Elizabeth Scott-Baumann & Ben Burton
Oxford University
Encoding Form: A proposed database of poetic form
1> In 1979, Heather Dubrow called for a new methodology in early modern studies. ‘We are still prone to separate literary history and generic analysis,’ she observed, ‘[b]ut the two modes of criticism can illuminate each other
“exploring why certain genres flourished when they did and how they shaped and were shaped by the temper of their age is one of the surest ways of tracing the complicated movements that inform literary history.”[1]
Thirty years later, it seems that literary scholars have been slow to respond to Dubrow’s appeal for a new kind of historical formalism, especially when developing electronic resources.[2] To be sure, the last four decades have seen exciting developments in digital literary studies, including e-texts and online corpora; multimedia electronic editions; full-text resources for biography, bibliography, and lexicography; as well as computer-assisted textual analysis. In the field of early modern studies, online resources such as Early English Books Online (EEBO), Chadwyck Healey’s Literature Online (LION), and Gale Cengage’s Eighteenth Century Collections Online (EECO) have enabled scholars and students to conduct literary historical research in novel ways, allowing users to discover obscure publications and pursue new lines of enquiry through keyword-searchable full texts.[3] Such resources as these have transformed the way we study words and phrases, but they are not designed to support sophisticated analysis of the formal properties of early modern texts. As scholars begin to reassess the relationship between formal and historical study, we need electronic resources that will enable scholars and students to study form in a historically-rooted way, stimulating new questions about parallels and differences across literary periods.
2> What might such a resource look like? This paper explores the motivations and challenges of creating a database of form for early modern English poetry. We are currently planning a database that will contain information on the genres, metrical structures and rhyme schemes of all poetry in early modern printed texts. Users will be able to search a variety of criteria, exploring specific poetic forms by name, or metre, or rhyme scheme and then refine search results within a time period, or according to author, or place of publication, or language. They would also be able to enter information about the formal properties of a poem, such as its rhyme scheme, and be able to view similar or identical forms in the database. The database would change the study of poetic form in the way that LION, EEBO and other keyword-searchable full text corpora have revolutionized our ability to study imitation, influence and tradition at the thematic and content level of words. In this paper we will outline the kinds of data this resource could provide, and the kinds of questions we would want to ask of it.
Purpose and Objectives
3> This database will contribute to an ongoing effort to combine literary history with the study of prosody and genre. In doing so, we hope to reassess the scope and purpose of prosodic analysis, and its relation to electronic resources, in shaping our understanding of the early modern period. The database would in no way replace scholarly analysis, but it can assist us in certain aspects of prosodic study, and will open up new areas for teaching and research.
4> The primary objective of the database is to provide an unprecedented quantity of information about poetic form – such as syllable count, rhyme scheme and rhyme words – which form the basis of prosodic and generic analysis. It does not aim to offer definitive or exhaustive analyses, but to stimulate further enquiry and to provide data for independent analysis. The database assists prosodic study by dramatically reducing the time taken to perform basic metrical analysis such as counting syllables and identifying rhyme schemes. More excitingly, it opens up areas of enquiry by enabling scholars to ask new questions about literary texts. At present, prosodic analysis is usually confined to isolated studies of an individual author or small groups of authors. One of the ultimate objectives of the database is to broaden the scope of prosodic analysis by providing the raw materials for a cultural history of poetic form and generic change across the early modern period. In this sense, the database will enable scholars to ask the same kinds of questions about form that our current resources allow us to ask about words, for example: What is the origin (or origins) of a given form? How does its structure, use and meaning change over time? Are there variations in use and meaning in different regions, or among different groups? How does a given form relate to others, and how does this relationship change over time? To pursue the analogy with lexical analysis, the database invites us to explore what might be called the etymology and morphology of poetic form, as well as the complex history of prosodic dialects, sociolects and idiolects.
Methodology
5> The database would encode additional information about the generic and formal characteristics of texts in our existing online corpora of early modern texts. At present, resources such as EEBO are designed as digital libraries and, as such, are not concerned with adding value by annotating the texts they compile.[4] Our strategy is to create an interface that would interrogate marked-up texts, as users present queries, so that the user’s search would call up certain poems and records of these poems’ formal features. Users would be able to see a short, searchable record of formal features for each text but also to search much more widely and narrowly according to their own research interests. We propose to annotate existing texts created by the Early English Books Online Text Creation Partnership (EEBO-TCP), enriching these texts by providing detailed information about their prosodic structure. We would use a Text Encoding Initiative (TEI) P5-compliant scheme to enrich the relevant encoded EEBO-TCP texts.[5] The TEI already provides a base tag set for encoding verse, including stanza divisions, caesurae, enjambment, rhyme scheme and metrical information, as well as a special purpose rhyme element to support the simple analysis of rhyming words. For rhyme and metrical analysis, the module for verse includes attributes that allow users to specify encoding for the conventional metrical structure of the element, as well as for the actual realization of the conventional metrical structure applicable to a given element of a text.[6]
6> What kind of information might be marked-up in this way? As David Chisholm and David Robey point out, ‘[t]here is...a general question as to whether text markup for scholarly interchange should be limited to broad formal conventions (e.g. meter, rhyme, stanzaic forms), or whether individual characteristics of verse texts may also be included (a step that often represents the foundation of a stylistic interpretation).’[7] What is more, since prosodic analysis necessarily entails an element of critical judgement, any encoder will need to formulate clear principles for marking up verse. For example, can we always distinguish between formal conventions and individual characteristics of verse? How should we indicate that the pattern indicated by our mark up is not the only possible pattern for the verse text under consideration?
7> We believe that some of these questions are best approached through practice. In our pilot project outlined below, we will test both methods of encoding on a sample of texts spanning a range of genres from the 1590s, including sonnet sequences, epic poetry and drama. At present, we can only speculate as to which approach will be most appropriate for a project of this scope. We suspect, however, that a simple mark-up of rhyme scheme and conventional metrical structure will provide more information about texts that involve a variety of stanzaic patterns – such as sonnet sequences, or collections of lyric poetry – than texts that use a single stanza, such as epic poems, or those which rely heavily on a single conventional metrical structure, like drama. If this is true, it seems likely that the project will need to provide information about formal conventions as well as a limited range of individual characteristics. Of course, this would be labour intensive. Manually marking up the actual metrical structure of poetic texts at line level could prove extremely cumbersome and time consuming. One possibility is that the database might develop in phases, beginning with rhyme scheme and conventional metrical structure, and then progressing to include further information about the actual features of these texts. This approach allows the flexibility to encode additional layers of complexity. These could include highlighting irregularities in rhyme scheme, or indicating any deviation in the real metrical structure of a line from the pattern defined by the conventional metrical structure. In the Pilot Project section (below) we discuss the possibilities of this kind of mark up, as well as its limitations in relation to our pilot project.
8> This pragmatic approach can only take us so far, however, as we attempt to address larger theoretical questions about how we mark-up verse texts. In a future article we will examine the theoretical assumptions that lie behind the TEI guidelines for encoding verse, and will situate our database in relation to recent historical studies of early modern prosody.[8] At this early stage, we can only highlight some of the large problems that remain to be answered by our project. For example, is the purpose of the database to examine poetry in the terms that early modern poets may have used, or to provide an analytical framework that draws on current prosodic theory? The former approach would allow users to see the period in something like its own terms, but it is open to criticism from theoretical prosodists who distinguish between ‘the conscious metrical principles to which poets may have adhered’ and ‘the principles which actually governed their practice.’[9] This is an important distinction, but it leaves us wondering how these ‘principles’ are to be defined in light of the ongoing methodological conflicts in verse studies. As Richard Cureton observes, ‘the failure of theoretical prosodists in [the twentieth] century to develop a generally accepted method of rhythmical analysis has made it impossible for literary historians and historically minded prosodists to take advantage of these advances in any coherent way.’[10]
9> These methodological questions are complicated further by practical and pedagogical concerns: should we adopt a traditional accentual-syllabic method of scansion that is already used in many textbooks and prosodic manuals, or should we eschew this arguably outdated system in favour of a more credible but less familiar method of analysis? A possible solution to this problem, adopted by Thomas Carper and Derek Attridge, is to combine modern methods of scansion and traditional terminology (such as ‘iambic pentameter’ and ‘trochaic trimeter’), with the proviso that these terms provide a ‘useful shorthand for describing particular sorts of meters’ rather than ‘keys to what’s really going on in metrical lines.’[11] This solution may be pragmatic, but it might also be seen as theoretically suspect. As Peter Groves argues, ‘the traditional description [of verse texts] lacks intellectual coherence and rigour, and should not be perpetuated merely by inertia, incumbency, habit or apathy.’[12] While we admire Groves’s refusal to compromise on this point, we hope to establish principles for encoding verse texts which address the concerns of theoretical prosodists as well as scholars, teachers, students and non-specialists who are already familiar with traditional methods of analysing poetic form.
Dissemination and Publication
10> Online databases and editions are now a central part of UK and US universities’ research resources. Oxford University, where we work as lecturers and researchers, currently subscribes to resources with a variety of search functions and searchability, such as The Bible in English (searchable by key word, or chapter and verse), the Poetry Archive and Past Masters (largely philosophical texts). Reference works are also increasingly used in their online forms. The Oxford English Dictionary and the Oxford Dictionary of National Biography are now largely used online in frequently updated and carefully managed versions, and scholarly journals increasingly make their articles accessible through JStor.org and Project MUSE.
11> Many databases providing primary texts, such as LION, EEBO and ECCO, are now commonly subscribed to by universities and widely used by undergraduates, postgraduates and senior scholars. These existing databases offer a sweep of literature from medieval to twentieth-century, and some provide both digital facsimiles and full-text versions of rare (and common) books. Digital facsimiles have enabled thousands of students and scholars to study rare, fragile books as they were originally published, including manuscript drafts (see Oxford’s First World War Poetry Digital Archive) and manuscript texts (the JISC-funded Digitizing Correspondence project under way at London’s Centre for Editing Lives and Letters; the Perdita Project providing facsimiles of early modern women’s writings). Full-text versions of swathes of poetry have been used for stylometric analysis to help ascertain authorship or date a work in a writer’s life or used to trace allusion, echoes and imitations of other authors (just as the availability of scholarship online has enabled many universities to use plagiarism software such as Turnitin to screen their undergraduates’ work).
12> The Text Creation Partnership provides texts to Chadwyck-Healey, who mount them online through their (otherwise facsimile-and-bibliographic-record only) EEBO database. EEBO therefore provides access to both facsimiles and an increasing number of full-text transcriptions of early printed books. We are keen to work with EEBO-TCP texts for the early modern phase of the database, using text encoding techniques to create enriched texts from which the searchable interface will glean and exhibit data on formal aspects of all poetry. We aim to negotiate with EEBO and Chadwyck-Healey to find a form of publication that both respects intellectual property and commercial interests, while also making this rich new material accessible to the widest possible audience.
13> At the current stage of planning we are considering several options for the dissemination of our research findings and the publication of the database. These include 1. An open access interface available to any internet user that provides both a searchable interface and access to marked up texts; 2. Access to searchable interface and marked-up texts to EEBO-TCP subscribing institutions, perhaps as one of the partners’ custom interfaces; 3. Open access to database and texts but not with mark up. This would mean that if we are not able to make the XML-encoded texts freely available, we would display the texts in their entirety (as users request them), but with the encoding invisible. The interface would interpret our encoding and display the verse with, for example, its stresses marked with accents, or its rhyme scheme colour-coded, rather than with visible tags; 4. Open access to database but no texts. We are most keen on the first option, but this will depend on how our negotiations with EEBO-TCP and Chadwyck-Healey progress. Option 3 would have obvious benefits over Option 4 in that institutions who do not subscribe to EEBO such as schools, where we are keen to see the database used in teaching, would have access to the texts as well as our data, just without the mark up visible.
14> This project would result in a sustainable resource that could be developed beyond the sixteenth and seventeenth centuries across every period, and could be connected to huge existing databases of poetry already subscribed to by universities, and ideally also be available to school students across the UK and USA.
Applications: Research and Teaching
15> We are currently in an exciting period of literary criticism, as scholars review the formal, political, theoretical and historical approaches of the previous century and call for perspectives that unite formalist literary and historical analysis.[13] Their work questions the critical tendency to conflate certain poetic genres such as lyric with outmoded critical methodologies of the early twentieth century, and encourages us to rethink the longstanding association between generic analysis and a rarefied concept of the aesthetic. Eschewing the New Critical ideal of formal unity or coherence, these scholars propose a freshly theorized ‘historical formalism,’ reassessing poetic form as historically specific, historically determined, and historically efficacious.
16> There has already been some excellent work in this field, concerned with the rise and fall of genres, changes in prosodic tastes and practices, how writers distinguished and differentiated the specific functions of literary texts, as well as a history of aesthetics, all of which have been sidelined by a cultural studies movement which has dominated literary scholarship in recent years.[14] At present, however, the scope of this research is restricted by the lack of electronic resources that would make prosodic information of all sorts more easily accessed, more quickly searched, and more widely shared. Indeed, we would imagine that many scholars interested in historical formalism are currently wrestling with the same kind of problems that frustrated a previous generation of scholars working on vocabulary and verbal allusion. As a result, scholars interested in formal and historical analysis have understandably tended to focus their energies on analysing a limited sample of texts or a small group of authors. Larger studies, spanning a broad time period and a wide range of authors, are often richly suggestive and point the way towards future research, but scholars need new resources to provide the kinds of evidence that could support these ambitious projects.
17> In creating an electronic database of poetic form for the early modern period, we would seek to provide a crucial tool in this emergent field, enabling scholars to explore the unpredictable, multiple interactions between form and history, while stimulating new questions about the cultural significance of specific forms as well as how particular genres changed and developed over time.
18> The multiple research implications of this database would change the way in which scholarship on poetic form is conducted. Scholars will be able to make new connections between authors of different periods, styles and cultural identities. For example, they would be able to trace the progress of a particular form, looking at how the heroic couplet was used before its most famous practitioner, Alexander Pope, and after him. It would also be possible to discern how common metrical irregularity was within certain forms, and whether this was dictated by subject matter, education or gender. Scholars would also be able to identify heterogeneity within certain forms, for instance, exploring how much formal variation existed within the sonnet boom of the 1590s, which variants endured into the seventeenth and eighteenth centuries, and which were abandoned but received a later revival.
19> Finding a poem difficult to categorise, it would be possible to use the database to investigate similar forms, exploring possible models and traditions, or revealing valuable innovations in form. Rhyme words would also be detailed in the database, providing rich new evidence of how rhyme could function: Were certain rhyme words always used for comic effect? Do particular rhymes appear more frequently in certain periods than others, and what might this reveal about the cultural significance of rhyme at these moments? It was a commonplace of the New Criticism of the earlier twentieth century that, in Pope’s words, ‘The Sound must seem an Eccho to the Sense’[15]; our resource would aim to provide the largest amount of data with which to explore the myriad aesthetic and political ramifications of this principle.
20> The Electronic Database of Poetic Form would also be an immensely useful teaching tool, at both school and university level. Recent research has confirmed that online resources such as EEBO are already transforming undergraduate teaching and learning.[16] Electronic resources are used by undergraduates more than any other group and our database would draw on enthusiasm for online material to challenge and develop their reading of poetry. Undergraduate students might be able to submit entries of rhyme schemes and syllable counts on a connected pedagogical website, developing their own skills at identifying certain metrical forms and exploring the implications of rhyme words and different kinds of rhyme.
21> In the same way, the database could be used to teach secondary school students about rhyme and metre, providing an accessible, hands-on introduction to key concepts of prosodic analysis, and enabling them to explore the variety of poetic forms used by a particular author. Students might be able to highlight and annotate specific features of their chosen texts, following online exercises organized by a teacher, and submitting work for further discussion in an electronic forum with class members. Here we find a particularly strong warrant for making our resource freely available to any internet user, especially to institutions which cannot afford the subscription fees which accompany many electronic resources in the humanities. Although we are currently considering several options for the publication of our database, we have been encouraged by the recent development of open access resources, such as the new Shakespeare Quartos Archive (http://www.quartos.org/), that are freely available for sophisticated, close analysis of literary texts for students and scholars across the globe.[17] The success of this resource convinces us that online resources can satisfy the demands of the scholarly community as well as the wider public. In our pilot project, described below, we will experiment with an interface that will be accessible to non-specialists while also providing complex search features for more advanced users.
Pilot Project: 1590s
22> The pilot project comprises two stages: 1. A small database containing information on the metrical structures and rhyme schemes of all verse in the first edition of 10 texts published between 1590 and 1599. 2. A larger database containing information on the metrical structures and rhyme schemes of all verse in first editions of texts published during this period.
23> The first stage of the pilot will illustrate the variety of texts contained in this database, ranging from those written entirely or predominantly in verse (such as epic poems, verse drama and sonnet collections), to those which include poetry and prose in varying proportions. The two principal investigators will work on the initial project, and the results of this work will help determine the search functions of the database, as well as the timescale and the resources needed for the second stage of the pilot. The second stage will demonstrate the complete search functions of the database across a sample period. We aim to provide information about poetry in first editions published between 1590 and 1599, ranging from works written entirely in verse, to brief quotations and epigrams that feature in larger works.
24> The 1590s was a unique and exciting decade in the development of poetry in English, with poets experimenting with classical metres, forging and theorising a vernacular literary tradition, reflecting continental influences, as well as the inflections of music and rhetoric. It saw the publication of Sidney’s Defence of Poesie, one of the most influential formulations of the purpose and responsibilities of poetry, as well as his sonnet sequence Astrophel and Stella. The decade saw the publication of Spenser’s major English epic, The Faerie Queene, Marlowe’s Ovidian Hero and Leander and Shakespeare’s two great narrative poems, Venus and Adonis and Lucrece. These latter two have been characterised as the most patterned, structured and even the most ‘literary’ of all Shakespeare’s works.[18] Heather Dubrow has talked of modern readers’ ‘distrust of elaborate devices’ and though these poems are widely studied, much remains to be explored in their formal patterns and innovations.[19] These poems enjoyed a much greater fame in Shakespeare’s lifetime than they do now, and it was these works that motivated contemporaries to write about the honeyed softness of Shakespeare’s verse. Our database would allow further exploration of what exactly such critical terms meant in the sixteenth and seventeenth centuries. Were these works defined by their formal regularity, by enjambment creating an admired sense of flow, or can we see ways in which irregularity and shifts in metre might also be admired? These terms of smoothness and softness became connected with the couplet in the eighteenth century and it would be fascinating to be able ultimately to compare the couplets of the sixteenth century with those of the eighteenth. Did poets’ use of the couplet become more uniform as it gained dominance?[20] Shakespeare’s narrative poems of course do not employ couplets, but for Venus and Adonis iambic pentameter sestets (rhyming ababcc) and for Lucrece rhyme royal (7 pentameter lines rhyming ababbcc). The Electronic Database of Poetic Form would allow users to input these schemes (either metre, rhyme scheme or both) and immediately see what other works were published in the period using the same stanzas, whether these became more or less common after Shakespeare’s poems were published, whether these forms were used for comic, tragic or other material, amongst numerous other research questions.
25> The 1590s also saw the rise of some of the greatest works in English verse drama, which this database will include, such as Marlowe’s Tamburlaine, Dido Queen of Carthage and Edward II, and Shakespeare’s Romeo and Juliet, Titus Andronicus, Richard II and Love’s Labour’s Lost. It has often been said that late sixteenth-century drama and poetry influenced each other, and this database would provide illuminating evidence of the exact nature of this influence. Certainly Love’s Labour’s Lost includes poetry which alludes to song forms such as madrigal verse, as well as strophic forms, and is the most heavily rhymed of Shakespeare’s plays. The rhymes are often semantically significant, such as the repeated pairings of ‘lord’ and ‘word,’ ‘eyes’ and ‘lies’. These rhymes were indexed in Helge Kökeritz’s Shakespeare’s Pronunciation of 1953, which used linguistic features such as rhyme words and puns as evidence of pronunciation. Our database will shed light on this issue by providing a far greater set of data on rhymes across first the 1590s and then the early modern period. It might also tackle and challenge in its methodology some of the assumptions that have been made about pronunciation on the basis of rhyme. As we develop the database, looking to ascertain which words constitute rhyming pairs, we will need to examine works on early modern pronunciation, but as these works have often used literary texts as their evidence, we face a thorny but fascinating methodological challenge in deciding the extent to which seeming rhymes can act as evidence of actual pronunciation.
26> The songs and sonnets in Sidney’s Astrophel and Stella constitute one of the most varied and sophisticated poetic sequences ever written, and were a major influence on the dazzling metrical and rhyme forms of the Sidney psalms. Much has been written about Astrophel and Stella as a political and social text, as well as about its Petrarchanism, anti-Petrarchanism and its playful manipulations of courtly lyric conventions. The sonnets indeed register tensions and competing impulses, between political activity and alienation, desire and frustration, authority and submission, the wishes of the individual and the rules of society. A database recording the structural rhyme and metre, rhyme words, and ideally the moments of irregularity in these poems would allow a far more nuanced analysis of whether these tensions and energies are embodied in the poems’ forms in ways that a reading of individual poems, or even reading all the poems in sequence, might not reveal.
27> The 1590s saw very few women publishing their works, or at least publishing under their own names. The translation of Robert Garnier’s Marc Antoine by Lady Mary Sidney Herbert, Countess of Pembroke, is a notable and fascinating exception. The Countess of Pembroke’s Antonius reveals her characteristic formal sophistication. Garnier’s drama was written more for reading or domestic performance than for public productions, and Antonius’ popularity (editions in 1592 and 1595) might be due to its suitability to informal performance.[21] Antonius marks an interesting point in English drama, as Pembroke adapts Senecan drama from a French work in alexandrines, to her own combination of blank verse (only in the 1590s becoming the dominant form for drama) with metrically diverse rhymed choruses. The excellent recent editions of the Countess of Pembroke’s works alert readers to the formal variety of the Countess’ works, but our database would allow her innovations to be studied in much greater detail and with greater scope in the context of her contemporaries.[22] The influence of Antonius on Shakespeare’s Antony and Cleopatra has been noted, but much remains to be explored in terms of the influence of translated drama, of drama by women and of the emergence of blank verse drama, and by allowing synchronic and diachronic comparisons of large texts our database would provide unprecedented quantity and quality of data from which to build such analyses.[23]
28> The nature of poetry in the 1590s also demands that we consider including works circulated in manuscript as well as those published in print. The Sidney Psalter, a series of 150 psalms composed by Philip Sidney and Mary Sidney Herbert, Countess of Pembroke, was famously praised by John Donne as ‘the highest matter in the noblest form.’[24] It was never published in the early modern period, and yet it was as highly esteemed and as influential as any printed work. If our database were able to include such a work, it would make it possible to ask new questions as well as to investigate current scholarly claims with an unprecedented data set. Fascinating work has been done on how both the Sidneys’ psalm translations were influenced by the forms of Sidney’s Astrophel and Stella, but the database would allow us to ask far broader questions about how manuscript poetry and print poetry cross-fertilised, how the psalter influenced its readers in their own writing, how the psalms compare formally to other works by both Philip and Mary Sidney. There are many reasons to include a work such as the Sidneys’ psalms in our database, not least its technical audacity and influence, but some of these qualities also make it a test-case for the difficulties of expanding our remit from first editions of printed works to manuscripts. The question of authorship for the psalter has largely been resolved, with the first forty-three psalms written by Philip Sidney and then revised by his sister, and the rest of the psalms her work alone. The case is not so clear for many manuscript works, where the form of publication and circulation itself led to a more flexible, collaborative process of writing and reading. This would pose questions for the database in terms of how to categorise authorship, possibly with levels of certainty indicated. While these questions are of course pertinent to printed texts, existing databases such as EEBO provide a fairly authoritative precedent for attribution. The question of dating is similarly vexed with manuscript texts. Perhaps most difficult would be the question of what to include. We would need access to full-text transcriptions of manuscript poetry that are being created currently in an active but dispersed field. We would need to address the possibility of working with a confined field, at least initially. This might mean starting with the poetry in manuscripts in one particular institution, for instance the Bodleian or British Library, or working on the texts included in a particular catalogue, such as Peter Beal’s Index of Literary Manuscripts.[25]
Conclusion
29> The digitization of books and archives has both stimulated exciting new scholarship and provoked concern that the digital image has lost many of the attributes and features of the printed page that make it a work uniquely of its period. This electronic database of poetic form would build on existing digitization to create new ways of thinking about poetry both in its moment of production and far beyond it. Each poem, whether lyric, epic, song, verse drama, will be open to analysis both for its own formal features, whether common or unique, and also in the context of its year, decade or century. We hope that this database will provide a framework in which scholars can find unprecedented ways of answering their research questions and, crucially, a tool that will provoke them to ask new questions.
Notes:
1. Heather Dubrow, ‘The Country House Poem: A Study in Generic Development,’ Genre, 12 (1979), 153-79 (p. 159).
2. For further discussion of these issues, excluding electronic resources, see Dubrow, ‘The Politics of Aesthetics: Recuperating Formalism and the Country House Poem. in Renaissance Literature and its Formal Engagements, ed. by Mark David Rasmussen (New York; Basingstoke: Palgrave, 2002), pp. 67-88.
3. For an introduction to these resources in early moderns studies, see Matthew Steggle, ‘“Knowledge will be multiplied”: Digital Literary Studies and Early Modern Literature’ in A Companion to Digital Literary Studies, ed. by Ray Siemens and Susan Schreibman (Oxford: Blackwell, 2007), pp. 82-105.
4. Steggle (2007), pp. 82-105, (p. 88). For further discussion of this aspect of humanities computing projects, see Susan Schreibman, ‘Computer-mediated Texts and Textuality: Theory and Practice,’ Computers and the Humanities, 36 (2002), 283-293.
5. By ‘P5-compliant,’ we mean that our database will follow the latest version (‘P5’) of the TEI Guidelines for Electronic Text Encoding and Interchange. The guidelines define and document a markup language for representing the structural, renditional and conceptual features of texts. An online version of the P5 guidelines is available at http://www.tei-c.org/release/doc/tei-p5-doc/en/html/, [accessed 26 April 2010].
6. P5: Guidelines for Electronic Text Encoding and Interchange.
7. David Chisholm and David Robey, ‘Encoding Verse Texts,’ Computers and the Humanities, 29 (1995), 99-111 (pp. 108-109); reprinted in Text Encoding Initiative: Background and Context, ed. by Nancy Ide and Jean Véronis (Dodrecht; London: Kluwer Academic, 1995), pp. 99-102.
8. Ben Burton and Elizabeth Scott-Baumann, ‘The Problem of Prosodic History: Towards an Electronic Database of Poetic Form’ (in progress).
9. Derek Attridge, The Rhythms of English Poetry (London: Longman, 1982), p. 147. Historical studies of early modern prosody tend to follow traditional methods of scansion. See, for example, Susan Woods, Natural Emphasis: English Versification from Chaucer to Dryden (San Marino, CA: Huntington Library, 1984); O. B. Hardison, Prosody and Purpose in the English Renaissance (Baltimore and London: John Hopkins University Press, 1989).
10. Richard D. Cureton, Rhythmic Phrasing in English Verse (London: Longman, 1992), p. 429.
11. Thomas Carper and Derek Attridge, Meter and Meaning: An Introduction to Rhythm in Poetry (London: Routledge, 2003), p. 87.
12. Peter L. Groves, Strange Music: The Meter of the English Heroic Line (Victoria, B. C.: English Literary Studies, University of Victoria, 1998), p. 33.
13. For a useful introduction, see Mark David Rasmussen, ‘Introduction: New Formalisms?’ in Rasmussen ed. (2002), pp. 1-16. On the shifting fortunes of formalism in literary studies, and for a lively defence of its continued importance to literary scholarship, see Richard Strier, Resistant Structures: Particularity, Radicalism, and Renaissance Texts (Berkeley and Los Angeles, CA: University of California Press, 1995).
14. For an overview of current work in this field, see the essays collected in Shakespeare and Historical Formalism, ed. by Stephen Cohen (Aldershot: Ashgate, 2007), as well as in Reading for Form, ed. by Susan J. Wolfson and Marshall Brown (London: University of Washington Press, 2006), originally published as a special issue, Modern Language Quarterly 61:1 (2000).
15. ‘An Essay on Criticism’, l. 365 in The Twickenham Edition of the Poems of Alexander Pope, gen. ed. John Butt, 11 vols (London: Methuen, 1938-1968) I, p. 281.
16. On the uses of EEBO in particular, see Thea Lindquist and Heather Wicht, ‘Pleas’d By a Newe Inuention?: Assessing the Impact of Early English Books Online on Teaching and Research at the University of Colorado at Boulder,’ The Journal of Academic Librarianship, 33:3 (2007), 347-60.
17. Louise Tickle, ‘Shakespeare goes Digital,’ The Guardian, Tuesday 26 January 2010. [Accessed 1 February 2010].
18. See, for instance, the introduction to Shakespeare’s Poems, ed. by Katherine Duncan-Jones and H. R. Woudhuysen (London: Arden Shakespeare, 2007).
19. Heather Dubrow, Captive Victors: Shakespeare’s Narrative Poems and Sonnets (Ithaca: Cornell University Press, 1987), p. 16; see also, looking at numeric patterning rather than form, Christopher Butler and Alastair Fowler, ‘Time-beguiling sport: Number Symbolism in Shakespeare’s Venus and Adonis,’ in Venus and Adonis: Critical Essays, ed. by Philip C. Kolin (New York: Garland Publishing, 1997), pp. 157-169.
20. Though scholars including J. Paul Hunter have productively questioned many assumptions about the couplet and its hegemony. Hunter, ‘Formalism and History: Binarism and the Anglophone Couplet,’ Modern Language Quarterly, 61 (2000), 109-129.
21. See Selected works of Mary Sidney Herbert Countess of Pembroke, ed. by Margaret P. Hannay, Noel J. Kinnamon, and Michael G. Brennan (Arizona Center for Medieval and Renaissance Studies, Tempe, Arizona, 2005), p. 42.
22. The Collected works of Mary Sidney Herbert Countess of Pembroke, ed. by Margaret P. Hannay, Noel J. Kinnamon, and Michael G. Brennan (Oxford: Clarendon Press, 1998), 2 vols; Selected Works, ed. by Hannay, Kinnamon, Brennan.
23. Michael Steppat, ‘Shakespeare’s Response to Dramatic Tradition in Antony and Cleopatra’ in Shakespeare- Text, Language, Criticism: Essays in Honour of Marvin Spevack, ed. by Bernard Fabian and Kurt Tetzeli von Rosador (Hildesheim: Olms-Weidmann, 1987), pp. 254-79.
24. ‘Upon the Translation of the Psalms by Sir Philip Sidney, and the Countess of Pembroke His Sister’, l. 11 in The Sidney Psalter: The Psalms of Sir Philip and Mary Sidney, ed. by Hannibal Hamlin and others (Oxford: Oxford University Press, 2009), p. 3.
25. Peter Beal, Index of Literary Manuscripts, 2 vols (London: Mansell Publishing Ltd, 1987).
_____
Elizabeth Scott-Baumann is a Research Fellow at Wolfson College, Oxford University and a Lecturer in Renaissance Literature at Oxford Brookes University. She has published articles in Literature Compass and Women’s Writing and co-edited The Intellectual Culture of Puritan Women, 1558-1680 (Palgrave Macmillan, forthcoming). She is currently writing a book called Forms of Engagement: Women, Poetry and Culture 1640-1680.
Ben Burton is a lecturer at Oxford University. He has published articles in Renaissance and Reformation / Renaissance et Réforme as well as entries for The Story of the Church in England (forthcoming). He is currently writing a book called Poetics of the Eucharist from Robert Southwell to John Milton.
_____
APPOSITIONS: Studies in Renaissance / Early Modern Literature & Culture, http://appositions.blogspot.com/, ISSN: 1946-1992, Volume Three (2010): Digital Archives
Oxford University
Encoding Form: A proposed database of poetic form
1> In 1979, Heather Dubrow called for a new methodology in early modern studies. ‘We are still prone to separate literary history and generic analysis,’ she observed, ‘[b]ut the two modes of criticism can illuminate each other
“exploring why certain genres flourished when they did and how they shaped and were shaped by the temper of their age is one of the surest ways of tracing the complicated movements that inform literary history.”[1]
Thirty years later, it seems that literary scholars have been slow to respond to Dubrow’s appeal for a new kind of historical formalism, especially when developing electronic resources.[2] To be sure, the last four decades have seen exciting developments in digital literary studies, including e-texts and online corpora; multimedia electronic editions; full-text resources for biography, bibliography, and lexicography; as well as computer-assisted textual analysis. In the field of early modern studies, online resources such as Early English Books Online (EEBO), Chadwyck Healey’s Literature Online (LION), and Gale Cengage’s Eighteenth Century Collections Online (EECO) have enabled scholars and students to conduct literary historical research in novel ways, allowing users to discover obscure publications and pursue new lines of enquiry through keyword-searchable full texts.[3] Such resources as these have transformed the way we study words and phrases, but they are not designed to support sophisticated analysis of the formal properties of early modern texts. As scholars begin to reassess the relationship between formal and historical study, we need electronic resources that will enable scholars and students to study form in a historically-rooted way, stimulating new questions about parallels and differences across literary periods.
2> What might such a resource look like? This paper explores the motivations and challenges of creating a database of form for early modern English poetry. We are currently planning a database that will contain information on the genres, metrical structures and rhyme schemes of all poetry in early modern printed texts. Users will be able to search a variety of criteria, exploring specific poetic forms by name, or metre, or rhyme scheme and then refine search results within a time period, or according to author, or place of publication, or language. They would also be able to enter information about the formal properties of a poem, such as its rhyme scheme, and be able to view similar or identical forms in the database. The database would change the study of poetic form in the way that LION, EEBO and other keyword-searchable full text corpora have revolutionized our ability to study imitation, influence and tradition at the thematic and content level of words. In this paper we will outline the kinds of data this resource could provide, and the kinds of questions we would want to ask of it.
Purpose and Objectives
3> This database will contribute to an ongoing effort to combine literary history with the study of prosody and genre. In doing so, we hope to reassess the scope and purpose of prosodic analysis, and its relation to electronic resources, in shaping our understanding of the early modern period. The database would in no way replace scholarly analysis, but it can assist us in certain aspects of prosodic study, and will open up new areas for teaching and research.
4> The primary objective of the database is to provide an unprecedented quantity of information about poetic form – such as syllable count, rhyme scheme and rhyme words – which form the basis of prosodic and generic analysis. It does not aim to offer definitive or exhaustive analyses, but to stimulate further enquiry and to provide data for independent analysis. The database assists prosodic study by dramatically reducing the time taken to perform basic metrical analysis such as counting syllables and identifying rhyme schemes. More excitingly, it opens up areas of enquiry by enabling scholars to ask new questions about literary texts. At present, prosodic analysis is usually confined to isolated studies of an individual author or small groups of authors. One of the ultimate objectives of the database is to broaden the scope of prosodic analysis by providing the raw materials for a cultural history of poetic form and generic change across the early modern period. In this sense, the database will enable scholars to ask the same kinds of questions about form that our current resources allow us to ask about words, for example: What is the origin (or origins) of a given form? How does its structure, use and meaning change over time? Are there variations in use and meaning in different regions, or among different groups? How does a given form relate to others, and how does this relationship change over time? To pursue the analogy with lexical analysis, the database invites us to explore what might be called the etymology and morphology of poetic form, as well as the complex history of prosodic dialects, sociolects and idiolects.
Methodology
5> The database would encode additional information about the generic and formal characteristics of texts in our existing online corpora of early modern texts. At present, resources such as EEBO are designed as digital libraries and, as such, are not concerned with adding value by annotating the texts they compile.[4] Our strategy is to create an interface that would interrogate marked-up texts, as users present queries, so that the user’s search would call up certain poems and records of these poems’ formal features. Users would be able to see a short, searchable record of formal features for each text but also to search much more widely and narrowly according to their own research interests. We propose to annotate existing texts created by the Early English Books Online Text Creation Partnership (EEBO-TCP), enriching these texts by providing detailed information about their prosodic structure. We would use a Text Encoding Initiative (TEI) P5-compliant scheme to enrich the relevant encoded EEBO-TCP texts.[5] The TEI already provides a base tag set for encoding verse, including stanza divisions, caesurae, enjambment, rhyme scheme and metrical information, as well as a special purpose rhyme element to support the simple analysis of rhyming words. For rhyme and metrical analysis, the module for verse includes attributes that allow users to specify encoding for the conventional metrical structure of the element, as well as for the actual realization of the conventional metrical structure applicable to a given element of a text.[6]
6> What kind of information might be marked-up in this way? As David Chisholm and David Robey point out, ‘[t]here is...a general question as to whether text markup for scholarly interchange should be limited to broad formal conventions (e.g. meter, rhyme, stanzaic forms), or whether individual characteristics of verse texts may also be included (a step that often represents the foundation of a stylistic interpretation).’[7] What is more, since prosodic analysis necessarily entails an element of critical judgement, any encoder will need to formulate clear principles for marking up verse. For example, can we always distinguish between formal conventions and individual characteristics of verse? How should we indicate that the pattern indicated by our mark up is not the only possible pattern for the verse text under consideration?
7> We believe that some of these questions are best approached through practice. In our pilot project outlined below, we will test both methods of encoding on a sample of texts spanning a range of genres from the 1590s, including sonnet sequences, epic poetry and drama. At present, we can only speculate as to which approach will be most appropriate for a project of this scope. We suspect, however, that a simple mark-up of rhyme scheme and conventional metrical structure will provide more information about texts that involve a variety of stanzaic patterns – such as sonnet sequences, or collections of lyric poetry – than texts that use a single stanza, such as epic poems, or those which rely heavily on a single conventional metrical structure, like drama. If this is true, it seems likely that the project will need to provide information about formal conventions as well as a limited range of individual characteristics. Of course, this would be labour intensive. Manually marking up the actual metrical structure of poetic texts at line level could prove extremely cumbersome and time consuming. One possibility is that the database might develop in phases, beginning with rhyme scheme and conventional metrical structure, and then progressing to include further information about the actual features of these texts. This approach allows the flexibility to encode additional layers of complexity. These could include highlighting irregularities in rhyme scheme, or indicating any deviation in the real metrical structure of a line from the pattern defined by the conventional metrical structure. In the Pilot Project section (below) we discuss the possibilities of this kind of mark up, as well as its limitations in relation to our pilot project.
8> This pragmatic approach can only take us so far, however, as we attempt to address larger theoretical questions about how we mark-up verse texts. In a future article we will examine the theoretical assumptions that lie behind the TEI guidelines for encoding verse, and will situate our database in relation to recent historical studies of early modern prosody.[8] At this early stage, we can only highlight some of the large problems that remain to be answered by our project. For example, is the purpose of the database to examine poetry in the terms that early modern poets may have used, or to provide an analytical framework that draws on current prosodic theory? The former approach would allow users to see the period in something like its own terms, but it is open to criticism from theoretical prosodists who distinguish between ‘the conscious metrical principles to which poets may have adhered’ and ‘the principles which actually governed their practice.’[9] This is an important distinction, but it leaves us wondering how these ‘principles’ are to be defined in light of the ongoing methodological conflicts in verse studies. As Richard Cureton observes, ‘the failure of theoretical prosodists in [the twentieth] century to develop a generally accepted method of rhythmical analysis has made it impossible for literary historians and historically minded prosodists to take advantage of these advances in any coherent way.’[10]
9> These methodological questions are complicated further by practical and pedagogical concerns: should we adopt a traditional accentual-syllabic method of scansion that is already used in many textbooks and prosodic manuals, or should we eschew this arguably outdated system in favour of a more credible but less familiar method of analysis? A possible solution to this problem, adopted by Thomas Carper and Derek Attridge, is to combine modern methods of scansion and traditional terminology (such as ‘iambic pentameter’ and ‘trochaic trimeter’), with the proviso that these terms provide a ‘useful shorthand for describing particular sorts of meters’ rather than ‘keys to what’s really going on in metrical lines.’[11] This solution may be pragmatic, but it might also be seen as theoretically suspect. As Peter Groves argues, ‘the traditional description [of verse texts] lacks intellectual coherence and rigour, and should not be perpetuated merely by inertia, incumbency, habit or apathy.’[12] While we admire Groves’s refusal to compromise on this point, we hope to establish principles for encoding verse texts which address the concerns of theoretical prosodists as well as scholars, teachers, students and non-specialists who are already familiar with traditional methods of analysing poetic form.
Dissemination and Publication
10> Online databases and editions are now a central part of UK and US universities’ research resources. Oxford University, where we work as lecturers and researchers, currently subscribes to resources with a variety of search functions and searchability, such as The Bible in English (searchable by key word, or chapter and verse), the Poetry Archive and Past Masters (largely philosophical texts). Reference works are also increasingly used in their online forms. The Oxford English Dictionary and the Oxford Dictionary of National Biography are now largely used online in frequently updated and carefully managed versions, and scholarly journals increasingly make their articles accessible through JStor.org and Project MUSE.
11> Many databases providing primary texts, such as LION, EEBO and ECCO, are now commonly subscribed to by universities and widely used by undergraduates, postgraduates and senior scholars. These existing databases offer a sweep of literature from medieval to twentieth-century, and some provide both digital facsimiles and full-text versions of rare (and common) books. Digital facsimiles have enabled thousands of students and scholars to study rare, fragile books as they were originally published, including manuscript drafts (see Oxford’s First World War Poetry Digital Archive) and manuscript texts (the JISC-funded Digitizing Correspondence project under way at London’s Centre for Editing Lives and Letters; the Perdita Project providing facsimiles of early modern women’s writings). Full-text versions of swathes of poetry have been used for stylometric analysis to help ascertain authorship or date a work in a writer’s life or used to trace allusion, echoes and imitations of other authors (just as the availability of scholarship online has enabled many universities to use plagiarism software such as Turnitin to screen their undergraduates’ work).
12> The Text Creation Partnership provides texts to Chadwyck-Healey, who mount them online through their (otherwise facsimile-and-bibliographic-record only) EEBO database. EEBO therefore provides access to both facsimiles and an increasing number of full-text transcriptions of early printed books. We are keen to work with EEBO-TCP texts for the early modern phase of the database, using text encoding techniques to create enriched texts from which the searchable interface will glean and exhibit data on formal aspects of all poetry. We aim to negotiate with EEBO and Chadwyck-Healey to find a form of publication that both respects intellectual property and commercial interests, while also making this rich new material accessible to the widest possible audience.
13> At the current stage of planning we are considering several options for the dissemination of our research findings and the publication of the database. These include 1. An open access interface available to any internet user that provides both a searchable interface and access to marked up texts; 2. Access to searchable interface and marked-up texts to EEBO-TCP subscribing institutions, perhaps as one of the partners’ custom interfaces; 3. Open access to database and texts but not with mark up. This would mean that if we are not able to make the XML-encoded texts freely available, we would display the texts in their entirety (as users request them), but with the encoding invisible. The interface would interpret our encoding and display the verse with, for example, its stresses marked with accents, or its rhyme scheme colour-coded, rather than with visible tags; 4. Open access to database but no texts. We are most keen on the first option, but this will depend on how our negotiations with EEBO-TCP and Chadwyck-Healey progress. Option 3 would have obvious benefits over Option 4 in that institutions who do not subscribe to EEBO such as schools, where we are keen to see the database used in teaching, would have access to the texts as well as our data, just without the mark up visible.
14> This project would result in a sustainable resource that could be developed beyond the sixteenth and seventeenth centuries across every period, and could be connected to huge existing databases of poetry already subscribed to by universities, and ideally also be available to school students across the UK and USA.
Applications: Research and Teaching
15> We are currently in an exciting period of literary criticism, as scholars review the formal, political, theoretical and historical approaches of the previous century and call for perspectives that unite formalist literary and historical analysis.[13] Their work questions the critical tendency to conflate certain poetic genres such as lyric with outmoded critical methodologies of the early twentieth century, and encourages us to rethink the longstanding association between generic analysis and a rarefied concept of the aesthetic. Eschewing the New Critical ideal of formal unity or coherence, these scholars propose a freshly theorized ‘historical formalism,’ reassessing poetic form as historically specific, historically determined, and historically efficacious.
16> There has already been some excellent work in this field, concerned with the rise and fall of genres, changes in prosodic tastes and practices, how writers distinguished and differentiated the specific functions of literary texts, as well as a history of aesthetics, all of which have been sidelined by a cultural studies movement which has dominated literary scholarship in recent years.[14] At present, however, the scope of this research is restricted by the lack of electronic resources that would make prosodic information of all sorts more easily accessed, more quickly searched, and more widely shared. Indeed, we would imagine that many scholars interested in historical formalism are currently wrestling with the same kind of problems that frustrated a previous generation of scholars working on vocabulary and verbal allusion. As a result, scholars interested in formal and historical analysis have understandably tended to focus their energies on analysing a limited sample of texts or a small group of authors. Larger studies, spanning a broad time period and a wide range of authors, are often richly suggestive and point the way towards future research, but scholars need new resources to provide the kinds of evidence that could support these ambitious projects.
17> In creating an electronic database of poetic form for the early modern period, we would seek to provide a crucial tool in this emergent field, enabling scholars to explore the unpredictable, multiple interactions between form and history, while stimulating new questions about the cultural significance of specific forms as well as how particular genres changed and developed over time.
18> The multiple research implications of this database would change the way in which scholarship on poetic form is conducted. Scholars will be able to make new connections between authors of different periods, styles and cultural identities. For example, they would be able to trace the progress of a particular form, looking at how the heroic couplet was used before its most famous practitioner, Alexander Pope, and after him. It would also be possible to discern how common metrical irregularity was within certain forms, and whether this was dictated by subject matter, education or gender. Scholars would also be able to identify heterogeneity within certain forms, for instance, exploring how much formal variation existed within the sonnet boom of the 1590s, which variants endured into the seventeenth and eighteenth centuries, and which were abandoned but received a later revival.
19> Finding a poem difficult to categorise, it would be possible to use the database to investigate similar forms, exploring possible models and traditions, or revealing valuable innovations in form. Rhyme words would also be detailed in the database, providing rich new evidence of how rhyme could function: Were certain rhyme words always used for comic effect? Do particular rhymes appear more frequently in certain periods than others, and what might this reveal about the cultural significance of rhyme at these moments? It was a commonplace of the New Criticism of the earlier twentieth century that, in Pope’s words, ‘The Sound must seem an Eccho to the Sense’[15]; our resource would aim to provide the largest amount of data with which to explore the myriad aesthetic and political ramifications of this principle.
20> The Electronic Database of Poetic Form would also be an immensely useful teaching tool, at both school and university level. Recent research has confirmed that online resources such as EEBO are already transforming undergraduate teaching and learning.[16] Electronic resources are used by undergraduates more than any other group and our database would draw on enthusiasm for online material to challenge and develop their reading of poetry. Undergraduate students might be able to submit entries of rhyme schemes and syllable counts on a connected pedagogical website, developing their own skills at identifying certain metrical forms and exploring the implications of rhyme words and different kinds of rhyme.
21> In the same way, the database could be used to teach secondary school students about rhyme and metre, providing an accessible, hands-on introduction to key concepts of prosodic analysis, and enabling them to explore the variety of poetic forms used by a particular author. Students might be able to highlight and annotate specific features of their chosen texts, following online exercises organized by a teacher, and submitting work for further discussion in an electronic forum with class members. Here we find a particularly strong warrant for making our resource freely available to any internet user, especially to institutions which cannot afford the subscription fees which accompany many electronic resources in the humanities. Although we are currently considering several options for the publication of our database, we have been encouraged by the recent development of open access resources, such as the new Shakespeare Quartos Archive (http://www.quartos.org/), that are freely available for sophisticated, close analysis of literary texts for students and scholars across the globe.[17] The success of this resource convinces us that online resources can satisfy the demands of the scholarly community as well as the wider public. In our pilot project, described below, we will experiment with an interface that will be accessible to non-specialists while also providing complex search features for more advanced users.
Pilot Project: 1590s
22> The pilot project comprises two stages: 1. A small database containing information on the metrical structures and rhyme schemes of all verse in the first edition of 10 texts published between 1590 and 1599. 2. A larger database containing information on the metrical structures and rhyme schemes of all verse in first editions of texts published during this period.
23> The first stage of the pilot will illustrate the variety of texts contained in this database, ranging from those written entirely or predominantly in verse (such as epic poems, verse drama and sonnet collections), to those which include poetry and prose in varying proportions. The two principal investigators will work on the initial project, and the results of this work will help determine the search functions of the database, as well as the timescale and the resources needed for the second stage of the pilot. The second stage will demonstrate the complete search functions of the database across a sample period. We aim to provide information about poetry in first editions published between 1590 and 1599, ranging from works written entirely in verse, to brief quotations and epigrams that feature in larger works.
24> The 1590s was a unique and exciting decade in the development of poetry in English, with poets experimenting with classical metres, forging and theorising a vernacular literary tradition, reflecting continental influences, as well as the inflections of music and rhetoric. It saw the publication of Sidney’s Defence of Poesie, one of the most influential formulations of the purpose and responsibilities of poetry, as well as his sonnet sequence Astrophel and Stella. The decade saw the publication of Spenser’s major English epic, The Faerie Queene, Marlowe’s Ovidian Hero and Leander and Shakespeare’s two great narrative poems, Venus and Adonis and Lucrece. These latter two have been characterised as the most patterned, structured and even the most ‘literary’ of all Shakespeare’s works.[18] Heather Dubrow has talked of modern readers’ ‘distrust of elaborate devices’ and though these poems are widely studied, much remains to be explored in their formal patterns and innovations.[19] These poems enjoyed a much greater fame in Shakespeare’s lifetime than they do now, and it was these works that motivated contemporaries to write about the honeyed softness of Shakespeare’s verse. Our database would allow further exploration of what exactly such critical terms meant in the sixteenth and seventeenth centuries. Were these works defined by their formal regularity, by enjambment creating an admired sense of flow, or can we see ways in which irregularity and shifts in metre might also be admired? These terms of smoothness and softness became connected with the couplet in the eighteenth century and it would be fascinating to be able ultimately to compare the couplets of the sixteenth century with those of the eighteenth. Did poets’ use of the couplet become more uniform as it gained dominance?[20] Shakespeare’s narrative poems of course do not employ couplets, but for Venus and Adonis iambic pentameter sestets (rhyming ababcc) and for Lucrece rhyme royal (7 pentameter lines rhyming ababbcc). The Electronic Database of Poetic Form would allow users to input these schemes (either metre, rhyme scheme or both) and immediately see what other works were published in the period using the same stanzas, whether these became more or less common after Shakespeare’s poems were published, whether these forms were used for comic, tragic or other material, amongst numerous other research questions.
25> The 1590s also saw the rise of some of the greatest works in English verse drama, which this database will include, such as Marlowe’s Tamburlaine, Dido Queen of Carthage and Edward II, and Shakespeare’s Romeo and Juliet, Titus Andronicus, Richard II and Love’s Labour’s Lost. It has often been said that late sixteenth-century drama and poetry influenced each other, and this database would provide illuminating evidence of the exact nature of this influence. Certainly Love’s Labour’s Lost includes poetry which alludes to song forms such as madrigal verse, as well as strophic forms, and is the most heavily rhymed of Shakespeare’s plays. The rhymes are often semantically significant, such as the repeated pairings of ‘lord’ and ‘word,’ ‘eyes’ and ‘lies’. These rhymes were indexed in Helge Kökeritz’s Shakespeare’s Pronunciation of 1953, which used linguistic features such as rhyme words and puns as evidence of pronunciation. Our database will shed light on this issue by providing a far greater set of data on rhymes across first the 1590s and then the early modern period. It might also tackle and challenge in its methodology some of the assumptions that have been made about pronunciation on the basis of rhyme. As we develop the database, looking to ascertain which words constitute rhyming pairs, we will need to examine works on early modern pronunciation, but as these works have often used literary texts as their evidence, we face a thorny but fascinating methodological challenge in deciding the extent to which seeming rhymes can act as evidence of actual pronunciation.
26> The songs and sonnets in Sidney’s Astrophel and Stella constitute one of the most varied and sophisticated poetic sequences ever written, and were a major influence on the dazzling metrical and rhyme forms of the Sidney psalms. Much has been written about Astrophel and Stella as a political and social text, as well as about its Petrarchanism, anti-Petrarchanism and its playful manipulations of courtly lyric conventions. The sonnets indeed register tensions and competing impulses, between political activity and alienation, desire and frustration, authority and submission, the wishes of the individual and the rules of society. A database recording the structural rhyme and metre, rhyme words, and ideally the moments of irregularity in these poems would allow a far more nuanced analysis of whether these tensions and energies are embodied in the poems’ forms in ways that a reading of individual poems, or even reading all the poems in sequence, might not reveal.
27> The 1590s saw very few women publishing their works, or at least publishing under their own names. The translation of Robert Garnier’s Marc Antoine by Lady Mary Sidney Herbert, Countess of Pembroke, is a notable and fascinating exception. The Countess of Pembroke’s Antonius reveals her characteristic formal sophistication. Garnier’s drama was written more for reading or domestic performance than for public productions, and Antonius’ popularity (editions in 1592 and 1595) might be due to its suitability to informal performance.[21] Antonius marks an interesting point in English drama, as Pembroke adapts Senecan drama from a French work in alexandrines, to her own combination of blank verse (only in the 1590s becoming the dominant form for drama) with metrically diverse rhymed choruses. The excellent recent editions of the Countess of Pembroke’s works alert readers to the formal variety of the Countess’ works, but our database would allow her innovations to be studied in much greater detail and with greater scope in the context of her contemporaries.[22] The influence of Antonius on Shakespeare’s Antony and Cleopatra has been noted, but much remains to be explored in terms of the influence of translated drama, of drama by women and of the emergence of blank verse drama, and by allowing synchronic and diachronic comparisons of large texts our database would provide unprecedented quantity and quality of data from which to build such analyses.[23]
28> The nature of poetry in the 1590s also demands that we consider including works circulated in manuscript as well as those published in print. The Sidney Psalter, a series of 150 psalms composed by Philip Sidney and Mary Sidney Herbert, Countess of Pembroke, was famously praised by John Donne as ‘the highest matter in the noblest form.’[24] It was never published in the early modern period, and yet it was as highly esteemed and as influential as any printed work. If our database were able to include such a work, it would make it possible to ask new questions as well as to investigate current scholarly claims with an unprecedented data set. Fascinating work has been done on how both the Sidneys’ psalm translations were influenced by the forms of Sidney’s Astrophel and Stella, but the database would allow us to ask far broader questions about how manuscript poetry and print poetry cross-fertilised, how the psalter influenced its readers in their own writing, how the psalms compare formally to other works by both Philip and Mary Sidney. There are many reasons to include a work such as the Sidneys’ psalms in our database, not least its technical audacity and influence, but some of these qualities also make it a test-case for the difficulties of expanding our remit from first editions of printed works to manuscripts. The question of authorship for the psalter has largely been resolved, with the first forty-three psalms written by Philip Sidney and then revised by his sister, and the rest of the psalms her work alone. The case is not so clear for many manuscript works, where the form of publication and circulation itself led to a more flexible, collaborative process of writing and reading. This would pose questions for the database in terms of how to categorise authorship, possibly with levels of certainty indicated. While these questions are of course pertinent to printed texts, existing databases such as EEBO provide a fairly authoritative precedent for attribution. The question of dating is similarly vexed with manuscript texts. Perhaps most difficult would be the question of what to include. We would need access to full-text transcriptions of manuscript poetry that are being created currently in an active but dispersed field. We would need to address the possibility of working with a confined field, at least initially. This might mean starting with the poetry in manuscripts in one particular institution, for instance the Bodleian or British Library, or working on the texts included in a particular catalogue, such as Peter Beal’s Index of Literary Manuscripts.[25]
Conclusion
29> The digitization of books and archives has both stimulated exciting new scholarship and provoked concern that the digital image has lost many of the attributes and features of the printed page that make it a work uniquely of its period. This electronic database of poetic form would build on existing digitization to create new ways of thinking about poetry both in its moment of production and far beyond it. Each poem, whether lyric, epic, song, verse drama, will be open to analysis both for its own formal features, whether common or unique, and also in the context of its year, decade or century. We hope that this database will provide a framework in which scholars can find unprecedented ways of answering their research questions and, crucially, a tool that will provoke them to ask new questions.
Notes:
1. Heather Dubrow, ‘The Country House Poem: A Study in Generic Development,’ Genre, 12 (1979), 153-79 (p. 159).
2. For further discussion of these issues, excluding electronic resources, see Dubrow, ‘The Politics of Aesthetics: Recuperating Formalism and the Country House Poem. in Renaissance Literature and its Formal Engagements, ed. by Mark David Rasmussen (New York; Basingstoke: Palgrave, 2002), pp. 67-88.
3. For an introduction to these resources in early moderns studies, see Matthew Steggle, ‘“Knowledge will be multiplied”: Digital Literary Studies and Early Modern Literature’ in A Companion to Digital Literary Studies, ed. by Ray Siemens and Susan Schreibman (Oxford: Blackwell, 2007), pp. 82-105.
4. Steggle (2007), pp. 82-105, (p. 88). For further discussion of this aspect of humanities computing projects, see Susan Schreibman, ‘Computer-mediated Texts and Textuality: Theory and Practice,’ Computers and the Humanities, 36 (2002), 283-293.
5. By ‘P5-compliant,’ we mean that our database will follow the latest version (‘P5’) of the TEI Guidelines for Electronic Text Encoding and Interchange. The guidelines define and document a markup language for representing the structural, renditional and conceptual features of texts. An online version of the P5 guidelines is available at http://www.tei-c.org/release/doc/tei-p5-doc/en/html/, [accessed 26 April 2010].
6. P5: Guidelines for Electronic Text Encoding and Interchange
7. David Chisholm and David Robey, ‘Encoding Verse Texts,’ Computers and the Humanities, 29 (1995), 99-111 (pp. 108-109); reprinted in Text Encoding Initiative: Background and Context, ed. by Nancy Ide and Jean Véronis (Dodrecht; London: Kluwer Academic, 1995), pp. 99-102.
8. Ben Burton and Elizabeth Scott-Baumann, ‘The Problem of Prosodic History: Towards an Electronic Database of Poetic Form’ (in progress).
9. Derek Attridge, The Rhythms of English Poetry (London: Longman, 1982), p. 147. Historical studies of early modern prosody tend to follow traditional methods of scansion. See, for example, Susan Woods, Natural Emphasis: English Versification from Chaucer to Dryden (San Marino, CA: Huntington Library, 1984); O. B. Hardison, Prosody and Purpose in the English Renaissance (Baltimore and London: John Hopkins University Press, 1989).
10. Richard D. Cureton, Rhythmic Phrasing in English Verse (London: Longman, 1992), p. 429.
11. Thomas Carper and Derek Attridge, Meter and Meaning: An Introduction to Rhythm in Poetry (London: Routledge, 2003), p. 87.
12. Peter L. Groves, Strange Music: The Meter of the English Heroic Line (Victoria, B. C.: English Literary Studies, University of Victoria, 1998), p. 33.
13. For a useful introduction, see Mark David Rasmussen, ‘Introduction: New Formalisms?’ in Rasmussen ed. (2002), pp. 1-16. On the shifting fortunes of formalism in literary studies, and for a lively defence of its continued importance to literary scholarship, see Richard Strier, Resistant Structures: Particularity, Radicalism, and Renaissance Texts (Berkeley and Los Angeles, CA: University of California Press, 1995).
14. For an overview of current work in this field, see the essays collected in Shakespeare and Historical Formalism, ed. by Stephen Cohen (Aldershot: Ashgate, 2007), as well as in Reading for Form, ed. by Susan J. Wolfson and Marshall Brown (London: University of Washington Press, 2006), originally published as a special issue, Modern Language Quarterly 61:1 (2000).
15. ‘An Essay on Criticism’, l. 365 in The Twickenham Edition of the Poems of Alexander Pope, gen. ed. John Butt, 11 vols (London: Methuen, 1938-1968) I, p. 281.
16. On the uses of EEBO in particular, see Thea Lindquist and Heather Wicht, ‘Pleas’d By a Newe Inuention?: Assessing the Impact of Early English Books Online on Teaching and Research at the University of Colorado at Boulder,’ The Journal of Academic Librarianship, 33:3 (2007), 347-60.
17. Louise Tickle, ‘Shakespeare goes Digital,’ The Guardian, Tuesday 26 January 2010.
18. See, for instance, the introduction to Shakespeare’s Poems, ed. by Katherine Duncan-Jones and H. R. Woudhuysen (London: Arden Shakespeare, 2007).
19. Heather Dubrow, Captive Victors: Shakespeare’s Narrative Poems and Sonnets (Ithaca: Cornell University Press, 1987), p. 16; see also, looking at numeric patterning rather than form, Christopher Butler and Alastair Fowler, ‘Time-beguiling sport: Number Symbolism in Shakespeare’s Venus and Adonis,’ in Venus and Adonis: Critical Essays, ed. by Philip C. Kolin (New York: Garland Publishing, 1997), pp. 157-169.
20. Though scholars including J. Paul Hunter have productively questioned many assumptions about the couplet and its hegemony. Hunter, ‘Formalism and History: Binarism and the Anglophone Couplet,’ Modern Language Quarterly, 61 (2000), 109-129.
21. See Selected works of Mary Sidney Herbert Countess of Pembroke, ed. by Margaret P. Hannay, Noel J. Kinnamon, and Michael G. Brennan (Arizona Center for Medieval and Renaissance Studies, Tempe, Arizona, 2005), p. 42.
22. The Collected works of Mary Sidney Herbert Countess of Pembroke, ed. by Margaret P. Hannay, Noel J. Kinnamon, and Michael G. Brennan (Oxford: Clarendon Press, 1998), 2 vols; Selected Works, ed. by Hannay, Kinnamon, Brennan.
23. Michael Steppat, ‘Shakespeare’s Response to Dramatic Tradition in Antony and Cleopatra’ in Shakespeare- Text, Language, Criticism: Essays in Honour of Marvin Spevack, ed. by Bernard Fabian and Kurt Tetzeli von Rosador (Hildesheim: Olms-Weidmann, 1987), pp. 254-79.
24. ‘Upon the Translation of the Psalms by Sir Philip Sidney, and the Countess of Pembroke His Sister’, l. 11 in The Sidney Psalter: The Psalms of Sir Philip and Mary Sidney, ed. by Hannibal Hamlin and others (Oxford: Oxford University Press, 2009), p. 3.
25. Peter Beal, Index of Literary Manuscripts, 2 vols (London: Mansell Publishing Ltd, 1987).
_____
Elizabeth Scott-Baumann is a Research Fellow at Wolfson College, Oxford University and a Lecturer in Renaissance Literature at Oxford Brookes University. She has published articles in Literature Compass and Women’s Writing and co-edited The Intellectual Culture of Puritan Women, 1558-1680 (Palgrave Macmillan, forthcoming). She is currently writing a book called Forms of Engagement: Women, Poetry and Culture 1640-1680.
Ben Burton is a lecturer at Oxford University. He has published articles in Renaissance and Reformation / Renaissance et Réforme as well as entries for The Story of the Church in England (forthcoming). He is currently writing a book called Poetics of the Eucharist from Robert Southwell to John Milton.
_____
APPOSITIONS: Studies in Renaissance / Early Modern Literature & Culture, http://appositions.blogspot.com/, ISSN: 1946-1992, Volume Three (2010): Digital Archives
1 comment:
One specific kind of search a user might undertake is not highlighted in your article but would, I think, be enabled by your database. It's searching for moments where the metre or rhyme scheme changes within a work. Even if different scholars don't quite agree on how to codify a particular metre or rhyme scheme, they are likely to agree on where the changes occur in the work--that is, they are likely to switch schemes at the same points in the work-- so no matter how you encode it the reader will be able to find the change.
I wonder, have you considered automated approaches to doing the markup, for example using a database (if one exists) of syllable counts in English words to automatically count syllables in whole lines? Or perhaps there is existing software that can count syllables by using mechanical rules regarding vowel and consonant placement. Such automated markup would be much less accurate than manual markup, but it would have two advantages: i) because the rules are mechanical and non-subjective, you'd know the markup was consistent (even if consistently wrong) across the entire database, and ii) you'd get a lot more texts marked up a lot more quickly.
You mention that in the pilot phase you'll be doing some small-scale marking up of specific texts to see how it goes. If you're looking for volunteers to just blindly follow your instructions and do the markup for a limited bit of literary writing, I'd be willing to give some time to this. Using an outsider who doesn't specialize in prosody might help get a sense of the range of possible interpretations (including erroneous ones) that can be made if you're not doing every bit of markup yourselves.
Gabriel Egan
Post a Comment