counting and scanning

Posted on by Eric Weiskott

This weekend, I presented a paper at Digital Britain: New Approaches to the Early Middle Ages in Cambridge, MA. My gratitude to Sam Berstler, Joey McMullen, and Erica Weaver for the invitation, and for organizing the conference. My paper, “Counting and Scanning: Qualitative and Quantitative Approaches to Early English Meter,” is an essay in the philosophy of science, focusing on the use of quantitative reasoning in a specialized field of literary study. Here’s the opening frame:

This is a paper in defense of counting in literary studies. It’s also a paper about why counting needs defending in literary studies in 2016. First, I’d like to connect these topics to the conference theme.

The so-called digital turn in academic knowledge production poses a material challenge to the methodological status quo of the humanities. Digital technology makes a bold promise: the quantification of human experience. For constituents of the status quo, the promise of digital technology seems more like a threat. Many humanists worry about what is lost when the richness of culture and the arts is reduced to data. For others, digital tech enables the humanities to live up to their potential. To take an egregious example, Franco Moretti’s ‘distant reading’ does not merely claim to be a good method of reading literature; it claims to be the best method. In promoting statistical approaches to literary history, Moretti taps into philosophical attitudes that predate the digital. His aggressive empiricism is at once the attraction and the scandal of distant reading.

Thus the digital turn irritates a long-standing tension in humanistic study between qualitative and quantitative methods. I propose to explore this tension by focusing on a field of inquiry that bears an unusual relationship to it. Metrics, depending on your perspective, is either the field entrusted with explaining what makes poetry poetry, or the field charged with inferring an organizational system from linguistic patterning. Metrical scholarship combines qualitative and quantitative reasoning in a way that has become uncomfortable in literary studies.

Discomfort with the modus operandi of metrics is acute in the case of early medieval literature, whose systems of formal organization lie at a great cultural remove from those of modern literature. In reconstructing early medieval meters, the value of modern experiences of poetics, and hence the network of assumptions underlying data collection, is always open to question. About the meter of Beowulf, for example, it is currently possible to hold any one of a number of mutually contradictory theoretical views. Moreover, Old English metrical theories cluster in two incommensurable research paradigms, one continuously elaborated since the nineteenth century, the other disclosed for the first time in 2008 (see Cornelius, “The Accentual Paradigm in Early English Metrics”). Metrists disagree, and they disagree about why they disagree. For many literary scholars, meanwhile, metrics is a field that deploys the rhetoric of accumulation to make exaggerated claims on historical truth.

counting and scanning header

In light of these problems, what I’d like to do this morning is to sketch the function of counting in metrics, with reference to my research on English alliterative verse. I have two goals: to affirm the role of quantitative reasoning in literary study, and then to set a certain limit on that role. In what follows, I identify and address two philosophical challenges to counting as a critical method. One challenge comes from the right, framed in the language of neopositivism; the other comes from the left, framed in the language of poststructuralism. I contend that the neopositivist and poststructuralist positions both miss the full significance of counting as a way of getting at “the heart of a poem,” in the words of Simon Jarvis. Ultimately, I’ll argue that metrical study explodes the distinction between qualitative and quantitative approaches to literature.

meter as a way of thinking

Posted on by Eric Weiskott

Yesterday, I gave an invited talk for the MIT Ancient & Medieval Studies Colloquium Series. My gratitude to Arthur Bahr for the invitation. My talk was entitled “Early English Meter as a Way of Thinking.” Here’s the opening frame of the talk:

This paper is about structures of thought that happen to take the form of poetry. So stated, my object of inquiry would seem to be intellectual history, to which poetics is subordinated. However, I will strive to demonstrate that verse form is never incidental to the thinking it performs. Apprehending meter as a way of thinking necessarily involves reimagining thinking itself.

My title echoes Simon Jarvis, who recommends approaching “prosody as cognition.” Jarvis had Alexander Pope and William Wordsworth in mind when he coined that phrase. I seek to test Jarvis’s concept against a different literary archive, exploring the particular kinds of thinking done by and through early English meter. I’ll focus on the second half of the fourteenth century, a stretch of decades that saw a large uptick in the production of literature in English. As we will see, in medieval England meter was a way of thinking about form and balance, translation and vernacularity, and the historicity of literary practice. I’ll present three case studies introducing three kinds of metrical practice: the half-line structure in Middle English alliterative meter, the interplay between Latin and English in Piers Plowman, and final –e in Chaucer’s pentameter.

The protagonists of the three case studies are the three biggest names in Middle English literature: the Gawain poet, William Langland, and Geoffrey Chaucer. The first of these is no name at all but a cypher: the Gawain poet, thought to have composed the four poems in British Library Cotton Nero MS A.X. For this poet, no external evidence for authorship or biography has been identified. William Langland is little more than a floating name in literary history: mentioned in a few contemporary documents, Langland probably belonged to the well-to-do Rokele family. The name ‘Langland’ itself may be a pseudonym. Chaucer, of course, is the Grand Poobah of medieval English literature. Like Gilbert and Sullivan’s character, Chaucer was chronically overemployed; at one time or another he was a clerk, controller of customs, diplomat, esquire, forester, page, and soldier. These three poets have garnered the lion’s share of scholarly attention, and this paper follows suit by placing them at the center of an essay in historical poetics. But I’ll continually emphasize how the metrical practice of a range of contemporary and prior poets shaped the structures of thought informing Sir Gawain and the Green Knight, Piers Plowman, and the Canterbury Tales.

My broadest aim this afternoon is to convince you that intellectual history and poetics can inform one another. Indeed, where poetry is concerned, the procedures of the two fields ought to coincide. Medievalists have made significant contributions toward understanding poetry as cognition: I’m thinking especially of the work of Ruth Evans, Alastair Minnis, Fiona Somerset, Nicholas Watson, and others under the banner of what Minnis calls “medieval literary theory.” This research program compares the explicit theories of authority and textuality propounded in Latin by medieval scholars with the often implicit theorization of literature performed by vernacular texts themselves. To date, few medievalists have considered the intellectual significance of English meter, though I am indebted to the work of Thomas Cable, a metrist who has always insisted that the study of meter is about “mental structures.” From the perspective of intellectual history, I propose to enrich the study of medieval literary theory by disaggregating the English literary field by metrical tradition. Alliterative meter does not think the same way pentameter thinks; the difference should matter in any account of medieval literary theory. From the perspective of poetics, I propose to redirect the philological procedures of the highly traditionalist field of metrics toward a phenomenological poetics. If meter lives in the mind, then it is part of the job of a metrist to discover what it is doing up there.

phonology and literary history

Posted on by Eric Weiskott

This past weekend, I presented a paper at the Medieval Academy of America Annual Meeting in Boston. My paper, “Metrical Phonology and Literary History in the Age of Chaucer,” introduces three English metrical traditions–alliterative meter, tetrameter, pentameter–and points to some phonological evidence for cross-pollination between them in the fourteenth century. Here’s the opening frame:

At the end of Book V of Troilus and Criseyde, Chaucer apostrophizes the poem:

And for ther is so gret diversite
In Englissh and in writyng of oure tonge,
So prey I God that non myswrite the,
Ne the mysmetre for defaute of tonge.

(1793-96)

This passage has attracted significant attention as a testament to linguistic, metrical, and textual variation in medieval England. The apostrophe appears to substantiate conclusions that medievalists are ready to accept anyway: that Chaucer’s language, meter, and texts were in flux around him, and that Chaucer was exquisitely aware of this situation. In the standard interpretation, the speaker of the passage is a lot like a modern editor, worried about language change, metrical decay, and scribal error.

In this paper I’d like to suggest a different cultural context for Chaucer’s hand-wringing: the mediated interaction between language and meter in fourteenth-century English verse. ‘Metrical phonology’ is my term for the linguistic forms that meter encodes. Think, for example, of the variation between monosyllabic and disyllabic scansions of the word heaven in Elizabethan poetry. I will argue that metrical phonology should be understood in terms of the larger historical formation in which it is embedded: the poetic tradition. In the fourteenth century, there were three major metrical options for poets working in English: the alliterative meter, the tetrameter, and the pentameter. The alliterative meter had been in continuous use since at least the eighth century. The tetrameter entered the English literary field in the mid thirteenth century. Chaucer invented the pentameter on the basis of French and Italian models in the 1380s: Troilus and Criseyde is his first substantial work in this new verse form. The three major Middle English meters thus had different histories. Yet these histories also inflected one another. Study of metrical phonology calls for triangulation between historical and comparative analysis. That’s what I’ll try to provide.

By mapping Middle English metrical phonologies, it becomes possible to attain some critical distance from the Troilus and Criseyde passage. Through apostrophe, Chaucer hints at the complexity of fourteenth-century vernacular poetics. However, writing before English became an academic subject, Chaucer necessarily expresses his position in literary culture symptomatically rather than analytically. Attention to the historicity of metrical phonology helps us understand what Chaucer meant by the neologism ‘mismeter’ but also how meter could appear to Chaucer as a poetic problem in the first place.

Neidorf, Dating of “Beowulf”

Posted on by Eric Weiskott

My review of The Dating of “Beowulf”: A Reassessment, ed. Leonard Neidorf (Cambridge: D. S. Brewer, 2014), appears in the Review of English Studies. Here’s the opening of the review:

This collection, published under the same title as the 1981 University of Toronto Press volume that it hopes to supersede, grew out of a conference at Harvard University. Thirteen contributors reconsider the dating of Beowulf, and each of them concludes or accepts that Beowulf was composed in the eighth century. As the singular noun of the subtitle suggests, this Dating of ‘Beowulf’ is less a report from the field than a concerted provocation.

quantity in the alliterative tradition

Posted on by Eric Weiskott

This past weekend, I presented a short paper at the MLA Annual Convention in Austin. My paper, “Quantity in the Alliterative Tradition,” summarizes current thinking on the topic. Thanks to Natalie Gerber and Tom Cable for including my paper in a panel on quantity in English verse. Here is a modified version of the paper:

Quantity matters in the meter of Beowulf and other early English poems. It matters in the form of a metrical principle known as resolution. Metrical resolution served alliterative poets as a way of counting; it can serve modern scholars as evidence for the cultural meanings of verse craft. This paper therefore has two sections: How it Works and What it Means.

How it Works

Metrical resolution operates at the conjunction of metrical stress and syllabic quantity. Resolution works like this: a metrically stressed, quantitatively short syllable plus the following syllable is equivalent to a metrically stressed, quantitatively long syllable. Short stressed + any = long stressed. In the wacky math of alliterative meter, 1 + 1 = 1. Both sides of the equation count as a single metrically strong position, or ‘lift.’ Short syllables are those with an etymologically short vowel followed by zero consonants. (I mark long vowels with macrons.) So for example, in Beowulf 3a hū ða æþelingas ‘how the noblemen,’ æþel- undergoes resolution, resulting in the metrical pattern xxSrSx, where ‘S’ represents a lift without metrical resolution, ‘Sr’ represents a lift with metrical resolution, and ‘x’ represents an unstressed syllable. The pattern xxSrSx is equivalent to xxSSx but not to xxSxSx. In Old English meter, resolution is quasi-obligatory.

The relevance of metrical stress is what distinguishes resolution from the quantitative principles of classical meters. In alliterative meter, two adjacent unstressed, short syllables never add up to one long syllable. In other words, the first of the two syllables undergoing resolution must be one that receives stress. (In alliterative verse, metrical stress is assigned by prosodic weight: content words, such as nouns, receive stress, while function words, such as pronouns, do not.)

In Old English meter, resolution works in harmony with a number of other principles in the metrical system. The quantitative principle is like one functionality of a multifarious and well-oiled machine. The experience of applying resolution in versification and scansion must have been something like this: once metrical stress is assigned to a syllable, check the quantity. If long, count the syllable as a lift. If short, look to the right and count the next syllable together with the first syllable as a lift.

So resolution is a way of counting. It is equally important in historical perspective, as evidence for the development of the alliterative meter. The standard narrative has been that resolution fell into disuse around the time of the Norman Conquest (1066), along with most other features of Old English meter.

More recently, however, Nicolay Yakovlev has demonstrated that resolution continued to be used in alliterative verse into the thirteenth century. In his 2008 Oxford thesis, “The Development of Alliterative Metre from Old to Middle English,” Yakovlev convincingly identifies the use of resolution in Lawman’s Brut, a twelfth-century alliterative verse chronicle. So for example, in Brut 144b & þene dēað þolien ‘and suffer death,’ þoli– undergoes resolution, resulting in the metrical pattern xxxSSrx. The pattern xxxSSrx is equivalent to xxxSSx but not to xxxSSxx. Obviously the judgment that Brut 144b shows resolution depends on an idea of what metrical patterns were acceptable to Lawman. It is only by assuming a certain metrical pattern that you can project resolution in the first place. The same is true, by the way, of Old English meter, but I didn’t mention it earlier because we know a lot about the metrical patterning of Old English verse. Before Yakovlev, we knew next to nothing about the metrical patterning of the Brut. In Early Middle English alliterative meter, resolution is optional rather than quasi-obligatory.

Yakovlev’s arguments about metrical resolution are one small part of a paradigm-shifting demonstration that the alliterative meter was in continuous use from the seventh to the sixteenth century. This conclusion flies in the face of 75 years of metrical and literary scholarship. You may have heard of the Alliterative Revival. According to Yakovlev, no such event occurred. For Yakovlev, metrical resolution is one vector of formal continuity in metrical history. In other words, resolution helps us see that the meter used in Lawman’s Brut is a later instantiation of the meter used in Beowulf: significantly changed, yes, but through continuous development rather than reinvention.

Finally, let’s bring our story to the end of the alliterative tradition: the fourteenth, fifteenth, and early sixteenth centuries, the age of Sir Gawain and the Green Knight. The story here is simple. Resolution really did die out. Everyone agrees that resolution is no longer functional in Middle English alliterative meter. So for example, in Gawain 6b and pātrounes bicome ‘and became overlords,’ –come does not undergo resolution, resulting in the pattern xSxxxSx. The pattern xSxxxSx is not equivalent to xSxxxSr or xSxxxS.

To sum up, resolution was a historically dynamic feature of the English alliterative meter. Resolution was in use from the earliest recorded poems in the seventh and eighth centuries down to the beginning of the thirteenth century. First it was quasi-obligatory; then it became an optional feature; then it disappeared. The alliterative meter incorporated quantity for at least five centuries but then continued to evolve without quantity for three more centuries.

What it Means

Most immediately, metrical resolution means that alliterative poets were thinking about quantity in the process of versification and scansion. I want to emphasize how odd that is. Resolution recapitulates equivalences that are thought to have obtained in prehistoric Old English, when quantity played a larger role in the regulation of syllables. Yet resolution remained a feature of alliterative meter as late as c. 1200. So one way to understand resolution is as a metrical vestige: a linguistic principle became encoded as a metrical principle, and the metrical principle then outlived the linguistic one by centuries. By the time Lawman employed it, resolution had become a highly artificial principle, only thinkable in the context of a durable poetic tradition.

Alliterative poets were thinking about quantity in vernacular versification, but this thinking lay on a different conceptual plane from theoretical knowledge about Latin metrics. English alliterative verse rose and fell before poets began experimenting with classical quantities in English verse. The last alliterative poems also predate the earliest treatises on English meter. Obviously these two historical developments, metrical and intellectual, are directly connected: you cannot employ classical quantities in English meter until English meter becomes a reputable object of academic attention. And that did not happen until the closing decades of the sixteenth century. By then, the alliterative meter had already gone defunct. Alliterative poets have left behind no ars poetica and very little explicit commentary of any kind on their own metrical practice in the vernacular. Poets probably learned alliterative meter tacitly, through repeated imitation of their predecessors. Therefore, metrical resolution in the alliterative tradition was almost certainly not a learned imitation of classical meter. So another way of understanding resolution is as a cultural phenomenon: an illustration of how metrical features can function and fall away in the absence of explicit prosodic theory.

Partly because of the lack of an ars poetica, it has proven difficult for modern scholars to reconstruct alliterative verse. Since the early nineteenth century, alliterative meter has most often been described as accentual, but this is an overstatement. The quantitative principle coexisted with the accentual principle in the eleventh, twelfth, and early thirteenth centuries. Alliterative meter in these centuries was a blended form, accentual-quantitative. Moreover, Yakovlev has made the stunning argument that Old English meter was not accentual at all. (For the significance of this claim, see Cornelius, “The Accentual Paradigm in Early English Metrics.”) So a third way of understanding resolution is as a historiographical corrective: a reminder that alliterative verse was more complexly organized than you might have heard.

Metrical vestige, cultural phenomenon, historiographical corrective: in promising new research, quantity in the alliterative tradition is all of these, and more.

I have also deposited the paper in MLA CORE.