Rhythmical Figures in Dante’s Commedia: A Study of Memory and Composition after Gianfranco Contini Ryan Robert Pepin Pembroke College This thesis is submitted for the degree of Doctor of Philosophy February 2020 This thesis is the result of my own work and includes nothing which is the outcome of work done in collaboration except as declared in the Preface and specified in the text. It is not substantially the same as any that I have submitted, or, is being concurrently submitted for a degree or diploma or other qualification at the University of Cambridge or any other University or similar institution except as declared in the Preface and specified in the text. I further state that no substantial part of my thesis has already been submitted, or, is being concurrently submitted for any such degree, diploma or other qualification at the University of Cambridge or any other University or similar institution except as declared in the Preface and specified in the text. It does not exceed the prescribed word limit for the relevant Degree Committee. Abstract Rhythmical Figures in Dante’s Commedia: A Study of Memory and Composition after Gianfranco Contini Ryan Pepin This thesis takes Contini’s 1965 study ‘Un’interpretazione di Dante’, and its discussion of repeated lines and part-lines, as its point of departure in order to pursue new understanding of the compositional processes of Dante’s Commedia. It shows how the repetition of similar items, or ‘figures’, relates to orality both in a transmissional, or textual-critical sense, as well as in a compositional, or oral-formulaic, sense. Recollected language in the poem is discovered to be more pervasive than hitherto thought, and this data is then interrogated, revealing a conception of heterogenous units of rhythm and syntax susceptible of re-use. This challenges the frameworks for prosodic description that have been used to date to describe the poem, while setting up a new historical understanding of rhythm as ‘shape’. These intertwined rhythmical- syntactical units, revealed through line-by-line comparison of the entire poem, are then shown to be historically emplaced – a product of the language-learning practices of the medieval grammatica. Figurae uerborum, a medieval conception of syntax and rhythm belonging to the grammatica, and Contini’s ‘figure ritmiche’ together raise an interpretive challenge: they ask for an autonomous way of speaking about achievements of syntax and rhythm, with reference to new critical categories – particularly memorability and authority. A final section proposes a rhythmical-critical experiment: a practical-critical analysis of a specific rhythmical shape, exploring how the new compositional dynamics discovered in the course of the thesis change how the critic relates to the newly conceived poetic object. From medieval grammarians to twentieth-century philologists, this thesis uncovers a tradition of thought that sees rhythm not as an accident of poetic language, but as a grammatical phenomenon, best exemplified in the poets. Acknowledgements My conversations about Dante and Contini with Robin Kirkpatrick, Claudio Ciociola, and Kenneth Clarke have been the inspiration for this work. I would also like to thank my advisor, Catherine Pickstock, for engaging with this work in its early stages, and for setting it enthusiastically going. I would like to thank my examiners, Simone Marchesi and Kenneth Clarke, for this exceptionally kind feedback. My wife, Rosa van Hensbergen, made finishing this project possible. I am grateful to my friends for their conversation: to Tom Weber, Christian Coppa, Andrea Lazzarini, Laura Kilbride. I would also like to thank mom and dad for so many years of support and encouragement. I am indebted to the Cambridge Trust, Pembroke College, and to Sam Blyth for making my education and my doctoral research possible. Contents Introduction 1 Textual Criticism 32 Orality 68 Figure et mémoire. Contini and Valéry 118 Figura and Grammatica 137 Dignum et iustum est 173 Conclusion 192 Bibliography 195 Introduction What does a decent memory of Dante’s Commedia hear as it reads? A few terzine from the fifth canto of Purgatorio are enough to give a sense of how memory uncovers echoes between parts of the poem: Che potea io ridir, se non « Io vegno »? Dissilo, alquanto del color consperso che fa l’uom di perdon talvolta degno. E ‘ntanto per la costa di traverso venivan genti innanzi a noi un poco, cantando ‘Miserere’ a verso a verso. Quando s’accorser ch’i’ non dava loco per lo mio corpo al trapassar d’i raggi, mutar lor canto in un « oh! » lungo e roco; (Purg. 5.19-27)1 At the top of this quote Dante responds to Vergil’s injunction to follow, and to disregard the whisperings about him: ‘Vien dietro a me, e lascia dir le genti’ (Purg. 5.13). Dante has been reprimanded by Vergil once before now for his prurient regard for gossip. He remarked at that point how the reprimand lingered still in his memory: ‘volsimi verso lui con tal vergogna, / ch’ancor per la memoria mi si gira’ (Inf. 30.135-6). At the bottom of the above quote we also 1 Dante, La Commedia secondo l’antica vulgata, ed. by Giorgio Petrocchi, 4 vols (Florence: Le Lettere, 1994). All further canto and line references will be given in this format: canticle (Inf., Purg. Par.), followed by Arabic numerals for canto and line. Line references only will be given in the case of successive references to a particular passage. All further references to Dante’s Commedia will be to Petrocchi’s text, in the Le Lettere edition of 1994. Subsequent citations often include, however, my amendments to Petrocchi’s text, which are always explained in accompanying footnotes. 1 find the second instance in Purgatorio so far of a scene that repeats throughout the canticle.2 Dante’s shadow, his body which is impermeable to light, has startled a crowd, bringing its previous activity to a halt. Both of these actions then already have a layered significance, or are more like a second night than a début. But narrative analogies of this kind are not the only ones to emerge. Lines and part- lines, too, recall other lines and part-lines of the Commedia. Textually then as well, the poet works by putting new wine in old wineskins. Line 21 above – ‘che fa l’uom di perdon talvolta degno’ – is a shape of line the poet often makes with ‘degno’, placing the adjective at the end of the line with ‘di’ + nominal before it. So this line recalls the poet’s description of purgatory, a place ‘dove l’umano spirito si purga / e di salire al ciel diventa degno’ (Purg. 1.5-6). Sordello’s appeal to Vergil a few canti on comes also to mind: ‘S’io son d’udir le tue parole degno, / dimmi se vien d’inferno, e di qual chiostra’ (Purg. 7.20-1). All three purgatorial lines in turn recall a half-line from Inferno, in which Pier della Vigna swears that ‘già mai non ruppi fede / al mio segnor, che fu d’onor sì degno’ (Inf. 13.74-5). These textual echoes can be found throughout the passage. Line 23 above, in which a new set of purgatorial souls appear – ‘venivan genti innanzi a noi un poco’ – is a narratival, matter-of-fact replaying of Vergil’s words to the bewildered souls of Purg. 2. There, the newly arrived ‘nova gente’ (58) turn to the pilgrims for direction, and Vergil replies: …Voi credete forse che siamo esperti d’esto loco ma noi siam peregrin come voi siete. 2 The first being Purg. 2.67-75. Seeing Dante breathe, ‘s'affisar quelle / anime fortunate tutte quante, / quasi oblïando d'ire a farsi belle’ (73-5). 2 Dianzi venimmo, innanzi a voi un poco per altra via, che fu sì aspra e forte, (Purg. 2.61-5)3 The second hemistichs of these lines are identical (‘voi’ becomes ‘noi’), but the movement of venire back makes a pause-less line out of a strongly bifurcated one. There are other part-lines that a decent memory recognises, perhaps with a subtler sense of acquaintance, from elsewhere. Sometimes the recall seems beyond comment, because reused bits of language can be merely phrasal or idiomatic. That is the case with the second hemistich of line 24, ‘a verso a verso’. Singing ‘Miserere’ ‘line by line’ or ‘part by part’ seems natural enough. But does one also hear, in this phrasing, the iracondi of Inf. 7 employing all their limbs, ‘troncandosi co’ denti a brano a brano’ (Inf. 7.114)? Or Vergil’s duty in leading Dante ‘per lo ‘nferno qua giù di giro in giro’ (Inf. 28.50)?4 The initial gerund of this same line – ‘cantando “Miserere” a verso a verso’ – might also mark out a kind of line that is characteristic of this poet. Wrestlers fight, ‘avvisando lor presa e lor vantaggio’ (Inf. 16.24). A metamorphosed snake-man, ‘fiera divenuta / suffolando si fugge per la valle’ (Inf. 25.136-7). The poet asks that Calliope ‘alquanto surga / seguitando il mio canto’ (Purg. 1.9-10). Counting up to this point in Purgatorio, there are over twenty- five lines that begin with a gerund. Can common syntactical organisation, of that kind, also make for memorial association? Sometimes a prior instance of a particular word, used before in some significant way, is what is recollected in reading. Stronger still, though, is the memory of hearing the word in the same spot, or the same location in the line. So Adam’s words on the fall in Paradise, 3 I use bold type throughout when calling attention to an argumentatively relevant part of a citation from the Commedia.