields on, for example, a record for a person, means that we are mixing FRBR entities in person records.In the 643 field we record information that is helpful to identify the series, namely, the place and publisher of the resource we have in hand when establishing the series, but it has always been understood that this may change, i.e. the information may not apply to the entire series asa work.
642. Series numbering example. Recording the information in this field is governed by RDA 24.6,an instruction for recording a relationship between a work as a whole and a part. The information in 642 is work-related. Although RDA 24.6 is worded in terms of works, it is contained in Chapter 24, which provides general guidelines for works, expressions, manifestations and items. It seems appropriate that the instructions about numbering of parts could apply to expressions as well. The presence of 642 in series authority records is not evidence that we mix FRBR elements in these records, since the information is about a relationship and is appropriate either at the work or expression level.
641. Series numbering peculiarities. Like 642, this information is work-related. It gives further details about recording the relationship between a work and a part, or problems with the whole-part relationship for a given series.
640. Series dates of publication, etc. The sort of information recorded in this rarely-used field gives information that helps identify the series as a whole—it was only published in 15 bibliographic volumes, so this work won’t have any more than that; it was created between 1864 and 1910; etc. This is work-related information.
022. It is permitted to record an ISSN in a series authority record, although this isbeing done somewhat inconsistently at the present time. The RDA instructions for recording ISSN are in Chapter 2 (2.12.8), but these instructions are for transcribing manifestation information in 490—2.12.8 is not about recording information in an 022 in a bibliographic or authority record. It is true that we are able to record the number in an 022 in an authority record and it has been our custom to do so, but we are not convinced that that practice is governed by 2.12.8. We think it’s a useful piece of information that helps identify the series (work). It’s also true that ISSN is given as an example under 2.15, “identifier for the manifestation” so in that sense it is a manifestation-level piece of information we’re recording, but again, it’s something that helps to identify the work—if we have two separate series with the same name, the fact that they have different ISSNs is a clue that they’re different works.
642, 644, 645.The local treatment fields, 642, 644, and 645, are non-RDA, and do not have anything to do with WEMI—they are about local record-keeping.
SCS: Should 642 be 646 instead? The latter seems more like “local treatment” but 642 does not.
050. 050 is also currently non-RDA, though information about classification may eventually appear the subject chapters. Classification information such as that found in 050 records a relationship of the work to a topic, and so is appropriate in a work-level record, but 050, like the local treatment fields, is really mainly about local record-keeping—where are we going to shelve resources that appear in this series?
Recommendation
Add the following language to DCM Z1 - Introduction - Series authority records, preceding the paragraph beginning “Until 1983 …”:
Recommended policy statement
DCM Z1 - Introduction - Series authority records . Catalogers should do their best not to mix FRBR entitieson the same series authority record.Elements that are specifically expression-related should not be addedinto work records and work-related elements should not be added into expression records (this specifically applies to type of work (380), language of expression (377), and content type (336)).
Note to PCC SCT/PoCo: we assume we will eventually be allowed to recordtheexpression-related element content type again. See also below, at I.3.
I.2. Adding “language of expression” to series authority records for works.
Discussion
Most series works, if they are expressed in a language, are expressed in only one language, inversely paralleling the typical bibliographic record which does triple duty (where the work and expression are implicitly represented because they are embodied in one and only one manifestation).One series authority record is usually created to primarily represent the work, but in current practice that series authority record may also “stand in” for the original language expression and the original manifestation of that series work. This practice (using the same record and access point to represent two different entities, the work and the expression), which originates with AACR2 practice for uniform titles, does not seem to be completely in accord with RDA, but it is parallel to the treatment of work-level records for non-series works. If a series exists in a single expression, is it appropriate to add “language of expression” (in MARC 21 field 377) to a series authority record for a work (essentially making the record perform double duty)? Or should “language of expression” be only added to series authority records that arefor expressions (e.g., a series authority record for a series consisting of all the plays of Shakespeare in French translation)?
Recommendation
The “Language of expression” element should not be addedto work-level records because this is not an element of the work description. We may not agree with FRBR’s ultimate decision (followed by RDA) not to allow an element called “original language of the work”, but that’s the way the model was set up. As discussed in I.1, our NACO series practice does not, in fact, typically mix elements of different FRBR entities within the same authority record. We should not at this point be introducing practices that mix FRBR/RDA elements in these records. A series is a type of work, an aggregate work. In the world of works, they’re a pretty small subset. Language of expression is not (or at least should not be) recorded in other kinds of work records (e.g. n78030104, Homer. Iliad). There is no compellingseries-related reason to record the original language of the series in the work record for the series when we don’t record it for other works.
Recommended policy statement:
LC-PCC PS 6.11.1.3. PCC practice: Series authority records. Record language of expression in series authority records only if the record represents an expression. Do not record language of expression in series authority records that represent works.
SCS: Reword as LC-PCC PS 6.11.1.3. PCC practice: Series Authority Records: Record language of expression in series authority records only if the record represents an expression other than the original language. Use field 377 for recording the language.
Note: it would probably be preferable to generalize this to apply to all work and expression authority records, not just series, but these recommendations apply only to series policy. A generalized PS should make it clear that the policy applies to series as well as to other types of work-level authority records.
See also below, under I.3, for an addition to this proposed policy statement.
I.3.Recording RDA/FRBRattributes
Discussion
Recording of many RDA work and expression attributes is optional. Should the PCC have a general policy in this area?
Regarding specific attributes:
a.Use 046 for work (RDA 6.4) and/or for expression records? (RDA 6.10)
b.Should the PCC have a recommendation about using “Form of work”? Also, should a list of terms for “Form of work” be devised?” (RDA 6.3; element applies to all works, not just series)
c.Use Content type on series authority records (RDA 6.9; element again applies to all expressions, not just series expressions; currently an RDA core element, but its use is currently restricted by LC-PCC PS to bibliographic records)
SCS: Questioning whether this is actually true? It was one of the recommendations from the PCC Expressions Task Group, but do not think it made it to the LC-PCC PSs.
d.Other attributes
e.Recording attributes when they already appear in an authorized access point?
Recommendation
In a linked data environment, it is desirable to encourage consistency whenever a given element is likely to be used for finding resources. This consistency should extend both to the terms used (drawn from a controlled vocabulary whenever possible)and to their assignment (assigned whenever the element is likely to be used for finding resources). If an element is unlikely to be used for finding resources but is likely to be used for selecting them from a result set, then this consistency is less critical but still desirable. Regardless of this, the inclusion of most non-core elements should generally remain at the discretion of the cataloger.
Ideally, the machine environment in which the authority record is created should be able to automatically generate the elements and more general values identified below under the LC-PCC PS for 6.3.1.3. For example, a 380 field containing the value “Series (Publications)” and its more specific variants might be automatically generated by the presence of specified values in 008/12 (Type of series): a = monographic series; b = multipart item; c = series-like phrase; z = other. However, we recommend recording such information in 3XX as well.
Recommended policy statements
DCM Z1 Introduction - Series Authority Records. Record any attribute used as a qualifier in an authorized access point both in theauthorized access pointand as a separate element (using fields 046/336/368-386). Recording attributes in other situations is strongly encouraged.
LC-PCC PS 6.4.1.3. PCC practice: Record Date of Work in series authority records using the 046 field whenever it has been used as a qualifier to the authorized access point. Otherwise, record date of work whenever it is readily ascertainable.
SCS: The work is the series, so the date of the work would be the date of publication of the component part that was published first, regardless of whether this part carried the lowest numbering or whether it carried a chronological designation that was earlier or later than its date of publication. The situation is clarified in CCM 10.4.3, which refers to the use of "first published issue" in AACR2 1.4F8.
LC-PCC PS 6.10.1.3. PCC practice: Record Date of Expression in series authority records using the 046 field whenever it has been used as a qualifier to the authorized access point for the expression. Otherwise, record date of expression whenever it is readily ascertainable.
LC-PCC PS 6.3.1.3.PCC practice: Record form of work in all work-level series authority records using the 380 field. At a minimum in all work-level series authority records, include
380$a Series (Publications) $2 lcsh
Recording form of work for the type of series is also strongly encouraged:
380$a Series (Publications) $a Monographic series $2 lcsh
or
380$a Series (Publications) $2 lcsh
380$a Multipart monograph
(Multipart monograph series require two 380 fields because “Multipart monographs” is not currently in LCSH.)
Additional forms may also be recorded as appropriate (e.g. Plays, Poems, Novels. LCSH and LCGFT are good sources, but the Art and Architecture Thesaurus sometimes has better form terms than LCSH does, e.g. “Poems” vs. “Poetry”).
Do not record “Series (Publications)” in records for series-like phrases. Instead, record
380$a Series-like phrase
NOTE to PCC SCT and PoCo:
On series-like phrases, see also below, III.6. While it could be argued that a series-like phrase does not represent a work, and so cannot have a “form of work,” this seems to be a practical place to record this information.
SCS: Agrees that an authority record is a practical place to put series-like phrases.
At least one member of the larger series training task force wasstrongly opposed to the above treatment for multipart monographs, arguing that multipart monographs were not series. However, RDA’s definition of series is “A group of separate resources related to one another by the fact that each resource bears, in addition to its own title proper, a collective title applying to the group as a whole.” This definition encompasses both monographic series and multipart monographs. Both are series as far as RDA is concerned. Additionally, the distinction is a somewhat artificial cataloger construct. Most users, including non-cataloging librarians, would see both as series.
SCS: Agree with the conclusion that multipart monographs and monographic series can both be considered series.
LC-PCC PS 6.5.1.3. PCC practice: Record Place of Origin of the Work in series authority records using the 370 field whenever it has been used as a qualifier to the authorized access point. Place of Origin of the Work may also be recorded in cases where it has not been used as a qualifier. Especially consider recording it if it would be helpful in retrieving or identifying the work, is particularly closely associated with the series, or would be otherwise helpful to users of the record.
SCS: The Place of Origin of the work is typically associated with a manifestation of the work rather than the work itself. For a serial, it may represent only a passing association (since conference proceedings often move from place to place). Recording it as an element in these cases is of little practical use, so we should limit this recording, perhaps only when you are gaining some identification here. For now, generally limit it to when it is used as a qualifier and the place is actually the origin of the work, not merely the manifestation.
LC-PCC PS 6.9.1.3. PCC practice: Record Content type in expression-level series authority records using the 336 field.
NOTE to PCC SCT and PoCo:
Content type (RDA 6.9) is a core element, and the prohibition against its use on authority records should be lifted. Under RDA Implementation Scenario 1, a series authority record will become a work or expression record. Content type is an expression element, so should only be recorded on expression level series authority records, not on the basic SAR work record. The purview of this task force is series, but we recommend that the lifting of the prohibition be general, not limited to series authority records. If so, we recommend that the PS be generalized; we do recommend that it be policy to record content type in expression-level series authority records; recording it in other types of expression-level records could be optional, if SCT/PoCo prefers.