The Rule of Three

"The Rule of Three" - I love the name, it sounds like a Sherlock Holmes story! Nothing so exciting, of course: it is the name commonly given to a cataloguing rule from the Anglo-American Cataloging Rules, and it's been there since the 1970s. (Wikipedia seems to think differently. "Inherently funnier, more satisfying"? hmmm). As AACR2 is on its way out, so too is this cast-iron rule, and it can't be allowed to pass without comment.

Here it is :

21.6C2
If responsibility is shared among more than three persons or corporate bodies and principal responsibility is not attributed to any one, two, or three, enter under title. Make an added entry under the heading for the first person or corporate body named.
[from AACR2, 1988, but the rule predates this edition].

RDA, the new code which is due to replace AACR2,  takes quite a different view, abolishing the idea of a "main entry", allowing all the authors to appear in their own right in the catalogue, and leaving far more policies up to "cataloguer's judgment".

What was the reason for the Rule of Three? It always seemed a bit mean to cut the authors off arbitrarily. A book with one, two, or three authors would include an entry in the catalogue for all three authors: more than that, and they were abruptly relegated to an added entry for the first author, and not even a whisper of a mention for the others. The trouble with it, at least for academic libraries, was trying to stick to the rule in the face of upset authors (often members of the academic staff known to the librarians) who didn't take kindly to being airbrushed from the catalogue.

Cataloguing rules seem to be a mystery to non-cataloguing librarians, who probably think that the cataloguers in their institutions make all these things up as they go along. We have had librarians of long standing (caught between intransigent cataloguers and miffed authors) insisting to us that they have never heard of this rule. Making things up might have been possible when each catalogue was a separate being in its own institution, but these days everything is visible on the web and many records are shared. This saves a lot of time and money and avoids the waste of duplication. It also means you need the rules even more - all that saving of time and avoiding of duplication goes down the drain if you are then constantly altering policies to suit local needs or tinkering with individual records on a random basis (i.e. when someone complains).

Like many rules, it dated from the days of the card catalogue, and it was probably partly designed to reduce the typing and filing of catalogue cards. Three was as good a cut-off point as any. Technology moved on long ago, and the idea of the "main entry" is no longer so important in the catalogue (it's still useful for allocating suffixes to classmarks for shelving purposes). While there is some additional work involved in including all the authors of a multi-author work, it is not the laborious drawer-filling process it would have been in terms of extra cards. There was a bit more to it than that, though: AACR2's definition of a personal author is "the person chiefly responsible for the creation of the intellectual or artistic content of a work" (21.1A1), and once you get beyond three authors you might well wonder whether you can say this of all of them. The "main entry" concept is also still as relevant as ever in bibliographic citation, which can these days be derived directly from the catalogue.

I think I can safely say that of all cataloguing rules the Rule of Three has probably been the cause of most complaints over the years, at least in the academic libraries I have worked in. Heddwch i'w llwch, as we say in Wales.

How reliable are our reference sources: or, how old was Bernice Rubens?

'As a child I was simply a liar by nature ... I was happily at home with mendacity. It was less boring than the truth. My natural home lay in fiction.' - Bernice Rubens, "When I grow up" (Little, Brown, 2005), pp. 2-3.


We cataloguers spend a lot of time trying to identify authors, usually by using dates of birth to distinguish them from others with the same name. I sometimes wonder whether we worry too much about this: some users do not want this level of detail. On the other hand, it is useful to be able to attribute works to the right author. We use Library of Congress authority records where possible.

The whole vexed question of dates of birth and age seems to be a much more sensitive issue to a previous generation. A few years ago when I was the branch librarian in a northern Cardiff suburb a new system was introduced and we were asked to add dates of birth of our users. What a can of worms that opened! The reason was to get a picture of which groups of people were using the library, either as a prelude to reaching out to less well-represented groups or, if that were not possible, to concentrate suitable resources on those who were using it. We were also discussing what age-specific discounts or benefits could be offered to those over 60, and we needed to be able to identify who was and who was not 18, for permission to use the Internet (new to us then). We were not interested in individuals’ personal details, and, as librarians among you will know, librarians do not ever give such information to anyone else. Several people refused point blank to give their date of birth, some left never to return in a rage at being so insulted, one man shouted “so how old are you then?” and looked flabbergasted when all the staff in turn told him their ages. (Far more men than women objected). One lady told me she was 46, which I duly recorded, and then, overhearing me explain to someone else that there was no charge to the over 60s for a particular service, she suddenly aged by 15 years.

Another former Cardiff resident, the author Bernice Rubens, who died in 2004, is recorded just about everywhere, including obituaries in The Times and The Guardian, as having been born in Cardiff on 26^th (or sometimes 28^th) July, 1928. She attended Cardiff High School and what is now Cardiff University, where she read English. The Library of Congress gives two references for the source of her date of birth: Who’s who (1983) and the Jewish Yearbook (also 1983).  These seem to be cast-iron authorities: but are they? Who is the source of dates in such books? Should we really regard such works as authoritative?

Elderly current and former residents of Cardiff alive today will indignantly tell you that Bernice Rubens was older than she said she was, that they remember her at school/synagogue/locally and that they are sure that the 1928 date is wrong. After her death, one such gentleman wrote a letter to The Times, setting out his recollections of her as a child older than he was, giving his own date of birth as 1926, and challenging The Times to prove that the birth date in her obituary was correct.

Among items recently added to stock in Cardiff University Library are copies of the school magazine of Cardiff High School in the 1930s, and the one for 1938 caught my eye, as it contains an account by Bernice, surname spelled Reuben, Form VA3, of a school trip to Dunkirk (“not particularly famous for anything” at the time). It does not read like the work of a 10 year old!

It may seem trivial to bother about a harmless lie, but she did upset her school contemporaries by claiming in interviews that they were anti-Semitic. There certainly was an anti-Semitic element in Cardiff (as elsewhere) and no doubt there were indeed some misguided girls at her school. In the same 1938 magazine, another girl describes attending a summer school in Berlin, arranged by the local government union NALGO – “as guests of high officials of the Nazi regime”.  The propaganda speeches are duly reported, with admiration for the “wonderful physique” of the German people. The previous year’s magazine, however, had included an article on Zionism by another Jewish pupil, of whom there were many at the school at the time.

If you are going to drop a few years from your age you would be better advised not to make public allegations about your classmates: those years may not seem significant in the great scheme of things, but it means you are talking about (and upsetting) a completely different cohort of people. A few years is a long time in the life of a school, and attitudes did change between the late 1920s/1930s and the 1940s.

I am a great admirer of Bernice Rubens’ work, so when her posthumous memoir, “When I grow up” appeared in 2005 I read it with much enjoyment. With the Cardiff rumours at the back of my mind I wondered how successfully she could publish a memoir without specifying her age (especially difficult if wartime comes into it, as that is a watershed event for most people who lived through it). She does manage to write her story in a way that would not lead readers to question the chronology, unless they had reason. The timing of her period at university is the main weak point: she completed her three-year course, graduated and went to teach in Birmingham before the end of the war by her own account, although a birth date of 1928 would have meant that she was only 17 in 1945. She also includes a photo of a page from a newspaper article about "twenty-three years old Bernice Rubens" - it's undated, but she supplies the date, 1946.

I’m an amateur genealogist in my spare time, and with access to the indexes of vital records through the CyMAL-funded public library subscription to such tools as Ancestry (library edition) and Find My Past, I thought I would try to prove the truth of the rumours one way or another. I had already failed to find evidence of her date of birth, but now I tried again, armed with the extra “e” and without the “s”.  Her  name at marriage in 1947 is registered as “Reuben” (in her memoir in fact she says that her father's surname was Reuben); she and all her siblings were registered at birth as Reuben. It took me just a few minutes to track down her birth in 1923 – not only 5 years earlier than every source states, but also not even at the same time of year (unless her parents were very late with the registration), as the birth is registered in the last quarter (Oct-Dec).  Splendid – not only did she shed five years, she also acquired a summer birthday! (Why didn’t I think of that? My birthday is in gloomy November!) Surprisingly, her date of birth given at the time of her death according to the index is the 1928 one, so she really did succeed in rewriting her own history, if I and older Cardiff residents have got this right!

Intriguingly, in her novel “The elected member”, for which she won the Booker Prize in 1970, the main character, Norman Zweck, is a linguistically talented young Jewish man whose mother lies about his age to make his gifts look more impressive, even delaying his bar mitzvah for three years. Was this Bernice Rubens’ own private literary joke, or a clue for her readers?