Index, A History of the 
by Dennis Duncan.
Allen Lane, 352 pp., £20, September, 978 0 241 37423 8
Show More
Show More

Indexes aretrouble. If you index your own work, you have to chew your cabbage twice, and then again, and again. You must reread the text that seemed so cogent when you sent it to the publisher – not to mention when you revised it, following the advice of your editor and referees, and when you answered the copy editor’s queries, and when you read the proofs. As you collect references across the span of your book, you are sure to find mistakes and inconsistencies. They are not only irritating: they provoke anxiety. Can they still be put right? It’s a race against the printer’s schedule. Yet if you pay someone else to make the index, you’re at their mercy. A professional will make a neater job of it than you can (or at any rate than I can), but can he or she really be trusted to find the most instructive categories and list the worthwhile references beneath them? Worse, you still face the annoyance of having your proofs read again, and your moments of ignorance, blindness and incompetence exposed and reported to you in bland, precise prose, by someone else.

Long ago I had dinner with six French copy editors. Authors, they told me, work by simple principles: never check a fact or a date, never verify a reference and never spell a proper name the same way twice. Would you rather have a professional expose these working practices or do it yourself? In either event, you’ll have to proofread the index too, with little time to spare. Friends are sure to turn up misspellings, and worse, while looking for their own names. Indeed, this has been a central function of indexes since the great days of white male writers, when William F. Buckley sent Norman Mailer a copy of his book The Unmaking of a Mayor and wrote ‘Hi!’ in the index next to Mailer’s name. Job’s comforters will want to be the first to tell you that the entry for ‘clocks and sundials’ in your index was printed without the first ‘l’ (this happened to a friend of mine) especially if they don’t find their own names there.

What is the making of an index to the consulting of an index? Scholars have recently become very interested in the historian of philosophy, Paul Oskar Kristeller (1905-99), who studied with the last great generation of German philologists. He himself told me two stories about those days: that he had learned to speak Latin properly in Werner Jaeger’s seminar in Berlin, and still admired Jaeger’s beautiful Latin; and that his teachers had trained him never to cite a book he hadn’t read from end to end. He viewed those who consulted indexes with a certain irony and would never have done so himself. But he also saw to it that his own books were equipped with detailed, intelligently constructed indexes, whose makers he always warmly thanked.

The index, as Dennis Duncan shows in Index, A History of the, has been around for a long time. Its roots stretch back to inventories, catalogues and the tags that identified scrolls, such as those of the Alexandrian Library, to tables of contents, which Pliny the Elder and other Roman authors of big books drew up to help readers find the scroll they needed, through to collections of biblical passages in which particular terms appeared, which 13th-century scholars composed to provide growing numbers of preachers, professors and students with ready reference tools so that they could produce theological treatises and sermons – the ‘impact’ component of high medieval scholarship – quickly and easily.

The index took what was nearly its definitive shape in the manuscript world of the 14th and 15th centuries. A table of contents could offer useful guidance to the structure of a work written on one or more scrolls, but there was no simple way to list and direct traffic to particular passages within them. The codex, which had elbowed the scroll aside in late antiquity, allowed for easy reference to every part of its contents. Scribes and scholars took some time to realise these possibilities. In a 1386 copy of Ranulf Higden’s Polychronicon, or history of the world, now in the library of St John’s, Oxford, the text is followed not only by an index, but also by an account of how to use it: ‘This number, 72, indicates that the heading from the table can be found on the leaf where 72 is written in the top corner.’ Unfortunately, the scribe copied the index – and its numbers – from another manuscript, in which the text had been laid out differently. The references didn’t work, until all of them except the one in the explanation were laboriously scraped away and replaced. The final innovation needed for the index to reach full power, accordingly, was a new form of book production: hand press printing, which produced more or less uniform codex books. The Nuremberg Chronicle, a massive compendium, splendidly illustrated with woodcuts, began with an index as its front page declared in immense Gothic type. The Latin version of the index, though not the German one, also came with instructions for its use: ‘The number that is on the same line with the name of each thing indicates the page of the leaves.’ All copies had numbered leaves, and the indexes sent all readers to the right places.

The index, like the antiquarian and the magus, was a creation of the later Middle Ages and the early Renaissance, and like them it had almost magical powers. Antiquarians claimed they could speak to the dead; magi claimed they could call them back to life; and indexes genuinely showed the reader where to find exemplary figures doing great things, as well as what they had said and written on any given subject. Antiquarians were satirised as pedants, magi exposed as charlatans. Good indexes actually worked as they were supposed to, but they still attracted plenty of criticism. Even the greatest experts on indexing – the scholarly print professionals who compiled them – feared that they might prevent readers from going through their texts with the proper energy and attentiveness. Konrad Pellikan, who compiled the index for Erasmus’s 1521 edition of the church father Cyprian, explained that he had gathered his lemmas quickly and keyed his references only to full pages, not to smaller segments of each page, ‘to avoid easing the way for gross laziness’. Conrad Gessner, who explained in detail how to compile an index on slips of paper, insisted that indexes were vital: without them, life would be too short to master more than one subject. But he also worried about those ‘who rely only on the indexes … and who do not read the complete texts of their authors in the proper order and methodically’. The index gave its users formidable power to find and quote adages and examples, narratives and poems, scriptural and patristic texts, whether or not they had actually read the full works they cited. That power in turn inspired anxiety, especially among those who had learned what they knew in the old-fashioned way, or claimed that they had. Would the index kill close reading?

By the 16th century, the index had established itself as what book historians would call an essential paratext, a standard piece of additional matter provided, in most cases, by a book’s publisher – a story that Duncan has told in the collection of articles on ‘book parts’ that he edited with Adam Smyth. Making them has remained one of the print professional’s crafts, as Virginia Woolf learned to her displeasure when she wore her eyes out compiling them. (Pulling together an index from a floor littered with little squares of paper wasn’t as gripping as setting type and sewing bindings.) But the index was also something more than part of a book: it was one of a number of tools for the crowds of semi-learned who wanted to gain quick mastery of languages, literatures and disciplines. These short cuts often played a paradoxical role in the life and work of scholars. Learned men, sometimes the most learned in Europe, compiled indexes and dictionaries. Erasmus indexed – and showed his secretaries how to index – his immense collection of ancient proverbs, the Adages, in multiple, complementary ways. Readers need only leaf through his copious aids to lay hands on proverbs that could be used to praise wealth or poverty, regal pride or pious modesty, as the occasion demanded. Henri Estienne, the 16th-century publisher and Hellenist, marked quotations in the margins of every Greek text known before assembling them in the Thesaurus graecae linguae, an index that I was taught to use in college (although I did not know then that he was the ‘Stephanus’ of whom my teachers spoke).

Sometimes the index could be a powerful epistemic tool. For ten months of 1601 and 1602, Joseph Scaliger put his own work aside to assemble 24 indexes to the enormous corpus of Greek and Latin inscriptions that his correspondent Janus Gruter’s publisher was putting through the press in Heidelberg. He didn’t want to do it. In fact, he claimed, his head exploded when Gruter’s publisher asked him to take on this enormous task. He told Gruter that index-making should be a job for ‘printers’ workmen or others with time on their hands’, not a great scholar without even a single research assistant, and dismissed the idea as typical of printers’ lust for profits. But the task grew on him as he worked. Reading through a thousand huge pages crammed with inscriptions, bent nib in hand, Scaliger came to see the special value of these idiosyncratic and often incomplete texts, carved by stonecutters who were not always highly literate and set in type by compositors who were sometimes less than competent. They preserved information about everyday life that wasn’t to be found in the stately narratives of the ancient historians. ‘There are so many fine wills and documents in these inscriptions,’ Scaliger told his pupils: texts that showed the way law and administration actually worked in provincial cities. The process of compiling taught him to imagine writing a new narrative about the ancient world, one built from the modest lives and careers recorded on stone.

Indexes continued​ to suffer disrespect, however. Great scholars took pride in not having taken short cuts to learning. Scaliger himself described in revealing terms the way he had mastered the languages of antiquity: ‘Beginning with a mere smattering of the Greek conjunctions, I procured Homer, with a translation, and learned him all in 21 days. I learned grammar exclusively from observation of the relation of Homer’s words to each other; indeed, I made my own grammar of the poetic dialect as I went along.’ A century later, Richard Bentley – the founder of historical philology who took Scaliger as his model – used his own granular mastery of Greek literature to show that the letters ascribed to the tyrant Phalaris must be a later composition. His critics at Christ Church replied not only by refuting his arguments but also by calling attention to what they insisted were his sources. Bentley ‘should have dug deeper for his materials, and consulted Original Authors’, they claimed. The sharp-tongued new Master of Trinity was not the heroic scholar he claimed to be, but a harmful drudge, dependent on thousands of slips that others had filled out and assembled into indexes – an accusation that must have stung all the more since its makers knew it to be false, even as they energetically repeated it. The paradoxical quality of reference tools – products of real erudition, they tempted their users to fake it till they made it – lent itself to polemical and satirical exploitation.

The same is true, as Duncan shows, of the index itself. As Thomas Babington Macaulay lay dying, he whispered one instruction to his publisher: ‘Let no damned Tory index my History.’ He knew what he was talking about. An index could distort, as well as facilitate, a reader’s understanding of a book. As early as the 16th century, the same scholars who knew the art of indexing backwards also learned how to turn an index into a distorting glass, which transformed the meaning of a text. And they combined it with other forms of learned writing to comic effect. When Erasmus read Alberto Pio’s polemics against him, he was especially annoyed by one of the indexes, which – in his view – pulled quotations out of context to misrepresent him as a heretic. Duncan describes the work Erasmus wrote in response as a satire in the form of an index. In fact, it’s more complex than that: the Brevissima scholia is a hybrid in which Erasmus summarised entries from the erring index and then added paratexts of his own – formal comments, which he called ‘scholia’. He did this, he said, for the many readers who read only indexes. An even better trick was played on Erasmus himself posthumously. The crafty and energetic Paris publisher Charlotte Guillard decided to reprint his edition of the Letters of Jerome. The Sorbonne had condemned Erasmus’s works, so Guillard (or a press professional in her shop) drew up an index of the passages in Erasmus’s commentary that most contradicted conventional Catholic wisdom. It took the form of a letter to the pious reader, warning that Erasmus’s paratexts ‘did more to cover the texts with darkness than to shed light on them’. Then it quoted a long series of passages in which Erasmus argued against practices including excessive veneration of the saints, supposedly in order to help the pious reader avoid them – but also, of course, to help the Protestant or critical Catholic reader discover them. A single index, creatively compiled, could make a big book respectable and subversive at the same time.

In the 17th century, these literary tactics moved from Latin into English. The literati who despised Bentley – Swift, Boyle, Pope – all enjoyed making fun of his dependence on indexes. As Duncan shows, they also turned the indexes of their own books into distinct satires, long series of mocking entries presented with a straight face. Bentley’s qualities, for instance, included ‘egregious dulness … Pedantry … familiar acquaintance with Books that he never saw’. No one did this sort of thing more skilfully than Pope. The index to his magnificent parody of the scholarly variorum editions of the classics, the 1729 Dunciad variorum, had fun with traditional literary history: ‘Pope (Mr) his Life] Educated by Jesuits, by a Parson, by a Monk, at St. Omers, at Oxford, at home, no where at all.’ Above all, it ridiculed the characteristic that Pope and his friends associated most with Bentley: the longest entry in the index was devoted to the goddess ‘Dulness’. Michael Baxandall pointed out long ago that those who have mastered complex tools often enjoy playing with them and applying them to new tasks. By the time everyone knew what an index was and how it worked, they also knew that it could serve many purposes for which it had never been intended.

In the 19th century, an age of vast scientific and historical systems, immense encyclopedias, and steampunk libraries with massive, quirky catalogues, the index took on what Duncan treats as a new set of ambitions: it tried to cover the universe, or at least great portions of it. He dwells on Sherlock Holmes’s fictional index: the multi-volume alphabetical compilation, built up by years of adroit work with scissors and paste, in which Holmes gathered all sorts of information, as well as narratives of his cases. But he also describes some of the actual projects of the period, such as the abbé Jacques-Paul Migne’s four-volume index to the Patrologia, his immense edition of the Greek and Latin Fathers, and William Frederick Poole’s Index to Periodical Literature. This suggestive chapter is necessarily incomplete: somewhere, near the Giant Rat of Sumatra, the ghost of Panizzi wails for recognition. But it also reveals one point on which Duncan’s lucid, lively account could be more informative – commonplacing.

Indexing flowered at a time when readers saw books as great mosaics of useful tags and examples that they could collect and organise for themselves. The chief tool they used to do this job was not the index but the commonplace book: a notebook, preferably organised by subject headings known as loci communes (common places), which provided both a material space in which to store material and a set of designators to help retrieve it. Like indexes, commonplace books were wildly popular. John Foxe, a corrector of the press during his exile years in Switzerland, printed two editions of a blank commonplace book designed to be filled in by users under Foxe’s various headings. One such user, Sir Julius Caesar, crammed his copy so full of notes and excerpts that the British Library classifies it as a manuscript. In his preface to the first edition, Foxe made clear that he knew he was riding a wave. He worked for Johannes Oporinus, the Basel printer who brought out the Magdeburg Centuries, a vast Protestant history built on commonplace books compiled by students. Foxe pointed out, with relish, that everyone who was anyone was making commonplace books and telling others how to do so. Humanists commonplaced to ready themselves for effective speech and writing, medical men, lawyers and theologians to prepare for their professions. Foxe argued that the commonplace book was vastly superior to the index. The index might point you to the materials you need, but the commonplace book actually provided them, handily organised. Holmes, like his contemporary Aby Warburg (another dab hand with scissors and paste), was a late master of the commonplace book, even if he called it an index. If printing made indexes easy to compile, commonplacing made them indispensable to printers and readers alike.

At one point, to illustrate the anxieties that attend new ways of encountering texts, Duncan wheels out (to use his term) the passage from Plato’s Phaedrus in which Socrates argues that writing makes humans inattentive and forgetful. He could have said a little more about the ambiguities of that dialogue and the ambivalences of its author. At one point, Socrates comes out with some ‘show-me-what-you’re-holding-under-your-cloak’ innuendo. He is both joking about the papyrus roll that Phaedrus has with him and admitting that he finds it charged with something like erotic appeal. In the Timaeus, a work less frequented by book historians than the Phaedrus, Socrates tells a contrasting story about Solon’s trip to Egypt. The priests, Solon said, mocked him and his fellow Greeks for their ignorance of the past. In Egypt, unlike Greece, the past, once inscribed in stone, wasn’t forgotten. Plato could see writing as a source of knowledge and a tool of criticism, as well as an object of lust and a threat to philosophy. Like writing and the printed book, indexes created excitement as well as anxiety, just as digital aids do now. Trouble is still their business – and we still can’t do without them.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 43 No. 19 · 7 October 2021

Indexes are trouble, Anthony Grafton writes (LRB, 23 September). He also makes clear how valuable they can be. Within a year of the invention of the Dewey Decimal System in 1877, W.H. Smith, then financial secretary to the Treasury, ‘was astounded’ to find several top graduates of the Staff College (usually sappers or gunners, temporarily in the fledgling Intelligence Division) combing obscure sources and subjects germane to their assigned part of the world and indexing it all into a 537-page quarto catalogue of seven columns with numerous cross entries to between five and nine divisions. This was a state of the art database capturing information ready to be turned into real-time intelligence for the Foreign Office (the War Office declined to involve itself with anything so tawdry as ‘intelligence’). Smith now understood why Lord Salisbury called the Intelligence Division ‘a most valuable department of state’ and offered to increase their grant whenever they asked. In the 1890s, when Edward Maunde Thompson, principal librarian and secretary to the British Museum, disdained to co-operate with the librarian of this unknown department, he received a gentle admonition from the Treasury; it was, he was told, ‘believed to be the best military library in the world’. Soon he, too, became a close, if confidential, advocate of the strange goings-on at 16-18 Queen Anne’s Gate.

William Beaver

Vol. 43 No. 20 · 21 October 2021

Anthony Grafton’s piece on indexes called to mind a particular favourite, the index to Hugh Trevor-Roper’s book from 1992, From Counter-Reformation to Glorious Revolution (LRB, 23 September). Some entries tell an entire story (‘Andrewes, Launcelot, learned and holy bishop: a cautious courtier … disappointed of Canterbury … cultivated by Grotius … orientalist … against Erastianism … but also high flying episcopacy’). Some are intriguing (‘Gawdy, Sir Thomas: his pew rudely curtailed’; ‘Ma T’ang, a terrible eunuch’; ‘Puddle, a rare bitch’; and ‘Richelieu, Cardinal de … believes he is a horse’). But the most striking entries are those that relate to Cambridge University. After his unhappy time in the Fens, Trevor-Roper must have enjoyed the entries for the university (‘protected from historical study … incurious of its own history … envious of Oxford’s library’) and for Peterhouse, the college where he had recently been master: ‘a noisy mafia there … unventilated Fellows … a liberal interlude’. The index as an instrument of instruction, intrigue, humour and revenge.

Tom Weisselberg
London EC4

Vol. 43 No. 22 · 18 November 2021

Anthony Grafton’s account of the rise of the index describes John Foxe’s printed blank commonplace book, first published in 1557, as ‘riding a wave’ (LRB, 23 September). That is being charitable: Foxe could also be accused of running a lucrative scam. In its expanded versions, his book consisted of a folio of six hundred leaves, each of which contained (on the recto side) printed headings, with the rest of the pages being completely blank, allowing purchasers to insert pithy quotations discovered in their reading. The book was then tied to a printed index in its final pages. In theory, this would allow users to create their own searchable archive – an analogue storage and retrieval device.

But, as users expanded the archive by inserting further blank leaves, the index as a finding device soon proved useless. Foxe’s system was also rigidly Calvinist: what would a Catholic user have entered on the pages reserved for Absurditates in doctrina Pontificia (‘Absurdities in papal doctrine’)? It’s true that the Master of the Rolls, Sir Julius Caesar, used Foxe’s book over a period of sixty years. But he was the exception. Many users seem to have abandoned it once its conceptual and ideological flaws became apparent, often using it merely as a source of paper.

What’s more, Foxe’s folio was competing with compact, readily available storage devices which were cheap, portable and fully functional. Printed almanacs were produced in their thousands in the period, and not only contained a wealth of useful information (when to plant crops, dates of holidays, phases of the moon etc) but were often sold at provincial markets and fairs with blank interleaved pages on which to jot memoranda. In effect, Foxe had produced a non-functioning mainframe device for an elite market, while all around him ordinary people were already using inexpensive hand-held devices.

Jonathan Sawday
Saint Louis University, Missouri

One troublesome use of indexes is their occasional weaponisation in academic feuds. A masterclass is provided by the Victorian medievalist J.H. Round, whose Feudal England (1895) was in large part a sustained and vitriolic attack on his fellow medievalist E.A. Freeman. The entry devoted to ‘Freeman, Professor’ is by far the longest in the index, taking up one and a half columns of its eleven double-columned pages, and has sub-entries such as: ‘his contemptuous criticism’; ‘when himself in error’; ‘his “facts”’; ‘confuses individuals’; ‘his pedantry’; ‘misconstrues his Latin’; ‘his confused views’; ‘his special weakness’; ‘his Domesday errors and confusion’. The final sub-entry might seem somewhat superfluous: ‘necessity of criticising his work’. And all this when Freeman had died in 1892.

Andy King
University of Southampton

An example of the woes of indexing known to many paediatricians was mentioned in the New York Times obituary of Waldo Nelson, author of the Nelson Textbook of Paediatrics, in March 1997.

The book was a family affair. Dr Nelson would call out items from each page as his wife and three children wrote them down on index cards. The children were not eager to help, but Dr Nelson insisted that it was a contribution to their education. In humorous retribution, his daughter Ann introduced a line in the index. Under ‘birds, for the,’ she listed the entire book, pages 1-1413.

Nelson insisted that the entry was removed from subsequent editions.

Norbert Hirschhorn
Minneapolis, Minnesota

In How to Make an Index (1902), Henry B. Wheatley, the founder of the Index Society, pointed to the dangers of a too-comprehensive index. William Prynne’s attack on the stage, Histrio-Mastix, might have escaped the notice of the attorney general had Prynne not supplied all the incriminating material in a convenient list. The entries include: ‘Devils, inventors and fomenters of stage plays’; ‘Heaven, no stage plays there’; ‘Herod Agrippa, smitten in theatre’; ‘Kings, infamous for them to act or frequent Playes’; ‘Plagues, occasioned by stage plays’; and, predictably, ‘Women actors, notorious whores’. The prosecutor quoted the index at Prynne’s trial.

Frances Post
London N10

For more than forty years I have been hoping for an opportunity to make the following index entry better known. It is from the Monthly Review, January-April 1802:

Interment, premature, of human bodies apparently dead, laudable institution for the prevention of, in Germany, 42.

Tony Lurcock

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences