Poetic theory

Medieval Irish vernacular poetics were hierarchical, reflecting a prescriptive and idealised vision of Irish society. The various metres and poetic forms were held to correspond to the status and functions of various types of poets (filid and baird) and their appropriate patrons. Poetry with a social function – praise and satire – was of paramount importance, at the expense of the now much-anthologised informal lyrics. This emphasis no doubt reflects some degree of institutional continuity from the ancient Celtic world, for the old Celtic term for poet, bardos, pl. bardi, survived in Old Irish as bard, baird (and in Welsh as bardd, beirdd). Three Celtic learned classes held in exceptional honour were generally noted by the classical writers: bards who were singers and poets, given to singing praise and satire to the accompaniment of lyre-like instruments, vates (oυατεις) or diviners (μαντεις), and druids. These correspond to Old Irish bard ‘poet (of a lower grade)’, fáith ‘prophet (pagan or biblical)’, and druí ‘druid’, although the fact that the Irish terms occur alongside a plethora of other terms and not as a discrete triad discourages speculation about the possible survival of a tripartite system.

The Old Irish laws, which were mostly written in the seventh and eighth centuries, give us a generalised and probably rather idealised picture of early Irish society, as seen by the writers of the laws, who were either professional lawyers or legally minded churchmen (or both). The laws are obsessed with rank and status and are careful to specify the exact rank of poets and other professional persons in a hierarchy ranging from the ‘free/noble’ (sóer) down to the ‘unfree/base’ (dóer).

Beginning in the late eleventh century and ending in the mid-fifteenth century, ars dictaminis consisted of a highly formalised code of rules governing the composition of prose letters in Latin and, sometimes towards the end of the period, in the vernacular. While acknowledging that letters could also be composed in verse or in prosametron, a mixture of prose and poetry, the manuals of the ars focused solely on prose communication. If they emphasised eloquent expresion, the goal of the instruction was eminently practical: to attain the purpose for which the letter was being sent.

The ancients never developed a theoretical approach to letter-writing, claiming as they did that the letter needed the flexibility of conversation. In practice, in the case of letters having the status of official or public communications, however, they seem to have followed the rules governing oratory for which textbooks like Cicero's De inventione and the Pseudo-Ciceronian Ad Herennium abounded. By the twelfth century, especially in Italy, the word dictamen came to be intimately associated with the rules of letter-writing while the professional writer of letters or teacher of the art became known as the dictator. Whereas in practice the ancients distinguished between public and private letters, medieval ars dictaminis tended to ignore any distinction and assimilated all letters to the oratorical model. In Italy the generation of Peter Damian (d. 1073) and Pope Gregory VII (d. 1085) was the last until the advent of humanism in the fourteenth century to give a personal tone to correspondence.

Dante Alighieri (1265–1321) reflected on matters of literature throughout his œuvre. His earliest compositions (the tenzone with Dante da Maiano, and the sonnets ‘A ciascun'alma presa e gentil core’ and ‘Com più vi fere Amor co’ suoi vincastri’ (Rime 39–47, 1, 62)), written before he was twenty, are lyric ‘correspondence’ poems, examples of a genre whose self-reflective concern with questions of literature has been amply documented. Dante's youthful rime are largely mechanical and conventional; yet, looked at in terms of his overall development and his general approach to literature, they acquire greater significance. It is suggestive that the poet should have begun writing in a ‘style’ which allowed him to establish direct textual contact with other poets and their works. From the very start, Dante was intrigued by the ‘mechanics’ of his art and by the ‘discourses’ which could accompany a literary text.

In the months before his death, Dante once again returned to the formulas of the poetic ‘debate’ (as he had on other occasions, most notably when he exchanged sonnets with Forese Donati and with Cino da Pistoia; see Dante, Rime 73–8; 110–15). Attacked in verse by the Bolognese magister Giovanni del Virgilio for having gone against traditional usage when he employed the vernacular instead of Latin, namely, a ‘low’ language instead of a ‘high’ one, to present the intellectually demanding subject-matter of the Commedia (see Dante, Egloghe I, 1–34), Dante defended his choice of language through a complex and multi-layered literary strategy. Although, especially in the first of his two replies, the poet explicitly and ironically rejected the idea that he had behaved improperly (Egloghe II, 51-4; pp. 44-5), he developed his full critique of Giovanni in a rather more subtle and implicit way: one, in fact, which underlined his poetic sensibility.

Written transmission of German texts begins in the Old High German (= OHG) period. The earliest texts of significance date from the last third of the eighth century, but the richest period of transmission is the ninth century. Though the OHG period witnesses the dawn of writing in German it was already a period of massive cultural transition. This was the time of the spread and consolidation of Christianity among the German tribes; consequently a few texts preserve tantalising traces of pagan thought. Virtually all the texts which survive do so thanks to the church. The church relied on Latin writing, while secular culture was orally transmitted in the vernacular. Some texts reflect the struggle to express the Latin culture of the church in German writing; others the attempt to record secular vernacular culture. And within the particular sphere of vernacular poetry we witness the giant leap from alliterative verse to rhyme.

The surviving OHG texts form a varied if not particularly extensive corpus. Doubtless many more texts were written than have come down to us, but the total written output of the period represents but the tip of the iceberg of the whole of OHG literary culture. That any secular literature has survived at all is remarkable given the church's monopoly of writing. As Alcuin declared in 797, there was not room for both: ‘The Word of God should be read while the clergy are eating. They should listen to the reader, not to the harpist, to the sermons of the Fathers, not to the songs of the pagans.

Though late-medieval Wales produced no Ars poetica it did produce texts which enunciated principles and standards relevant to literary composition and which highlighted matters at issue in contemporary literary life. Foremost among these were the various recensions of the bardic grammar (gramadegau'r penceirddiaid) and the contentions (ymrysonau) involving some of the leading poets of the age.

The bardic grammar

Copies of the bardic grammar survive in four medieval manuscripts: Aberystwyth, National Library of Wales, MS Peniarth 20 (c. 1330); Oxford, Bodleian Library, MS Jesus College 111 [the Red Book of Hergest] (c. 1400); Aberystwyth, National Library of Wales, MS Llanstephan 3 (c. 1425), and Bangor, University of Wales, MS Bangor 1 (c. 1450). The relationship between the texts is complicated: the manuscript copies represent different recensions with considerable variation in wording, order and substance, the Red Book, Llanstephan 3 and Bangor 1 texts displaying an affinity which is not shared with the Peniarth 20 version. Paradoxically, in view of its date, it is the Red Book text which preserves the earliest recension of the grammar; notwithstanding its status as the earliest manuscript text the Peniarth 20 grammar represents a later, more developed recension. Peniarth 20, however, provides a terminus ante quem for the composition of the grammar, and this, together with internal evidence – a metrical example dated to 1316–17 – makes it likely that it was composed during the second decade of the fourteenth century.

The history of medieval literary studies properly begins in the early fourth century, and is heralded by a renewal of interest in the study of the great Latin writers of the past. With the exception of Horace's Ars poetica, the major texts of ancient literary criticism seem to have been little known during this period, and critical activity is largely confined to commentaries directed to teachers and students of grammar and rhetoric. It was in the classroom that the emergent Christian culture came to terms with its pagan heritage, and the negotiations this involved largely determined the shape of education and literary criticism throughout the medieval period.

The role of the great Latin authors of the republic and early empire in preserving the authority of ancient culture during the third and fourth centuries is hard to overestimate. The study of literature had always been an important part of the imperial system of education, and the prestige of literary culture seems to have grown during this period. In the face of the growing influence of Christian apologists sceptical of all pagan institutions, education became an important preserve of national culture, and literature was increasingly valued as a repository of ancient traditions of all kinds. The later fourth century is an important time in the history of commentary on classical texts and of textual criticism, and Virgil, Livy and other major figures seem to have enjoyed in some circles a virtual cult status.

Among German poets of the early thirteenth century there emerges a new literary self-consciousness which manifests itself in references to the author's own person in the body of his poem and in allusions to other writers and their work.

Such references to other poets may be implicit. That is the case, for example, when the narrator in Wolfram von Eschenbach's Parzival (c. 1210) claims in 436, 4ff. that Sigune's love for the dead knight Schionatulander was such that, if the couple had been married, lady Lunete (a character in Hartmann von Aue's Iwein, c. 1200) would have been more reticent in advising Sigune to remarry than she was towards her bereaved mistress Laudine. The critical comment on the behaviour of Lunete, who presses her mistress to remarry on the very day of her husband's death, remains at the level of a playful critique of a character in another poem (as distinct, say, from being a comment on Hartmann's character motivation). By the contrast between his own characters' behaviour and that attributed to the characters of another fictional work Wolfram's narrator claims a special seriousness for the qualities of triwe (‘fidelity’) and minne (‘love’) displayed by Sigune (and later also by the hero Parzival); he marks out the divide between the values of his own literary world and those of ‘traditional’ Arthurian romance. At the same time the author establishes a relationship between the ethical constructs of his literary fiction and the real world, the world in which the composition and reception of Iwein and Parzival is to be situated.

In his Biographia literaria (written 1815, published 1817), Samuel Taylor Coleridge gave his theory of the literary imagination its fullest exposition. The ideal poet, he declares, ‘brings the whole soul of man into activity, with the subordination of its faculties to each other, according to their relative worth and dignity. He diffuses a tone and spirit of unity that blends and (as it were) fuses, each into each, by that synthetic [i.e. synthesising] and magic power to which we have exclusively appropriated the name of imagination’ (pp. 173–4). This connection between imagination and the value of creative genius persists in modern use of the term ‘imagination’ and especially in use of the adjectival form ‘imaginative’. But it is the tradition which Coleridge was breaking away from, the empiricist-materialist view of literary imagination as memory images brought together by association, which has more in common with medieval views on the subject. Moreover, while Coleridge was interested in the psychology of composition, medieval thinkers were more interested in the psychology of audience-response, images being common property of author and audience, having a life beyond the psyches of their creators.

Standard late-medieval theory of imagination is cogently summarised in the encyclopaedic De proprietatibus rerum of Bartholomew the Englishman (compiled before 1250), which was translated into several European vernaculars. The brain, he explains (following a description which goes back to Galen), is divided into three small cells, the first being ymaginatiua, where things which the exterior senses perceive ‘are ordered and put together’; the middle chamber is called logica, where the power of estimation is master; and the third and last is memorativa, the power of remembrance, by which things which are apprehended and known by imagination and reason are held and preserved in the treasury of memory (3.10).

This is the first general history of medieval literary theory and criticism. It has been achieved through a long process of selection and compromise. When the project originally was conceived, we did not know (we could not have known then) what quantity and quality of materials awaited us, what would happen when scholars from a wide range of disciplines, segregated within the modern academy, would come together and pool their expertise – and, indeed, be encouraged to work on materials which had hitherto been ignored, or unexplored from the perspective of literary theory and criticism.

For a long time this subject has suffered from a refusal to believe in its very existence. George Saintsbury, in his History of Criticism and Literary Taste in Europe (1900–4), declared that ‘the Middle Ages were … certainly not Ages of Criticism’; ‘their very essence was opposed to criticism in any prevalence’ (I, p. 373). Writing some forty years later, in his English Literary Criticism: The Medieval Phase, J. W. H. Atkins challenged Saintsbury's claim, yet supposed that the period was ‘one of confused thinking in literary matters’ (p. 3). The 1957 short history of literary criticism by W. K. Wimsatt and Cleanth Brooks felt obliged to seek aesthetic, rather than distinctively literary, theory in the Middle Ages, and came away disappointed that ‘no new theory of beauty, of fine art in general, or of poetry’ is offered by St Thomas Aquinas or ‘other theologians of the high Middle Ages’ (p. 126). Giovanni Boccaccio's account of poetics in his Genealogy of the Gentile Gods is, however, given honourable mention.

This chapter is concerned with the arguments put forward by Italian humanists in the fourteenth and fifteenth centuries as to how and why poetry should be read, and therefore with one of the central themes of literary criticism, or more exactly literary theory, in the early Italian Renaissance. (The date from which the Renaissance starts is a matter over which opinions legitimately differ; here, as far as Italy is concerned, I am using the expression ‘early Renaissance’ to cover the fourteenth and fifteenth centuries.) The study and imitation of the classical poets was, of course, a major humanist interest, and poetry occupied an important place in the new curriculum of secondary education. But of all the subjects with which the humanists were concerned it was also the most contentious. In the fourteenth and fifteenth centuries Italian humanists were attacked from a number of quarters because of their poetic interests. The attack came from theologians, monks and the clergy, as well as from members of the established professions of law and medicine. Its main substance was the argument that classical poetry distracted men's minds from better things with stories that were not only pagan and therefore mendacious, but also lascivious and immoral. Moreover, they had been condemned both by classical figures such as Plato and Boethius and by Fathers of the church; ‘the songs of the poets are the food of devils’, St Jerome wrote for instance (Epist. 21.13 [CSEL 54]; see also Plato, Rep. 2.376-3, 403 and Boethius, De cons. phil. 1 met. 1).

Copies of Inferno and Purgatorio were already circulating in northern Italy when Dante died in September 1321, and these lost early exemplars were the forerunners of hundreds of fourteenth-century manuscripts of the Commedia, complete and partial. Few vernacular texts achieved so rapid or widespread a diffusion anywhere in medieval Europe. But Dante's poem was not long allowed to go about unaccompanied. By 1322 commentators were working on Inferno; by the decade's end a commentary on the whole Commedia had appeared, and the first century of Dante criticism eventually yielded a vast crop of exegesis. It includes full-scale commentaries, in which theoretical prologues, proems to each canto and textual glosses are combined to form an organic whole; sets of individual glosses (chiose), either discontinuous or in connected prose; and a variety of paraphrases, summaries, introductions, biographies, and other prolegomena, frequently in verse, which flourished on the margins of commentary proper, especially in the 1320s. New material continues to come to light: a long-lost Neapolitan commentary on Inferno, of 1369–73, was published in 1998. Commentators wrote in Italian and Latin, all over Italy (Naples, Milan, Bologna, Venice, Verona, Pisa) and abroad (Germany). Even Dante's loved and hated native city paid tribute to the poem that so ruthlessly dissects it: Giovanni Boccaccio and Filippo Villani lectured and wrote on Inferno in Florence, while the Ottimo commento and the work ascribed to the ‘Anonimo Fiorentino’ almost certainly originated there.

No single medieval discipline embraced all that we call literary criticism or theory today, but the discipline closest to literary criticism – in the sense of the interpretation of a traditional literary canon and the description of literary language – was grammatica. The scope and cultural effects of grammatica are large topics, embracing literacy, linguistic theory, traditions of commentary and exegesis, and the development of a literary canon, but here we must limit the field to those aspects of grammatica that had a direct bearing on the practice of literary criticism and interpretation.

Grammatica was traditionally defined as having two main methodological divisions and subject-areas: ‘the science of interpreting the poets and other writers and the systematic principles [ratio] for speaking and writing correctly’, that is, the methods for reading, interpreting and evaluating literary works, especially the canon of classical poets, and the rules or principles for speaking and writing according to normative Latin conventions. The literary division of the discipline, scientia interpretandi, was understood to have four main parts or methodological divisions – lectio, the principles for reading a text aloud from a manuscript, including the rules of prosody; enarratio, exposition of content and the principles for interpretation, including the analysis of figurative language; emendatio, the rules for establishing textual authenticity and linguistic correctness, and iudicium, criticism or evaluation of writings. In the linguistic division, the object of analysis was the language of classical literary texts, the auctores, not ordinary speech. From its beginnings, then, grammatica was a science of the text, embracing a systematic description of the authoritative textual language (Greek or Latin) and the methods for reading and interpreting an established literary canon.

One of the most dazzling and influential vernacular literary traditions of the twelfth and thirteenth centuries is that of the troubadours. The troubadour lyric quickly developed its own sophisticated poetics that was in turn to have a great impact on the development of lyric poetry and love casuistry in the literary traditions and cultures of Italy, Spain, northern France, Germany and England: Dante, for example, devoted a good portion of his literary criticism to the troubadours and similarly expounded his literary theory in part at least on the basis of his reading of troubadour chansonniers. The troubadours composed in a language that today we call Occitan, though earlier in the twentieth century it was (somewhat erroneously) known as Provençal and in the Middle Ages it could be designated by a variety of terms, some reflecting dialect (e.g. Lemosi), others merely vernacularity (Romans). Occitan was spoken until the early twentieth century in roughly the southern half of present-day France, but in the Middle Ages it was also used as a literary language at least in the twelfth century at the culturally crucial court of Poitiers and at a variety of courts in Catalonia, other parts of northern Spain, and then, in the thirteenth century, in Italy. Occitan was therefore a literary language with an international following, a language that some foreigners learnt, again particularly in the thirteenth century, with the specific intention of composing poetry.