When is a poem not a poem?  This week sees the final of a national poetry recitation competition, Poetry By Heart, for 14 to 18 year olds.  Encouragingly, given that medieval literature is these days not so often taught in schools, Middle English poetry is featured on the competition’s timeline of pieces from which to choose.  Indeed, last year’s winner, Kaiti Soultana, chose the extract from Sir Gawain and the Green Knight, in the original Middle English, as one of her recitations.  Here‘s her winning performance.

As well as Gawain, the Middle English pieces in the timeline are Chaucer’s description of the Wife of Bath from the General Prologue of the Canterbury Tales (a great piece of  narrative description, full of knowing comment) and the lyric ‘I sing of a maiden’ (scroll down for text and translation).  Whilst I am hugely in support of this marvellous competition, I was somewhat surprised by this last choice.  It is, of course, a famously delicate, somewhat enigmatic, much anthologised text, capturing the mystery of the Annunciation.  Though it seems deliciously sparse, much scholarship has shown how it ingeniously reworks scriptural and liturgical reference and imagery.

Its fame and beauty may well have led to its inclusion in the timeline, but the assumption that it is a poem in all of the senses of that word might give some of the competitors, if they chose this poem, a little difficulty.  Some of the things we think of as Middle English ‘poems’ (and might try to scan as poems, if we were attempting to recite them) are song lyrics, lyrics written to be sung and not imagined as being spoken metrically. To present them as poems runs the risk of leaving them high and dry without the necessary clues to help the reader puzzle out their form and poetics.

Thomas G Duncan has a useful appendix on ‘Music and Metre’ in his anthology of Medieval English Lyrics, 1200–1400.  As its first line indicates, this lyric too may well have been composed as song lyric rather than stand-alone poem, even though it is not accompanied by music in its manuscript.  It is found in British Library MS Sloane 2593, a collection of mostly carols and some songs, most in English, but some in Latin and some macaronic.  An unrelated fifteenth-century sermon alludes to lines from the lyric as something which would be sung, but it also shares some lines with an earlier poem (perhaps suggesting the Sloane version is a song based on an earlier poem).

It’s not that you can’t read out song lyrics as if they were poetry, but, if you do, you are, it seems to me, doing something rather different, creating an alternate realisation of a song’s words in speech rather than feeling one’s way towards something the poet himself might have intended.  Songs are written for melody and rhythm whilst poems are written for metre.  (Metre of course means different things in different sorts of Middle English verse, syllabic, accentual, accentual-syllabic).

Derek Pearsall, who recently edited ‘I sing of a maiden’ in his Chaucer to Spenser anthology, notes that it is written in ‘short three-stress lines’ (which are laid out as long-line couplets in the manuscript with mid-line punctuation, so four lines in the layout given below becoming two lines in the manuscript).  This makes it sound unproblematically poetic, but perhaps this in fact describes its song rhythms in verse terms.  It can indeed be read rhythmically with three (but sometimes two) stresses per line (a speculative scansion, trying to give a sense of the rhythm, is marked by bold text below).  The rhythm seems in many places to require the possibility of a very brief beat or breath mid-line and/or between lines.  It is not regularly metrical but (as would make sense if it were written to follow a melody) it is full of varying and repeating rhythms, especially the repeated dactylic patterns of As dew in Aprylle that ( / x x / x x).  It’s a tough but worthwhile challenge for a teenager reciter.