Reconstructing Alliterative Verse: The Pursuit of a Medieval Meter by Ian Cornelius (review)

Recent years have seen a flourishing of publications on the subject of medieval English metrics. The Cambridge Studies in Medieval Literature series alone has produced, along with Ian Cornelius's Reconstructing Alliterative Verse, books by Geoffrey Russom (2017) and Eric Weiskott (2016) on medieval English alliterative poetry. Like these books, Reconstructing Alliterative Verse operates across a long temporal scale, from the seventh through sixteenth centuries, with occasional forays into comparisons with late antique Latin poetry. And like Weiskott and Russom, Cornelius rejects the idea of the so-called "Alliterative Revival" of the fourteenth century, arguing for a continuous tradition of alliterative meter spanning the English Middle Ages, despite the lack of written evidence of such poetry composed c. 1250–c. 1350. The book asks how alliterative meter arose; how it developed over time; and, finally, why it fell out of favor in the fifteenth and sixteenth centuries.

In support of its claims, the book also offers a radical revision of how we theorize alliterative meter. Building on the 2008 dissertation of Nikolay Yakovlev, Cornelius describes a system of Old English alliterative meter constructed not around four stressed syllables per line (as in the Sievers five-type system), but rather around four metrical "positions" per line, which are filled by morphemes. Morphological units are not categorized as stressed and unstressed, but rather "strong" or "weak," depending on the part of speech to which they belong and their position in the line. In Cornelius's example, Beowulf, 54a—"leof leodcyning"—would be scanned according to Yakovlev's morphological system as SSSS, where "S" indicates a strong and "x" a weak position. In more traditional scansion, informed by phonological stress patterns, we would expect leodcyning to have primary stress on leod- and secondary stress on -cyn-, with the suffix -ing unstressed. Yakovlev's theory does away with the distinction between primary and secondary stress in polysyllabic Old English words; positions are strong or not, with no gradations between them. Half lines have a minimum of two strong positions but may have as many as four strong positions, and in some metrical types the number of morphemes in some weak positons can or must be more than one. Yakovlev's system results in eight possible types of meter (61). Stress (or accent) remains possible in Old English verse (and existed phonologically in the spoken language), but is not the primary determiner of meter. Much of the book's argument about both Old English and Middle English meter relies on the reader accepting Yakovlev's ideas, as presented and retested by Cornelius. Not all will. Yakovlev's theory is gaining interest and attention among metrists—Weiskott and Russom also refer to his work, and it has been reviewed by Thomas Cable—but it is not necessarily widely accepted among early medievalists. Early English medievalists who study vernacular poetry but do not identify as metrists may be surprised to learn of its existence.

Reconstructing Alliterative Verse is divided into an introduction, five chapters, and a brief epilogue on Spenser's Shepheardes Calender (1578). Chapter 1 argues that even if an ars poetica for medieval English had survived, we would still need to turn to the poetry itself to recover a full understanding of alliterative verse. Not only does practice always deviate from theorized norms, but also the evidence we have for poetic theory from earlier Latin sources and from medieval Iceland demonstrates that medieval literary theory would not have had the vocabulary to address poetic form sufficiently. This is not a critique of medieval thinkers, but an acknowledgment that poetry is very difficult to write about. As Cornelius notes, "[t]he theorists are still playing catch-up today" (25). Moreover, medieval people did not inherit a useful set of terminology from Latin (or Greek) for alliteration, because alliteration would have functioned differently in English than in Latin and because many authors of Latin treatises either ignored alliteration or saw it as hopelessly old-fashioned.

One...