CHAPTER TWO
0 INTRODUCTION
This chapter reviews the historical development of basic concepts in descriptive cataloguing and provides an analytical overview of selected cataloguing codes affecting the evolution of cataloguing principles. The main purpose is to explore the basic concepts underlying different codes of cataloguing rules and to provide an understanding of how these codes have developed in accordance with forms of the catalogue.
A clear grasp of the historical development of modern cataloguing codes and an understanding of the evolutionary growth of catalogues are both essential to an understanding of the relevance of cataloguing codes to the new electronic environment. In this context, another aim of this chapter is to acknowledge the role of the catalogue environment on the cataloguing rules and principles, i.e., to see how far the environment in which a catalogue is constructed might influence cataloguing principles. This chapter is intended to be an introduction to later chapters concerning the conceptual analysis of theories and principles of descriptive cataloguing with regard to the possible influences of the online environment on them. To understand the historical background and to examine the past is very important to any study of cataloguing principles and rules and to the future of code revision. To quote Svenonius (1989: 43):
A well-founded philosophy of catalog code development must take cognizance of the past, particularly of those factors which have affected and may continue to affect code design.
Although cataloguing codes have a history as long as the history of libraries, the focus of the present chapter is on the development of codes in the Anglo-American context throughout the nineteenth and twentieth centuries. Two non Anglo-American codes (i.e., the German code: Regeln für die alphabetische Katalogisierung (RAK) and the Japanese code: Nippon Cataloging Rules (NCR)) are also introduced to illustrate some other cataloguing traditions which have been different in some basic aspects.
The nineteenth century has been characterised as the formative era of modern catalogues and cataloguing codes. It was a time that revealed a keen interest in catalogues and the involvement of individual libraries and librarians in the compilation of cataloguing codes. According to Buckland (1988: 301-2),
The transformation of library cataloguing to its present form occurred in the nineteenth century, when it was argued that simple author access was not enough and that a different, more sophisticated, and more elaborate approach was needed.
Although the printed book catalogue was widely in use throughout the nineteenth century, the slip catalogue which was primarily created for the preparation of the printed book catalogue, was gradually found to be more flexible and more suitable. Book catalogues, which were found to be inflexible and in need of constant updating and reprinting, grew less popular in a short space of time. In addition, the limited number of access points per entry was a major disadvantage of book catalogues. By the end of the nineteenth century the printed card catalogue, although not in a standard form, became the alternative to the printed book catalogue and its use became widespread in the twentieth century throughout the United States as well as in many other countries (see section 1.4 in this chapter).
The two most common catalogue forms in the nineteenth century were the classified catalogue and the alphabetical catalogue. The classed or classified catalogue, which was a type of subject catalogue depending on a systematic classification, was used in many British, French and American libraries. In this form of the catalogue, entries were arranged in a systematic order to group together books with subject proximity. Authors and titles made up accompanying alphabetical indexes. Although the classified catalogue continued its development during the nineteenth century, the difficulties of maintaining it limited its application to a relatively small number of libraries. On the other hand, a major reason for the development of the alphabetical catalogue lies in the fact that alphabetical order has long been common knowledge and best suited for finding items in the catalogue. Most cataloguing codes developed by individuals or national agencies in modern times have been formulated for alphabetical catalogues rather than classified catalogues.
The nineteenth century was a time when the objectives of the catalogue became more clearly defined. As library catalogues had to be developed in accordance with the users' needs, a first step was to clarify the functions of the catalogue. Thus an essential principle on which rules for entries had to be based was the formulation of objectives for the catalogue.
The concept of one full record per book (i.e., one 'main entry' for the publication described) which was dominant in classified and alphabetical catalogues evolved into multiple entries in the dictionary catalogue. Although the dictionary catalogue had been first introduced by Andrew Maunsell in 1595 in the form of a simple list including authors, added entries (e.g., for translators), and subjects in a single alphabetical sequence (Osborn, 1963: 6), it had its flourishing in the nineteenth century, particularly through the works of Charles Ammi Cutter.
In effect, the development of our current cataloguing codes has its roots in the landmark works of a few librarians who formed the theoretical foundations of descriptive cataloguing:
The beginning of the development of modern cataloguing codes is generally attributed to Sir Anthony Panizzi (1797-1879) whose influence in descriptive cataloguing in general, and the theory of cataloguing in particular, is well known to the profession. The British Museum code of 1841 had its roots in preceding rules, and in fact continued practices common in libraries of medieval monasteries (Carpenter and Svenonius, 1985: 2; Hyman, 1989: 13). Panizzi took the most useful principles, reshaping and widening them to offer solutions to most of the cataloguing problems of his time.
The Ninety-One Rules were drafted by Panizzi for the printed book catalogue of the British Museum and were modified substantially by the Trustees of the British Museum. The rules were a monumental achievement, as they represented the first attempt to codify rules for the compilation of an author catalogue with the inclusion of logical guides for cross references. This system was suited to book catalogues in which only one full entry was provided for each book and other entries were in the form of cross references.
Panizzi's Letter to the Earl of Ellesmere (reprinted in Carpenter and Svenonius, 1985: 18-47) which defined the alphabetical catalogue; the concept of 'heading' as the most important bibliographical element in alphabetical catalogues under which other bibliographic data was subsumed in the catalogue (Ibid: 25-29); and the objectives and problems of the catalogue of a large library such as that of the British Museum are considered as the first theoretical foundations of descriptive cataloguing. Throughout his Letter Panizzi emphasised the concept of uniformity, whether in the form of author heading, the alphabetical order of headings or in the form of slips or cards, and warned of the difficulties resulting from the lack of uniformity in a catalogue. Panizzi's approach to this concept can be seen in his rules for choice and form of a name.
The focus of Panizzi's code was primarily on the creation of entries based on the information found on the title-page. He selected the title-page as the authoritative source of cataloguing data in the sense that it would offer the same information to all cataloguers (rules 20, 1, 3). This was the first approach to standardisation in descriptive cataloguing and has been continued through some of the subsequent codes. Title-page cataloguing was also an indication of the focus of the code on the concept of 'bibliographic unit' and on the finding-list function of the catalogue in that it made for easier retrieval of the item for the user who searched the catalogue with information already obtained from the title page.
The principle of authorship was emphasised as the principal 'organising element' in the catalogue and entries were filed alphabetically by author. When compared to older codes, the treatment of various conditions of authorship that appear on the title-page is, to a considerable extent, complete in Panizzi's code (rules 2-17). The principle of multiple authorship, although without any differentiation between joint and collective authorship, is apparent in the code. In fact, the type of publications of that time did not require a clear distinction; however, the 'Anglo-American Code of 1908' and the 'American Library Association Code of 1949' later found this to be necessary. In terms of cataloguing of publications by corporate bodies, the 91 rules do not envision corporate authorship. Instead, corporate entries could be used as a default (rule 34) or as an organising element (rule 80). According to rule 80:
All acts, memoirs, translations, journals, minutes, &c., of academies, institutes, associations, universities, or societies learned, scientific, or literary, by whatever name known or designated, as well as works by various hands, forming part of a series of volumes edited by any such society, to be entered, according to the English name of the country and town at which the sittings of the society are held, in the following order...(Panizzi, 1841, reprinted in Carpenter and Svenonius, 1985: 12).
In contrast to Needham (1964: 38), Carpenter (1981: 8) states that this is a misconception to consider the 91 Rules as the first code which explicitly recognised the possibility of corporate authorship.
Panizzi suggested the use of form headings as the main entry for certain types of publications. Publications such as acts, memoirs, transactions, and minutes published by institutes, universities and learned societies were entered under ACADEMIES; newspapers, magazines, and annuals under the heading PERIODICAL PUBLICATIONS; and dictionaries under DICTIONARIES. These form headings, which can be regarded as akin to today's form subject headings, were a means for the bringing together of works of a common nature. Panizzi used form headings also for anonymous works, which he entered under some catchword from the title; for example, in his rule 33, he prescribed that: "Anonymous publications relating to any act, or to the life of a person whose name occurs on the title of a work, are to be catalogued under the name of such person" (Panizzi, 1841, reprinted in Carpenter and Svenonius, 1985: 7).
It is generally agreed that Panizzi's Ninety-One Rules for the construction of the British Museum catalogue became the source for all subsequent Anglo-American codes. According to Hyman (1989: 14), "All later catalogers started with Panizzi to compose their codes, and though the debates of his time are still with us, suggested solutions are largely expansions or refinements of the Ninety-One Rules."
It was Panizzi's code that influenced American cataloguers to compile their own codes of cataloguing rules. The rapid growth of libraries in the United States brought the same need for compiling cataloguing codes. Charles Coffin Jewett (1816-1868), then the librarian of the Smithsonian Institution, prepared the first distinct code of cataloguing rules in the United States.
Jewett's code of 1852 entitled On the Construction of Catalogues of Libraries and of a General Catalogue, and Their Publication By Means of Separate Stereotype Titles; With Rules and Examples (Jewett, 1853, reprinted in Harris, 1988: 131-155) consisted of thirty-three rules, mainly based upon Panizzi's code (Hyman, 1989: 14; Tait, 1969: 31). Through minor modifications to Panizzi's rules, Jewett attempted to compose a code of cataloguing rules which would be useful for all American libraries. Another distinct feature of Jewett's code was that it was to be used to prepare both an author listing and an alphabetical listing of subjects.
Jewett continued the work of Panizzi and laid the foundation for the later Anglo-American concept of authorship based upon the principle of intellectual responsibility (Tait, 1969: 31). Unlike Panizzi, who did not envision corporate authorship (see section 1.1), the concept of corporate authorship was reinforced and established more clearly in Jewett's Code. For example, rule 22 prescribed that:
Academies, institutes, associations, universities, colleges; literary, scientific, economical, eleemosynary and religious societies; national and municipal governments, boards, corporations, and other bodies of men...issuing publications...are to be considered and treated as the author of all works issued by them, and in their name alone (Jewett, 1853, reprinted in Harris, 1988: 149).
However, as can be seen, this rule implies that corporate bodies are not authors but should be considered as such. Carpenter (1981: 11) has a similar interpretation.
To provide effective access to the author/title catalogue, Jewett composed rules for the choice of headings (rules 13-29). An interesting and new concept was that, in rule 13, he prescribed that the heading was to be written above the transcribed title. According to Tait (1969: 32), what makes Jewett modern and in line with cataloguing codes of recent times, such as CCR (1960) and AACR (1967) is in the extent to which he separated rules for titles (rules 1-12) from those for headings (rules 13-29). Another departure from Panizzi is in the entering anonymous works under the first word of the title rather than under catchwords contained in the title. Unlike Panizzi, Jewett did not use form headings for anonymous works. He suggested that anonymous works should be entered under the first word of the title, not an article or preposition (Jewett, 1853, reprinted in Harris, 1988: 133-134).
Jewett emphasised uniformity as an essential issue in cataloguing and suggested all libraries should adhere to same set of rules (Jewett, 1853, reprinted in Harris, 1988: 135).
Jewett's contribution to author/title cataloguing and the idea of centralised and cooperative cataloguing was in his proposal to prepare stereotype entries for a printed book catalogue. The stereotype blocks that Jewett proposed could be used to print book catalogues for other libraries. In addition to being economically advantageous, this scheme necessitated uniformity in bibliographic description and an agreed code of cataloguing rules. According to Carpenter and Svenonius (1985: 49):
A technological advance that promises to reduce cataloging effort tends to be accompanied by a concern for standardization. If the bibliographic records of one institution are to be used by another, they must be constructed according to a uniform style. For Jewett, a code of cataloging rules exists to promote uniformity, and he formulated what might be called a principle of standardization.
The year 1876, besides being the year in which the American Library Association was established, is also a significant date in the Anglo-American history of descriptive cataloguing; in this year Charles Ammi Cutter (1837-1903) published his famous Rules for a Dictionary Catalog, a landmark in the continuum of the Anglo-American cataloguing code tradition. Cutter's Rules are undoubtedly the most comprehensive set of rules ever produced by an individual (Quigg, 1966: 19).
Cutter's influence on descriptive cataloguing can be seen in the later design, objectives and principles of cataloguing codes. In his "Preface to the Fourth Edition" of Rules for a Dictionary Catalog, Cutter illuminated cataloguing theory by considering the principle that "The convenience of the public is always to be set before the ease of the cataloger" (Cutter, 1904: 6). According to Svenonius (1989: 41), "Unlike Jewett, Cutter preferred to view cataloging as an art needing only a few high-level rules or principles that could be applied by analogy to a variety of situations."
Through his 'Objectives of the catalogue' Cutter reinforced the contention that the catalogue should not only function as a finding list of what exists in a library, but also should assemble 'literary units' by showing what the library has under a given author. Cutter (1904: 12) formulated the objectives of the catalogue as:
1. To enable a person to find a book of which either
(A) the author )
(B) the title ) is known
(C) the subject )
2. To show what the library has
(D) by a given author
(E) on a given subject
(F) in a given kind of literature
3. To assist in the choice of a book
(G) as to its edition (bibliographically)
(H) as to its character (literary or topical)
To achieve these objectives, Cutter (Ibid: 12) offered as practical means:
1. Author-entry with necessary references (for A and D)
2. Title-entry or title-reference (for B)
3. Subject-entry, cross-references and classed subject table (for C and E)
4. Form-entry and language entry (for F)
5. Giving edition and imprint, with notes when necessary (for G)
6. Notes (for H)
Cutter anticipated a need to go beyond the limits of the finding-list function and assumed that a library catalogue should assemble all the editions of a work. This can be considered to be an evolutionary step toward the further development of the concept of the functions of the catalogue. Pettee (1936, reprinted in Carpenter and Svenonius, 1985: 84) asserted that:
We find in Cutter, fixed for all time, I believe, the two fundamental principles of the modern author catalog: 1. The author catalogue is more than a finding list of separate and particular books. It deals with literary unit and its function is to assemble under a convenient heading all issues or forms of the same literary unit. 2. the most satisfactory method of doing this is through the attribution of authorship, using as heading the name, of the person, or corporate body responsible for the work, or using as a substitute for author heading, a conventional name not derived from the title-page but from the literary source of the book or document.
Rules for a Dictionary Catalog was a relatively comprehensive code that included provision for author, title, subject, form headings, description, and filing.
Cutter made one of the most significant contributions to the development of the author catalogue and the establishment of the authorship principle. He enriched the concept of authorship by using 'main entries' for the fullest bibliographic record and 'added entries' in abbreviated detail. 'Main' and 'added entries' would fulfil both the finding and the assembling functions of the catalogue. To Cutter, the assembling or collocating function of the catalogue, which brought together all works of an author and all editions of a work, was as important as the finding function.
A significant feature of Cutter's Rules was that he extended the concept of corporate authorship. This is an indication of the flexibility of the code towards changing conditions in the authorship concept. Due to an increase in the number of corporate bodies and consequently the number of publications created under their responsibility, it was natural for the code to provide rules for them. Corporate bodies were considered as the authors of works that were published by them: "Bodies of men are to be considered as authors of works published in their name or by their authority" (Cutter, 1904: 41). This concept is a restatement of what Jewett embodied in his Rule 22 (see section 1.2). Carpenter (1981: 12) identifies Cutter as the first theoretician of corporate authorship. Cutter's principle of corporate authorship, as Lubetzky (1980: 22) points out, dominated all subsequent Anglo-American codes up to AACR2, which attempted to restrict it to only a few specified categories.
Cutter made no use of form headings for main entry but rather transferred this type of heading to the area of subjects. Like Jewett, he entered anonymous works under the first word of the title, but with an interesting exception in the case of biographies, which he entered not under the first word of the title but under the subject (Strout, 1957: 21-22).
Cutter's Rules were designed for the construction of a dictionary book catalogue suited for small and medium-size public libraries with a collection of materials mainly in monographic format. His approach to the dictionary catalogue was to design a code consistent with the dominant printed book catalogues of the time. Thus, the types of entries he proposed and the kinds of access points he prescribed were relevant to this form of catalogue. The use of a full main entry with abbreviated added entries for the retrieval of bibliographic information was suited to the printed book catalogue. In a book catalogue, the complete bibliographic record would be identified under the main entry and added entries would refer the user to the main entry. The added entry function before Cutter was performed by cross-references which referred the reader to the main entry record (Strout, 1957: 25). From the fourth edition of Cutter's rules onward the boundary between the use of added entries and cross references became clearer: added entries to provide added access to a bibliographic record, and cross references for name variations (Tillett, 1989a: 156).
Another significant feature of Cutter's Rules was in the provision for short, medium and full entries. This was a requirement that Cutter considered necessary for different library catalogues of that time. It should be noted that a library could choose one option and construct its catalogue according to only one of the prescribed levels of entries. Cutter took a pragmatic approach toward rules and suggested alternative rules based on the nature and size of the library and the physical form of the catalogue (Needham, 1964: 39). The three styles of cataloguing (short, medium, and full entries) that he formulated were a response to such considerations.
Rules for a Dictionary Catalog became the chief source for later codes in the English language (Ranganathan, 1955: 14). Lubetzky, like many others, was influenced by Cutter and the principles he proposed in his Code of Cataloging Rules (1960) and to the International Conference on Cataloguing principles (1961) are an indication of this influence. According to Tait (1969: 38), "The principles laid down by Cutter in 1876 remained almost without challenge until almost the present decade -- a testimony to their value."
Parallel to the development of cataloguing codes in the nineteenth century in the Anglo-American countries, an innovation -- the introduction of cards for the construction of library catalogues -- changed the whole face of American libraries. Although the idea of using slips and cards as a basis for creating the catalogue is not an American invention and goes back to the eighteenth century, when Abbé Rozier employed cards for the compilation of an index of publications of the Paris Académie des sciences in 1775, it was only much later and in the United States that card catalogues became common in libraries for public use (Hanson and Daily, 1970: 266). It should be added that slip or hand-written card catalogues were used for the preparation of book catalogues, but not for public use.
After a number of scattered attempts to use printed slips in public catalogues in the second half of the nineteenth century, the card catalogue became commonplace in the United States in the late nineteenth century. In 1898, the Library of Congress began to provide printed catalogue cards for distribution. Although the service became available to all American libraries by 1898, it was not widely accepted by librarians, due to the various approaches and practices of individual libraries toward author headings, the size of the card, and long delays in shipment (Jordon, 1904: 318). In the last edition of his rules in 1904, Cutter (1904: 5) recognised the card catalogue as the way of the future.
By the late nineteenth century, the dictionary card catalogue had been universally accepted as a practical and responsive catalogue. According to Hanson and Daily (1970: 297-8), "By the end of the nineteenth century the pattern of catalog construction was well defined with subject headings, main entry, literary unit principle, dictionary catalog, classification, unit cards, added entries, and adequate bibliographical descriptions all well-developed elements."
Cataloguing in the second half of the nineteenth century was characterised by the compilation of a great number of codes which followed more or less the same trend in establishing principles for the construction of library catalogues. There was agreement on a number of general principles, particularly entry under author.
In the twentieth century, a number of significant factors have affected libraries in their operations and practices, including descriptive cataloguing. These factors are: the steady growth in the number of libraries and the size of collections, as an indirect indication of social and technological change; a rapid increase in the number of publications in book and non-book form which led to the 'information explosion' in the twentieth century and the tendency to, and need for, more and closer international relations. There was a strong trend towards international cooperation and the exchange of ideas in the early twentieth century. The role of national libraries and library associations in formulating bibliographic standards is considered to have been a significant factor in the development of cataloguing codes and the move toward national as well as international standardisation. In the Anglo-American world, the move toward closer cooperation and formulation of joint codes is a clear expression of such trends in the cataloguing community.
Also, availability of printed catalogue cards from the Library of Congress, which was considered by Cutter (1904: 5) as "...a great change...[in] the status of cataloging in the United States" was an important move toward bibliographic uniformity at the national, and later at the international, level.
As the first attempt at joint compilation of a code of cataloguing rules at an international level, the Anglo-American Code, known as AA (1908) or the Joint Code, was a response to the common interests and problems in descriptive cataloguing that existed in the United States and the United Kingdom in the late nineteenth and early twentieth centuries. It was the result of cooperation between the [British] Library Association and the American Library Association in the production of "...a joint code which would bring uniformity into the cataloging practice of the English-speaking countries" (Quigg, 1966: 21). Although the idea came first from Melvil Dewey in 1900, the proposal was made by the [British] Library Association to ALA in 1904 (Ibid: 21).
Unlike most older codes, AA (1908) was not formulated for an individual library, rather it was designed for larger libraries of scholarly character (American Library Association, 1908: Preface). It was, therefore, naturally influenced by other codes, such as Cutter's Rules, the Prussian Instructions, the British Museum code and also the Library of Congress rules, as shown in its preface and throughout the text. According to Daily (1968: 417), the code "...was a largely successful effort to rationalize the rules of several different libraries into one readily followed standard."
It can be said that the Anglo-American Code of 1908 was designed for the construction of author/title card catalogues. The printed card service of the Library of Congress, started in 1898, is considered to have been a significant factor in the rapid proliferation of card catalogues among libraries of different sizes and types. In this situation, the compilation of a joint code which would bring uniformity to the catalogues of different libraries was a welcome event, particularly in the United States.
The code recognised the 'literary unit' as the basis for cataloguing. In this regard, the code preferred as the heading the full, real, baptismal name of the author rather than accepting the name by which the author was identified most commonly on the title pages of works or in reference sources (Tait, 1969: 15).
A lack of definition of the concept of corporate authorship and also the undefined distinction and inconsistent treatment of institutions (under place) and societies (under name) were among some of the criticisms made of the code (Quigg, 1966: 22; Tait, 1969: 68; Carpenter, 1981: 17). For example, a difficulty with the rules concerning corporate entry was that buildings and ships were considered as authors?? Most importantly, the principles on which it was formulated were not stated in the code itself. In addition to some inconsistencies and many omissions, its numerous rules made it difficult to understand and use.
Due to a perceived need for a more detailed and comprehensive code which would cover the various problems encountered by cataloguers, there was increasing demand for the revision of the 1908 code. In 1930 the Library of Congress appointed a subcommittee to study its revision. Based on cataloguers' experiences in the actual use of the code, work was begun by ALA with the cooperation of the [British] Library Association. Because of the outbreak of World War II, the British could not continue their cooperation. A preliminary edition was published in 1941 and a second American edition was published in the same year as: ALA Catalog Rules; Author and Title Entries. However, because of its adherence to American perspectives and cataloguing tradition, ALA (1941) is not considered an international code.
An extremely complex code with an enumeration of cases and what many regard as an over-elaboration of rules, ALA (1941) was not welcomed by the cataloguing community (Quigg, 1966: 24; Brown, 1991: 82). Rather than defining principles, it supplemented and amended the 1908 Code. For example, the definition (ALA, 1941: xv-xvi) and the treatment of corporate entry are almost identical to AA 1908. the code was so severely criticised that the situation encouraged Andrew Osborn to write his famous article entitled "The crisis in cataloging" (Osborn, 1941). This article is considered a classic within the cataloguing community and is very much cited in the literature. Osborn criticised the code's approach towards the provision of various rules for the coverage of different cataloguing cases. Daily restates Osborn's criticism of the code that "What is needed is a set of guiding principles, worked out for the most common cases, which accomplish the basic task of cataloging without formalistic constraints that perplex the user without adding materially to precise description" (Daily, 1968: 419).
Due to the severe criticisms of ALA (1941), a new edition, i.e., the rules for entry (known as the 'red book') was published by the American Library Association in 1949. The Library of Congress had already decided to prepare its own rules for descriptive cataloguing (which was published in the same year and was known as the 'green book'). Therefore, ALA requested its Catalog Code Revision Committee to deal only with the revision of part 1 of ALA (1941), taking into account all the criticisms which had been made of it. Thus, the 1949 code covered only rules for entries. In this respect, the code was basically influenced by Cutter's rules for main entry, and his theory that "...the catalogue user was the final arbiter of the form that any specific rule should take" (Tait, 1969: 77).
Although ALA (1949) basically followed AA (1908) and ALA (1941) (see ALA, 1949: v), it had some new features such as:
1. The code chose 'work' as the basis for description. In this regard, it followed AA (1908) and took the concept of 'literary unit' (as opposed to 'bibliographic unit') from Cutter's rules.
2. With regard to the choice of main entry, the code prescribed that the entry was to be made under "...the person or body chiefly responsible for the intellectual content of the book, literary, artistic or musical" (ALA, 1949: 3). Intellectual responsibility was not confined to title page information, rather, the author was usually to be chosen from either the work itself or other sources. In this regard, the concept of principal responsibility was a new idea in ALA (1949).
3. It emphasised the concept of corporate versus personal authorship by giving more guidance on the construction of corporate headings. For example, Rule 1 prescribed that: "Enter a work under the name of its author whether personal or corporate" (ALA, 1949: 3).
In the introduction to the code (ALA, 1949: xx) it is stated that, as the dictionary card catalogue was the dominant form of the catalogue in most American libraries, the rules were prepared with that type of catalogue in mind. The main thrust of the Code's rules was the choice of main entry for the author, personal or corporate, who was considered to be chiefly responsible for the intellectual or artistic content of the work. Added entries were to be made "to enable the user of the catalog to find a work when incomplete knowledge or imperfect memory of the work, or unfamiliarity with the rules of entry, would prevent ready access to the main entry" (ALA, 1949: 220). References (see and see also) were to be used to provide uniformity and effectiveness in the catalogue.
ALA (1949) was criticised for not continuing the trend towards internationalism which AA (1908) had tried to achieve at the beginning of the century, especially as the time was more convenient for the code to approach international uniformity and the postwar condition also seemed to be more in need of international cooperation. By abandonment of the agreement reached in 1908 and by adherence to a national tradition, ALA missed a good opportunity for getting closer to uniformity and standardisation (Strout, 1957: 2; Osborn, 1957: 31).
In his Headings and Canons; Comparative Study of Five Catalogue Codes, Ranganathan (1955: 161) criticised the ALA Code more than other codes because of redundancies and inconsistencies in its rules. In relation to the lack of sufficient answers to different cataloguing problems, there were many objections to the ALA rules (Hyman, 1989: 30). Chan (1981: 16) asserts that "The criticism of Osborn did not seem to have a great deal of effect on ALA 1949, for the rules in this code, in the opinion of many, are as pedantic, elaborate, and often arbitrary, as those in the preliminary edition of 1941." The code was constructed on the same formula as AA (1908) and ALA (1941) but lacked a clear statement of basic principles. Among major critiques of the code was Lubetzky's report, Cataloging Rules and Principles: A Critique of the A.L.A. Rules for Entry and a Proposed Design for their Revision, 1953. Lubetzky (1953: 61) stated that a major shortcoming of the code was that it was "vague in design and weak in structure." According to Tait (1969: 81) Lubetzky criticised the code because of its "proliferation of rules to cover specific cases, and almost complete absence of any principles of entry, in spite of the statement in the introduction. The result is that proliferation also results in inconsistencies."
Lubetzky's Cataloging Rules and Principles (1953) which was prepared for the Board of Cataloging Policy and Research of the American Library Association received general approval (Tait, 1969: 92). Because of his approach towards the design of an effective code of cataloguing rules, Lubetzky was invited by the ALA Catalog Code Revision Committee to prepare a new edition of ALA. Being influenced by Cutter's 'Objectives', Lubetzky produced his Code of cataloging rules, author and title entry: an unfinished draft, 1960, known as CCR (1960). Lubetzky (1960: ix-x) stated the objectives on which the rules were formulated as follows:
The objectives which the catalog has to serve are two:
1. To facilitate the location of a particular publication,
i.e., a particular edition of a work which is in the library.
2 To relate and display together the editions which the
library has of a given work and the works which it has
of a given author.
The two functions are complementary, but both are essential
to the effectiveness of the catalog.
Lubetzky adopted Cutter's 'Objectives' with minor changes. Here, the word 'Book' is replaced by the words 'Work' and 'Edition' to give priority to the literary unit concept and to cater for the inclusion of various types of 'non-book' items in the code. By a clear definition of the main entry concept, Lubetzky tried to remove inconsistencies within Cutter's Code, AA (1908), and ALA (1949) in which "the main entry sometimes represents 'work' and sometimes 'edition'" (Tait, 1969: 93). His emphasis was on the 'work' and he considered that a major function for the main entry was the assembling of the editions of a certain work by a certain author. To Lubetzky, the 'work', rather than the 'edition,' was the primary unit. He assumed that "...the catalogue user is interested in the work represented by the particular publication rather than in its embodiment in any particular edition" (Tait, 1969: 94). In determining the basic principles of his rules, Lubetzky followed Cutter's "convenience of the public" and considered the reader's approach to be an important factor (Dunkin, 1957: 40).
As CCR (1960) was written primarily for the construction of a card catalogue using the concept of a unit cataloguing method, added entries carried far more information than in older Anglo-American codes. Main entry was not considered as the most important entry for a work and lost its importance significantly in the sense that the code avoided the use of the term "main entry". Instead, it used terms like "entry is made under..." or "...is entered under the person..." However, the significance of the main entry concept in single-entry catalogues, such as union catalogues in hard copies, could not be disregarded.
In comparison to the older Anglo-American cataloguing codes, the concept of authorship was broadened and extended to all types of materials in CCR (1960). The utility of such an approach to authorship in the case of non-book materials has been challenged. In this regard, Tait (1969: 95) pointed out that:
Certain types of material may not be amenable to or require author entry as such, and one has the impression that CCR (1960) occasionally forces author entry, presumably for the sake of consistency....This would appear to be pushing the authorship concept rather too far, but is almost inevitable within the general framework of the philosophy of CCR (1960).
The code emphasised the concept of corporate authorship. Rule 22 was a general rule for works of corporate bodies.
Another concept emphasised in the code was a reliance on title-page information in providing standard cataloguing data. In this respect, Lubetzky followed Panizzi, who was the first to recognise the importance of the title-page. This is in line with the users' knowledge of or familiarity with books as known or seen by them. In this regard, it is generally said that the cataloguer can avoid the difficulty of determining the intellectual responsibility of the item in hand (Tait, 1969: 96). However, this approach may be in conflict with the uniformity which is necessary in the assembling of different editions of a work under one form of heading. In relation to the above-mentioned discussion it should be added that, unlike ALA (1949), CCR (1960) avoided the subjective judgements required of the cataloguer in distinguishing between different types of multiple authorship. As noted earlier, this was possible by relying on title-page criteria.
Because of its logical consistency and its compilation on a solid theoretical basis, CCR (1960) received favourable reaction (Tait, 1969: 106). However, in terms of the alterations that would result in catalogues from the possible implementation of the code, it was criticised by reference librarians who claimed that they would find answering reference questions difficult and by library administrators who preferred the simplification of cataloguing rules and less costly practices (Tait, 1969: 106, 107).
CCR (1960) was very influential at the time and Lubetzky was hailed as one of the great theoreticians in descriptive cataloguing. Suffice it to say that the principles adopted and internationally agreed upon at the International Conference on Cataloguing Principles, Paris, 1961 were to a large extent influenced by CCR (1960).
As one of the most important events in the history of descriptive cataloguing, the International Conference on Cataloguing Principles (ICCP), Paris, 1961 has had a great impact on the development of current cataloguing codes. ICCP was a response to the need for international agreement on the principles of cataloguing. The conference aimed at providing basic agreement for the compilation of national bibliographies and national codes (ICCP, 1963: 91). In this regard and in order "to facilitate the international communication of knowledge by achieving the widest possible uniformity in library catalogues and other means of bibliographical communication" (ICCP, 1963: 13), the Paris Conference has been considered to be one of the most effective attempts in the direction of universal bibliographical control and international standardization in cataloguing. Due to a number of conflicting issues in different cataloguing codes and practices, which made the exchange and understanding of bibliographic information difficult at an international level, there was a strong need for some kind of uniformity in national bibliographies and, at a higher level, in the bibliographic universe at large. According to the ICCP Report (1963: 91), the Conference intended:
... to take the necessary action to ensure (a) that cataloguing rules in their countries are established or revised as soon as possible in conformity with the principles laid down by the Conference, and put into practice; (b) that the same principles are taken into account in the compilation of national bibliographies.
Although the physical form of the catalogue was not directly specified in the Principles, it can be concluded that they and their allied recommendations were based on the concept of the card catalogue, the most widely used form of catalogue at the time. It should be added that, by this time, the card catalogue had become an internationally accepted form for the construction of library catalogues.
The principles discussed and agreed upon at the ICCP (1961) applied mainly to the choice and form of headings and entry words as the most important organising elements in author/title catalogues, which needed uniformity at an international level. The ICCP (1961) continued the principles of the older codes in the sense that the emphasis was on alphabetical author/title catalogues and lists of books. As the first issue in the Statement of Principles (ICCP Report, 1963: 91) the functions of the catalogue were agreed upon as:
2. Functions of the catalogue
The catalogue should be an efficient instrument for ascertaining
2. 1 Whether the library contains a particular book specified by
(a) its author and title, or
(b) if the author is not named in the book, its title alone, or
(c) if the author and title are inappropriate or insufficient for
identification, a suitable substitute for the title; and
2. 2 (a) which works by a particular author and
(b) which editions of a particular work are in the library.
In relation to the functions of the catalogue and their influence on other cataloguing principles two working papers were presented at ICCP. The two papers, in fact, illustrate the two contradictory approaches which have their root in the long history of cataloguing codes in relation to the functions of the catalogue and the choice and form of main and added entries. In his working paper, Lubetzky (1963: 139-143) emphasised the second function (i.e., the collocating function) as the principal element for bringing together different editions of the same work. He considered the 'literary unit' (i.e., the 'work') as the basis for description (Lubetzky, 1963: 139-140). In this context, different editions and translations of a work should be found together rather than under their own titles. The name of an author must be uniform to bring together all publications by that author. In terms of titles, Lubetzky (1963: 141-142) put emphasis on entering publications under the title as found in the first publication of the work, or where this was not possible, under the accepted 'conventional title.' This would bring together different appearances of exactly the same work. This approach, of course, has some problems particularly for public library users who usually look for a specific edition or translation of a work and not the original work with a different title they are not familiar with. Verona (1963: 145-157), who presented the second approach to the function of the catalogue, emphasised the publication in hand as the basis for description. She stressed that the function of main entries is: 1) to represent particular publications, and 2) to bring together in the catalogue all publications by one author" (Verona, 1963: 157). In this context, a certain form of the author's name would group together all publications by the author. To Verona, not the main entries, but added entries would link all editions of a certain literary unit (Ibid: 157).
In Statements 3 to 12, the ICCP specified the structure of the catalogue and the kinds and forms of entries that were necessary for the effective discharge of the catalogue's functions. Although Principle 2.1 considered the single item important, Principle 2.2 put emphasis on the work. This was a contradictory situation to which the ICCP (1961) did not offer an effective solution and to which there were objections.
A significant improvement of the ICCP (1961) over AA (1908) and ALA (1949) was in the definitions that it provided for personal and corporate authorship. In this respect, ICCP (1961) put the emphasis on the intellectual responsibility concept of authorship (Tait, 1969: 109). As opposed to some European traditions, Principles 9.1 to 9.3 (see Appendix 1), that regarded corporate bodies as authors, approached the Anglo-American practice and were accepted by a majority of the national cataloguing committees. Representing the point of view of Hungarian cataloguers, Domanovszky argued against this practice and considered it to be opposed to common practice in everyday speech (ICCP, 1963: 43). The German cataloguing tradition also rejected the concept of corporate authorship in most cases and accepted it only when there was no personal author named in the work and when the title of the work was explicitly related to the corporate body as the originator (Ibid: 43). On the other hand, Verona was one of the proponents of corporate authorship and later stated that:
A work should be considered to be of corporate authorship if it may be concluded by its character or nature that it is necessarily the result of the creative and/or organizational activity of a corporate body as a whole, and not the result of an independent creative activity of the individual(s) who drafted it (Verona, 1975: 155).
An interesting issue in the list of topics discussed at ICCP that was not fully acknowledged at the time was the potential impact of electronic information systems upon the process of cataloguing. In his working paper presented at the Conference, Gull discussed the potential impact of electronics, such as machine readable texts and the concept of 'automatic authorship' on cataloguing rules and suggested that the issue be addressed in the design of cataloguing codes (Gull, 1963). For example, Gull (Ibid: 287) anticipated that the concept of main entry would be influenced by electronic texts and main entry headings would not be needed for machine readable texts. Gull (Ibid: 283) concluded that: "...cataloguers must fit cataloguing rules to human capabilities and adapt them to changes in technologies; indeed, they must enlist the aid of new technologies to assure that cataloguing rules are based on substance rather than form." In the early 1960s, however, the issue was far from clear and did not arouse much comment within the profession.
Most subsequent current national cataloguing codes have been compiled on the basis of the ICCP Statements. Although ICCP (1961) was regarded as being highly successful in its resolutions and objectives at the time, it was later criticised for some inconsistencies and vagueness (Verona, 1971: ix; Carpenter, 1994: 109). As reported by national cataloguing committees, full conformity to the Statements proved not to be practical, due to variations in cataloguing traditions and practices.
After ICCP (1961) it was time for the English-speaking world to prepare a new set of rules, a second Anglo-American joint code, which could provide for greater uniformity both in library catalogues and in the exchange of bibliographic information between libraries in the English-speaking countries. Based on ICCP (1961) and their own cataloguing traditions, the library associations of the United States, Canada and the United Kingdom, with the collaboration of the Library of Congress, provided the groundwork for the compilation of a new joint code which would be applicable at an international level.
Due to some variations in cataloguing practices, the code was published in 1967 in two separate texts: North American and British. One difference between these texts was the departure of the American text from the Paris Principles in its rejection of the entry of a collection under title when such a work had a collective title. Furthermore, due to the recommendations and requirements of the Association of Research Libraries (ARL) the impact of the Paris Principles on existing entries for corporate bodies was reduced and there were a few cases in which the code departed from those principles (AACR1, 1967: 3). The most significant departure was the exemption of certain bodies of an institutional nature from the principle of entry under name and a substitution of a rule of entry under the place in which the institution was situated. Based on the American cataloguing tradition of entering a work under author rather than title, AACR1 gave entry of collections under compiler when the compiler's name was on the title-page (AACR1 rules 4 and 5). These exceptions, however, were changed in 1974 and 1975 to bring the code into line with the Paris Principles and to provide uniformity between the two versions (Maxwell, 1989: 4-5).
Unlike AA (1908) and ALA (1949), AACR1 was primarily drawn up to respond to the needs of general research libraries and had some features that distinguished it from its predecessors (AACR, 1967: 4). The needs of smaller public libraries were addressed by the provision of alternative rules as well as more direct headings. Because of a basic difference in approach to the practice of cataloguing, the character of the rules for entry and heading was considerably different. These differences were summarised as: 1) The rules were based on internationally agreed principles, 2) choice of entry and construction of heading were treated as separate problems, except when form subheadings were involved, 3) choice of entry was treated as a problem of determination of authorship responsibility, and 4) construction of heading was treated as a problem of name (AACR, 1967: 5).
A thorough examination of the code reveals that the rules for entry and heading are substantially based on ALA (1949), Lubetzky's CCR (1960), and ICCP (1961). However, rules for description in AACR1 were formulated on the basis of Rules for description in the Library of Congress (1949) and its supplementary rules (AACR, 1967: 6).
The principle of authorship was more clearly enunciated in AACR1, where the primary criterion for determining authorship was 'intellectual responsibility'. As will be discussed later, its handling of this issue was based on conditions of authorship rather than on types of work. According to Lubetzky (1980: 22), "...the governing principle in AACR1 is that entry is to be based not on type of publication or work but on the varying conditions of authorship." With regard to corporate bodies, as can be seen from the AACR's definition, they were considered as authors in an explicit way (AACR, 1967: 9, 243).
Although the title-page was chosen as the chief source for determining entries (i.e., main and added entries), AACR1 chose the 'work' as the basis for cataloguing. This was apparent throughout the statements of the rules in part I. Lubetzky (1969: leaf 26) criticised the definition of authorship in AACR1 in that it based the chief responsibility for a work on its intellectual or artistic content rather than considering the producer as the person chiefly responsible for a work (Wajenberg, 1989: 22).
Through their emphasis on entries under uniform headings and the use of references, the rules were oriented to multiple-entry alphabetical catalogues. However, AACR1 still distinguished between main and added entries. In explaining the significance of this concept, the framers of the code stated that, both for identification of a single-entry work in multiple entry catalogs as well as for the general needs of libraries and bibliographical and book-trade activities, a standard mode of identifying bibliographical entities was necessary (AACR, 1967: 2). This standard mode is one of the often stated justifications for the use of main entries in such situations as single-entry bibliographies, book lists, order lists and bibliographic citations.
At the time that AACR1 was compiled, the card catalogue was the dominant form of library catalogue and the framers of the code did not take into account the potential impact of automation on library operations, particularly the application of the computer in providing public catalogues. Daily (1968: 421) believed that it was the fate of the code that it was designed for practices of that time and even previous years and not for the coming decades and for the computerised listing of entries. It is worth noting that in the 1960s, the attitude of most librarians towards computerised catalogues was limited to a superficial understanding of the concept of automation, e.g., the ability of computers to perform repetitive tasks with much more precision and speed.
An important factor that disappointed cataloguers in the full adoption of AACR1 was the policy of 'superimposition', declared by the Library of Congress, of continuing the usage of headings which had been constructed according to ALA (1949) rules. This policy made trouble for some cataloguing departments in maintaining their catalogues and resulted in numerous inconsistencies in name headings.
Soon after the adoption of AACR1 in large libraries, it became apparent that the code was in need of a number of additions and changes. The Library of Congress and the ALA's Division of Cataloging and Classification approved the need for the addition or change of a number of rules. The British did not welcome the continuous revisions of the rules that were being undertaken in the United States. However, cataloguing trends in the 1970s, such as the introduction of the International Standard Bibliographic Description (ISBD), necessitated that the code should be brought into line with these new trends.
As one of the major products of the train of activity set in motion by ICCP and elaborated by a Working Group set up by the International Meeting of Cataloguing Experts (held at Copenhagen in 1969), ISBD was developed to specify requirements for the description of publications for the purpose of international communication. The standard assigns an order to the descriptive elements, and specifies a system of punctuation for the description. It is designed primarily as an instrument for the international communication of bibliographical information not only in library cataloguing but also in book trade activities (IFLA, 1971: 1). ISBD is not a set of cataloguing rules: it does not deal with access points and "does not include any prescription for the heading underwhich a description should appear in a catalogue or other bibliographical list" (IFLA, 1971: iv). In addition to the General International Standard Bibliographic Description (ISBD (G)), IFLA has developed specialised ISBDs for specific types of material.
Many national cataloguing codes have based the rules for description on the general framework of ISBD (G). Also most national bibliographies are prepared according to this standard. Adherence to these descriptive standards by many cataloguing agencies, whether working in a manual or an automated environment, has provided greater uniformity in the communication and understanding of bibliographical information within countries and between countries. This is essential to Universal Bibliographic Control.
In respond to the requirements of automated systems, which were developing rapidly in the early 1970s, ISBD was designed in a way to facilitate the conversion of bibliographic records into machine-readable form, or as Gorman (1979: 134) points out, "to maximize the interchange and processibility of data in the machine environment." This was done through the systematisation of the order of data elements and the punctuation that is used to separate these elements in a record.
Although ISBD is a computer compatible format, many automated systems do not use the ISBD order and/or punctuation, especially when displaying bibliographic records. This is in contrast to the primary aim of the standard. Many online catalogues rearrange the data elements, causing confusion from one system to the next (Tillett, 1992a: 112). Much of the criticism of ISBD has been in relation to problems other than automation. Ricard (1977: 431), Lubetzky (1979: 157), Ayres (1981: 4) criticised ISBD mostly for the punctuation, the redundancies and repetitive author statement.
The second edition of AACR was the result of four years' work by the Joint Steering Committee for Revision of AACR (JSC) which had been set up in 1974 by the American Library Association, the [British] Library Association and the Canadian Committee on Cataloguing, with the support and cooperation of the British Library and the Library of Congress. According to Tikku (1983: 156), AACR2 was "the result of the culmination of several factors, which appeared in the 1970's and thus necessitated the production of a revised edition of cataloging rules." These factors were: piecemeal revision of AACR1; the development of the International Standard Bibliographic Description (ISBD); LC's interest in 'desuperimposition', i.e., allowing for changes to headings established under pre-AACR rules; and the use of computers in libraries.
During the period between the publication of the first and second editions of AACR (1967-1978) many changes had taken place, such as the development of MARC formats, the development of computerised catalogues, the growth of centralised and cooperative bibliographic services and the proliferation of new media and nonbook materials, which had significant effects on cataloguing and bibliographic control. Also, the introduction of ISBD (M), which was incorporated in AACR2, was considered to be a major factor in the move toward Universal Bibliographic Control (UBC).
In spite of these factors, AACR2 maintained the same principles and objectives as AACR1 and was based on the work of those who created the first edition (AACR2, 1978: v). According to Gorman, coeditor of the code (1980: 41): "AACR2 should be judged not just on its own merits but also as part of a continuum, as part of the 'great tradition' in Anglo-American cataloging which stretches back to Panizzi, Jewett, and Cutter." The aims of the second edition of AACR as set out by JSC were (Gorman, 1978: 209):
to incorporate already agreed revisions to AACR1;
to harmonize the British and North American texts of AACR1;
to incorporate international standards and international agreements;
to take developments in library automation into account; and
to incorporate changes arising from proposals for change coming
from any source.
In comparison to AACR1, the second edition had the following features:
1. AACR2 maintained a more logical and consistent structure than AACR1. In order to provide ease of use for cataloguers, it followed the sequence of the cataloguing process. The rules were divided into two parts: bibliographic description coming as the first part and choice and form of headings as the second. Also the wording, order, and method of construction of the rules were different.
2. AACR2 incorporated fully the General International Standard Bibliographic Description (ISBD (G)) in the rules for description, in that it prescribed the same order of elements and punctuation. As "an instrument for the international communication of bibliographical information" (ISBD, 1971: III), this standard is an important factor in achieving bibliographic standardisation and, at a higher level, Universal Bibliographic Control.
3. AACR2 was designed for use in the construction of catalogues and other lists in general libraries of all sizes (AACR2, 1978: 1). With regard to the varying requirements of detail in bibliographic records in different libraries, three levels of description have been provided. These levels, viz., minimum, medium or standard, and detailed or maximum were, in fact, a revival of Cutter's short, medium, and long entries and are applicable to all types of materials. The choice of a level of description was based on "the purpose of the catalogue or catalogues for which the entry is constructed" (AACR2, 1978: 14).
4. Unlike AACR1, the rules for entry and heading were applicable to all types of library materials. According to Gorman (1978: 219, 1980: 42), the rules and the examples in AACR2 Chapter 21, unlike the book-oriented rules in AACR1, presented an integrated approach to all library materials.
5. In AACR2 the concept of authorship was shrinking. This is apparent in the following aspects:
5.1. In some cases the code replaced the term 'authorship' with the term 'responsibility', which is more comprehensive. A 'statement of responsibility' appeared after the title to indicate the type of responsibility of any one involved in the creation of the item. Also the terms 'Mixed authorship' and 'Shared authorship' that were used in AACR1 were replaced by 'Mixed responsibility' and 'Shared responsibility' respectively.
5.2. A major difference of AACR2 from the older Anglo-American codes was that editors of collections were no longer regarded as authors. The decision was made in accordance to Statement 10.3 in the Paris Principles. This shrinking of the concept of authorship was a new trend in the family of the Anglo-American codes.
5.3. In AACR2 the concept of corporate authorship was no longer prevalent, as can be seen in the restriction of cases where main entry was to be under the names of corporate bodies and the increase in the number of cases where serials were given title main entry. In terms of serials, all special rules for entry of serials were removed from the code with AACR2. Thus, rule 6 (entry for serials) is not based on authorship. This approach has been criticised by some serials librarians and others on the grounds that it was hard to create relationships among the different publications of a corporate body (Malinconico, 1980: 39; Fasana, 1979: 468).
6. In accordance with the requirements of the MARC format, AACR2 abolished the dash entry. This was done by prescribing separate entry or multi-level description for items such as supplements, indexes, detached copies and offprints. The dash entry had been used in book and card catalogues and, once the MARC format was introduced, it was abandoned (Tillett, 1991b: 396; 1992b: 25).
AACR2 followed its predecessors, mainly CCR (1960) and AACR1, in that the rules for choice and form of access points (the rules in part II) "apply to works and not generally to physical manifestations of those works, though the characteristics of an individual item are taken into account in some instances" (AACR2, 1978: 277).
In spite of the fact that a computer catalogue permits a number of equal access points, the concept of main entry was retained in AACR2. The reasons for this, as presented by the proponents of the main entry concept in JSC's arguments, were on the basis that it is a central principle in conventional cataloguing theory and that it has practical utility in printed book catalogues, in shelf-listing, and in single-entry listings (Gorman, 1978: 218). However, for those catalogues which are not based on the main entry principle (i.e, they do not distinguish between main and other entries), AACR2 instructed the cataloguer to use the rules in chapter 21 as a guidance for making necessary entries (AACR2, 1978: 2). Although AACR2 deemphasised the traditional concept of main entry and encouraged the use of 'access point' for any entry, it did not ignore its usefulness and states that (Ibid: 2):
It will be necessary, however, for all libraries to distinguish the
main entry from the others when:
a) making a single entry listing
b) making a single citation for a work (as required for entries for related works and for some subject entries).
In addition, the concept of main entry is considered to be useful
in assigning uniform titles and in promoting the standardization of
bibliographic citation.
Despite its flaws in a number of aspects, such as not being based on coherent principles (Shinebourne, 1979), a lack of defined objectives and a lack of integrity (Lubetzky, 1980), a lack of separate rules for deciding on serials publications access points and inconsistent treatment of the various media (Butchard, 1980; Willard, 1981), ambiguities which make the rules open to misinterpretations (Ayres, 1981, 1990), irrelevance to computerised catalogues (see Review of the Related Literature, section 1, Chapter 1, or Fattahi, 1995a, for a more detailed account), AACR2 obtained overall support among cataloguers and cataloguing educators (Musavi, 1986: 137, 140). It undoubtedly went further toward achieving standardisation and internationalisation than AACR1 and even more than any other code (Delsey, 1989: 53; Rowley, 1989: 9; Muller, 1992: v). According to Smiraglia (1992: ix), "AACR2 is an important partner in the international effort to share bibliographic data." The code was considered by UNISIST Guide to Standards for Information Handling (1980: 152) as one of the standard codes to be used in a multinational context.
The declaration of 'desuperimposition' and the new policy of the Library of Congress towards the closing of its card catalogues beginning from 1980, which later was postponed to 1 January, 1981 was considered an important issue for American libraries. There was a great deal of frustration about the extent to which headings in existing catalogues constructed under older codes would be affected. At the same time, the advent of online catalogues, as well as the difficulties of maintaining manual systems, encouraged many libraries of different sizes to close or freeze their card catalogues, to adapt themselves to the Library of Congress' policy, and to adopt fully AACR2.
The implementation of AACR2 engendered frustration in many types of library and generated a large amount of literature concerning its problems, shortcomings and flaws. To many cataloguers, the code did not fulfil their cataloguing requirements in a time of rapid movement towards information networks and the international exchange of bibliographic records. Based on these criticisms and recommendations, the JSC (this time with a new member, the Australian Committee on Cataloguing (ACOC)) published three separate groups of rule revisions in 1982, 1983, and 1985. These revisions and the complete revision of chapter 9 (Machine-readable data files), published in 1987, made it necessary to provide the profession with a new revision of the code. Thus, the Anglo-American Cataloguing Rules, Second Edition, 1988 Revision (AACR2R) was published under the editorship of Michael Gorman and Paul Winkler.
The appearance of the 1988 revision of the Anglo-American Cataloging Rules did not arouse as much enthusiasm as its predecessor (AACR2) had in 1978. Although AACR2R is the result of ongoing revisions, it maintains the same principles and guidelines as AACR2. The major differences relate to the rules for description of some types of materials, notably computer files (in chapter 9). Part II (Headings, Uniform titles, and References) has some changes, of which the most important are in the areas of pseudonyms and geographic names as well as the headings for some subordinate corporate bodies and uniform titles for music.
Two major changes in the authorship concept have been addressed in AACR2R. Separate bibliographic identities (single authors writing under two or more identities) are to go under their own headings in the catalogue. AACR2R rule 22.2B2 instructs the cataloguer to choose two or more headings for a person who uses a form of his or her name (for example, the real name) for one type of work and another name (for example, a pseudonym) for another type of work. Rule 22.2B3 (concerning contemporary authors) takes the same approach: the heading for contemporary authors is to be chosen according to the name which appears in the manifestation of the work being catalogued. The aim is to simplify the cataloguer's task and to provide convenient access for the catalogue user (Gorman, 1989: 184).
As the fact that the direct name of a corporate body is more convenient to the catalogue user, AACR2R has taken this approach and suggests that the direct forms of heading be preferred. Rules for geographic names (chapter 23) have also been drastically simplified to identify geographic names through addition to the name of the state, province, region, etc. in certain countries (Rule 23.4)
The framers of AACR2R have tried to prepare a set of rules that could be applicable to all types of library materials in all formats and all languages. Thus, in comparison to its predecessors, the code moves further toward internationalism. The code has been adopted by many countries as their national cataloguing standard (Delsey, 1989: 53). It has also been translated into Arabic, Finnish, French, Japanese, Korean, Malaysian, Norwegian, Portuguese, Spanish, Swedish, Urdu (Smiraglia, 1992: ix), and into Afrikaans, Chinese, Italian, Russian and Persian.
A number of national cataloguing codes have either been independently written or contain some major departures from the Paris Principles. In this section two non Anglo-American cataloguing codes, i.e., the German cataloguing code (Regeln fr die alphabetische Katalogisierung (RAK) and the Japanese cataloguing code (Nippon cataloguing rules (NCR)) will be briefly studied to acknowledge some of the basic concepts underlying cataloguing traditions outside the Anglo-American context.
The Instruktionen fr die alphabetischen Kataloge in der preussischen Bibliotheken [Instructions for alphabetic catalogues in Prussian libraries, the PI or The Prussian Instructions], which had been in use since 1899, no longer satisfied the requirements of descriptive cataloguing in the second half of the twentieth century. There was a strong need to adapt, as far as possible, to international cataloguing practice and to take into consideration a more precise treatment of publications, for example, to comply with modern forms and titles of publications and to take into consideration the impact of electronic data processing (Bouvier, 1982: 47).
After the Paris Conference (1961) and in moving toward an international consistency of cataloguing codes, a new code began to be prepared by the Verein Deutscher Bibliotekare of the Federal Republic of Germany. Among the concepts covered by the new code, two concepts were new to the German cataloguing tradition: entry under corporate name and the filing of entries by computers (Kaltwasser, 1974). In 1965 cooperation began with the Bibliotheksverband der D D R (German Democratic Republic) which was working along the same lines. In 1969 two parts of the code were published as preliminary editions. RAK was produced jointly by commissions of the Verein Deutscher Bibliothekare (German Federal Republic), of the Deutscher Bibliotheksverband (now Bibliotheksverband der DDR), and of the Vereinigung Osterreichischer Bibliothekare (Austria) in 1969. The Rules in RAK apply to libraries of every size and of varying functions. They contain many alternatives for different cases.
RAK is based on the Paris Principles and ISBD. It is said that the code is closer to international recommendations than other cataloguing codes (Kaltwasser, 1974: 283; Johnson, 1975: 13). Johnson (1975) reports some of the major differences between RAK and AACR. In terms of authorship, however, RAK continues the German tradition of attributing this concept only to persons. For corporate bodies, it introduces the concept of 'Urheber' (originator) to identify such bodies as either creators or sponsors of anonymous works. The term 'Urheber' came with the introduction of the concept of corporate authorship and was taken over from legal theory and adapted for cataloguing purposes (Verona, 1974: 3). In this context, the main entry is made under the 'Urheber' only when no personal author is named in the work and only when the title of the work is explicitly related to the corporate body as the originator.
The filing principle in the Prussian Instruction was different from that of other codes in that it was grammatical and not mechanical (i.e., titles were arranged according to the grammatical dependencies of words on each other and not on the order in which they appeared in the title). In RAK this approach was changed to take advantage of electronic data processing in the filing of large groups of entries in automated catalogues. It is claimed that the RAK filing rules are the first filing rules for catalogues with the application of automation in mind (Kaltwasser, 1974: 283).
To adapt the German code to the online environment and to consider possible changes to RAK, an expert committee (Expert Group RAK) was established in 1994 to continue the work of a special Online Expert Group (personal communication of Monika Mnnich<MCH@ubhd2.ub.uni-heidelberg.de>, Chair, Expert Group RAK, to the writer, 2 July 1994). It is said that the Expert Group is working on possible modification of RAK with radical ideas and concepts which may be challenging the Anglo-American cataloguing tradition. Among the issues for possible development for online catalogues are:
1) The abandonment of the concept of main entry and the relegation of its function to rules for output format. In their present structures the rules for determining main and added entries are said to be too restrictive for online catalogues. For the online catalogue, none of the classic access points such as corporate names, personal names and titles need to be weighted as primary or secondary by means of tags or indicators (Munnich, 1995: 1269).
2) Indexing for online catalogues, for example, to avoid redundancies in title in the bibliographic description and in the tracing. It should be ensured that rules do not hamper existing data structures, as the many present redundancies do (Mnnich and Zillman, 1994: 1423).
3) Formatting possibilities and display of bibliographic description. The desired form of lists can be prepared according to the needs of the clientele for whom they are intended and not according to academic citation methods which are not consistent with OPAC displays (Mnnich and Zillman, 1994: 1424)
It remains to be seen what would be the consequences of such changes for the international exchange of bibliographic data and how far such new rules affect online retrieval of records in a global online environment.
The Japanese code, i.e., Nippon Cataloguing Rules (NCR), which was compiled in 1977, is not based on the Paris Principles. It is basically for publications in the Japanese and Chinese languages and is based on a description unit card system (DUCS) and does not conform to the principle of main entry. In a catalogue constructed according to this system, multiple entries are created for an item by reproducing the unit card made for it. There are no main entry headings and as many headings as are needed are recorded in the tracing. The descriptive part of the record is created independently of the headings (Takawashi, 1989: 67).
Four revisions were carried out in the 1980s and it was revised to be compatible with both the ISBDs and UNIMARC. The last edition, the 1987 edition, is still based on the no main-entry principle.
A most notable feature of NCR is in its treatment of bibliographic hierarchy. This novel feature has been incorporated in preparation for online catalogues (Takawashi, 1989: 70). Based on the concept of bibliographic hierarchy as derived from the Reference Manual for Machine-Readable Bibliographic Description (Dierickx and Hopkinson, 1981) and unlike its 1977 edition, in which the 'physical unit' was taken as the basis for description, this edition has taken the 'work' as the basic unit for description (Takawashi, 1989: 70). In this situation, the physical manifestations of a work are considered subordinate to it. With this approach there is much flexibility for the treatment of bibliographic relationships through a multi-level record.
In studying the historical development of modern catalogues and cataloguing codes, particularly with respect to the physical form of the catalogue, a range of factors and elements have been highlighted as the most important concepts. The development of these concepts, which form the underlying principles of descriptive cataloguing and the design of cataloguing codes, has been substantially based on two fundamental factors: the form of the catalogue and the nature of bibliographic entities. These concepts have been subject to constant assessment, reassessment and change throughout the history of modern cataloguing:
1. The alphabetical catalogue. The classified catalogue of the early nineteenth century gradually gave way to the alphabetical catalogue, particularly the dictionary catalogue, which has been one of the most popular forms of the catalogue throughout the twentieth century. The development of cataloguing rules in the Anglo-American context for choice and form of entry has been influenced by this type of catalogue.
2. Objectives and functions of the catalogue. The principles on which modern cataloguing codes have been formulated have developed in accordance with the objectives and functions of the catalogue. The older inventory function of the catalogue evolved into a finding function in the early nineteenth century and later also included an assembling function in Cutter's Rules and afterwards. Provisions for the choice of entry in cataloguing codes are based on the objectives and functions of the catalogue and how they should be discharged.
3. The concept of a work. The controversial concept of 'work', as opposed to the concept of 'book' has had significant influence on other concepts and principles in descriptive cataloguing. By giving priority to one concept, other concepts such as the concept of authorship, choice and form of entries and uniform titles are influenced. Inconsistencies are present in some cataloguing codes due to the confusion between 'work' and 'book'.
4. The concept of entry. As the most important means of approaching surrogates of library materials, the concept of entry (the bibliographic record of an item) has become fixed in bibliographic practice. Thus, a significant portion of any cataloguing code is devoted to rules for the construction of entries, which discuss what constitutes the entry and the choice of entry words and access points (i.e., main and added entries) as well as the form of entry words (headings). Thus, this concept implies the significant notion that there is more than one approach for finding an item in a catalogue.
4.1 The concept of main entry. The question of determining principal responsibility, i.e., the concept of main entry under which the complete catalogue record of an item is entered, has been one of the most important themes in Anglo-American cataloguing codes. With the advent of online catalogues, the value of the main entry has been challenged but with no satisfactory resolution of how to fulfil its functions in its absence.
4.2. Number of access points per work and the reference structure of the catalogue. As the medium through which the bibliographic record is accessible to the user and the technology of catalogue production has developed from the book catalogue to the card catalogue to the computerised catalogue, there has been more potential for increasing the number of access points by which the complete record is accessible. The author, main entry catalogue, with references, that Panizzi and Jewett found to be the most usable and reasonable eventually evolved into the dictionary catalogue. Development in the application of main and added entries is an indication of better, more flexible and more comprehensive ways of access to bibliographic information.
5. The concept of authorship. This concept is one of the most important organising (as well as identifying) elements in catalogues and bibliographic lists. A considerable portion of most codes has been devoted to the concept of authorship and to addressing its numerous conditions and problems. Due to the various conditions and complexities of attribution (for example, in the case of multimedia) what constitutes authorship in cataloguing (the originator or the producer?) is a controversial issue with no satisfactory and comprehensive definition having been established.
5.1. Personal authorship. Although there has not been consensus as to the definition of what constitutes authorship, this concept has been accepted as one of the most important organising and identifying attributes of a work in the Anglo-American cataloguing tradition, as opposed to oriental traditions. Throughout the evolution of types of authorship, the term has been extended to persons other than writers of works such as editors, compilers of collections, artists, photographers, cartographers and composers of music. There have, however, been changes in the last two decades so that editors and compilers no longer have the same authorship status as previously.
5.2. Corporate authorship. It was the Anglo-American tradition that extended the concept of authorship to corporate bodies. The concept, though sometimes controversial in definition, was retained for more than a century through the work of Panizzi, Jewett, Cutter, AA (1908), ALA (1941), ALA (1949), ICCP (1961), and AACR1 (1967). Although not completely agreed upon, the concept was included in ICCP. The concept has since lost some of its force and, in AACR2 (1978), the term 'corporate authorship' has been changed to 'corporate responsibility'.
6. Forms of heading. A major portion of each cataloguing code is concerned with the form of the name chosen as heading. This has been a significant contribution, in most codes over the past one hundred and fifty years, towards the achievement of uniformity in bibliographic description and access. In most modern cataloguing codes, it is suggested that a single form of name (i.e., uniform headings) be used for all the works of an author. Uniform titles are another approach to bringing uniformity in the treatment of the various manifestations of a work.
7. The principle of standardisation. A major aim of modern cataloguing codes has been the achievement of standardisation in bibliographic information. This concept has been embodied in different codes in the provision of rules for transcription of cataloguing data from uniform sources (e.g., title pages or chief source of information), standardisation of bibliographic description (e.g., ISBDs), and uniform rules for different types of material. Standardisation has evolved from an internal uniformity within individual catalogues into an international context.
8. Form of the catalogue. In relation to the impact of the physical form of the catalogue on the design of cataloguing codes, it can be concluded that, in most cases, the form of the catalogue is either explicitly indicated in the code or can be implied from the predominant form of the catalogue at the time a code was being prepared.
In conclusion, while there have been considerable changes and revisions in rules, cataloguing principles have changed little. Most of the principles formulated in the nineteenth century by Panizzi, Jewett and Cutter have been adopted in our present codes. How does the online environment differ from the manual environment and what are the capabilities of the online catalogue which, as many writers claim in the literature, demand changes in both cataloguing principles and rules? Such questions and concepts require that, in the first place, a thorough comparison of the two environments be carried out to reveal how far the process of input, storage and output of bibliographic data is different in the new electronic environment, a subject which will be treated in the next chapter. The study of these differences will help to understand better the possible influence of the catalogue environment on cataloguing principles and rules.
Back to: