What was T.S. Eliot’s favourite colour? Which season – summer, autumn, winter, spring – would you expect to feature most often in the works of Philip Larkin? And which of these two poets would you reckon was the more self-centred, fond of flowers, susceptible to hyphens, keen on using the word mother?

Such are the questions that can spin off from too many hours spent browsing in the realms of the Concordance. It so happens that both Larkin and Eliot have lately had their works ‘concorded’* – that’s to say, worked over with a scanner in order that the world might know how many and which words each of them used, how often, when and where. Don’t ask me why. A concordance is, I think, meant to assist us in the checking of quotations, the remembering of half-forgotten lines. In my hands, it becomes a kind of toy.

But then, I love such toys. I love compiling lists. A few years ago, when I was editing an Oxford Companion to 20th-century poetry, one of the rare pleasures of the job was in trying to work out how many of my poets had done what – apart, that is, from writing verse. How many had been policemen, say, or boxers, or management consultants? How many had played hockey for their country, been murdered, died in battle? How many had served time in jail? Who cares? you will retort, and I’d agree. Back then, though, for a day or two, I had to know these things. Again, don’t ask me why.

With some of my lists, I tried to kid myself that I was not just playing games. For instance: surely it was worth knowing how many modern poets died young, were alcoholic, or went mad? In these areas, it must be said, the lists I ended up with revealed little that could not have easily been guessed. And yet this hardly seemed to matter. Thanks to me, guesswork had been hardened into fact, statisticised. From now on, I could say, it was a matter of official record that the majority of modern poets had been not only sane and sober but also, survival-wise, extremely shrewd. Most of them, my final list disclosed, had spent their lives in universities and, come pension-time, been good for at least one further decade of ‘productive work’.

And this list, I need hardly say, led to another list, a list-within-a-list. Now and then, during the course of my labours, some churlish friend would remind me that only a small handful of my listed poets was, had ever been, could ever be, well, ‘any good’. Was it statistically correct to call these people poets, ‘just because they said they were’? In retaliation, I took to totting up the century’s ‘real talents’, as decreed by me. And this list, when I finally got through with it, ran to about two dozen names – from a total of something like a thousand. Disconcertingly, about a third of these real talents had, alas, gone mad, been alcoholic, and/or died at a (statistically) young age.

So where did that get us? Not very far – and in my case especially not far, since I had no wish to be persuaded, by arithmetic, that great wits are, or have to be, unstable. There was a clear need for yet another list, a list-within-a-list-within-a-list: one that would demonstrate, to my personal satisfaction, that insane/drunk/ill poets wrote their best work when sane/sober/in good nick. Yes, yes, I know. My statistics had turned into a chaos of begged questions. There was madness in my methodologies, and it was beginning to show through.

All the same, I can’t help loving these concordances. The mere look of them – page after page of headwords, numerals, page refs, original pub. dates and all the rest of it – is marvellously soothing. The Eliot concordance, done by a Cambridge team of literary and linguistic statisticians, is a giant volume of some 1250 pages (twice the length of T.S.E.’s Collected Works) – and this without full listings of and, the, this, that and other functional connectives (though it is vouchsafed that Eliot employed the – 6940 usages – nearly twice as frequently as he had use for and – 3926 – and that his thats – 2085 – almost contemptuously outrank his thises – 538. Who would have thought?). Larkin, whose output was about a third the size of Eliot’s, nonetheless concords at a hefty 660 pages. His Collected Works weigh in at 338.

When books as big as these land on your desk, you can’t just sit and smile at them. You feel a duty to get in there and start counting. Memory – 37 usages; desire – 25; waste – 26; land – 31; Jew – three; and so on. And since there are two poets on parade here, there is also a temptation to compare and contrast. Thus Larkin, you can now be told, used sex 11 times to Eliot’s none. Eliot, on the other hand, has three lusts where Larkin can manage only one. Mad is much the same: Larkin here scores three to Eliot’s nine. Of course, with lust and mad you have to bear in mind that Larkin’s figures ought really to be multiplied by three, so as to reflect his smaller output. There is no knowing what Larkin’s vocabulary might have got up to had he been three times more prolific; as things stand, though, the two poets must – on lust and mad – be rated neck-and-neck. Not so with know, which turns out to be Eliot’s all-time number-one word, scoring an impressive 677 to Larkin’s 42. Time is T.S.’s second-favourite. Here Eliot clocks in with 357 to Larkin’s laggard 74.

At a brisk flick-through, it might seem that Eliot’s word-hoard was on the whole more prosy and various than Larkin’s. To take just the ‘A’ section of the two concordances, there are Eliot-words like abnormal, absent-minded, agitate, absorption, allowance etc which don’t appear at all in Larkin’s poems. But nor do they appear in Eliot’s. These words are all taken from the plays. And the inclusion of his plays does somewhat falsify the Eliot picture. For instance, the Eliot concordance reveals 93 uses of the word mother (compared to Larkin’s six). But only three of Eliot’s 93 appear in poems. The others are spoken by characters in plays: ‘Is this your mother?’, ‘The mother of my child was Mrs Guzzard’s sister,’ or ‘I’ve never had a father and a mother.’ Similarly, about 50 per cent of Eliot’s 102 God mentions come from Murder in the Cathedral. God appears five times in the Four Quartets, and four times in ‘Ash Wednesday’ – which in each case is somewhat less often than I would have guessed. If, as is sometimes claimed, concordances expose authorial obsessions, we need to tread carefully with Eliot, the bulk of whose lexicon was deployed as dramatic speech.

Certainly, we need to be more nervous than I’ve been when comparing him with Larkin. In truth, even joke-comparisons don’t actually stand up, not least because the two concordances don’t follow the same editorial guidelines. The Eliot, for example, is much stricter than the Larkin in its exclusions of the poet’s ‘small words’ prepositions, conjunctives, articles, dead-common verbs. It leaves out high-scoring Larkin-words like you and I, could, come and did. Thus, the best I can offer, using words common to both listings, is the information that Larkin’s most-frequent usages – compared to Eliot’s know and time – are like (175), their (112), up (92) and down (75), which could I suppose be juggled into the beginnings of a sombre Larkin line.

In his Preface to the Eliot book, co-editor Peter Holland says that it is not his task ‘to in dicate the potential uses of the concordance’. He does suggest, though, that concordances can have ‘a peculiar magic in their consequential revelations’. He speaks also of ‘the serendipity of scholarship, the critical chance of conjunction’ and points with pride to ‘the intriguing juxtaposition of god-given and god-shaken’, as revealed by ‘our index of Words Containing Hyphen or Apostrophe’. He could happily have pointed elsewhere – to his ‘Reverse Index’, for example, where we get ‘investigate/margate/moorgate/gallowgate/hate/appreciate/associate/immediate/repudiate/humiliate/multifoliate/expiate/appropriate’, which makes Eliot sound like a mountingly ecstasised rap poet. Or, under ‘Statistical Ranking List of Word Forms’ there is (each word has 11 usages) the rather poignant observation that ‘perpetual/powers/prefer/priests/problem’. Mr Holland also directs us to a surreal pile-up under bark. ‘Computers have no sense of humour,’ he remarks, but ‘their output is sometimes amusing.’ So it is:

222Pekes Pols 8 they join in to the fray/And
they/Bark bark bark bark/Bark bark BARK BARK
222Pekes Pols 8 join in to the fray/And they/Bark
    bark bark bark/Bark bark BARK BARK/Until
222Pekes Pols 8 in to the fray/And they/Bark bark
     bark bark/Bark bark BARK BARK/Until you
222Pekes Pols 8 to the fray/And they/Bark bark
 bark bark/Bark bark BARK BARK/Until you can
222Pekes Pols 8 fray/And they/Bark bark bark
 bark/Bark bark BARK BARK/Until you can hear
222Pekes Pols 8/And they/Bark bark bark bark/
 Bark bark BARK BARK/Until you can hear them all
222Pekes Pols 8 they/Bark bark bark bark/Bark
bark BARK BARK/Until you can hear them all over

And so it continues for a further 20 lines, barking all the way. Glanced at on the page it has a crazed and frantic look – an impression which, close-up, the intense pedantry serves happily to underscore. This bark item is pretty funny, I’d agree, but it is from Eliot’s Practical Cats book and was in the first place pitched for laughs. My own favourite, in terms of serendipitous eeriness, comes under ‘knock’. The entry begins: ‘Pressing lidless eyes and waiting for a knock upon the door/When Lil’s husband got’. And then proceeds:

I’d like to know about that coffin/KNOCK KNOCK KNOCK/KNOCK
know about that coffin/KNOCK KNOCK KNOCK/KNOCK KNOCK

careers on through a further 30 KNOCKS and ends:

KNOCK/KNOCK/KNOCK/KNOCK/Coriolan Triumphal March/Stone.
bull by the horns/And this is one/(Knock: and enter DOWNING)/Good.

Concordances needn’t be all fun-and-games. Indeed, the Larkin editor, R.J.C. Watt, makes a good case for the critical usefulness of his word-lists. He points, for instance, to ‘the extraordinary series of words beginning with the prefix un-’ that can be found in Larkin’s work, and certainly the series, ploughed through at one sitting, over seven pages, does have a deeply dampening relentlessness. At the same time, though, Watt is correct to emphasise how often Larkin used the un- prefix not simply in order to negate but rather, on occasions, to restrain or sorrowingly mitigate a positive: ‘the unraised hand’, ‘the apple unbitten’, ‘love unused, in unsaid words’, ‘unchilded and unwifed’.

Watt’s compilation also has the merit (not shared by the Eliot) of dating each of its word-entries – on the page, in an adjoining column, ‘making it possible’, he says, ‘to see at a glance whether Larkin favoured a particular word at a particular stage of his career’. Hence, there are Auden-words during his Auden phase, Yeats-words when he was influenced by Yeats. There is also a discernibly deepening distrust of language that might be thought of as too ‘poetic’ or too overtly fervent. Blood never crops up after 1950. Bright stops at 1959. Heart makes 26 appearances pre-1958 and is used again but twice (and one of these is in a line about ‘a boy puking his heart out in the Gents’). Nine of Larkin’s 14 God-mentions are in early poems. Post-‘Church Going’, we get ‘No God any more’, ‘Talking to God (who’s gone too)’, and the like.

All this is of real interest, and one can easily imagine Watt’s concordance serving as the basis of some future, and perhaps valuable analysis of Larkin’s changing way with words. For me, though, the toy-appeal of these two mammoth list-books seems likely to remain paramount. So: to go back to the questions I put to you at the beginning. The answers to the first two, since you press me, are grey (or gray) – 22 usages – and summer – 34. As to the others, even I can’t be bothered to check them out right now. As T.S. Eliot might have said, I’ll let you know some other time.

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