I dreamed I saw Joe Hill last night,
Alive as you or me:
Said I, but Joe you're ten years dead;
I never died said he.
I never died said he.
And standing there as big as life
A-smiling with his eyes.
Said Joe, what they forgot to kill
Went on to organize,
Went on to organize.
The Ballad of Joe Hill
by Alfred Hayes and Earl Robinson (1925)
The story of cataloguing in the 20th century is the story of two structures. The first is that of codes and standards—from the 4th edition of Cutter's rules in 1904 through the Red Book, ISBDs, MARC, and AACR2. The second is that of the means by which catalogue records are communicated—from the book catalogues and cards of the turn of the century through microfiches, online catalogues, and Web-based catalogues. Both are a story of onward and upwards, but both are threatened by the bizarre millenarianism of "the end of history" crowd. In their view, forms of catalogue are irrelevant since all forms of human communication will be swept away in favor of digital communication (and those digital documents will, mysteriously, catalogue themselves). Also, to them cataloguing standards are unimportant since they believe they do not apply to Web sites and the rest. Such views are not only wrong but also noxious because, though masquerading as progressive, they are impeding progress. Digital communication is an important development, but it is not a unique and obliterating development, as any historian of communication will attest. Contemporary cataloguing standards not only can be used for digital resources, but are also greatly superior to the generally ill-considered proposals that are advanced as answers to the wrong question. "How should we catalogue electronic resources?" is not an important question. "Which electronic resources should we catalogue and how shall we preserve them?" is. The reason is that effective cataloguing involves controlled vocabularies and adherence to the standards that have evolved in the past 100 years.
What, fundamentally, is the topic of this conference? It is the idea of describing assemblages of recorded knowledge and information in terms of their titles, editions, issuers, date, extent, etc.; of adding formalized names and titles to those descriptions that allow library users to retrieve and collocate those descriptions; and relating them to a location—physical or in cyberspace. That is it, but one might as well describe chess as a game in which 32 pieces of wood or plastic are moved on a cardboard checkerboard according to prescribed rules. True, but scarcely an explanation of the fascination of cataloguing or chess or of the unlimited permutations and problems to be solved. Charles Ammi Cutter knew this, when he wrote, in 1904, of cataloguing's "… difficulties and discussions which have furnished an innocent pleasure to so many …" Cutter made that remark in the context of his idea, widely shared, that the advent of the LC printed catalogue would resolve all those difficulties and discussions and that the energy hitherto put into cataloguing would be diverted into "other parts of the service—the children's room and the information desk, perhaps." We all know that his idea did not exactly come true and cataloguing in the 20th century turned out to be full of difficulties and discussions.
Just as in the early 1900s, there is a tendency today to belittle the importance of descriptive cataloguing, even by people, who should know better (I shall deal with them later.) The difference is that Cutter and others thought that cataloguing had been perfected, whereas the naysayers today believe that cataloguing is irrelevant.
Issues in 1901
Let us begin by looking back 100 years to the cataloguing issues that preoccupied our long-dead colleagues in 1901. C.W. Andrews' paper at a meeting of the Illinois Library Association in February 1901 dealt with issues raised by the emerging cooperative cataloguing system. He welcomed the economy, fullness, uniformity and legibility of the printed cards. In a back to the future moment, Mr. Andrews opined "… the effect of this plan will not be to deprive catalogers of their work, but to substitute the intellectual for the mechanical …"—a point that I found myself making over and over again, also in Illinois, some 75 years later. The Advisory Committee on Cataloging, formed by the ALA Publishing Board, met in Atlantic City in March 1901, Cutter and other luminaries in attendance. They considered the typography of printed catalogue cards, discussed and rejected a proposal that the contents note be given after the title (a far more radical change in order of bibliographic data that anything envisaged by MARC or the Dublin Core), and stated that full names were more important in headings for English authors and in large libraries. The committee agreed that headings should be those to be found "…where the average person using a library is apt to look …" The report gives this as a throwaway line, seemingly without consciousness that that the great fault line between "correct" headings and "sought" headings was being established by this reference to the needs of the "average person." They then moved on to wrestle with corporate entries—generally agreeing with Cutter's rules (then in their 3rd edition), which stipulated that some corporate bodies were to be entered under name and some under place. This innocent seeming decision was to bedevil cataloguing and cataloguers for the next 77 years. They decided that there should be a more extensive use of birth and death dates in headings. The most divisive issue was that of measuring the size of books. The Committee could not choose which of three proposed methods (letter symbols, fold symbols, or exact size in centimeters) should be used, but did decide to write minority reports on each. The report states: "The committee has been impressed with the practical agreement of its members on cataloging rules, upon the willingness to yield on inessential points, and upon the idea that the catalog should be made for the user, not for the cataloger."
Before we succumb to the temptation to see Cutter and his colleagues as quaintly old-fashioned, let us remember that they were far more advanced in the standardization of cataloguing the materials of their age, and in cooperating, than we are in dealing with the electronic documents of our day. In fact, the parallel with our time is the situation in the late 18th century when the French revolutionaries hit on the idea of using the blank backs of playing cards to record the holdings of the aristocratic libraries taken over in the name of the people. (On reflection, that is a far more organized and coherent scheme than anything we have done for electronic documents to date.) The late 18th century was a time of chaos, to which a few brave souls tried to bring order, one small step at a time. Ours is a culture in chaos—a time of beleaguered learning and of threats to the records of humankind. We too need a few brave souls, and should applaud those who try to bring real cataloguing to bear, while defying those who want to capitulate to the fecklessness that disregards standards and bibliographic control, on the irrelevant and dubious grounds that electronic documents are transcendent and transformational.
In 1908, committees of the ALA and the [British] Library Association published Catalog rules: author and title entries in two editions, thus setting an unfortunate precedent that lasted until 1968. The committees were unable to agree on all rules, both between the US and the UK, and between themselves and LC practice. LC was more robust in those days and, rather than issuing Rule interpretations that contradict the rules, you will find in the 1908 rules flat statements of LC practice that differs from the rules in matters great and small. Thus, a British or American cataloguer would, for the next 40 years, have to choose, in many instances, between British rules, American rules, and LC rules. In North America, in which the LC card was to dominate cataloguing for at least the next 80 years, there was a strong tendency to follow LC practice and to hell with the rules to which LC practice was an alternative.
The 1908 code was dominated by cases not conditions and principles. This arose because of the bilateral nature of its origins and construction. Nineteenth century codes were almost all the product of single individuals (Panizzi, Cutter, Jewett, etc.). Despite the fact that Cutter was active on the American committee until his death in 1904, his was but one voice among many, often eminent others. Lacking a guiding hand and a single set of unified principles, this, the first of the committee codes that have lumbered through the 20th century, was an assemblage of the best practices of Anglophone libraries. It was inevitable that such a code would be based on cases and ever more minute distinctions between cases. The latter reached its reductio ad absurdum in the full page devoted to the rule on Exploring expeditions, with its two subrules, the second of which has 6 sub-sub-rules. The 1908 code set in train a period of code making that was to lead, inevitably, to calls for reform from Andrew Osborn and others.
There was an interesting statement of American cataloguing practice in the inter-war years. The Vatican Library published a code of rules in 1931 that was later stated to be "… the most complete statement of American cataloging practice." The Vatican code was notable for a number of reasons. It included rules on name and title entry, description, subject headings, and filing—the only code since Cutter to do so. Had it not been for the Second World War, it is quite possible that work on the Vatican code would have taken the place of the work that led to the abortive 1941 draft rules and the unmitigated disaster of the 1949 ("Red Book") rules. Perhaps, however, it was not just the course of cataloguing that was changed irrevocably by WWII.
1941 and Osborn
The tenuous connection between North American and British cataloguing committees broke entirely with the publication of the 1941 ALA cataloging rules, preliminary American second edition (the British being largely occupied with other matters in 1941). The main contribution of the 1941 draft rules to cataloguing history was the reaction it provoked in the Australian librarian Andrew Osborn. His The crisis in cataloging called, in essence, for fewer, simpler rules based on principles and ignoring non-essentials. He also called for codes that allowed cataloguers to use their judgment based on experience and, again, on principles. Perhaps naively, he thought that such cataloguers would win more respect from library administrators. Osborn's important article appeared to have faded into oblivion as the attention of the United States turned to World War II, which the Americans entered a little over a month after The crisis was published. After the war, the ALA/LC cataloguing committees carried on their work as if Osborn had never spoken. The result was "The Red Book" and "The Green Book" of 1949.
1949: ALA and LC
Almost all you need to know about the Red Book (The ALA cataloging rules for author and title entries) is summed up in the fact that one rule, 116A(3), is devoted to and only to the Basilian Monastery at Mount Sinai. This is the logical inevitable result of piling case upon case and splitting ever thinner hairs, all the while ignoring the principles or even the need for principles and ignoring the needs of catalogue users for clarity and consistency. After 1949, there were only two possible directions. Reform or progress toward a code that consisted of nothing but cases applying to tiny numbers of documents. Thank the Lord and Lubetzky, we embarked on the road—the long, twisty, and obstacle-ridden road—to reform.
Seymour Lubetzky, employed at the time by the Library of Congress, was the most prominent critic of the 1949 rules. His seminal work Cataloging rules and principles was subtitled "a critique of the ALA rules for entry," but might, like a latter day John Knox, have been entitled "A blast against the monstrous regiment of cataloguing rules." With the simplicity of genius, Lubetzky stepped away from the trees of exploring expeditions and Basilian monasteries and saw the forest of the cataloguing code in asking his famous question "Is this rule necessary?" Further, he asked of every rule is it consistent with principles and is it properly related to other rules. These questions gave rise to a draft code that was as spare and coherent as the 1949 rules were sprawling and incoherent. The Lubetzkyan revolution spilled over into the "Paris principles" which were thought, at the time, to be the framework for a universal cataloguing code that would revolutionize international bibliographic cooperation. Alas, reality intervened in the unholy alliance of traditionalist cataloguers (some not a million miles away from the Library of Congress) and library administrators (the forces of darkness then and in the War of AACR2) that caused Seymour Lubetzky to be replaced as editor of the code that was aborning—the code that was intended to unite North American and British cataloguing and usher in the Lubetzkyan age of global cooperation.
1968: two codes
The sad fact is that the code that resulted from this reactionary tide—the first AACR--was not only a major fudge betraying Lubetzky's ideas in many instances but also could not even reconcile British and American practice (the prophet Lubetzky was honored far more in the UK than on his native heath). Also, and crucially, AACR failed to deal adequately with what we used to call "non-books." The first AACR did have many strengths, and was a great improvement over its predecessors, but, ultimately, it represented a failure of nerve that has consequences to this very day. I cannot now give details of the good and bad rules it contained, but wish merely to emphasize that its failure was truly historic in that it failed to live up to its time—an era in which Universal Bibliographic Control became more than a dream. In that era, a unified English-language cataloguing code based on coherent principles would have saved us from a couple of decades of squabbles and confusion.
1968: MARC and ISBD
So, what happened? MARC and ISBD happened—both in the late 60s and early 70s. It should be unnecessary to point out that MARC is not a cataloguing content standard—it is a framework standard to which cataloguing content has to be added. I mention this only because I have heard and read so many of the metadata boys talking about "MARC cataloguing," a shockingly ignorant phrase that betrays the fact that the very nature of cataloguing is not understood in such circles. In MARC, the electronic version of the catalogue card as a carrier of bibliographic data, we had a means of communicating bibliographic data between nations and continents—a means that would be utterly ineffective without universally agreed standards for the content of bibliographic records. The International Standard Bibliographic Description (ISBD) provided a complementary framework to MARC, and some guidance on the international standardization of descriptive content. The stage was set for a totally new code for a new bibliographic world.
It is quite unnecessary to rehearse all the events and circumstances that led to the publication, in 1978, of AACR2 beyond saying that was modified and resisted because of pressure from the same reactionary elements that stalled the Lubetzkyan revolution. AACR2's very name is a fraud. This was by no means a second edition of the 1968 code, but a new code that should have had its own name, something that would have spared us much subsequent grief. Concern about the amount of change the new code would mandate mingled with a witches' brew of nonsense about the ISBD being "unnatural" and "foreign" (nonsense that is being repeated to this day) and pressure from special classes of cataloguer to produce a code that is encrusted with the unnecessary elaborations about which both Osborn and Lubetzky complained. There is a major difference, however. AACR2 is barnacled with unnecessary rules, but has a core of Lubetzkyan truth in its rules on entry and heading. Moreover, the descriptive rules, based as they are on the ISBD, have a coherent structure that is fair in its treatment of all media, and capable of incorporating any new media. These strengths were shown in the publication of AACR2R--a publication that caused none of the excitement of AACR2 but quietly demonstrated the strength of the underlying structure of the 1978 code.
In the 19th century, we had codes that were the product of individuals, for most of the 20th century, we had codes that were the product of committees. Then, in 1998, we reached the next step—a code that was the product of nobody at all! The 1998 publication (I scorn to call it an edition or revision) of AACR2 consisted of the 1988 revision with agreed amendments slipped in without regard for their implications elsewhere in the text. I suppose a virtual code was inevitable. Perhaps a robot called Hal will produce the next? Whatever that future may be, the 1998 publication is indicative of the parlous state of cataloguing codes at the close of this century—no battles, no excitement, but no progress either.
Perhaps metadata is the reason why progress on refining the cataloguing code appears to be at a standstill? Now what? It used to be AACR3? Now, the chatter is that all that matters are electronic documents and "traditional" standards cannot deal with electronic documents. Neither happens to be true, but that has not stopped the most abundant supply of hot air (from proponents of metadata) since the War of AACR2 and the great "ISBD is a foreign plot" debate.
Metadata, as I said previously, is a subset of the MARC record without the instructions on standardization of content necessary to create a bibliographic control system. Amusingly, the Dublin Core (like MARC) is based on the catalogue card in that it preserves the main entry (see the placement of the principal creator and subsidiary names). Metadata (a fancy name for an inferior form of cataloguing) also ignores the central question—what should be catalogued? No one is interested in taming all the vast wasteland of the Internet because most of that vast wasteland is worthless. The task is to identify the oases and to apply some version of real cataloguing to them. My belief is that a modicum of common sense is beginning to dawn. A recent discussion in the California State University libraries centered on the fact that many of our patrons were unaware of the many journals that are available online (at some considerable expense to the taxpayer). My mind had wandered to some more agreeable topic (utterly unconnected with libraries) but I was brought back abruptly when a consensus emerged that use of these journals might well be increased if they were … wait for it, listed individually in the catalogue!
So, here we are at the end of a century of erratic progress in cataloguing, facing the possibility that we will settle for tenth best, a weird amalgam of free text searching and unstandardized, uncontrolled, ersatz cataloguing masquerading as a branch of information "science." However, there is another possibility— that we will incorporate electronic documents into Universal Bibliographic Control, as we have incorporated all other forms of human communication and thus usher in another golden age of cataloguing that supports our unique task as librarians—the preservation and onward transmission of the human record.