The photograph below is of the verso of the titlepage of a reprint of the Indian Army Department’s 1911 Instructions for Armourers. Don’t worry – that is the first and last you are going to hear about this excruciatingly unexciting book. The original was published by His Majesty’s Stationery Office. The publishers, D.P. & G. of Doncaster, want you to think they had permission to publish a reprint. How better than to show an ISBN?
If you enter this ISBN in a search engine it will return the message “Your search – isbn 1903972639 – did not match any documents.” How can this be? Are your fingers fatter than you think they are? The answer, dear reader, is that the ISBN is a fake. Perhaps the publishers didn’t write to the Controller of HMSO for permission to reprint. Perhaps they were too mean to send copies of the book to the four UK copyright libraries.
Other enemies of the ISBN are search engines like Amazon. Look for a book on Amazon – say one of Carman’s Indian Army Uniforms – and you will find two different books
Note the publication dates and other variations. One of these is wrong. But I wouldn’t waste your time getting in touch with Amazon to suggest a change to their catalogue. They are famously unresponsive. The problem is they copy other people’s errors. Thus, it’s not too difficult to invent the details for four exciting new books entitled Total Bollocks, Total bollocks, total bollocks, and TOTAL BOLLOCKS. You can’t any longer enter them on to the Amazon website, but you could in the past.
You could also give them fake ISBNs. Things really get interesting when a fake ISBN clashes with a real one. By mischance someone inventing an ISBN used a real one. Or someone intending to enter a real ISBN mistyped it. Either way the slot is now taken. The real ISBN can no longer be used. And if you are one of Amazon’s lawyers, I suggest you read some of the threads about this on the Amazon website which you can most easily get to by feeding ‘wrong isbn on book listing’ into Google.
The ISBN properly used is an invaluable computerised system matching the details of a book with a unique numerical identifier. The final digit of a ten or thirteen digit ISBN is a check digit enabling you to check whether as ISBN is true or false. Simply:
- Take the first 12 digits of the 13-digit ISBN
- Multiply each number in turn, from left to right by a number. The first digit is multiplied by 1, the second by 3, the third by 1 again, the fourth by 3 again, and so on to the eleventh which is multiplied by 1 and the twelfth by 3.
- Add all of the 12 answers.
- Do a modulo 10 division on the result from step 2. (Don’t know what a modulo 10 division is? It’s easy. It’s just the remainder from a whole number division by 10. I bet you learned to do that in junior school, before you even learned about decimal fractions.)
- Take that remainder result from step 4. If it’s a zero, then the check digit is zero. If the remainder isn’t zero then subtract the remainder from 10. The answer to that is your check digit.
Seriously – you and I can’t do all that; Amazon can. It’s a pity they don’t clean up their site by doing so for existing entries. The six steps shown above are lifted from https://isbn-information.com which is filled with interesting information.
And so to book notation which is a posh term for a set of symbols representing the classes into which you sort your books. On the domestic front most of us use letters to form simple words that we keep in our heads. Thus in my case I separate ‘history’ from ‘geography’ and have separate shelves for separate periods or regions. You could call it a notation system if I went to the trouble of writing the separate classifications on little bits of paper that I then stuck to the shelves. There are problems with (for example) my sister-in-law who, as a demographer, writes books that are neither fish nor fowl. But the system works well enough on a small scale like mine. Fiction (including poetry, lit crit and what used to be called belles-lettres) is arranged alphabetically.
The villain of the decision to put our fiction into alphabetical order, hereinafter referred to as Management, is depicted above. She decided to rearrrange our fiction shelves at about 2 am twenty odd years ago. I remember it well. Biography – arranged alphabetically by subject (so you need to know what you’re looking for) lives in her study. Her cookery books in the kitchen are in no order at all.
The case for the prosecution rests. Myself, within the confines of the notational system outlined above, I tend to arrange books by size, big ones at either end of the shelves, smaller ones towards the middle. I’ve known people who arrange books by colour and some even who use Dewey numbers (which we’ll be coming to in a while).
We have about 3500 books in our flat. So it should be easy to find any one of them especially since we have a classification system of sorts. Management currently wants to look at the Observer Book of British Mammals. Can the working classes find it? Can they hell. It might be under British Topography. Then again it might be under Science. Worst of all it might be hidden away in Miscellaneous.
Ashurbanipal’s scribes had the same problem back in 627 BCE. The King of Assyria had just conquered Babylon and come away with a fine haul of clay tablets. By the time the Babylonian loot had been incorporated in it, Ashurbanipal’s library had about 10,000 works on 30,000 tablets. I’ve never tried browsing tablets, but I’d guess it’s none too easy. The one below, now in the British Museum, gives the recipe for making blue and purple dyes for wool. It’s in remarkably good nick because Ashurbanipal’s library was burnt only a few years later, and most of the tablets in it were baked.
Fortunately, centuries earlier, Assyrian scribes had devised the world’s first known library classification system. Probably they improved a Babylonian system which derived originally from the Sumerians. We don’t know how they stored literature. Everything else, according to Thomas M Dousa in The Encyclopedia of Knowledge Organization, was “often listed in the following order — Astrological omens, Haruspical omens, Terrestrial omens, Teratological omens, Exorcists’ lore, Medical recipes, and Dream omens — though this sequence was by no means obligatory….” So heavy on science. You can find the rest of the article at https://www.isko.org/cyclo/library_classification
As you can read for yourself in Thomas M Dousa, there’s more to library classification than you might have thought. Still – and this is the point – the basic idea hasn’t changed. Confront an Assyrian scribe with the latest version of the Dewey System and he’d understand it fast. Of course, looking at classes 000 to 006 he wouldn’t begin to understand what “computer science” was let alone what any of the tens of thousand of books classified thereunder were about. But he would understand the system.
Time passed. Life became more complex. Literature proliferated. The Great Library at Alexandria was a hundred times the size of Ashurbanipal’s. It probably contained the best part of half a million texts. There was nothing comparable to it outside China. Almost all its texts were written on papyrus scrolls. There are limits to how many scrolls you can stuff into a pigeon-hole. Above is a picture of the old Mount Pleasant sorting office from learningroyalmail which used to be a wordpress blog. It illustrates just how much space pigeon holes take up. There are fewer than three hundred on view in this picture. With tens of thousand of scrolls in each of their main classes you can see why the librarians needed to invent sub-classifications and indeed sub-sub-classifications.
Callimachus wrote a detailed guide to the Great Library at the beginning of the Fourth Century CE. His guide is lost but is cited often enough by other writers for Encyclopaedia Britannica to assert that it contained the “following divisions: rhetoric, law, epic, tragedy, comedy, lyric poetry, history, medicine, mathematics, natural science, and miscellaneous.” Callimachus’ guide is called Pinakes which is the plural of the word pinax meaning tablet. Presumably when that grizzled old Macedonian general Ptolemy Soter first conceived the idea of copying some of the texts Alexander the Great had looted from Persia, he brought some Persian scribes to look after them. And they used the notation system with which they were familiar, putting clay tablets with classification marks on top of each set of pigeon-holes. Below is an image of Ptolemy’s son, Ptolemy II Philadelphus, the man who most people think founded the Great Library, from a bronze found in the Villa of the Papyri at Herculaneum, and now in the Naples Archaeological Museum where it was photographed by Marie-Lan Nguyen.
There are many modern libraries bigger than the Great Library but they are of the same order of complexity. The problem modern librarians have is the same as the one that confronted Callimachus and his friends – how do you find an individual book within a given sub-sub-classification?
Almost certainly the full text of Callimachus’ Pinakes would have provided an answer. Unfortunately mediaeval monks, whose monasteries had libraries numbering in the hundreds rather than the hundreds of thousands, had little interest in the parts of his book where the question is discussed and so didn’t preserve them by copying them. Nor, for reasons that are more complicated to explain, did the Arabs. In our dark ages the Arabs had libraries approaching the size of the Great Library. The library at Cordoba is reckoned to have had 400,000 books and a 44 volume catalogue.
We do know a bit about how private libraries were arranged at various times during the time of the Roman Empire. Augustine of Hippo had a vast library which is mentioned frequently both in his own writings and in the writings of his biographer Possidius. Cicero is another famous Roman who amassed scrolls. Here is the statue of the great orator outside the Palace of Justice in modern day Rome.
But actually the man whose statue I should be showing you is Tiro, Cicero’s trusted slave, later freed. Tiro, Wikipedia will tell you, is the inventor of shorthand. But –
πρῶτός τε κολυμβήθραν θερμοῦ ὕδατος ἐν τῇ πόλει κατεσκεύασε, καὶ πρῶτος σημεῖά τινα γραμμάτων πρὸς τάχος ἐξεῦρε, καὶ αὐτὰ διὰ Ἀκύλου ἀπελευθέρου συχνοὺς ἐξεδίδαξε
Sorry about that – I couldn’t resist it – and yes, you’re right, I can’t read Greek either. I lifted the quote from http://en.antiquitatem.com/tironian-notes-shorthand-ampersand/ where Antonio Marco Martinez discusses Cicero. Martinez, who can read Greek, took it from Dio Cassius, who is telling us here that Cicero wasn’t the inventor of shorthand at all, and that really it was invented by Aquila, a freedman of Maecenas.
No matter whether he invented it or not, Tiro was highly proficient in shorthand. He used it to note down Senate proceedings, to record Cicero’s speeches in court and to copy the personal letters he dicated. Then he wrote the whole lot up on parchment scrolls which he pigeon-holed. And – what is material to my argument – he could locate any one of these hundreds of scrolls whenever Cicero wanted to check a fact. We can’t be sure how he did it. But I’m going to follow Robert Harris’ lead and suggest he arranged them in date order according to who were the two consuls in the relevant year. Any educated Roman knew the list of consuls. And if you don’t know Robert Harris’ trilogy, you ought to. Unlike me, Harris is exciting, well researched and coherent.
About the above – Management has instructed me to point out that the reason I was able to find these books and photograph them was because they were arranged alphabetically. Sorting alphabetically by title within sub-classifications was done in the Sorbonne in the 13th Century but grouping book in any sort of alphabetical order wasn’t at all common until the 18th Century. Samuel Taylor Coleridge didn’t like it and who is Management to argue with him? (Collected Works – Marginalia: Part 6. Valckenaer to Zwick – find the reference yourself; life is too short.)
To continue – Cicero was hugely influential. Already, by the time Vesuvius erupted in 79CE, miserable school children were writing graffiti such as ‘like Cicero or be whipped’. As Shrikant Yelegaonkar notes in Western Thinkers in Political Science, St Augustine credits his conversion to Christianity to reading Cicero’s Hortensius while St Jerome had a nightmare “in which he was accused of being ‘follower of Cicero and not of Christ’ before the judgement seat.” When the printing press was invented, Cicero’s De Officiis was second only to The Gutenberg Bible to be set in type.
Little wonder then that Cicero’s (or rather Tiro’s) ideas about arranging libraries were equally pervasive. Listing titles in order of accession became the norm. The substitution of books for scrolls made little difference. Instead of sets of pigeon holes with notations indicating the sub-classes that were contained in them, there were book cases with similar notations. Richard of Fournival who died 1260 amassed 162 manuscripts, one of which is depicted below. In his Biblionomia he classified them into three main classes – Philosophy, Lucrative Sciences and Theology. Within Philosophy were nine sub-classes – Grammar, Dialectic, Rhetoric, Geometry and Arithmetic, Music and Astronomy, Physics and Metaphysics, Metaphysics and Morals, Melanges of Philosophy, and Poetry. The lucrative sciences were medicine, civil law and canon law. There is more about Bibionomia at http://www.historyofinformation.com/detail.php?id=253
And so to Umberto Eco and The Name of the Rose. The maze-like library on the top floor of the Aedificium in the fictional unnamed abbey he invented is very much larger than any real library of the time. But it is organised in a way that any mediaeval librarian would find easy to comprehend. William of Baskerville, investigating the death of an illuminator, has to deduce that “the books are arranged according to the country of their origin” but the librarian, Malachi of Hildesheim, has been taught how to navigate the maze and thus understands the notations in the catalogue. There are several main classifications by region. You’d expect to find books by the Anglo-Saxon Bede somewhere in the area called Anglia. Other books in Anglia are less obviously placed. An earlier librarian, Jorge de Burgos, whose name – not at all coincidentally – resembles that of Jorge Luis Borges, has also placed in Anglia books by authors he thinks should have been born in Anglia. Tricky. But entirely typical of librarians who enjoy nothing more than telling you why the book you can’t find isn’t where you expected it to be.
Eco’s classification by region is a little more complicated than I’ve said, but I don’t want to say any more in case you haven’t read his book. It works pretty well. It is what is technically called hospitable meaning that each classification can be expanded by the addition of extra sub-classifications. Nuclear physics, thanks to Rutherford, would probably have a room of its own in Anglia. The discovery of America might have posed bigger problems. But I’m joking. Eco’s system works.
Within the weird classifications the library is arranged exactly as a mediaeval librarian would expect. Different rooms in Anglia represent different sub-classifications. Different bookcases – most rooms have four, and unusually for the period, they are placed against walls – contain different sub-sub classifications. The bookcases have separate shelves. Presumably the books are kept flat but we are not actually told.
The difficulty then is not knowing where a book is. The catalogue will give you that information, though you may be discouraged from looking there. Let’s suppose you’ve been given permission to do so. A book in Anglia may be referenced A ii b iii in the catalogue. Hey presto – you know exactly where it is. Reading backwards it is on the third shelf (iii) of the second bookcase (b) of the second room (ii) of the region known as Anglia. The difficulty lies in finding the room. Therein lies the skill of the librarian.
We’ll leave Umberto Eco’s library with a view of the portal to the abbatial church. This is actually the Romanesque portal of Moissac on which Eco based his fictional portal.
Clearly the primary function of Umberto Eco’s library wasn’t the dissemination of knowledge. Forty years earlier, Jorge Luis Borges had invented his Library of Babel where, practically speaking, it is impossible to find a book at all. Borges’ library consists of an infinite number of interlinked hexagonal rooms which between them contain all possible 410 page books using the same characters and the same format. His maths was slightly awry since this library wouldn’t formally speaking be infinite, but this is a pedantic objection since it would be considerably larger than the known universe.
I have neither the time nor the patience to explore all the nuances of Borges’ library. (If you are interested in this sort of thing go to https://libraryofbabel.info/ and press random.) But I do want to introduce you to one of its denizens. This is the searcher, who knows that somewhere amidst the meaningless drivel there has to be a perfect catalogue which will enable the finder to navigate the library. Alas, there are also any number of imperfect catalogues.
Which brings me to The Dewey System. Melvil Dewey (1851-1931) was the most influential person in American librarianship in the late 19th and early 20th Centuries. He had many other interests, including spelling reform, hence Melvil rather than Melville as he was christened. His interest in librarianship was already apparent while he was an undergraduate at Amherst when he founded The Library Bureau to furnish “libraries with equipment and supplies of unvarying correctness and reliability.” The dimensions of the index cards sold by the Library Bureau are still standard. He also founded a less successful American Metric Bureau.
Amherst was sufficiently impressed by Dewey to ask him to undertake the reorganisation of its library. He devised a new classification scheme by freely adapting the taxonomy Sir Francis Bacon had proposed in 1605 in The Advancement of Learning and forcing it into the procrustean bed of a rigid decimal structure. Dewey copyrighted his system in 1876 and later expanded and improved it. He never denied his debt to Francis Bacon who is thanked in all the early editions of the System. Equally he never mentioned his debt to William Torrey Harris, Superintendent of the St Louis Public Schools, who had proposed a similar classification system based on Bacon and published it in the 1870 Journal of Speculative Philosophy.
Dewey’s USP was his decimal classification. He was obsessed with 10s. As you can read at https://americanlibrariesmagazine.org/2014/03/24/melvil-dewey-compulsive-innovator/
“At the Lake Placid Club, the cooperative resort in the Adirondacks that he established in 1895, the number 10 ruled. He charged members a $10 annual fee. (Lifetime memberships were $1,000, or $103.) He also required that guests turn off their lights at exactly 10 p.m.—the same time that the overnight train from the club left for Manhattan. This was a man who was addicted to writing 10-page letters and even preferred to “sleep decimally”—that is, 10 hours each night.”
All knowledge, Dewey decided, could be squashed into ten main classes. Each of the classes could be divided into ten sub-classes (divisions) and each of those could be divided into sub-sub-classes (sections). Here is the titlepage of his second edition
And by-the-by we have Dewey to thank for the horrid American spelling of catalogue. The main Dewey classifications now are:
- 000-099 Computer Science and General Knowledge
- 100-199 Philosophy and Psychology
- 200-299 Religion
- 300-399 Social Science
- 400-499 Languages
- 500-599 Science
- 600-699 Technology
- 700-799 Arts and Recreation
- 800-899 Literature
- 900-999 History and Geography
All well and good. The trouble is Dewey was a racist – later in life he founded the Lake Placid Club which banned any “member or guest against whom there is physical, moral, social or race objection….” He was also anti-semitic, profoundly anglo-centric, and a serial sexual harrasser of women who should have been outed long before 2019. These charmless attitudes show in his sub-classifications (divisions).
Take religion. Sub-sub-classifications (sections) 200 to 219 concern religion in general and the philosophy and theory of religion. Sub-sub-classifications 220 to 289 concern Christianity. Sub-sub-classifications 290 to 299 concern other religions. You have to go to 297 before you find a single slot for Islam, Babism and Bahai faith. Defenders of the Dewey System will say that the system is hospitable and accommodates many more sub-divisions within 297 by the use of a decimal point. Thus the Koran appears with other textual sources under 297.1. So the Koran occupies part of sub-slot 297.1 while the Bible occupies all the slots between 220 and 229. A problem there, I think.
Take literature. The 810s belong to American literature in English, the 820s to English, the 830s to German, the 840s to French, the 850s to Italian, the 860s to the Iberian languages, the 870s to Latin, the 880s to Greek. The literature of the whole of the rest of the world is squashed into the 890s. Chinese literature appears under 895.1. You can see all 8509 works listed there by going to https://www.librarything.com/mds/895
Until his death Dewey retained control of revisions to his system, personally until the sixth edition published in 1919, thereafter through the Lake Placid Club Educational Foundation. In 1988 the Online Computer Library Center (OCLC) of Ohio acquired the trademark and associatted copyrights when it acquired Forrest Press. The last physical edition was the twenty third edition of 2017. Now DDC is only available as WebDewey.
It’s lucky the geeks at OCLC have control. Nothing has given the Dewey System more problems than the attempt to incorporate computing. In particular, Section 006 is in the middle of a schizophrenic episode, since it covers both artificial intelligence and natural computing. Thus 006.3824 is captioned ‘swarm intelligence’ while 006.3825 is captioned ‘artificial immune systems’.
Dewey really won’t do. It needs replacing. Or does it?
Do we really need a notation system as Byzantine as OCLC? The only books that need to be kept in classified order are books that are physically browsed or retrieved. Since the invention of the International Standard Book Number every new book already has a unique annotation, its ISBN number. The ISBN lists books in groups published by individual publishers. That’s not much help, but so what: you can attach any extra information you like to an ISBN. So you can use imperfect classification systems such as the Library Classification for Medium and Small Libraries or the Book Classification of Chinese Libraries or even the one invented by Brother Jorge of Burgos. These aren’t catch-all as OCLC still hopes to be, but it doesn’t matter; ISBN will come to the rescue when a book falls between the cracks.
And to end here is a nightmare for all librarians. Newmarket Library took advantage of Coronavirus to have their library deep-cleaned. The diligent cleaner rearranged their books into size order.
Thanks (I think) to the BBC for this photo. You can read the full story at https://www.bbc.co.uk/news/uk-england-suffolk-52412655
Breaking news – my son Thomas, who has Aspergers, has just blogged a cognominal cricket team. In it he explains why Marcus Tullius Cicero was called Cicero. Read all about it at https://aspi.blog/2020/04/28/all-time-xis-the-cognominal-clash/
To receive other blogs in this series go to the top of this page and press ‘BLOG’. On the new page go to the bottom right and press ‘follow’.