Bibliographic Standards Committee
1979 IRLA Report
This report is the foundational document for the Standards Committee (later, Bibliographic Standards Committee) of RBMS. We take pleasure in making this seminal report widely available, with permission of Marcus McCorison.
Proposals for Establishing Standards for the Cataloging of
Rare Books
and Specialized Research Materials in Machine-Readable Form
Independent Research Libraries Association
185 Salisbury Street
Worcester, Massachusetts 01609
December 1979
INDEPENDENT RESEARCH LIBRARIES ASSOCIATION
Ad Hoc Committee on Standards for Rare Book Cataloging
in Machine-Readable Form
M. A. McCorison, Chairman
American Antiquarian Society
Joan M. Friedman
Yale Center for British Art
Barbara A. Gates
John D. Rockefeller, Jr., Library
Brown University
Kay Guiles
Library of Congress
John Knapp
Blackwell North America, Inc.
Alexandra Mason
Kenneth Spencer Research Library
J. William Matheson
Library of Congress
Marion Schild
Brooklyn, New York
(January-September 1979)
G. Thomas Tanselle
John Simon Guggenheim Memorial Foundation
Stephen Paul Davis, Secretary
Library of Congress
The work of the Committee has been funded, in part, by a grant
From the National Endowment for the Humanities
Table of Contents
1. Proposed MARC Format Change: Access by Genre
2. Proposed MARC Format Change: Place of Publication/Printing Access (Indirect)
3. Proposed MARC Format Change: Place of Publication/Printing Access (Direct)
4. Proposed MARC Format Change: Copy-Specific Added Entry Fields
5. Proposed MARC Format Change: Access By Publishing/Physical Aspects of Item
6. Proposed MARC Format Change: Extent of Item (Copy-Specific)
7. Local and Copy-Specific Information
9. Additions to Expanded Library Character Set
11. Draft Standards: Genre Terminology, Illustration/Graphic Technique
12. Standard Citation Forms for Bibliographical References
13. Library of Congress Local Data
14. Authority Information for Retrospective Name Authorities
15. Library of Congress manual for cataloguing older imprints
In late 1978, the Independent Research Libraries Association established an Ad Hoc Committee on Standards for Rare Book Cataloguing in Machine-Readable Form, with funding from the Research Division of the National Endowment for the Humanities. The Ad Hoc Committee met in Washington in March 1979 with a preliminary agenda. Tasks were outlined and assigned to members, and work progressed through the spring and summer. An interim report containing seventeen draft proposals was issued in September 1979 and circulated to more than 150 libraries and individuals in the United States, Canada, the United Kingdom, France, and New Zealand. The Ad Hoc Committee met again in October 1979 to review the interim report and the many comments on it that had been received.
The Committee herewith presents its final report in the form of fifteen proposals and recommendations for the consideration of our colleagues. If adopted, they will provide means of access to information about the characteristics of specialized research materials that at present cannot be recorded in standard ways in machine-readable records. To this end, the Committee has suggested a number of options in forms that are compatible with and parallel to current practices in the standard MARC II format. These options have been constructed and drafts of thesauri prepared in order that opportunities for access be achieved through mutually agreed-upon standard practices among institutions that choose to catalogue specialized research materials by machine-readable methods.
Some proposals deal with physical characteristics of a publication that were acquired at time of production or, on the other hand, with data that are relevant to elements acquired by the copy following production. These proposals are designed to lead the cataloguer to focus on the distinction between an entry that is based on a single copy and one that generalizes about an edition or impression as a whole—and thus to indicate more clearly to the users of the catalogue what sort of description has been attempted.
In all cases, the Committee wishes to emphasize that the use of these options, should they be adopted by the American Library Association, the Library of Congress, and the bibliographic networks, would be entirely optional on the part of any institution. That is to say, although standards are proposed for the recording of certain data in machine-readable form, the decision to include or not to include specific fields within a cataloguing record is entirely a matter of choice for each institution. We assume that the choice is to be made in light of the needs and interests of an institution—not on the basis that a set of standards exists.
Three additional points are worth noting. At the time of this writing the MARC format proposals (Proposals One through Six) have been submitted to the Library of Congress's Automated Systems Office. They were discussed at the joint Networks/National Library of Canada/Library of Congress meeting of November 8-9, 1979, and will be presented formally to the RTSD/RASD/LITA Committee on Machine-Readable Bibliographic Information (MARBI) of the American Library Association at its January 1980 meeting. Unless these proposals are promulgated as acceptable standards, they remain only proposals.
Second, in Proposals Seven through Fifteen the Committee has considered other problems and has, in some cases, compiled preliminary drafts of lists of terminology with the hope that they will lead to the early standardization of access language for the description of rare and specialized materials. For, unless standard terminology is adopted, as we presently accept uniform subject headings, access to information in machine-readable form will be impossible except in purely local situations. These recommendations have been forwarded to the Standards Committee of the Rare Books and Manuscripts Section (RBMS), Association of College and Research Libraries. Ms. Helen S. Butz, Head of Rare Book Cataloguing, University of Michigan Libraries, presently is the chair of that committee.
Third, the Committee is pleased to report that a handbook on cataloguing specialized research materials is being prepared under the auspices of the Library of Congress. It will coordinate rules found in the second edition of the Anglo-American Cataloguing Rules with recommendations contained in the latest draft of the International Standard Bibliographic Description for Antiquarian Materials (ISBD[A]). Ben R. Tucker, Head, Office of Descriptive Cataloging Policy, has the responsibility for preparing this important document which will supply the other part of the Ad Hoc Committee's charge—an interpretation of AACR 2 for rare book cataloguing.
The Committee thanks Mr. William J. Welsh, Deputy Librarian of Congress, and his colleagues at the Library of Congress for their cooperation and invaluable assistance with the work of the Ad Hoc Committee. The Research Resources Program of the National Endowment for the Humanities, Mrs. Margaret S. Child, assistant director, generously provided funds that defrayed the not inconsiderable costs of the Committee's work. We are grateful for this help.
To Stephen Paul Davis, the Secretary of the Committee, who expended an enormous amount of energy in the tasks of formulating and revising these proposals and recommendations, the Committee gives its heartiest thanks and warmest appreciation.
M. A. McCorison
Chairman
American Antiquarian Society
December 17, 1979
Many of the following proposals have changed substantially since the Interim Draft was issued in September. On the basis of discussion at the joint Networks/National Library of Canada/Library of Congress meeting in Washington, D.C., November 8-9, 1979, Proposal One of the Interim Draft ("Auxiliary Access Fields—Contributor to Publication, Distribution, or Production Aspect of Work") was withdrawn. It was recommended that such added entries be recorded in the regular 700/710/711 fields with the use of a relator and/or relator code to distinguish the particular function of the person or body. The Ad Hoc Committee feels that this is an acceptable alternative. This decision, however, makes the standardization of relator terms and the implementation by the networks of relator codes (see Proposal Eight) all the more pressing a matter.
Proposal Three of the Interim Draft ("Place of Publication/Printing Access") was divided into two separate proposals to allow for both hierarchical and non-hierarchical place names as access points.
The first six proposals are now also numbered as LC-216 through LC-221, reflecting the fact that these have become formal Library of Congress proposals to the networks and the American Library Association.
A matter that came up with great frequency in the comments on Proposal Seven of the Interim Draft was the desirability of displaying local data in a network environment. While the Committee was sympathetic to this need, it nonetheless felt that it was more important at present to concentrate on defining the problem of copy-specific information and developing ways of recording such data in the MARC format with adequate content designation. The display (or suppression) of particular kinds of data depends on system design, economics, and politics, and will need to be addressed by users of networks through other channels, such as users' groups.
Proposal Eleven of the Interim Draft originally distinguished between "intellectual" and "physical" genres. As a result of comments on the interim draft and the change of the original Proposal Two from a fixed to a variable field it was thought that the two lists could be merged. Accordingly, the Draft Thesaurus of Generic Terms now contains both physical and intellectual genres. Its length has been greatly increased by suggested additions from outside the Committee, and the result is really a list of candidates for a thesaurus. This will be presented to the RBMS Standards Committee for further refinement in the hope that work on it might be completed by mid-1980 and publication follow shortly thereafter.
Proposal Fourteen of the Interim Draft ("Distribution of LC Authority
Information") was deleted from the Final Report after discussion at the
Committee's September meeting indicated that the distribution of LC authority
data in card form was felt to be an unwieldy solution for most rare book and
special collection libraries.
Since this proposal was originally formulated, all major bibliographic networks have announced plans to provide access to LC's automated name authority file on-line before 1981. This will substantially reduce the difficulty of obtaining AACR2 forms of names for those headings that LC has added to its automated system. (However, this does not in any way solve the problem of obtaining AACR2 forms for those headings that LC has not added to its automated system.)
The Committee is grateful for the large number of comments it received on the Interim Draft. Many of these comments are reflected in the revisions made for the final report. All comments—including those received too late to be considered by the Committee as a whole—will be forwarded to Ms. Helen Butz, Chair of the new RBMS Standards Committee. We ask that all comments on this our final report also be sent directly to Ms. Butz. (Her address: Rare Book Cataloging, Technical Services Dept., University Library, University of Michigan, Ann Arbor, Michigan 48109.)
S.P.D.
LC-216
December 4, 1979
Proposed MARC Format Change
1. Formats affected: Books, Maps, Music, Serials
2. General description of change: Define one additional added entry field for the Books, Maps, Music, and Serials formats
3. Specific description of change:
Add the following field to the Books, Maps, Music, and Serials formats:
655 Auxiliary Access Field (Genre) [R]
$a Access Term for genre [NR]
$x General subdivision [R]
$y Period subdivision (chronological) [R]
$z Place subdivision [R]
$2 Code source [NR] [mandatory]First indicator - Blank
Second indicator - Source of term
7 - Source is specified in subfield $2 [mandatory]
4. Source of recommendation: IRLA Ad Hoc Committee
5. Reason for change: There is at present no suitable field in the format to accommodate, for purposes of access, terms indicating the genre of a work.
Many specialized libraries have extensive manual files for particular genres of works represented in their collections. These files have in the past been maintained in card form, usually by simply supplying the desired generic term as an added entry, with appropriate subdivision when needed. As such libraries have begun entering records into the bibliographic networks, it has become apparent that the MARC format does not provide suitable content designation for such generic information.
6. Comments
The need for generic access is most conspicuous in those libraries with special
research collections and older printed materials. Unlike most modern works,
for which the chief approaches are subject, author, or title, many types of
special research materials are most usefully accessed through terms indicating
their genre, often in combination with place and date of publication.
One might, for example, want to investigate sermons published in Boston in
the 1690's, or almanacs published in Philadelphia in 1774.
That this kind of access is useful for modern materials as well, can be seen by the fact that the term "Almanacs" is indeed an LC subject heading, but one which is applied—against the general logic of subject headings—to almanacs themselves, rather than to works about almanacs. Library of Congress Subject Headings, 8th ed., contains a number of such anomalous generic terms, although they are not consistently or exhaustively covered there. In addition, the use of such form subheadings as "Liturgy and ritual" and "Laws, statutes, etc." in pre-AACR2 cataloging codes was an attempt to fulfill the same kind of access need. The Ad Hoc Committee believes that a consistent technique for providing generic access is called for.
It is the consensus of the Committee that the indication of a work's genre differs significantly from subject analysis as it is usually defined (despite the above-mentioned anomalies in the LC subject heading scheme), and that genres should therefore be kept distinct from subjects in MARC content designation. A further reason for distinguishing generic terms from subject terms is the fact that generic access is not mandated by AACR2 or LC subject practice.
A draft standard for a Thesaurus of Generic Terms for Older Materials is being proposed by the IRLA Ad Hoc Committee to the appropriate bodies in ALA (see working draft of the list in Proposal 11). It is expected that the ACRL RBMS Standards Committee will further the development of this proposed standard and that it will become the preferred list for use by rare book and other special collections. A working draft of this Thesaurus is expected by June 1980. The terminology included in the working draft is oriented to books and serials; terminology oriented to maps and music needs to be developed by the appropriate specialists in these areas.
Subfield $2 has been included in this proposal to be in accord with the technique recently established in the format which requires that the list or thesaurus from which an access term has been taken be designated by a coded value. Indicator value '7' is the only indicator value to be defined. The use of both subfield $2 and indicator value '7' would be mandatory in this field. The RBMS Standards Committee will be advised to request a coded value for subfield $2 from the Federal Library Committee as soon as the rare book and special collections thesaurus has been completed.
The Committee feels that field 655 should not be used for "uncontrolled headings" (as field 653 has been defined).
The MARC format does include several fixed field positions to indicate information similar to that dealt with in this proposal. For instance, 008, positions 24-27 (Form of Content Codes); position 30 (Festschrift Indicator); position 33 (Fiction Indicator), etc. It would perhaps be possible for generic information to be handled in a similar way. The Committee felt, however, that a variable field would be a more appropriate solution for the following reasons:
a. All existing fixed field positions in 008 (Books format) have already been define, and the Committee was advised that it was not practicable to expand 008 at this time.
b. The expected number of terms needed to indicate genre is greater than the 35 values that can be coded in one fixed field byte. In addition, the case does arise in which two or more generic terms need to be assigned to one work. To accommodate both of these contingencies, one would need to define a two-byte fixed field value capable of being repeated up to three times, at least, requiring a minimum of six fixed field positions. This was thought to be an unwieldy and probably inefficient solution.
c. No firm standard for the application of generic terms yet exists—although the need to enter generic terms in machine records is immediate (especially for some of the large retrospective projects now in their early stages). Although the IRLA Ad Hoc Committee is proposing a draft thesaurus of generic terms, it is likely that it will be some time before this terminology is completely standardized. If a set of coded fixed field values were defined now, it is likely that there would be a good deal of revision needed in the future.
d. Libraries other than those that collect older materials may also wish to employ some degree of generic access. The IRLA Ad Hoc Committee feels itself competent to recommend generic terminology only for pre-1900 materials. Other interested groups may wish to expand the list of generic terms developed by the Committee or develop additional thesauri of their own. If a fixed field solution were adopted, this again would necessitate continual revision of any list of coded values initially defined.
e. The general feeling of the Ad Hoc Committee (and of groups outside the Committee that contributed comments on the Committee's work) was that a fixed field solution would effectively render the data thus coded unusable in the immediate future by libraries. Most libraries outside of LC that use the MARC formats receive their processing services from utilities whose output programs do not provide the print constants for fixed field data that would make such data accessible to users.
7. Additional comment
The Thesaurus of Generic Terms for Older Materials will, when completed, include complete instructions for subdivision of terms. (It is felt that subfields $x, $y, and $z will accommodate all the kinds of subdivision required by this particular thesaurus. Subfields $y and $z will be used in much the same way as they are used for LC subject headings; $x will be used—also paralleling its use for subject headings—for any kind of subdivision not encompassed by $y or $z as defined.)
Examples of the possible use of this field follow:
655 _7 $a Emblem book $z Germany $y 17th century $2 [xx]
655 _7 $a Dictionary $x French $y 18th century $2 [xx]
655 _7 $a Catalog (auction) $z Great Britain $y 19th century $x Sotheby's $2 [xx]
[For a preliminary draft list of generic terms, see Proposal Eleven]
LC-217
December 4, 1979
Proposed MARC Format Change
1. Formats affected: Books, Music, Maps, Serials
2. General description of change: Add one field to Books, Music, and Maps formats; change definitions of field in Serials format.
3. Specific description of change:
Extend definition of field 752 (Serials format) as follows. Add field 752 to Books, Music, and Maps formats.
752 Place of publication/printing access (Indirect) [R]
$a Country of publication/printing [NR]
$b State or province of publication/printing [NR]
$c County of publication/printing [NR]
$d City of publication/printing [NR]
$j Linking data (Tag and sequence number of the 700/710/711 field to which this field is linked) [R]Indicators - blank
4. Source of proposal: IRLA Ad Hoc Committee
5. Reason for change: The need for rare book and special collections libraries to maintain various kinds of imprint files for books, printed music, and maps. (Such imprint files are typically arranged first by place of publication, then chronologically; or, alternatively, by place of publication, then by publisher/printer, then by title or date.) At present this cannot be done by indexing on the $a subfield of 260 because of the lack of normalized forms of place names there and the ambiguity of data. This will continue to be true under AACR2 or ISBD(A).
6. Comments
Field 752 would be used in those cases where a hierarchical form of place name was desired. Use of the field would be according to the following pattern (following the use already established in the Serials format):
a. For United States imprints:
name of country, state, and city of publication
b. For Canadian imprints
name of country, province, and city of publication
c. For other foreign imprints
name of country and city of publication
In those cases in which a direct form of entry was desired, field 751 (proposed) would be used (see LC-218).
This field should be repeatable for those items with more than one place of publication/printing. In cases of multiple 752s, the $j subfield would be necessary in order to link each place of publication to the added entry field for the printer/publisher located in that place. The tag and sequence number of the appropriate 700/710/711 field(s) would be entered in each $j subfield. For example:
700 11 $a Franklin, Benjamin, $e printer
700 11 $a Dalton, James, $e bookseller
752 _ _ $a United States $b Pennsylvania $d Philadelphia $j 700/1
752 _ _ $a United States $b New York $d New York $j 700/2
Subfield $j is repeatable when more than one place of publishing/printing needs to be linked with more than one publisher/printer:
752 _ _ $a United States $b Massachusetts $d Boston $j 700/1 $j 700/2
752 _ _ $a United States $b Rhode Island $d Providence $j 700/3 $j 700/4
Because such complex imprints occur in up to 15% (roughly estimated) of pre-1850 imprints, the use of a linking technique is necessary to maintain the associations between various added entry names and the places in which they produced their contributions. If these associations are not somehow retained, certain kinds of imprint files and indexes will not be able to be generated programmatically. An example of such an imprint index would be the type not infrequently found in published inventories of 16th and 17th century books, arranged first by city of printing/ publication, then subarranged by the printer or bookseller located in that city.
In most cases, however, there will be only one place name associated with a work and all 700/710/711 added entries could be subarranged under the one 752 field present in the record. In such cases no $j subfield would be used.
7. Additional comment
Place names in field 752 would be formulated according to AACR1 (after 1/2/81: AACR2) except that the qualifiers would be omitted. For example:
752 _ _ $a France $d Paris
not 752 _ _ $a France $d Paris (France)
LC-218
December 4, 1979
Proposed MARC Format Change
1. Formats affected: Books, Maps, Music, Serials
2. General description of change: Add one field to the Books, Maps, Music, and Serials formats.
3. Specific description of change:
Add the following field and definition to the Books, Maps, Music, and Serials formats:
751 Place of publication/printing access (Direct) [R]
$a Place of publication/printing [NR]
$j Linking data (Tag and sequence number of the 700/710/711 field to which this field is linked) [R]Indicators - blank
4. Source of proposal: IRLA Ad Hoc Committee
5. Reason for change: [See LC-217]
6. Comments:
Field 751 would be used in those cases where a direct form of the place name is desired as an access point. Place names in this field would be formulated according to AACR1 (after 1/2/81: AACR2) including any qualification required by the cataloging code or by LC practice. The name of the city of publication/printing would be used when known. If the city is unknown, AACR2 1.4C6 would be applied.
EXAMPLES
751 _ _ $a Philadelphia (Pa.)
751 _ _ $a Paris (France)
7. Additional comments:
For use of subfield $j, "Linking data," see LC-217, p. 2.
LC-219
December 4, 1979
Proposed MARC Format Change
1. Formats affected: Books, Maps, Music, Serials
2. General description of change: Define three additional fields
3. Specific description of change:
Explicitly define the following fields for the Books, Maps, Music, and Serials formats:
790 Copy-specific added entry - Personal name [R]
$a Name
$b Numeration
$c Titles and other words associated with the name
$d Dates associated with the name
$e Relator
$f Date (of a work)
$g Miscellaneous information
$h Media qualifier
$k Form subheading
$l Language
$m Medium of performance (for music)
$n Number of part/section (for music, this may be the serials, opus, or thematic number or date used as a number
$o Arranged statement (for music)
$p Name of part/section (of a work)
$q Qualification of name (expansion of initials)
$r Key (for music)
$s Version
$t Title (of a work)
$4 Relator code
$5 Library (and copy as needed) to which this field applies [R]First indicator - Type of personal name
0 - Forename
1 - Single surname
2 - Multiple surname
3 - Name of familySecond indicator - Blank
791 Copy-specific added entry - Corporate name [R]
$a Name
$b Each subordinate unit in hierarchy
$c Place
$d Date
$e Relator
$f Date (of a work)
$g Miscellaneous information
$h Media qualifier
$k Form subheading
$l Language
$m Medium of performance (for music)
$n Number of part/section (for music, this may be the serial, opus, or thematic number or date used as a number)
$o Arranged statement (for music)
$p Name of part/section (of a work)
$r Key (for music)
$s Version
$t Title (of a work)
$u Nonprinting subfield
$4 Relator code
$5 Library (and copy as needed) to which this field applies [R]First indicator - Form of corporate name
0 - Surname (inverted)
1 - Place or place and name
2 - Name (direct order)Second indicator - Blank
752 Copy-specific added entry - Conference or meeting [R]
$a Name of meeting or place element
$c Place
$d Date
$e Subordinate unit in name
$f Date (of a work)
$g Miscellaneous information
$h Media qualifier
$k Form subheading
$l Language
$n Number of part/section, etc.
$p Name of part/section (of a work)
$q Name of meeting following place element
$t Title (of a work)
$4 Relator code
$5 Library (and copy as needed) to which this field applies [R]
First indicator - Form of conference or meeting
0 - Surname (inverted)
1 - Place or place and name
2 - Name (direct order)Second indicator - Blank
4. Source of recommendation: IRLA Ad Hoc Committee
5. Reason for change: As more rare book and special collections libraries have begun inputting data into national networks, the need for a closer definition of so-called local fields has become apparent. In the past, many libraries have routinely maintained manual files (some going back for decades) of information having to do with the particular copies of a work which the library holds—but not necessarily to any other copy of the work. Those libraries which now maintain provenance, binding, donor files, etc. will wish to retain this capability when building their machine-based files. Special access files are useful tools for scholars of the history of books and printing, as well as for historians of other kinds. In many cases, such files are the principal means of providing access to the most important features of older imprints.
Unfortunately, the capability does not exist at present for copy-specific added entries to be entered into bibliographic records in a coherent, standardized way which will assure the long-range integrity of the files now being constructed. Because fields for copy-specific information have not yet been defined in the format, many libraries have taken to dumping all kinds of disparate data into those fields which are available, either in the format or in vendors' expansions of the format. An example of this would be the use of the 690 field in OCLC, into which many libraries are inputting various kinds of data simply in order to have it retained somewhere in the record and to be able to have overprinted catalog cards generated. It is regrettable that, when the time comes for these libraries to close their catalogs or produce bibliographies and indexes, the machine records which remain will, in many instances, be of little use.
A more systematic approach to the definition of local or "copy-specific" fields in the format would be of benefit to many libraries of all kinds and, in addition, would foster inter-institutional compatibility in the files now being created by various institutions in isolation.
6. Examples
790 1_ $a Jefferson, Thomas, $d 1743-1826, $e former owner $5 CtY c. 3
790 1_ $a Bridgewater Library, $e former owner $ 5 MH (Houghton)
790 1_ $a Dawson, Thomas, $d b. ca. 1712, $e binder $5 MWA c. 2
790 1_ $a Marsden, Peter George, $e donor $5 DLC c. 1
790 1_ $a Benson, Elizabeth, $d 1903-1975, $e donor $5 IdU
7. Comments
Note that the 790/791/792 fields would be defined only for copy-specific added entries. What is now imprecisely referred to as "local information" should in all cases be evaluated as to whether it is copy-specific or copy-general and then entered into the 79x fields, or the 700/710/711 fields as appropriate, depending on the nature of the added entry.
Identification of specific kinds of copy-specific added entries (for searching and listing) would be accomplished by the use of appropriate designations in the relator subfield ($e) or by use of the relator code ($4).
Subfield $5 would be used to associate the particular copy-specific added entry with the library and copy to which it applies (a technique corresponding to that used in the authority format for series). The library is identified by the NUC symbol. The NUC symbol is followed by the copy (or set) number when required. If a copy number applies only to a particular administrative unit or collection within a library, the copy number may be followed by an appropriate qualifier in parentheses.
Proposal Five
LC-220
December 4, 1979
Proposed MARC Format Change
1. Formats affected: Books, Maps, Music, Serials
2. General description of change: Addition of one field to the Books, Maps, Music, and Serials formats
3. Specific description of change:
Add the following new field and description:
755 - Auxiliary Access (Publishing/physical aspects of item) [R]
$a Access term
$j Linking data (Tag and sequence number of the added entry field to which this field is linked) [R]
$x General subdivision [R]
$y Period subdivision (chronological) [R]
$z Place subdivision [R]
$5 Library (and copy as needed) to which this field applies [R]First indicator - Aspect being described
0 - Publishing/bookselling information
1 - Paper and papermaking information
2 - Printing information
3 - Typographic information
4 - Illustration/graphic information
5 - Binding information
6 - Provenance evidence
7-8 - [undefined]
9 - OtherSecond indicator - Blank
4. Source of proposal: IRLA Ad Hoc Committee
5. Reason for change:
Libraries and special collections which foster research into the history of books and printing have in the past needed and will continue to need to create special files arranged according to various aspects of the publication or physical production of individual books. Some books are of interest chiefly because of their physical form.
An otherwise unimportant book might, for instance, demonstrate eccentricities in its printing that illuminate a particular practice in the history of printing. Or it might be clothed in a particularly fine binding or be decorated by a fore-edged painting. Furthermore, every book may be approached as a physical object, a product of a particular moment in printing and publishing history. The study of such history is greatly facilitated by records organized so as to provide access to examples of books illustrating particular features (whether routine or apparently unusual) of book production.
At present it is not possible to accommodate such information in the format with sufficient content designation and structure to generate coherent files. Many libraries needing the capability to retrieve or generate files from such information have been forced to develop makeshift solutions to this problem. A standard approach in this area is needed, both for local processing needs and for eventual national applications, such as union catalogs, remote access systems, etc.
6. Comments
Some features of the publication/production of a book that need to be recorded and structured for access may apply only to the copy of the book in hand, whereas others may apply to all copies of an impression. The number of copies involved, however, has nothing to do with the relevance of a given feature to book production history. The crucial distinction, rather, is between those characteristics of a copy that result from the process of production and publication and those characteristics that result from what has happened to the copy after the time it left the hands of its producer. Important as this distinction is, it is not one that can always be made on the basis of the examination of a single copy—as, for instance, when manuscript corrections were made in copies of an edition before they were released from the printer's or publisher's office.
In view of these factors, the Committee felt that no distinction should be made in the format in this case between copy-specific and copy-general features (or between features originating before and after publication). Such distinctions would, if necessary, have to be made in the general note areas (500/590) or in the access terminology itself.
The terms to be used in these fields have only a fair degree of consistency in the rare book and special collections community. Draft standards for some of the terminology involved in such entries are being proposed by the IRLA Ad Hoc Committee to ALA. The RBMS Standards Committee will be asked to continue working on standard thesauri of such terms, but until such time as a standard exists, the access terminology in this field would be uncontrolled.
7. Examples
[Publishing/bookselling information]
755 0_ $a Privately published book
755 0_ $a Large paper edition
[Paper and papermaking information]
755 1_ $a Watermark $x Lion $5 MWA c. 1
755 1_ $a Unopened $5 MH c. 2
[Printing information]
755 2_ $a Cancellans $5 CtY c. 3 (Mus. Lib.)
755 2_ $a Press figures $5 MWA c. 1
[Typographic information]
755 3_ $a Fraktur $z Nuremberg $5 MH c. 2
755 3_ $a Romains du roi $z Sudan $y 1624 $5 MWA c. 1
[Illustration/graphic information]
755 4_ $a Lithograph $5 CtY c. 3
755 4_ $a Hand colored $5 CtY c. 2
[Binding information]
755 5_ $a Boards $5 CtY c. 3
755 5_ $a Inlay $x Mother of pearl $5 DLC
[Provenance evidence]
755 6_ $a Autograph $5 MWA c. 4
755 6_ $a Bookplate $5 MWA c. 5
[Other]
755 9_ $a Sophisticated copy $5 DLC
6. Additional comment
If a library has more than one copy of a book which requires access by such copy-specific information, it is recommended that the field be repeated for each copy.
Subfield $5 would be used to associate the particular copy-specific added entry with the library and copy to which it applies (a technique corresponding to that used in the authority format for series). The library is identified by the NUC symbol. The NUC copy number applies only to a particular administrative unit or collection within a library, the copy number may be followed by an appropriate qualifier in parentheses.
In those cases in which it was desirable to link particular aspects of the publication/physical aspects of item with one or more name added entries, the $j subfield, "Linking data," would be used. With this technique, important associations between physical features and those persons linked in some way to these features will not be lost. For example, when a book has an autograph indicating it was once in Thomas Jefferson's library and also a bookplate indicating it was in the Library of Congress, the following entries would be made:
755 6_ $a Autograph $j 790/1 $5 MWA c. 2
755 6_ $a Bookplate $j 791/1 $5 MWA c.2790 1_ $a Jefferson, Thomas, $e former owner $5 MWA c. 2
790 1_ $a United States. $b Library of Congress, $e former owner $5 MWA c. 2
The definition of subfields $x, $y, and $z would provide a flexible structure parallel to subject headings, which corresponds to the kind of file structure to be found currently in many rare book libraries.
It is felt that the definition of these subfields is adequate as they are now; their specific application, however, would be uncontrolled for the time being and left to local processing needs. The RBMS Standards Committee will be asked to study the need to establish guidelines for such subdivision.
LC-221
December 4, 1979
Proposed MARC Format Change
1. Format affected: Books
2. General description of change: Add a new field
3. Specific description of change:
Add the following field and description to the Books format:
309 Extent of item - copy-specific [NR]
$a Collation/pagination [NR]
$5 Library (and copy as needed) to which this field applies [R]Indicators - Blank
EXAMPLES
300 _ _ $a [2], 76, [16] p. ; $c 19 cm.
309 _ _ $a [2], 76 p. $5 MH c. 1300 _ _ $a xxxviii, 220 p. ; $c 23 cm.
309 _ _ $a iii-xxxviii, 1-48, 51-200 p. $5 MWA c. 2
4. Source of proposal: IRLA Ad Hoc Committee
5. Reason for change: The collation/pagination of an item which is entered into the 300 field is required by AACR2 (by implication) to be that of an "ideal" copy of a work (cf. 2.5B16). Information about the "actual" copy is relegated to the note area.
The description of an "ideal" copy may be a matter of scholarly investigation or even, sometimes, speculation. For those libraries that wish to include the actual collation of an item in hand, in addition to a speculative "ideal" description, it is desirable to have another, parallel extent-of-item field.
6. Comments
This field would be applied in addition to the 300 field, not in place of it (at least at the network level).
It is recommended that this field be repeated for multiple copies of the same item if it is necessary to distinguish differences in the collation/pagination.
Subfield $5 would be used to associate a particular copy-specific extent of item field with the library and copy to which it applies (technique corresponding to that used in the authority format for series). The library is identified by the NUC symbol. The NUC symbol is followed by the copy (or set) number when required. If a copy number applies only to a particular administrative unit or collection within a library, the copy number may be followed by an appropriate qualifier in parentheses.
SUBJECT: Local and Copy-Specific Bibliographic Information
TO BE PRESENTED TO: Library of Congress, Council on Library Resources, the bibliographic
networks, the RBMS Standards Committee
The IRLA Ad Hoc Committee strongly urges the above agencies to work towards the development of a technique to accommodate local and copy-specific information within the MARC format.
Proposals 4-6 of this report address several specific areas in which the rare book and special collections community required copy-specific information and access points. It is clear, however, that a generalized technique within the MARC format needs to be developed also.
The Committee recommends that this problem be given high priority in the general review of the MARC format currently being planned by the Library of Congress and the Council on Library Resources.
Because there is at present no suitable mechanism for such data in the MARC format or in the programs maintained by the bibliographic networks, it is the opinion of the Committee that it would be wise for rare book and special collections libraries to postpone joining a bibliographic network (or, if they have already joined, to postpone entering rare book cataloging) until such a mechanism is developed and implemented by the networks.
SUBJECT: Relators and Relator Codes
TO BE PRESENTED TO: RBMS Standards Committee
I. RELATORS
The once common practice of appending relators (e.g. joint author, editor, illustrator) to names used as access points in bibliographic records has gradually dwindled in generalist libraries. AACR1, as revised, provided only for the use of "joint author"; AACR2 allows optional use of compiler, editor, illustrator, and translator only, and the Library of Congress has decided not to apply even this option. (This despite the fact that UNIMARC, the international umbrella format for all the national MARC formats, calls for relators in all personal and corporate access points in the bibliographic record.)
Discontinuing the use of relators may make sense in the machine and card catalogs of the large general research libraries, in which name entries are not distinguished or segregated according to their varying types of contribution to or association with published works. However, many rare book and special collections libraries will need to continue to maintain special files and have the capability of generating lists and bibliographies according to the particular types of contribution of persons to the intellectual content, physical form, or history of an item. Thus many libraries need to distinguish bibliographic records for works that Benjamin Franklin wrote from those that he printed, or, similarly, those works that George Cruikshank illustrated from books that he once owned personally.
These needs can be met only through the use of relators (or relator codes—see section II of this proposal). Those rare book and special collection libraries that adopt AACR2 will suffer a serious disruption in the services they provide to scholars and researchers if they dispense with the relating terms presently in use.
Therefore, the IRLA Ad Hoc Committee recommends that the RBMS Standards Committee develop a proposal to the RTSD Committee on Cataloging: Description and Access for the revision of AACR2 to permit cataloging agencies the option of using relators when needed. The Ad Hoc Committee further recommends that the RBMS Standards Committee develop and maintain a standard list of relator terms to be used by rare book libraries and special collections.
A suggested starting point for the RBMS Committee's work is provided here
in the attached draft list of relator terms. We have attempted to include in
this list terms that would have widest applicability. We have not tried to be
exhaustive. In addition, the Committee has tried to identify only those relators
by which libraries might actually want to provide special access. For this reason,
relators such as editor, joint author, etc., and the more esoteric process engraver,
seriographer, etc., have been omitted, because few if any libraries would arrange
access solely by such types of contributors. It is recommended that the RBMS
Standards Committee consider taking a similarly
pragmatic approach in developing a standard list.
The need for a standard list of relators stems especially from the increasing use of computers and networks in rare book and special collections libraries. Until the technique of relator codes (see section II of this proposal) is fully implemented, relator terms will provide the only means of access by type of function for the creation of special files, catalogs and printed lists (by string matching on the $e subfield of name fields).
DRAFT LIST OF RELATOR TERMS FOR
RARE BOOK AND SPECIAL COLLECTIONS
Assignee
Binder
Binding designer
Bookseller
Calligrapher
Cartographer
Censor
Compositor
Copyright holder
Dedicatee
Dedicator
Donor
Engraver
Etcher
Executor/Executrix
Forger
Former owner [for provenance]
Illustrator
Illuminator
Licensee
Licenser
Lithographer
Papermaker
Printer
Publisher
Signer [not a former owner; for provenance]
Translator
Type designer
Typographer
Wood engraver
[When translating these into relator codes, there also needs to be a category of Association for provenance, where the relationship is not that of signer, former owner, etc. There is no acceptable relator term for this category.]
II. RELATOR CODES
A recent LC proposal to MARBI (accepted in principle at the summer 1979 ALA annual meeting) outlined a system to provide an indication of the functions of any personal or corporate access point in the bibliographic record by a so-called relator code. This code would be a two-character numerical code taken from a fixed list and could, in the future, replace the relator term completely. The advantages of this technique were seen to be: 1) an increased efficiency in machine processing, since a two character numeric is easier to process than a variable length character string; 2) the possibility of international exchangeability, since the numerics could generate print constants (i.e. the verbal relator terms themselves) in whatever language was needed; and 3) a reduction in the number of specialized added entry fields needed to satisfy specialist needs in the MARC format.
Only four relator codes have been established so far, namely those that were needed for the new technical reports subset of the Books format. These were: performer, originator, monitor, and funder. These will be coded in the $4 subfield (a numeric rather than alphabetic subfield) in the 1xx and 7xx fields.
The IRLA Ad Hoc Committee believes that such a concept could be used by rare book and special collection libraries either in addition to or in lieu of verbal relators. The advantages cited above probably would hold true for the rare book community also.
At present, however, there are at least two difficulties that need to be considered. One is the lack of a standard set of relator terms to which coded values could be assigned. The first half of this proposal makes recommendations in that area. After such a standard list has been developed, the terms should then be submitted to the Library of Congress for coding.
A second difficulty might come into play if these codes were implemented to be used in lieu of the actual relators without appropriate provision for the display and output of print constants (i.e. the verbal relators). Therefore, the Committee urges the bibliographic utilities to plan to provide print constants when implementing the relator code subfield.
In the future, as rare book and special collection libraries move to machine-based catalog and retain "custom" processing services that tailor output products to the libraries' own specification, it will be possible to use relator codes flexibly, thus making unnecessary the storing of verbal relators in the bibliographic records themselves.
SUBJECT: Additions to the Expanded Library Character Set
TO BE SUBMITTED TO: The Library of Congress, RBMS Standards Committee
To accommodate detailed bibliographic description in MARC records (cf. the second example in AACR2, 2.18B) some additional special characters are needed in the Expanded Library Character Set. Characters which are lacking, and which have no widely recognized equivalents among the already-defined characters, are the following:
1) superscript a and b (to designate the columns of a page)
2) superscript r and v (to designate recto and verso)
3) Greek letters pi and chi (to stand for unsigned gatherings)
4) the vertical bar (to indicate line endings in title transcription)
If it is found to be impossible to add these to the character set in the near future, the Committee recommends that the following equivalents be made in interim standard:
1) superscript a and b: record in lower case on the line
2) superscript r and v: record in lower case on the line
3) Greek letters pi and chi: use empty square brackets, thus: [ ]example: [ ][superscript8], A-Z[superscript8], Hh (note that superscript numerics are available in the Expanded Library Character Set)
4) the vertical bar: use a double slash (//)
The Committee thinks that the above are workable substitutes.
SUBJECT: A Flexible Filing Override Mechanism
TO BE SUBMITTED TO: The Library of Congress, MARBI, the networks
Research libraries in general and rare book and special collections libraries in particular increasingly will have to produce sorted output files from machine-based catalogs. The need for a filing override mechanism can be seen in the following:
1) Most rare book libraries find it a matter of importance to have lists and catalogs ordered in a way that they consider to be bibliographically correct; the precise kind of sorting desired, however, may vary from list to list or even within one list.
2) Special collections, because of the narrower focus and greater depth of their holdings, frequently have several bibliographic records with the same author, uniform title, and date of publication (representing different editions, issues or states). Without a mechanism to influence filing in a way that is appropriate to the material, most filing programs will simply arrange the records randomly below the date level. This would not be useful for most rare book and special collections libraries.
3) The information by which output files of rare and special materials must be arranged to be most useful to scholars and researchers may sometimes be present in the bibliographic record, but not in a regularized form that could aid in filing. However, more often than not the information that should determine the filing order of the records is completely extraneous to the bibliographic data present in the record. For instance, a standard bibliography may propose a particular ordering of all the editions and issues of a work based on the physical evidence of the volumes or on historical deduction.
4) Many rare book and special collections libraries will want to control the arrangement of non-alphabetic characters, or retain initial articles in access points—niceties that the largest research libraries have dispensed with.
In light of these requirements, the IRLA Ad Hoc Committee recommends that the Library of Congress and the networks develop a filing override mechanism. Such a mechanism should be standardized and made available in the MARC format for use by all libraries with computer-based systems. This would avoid the wasted time and expense—and frequently inadequate products—that will result from individually developed filing techniques or, what is more likely, from standard vendor sort/merge routines.
The British Library's Eighteenth Century Short Title Catalogue staff has shown a good deal of ingenuity in this area, and a similar—if perhaps more generalized—technique is needed in this country.
In the development of such a mechanism, we believe that the following points should be taken into consideration:
1) the technique should be capable of being applied flexibly to any field upon which sorting may be required;
2) the technique should be capable of being applied manually either at the time a bibliographic record is created, or later on;
3) the technique should be system-independent (that is, transferable) so that it may be useful to libraries without regard to the source of their computer processing services.
SUBJECT: Draft Standard for Terms Indicating Genre; Draft Standard for Terms
Indicating Illustration/Graphic Technique
TO BE PRESENTED TO: RBMS Standards Committee
Many rare book libraries and special collections maintain special files to achieve access to materials that display certain physical or intellectual features that may be extraneous to their subject content. In manual systems this has been easy to accomplish, simply by overprinting the desired access terms and by establishing distinct card files for various classes of materials. In machine-based systems this capability has been difficult to achieve, primarily because the MARC format has not included access fields for such specialized information.
The IRLA Ad Hoc Committee has prepared proposals to the Library of Congress and MARBI for the definition of such access fields within the MARC format (see Proposals One and Five).
Meanwhile, it is imperative that a standard vocabulary of access terms be developed, both for copy-general and copy-specific information. A controlled access language will be of use in the short run for the interchangeability of cataloging records (chiefly for more modern rare books) and in the longer run for the production of machine-based union catalogs and for remote access to the local files of individual institutions.
The draft lists of proposed access terms which form part of this proposal have been developed in consultation with a number of rare book and special collection librarians and are put forward as working drafts towards a standard. The Ad Hoc Committee recommends that the RBMS Standards Committee study these drafts with a view to developing lists that can then be published and maintained by an agency within ALA.
In addition to the two areas addressed by the Ad Hoc Committee (i.e. genre and techniques of illustration), there are several more areas in which standardized access languages might well be developed, including publishing/bookselling, binding, typographic, papermaking, and provenance information (cf. Proposal Five). The Ad Hoc Committee recommends that the RBMS Standards Committee study the feasibility of developing and maintaining lists of access terms in these areas.
The Ad Hoc Committee realizes that not every library will wish to make use of such controlled access language. This attempt at standardization is meant to address the needs of those libraries that have in the past or will in the future want to provide these kinds of access to their collections.
DRAFT THESAURUS OF GENERIC TERMS FOR OLDER MATERIALS
rev. 12/1/79
Advertisement
Almanac
Alphabet
Annals
Annual
Anthology
Army/navy list
see Register (military)
Atlas
Autobiography
Ballad
Battledore book
see Hornbook
Bibliography
Bill (legislative)
Bill or mortality
Biography
Blank form
see Form (blank)
Block book
Book of hours
Bowdlerized book
Braille book
Breviary
Broadside
x Handbill
Calendar
Calling card
Captivity
Carrier's address
Catalog
Catalog (auction)
Catalog (Bookseller's)
Catalog (exhibition)
Catalog (publisher's)
Catalog (library)
Catalog (trade)
Catechism
Chapbook
Chronicle
Cipher(s)
see Code(s)
Code(s)
x Cipher(s)
Coloring book
Concordance
Conjuring book
Coranto
Courtesy book
Curriculum
Dialogue
Diary
x Journal
xx Memoir
see also Memoir
Dictionary
Dime novel
Directory
Dissertation (academic)
Division list (legislative)
Drama
see Play(s)
Elegy
Emblem book
Ephemeris
Epithalamium
Erotica
Essay(s)
Eulogy
Exhibition (museum)
Facetiae
Facsimile
Festival book
Festschrift
Fiction
Flora
Forgery
Form (blank)
x Blank form
Formulary
Gazetteer
Genealogy
Grammar
Greeting card
Guidebook
Handbill
see Broadside
Handbook
Herbal
Hieroglyphic(s)
History
Hornbook
x Battledore book
Hymnal [texts only]
Index
Indulgence(s)
Instruction book/textbook
Juvenilia
Journal
see Diary
Journalism
see Reporting
Laws, statutes, etc.
Lectures
Letter(s)
Libretto
Liturgy and ritual
Masque
Mazarinade
Memoir
xx Diary
see also Diary
Miniature book
Miracle play
x Mystery play
xx Passion play
see also Passion play
Missal
Monologue
Mystery play
see Miracle play
Newsletter
Novena
Oral history
Order in council
Oration(s)
Pantomime
Pasquinade
Passion play
xx Miracle play
see also Miracle play
Pattern poetry
Penny poem
Petition
Pharmacopeia
Phrase book
Play(s)
x Drama
Playbill
Poem(s)
Poll book
Prayer book
Prices current
Primer
Proclamation
Program
Prompt book
Prospectus
x Subscription proposal
Rappresentazione
Register (military)
x Army/navy list
Regulation(s)
x Journalism
Roadbook
Satire
Scroll
Sermon
Session laws
Song(s) [texts only]
Song sheet [text only]
Standing orders
Subscription list
Subscription proposal
see Prospectus
Thesaurus
Trade card
Travel book
Trial proceedings
Type specimen(s)
Vocabulary
Volvelle
Voyage(s)
Writing book
Yellow back
DRAFT STANDARD FOR TERMS RELATING TO ILLUSTRATION/GRAPHIC PROCESSES
(working draft)
Many of the following processes could be done in color, which may be indicated by the addition of the work color to a term, e.g., "Lithograph, color". Some of the following terms already have "color" incorporated into them because of the importance of color in that process.
The addition of the work colored to any of the following denotes the addition of hand coloring.
These terms would be entered in the field recommended in Proposal #5 (which for the time being has been labeled 755, though the particular tag number may be changed if and when it is accepted into the MARC format).
Asterisks denote little-used processes with proprietary names.
I. Non-photographic processes
A. Relief
1. Relief cut
Woodcut
Metal cut
Chiaroscuro woodcut2. Wood engraving
Chromoxylograph
*3. Graphotype
4. Relief etching
*Chemitype
5. Stereotype
6. Electrotype
*Glyphograph
*Kerograph
*Gypsograph7. Linocut
B. Intaglio
1. Drypoint
2. Engraving
Steel engraving
Nature printing3. Mezzotint
4. Etching
Soft-ground etching
Stipple engraving
Aquatint
Sugar aquatintC. Planographic
1. Lithograph
Pen lithograph
Wash lithograph
Scratch lithograph
Transfer lithograph
Anastatic printing2. Zincograph
3. Tinted lithograph
Lithotint
4. Chromolithograph
5. Monotype
6. Counterproof
D. Other
1. Silkscreen
2. Stencil
3. Natural illustration
II. Direct photographic processes
A. Photograph
B. Daguerrotype
C. Calotype
D. Carbon print
Autotype
III. Photomechanical printing processes
A. Intagio
1. Photogalvanograph
2. Heliogravure
*Positive etching
3. Photomezzotint
4. Photogravure (screened or unscreened)
Rotogravure
*"Rembrandt" prints
Color gravureB. Planographic
1. Collotype
*Heliotype
*Albertype
Ink photo2. Photolithograph (unscreened)
Photozincograph
Photo-chromolithograph3. Photo-offset lithograph (screened)
Color photo-offset lithograph
C. Relief
1. Woodbury type
2. Line block
*Hyalograph
3. Halftone
Color halftone
SUBJECT: Standardized Citation Forms for Bibliographical References to be Used
As Access Points
TO BE PRESENTED TO: The Library of Congress, the RBMS Standards Committee
Traditionally libraries have given bibliographic references in the note area of cataloging records. For older imprints this can be an especially important form of identification because LC card numbers and ISBNs are not generally available for works published before 1900. In addition to simple identification, bibliographic references are often used to indicate the location of a fuller description of an item than the cataloger can supply. AACR2 provides for the use of such references (cf. 1.7B15, 2.18C).
The development of machine-based processing raises the possibility that citation notes will be able to be used as access points, both for searching data files for specific items and for creating output lists. A recent addition to the MARC format for books defines an optional structured field for such citations. (This format change was accepted at the summer 1979 ALA annual meeting but it has not yet been issued as an addendum to the MARC format, nor implemented by any of the networks at the time of this writing.) The Ad Hoc Committee gave advice toward the development of this format change and strongly endorses the implementation and use of the new field.
The Committee has several further proposals to make in this area:
1) We recommend that the field (tag 510) be used whenever a simple reference from a numbered bibliography is to be recorded. If a citation needs to be amplified, a general note in the 500 field should be given instead of or in addition to the 510 entry.
2) Following AACR2, such citations should always be made for incunabula.
3) For other printed works, the Committee recommends that citations always be given if the work is found in one of the following general bibliographies:
Bristol, R. P. Supplement to Charles Evans' American bibliography
Evans, C. American bibliography …
Pollard, A. W. & Redgrave, G. R. Short-title catalogue …
Pollard, A. W. & al. Short-title catalogue … 2nd ed.
Wing, D. G. Short-title catalogue …
If all libraries that are members of a bibliographic utility recorded citations for these works in the 510 field, access to bibliographic records for older materials (especially those with long and non-distinctive titles) would be greatly improved.
4) The citation of other bibliographies would be optional—though strongly recommended when a particular book is described in a generally recognized, standard bibliography.
When it is desirable to cite a more specialized bibliography that describes an item contained also in one of the five bibliographies named above, it is strongly recommended that the specialized bibliography be cited in addition to the general bibliography listed above—not instead of it.
5) A standard list of brief citation forms should be developed, maintained, and published by the RBMS Standards Committee (or another unit within ALA). The IRLA Ad Hoc Committee has prepared a working draft of approximately 250 bibliographies that are considered standard in the field. We will forward this list, along with suggested citation forms, to the RBMS Standards Committee as a starting point for their work on this task.
The list maintained by the RBMS Standards Committee should not be considered to be a list of recommended bibliographies; rather, it should be a record of agreed-upon citation forms for whichever bibliographies might be cited. Further, only numbered bibliographies should be included on this list. Unnumbered bibliographies do not provide unique access to particular bibliographic items.
Unnumbered bibliographies could, of course, continue to be cited in the 500 field.
SUBJECT: Library of Congress Local Data in Distributed Records
TO BE SUBMITTED TO: Library of Congress
The IRLA Ad Hoc Committee recommends that the Library of Congress reformat its distributed bibliographic tapes so that information pertaining to LC's copy of a work is clearly identified as being copy-specific data; that is, we recommend that LC make use of—and distribute—590 fields (instead of 500 fields) for LC local notes falling into the categories described in AACR2 2.7B20-21.
Most libraries would like to see LC's local notes on the distributed tapes—they are valuable as information about LC's collections. However, not all libraries will want to analyze each 500 note in a record to ascertain whether or not it may apply to LC's copy only or to all copies of an item. By using the 590 field for copy-specific data, local notes would be flagged for evaluation.
The Ad Hoc Committee recommends that those libraries that are members of bibliographic utilities also pay close attention to this distinction when creating original bibliographic records.
SUBJECT: Authority Information For Retrospective Name Authorities
TO BE SUBMITTED TO: The RBMS Standards Committee, the Library of Congress, the
Council on Library Resources
I. The impact of AACR2 on retrospective collections
Much thought is being given at present to the impact of AACR2 on the catalogs of the large research libraries. Thus far, the consensus seems to be that the effect of the new code will not be insignificant, not only financially, but in terms of staff time lost in re-doing authority work and educating users. It is likely that the impact of AACR2 on rare book and special collections libraries will be much the same.
However, rare book and special collection libraries will have at least two advantages:
1) Corporate names, which will often be quite different under AACR2, seem not to be as common in rare book libraries as in the large research libraries.
2) The introduction of new AACR2 headings will take place in rare book libraries over a longer period of time than in the largest libraries because of the lower rate of acquisition in rare book and special collections.
Beyond these two factors, it has become clear that many of the new AACR2 forms will be able to be interfiled with the older headings without too much difficulty. Those AACR2 headings which do not readily interfile with the older version of the headings might need to be handled by splitting the files and connecting them with see-also references, though this detracts somewhat from the usefulness of the card catalog.
In general, however, the prospect is that rare book libraries will not need to close their catalogs in order to adopt AACR2 unless they want to.
II. Availability of AACR2 authority information
The Committee believes that, for various reasons, rare book libraries will want to adopt AACR2 (and the LC manual described in the following proposal)—along with most of the other research libraries in the United States, Great Britain, and Canada—on January 2, 1981. The availability of AACR2 name forms may be the chief barrier to the complete adoption of the code.
In theory, each library can create is own AACR2 forms of names at the time
of cataloging or re-cataloging. The new code, and LC's interpretation of it,
specifies that forms of names are to be based on the need in individual files
to distinguish one name from another (cf. AACR2 22.16A).
In addition, AACR2 calls for headings based on the form of name by which an author "is most commonly known", or the name appearing most frequently in the person's works. Most libraries, including the Library of Congress, will base the decision about predominance on the forms of name on the works in their own collections. (Further, the Library of Congress is basing its decisions about predominance on the percentage of cards under a given heading in its official catalog that have author statements, and author statements tend to be rare on older LC catalog cards.) All this could result in a large margin for disagreement about the proper AACR2 forms of name. In practice, however, it is likely that the Library of Congress form of name will remain the de facto standard as it has through the years and as economic realities suggest it should. As a library eventually learns of the LC form, it would in most cases routinely change its file to match it.
It may be some time before LC's AACR2 authorities are available, however. The Library of Congress will, under its current plans, provide name authorities for retrospective names other than those already represented in the (post-1968) MARC data base only as they happen upon them in the course of cataloging newly acquired materials. While this may not be a problem in the case of the popular earlier writers—who are reprinted with some frequency—most of the authors who wrote before the year 1900 will probably not again come up in the course of LC's cataloging for many years, if ever.
While this may at first seem like a reprieve to rare book libraries, it is not really. Any library expecting to use RLIN, OCLC, or any other network for original input of retrospective material after January 2, 1981, probably will be obliged to use AACR2 name forms in order to meet system-wide standards.
Thus, libraries that are committed to building machine-based catalogs face the perplexing task of trying to learn if LC has yet established AACR2 forms for older names and, if not, how it should be established according to AACR2. If a library establishes an AACR2 form on its own, this decision will, as indicated before, have no validity for any other library, and may well be invalidated if and when the Library of Congress does eventually establish it.
III. The situation for current large-scale retrospective cataloging projects
A number of important efforts in the area of retrospective bibliography have begun in the last year or so. They include the Eighteenth Century Short Title Catalogue (at the British Library and at Louisiana State University), and its counterpart for U.S. 18th century materials, the North American Imprints Program (at the American Antiquarian Society). Other projects are in the planning stages and most involve the creation of large-scale machine-readable data bases.
The two large retrospective projects (ESTC and NAIP), which are committed on paper to the use of AACR2, face the same problems with authorities as do individual libraries. At least one, ESTC, has reportedly decided to postpone the attempt to assign AACR2 headings until some future date—the "editorial stage". (This will be possible because the ESTC is creating its own machine-readable file which, at least for the time being, will not have to be in complete conformity with AACR2.)
IV. LC's role in the provision of authority information
In the best of all possible worlds, libraries and special projects involved in cataloging or recataloging would simply make use of CRT terminals to access LC's authority file on-line. If a needed name were established there, it could be used immediately. If it were not, a request would be lodged with LC to establish it as soon as possible and transmit it back to the requesting library.
Although RLIN and OCLC have both announced plans to mount the LC authority file on-line, no system is anticipated by which LC would establish authorities on request.
At present, the Library of Congress is involved in cooperative name authority projects with three outside agencies: the U.S. Government Printing Office, Northwestern University, and the University of Texas. The sense that one gets now from LC, however, is that it is reluctant to take on too many more of such cooperative relationships—though it might if sufficient funding became available. For the moment, LC's own staff for inputting, processing, and verifying retrospective name authorities is badly behind schedule.
V. The need for a national or multi-institutional program for creating and distributing retrospective name authorities
A national or multi-institutional system is needed to generate and make available name authorities in the MARC authority format. Some planning in this area has been undertaken and the Council on Library Resources is now engaged in developing specifications for such a system.
If such a system were to be developed, some of the burden could be lifted from LC's shoulders, especially with regard to retrospective authorities. For a national authority system to be of greatest benefit to the rare book and special collections community, it should have at least the following features:
1) The capability for de-centralized input of AACR2 authorities, preferably
in an on-line mode
2) A mechanism for establishing a provisional AACR2 form for use until such
time as LC supplies its version of the heading
3) A mechanism for quality control and authentication outside of LC
4) A method for LC to consult and make use of authority work done by other libraries.
This last feature could be especially important in light of the fact that some
rare book libraries have larger collections of the works of earlier authors
than LC does and, hence, more works upon which to base decisions about predominant
forms of name.
VI. Recommendation
The Ad Hoc Committee strongly endorses the concept of a nation-wide authority system and urges the Council on Library Resources, the networks, and the Library of Congress to move quickly to develop one. The Committee recommends that the RBMS Standards Committee follow these developments closely with a view to supporting the implementation of such a system and informing the rare book and special collections community about progress in this area.
Further, the Ad Hoc Committee recommends that funding agencies for retrospective cataloging projects (CLR, NEH, etc.) consider this problem when entertaining proposals for new retrospective projects. Without such a system of authority cooperation, the updating of machine-based files created with out-of-date, pre-AACR2 headings will become an additional cost to be figured into the total expense of such retrospective projects.
The Committee also asks the Library of Congress and the bibliographic networks to propose some interim solution to the problem of retrospective authorities that have not been assigned AACR2 forms by 1981, when the new code is adopted.
SUBJECT: The Library of Congress Manual for the Cataloging of Older Imprints
TO BE SUBMITTED TO: The RBMS Standards Committee, the Library of Congress
When the IRLA Ad Hoc Committee met in March, 1979, it found itself in agreement that the early imprints section of AACR2 was not an adequate guide for the cataloging of older materials. On the other hand, the Committee was encouraged by the latest draft of the new international Standard Bibliographic Description for Antiquarian materials, ISBD(A), which was studied by individual members and by a de facto subcommittee made up of members of the "Athenaeum Group".
The Committee asked the Library of Congress for advice on this matter. As a result of the Committee's interest, in part, LC agreed to prepare a manual for the cataloging of older imprints which would fill out and perhaps adjust chapters 1 and 2 of AACR2 according to some of the provisions of ISBD(A).
This manual is now in draft form and should be published in a "first edition" during the first half of 1980. Libraries wishing to see a draft of the manual may request one from the Office of Descriptive Cataloging Policy, Processing Services, Library of Congress, Washington, D.C., 20540.
The Ad Hoc Committee supports the principles behind the manual and sees it as fulfilling a need in the rare book and special collections community. The Committee extends its thanks to the Library of Congress for this important service.
The Committee recommends that, when the manual has been completed, it be adopted for use by rare book libraries and special collections in lieu of the first two chapters of AACR2 for cataloging, beginning in 1981.
The Committee also recommends that the RBMS Standards Committee, national funding agencies, and parent institutions that have special collections associated with them, encourage the use of this manual—in conjunction with AACR2—as a way of promoting standardized cataloging practices in rare book and special collections libraries.