After the 1997 International Conference on the Principles & Future Development of AACR, there were two action items, in particular, that began a process of revision that resulted in a complete deconstruction of AACR2, and the development of the new successor standard, RDA: Resource Description and Access:
Tom Delsey, who had delivered the paper, "Modeling the Logic of AACR", at the International Conference, and was a member of the IFLA Study Group who had developed the FRBR model, was asked to carry out the logical analysis. He used an entity-relationship model, and, drafted, with assistance from others,52 a detailed analysis of both Parts 1 and 2 of AACR which brought to light certain fundamental problems in the logical structure of AACR. The analysis is reported in the document The Logical Structure of the Anglo-American Cataloguing Rules.
The first key issue addressed was the question: "Does the concept of class of materials as currently reflected in the code serve as a viable basis for an extended structure accommodating new forms of digital materials?" 53 To approach the question of whether AACR could easily extend to describe new types of bibliographic resources, Delsey examined the assumption underlying the concept of "class of materials". Rule 0.24 assumed that class of material was defined on the basis of a physical carrier:
It is a cardinal principle of the use of Part I that the description of a physical item should be based in the first instance on the chapter dealing with the class of materials to which the item belongs ... In short, the starting point is the physical form of the item in hand...
However, in the course of modelling the data and looking in more depth at the chapters in Part I, Delsey demonstrated that only five classes of material were defined by physical carrier: sound recordings, motion pictures, videorecordings, computer files and microforms. Cartographic material, graphic materials, and three-dimensional artefacts and realia were not in fact defined according to physical carriers, but according to the type of content. Music, in the AACR context, is only music as expressed in musical notation, i.e., scores. Taking the FRBR model as a way to clarify the problem, one sees that some classes of material are defined according to attributes at the manifestation level, such as videorecordings, one is defined according to an attribute at the expression level, i.e., music scores, and some are defined according to attributes at the work level, such as cartographic material. It is little wonder that it was difficult to extend AACR2 to describe new types of resources and difficult to describe resources consisting of different types of material. The categories were assumed to be equal, but were in fact disparate and inconsistent. In light of this inconsistency, there were two pressing questions: if a resource belongs to two or more classes of material, rules associated with which class should take precedence? And similarly, for a new type of resource, if it falls into two or more classes of material, which class of material should take precedence?
The recommendation in answer to this key issue was to consider the possibility of "deconstructing" class of materials and developing a more flexible approach so that AACR could easily extend to the description of new types of material. Since the concept of class of materials dominated the structure of Part I, the recommendation was worded as "Use the model developed for this study to assess the options for restructuring Part I of the code." 54 It also goes on to suggest the possibility of reorganizing Part I according to the ISBD areas.
The second action item, "Solicit a proposal to revise rule 0.24 to advance the discussion on the primacy of intellectual content over physical format", was assigned to the ALCTS CCS Committee on Cataloging: Description & Access (CC:DA). They were asked to examine 0.24 and prepare a rule revision proposal. The CC:DA task force identified two aspects of the problem: (1) how to describe a bibliographic resource that has multiple characteristics and (2) how to deal with identical intellectual content existing on a variety of carriers, also called the format variation problem in their report.55 They prepared a revision proposal that led to the amendment of rule 0.24 in 2001.
It is important to bring out all aspects of the item being described, including its content, its carrier, its type of publication, its bibliographic relationships, and whether it is published or unpublished. In any given area of the description, all relevant aspects should be described.
The revision responded to the problem of how to describe a bibliographic resource that has multiple characteristics. It did not indicate any precedence among the classes of materials, nor did it address the inconsistency in categorization of the classes of material. Any changes to classes of material would necessarily entail changes to the structure of Part I of AACR2.
The task force also explicitly supported the recommendation made by Tom Delsey to restructure Part I, and recognized that the full resolution of the 0.24 problem could not be done simply as a revision of one rule. The task force clearly indicated that the wording of the revision proposal was intended as an interim step:
It is recognized that the editing required to reorganize AACR2 by ISBD area will be extensive. Thus, CC:DA recommends that a staged approach be taken such that the text for rule 0.24 would be changed immediately, but more time would be devoted to reorganizing the chapters of AACR2R according to ISBD area.56
The Task Force also predicted that "the format variation problem" would have far-reaching consequences and recommended that the Joint Steering Committee appoint another task force to explore this problem.
The Joint Steering Committee appointed the Format Variation Working Group (FVWG). As Jennifer Bowen, its Chair, summarized:
While the group had several different terms of reference requesting that it undertake various specific tasks for the JSC, all of these tasks had the common element of dealing with the FRBR entity expression. 57
The first term of reference was to examine the viability of expression-level cataloguing. At first glance, one might wonder why JSC framed the term of reference in this way, rather than asking the Group to look at the "format variation problem", as described by the ALCTS Task Force. The ALCTS Task Force had defined the format variation problem as the multiple versions problem: "where the same expression of the same work can exist on different carriers (the "multiple versions" problem)" 58 However, given the insights from the FRBR model, the format variation question is not solely a question about how to approach the description of identical content on different carriers. It is also a question of how to approach the description of different expressions of the same work. The FRBR model identifies form of carrier as an attribute at the manifestation level, and form of expression as an attribute at the expression level. It is in taking the perspective of the FRBR model that one can see how "format variation" is a question of how to describe and how to give access to different expressions of the same work, as well as different manifestations of the same expression. The focus of the group was on the expression entity, and on the bibliographic relationships between expression and the other group 1 entities, especially between work and expression.
Pat Riva, a member of the FVWG, summarized the group's work:
The expression entity was the Group 1 entity least explicit in AACR2. The FVWG engaged in a three-year journey of reflection that eventually prompted proposals about headings for works and expressions that could serve as citations or identifiers for those entities, to clarify the structure among manifestations of a single work. The committee also considered how to provide a meaningful basis for grouping manifestations by form of expression or basic type of content; this work was intended to provide background towards reassessing the role, function and form of the General Material Designation (GMD).59
By 2002, the JSC was rethinking the general material designation (GMD). In September 2002, the JSC gave an additional term of reference to the FVWG that clearly indicates this intention:
JSC has been exploring the feasibility of "deconstructing" the GMD through the use of a term or device in the bibliographic record to represent the mode of expression, and moving terms representing physical format and form of carrier to area 5 or to notes ... It has been suggested that the expression-level indicator could be an element in the expression-level citation that is being proposed by the Format Variation Working Group. The Working Group is asked to develop these ideas and to make recommendations on how the mode of expression could be represented in the bibliographic record.60
The GMD is "a term indicating the broad class of material to which an item belongs" (AACR2 glossary). Once the concept of class of material was under scrutiny, as a result of Delsey's logical analysis, it was evident that the same inconsistencies also applied to the general material designation. As with class of material, if one looks at the terms used as GMDs, there are some terms that describe attributes at the manifestation level, such as microform, one at the expression level, music, and some at the work level, such as realia. Like the class of material concept, the categories are disparate and inconsistent.
The FVWG did some initial work on categorizing "modes of expression". It was evident that an extensible grid or framework was needed to replace the GMDs. The FVWG focused on the form of expression, and its relationship to the form of content. This work was happening at a time when the JSC was making the decision to change from amending AACR2 to developing a new edition, to be called AACR3. With the decision to move to a new edition, the JSC appointed Tom Delsey as the editor. The work of the FVWG, both on expression-level citations, and the preliminary work on ways to replace the GMDs, was forwarded, to the new editor and subsumed in the work on AACR3.
Tom Delsey had suggested in recommendation 1 of his Logical Analysis, Part 1, that a possible way to re-organize Part I of AACR2 might be to use the ISBD areas as the underlying structure for the rules, instead the existing structure of chapters based on class of material. As early as 1999, work had begun on an experimental "alpha" prototype of a reorganized Part I of AACR2, created by Bruce Johnson and Bob Ewald. The prototype simply rearranged the rules, but it was the first step in the "deconstruction" process. It took the rules out of the structure based on class of material chapters, and organized them according to the ISBD areas. The rearrangement highlighted some problems and discrepancies. An ALA task force, ALA Task Force on Consistency across Part I of AACR2, was asked to analyze the consistency of rules across the chapters in Part I. It took the process of "deconstruction" a step further. Taking the prototype of rearranged text, it was asked to look at the degree of overlap between similar rules originating from different chapters, and to identify inconsistencies, discrepancies or conflicts between these similar rules. The Task Force proposed revisions to increase consistency. The Task Force then consolidated their work on the rearranged rules and prepared another prototype for a reorganized Part I.61 The rearrangement of the text of the rules began a process of visualizing a new organization for the structure of the cataloguing code.
In 2002, a member of the Format Variation Group, Pat Riva, was asked by the Joint Steering Committee to examine the terminology used in AACR2 and to propose ways to make AACR2 terminology consistent with FRBR terminology. This process of incorporating FRBR terminology into AACR2 highlighted problem areas where FRBR terminology could not simply be grafted onto AACR2, but required rethinking a new approach. The task of incorporating FRBR terminology was an early catalyst for the alignment of the cataloguing code with the FRBR model, and an early step in the transformation of AACR2 into RDA. Work on this task continued until 2004, when the newly appointed editor of AACR3 was charged with continuing the work to incorporate FRBR terminology and concepts.
In April 2004, the Committee of Principals (CoP) and JSC decided that the degree of reorganization and changes required had surpassed the level of "amendments" and warranted a comprehensive revision of the rules. They named the new revision AACR3: Resource Description and Access.
A new draft of Part I was prepared. In the background comments from the JSC, there was a summary of the goals for the new edition:
The revisions being undertaken will entail a re-articulation of the function of the catalogue and a logical "deconstruction" of many of the concepts that underlie the current structure and formulation of the rules. In that context, the objectives established for the revision will entail aligning the rules with the concepts and terminology used in the Functional Requirements for Bibliographic Records, and re-examining underlying assumptions pertaining to concepts such as class of materials, main entry, authorship, and uniform titles. The adaptation of the concepts underlying the rules to accommodate the description of newly emerging forms of digital resources is a major issue to be addressed with the new edition.62
The proposed changes for AACR3 increased the integration of FRBR into the cataloguing rules and explored a new organization and structure. Objectives and principles were articulated and there was an intention to incorporate a theoretical framework in order to build cataloguer judgment. The division into Parts I and II continued to mirror AACR2's structure, with the addition of a third part for authority control. But there was a new structure for the chapters within Part I. The draft also demonstrated a new approach to class of materials and GMDs, where there was a conscious differentiation between the type of content and the type of medium. The class of material concept was in the process of being removed and replaced with a new, more logically rigorous and extensible framework for the technical and content description of resources. This was the first stage in a new approach to describing content, medium and carrier. But as the new changes were proposed and discussed, it became evident that the standard was moving in the right direction, but it had not yet gone far enough.
From Library of Congress's response to the AACR3 draft, March 2005 (5JSC/AACR3/I/LC response):
The Library of Congress supports the JSC in the initiative to improve the cataloguing rules and offers the following constructive comments towards that end. We would like to see the next edition of the rules go further towards achieving the objectives and principles stated for the rules. We would like to see more aggressive changes that respond to Web catalogues and future capabilities ...63
The British Library also voiced the need to move ahead more radically in their response to the draft (5JSC/AACR3/I/BL response):
The text as drafted in Part I is familiar to AACR2 users and there is great validity in generalizing, clarifying and improving this text to answer the continuing needs of these cataloguing users in our changing environment. However, in order to reach beyond the traditional AACR community and have the standard recognised and used by others, we think an additional, more radical product is required. One that is available on the Web, and that can be easily customised by whichever community needs instant content guidelines.64
In April 2005, the CoP and JSC announced a further change in approach that would shed the constraints of the AACR structure, move to a more complete alignment with the FRBR model, and aim to operate fully in the digital environment. The new standard was named RDA, Resource Description and Access, a new standard designed for the digital world.
52. For Part 1, the Logical Structure was drafted with the assistance of Beth Dulabahn, Michael Heaney and Jean Hirons, for Part 2, with the assistance of Beth Dulabahn and Michael Heaney.
54. Ibid, Part 1, recommendation no. 1.
55. ALCTS CC:DA Task Force on Rule 0.24. Overview and Recommendations Concerning Revision of Rule 0.24, (JSC document, 4JSC/ALA/30, August 16, 1999):
56. Ibid, 5.
57. Bowen, "FRBR: Coming Soon to Your Library?" 176.
58. ALCTS CC:DA Task Force on Rule 0.24. Overview and Recommendations Concerning Revision of Rule 0.24, 3.
59. Riva, "Introducing the Functional Requirements for Bibliographic Records," 10.