Cataloging & Classification Quarterly

Volume 29, Number 1-2 2000


Special Guest Editorial 

THE LCSH CENTURY:

 A BRIEF HISTORY OF THE LIBRARY OF CONGRESS SUBJECT HEADINGS, AND INTRODUCTION TO THE CENTENNIAL ESSAYS

By Alva Stone

             The summer of 1998 witnessed a different sort of celebration in Washington, DC than the usual Fourth of July fireworks on “The Mall.”  At the Library of Congress speeches were made, awards presented, and a cake designed to look like a big red book was ceremoniously sliced and consumed by invited guests.[i]  It was the 100th birthday of one of LC’s perennial bestsellers, the Library of Congress Subject Headings (LCSH).

            Technically speaking, LC’s publication of its subject headings list did not really begin in 1898.  That was instead the year in which the Library of Congress converted from an author- plus a classed-catalog to a dictionary catalog, which incorporated author, title, and subject entries into a single file.  For its first subject headings, LC used the American Library Association’s List of Subject Headings for Use in Dictionary Catalogs (1st ed., 1895; 2nd ed., 1898), to which the LC catalogers added new headings as they were needed.  Meanwhile, with the ALA community clamoring for greater standardization as well as more cooperative cataloging, there arose a ready and appreciative market for the LC catalog Card Distribution Service, which began in 1902.  To the larger and “outside” library world, then, that was when subject headings formulated and assigned by LC began to be noticed and utilized; hence, the year 2002 might be a reasonable candidate for the centennial of LCSH.  The first actual printing of Subject Headings Used in the Dictionary Catalogues of the Library of Congress (later to be titled Library of Congress Subject Headings) began in the summer of 1909, was issued in parts, and was completed in March 1914.  Based strictly on first-edition data, therefore, one could argue that the 100th year of the LCSH will not occur until 2009, or even 2014.  However, there was some publication history prior to 1909.  As shown in the Chronology of Official LCSH-Related Publications appended to this essay, in 1906 the Library of Congress published a preliminary list of subject subdivisions for place names and of subjects that could be subdivided by place.  With all of these different dates to mark a beginning, perhaps it is the most sensible thing to choose 1898, assuming this to have been the year that the very first modification to the ALA List was made for LC use, and hence, the evolution of the Library of Congress Subject Headings was begun.  At any rate, the span of dates certainly coincides closely enough to those belonging to the 20th century, and that is why we have chosen for the present collection the title, The LCSH Century.

In this collection of papers there are frequent references to the guidelines for subject access delineated by Charles Ammi Cutter.  This is remarkable because his Rules for a Dictionary Catalogue was originally published in 1876.  A textbook on cataloging published in 1963 attested that “nothing has been written since that is any clearer than Mr. Cutter’s explanation of this aspect of cataloging.”[ii]  Certainly, we are indebted to Cutter for the concepts of direct entry, the use of natural language, and the syndetic structure (cross-references) meant to compensate for the "scattering" of related topics in the dictionary catalog.  And yet, the influence of Cutter on the LCSH was not acknowledged by the Library of Congress until 1972, when that acknowledgment was made unofficially by the Chief of the LC Subject Cataloging Division.[iii]  However, during the early development of the LCSH,  J.C.M. Hanson at the Library of Congress adopted a more pragmatic approach than what Cutter had intended, resulting in some inconsistencies in the forms of subject headings and the choice of cross-references.[iv]  Inverted headings, meant to bring some subjects together in logical groupings, were established in the early LCSH lists to a much greater extent than had been recommended by Charles Cutter.

It has also been noted that, whereas Cutter’s Rules eventually led to much greater elaboration of rules--and examples--in the area of descriptive cataloging, in subject cataloging the efforts resulted in lists rather than codes.[v]  Paul Dunkin believed that our excessive focus on the lists and their growth actually hindered the further development of the basic philosophy of the rules for subject headings.[vi]  Perhaps this explains why throughout the century the library community made numerous requests for explanations of the fundamental rules or theory behind LC’s subject heading lists.  Let’s continue to sketch the history of LCSH, and see how this theme arises again and again.

            The Library of Congress developed the LCSH for use in its own catalogs.  The list was also considered appropriate for the very largest public libraries, some colleges and many university libraries.    Smaller libraries continued to use the ALA List, last published in a 3rd edition (1911), and later, the Sears List of Subject Headings, which first appeared in 1923.  LCSH was criticized  for its aforementioned inverted headings, as well as its subdivided headings, which seemed to be a carryover from the days of the alphabetico-classed catalog.  Also, in the 1920’s and 1930’s the LC list lacked some of the syndetic structure found in ALA or in Sears.  (It was not until 4th LCSH (1943) that “refer from” references were included, in addition to the “see” and “see also” cross-references, but printed in a separate volume; the 5th LCSH (1948) incorporated the “refer from” references in the main list.[vii])  Nevertheless, by the 1930’s more and more libraries found themselves doing a retrospective conversion of subject headings.  They abandoned their homegrown lists, the outdated (1911) ALA List, or Sears, and converted to the LCSH for several compelling reasons: as their library collections grew, the desire for more precise subject headings increased; it was no longer economically feasible to continually revise subject headings appearing on LC catalog cards; and, the LCSH was the only general list that made a consistent practice of keeping up to date by creating new subject headings for new topics.[viii]  LC had been publishing supplements from time to time, but many outside catalogers reported using the card distribution service as an informal notification about the newest subject headings.  (In the 1940’s a regular column in Library Journal also listed the newest LC subject headings, until finally, beginning in 1948, LC started issuing its printed supplements on a quarterly basis.) 

Simultaneous to the increased adoption of LCSH in the 1930’s, the library community began to demand a textbook on understanding and applying the LCSH.  As a compromise, in the early 1940’s “notes on L.C. methods and authorities used to establish certain types of headings, geographic headings, names of Indian and African tribes and languages, were obtained” and published in the aforementioned column of Library Journal, largely through the efforts of the ALA Committee on Subject Headings and the “friendly cooperation of the Library of Congress.”[ix]   Finally in 1946 a treatise on subject headings by Julia Pettee, librarian of the Union Theological Seminary in New York, was published; later that same year the Library of Congress indicated that it was also working on a manual.[x]  The latter was not published until 1951.  Written by the Chief of the LC Subject Cataloging Division, David J. Haykin, it was a 140-page book that basically reiterated most of Cutter’s principles and tried to explain or rationalize some of the seeming inconsistencies in LCSH. (Haykin’s work is discussed in more detail by a couple of the authors in this collection.)  Neither of these books, however, provided a true manual of practice for using and applying the LCSH.[xi]  Another controversy that flared up in the literature of the 1940’s was a debate over whether or not published subject bibliographies were superior to catalog subject headings as a means of access to subject information.[xii] 

At mid-century there was a growing dissatisfaction with LCSH.  Libraries were feeling the pressures of the increased publication and acquisition activities of this postwar period.[xiii]  There was a need to improve subject access but also to make subject cataloging less costly, goals which are often in conflict with each other.  Critics complained that “the basic rules and techniques for the construction of subject catalogs and the development of subject heading lists have undergone virtually no change in the last 75 years,” and that such practices did not reflect any clear understanding of function or purpose.[xiv]  One trend that gained popularity in the 1950’s and 1960’s was a move to the divided catalog.  Considering the complexity of catalog filing rules, along with the ever-lengthening, more precise subject headings, the rise of the “subject catalog” (separated from the author/title catalog) undoubtedly did benefit the library staff, but some librarians worried that it often diminished access--especially access of the serendipitous sort--for the catalog user.  Pettee observed the irony that libraries had gone “full circle”—from author and classed catalogs to dictionary catalogs and then back to divided catalogs.[xv]  (Interesting comment, but, an alphabetical, specific-entry subject catalog is not the same as a classed catalog.)  As another result of the Cold War period’s rapid expansion of knowledge and information, many catalogers felt that the Library of Congress was too slow in adopting new subjects or revising antiquated terminology.  Nevertheless, LC did demonstrate a greater sense of responsibility in educating the library community during this time, principally by including in its Cataloging Service Bulletin explanations and guidelines on the use of certain types of subject headings or subdivisions, as well as notices of major revisions.  The reporting concerning LCSH printed in Cataloging Service Bulletin has continued and increased from the 1950’s to the present day.  In 1957 the 6th edition of LCSH was published; its innovation was the increase from two to three columns of subject headings per page, a format which would remain constant through to 1999.

Another development of the latter half of the 20th century was the experimentation with automation as a means for resolving the problems of subject access; this resulted in several efforts to create or discover a viable alternative to the LCSH.  In the 1950’s there was much discussion of “uniterms” (later called “descriptors”) designed for coordinate indexing via machine manipulation.  A proliferation of keyword indexes (KWIC, KWOC, etc.) was witnessed in the 1960’s-1970’s, succeeded in later years by the keyword/Boolean search techniques utilized in most online public catalogs (OPACs).  Thesauri became popular, particularly in special libraries, during the same period.  And, in the 1970’s-1980’s a powerful string index language, PRECIS, was used by the British Library (and even studied briefly by the Library of Congress as an augmentation to LCSH).   These new techniques, experiments or functionalities proved to be useful, to a point.  But ultimately it was generally concluded that none of these means for subject access could fully match the effectiveness of a precoordinated, controlled vocabulary such as the LCSH system.[xvi]  In the 1980’s and 1990’s the emphasis shifted to enhancing, rather than eliminating, the LCSH access, e.g., through mapping to other vocabularies or linking to classification numbers.

 Social changes also had an impact, resulting in complaints in the 1960’s-1970’s about bias in the LCSH.  Gradually, and often after much study and input from the ALA Subject Analysis Committee or other special-focus groups, the Library of Congress did revise many of the offensive terms or subtly-discriminatory wordings of headings related to racial, ethnic, religious, or gender groups.  More technical changes occurred in the 1970’s: the LC practice of using free-floating subdivisions was named in 1974 and later was expanded to new types or categories of headings; the 8th LCSH (1975) grew to two volumes and adopted its modern title, Library of Congress Subject Headings; in 1976 LC discontinued  its practice of dual methods for geographic subdividing ("direct" and "indirect" instructions); and the LCSH began to print some headings that had previously been omitted (legendary and fictitious characters, works of art, chemicals, ancient cities, structures, buildings, parks and reserves, and others).  Also in 1976, the Library of Congress offered the list in a microfiche version for the first time.  This made it affordable for libraries desiring to have multiple copies; the microfiche was portable, and very convenient for libraries that had converted to or were planning for COM (computer-output microform) catalogs. 

Practitioners and professors writing in the literature began to propose the creation of a subject cataloging “code” to do for subject analysis and practice what the Anglo-American Cataloguing Rules 2nd ed. (AACR2) did for descriptive cataloging.  Although technically not a general code, the expressed need was partially fulfilled by the Library of Congress in 1984 through the publication of its Subject Cataloging Manual: Subject Headings (SCH:SH).  It was necessary, it may be argued, to publish such a manual in order to promulgate the rules for assigning the numerous and complicated free-floating subdivisions to be used under various categories or pattern headings.  But the manual provided much more information, including LC policies and procedures, scope notes, and detailed guidance on applying subject headings for particular disciplines or forms of material.  For the first time, “outside” catalogers were able to read the same memoranda, or instruction sheets, available to LC catalogers, and thus acquire the opportunity to become more consistent and improve the quality of the LCSH strings that they assigned to original-cataloging records for the bibliographic utilities.  From this point forward, the LCSH ceased being simply a list, and became the LCSH system; a cataloger could no longer select or verify a subject heading using the LCSH alone, but instead must use the list in conjunction with the Subject Cataloging Manual: Subject Headings.  The SCM:SH has been issued in a looseleaf format, to allow for the ease of updating (generally there have been two releases of added or revised pages each year).  Updating of the LCSH also became more timely during this period.  Subscriptions to the “Weekly Lists” of new and changed subject headings and references were first offered in 1984.

 In 1985 the Library of Congress began to regularize its practices regarding the provision of cross-references for new subject headings, and hoped to conduct a review and revision of the syndetic structure under older headings, as time allowed, to bring them into conformity with the new criteria.  The criteria called for making links between subject headings only under certain circumstances, e.g., when the concepts or objects represented by the terms had a whole/part or genus/species relationship, among other rigorous conditions.  At this time, the LCSH converted its notation devices to those most often used in thesaurus construction, i.e., BT for “broader term,” NT for “narrower term,” and so forth.  The late 1980’s also saw the introduction of two new formats for LCSH: one was the CDMARC Subjects for use on personal computers with CD ROM drives; the other was the Subject Authorities in MARC format issued on magnetic tape (later also available via FTP transfer).  The latter product enabled bibliographic utilities such as OCLC and RLIN to load the LCSH authorities in their databases and make searching of them available to their member libraries.  While these improvements in the sharing of guidelines and dissemination of additions and changes were occurring, the LCSH continued to grow, expanding to three volumes beginning with the 11th ed. (1988).  It was also in that year that new editions of the printed LCSH began to be published on an annual basis.

The 1990’s ushered in a new era of cooperation, seeming to suggest that the Library of Congress no longer took a proprietary view of the LCSH, but rather, was finally acknowledging its responsibility to all users of the LCSH system.  In May 1991, LC hosted an invitational Subject Subdivisions Conference, which resulted in many changes to LCSH, still ongoing and regularly summarized in Cataloging Service Bulletin.  Perhaps even more significant has been the development, under the auspices of the Program for Cooperative Cataloging, of the Subject Authority Cooperative Program (SACO).  The SACO process allows specially-trained “outside” catalogers to propose new subject headings based on their libraries’ new acquisitions or their users’ needs, rather than those exclusively of the Library of Congress.  Clearly, the Library of Congress has accepted its leadership role, while encouraging the wider cataloging community to contribute to the list.  That list, by the way, grew to four volumes by 1992, and reached five volumes beginning with the 21st ed. (1998).

There are many other issues and important moments in The LCSH Century which have not been addressed in this brief overview.  Some are described or critiqued in the papers of this collection.  The collection begins with three essays that focus on the history or theoretical aspects of LCSH practice.  Elaine Svenonius presents an analysis of the LCSH system as a language, describing some of its semantics and syntax as well as the pragmatics of specificity, and the effect of these on precision and recall.  In the paper by Hoerman and Furniss, many of these same principles are investigated in light of recent guidelines developed and approved by the International Federation of Library Associations and Institutions (IFLA). The authors discover a good deal of similarity between the IFLA Principles and those of the LCSH, but also indicate two major defects: the principles are often in conflict with each other; and, the principles are derived from tradition and practice rather than being based on users’ searching behaviors and needs.   The essay contributed by Hope Olson is a successor to earlier books and articles which blew the whistle on the apparent or subtle biases inherent in the terminology found in LCSH; she uses the theories of a postcolonial critic as a framework to argue that LC could choose to eschew the exclusion or marginalization of certain peoples or cultures (even though such biases may in fact be upheld by American literary warrant), and instead become a change agent aspiring to the enrichment of the lives of all library users.

Since these papers were all written in 1999, it is only natural that some focus be given to the prevailing form of current library catalogs.  The next three authors examine various aspects of the LCSH as it has functioned in the online environment.  Pauline Cochrane first covered this topic in two works published in 1981 and 1986, and now offers an “update” on the progress made and the issues that still remain in improving the LCSH for use in OPACs. A topic which has seldom been addressed in the literature is examined by Gregory Wool.  Because librarians have relinquished their control over the traditional catalog’s filing rules by meekly accepting the limitations imposed by programmers and designers of OPACs, the result has been an accidental (or, unintended) deregulation of standard arrangements of subject headings in the indexes.  Wool illustrates in particular how the Library of Congress Filing Rules, which arranges entries differently according to the punctuation that is present (e.g., commas for inverted headings, parentheses for qualified headings, etc.), takes advantage of the highly developed syntax and semantic features of the LCSH and result in logical groups that can benefit the searcher.  These structured but perceivably helpful collocations are lost, however, in most online catalogs that simply arrange LCSH strings in a word-by-word fashion, causing Wool to wonder if LC and the library community as a whole have virtually abandoned their faith in a precoordinated controlled vocabulary.  On a more positive note, Stephen Hearn reports on the potential for online validation and updating of LCSH strings; he examines three approaches being studied, and discusses improvements to the LCSH authorities, including the establishment of authority records for all types of free-floating subject subdivisions.

Next, we hear from practitioners about how they use the LCSH in specific environments or for particular forms of material.  Most significant of all, of course, is the perspective of the reference librarian who interacts with the user and interprets the LCSH for patrons every day.  Thomas Mann explains how he uses and teaches others to use the LCSH, for the most effective and thorough results in subject searching.  His techniques--illustrated by cogent examples--combine explorations of adjacencies, cross-references and precoordination in the LCSH “red books” with redirections and browsing within the subject catalog itself.  He also addresses the differences between keyword and LCSH searches, pointing out several advantages to the controlled vocabulary approach.

For the public library perspective, we could hardly choose a better representative than the Cleveland Public Library, whose catalog was among the five library catalogs examined in order to compile the first ed. (1895) of the ALA List later used as the foundation for developing the Library of Congress Subject Headings.[xvii]  Louisa Kreider looks at a sample of subject searches executed at the urban library, and finds that LCSH terms and phrases match fairly well on the vocabulary used by the users.  She also discusses the library’s participation in SACO (through which they have proposed new subject headings that have indeed been established in LCSH), the application of subject headings to works of fiction, and certain accommodations made locally for differences between the LCSH and the standard LC Annotated Card program headings (i.e., subject headings for juvenile literature).  Our entry in the special libraries category comes from the field of music.  Co-authors Hemmasi and Young present a comprehensive history of the development of music-related headings in LCSH.  They include discussions on the LC Music Division’s long-held preference for a classed catalog, problems such as a bias in favor of Western “art” (or, classical) music, and the potential for the ongoing Music Thesaurus Project to assist in future improvements to LCSH.  Perhaps most instructive to other “special interest” groups is the authors’ account of the proactive role taken by the Music Library Association in furthering the additions and revisions to LC subject headings for music.

 Although designed for library catalogs, the LCSH have also been used over the years in published or online bibliographies, abstracts, and periodical indexes too.  Miller and Kuhr describe the vocabularies used in periodical indexing at the H.W. Wilson Company, with examples of various adaptations to LCSH that have been necessary.  Their analysis highlights the fact that, whereas the LCSH is an excellent tool as a general system, for highly specialized indexes (such as those geared to children, or lawyers!) the LC terminologies are often either too specific or not detailed enough.  David Miller treats the issue of LCSH’s handling of access for different formats of materials, an area of concern for libraries with extensive audiovisual collections, or even, for those desiring better access to the genres/forms found in collections of literature.  Recent developments such as the implementation of a new MARC code for subject “form” subdivisions have made this a hot topic of the 1990’s.  His paper contends that, despite the LCSH inclusion of such terms, the form of a work is distinct from its subject; it is therefore questionable whether or not a subject heading list is really the proper venue for documenting practice.

Stepping aside from LCSH use in special environments or with particular forms of materials, the next three essays focus on the use of LCSH by libraries outside of the United States.    Heiner-Freiling reports the results of an IFLA-sponsored survey of national libraries, revealing that the LCSH has become the leading subject heading language for national bibliographies worldwide, most particularly in English-speaking countries, while it is also widely used in translation or adaptation into other languages.  Andrew MacEwan, of the British Library, describes some challenges discovered in the preliminary study that led to the MACS project (Multilingual Access to Subjects), in which four European national libraries will use authority records to link equivalent subject headings of the three predominant subject heading languages in English (LCSH), French (RAMEAU), and German (RSWK).  Although the context of European unity drives this particular desire to surmount linguistic barriers, there are wider implications for other regions or countries whose citizens use different languages.   A bilingual project on this side of the Atlantic is discussed by co-authors Quijano, Moreno and Figueroa.  First, they provide a survey of Spanish-language subject authorities that either translate from or link to LCSH terms, particularly those in Spain, Colombia, Chile, and Mexico.  Next, the authors describe the planning and initial efforts of a project overseen by the library of El Colegio de México which will translate LCSH into Spanish, for the purpose of supporting subject access to a projected online union catalog of ten (mainly university) libraries in Mexico City, and to offer such a tool to U.S. libraries serving bilingual communities.  An interesting sideline here is the fact that the catalogers at El Colegio de México have received NACO/SACO training from the Library of Congress, and intend to propose new LCSH terms or phrases reflecting concepts that are unique to the Latin American culture. 

In the concluding piece of this collection, Lois Mai Chan and Theodora Hodges present a summary of the early development and later improvements of the LCSH and the reasons for its success.  Then the authors speculate about the possible directions the LCSH might take in the future, with a particular view towards adapting the LCSH for optimum effectiveness in the “web” environment (information resources accessible via the Internet). 

The 100th birthday of the LCSH is the raison d’être of this collection of papers, similar to those books of essays written in honor of a scholarly and esteemed colleague.  As in any good Festschrift, one would expect to see included a comprehensive bibliography of the honoree’s own published works.  Here this takes the form of a chronology of LCSH-related publications issued by the Library of Congress over the years.  However, this is where our collection’s resemblance to a Festschrift ends.  For, typically the Festschrift is planned to celebrate the honoree upon her retirement, or near the end of her career.  It is our hope and expectation that this best of all generalized subject heading systems, the Library of Congress Subject Headings, will expand, evolve and endure for many years to come.


[i] Gail Fineberg, “Happy Birthday, ‘Big Red Book’: ALA Reception Honors LCSH Centennial,” Library of Congress Information Bulletin, 57 (1998), p. 202.

[ii] Thelma Eaton, Cataloging and Classification: An Introductory Manual, 3rd ed. (Champaign, Ill.: Illini Union Bookstore, 1963), p. 125.

[iii] Richard S. Angell, "Library of Congress Subject Headings--Review and Forecast," in Subject Retrieval in the Seventies (Westport, Conn.: Greenwood Press, 1972), p. 143.

[iv] A. C. Foskett, The Subject Approach to Information.  5th ed.  (London: Library Association Publishing, 1996), p. 336-7.  For a more complete history of the early development of the LCSH list, see: Francis Miksa, The Subject in the Dictionary Catalog from Cutter to the Present. (Chicago: American Library Association, 1983).

[v] Doralyn J. Hickey, “Subject Analysis: An Interpretive Survey,” Library Trends 25 (1976): 273.

[vi] Paul S. Dunkin, Cataloging U.S.A. (Chicago: American Library Association, 1969), p. 83-84.

[vii] Eaton, Cataloging and Classification, p. 116.

[viii] Harriet Dorothea MacPherson, Some Practical Problems in Cataloging (Chicago: American Library Association, 1936), p. 18.

[ix] American Library Association. Division of Cataloging and Classification, In Retrospect: A History of the Division of Cataloging and Classification of the American Library Association, 1900-1950 (Chicago: A.L.A., 1950), p. 11.

[x] The two manuals were: Julia Pettee, Subject Headings: The History and Theory of the Alphabetical Subject Approach to Books (New York: H.W. Wilson, 1946); and, David Judson Haykin, Subject Headings: A Practical Guide (Washington, D.C.: U.S. Govt. Print. Office, 1951).  Selections from both works are reprinted in: Lois Mai Chan, Phyllis A. Richmond and Elaine Svenonius, eds., Theory of Subject Analysis: A Sourcebook (Littleton, Colo.: Libraries Unlimited, 1985), p. 94-111.

[xi] Jessica Lee Harris, Subject Analysis: Computer Implications of Rigorous Definition (Metuchen, N.J.: Scarecrow Press, 1970), at p. 18: "... Miss Pettee was too occupied, as was Haykin later, with the need to arrange inherited practice into a system with some meaning."

[xii] See, for example: R.C. Swank, “Subject Catalogs, Classifications, or Bibliographies? A Review of Critical Discussions, 1876-1942,” The Library Quarterly 14 (1944): 316-322.

[xiii] Hickey, “Subject Analysis,” p. 273-74.

[xiv] Carlyle J. Frarey, “Subject Headings,” v.1, pt.2 of The State of the Library Art, ed. Ralph R. Shaw (New Brunswick, N.J.: Graduate School of Library Service, Rutgers—The State University, 1960), p. 57, 63.

[xv] Pettee, Subject Headings, p. 42.

[xvi] Indeed, as recently as 1998, in Martha M. Yee and Sara Shatford Layne, Improving Online Public Access Catalogs (Chicago: American Library Association, 1998), the authors provide much evidence that, for the best subject-searching retrievals, "the default subject search should be a search of the controlled vocabulary and that free text searching should be used only as a backup if controlled vocabulary searching fails, and even then it should be used as a gateway to the controlled vocabulary" (p. 155).

[xvii] Eaton, Cataloging and Classification, p. 113.


Back to Vol. 29 Nr.1-2 | CCQ Homepage | Tables of Contents |Informaworld |