The origin of the idea that literature aspires to canonical immortality has proved difficult to pin down, but it clearly began with religious writings. Anything that supported your chosen faith was immediately canonical - anything that didn’t was heretical or blasphemous. So the canon was, at first, a collected body of literature that adhered to the tenets of your religion. Which makes the Bible the best known canon there is, because it’s a compilation of the work of many different writers. Moreover, its meaning is immortal by default, being the Word of God that carries its own celestial copyright. Writings that different religions have expelled from versions of their holy texts are placed in various ‘Apocrypha’ (literally “hidden writings”), or stuff not considered to be authentically ‘inspired’ enough to pass the meaning test. The Koran has them (the “Satanic Verses”), and so does the Book of Mormon. It’s like editing a text; some bits are genuine, others aren’t, and the latter are discarded.

It’s not a great leap from this to the secular definition of canonical literature - it’s those works that most successfully embody your aesthetic rather than your religious creed. (1) I dare say any medieval or Renaissance scholar worth his salt would have jotted down a ‘must read’ list of non-religious reading from his tutor and kept it in his doublet pocket. Ferdinand, the King of Navarre in Shakespeare’s ‘Love’s Labour’s Lost’ and his coterie of young wits certainly would have. They aspire to be “heirs of all eternity” by setting up their own “little academe” to study the finest texts literature and philosophy can afford them. Sir Philip Sidney had a similar ambition; in his ‘Defense of Poesie’ we looked at earlier, the following writers are quoted in support of his arguments: Pliny, Musaeus, Homer, Hesiod, Orpheus, Linus, Amphion, Livius Andronicus, Ennius, Dante, Boccaccio, Petrarch, Gower, Chaucer, Thales, Empedocles, Parmenides, Pythagoras, Phocilydes, Herodotus, Virgil, Xenophon, Tremellius, Junius, Tyrtaeus, Lucretius, Manilius, Pontanus, Lucan, Cicero, Heliodorus, Plato, Aristotle, Cornelius Agrippa, Horace, Terence, More, Erasmus, Plautus, Euripides, Phocion, Sannazaro, Boethius, Persius, Plutarch, Pindar, Tasso, Ovid, Dio Cassius, Ariosto, Scaliger, Bembo, Bibbiena, Beze, Melancthon, Fracastorio, Muret, Buchanan, Hurault, Juvenal, Surrey, Spenser, Sackville, Norton, Apuleius, Demosthenes, Landino, and both Old and New Testaments. Someone must have given him a hand with that list of worthies.

Notwithstanding these early examples,  the sense of the word ‘canon’ used in the non-religious sense of ‘a body of literature’ doesn’t gain general currency in English until the mid-18th century, and even then, all you had to do was read widely and have firm opinions to compile your selection. There were no objective criteria involved. Just an intellectual cocktail of precedent, peer pressure and dogmatism - and it’s these three elements which have characterized the judgments of the self-appointed gatekeepers of the canon ever since. Take Joseph Addison’s ‘An Account of the Greatest English Poets’ as an example, in which he notes that the canon changes as tastes do. No-one likes Spenser any more, he reckons, because he’s so old-fashioned;

Old Spenser next, warm'd with poetic rage,

In ancient tales amus'd a barb'rous age;

An age that yet uncultivate and rude,

Where'er the poet's fancy led, pursu'd

Through pathless fields, and unfrequented floods,

To dens of dragons and enchanted woods.

But now the mystic tale, that pleas'd of yore,

Can charm an understanding age no more;

The long-spun allegories fulsome grow.

While the dull moral lies too plain below.

We view well-pleas'd at distance all the sights

Of arms and palfreys, battles, fields, and fights,

And damsels in distress, and courteous knights;

But when we look too near, the shades decay,

And all the pleasing landscape fades away. (2)

Away with tradition, says Addison, we’ve made progress since then. So Spenser was out of his canon, then.

From the mid 1790’s onwards, anthologies of (mainly) poetry came thick and fast. And these were not just Miscellanies which were basically collections of verse: these were sold as selections of good stuff that you needed to read to be thought ‘up’ in English literature. And it’s from these times that English Literature (in capitals) begins to coalesce as a subject not for university study (that, as we’ve seen, came nearly a hundred years later), but as an ‘improving’ pastime. Joseph Ritson’s ‘The English Anthology’ was published in three volumes in 1793-4; hot on its heels was Samuel Pratt’s ‘The Cabinet of Poetry’ at double the length. Among the better known examples is William Hazlitt’s ‘Select Poets of Great Britain’ published in 1825, a feat of prodigious reading which includes detailed selections from the works of 41 writers from Chaucer to Burns. And, of course, there’s been many hundreds since, separating the wheat from the chaff so that we grateful Recreational Readers don’t have to ruin our eyesight compiling our own ‘Now That’s What I Call Literature’.

Then came those critics who chose to write about the canon, who performed a similarly thorough editorial role; so when TS Eliot, in his essay ‘The Perfect Critic’ rhetorically demands that we “Compare a medieval theologian or mystic, [or] a seventeenth-century preacher, with any liberal sermon since Schleiermacher” he’s not only relieving us of the responsibility, he’s letting us know that, scarily, he actually has. And if you look at Harold Bloom’s incredible list of canonical literature published in 1994, you wonder how the poor bugger had time to eat or sleep.

The members of the literary police force who, like Eliot, were taking the job seriously were keen that  the canon should not only foster a reverence for past achievements, but also demonstrate a range of “quality” writing (which was usually identical to the stuff they liked). Some had more pretensions to impartiality than others; Johnson, in his ‘Lives of the Poets’ tried to cloak his partiality by appealing to a notion of taste shared by all right-thinking people (of whom he, of course, was the leader); Eliot thought that meaning worth its salt didn’t need any personal endorsement to help it survive, and tried, wherever possible, to write himself out of the selection procedure. The canon, in fact, selected itself on intrinsic merit.

I think Eliot’s arguments still form the best impartial introduction to an understanding of how a canon works, so I reckon a quick precis wouldn’t be a bad idea.

The Canon, says Tom,  shouldn’t be conservative - it should never blindly follow tradition. And tradition, anyway, is a far broader church than that represented by a blind obedience to past successes. If you’re going to try and write for the Canon, you need to educate yourself both in “the historical sense” (a conception of “the pastness of the past”), but also be aware of the presentness of the past. And in order to get that sense, you have to read one hell of a lot.

That being the case, “no . . . artist of any sort, has his complete meaning alone.” His meaning is built on both the vital and decomposing meaning of every previous artist. Then he goes all Emersonian/ Platonic on us. Every new work of worthwhile art adds meaning to the sum of that which preceded it, altering the nature of all meaning “ever so slightly”. But the sort of literature that exerts the greatest influence on this sum of meaning is that which  demonstrates the greatest “intensity of the artistic process”. The component parts of meaning don’t matter that much when set against the power of the artist’s vision that fuses them all together into meaning. And the more diffuse and startling those components are, the greater the force of that imaginative power that shapes them needs to be. So it’s not necessarily the period trappings of a work of art which will date it - it’s rather a failure of skill. So only second-rate, conservative art will date. Neat, isn’t it?

So at least Eliot thought the matter through for himself, although he declined to produce a definite canon; as we’ve seen, he didn’t think there was any point because it would be too personal and, therefore, inconsequential. It’s a pretty safe bet that Dante would be #1 though.

But most compilers of canons and anthologies not blessed with Eliot’s intellectual conscience and rigour usually hug a well-tried formula: a mixture of the usual suspects with a few idiosyncratic, cult and/or contemporary selections thrown in to either ferment debate or to show how magnanimous or hip they are. And so the canon creeps inexorably forward over time, shedding its dead wood and enshrining new reputations as it proceeds.

But in practice it’s all so arbitrary, particularly if you’re remotely inclined to favour any form of rational or scientific analysis in the way you assess excellence; meaning, by and large, still lives or dies courtesy of its perceived merit and not its intrinsic merit, no matter what Eliot says - even though the unwritten criteria governing that merit are impossible to objectify. But those texts that make it into the canon are necessarily ‘representative’ of those shadowy criteria. 

Perhaps it would be better just to look at popularity; there’s nothing (except as we’ve noted, snobbery) that says meaning can’t at least in part be the product of concensus. And it’s what’s tending to happen anyway.

In recent times, the canon has become a publisher’s dream - any marketing director who wants to stoke up interest in the back catalogue canvasses readers every few years for their personal Top Five books and a chart is compiled. Which of course shifts the idea of meaning further away from quality in the direction of opinion. It’s all much more democratic. And this change of emphasis away from the traditional guardians of the canon and their Masonic criteria of excellence that even they don’t fully understand towards the Recreational Reader and his pursuit of pleasure is no bad thing. “Favourite” and “Best” can cover a huge range of potential responses which are just about as empirically sustainable as anything academe has thus far offered, after all.

But what’s surprising is how closely the respective canons of the Great & the Good and the Recreational Reader often resemble one another.

One of the most recent polls of favourite novels (by the sponsors of  the Orange Prize for Women’s Literature) found it’s still ‘Jane Eyre’, ‘Wuthering Heights’ and ‘Rebecca’ that push the distaff meaning buttons. Of the Top 10 authoresses, only two were living - Carol Shields and Harper Lee. (3)

The BBC’s ‘Big Read’ of 2003 (once again, novels only) was less conservative, but nevertheless boasted Tolstoy (2 entries), Dickens (5), Dumas, Dostoyevsky, Hardy (2), Charlotte Bronte, Louisa May Alcott, George Orwell (2), Jane Austen (2), Robert Louis Stevenson, James Joyce, Wilkie Collins, and Emily Bronte on its list, rubbing shoulders with JK Rowling (4) and Terry Pratchett (5). Any of these authors can still be found on reading lists and syllabuses in our schools and colleges.

So maybe all that canonical brainwashing worked after all. Or maybe they’re just damn good books whose meaning has endured. It could even prove that academics’ tastes aren’t governed by such different criteria as they’d like to think. Or, even better, than the public’s better read than they imagined.

Talking of the academic community, democracy within the canon in that neck of the woods is being interpreted in rather different ways. In fact, it’s currently hugely unfashionable to talk in terms of ‘canons’ since, to many of the more socially or culturally-influenced critics, the very idea of a hierarchy of meaning smacks of  good old-fashioned colonialism. And there’s no denying that some groups of writers have been neglected (or as we say these days, ‘disempowered’) in what’s come to be known as the ‘Western Literary Canon’. The American critic George P. Landow is one of the more dispassionate commentators on the subject:

Like the colonial power, like, say, France, Germany, or England, the canonical work acts as a center -  the center -  of the perceptual field, the center of values, the center of interest, the center, in short, of a web of meaningful interrelations. The noncanonical works act as colonies or as countries that are unknown and out of sight and mind. That is why many object to the omission or excision of female works from the canon, for by not appearing within the canon works by women do not appear anywhere.

Landow’s using women’s literature as his example, but he could, of course, equally have used Black or Gay in its place. And he’s right. There is a strong case to be made that the WLC is overly male and bourgeois. But was this the result of a deliberate excision of other sources of writing or their omission through ignorance? Conspiracy or cock-up? The argument goes as follows: the canon of "great literature" has thus far ensured that it is only what is called “representative experience" (one selected by male bourgeois critics) that is transmitted to future generations, rather than those deviant, unrepresentative experiences to be found in much female, ethnic, and working-class writing. So meaning is a white middle-aged, middle class bloke. Which still adequately describes many of the senior teaching staff in academia, but isn’t an adequate reflection of the wide variety of literature that’s currently being written, or, indeed, who’s reading it.

So what can you do about this? Landow reckons there’s three options - (1) make the canon bigger, (2) change the entry criteria or (3) start an ‘alternative’ canon. Well, for what it’s worth, I’d jettison (3) for two reasons: you’ll ghettoize the very writers you’re intending to help, and, as typically happens in anything set up as a reaction to something else, you’ll replace one hierarchy with another that’s usually as tyrannical as its predecessor. With (2), it’s easy to level charges of special pleading or positive discrimination that hinder the reputation of those very works you’re trying to help. So we’re left with (1), and then someone comes along and accuses you of diluting the quality of the existing canon. Lose/lose.

But what the issue ultimately reduces itself to is whether you want the canon to draw on a wider list of influences for the sake of variety and inclusiveness, or whether you think these minority genres aren’t represented because they simply can’t cut the mustard in the company of Shakespeare, Milton et al.

If your sympathies are with the former, you won’t mind that my nephew’s studying the poetry of Emily Bronte (which is basically badly-written juvenilia) for his AS Level and not her classic ‘Wuthering Heights’; or that his other set text is ‘Top Girls’, a badly dated critique of the Thatcher era by Caryl Churchill; if you’re in the latter group, this attempt to redress the paternalistic/ hierarchical composition of the canon using students as its lab rats will make your blood boil.

I could go on, but I won’t, for obvious reasons. Suffice to say the debate is very much alive and kicking.

Footnotes:

1. And let’s not forget that the works of Shakespeare form the best-known secular canon there is. Never collected and published in his lifetime, attribution of individual plays and poems keeps a whole industry in a job. 

2. In his ‘Preface to Addison’, Johnson teases the poet and essayist about this particular passage: “In this poem is a very confident and discriminate character of Spenser, whose work he had then never read. So little sometimes is criticism the effect of judgement.” Johnson, of course, never fell into this trap.

3. Since that poll was conducted, Carol Shields has, of course, died.