Since Antonio Panizzi published the “91 Rules” for compiling the book catalog of the British Museum in 1841, cataloging codes have been in a continuous state of change.⁸ Charles Coffin Jewett adapted most of Panizzi’s rules for the Smithsonian Institution in 1850, Charles A. Cutter devoted several decades of the second half of the nineteenth century to developing Rules for a Dictionary Catalog, and Melvil Dewey’s Library School Rules (1888) reflected his work directing the Columbia School of Library Economy.⁹ In addition, codes by Klaus August Linderfelt and Frederick B. Perkins and a pamphlet of suggestions by the Library Bureau were also in circulation in the late nineteenth century.¹⁰

Contemporaneously, a committee (which included Cutter) of the American Library Association (ALA) prepared a “Condensed Rules for an Author and Title Catalog” in 1883.¹¹ However, within a couple decades, the rules had not prevented “considerable divergence in the practice even of libraries organized subsequent to 1883.”¹² Between 1901 and 1908, a second committee (again including Cutter) worked to develop a revised cataloging code “to bring about uniformity between its revision of the A.L.A. Rules, the 4th edition of Cutter’s Rules for a Dictionary Catalog … and the Library School Rules.”¹³ The committee also worked with the Library Association in Great Britain to harmonize the cataloging codes used in the United States and in the United Kingdom. The resulting Catalog Rules: Author and Title Entries (American edition) and Cataloguing Rules: Author and Title Entries (British edition) were jointly adopted by the ALA and the British Library Association in 1908.¹⁴

The necessity for all libraries to adopt a shared set of cataloging rules had become steadily more apparent as early union catalogs were created, and had taken on added urgency in 1901, when the Library of Congress (LC) issued printed catalog cards for titles it had received.¹⁵ As libraries across the country took these printed cards into their catalogs, their locally cataloged materials required entry and description according to the same rules as the titles cataloged at LC. Hence, the adoption of the 1908 rules was achieved after only a short review period, and with near-unanimity between the two largest library associations in the English-speaking world (separate British and American editions were issued, but with only minor differences in the rules). It was the first set of cataloging rules to achieve widespread acceptance in libraries in the United States.¹⁶

The 1908 code, and each code that followed, was limited to rules for descriptive cataloging. Although some of the earlier codes, including Cutter’s, included rules for subject entry, an English-language subject cataloging code for universal application has not yet been developed as of 2008.

After 1908, the LC introduced many changes and additions to the rules on an ad hoc basis, to address cataloging issues not covered by the 1908 Catalog Rules: Author and Title Entries. These changes and additions were issued to libraries that subscribed to the LC’s catalog cards, but “in the absence of any supplementary rules from the American Library Association since 1908, libraries … had to formulate their own rules, relying chiefly for guidance on rules issues occasionally by the Library of Congress, added to such deductions relating to practice as could be made from the printed cards as examples.”¹⁷ By 1930, librarians felt a need for a revised code to incorporate the LC’s revisions and reduce local variation in cataloging, so work began on an updated set of cataloging rules.¹⁸ The motivating idea for the revisers was the feeling that the 1908 rules had not been extensive enough, so that the revised rules would cover more circumstances that proved troublesome to catalogers—such as serials, anonymous classics, music, maps, pseudonymous works, and corporate authorship. The coverage of such fine details meant that the 88 pages of the original Catalog Rules: Author and Title Entries became 408 pages in the revision published in 1941.¹⁹ Furthermore, most of the justification for the rules came from prior use, or “precedent,” rather than any logical reasoning; many rules had exceptions; and inconsistencies in the treatment of different types of material were noted. The structure of the rules, consisting of compound complex sentences with few illustrative examples, made the application of the rules difficult.²⁰

The complexity of the 1941 edition (which applied only in the United States, as British librarians could not participate in the revision due to World War II) lead to dissatisfaction in some quarters. Osborn published a famous article, “The Crisis in Cataloging,” in 1941, in which he lamented the large backlogs in cataloging departments and predicted that an even more complex set of rules would further slow down catalogers—an irony in light of the new rules’ purpose of easing catalogers’ work by providing rules for more types of publication and issues of entry.²¹

Cutter had attempted to generate his rules according to the objectives of a catalog: namely, to allow the user to find a book, to show what the library has, and to assist in the choice of a book.²² However, the 1908 rules and their subsequent revisions had excluded the statement of these objectives. In response to the criticism of the 1941 rules, the ALA commissioned another revision (A.L.A. Cataloging Rules for Author and Title Entries) to simplify the rules and arrange the presentation so that the principles behind the rules would be more apparent.²³ This 1949 revision also eliminated rules for description of the book; only rules for entry were included. Despite these changes, the 1949 revision was also criticized for its complexity and unwieldiness.²⁴

Although the 1949 ALA cataloging rules had omitted rules for description, the LC published its own Rules for Descriptive Cataloging in the Library of Congress in 1949; the rules were originally prepared for internal use, but were published for the wider library community in order to provide guidance for librarians using catalog cards printed by the LC.²⁵ The 1949 ALA rules for entry and 1949 LC rules for description became known, respectively, as the “Red Book” and the “Green Book” from the colors of their bindings. The Rules for Descriptive Cataloging at the Library of Congress (RDC) were considerably simplified from the 1908 and 1941 codes, and were largely praised for this fact.

In light of the praise for RDC and the less positive reception of the ALA rules for entry, the Board on Cataloging Policy and Research determined to approach rules for entry in the same fashion that the LC’s rules for description had been developed: namely, “prepare the simplest code of descriptive rules which could meet the established needs.”²⁶ To begin the work of preparing the simplest code, the ALA engaged the services of Seymour Lubetzky, a librarian at the LC who had also worked on the RDC. Lubetzky first prepared a critique of the 1949 ALA rules for entry, called Cataloging Rules and Principles: A Critique of the ALA Rules for Entry and a Proposed Design for the Revision.²⁷ Lubetzky’s critique not only pointed out the flaws in the existing rules for entry, but laid out the need for establishing a set of principles from which an improved code could be built. It included his famous question, “Is this rule necessary?” to which the answer was often “no” because the determination of the form of heading or rule of entry could be discerned from a larger principle, without need for a specific rule.²⁸

According to Tillett, Lubetzky felt that the cataloging rules had become so complex because catalogers had lost sight of the reason for the catalog: to help users identify and distinguish among works that meet their needs.²⁹ Cataloging rules that expressed the principles defined by Cutter (and refined by Lubetzky) would of necessity be simpler, and would allow catalogers to create better catalogs.

By 1954, the ALA had decided to prepare a complete revision of both the Red and Green Books, and appointed a Catalog Code Revision Planning Committee to the task of overseeing and advising Lubetzky’s drafting of a revised code.³⁰ Over the next decade, many discussions about the revised code were held in symposia and in the pages of journals. Almost all discussion focused on the approach to cataloging presented by Lubetzky, whose work became the sine qua non of the new cataloging code. As Dunkin wrote in 1959, “The genius of Seymour Lubetzky now dominates our thinking about the catalog as completely as Cutter once did.”³¹

In this environment of serious contemplation of the principles by which works should be cataloged, LRTS was launched in 1957. Debate over cataloging code reform was not limited to the pages of LRTS, but the pieces presented in that journal form a useful record of the voices for and against reforming the cataloging code along Lubetzky’s plan. Although both pro- and anti-reform articles appeared in LRTS between 1957 and 1966, this paper concentrates on articles composed by librarians who had reservations about the Lubetzky code, as they expressed a concern for values that have been considered of less importance than those that motivated AACR. Because the articles discussed Lubetzky’s proposed code on its merits, a variety of perspectives (including some commendation of aspects of the proposed reform, as well as reservations about the changes) can be traced through the pieces under consideration.

LRTS in its first decade was not the research-oriented journal it is today. Rather, it was a forum for news and debate over the latest trends in technical services librarianship.³² Articles were frequently fewer than five pages long, and included reports from various ALA committees, opinion pieces, and even humor. Because of its nature as a professional round table, LRTS provided an opportunity for librarians to voice their concerns about developments in the revision of the cataloging rules outside of the formal structure of a research article or literature review. Throughout the 1950s and 1960s, discussion of current cataloging issues was published regularly.

The very first volume of LRTS, published in 1957, features a lengthy review of a symposium held at the University of Chicago in June 1956, called “Toward a Better Cataloging Code: A Review.”³³ An “unusually large” number of attendees (148) testified to the interest in the revision of the cataloging code among librarians, and the concerns voiced by some of the speakers foreshadowed the debate that would follow for the next decade.³⁴ While a number of speakers expressed enthusiasm for Lubetzky’s proposals to return to the basic principles of cataloging, most were concerned about the cost of recataloging items already entered. Angell delivered a more philosophical demurral.³⁵ Where Lubetzky wished to do away with all entries other than author and title, Angell preferred to retain form headings (for example, “Laws, statues, etc.” or “Anonymous classics”) as a natural entry (that is, an entry that a user would think to look under using his or her native intelligence). Angell also raised the point that both the Red (ALA rules for entry) and the Green (LC rules for description) Books needed to be revised despite the general acceptance of RDC because the choice of entry influences how the name of the author may be described and because the descriptive rules should include provisions for media other than books. Osborn also urged that the new code seek to achieve harmony with codes of other countries.³⁶ Henkle (Lubetzky’s former supervisor at the LC) raised the issue of user studies; some librarians felt that data from the observation of nonlibrarian catalog users should influence the code.³⁷ All these issues would continue to be important topics in LRTS for the next two decades.

Another early article in LRTS supported Angell’s proposal to revise rules for description along with rules for entry. “The Red and the Green” by Waters of Georgetown University used a sample of publisher statements (of the RDC) to demonstrate the difficulty in determining proper description of that field according to RDC.³⁸ Waters felt that a review of the principles and goals that descriptive rules served should accompany the review of principles for entry, and that both set of rules should be revised simultaneously.

By 1958, Lubetzky had prepared a draft of a revised cataloging code, which was discussed by more than 175 librarians at the “Institute on Catalog Code Revision” at Stanford in July 1958.³⁹ As promised in his earlier works, Lubetzky laid out the objectives of the catalog as the first statement of the code: “1) To facilitate the location of a particular work; and 2) To relate and bring together the works of an author and the editions of a work.”⁴⁰ The similarity to Cutter’s objectives was noted––but Lubetzky had done away with another of Cutter’s principles: serving “the convenience of the public” (in the sense of deferring to the searching practices of users).⁴¹ This ambiguous phrase had led to many of the awkward, contradictory, and unintuitive rules in the 1941 and 1949 codes, such as entering certain types of corporate body under their location and the use of form headings. Instead of “the convenience of the public,” Lubetzky relied on logic in the observation that a simple rule, strictly followed, will become apparent to the catalog user and therefore serve him or her better than a maze of unexplained and inconsistent rules with ad hoc exceptions for particular circumstances. In this way, it was believed that the convenience of the public was served more effectively.⁴² To achieve the stated objectives, Lubetzky insisted on main entry under a name or title. No entries under location or form were to be made. Lubetzky’s draft code also addressed the contentious issue of corporate authorship by calling for entry of serials titles and corporate bodies that changed name under their successive names. A number of critics felt that this policy undermined the second objective.⁴³

According to a report on the Stanford Institute, which was the first public discussion of the draft code, a number of attendees questioned the value of Lubetzky’s second objective (“to relate and bring together the works of an author and the editions of a work”).⁴⁴ Many at the institute felt the draft code promoted excessive cross-entry, requiring more complex rather than simpler rules for entry. Wright questioned whether the code should consider subject entry as well.⁴⁵ The issue of the cost of converting the catalog to a new code was raised, along with the necessity of international cooperation on cataloging rules. However, the institute achieved a consensus on the notion of preparing the best code and then finding methods to achieve cost savings or international agreement afterward as the most productive approach. Further issues were raised, but left unresolved. These would continue to occupy the minds of catalogers as revision continued—the problems of corporate author entry (which circumstances require corporate, rather than personal, authorship; under successive or latest name; under subdivisions; under location) and serial title entry (successive versus latest title). During this period, Lubetzky wrote an article for LRTS explaining the process of code revision and his own reasoning behind the principles and rules used in the code.⁴⁶ After this, the task of defending the code in the page of LRTS against its critics fell to other writers. Following the Stanford Institute (and in preparation for another held in Montreal at McGill University in 1960), Lubetzky prepared yet another draft of the code.⁴⁷ In response, the pages of LRTS featured many critiques of the proposed rules changes, as reviewed below.

Although some of the issues debated were of a technical or practical nature, the issue of corporate authorship was one in which the values of the Lubetzky code stood in strong contrast to the values of the earlier codes. In particular, logic was pitted against tradition and user convenience, the latter referring to the sense that a catalog should have entries for corporate bodies where a user would look for them.

Draper of the University of California, Berkeley was dismayed that discussions of cataloging code reform (and the 1960 Draft Code) had not sufficiently addressed the problem of determining under which circumstances an entry should be made under a corporate author as opposed to a personal author.⁴⁸ He found the rule for entry under corporate body to be “highly vulcanized, i.e., full of rubber which can stretch in any direction at will,” because the wording of the rule allowed for much latitude in interpretation.⁴⁹

Haskins of Harvard University defended the to-be-discarded rules requiring entry of local or civic institutions under place by referring again to the convenience of the user:

From the standpoint of the use of the catalog the most direct approach would appear to be by the place where [the institutions] are located. Also, there would seem to be a real advantage in bringing together the schools, hospitals, churches, museums, etc., that generally may be of slight interest individually, but which play such a large part in the life of a city. If an institution bears a name that has little significance without the place where it is located, whether it be the Free Public Library, the First Church, Unitarian, Saint Paul’s Church, or Saint Luke’s Hospital, is it not logical to record it under the name of the place?⁵⁰

The Summer 1961 (vol. 5, no. 3) issue of LRTS featured a series of articles on the effects that implementing the 1960 draft code (now called Code of Cataloging Rules: Author and Title Entry, an Unfinished Draft, or CCR) would have on the operation of libraries. Dunkin of Rutgers presented an overview of the changes catalogers would have to make in the switch from the 1949 rules to CCR.⁵¹ He called it “Guesstimates Unlimited,” but only pointed out three major areas that would require significant changes in the form of entry: the use of a uniform title following a personal name main entry (a new idea first proposed in CCR); the elimination of the distinction between “institutions” and “societies” among corporate bodies, and the entry of all corporate bodies under name rather than place; and entry of anonymous works under title, rather than form. Dunkin offered suggestions for adapting the catalog to the new rules, such as using guide cards to provide cross-references from the older form of entry to the CCR form.

Wright of Williams College presented the results of a survey of catalogers who were asked to examine entries currently in their catalogs and determine if CCR would require changes in form of entry.⁵² Under the rules 70 percent of headings would remain unchanged, 13 percent would require minor changes, and 17 percent of headings would be different. Most respondents reacted favorably to the new rules as “more explicit, more reasonable, and easier to use,” although some expressed reservations about making such a large number of changes.⁵³

Haskins wrote—on behalf of the librarians at Harvard—in defense of many of the old ALA rules, including form headings and entry under place for corporate bodies, “What is to be gained by giving up this type of heading which has been in use over a long period and is generally understood and liked?”⁵⁴ She also found much to object to in the imposition of new rules, such as uniform titles combined with author main entry, changes in the form of foreign names, and successive entry for corporate bodies that change names—mainly on the grounds of the need to revise and update thousands of catalog cards, with little gained (in the opinion of Harvard’s librarians). She concluded with several thoughts about the flurry of cataloging rules changes that had come in the 1940s and 1950s:

I am beginning to wonder if we, as librarians generally and as catalogers specifically, know what we really want in the way of a cataloging code. We became dissatisfied with the 1908 code. For one thing it was too general. So a large committee made up of extremely able people worked for many years to revise the rules. The result was a very detailed code. In that respect it should have been the answer to a cataloger’s prayer. Perhaps it was for many. But within a short time, even before the second (1949) edition was published, it was on the carpet and was severely criticized for its complex rules, when the trend was toward simplification, for its lack of organization, its lack of basic principles, and so on. So once again we set to work. This time we started from scratch. … But from there on have we gone far enough or have we gone too far? Are we going to be successful this time? … We also started this revision by shutting out the past, closing our eyes to all the water that had gone over the dam. We have now come to the point where we can no longer disregard what has gone before. … How much can the large research library afford in order to implement rules that call for so many changes in practice?⁵⁵

Brown, of the Free Library of Philadelphia, wrote with concerns about the rules requiring uniform titles for works that appear under various names.⁵⁶ She preferred the entry as it appears in the work, whether it is title, corporate name, or personal name. Although it would create a “mongrel catalog,” her opinion was that users would be better served (particularly in a large public library) by reducing the number of “two-step searches,” which would be caused by the creation of uniform headings (step one was finding the proper heading from cross-references, step two was searching under that heading—a lengthy process when using a large card catalog).⁵⁷ Further, she found that a rigid application of principles should give way to a consideration of user behavior: “The Nibelungenlied, whether considered from the point of view of bibliographical characteristics or from the point of view of use, differs significantly from a recently published government document on jet propulsion. Consistency is a virtue in developing a catalog, but … [i]t need not be interpreted to mean that the same policy must be applied to all material regardless of that material’s bibliographical characteristics.”⁵⁸

Hines of Rutgers wrote with concerns about Lubetzky’s use of the term “work” instead of “book”:

The implication is that the work is to be considered as an intellectual rather than as a physical entity. … This distinction between the physical and intellectual cannot be pushed too far. It is clear that … Lubetzky does not mean that we should have a single main entry for Nine Plays of Bernard Shaw which would file with editions of Caesar and Cleopatra issued as physically separate bibliographic units. … It is here that a qualifying phrase seems to be needed in the draft code. It would appear that the code tacitly accepts the long-existing premise that the cataloger deals with physical bibliographic units, and that he catalogs them as such. … This preference for the physical bibliographic unit in cases of conflict [with intellectual units larger or smaller than the physical units] should be explicitly stated in the code.⁵⁹

Beckman reported the results of an experiment at the University of Waterloo in which CCR was used to catalog new acquisitions.⁶⁰ Although she found the “revised code a pleasure to work with,” and noted the ease with which her catalogers now addressed the names of authors, she did describe some difficulty in applying the rules for works of changing authorship, such as yearbooks and dictionaries.⁶¹ “The most difficult problem with this rule is that it is impossible to tell when handling a first edition of a reference work whether or not it will go into successive editions.”⁶² As well, the rules in this section diverged so far from current LC practice that Waterloo was unable to use, even in modified form, printed cards from the LC for those titles.

All such criticisms of CCR would no longer be addressed by Lubetzky; in 1960, he left the employ of the LC and accepted a professorship at the University of California, Los Angeles. The Catalog Code Revision Planning Committee turned over the job of editing CCR to C. Sumner Spalding. Lubetzky made one more important contribution to the revised code in the form of his role in formulating what became known as the Paris Principles.⁶³

As Osborn and others had noted, the American cataloging code revision was taking place during a time when librarians in other countries were also contemplating cataloging code reform. The destruction of many libraries in Europe during World War II made the possibility of revising cataloging codes more feasible because the number of books requiring recataloging was much reduced.⁶⁴ Although the possibility of international agreement on cataloging rules had been explored at the International Congress of Archivists and Librarians at Brussels shortly after the publication of the 1908 rules, those in attendance determined that differences between Anglo-American and continental (particularly German) rules were too great.⁶⁵ During the 1950s, a number of library associations—including those of France, Poland, Japan, Spain, Italy, Switzerland, the U.S.S.R., and India—worked on revised cataloging codes. The Library Association (of the United Kingdom) determined that it would work with the ALA so that the revised code being prepared by Lubetzky would be Anglo-American. In light of these developments, the International Federation of Library Associations (IFLA) convened conferences in 1958 and 1959 to discuss the possibility of an international agreement on cataloging principles. The result of these discussions was the IFLA International Conference on Cataloguing Principles (ICCP), held in Paris, October 9–18, 1961.

In Paris, representatives from thirty-four national library associations met and agreed on the Paris Principles, which served as the basis for future cataloging codes in most countries. Lubetzky’s contributions included articulation of the principle of main entry. Another important point of agreement was the principle of corporate authorship, which had previously not been observed in Germany.

The importance of the Paris Principles to catalog code revision was that revision could go forward with an internationally accepted set of principles underlying it and also provide strict guidelines. As Kebabian of the University of Florida commented, “There is no doubt that American librarianship will be under world-wide scrutiny as our new code reaches completion.”⁶⁶ As with Lubetzky’s code, the Paris Principles stirred up some criticism. For example, Scott of the University of Oklahoma found some of the guidelines to be too vague: “Consistent catalog entries for current materials cannot depend on ‘best known’ or ‘most frequently used’ form. Such criteria are useful only in retrospect.”⁶⁷

The question of whether the proposed Anglo-American code would go forward truly based on the Paris Principles was another concern. Kebabian noted,

Though the concept of the corporate author has been finding its way during the past ten or more years into French catalogs and bibliographies, for most national delegations this constituted the most fundamental break with tradition, and there was considerable debate at Paris before its final acceptance. In the form which was approved, moreover, it includes at least two provisions which contradict current United States practice. … It is ironical to reflect that these two principles were among those suggested by Seymour Lubetzky in his critique, Cataloging Principles, and that his studies and the preparation of that document stemmed from our desire to seek solutions to the inconsistencies of the “corporate complex” as structured in some seventy rules in the 1949 ALA code. While acceptance abroad at the ICCP was achieved, at home these principles constitute a problem of considerable consequence to achieve their reconciliation with existing entries in our long-establish, monolithic card catalogs; they are the one major source of yet unresolved compromise efforts in the current preparation of our code of cataloging rules. It is thus that the dead hand of history plagues us.⁶⁸

With the Paris Principles in place, the Catalog Code Revision Planning Committee continued to revise the cataloging code. An important agreement was settled upon at the 1963 ALA Midwinter Meeting in Miami. After the LC and the Association of Research Libraries “complained that they would be unable to pay the cost of changing the headings on cards already in their catalogs if the Committee followed the IFLA Paris Statement of 1961 which called for the entry of all corporate bodies directly under their names,” the Committee “decided to say plainly that the ‘institutions’ rule is an exception to the Paris Statement name-entry principle.”⁶⁹ Essentially, the parties agreed that entry for corporate bodies could continue under place. Without that agreement (which Dunkin called the “Miami Compromise”), the LC and large research libraries might not have adopted the new code, as it would have created an immense burden of recataloging.⁷⁰ In 1963, the committee decided that rules for description should be revised to encompass all media.

These steps toward completion of the new code may have alleviated for some catalogers the weariness with the lengthy process of code revision. As Symons wrote in 1962,

Any cataloging code must be a compromise between the principles of consistency and convenience (but whose convenience? Surely not the cataloger’s). There are bound to be areas of conflict. The exact place where the compromise is made seems to me not to matter very much. Rather than waste several more years of time and emotion and inaction, I suggest we encourage the publication of a Revised Code as soon as possible, so that we can all get on with applying in our libraries (or not applying it, if we really dislike it heartily).⁷¹

After the Miami Compromise, the Committee (working closely with British and Canadian representatives), labored feverishly to prepare the final edition of the rules, which was published in early 1967 as AACR.⁷² Although the rules conformed mostly to Lubetzky’s principles, some exceptions were present, particularly those involving corporate entry under place.⁷³ The committee recorded its regret “that, because of the great size of many American card catalogs, it was necessary for the Catalog Code Revision Committee to agree to the suggestions of the Association of Research Libraries that certain incompatible American practices be continued in the present rules.”⁷⁴ Lubetzky himself was disappointed that AACR omitted a statement of principles, on which he had based his draft codes.⁷⁵ For the most part, the catalogers accepted the new code and found its revisions worthwhile and useful.⁷⁶

This does not mean that criticism of the cataloging rules ceased in 1967. Indeed, a paper twice this length could be written about the critiques of AACR that led to its revision in 1978.

Most librarians using AACR today entered the profession after the code was published, and therefore accept it as the fundamental basis for cataloging. Further, many librarians believe that cataloging rules have improved over time so that the current rules most closely approach logical, principled cataloging. Nonetheless, AACR was controversial in its day, not least for the major upheaval it caused to previously created catalogs. The process of superimposition followed by many libraries in order to accommodate AACR attests to the wide-ranging consequences of such a thorough revision of cataloging rules.

The historical view of steady improving cataloging codes also feeds the library community’s own self-image as leaders in “modernity, progress, and science.”⁷⁷ However, many thoughtful librarians working during the days when AACR was being developed did not necessarily find the principles espoused by Lubetzky to be an improvement over then current practices. Some librarians felt that the values of tradition and user convenience were being disregarded.

An appraisal of the record will show that LRTS served as an important forum for discussing just how, why, and whether catalog code revision would truly make the catalog a better guide to a library’s collection, and that the library community was far from unanimous in regarding AACR as progress. As the cataloging community moves forward with revision of the current catalog code, it would be well-served by an examination of those values, both stated and unstated, that motivate such revision.