About the Literature Collections
Victorian Poetry and the English Poetry Full-Text Database: A case study
The Penguin Book of Victorian Verse was published in 1997. It was commissioned to replace an older anthology dating from 1969, edited by the poet George MacBeth; the new volume was to be significantly larger, both in terms of the quantity of poetry included and the length of the editorial introduction. MacBeth's selection – not the worse for being eccentric and arguable in places – gained its flavour from his own wide and admirably miscellaneous reading, but like most anthologists he recognised a core of poems which had to be included whether he liked them or not. I began compiling my anthology with the same division between, so to speak, impersonal and personal choices. The impersonal choices in Victorian poetry are not hard: a standard, representative anthology which left out Tennyson's 'The Charge of the Light Brigade', Robert Browning's 'Home-Thoughts, from Abroad' ('Oh, to be in England, / Now that April's there . . .'), or Emily Brontė's 'Remembrance' ('Cold in the earth – and the deep snow piled above thee . . .') would be thought not eccentric but perverse; in fact I allowed myself one such perverse decision, the omission of Gerard Manley Hopkins's The Wreck of the Deutschland, a poem which I cannot stomach, but which I now regret not forcing myself to include. As well as an obligation towards famous single poems, there is an obligation towards famous poets generally, and here too, though differences of proportion and emphasis are of course significant and form part of the critical interest of the anthology, the conditions in which selection is made are unproblematic. The works of all the first- and most of the second-rank Victorian poets are readily available and, it is charitable to assume, reasonably familiar to the anthologist. I started, therefore, with a list of poems I already knew, and poets whose books I owned; of the 416 poems which eventually made up the anthology, I calculate that roughly half were either chosen for me in advance, or consisted of choices from a prescribed group of poets.
The problem – and, at any rate for the editor, the interest – lay in the other half. I found myself reflecting on two linked facts. The first was the extreme limitation of the conventional 'field' of Victorian poetry, even among so-called specialists. Indeed, specialisation increased this limitation, because of the nature of academic teaching and learning. My general notions of the poetry produced from the 1830s to the 1890s were derived from my intensive reading of a small number of great poets, not the broad, miscellaneous reading which had nourished MacBeth's volume. The second, corresponding fact concerned the vastness of the actual 'field' – which was more like an ocean, mysterious and dark. An anthologist such as myself – professionally trained to specialise in a few writers, yet also required to make broad historical judgments as though these writers were always representative as well as exceptional – might well feel, on leaving the safe confines of Tennyson and Browning, or even the more doubtful haven of Swinburne, that he was venturing not just into the unknown but into the illimitable.
Readers of anthologies look for two kinds of pleasure: the pleasure of recognition, and the pleasure of discovery. I was reasonably confident of being able to supply the first, but much less confident about the second. I knew I needed to do a different kind of reading from that which had absorbed my professional career to date, but I also knew I did not have all the time I needed; teaching and other publishing commitments would see to that. I needed a short-cut, and the English Poetry Full-Text Database provided it. The database solved many of my difficulties, but it also, inevitably, had effects on the anthology itself. To my mind the nature and variety of these effects are the most interesting part of the whole business, but before I discuss them I should say some more about my own knowledge of the database and how I used it.
I had been on the editorial board of English Poetry, which was published on CD-ROM by Chadwyck-Healey in 1994, and I was in a good position to know both its strengths and weaknesses. In what follows I am going to take these weaknesses for granted, not because they are negligible but because they do not directly impinge on the use I made of the database itself. I knew, for example, that I would need to go outside its contents for an adequate coverage of women poets, that several significant male poets such as George Meredith or James Henry had been mistakenly omitted, that it was not possible to search the database by poetic genre or form other than rhymed and unrhymed verse (serious failings had I been compiling an anthology of sonnets), and that some categories of verse, such as hymns, had been omitted in toto from the nineteenth-century portion of the database on grounds of space. However, the second edition of the database, published in 2000, goes a long way towards remedying these weaknesses and would have saved me even more time. The primary bibliographical source of the English Poetry database was the New Cambridge Bibliography of English Literature (NCBEL), edited by George Watson and published in 1969; English Poetry, Second Edition includes the poetry of many writers who are designated primarily as 'novelists' or 'dramatists' by NCBEL and were therefore missed in the original sweep, such as George Eliot and George Meredith, and it adds many women poets left out entirely by NCBEL, such as Amy Levy and Augusta Webster.¹
The first, and in some ways the principal, strength of the database was accessibility. English Poetry brings together a collection of electronic texts whose physical originals could not all be found in any individual library, not even copyright libraries such as the British Library or the Bodleian. Even if they could, consulting them in quick succession would be arduous and time-consuming; having them all simultaneously available would be simply impossible. It seems a banal point to make, but the dematerialisation of texts, the removal of their physical bulk, has enormous implications for the scholar who is dealing with large amounts in short periods of time. The British Library has no limit, theoretically, to the number of volumes you may consult at a day's sitting; but between one sitting and the next you may reserve only six. English Poetry places hundreds of volumes on permanent reserve, without competition from other readers and without danger that the particular volume you want has been 'sent for conservation', an inconvenience which deprives you even of the luxury of complaining.
I had decided that the date parameters of my anthology would be those of Queen Victoria's reign (1837–1901) with a very few exceptions, such as Tennyson's 'The Lady of Shalott' (written 1832, but un-omittable). In terms of the database (I am speaking, remember, of the first edition, which was available to me only on CD-ROM) this meant mainly using discs 3 and 4, which covered the period 1800–1900. I sat down, therefore, one summer and the sabbatical term that followed, to read my way through the contents of these discs. (A caveat attaches to the phrase 'read through', which I will come to later.) Among the poets I encountered of whose existence, let alone works, I was totally ignorant were the following: Henry Alford, *Jane Barlow, Wathen Mark Wilks Call, *Digby Mackworth Dolben, *Anne Evans, Sir Samuel Ferguson, *David Gray, *Thomas Gordon Hake, John William Inchbold, Robert Dwyer Joyce, Charles Rann Kennedy, *William Larminie, Herman Charles Merivale, John Mason Neale, Henry Nutcombe Oxenham, Sir Joseph Noel Paton, *William Brighty Rands, *Menella Bute Smedley, *John Todhunter, and Frederick William Orde Ward. I have marked with an asterisk those whom I eventually included.
I could, of course, have compiled this list from NCBEL itself. It might or might not have been more error-prone – English Poetry, needless to say, is not free from error – but it would certainly have been more time-consuming, and the results would not have been as comprehensive. First I would have had to compile a list of poets and volumes to be consulted; then locate them; then lay hands on them. The database collapsed those stages into one. The list had already been compiled, and each volume could be summoned immediately and at will.
Although I had made a good deal of 'casual' use of the database since it first came out, I had not used it for a sustained research project and didn't quite know what to expect. I think the basic physical conditions of using any technology are often left out of account in thinking about its purpose and value, especially when those conditions involve continuous and intensive use. Like most literary scholars I am thoroughly habituated to spending an entire day with books – objects you pick up and put down, browse through, mark your place in, annotate (if they belong to you!); which have texture as well as text, a particular weight, feel, smell; whose physical characteristics (size and age, quality of paper and state of conservation) dictate the pace at which you can absorb them; above all, whose fundamental mode of use is a physical and manual act, the turning of each page. The different physical conditions of using any database for prolonged reading are obvious enough, but English Poetry also had some distinctive features. Not only are the texts dematerialised, they are standardised in look and layout. The database consists of transcriptions, not facsimiles; the texts were not scanned or photographically reproduced, but keyed in by copyists (like monks in a virtual electronic scriptorium); the result is a homogenous surface which elides every difference of look and feel in the volumes which make up the corpus. With the exception of orthography, there is nothing to distinguish a Renaissance folio printed on a hand-press from a Victorian octavo printed from stereotype plates; the huge diversity within the Victorian period itself is flattened out. All graphic features are missing (portraits, facsimiles of manuscript, plates), but the ruthless economy of the database eliminates 'extraneous' prose, too, such as prefaces and introductions, not to mention editorial or bibliographical details such as headlines, running titles, and publishers' advertisements: the social identity of the volume, its material and commercial specificity, are subordinated to its appearance simply as text on a screen.
I know that 'simply' is a misnomer, and that the apparently inert textual surface of the database is an illusion. The point I am making here, however, is to do with the basic, physical difference between reading a hundred, or two hundred, or five hundred volumes of poetry in electronic as opposed to material form. There is no doubt that the sameness of appearance of the texts made me more open, less prejudiced than I might otherwise have been; English Poetry doesn't allow you to judge a book by its cover. If a volume of verse had been produced with twee vignettes, I missed the vignettes; if it was a library copy, 'crumpled, dogseared and defaced – boys' way' (as Browning says), I didn't see this dilapidation, since items on the database remain fresh and pristine through years of use. Neither luxury nor poverty register as 'production values'. This democratic quality extends to the whole project of the database, in which markers of privilege are practically non-existent. The poets file past in alphabetical order, without notation of hierarchy, without biographical or critical tags, without distinction in the way they are treated. Since I hadn't the least idea who most of them were to start with, I came to their poetry without preconception except in terms offered by the poems themselves – their own declarations of intent as to genre, or verse-form, or subject, their proclamations of ideological allegiance, their more subtle signs of stylistic affinity. Sometimes after reading the poems I would look up the poet, to have my guess as to their age, social origins, career, and in several cases gender, confounded.
Above all, however, the experience of reading large amounts of text on screen, day after day, was radically different from reading the same amount in book form. (I should add at this point that the option of printing directly from the database and reading the poems in print-out would have been too time-consuming; saving the poems to disk also had the effect, in this early version, of stripping out all the character and paragraph formatting, so that too was ruled out.) To begin with you cannot browse a database in the same way as you can a volume. English Poetry allows you to view tables of contents for each volume, and I found this one of the most useful features for my purpose, but rapidly scanning the contents of a volume, flicking between different items, remembering a similar item in a different volume and looking that one up – these actions took longer in electronic form, were subject to the uniform pace and systematic logic of the software, and felt both artificial and ponderous. Second, the basic unit of the database is the screen, not the page. Again, English Poetry very usefully indicates page-breaks – this proved invaluable for the checking I did in the later stages of the anthology – but a 'page' in the database is notional, since the amount of text on screen at any one time may contain portions of two or more pages, and the action of scrolling automatically creates different combinations. Moreover, using the full-screen command makes a big difference to the number of lines displayed, which can now occupy the space taken up on the left of the screen by the table of contents or summary of matches. (Although it is possible to alter the font size of the texts, I never learned how to do this throughout the time I used the database for this particular project; software designers often don't acknowledge just how lazy, dim, or timid users can be.) The database breaks down the basic rhythmic and visual unit of reading, the page, and substitutes a mobile, provisional unit, whose orientation within the larger structure is vertical rather than horizontal. You move forwards and backwards in a book; on screen you move up and down. All these differences (there are many more of them, too minute and fleeting to record here) impose a different pattern to reading and study, and suggest different responses to the object of study itself.
The nature of the database therefore influenced how I made use of it, but the traffic was not all one-way; my own needs affected the way in which I approached this enormous, unprecedented resource. For example, there was the length of the poems I was prepared to include. Period anthologies are almost always biased towards shorter pieces, simply because you can represent more poets and more poems that way. Anthologists often compromise by including selections from longer works; sometimes this is unavoidable (Tennyson's In Memoriam, for example, or Robert Browning's The Ring and the Book, the longest single poem in the language) but I have always disliked the bittiness of this method and wanted my anthology to have space for some complete longer works, particularly sequence-poems. But these were almost all from the 'front half', so to speak – works such as Elizabeth Barrett Browning's Sonnets from the Portuguese, or Arthur Hugh Clough's Amours de Voyage – poems which I already knew and had determined to include from the start. I ended up with only a couple of longer works from less well-known poets – Augusta Webster's Mother and Daughter sonnet-sequence is the longest – and my policy in reading the database was to skip longer works unless there was some compelling reason to consider them.
The second example of the way in which my own needs influenced my use of the database concerns time – or the lack of it. I have said that I set out to 'read my way through the contents' of discs 3 and 4 of English Poetry, but this phrase carries a hefty dose of poetic licence. The way I actually worked was more summary, and resembled (at the extreme end of the scale) the scene in Robert Altman's Hollywood satire The Player in which aspiring screenwriters are given 25 words to pitch their idea to a studio executive. Victorian poems – preferably short ones – had about ten lines, or two or three stanzas, to catch my attention. Slow starters were penalised, in all likelihood; at any rate, if Sir Joseph Noel Paton's 'Ulysses in Ogygia' got much better after its opening lines –
Was it in very deed, or but in dream,
I, King Odysseus, girt with brazen spears,
Princes, and long-haired warriors of the Isles,
Sailed with the dawn from weeping Ithaca,
To battle round the god-built walls of Troy
For that fair, faithless Pest---so long ago?
– I never got to hear about it. The same applied to volumes of poetry, though here I tried hard to sample poems from different sections of a volume, so that I gained some idea of the whole and was able to spot obvious differences in quality.
My experience of reading, for the first time, Victorian poetry in the mass, rather than the poetry selected by posterity and which formed my inheritance as a literary scholar, was therefore conditioned by the physical and, so to speak, metaphysical properties of database and computer technology as it existed nearly a decade ago. It was, even so, an extraordinary experience, and one which permanently changed my understanding of literary history. This change, however, did not take place all at once and the end was not of a piece with the beginning. My first reaction was to doubt the validity of any generalisation about literary periods or genres based on what seemed like absurdly small and unrepresentative samples. My confidence in being able to say anything sensible about 'Victorian poetry' vanished, since it had been based on my reading of the few survivors from the sea of verse still clinging to the rock of the university syllabus. Arguably those survivors had endured not because they were representative Victorians but because they were not. English Poetry made me aware of a split between our notion of the 'average' or 'characteristic' product of any period, and the image of that period which is conveyed to us by its masterpieces; or perhaps it would be truer to say that though I was aware of this distinction I had not realised its actual force. 'Axioms in philosophy are not axioms until they are proved upon our pulses,' said Keats; English Poetry allowed me to take the pulse of Victorian poetry for the first time.
I did not stop there, however. My absolute scepticism evolved into something more nuanced and complex. The database told me that 'Victorian poetry' was not the distillation of a few great names. But on reflection I saw that this fact was not single, but composite. We make generalisations for different purposes, and these purposes affect the kinds of generalisation which we find useful. Although I found many interesting and beautiful poems on the database, I did not discover another great poet. If there is an undiscovered Victorian poetic genius, he or she is not included in English Poetry (I didn't find one anywhere else, either). There is room for differences of emphasis and for shifts of taste, but posterity is not going to replace Alfred Tennyson with Charles Tennyson Turner, no matter how excellent some of Turner's sonnets are; Christina Rossetti may have written a poem beginning 'Give me the lowest room', but she is not going to surrender her high status to Frances Ridley Havergal. English Poetry confirms our collective judgment about what constitutes 'greatness' in Victorian poetry even as it challenges the narrowness of that judgment in other contexts. The study of the best poetry is not all that the study of poetry can offer us; we cannot afford to discard our notion of 'the best' in favour of cultural relativism or fashion, but we can afford to enlarge our sense of literary history and, for that matter, literary pleasure.
¹ NCBEL itself has since been updated, and would now be a more reliable guide for the compilation of a database such as English Poetry. In addition, English Poetry is now available in the more flexible web-based format of Literature Online as well as in CD-ROM; further additions and enhancements are therefore more feasible, such as searching by poetic genre, which, I am told, is planned for Literature Online, Third Edition.