Belgian Journal of Linguistics - Volume 15, Issue 1, 2001

Some concepts — and hence the thoughts that contain them — are relatively ineffable. Some literary communication nevertheless attempts to eff these concepts. This article is interested in the nature of such concepts and the extent to which pragmatics can deal with them. I discuss the idea, familiar from Relevance Theory and developed in Carston (1996; forthcoming), that pragmatic inferencing is involved in on-line ad hoc concept construction, certainly in the case of concept narrowing, but also possibly in the case of concept loosening. I then discuss the relative effability of non-lexicalised concepts, borrowing from Sperber and Wilson (1998) and focussing on phenomenal concepts (or concepts with a significant phenomenal component). I then define poetic thoughts as thoughts containing such concepts.

This article explores the extent to which the Relevance Theory concepts of interpretive and echoic use can help to explain the complexities of the use of voice in poetry. Echoic use in Relevance Theory is a sub-type of interpretive use, a use which can allow a speaker to communicate one of many possible attitudes towards a proposition, ranging from endorsement through disapproval to ridicule. My argument is that this model could be extremely powerful in accounting for the differences and relationships between perceived poets’/authors’ views and views presented directly in literary works. This approach goes some way towards integrating the study of poetry into a general account of communication. The article develops these arguments by using the Relevance Theory model in analysing the use of voice in a selection of poems by Dorothy Parker, Robert Browning, John Keats, Gerard Manley Hopkins, Fleur Adcock and Tony Harrison, and raises the question of whether all poetry might be considered interpretive.

By employing a mathematical characterisation of the distinction between prose and verse, namely the randomword length features of English prose and the non-random features of verse, it is possible to detect mathematical lineation in writings that are not typographically lineated. For example, such lineation can be shown to be present in T. S. Eliot's poem Burnt Norton (1941), and Jim Crace's prose ction Quarantine (1997). In the rst of these cases we show that the verse is lineated in units of four syllables, while the other sections of The Four Quartets are not lineated. In the second we show that Crace's text is lineated in syllabic groups of two, four, six, eight, ten, and subsequent multiples of two. Quarantine, we demonstrate, is non-randomly segmented, and while it does not employ a core isometric line length, and its lines do not follow on one from another, it is still, and in a novel and important sense, lineated. In this paper we offer further comments on appropriate statistical methods for such work, and also on the nature of formal innovation in these two texts. Additional remarks are made on the roots of lineation as a metrical form, and on the prose-verse continuum.

In this paper, we introduce the basic concepts needed for a pragmatic analysis of a specific kind of magic poetry, viz. Romanian love charms. We first distinguish between five types of charms: charms for love and beauty; charms for knowing one’s fated spouse; charms for bringing one’s fated spouse; charms for hate; charms for undoing hate. Our study concentrates on charm scenarios, i.e. on the instructions transmitted by charm-performers. Each charm scenario combines two sets of textual data: charm formulas (verses or poems) and charm techniques. The technique texts contain prose directions concerning the nonverbal component of the charm. After analyzing an example of each type of charms, we suggest that some rhetorical topoi of love charms (generally, metaphors) should be understood as modal indications that “guide” the magic process by illustrating its ideal result.

Corpora show that people are less original in using language than is generally believed. We routinely employ an immense repertoire of semi-preconstructed phrases, though we also adapt them: creative extensions and adaptations of institutionalized locutions sometimes occur more frequently than the ordinary form. Corpora also reveal that fiction uses verbal idioms rarely found in other forms of writing or in conversation, which suggests that novelists draw on their own experience of stereotyped fictional dialogue more than on real-life conversation. Oral epic poetry, from Homer to Beowulf, was, of course, also formulaic, but the received view is that written poetry should be quite the opposite: it should consist of new combinations of words. While it is easy to find poetry that does contain fixed expressions and poetic transformations of them, such as the ‘conversational’ (and occasionally prosaic) poetry of Wordsworth, Frost, Auden and McDiarmid, it is harder to argue that the poetry of Shakespeare, Donne, Shelley, Keats, Hopkins, Stevens or Ashbery is made up of formulaic language. Conversely, however, it can be shown that canonical poetry is the source of hundreds of phrases in our active verbal lexicons.

Lucretius wanted his poetry to convey the scientific truths of Epicurean doctrine. In addition, he thought that one single generative and organic principle unites atomistic physics and the combinatory production of meaning. Thanks to the interplay between its “outer metric” (the distribution of dactylic/spondaic feet and caesuras) and its “inner metric” (the collocation of metrical word-types), the Latin hexameter allowed him to create intricate networks of similarities and differences, aimed at expressing the central tenets of the Epicurean system in such a way that they acquire the “evidence” of sensory impressions. As shown by the detailed analysis of two excerpts (II, 308-332; I, 1-20), systematic choices operating at the level of both “outer” and “inner” metrics provide a mimetic representation of the very essence of Epicurean reality, whose constant change and movement produce order and complexity based on the functional properties of its pro-forms.

Recent work on Arabic metrics aims at developing a coherent research programme which relies on the systematic analysis of electronic corpora. The computer program XALIYL performs, for any line of ancient Arabic poetry, an automatic recognition of the metre used. This operation takes place whatever the length of the verses, and regardless of whether they are encoded in ordinary Arabic script (with the addition of vowels) or by means of the TRS system, which relates functionally to ordinary Arabic script. XALIYL produces a textual database that contains the syllabic decomposition for each hemistich of each line, as well as its metrical analysis. It can cope not only with the general problems linked to re-syllabification and sandhi, but also with problems of syllabification specific to Arabic metrics. Errors due to the metrical scanning or to the editing of poems can be located automatically. Moreover, by allowing a computerised search for formulae, XALIYL provides significant information on the “formulaic systems” of ancient Arabic poetry.

This paper deals with the metrical and rhythmical foundations of the formulaic style of ancient Arabic poetry. It is first shown how proper formulas can match different verse-patterns, by means of slight modifications such as the adjunction, deletion or substitution of conjunctions, prepositions, interrogative pronouns or aspectual markers, which partly behave like “stop-gaps”, keeping the meaning unchanged while modifying the metrical pattern of the formula. The analysis is then extended to “rhythmical formulas”, i.e. to combined metrical and word-stress patterns which serve as models for a great number of “formulaic expressions”. Word boundaries may be specified, as well as some morphological and syntactical informations, so that expressions derived from a same rhythmical formula can be classified into a number of more or less abstract subcategories. Finally, the syntagmatic combination of rhythmical formulas into lines leads to the identification of a small number of prototypical verse-instances underlying the various actual instances of a same verse-pattern.

This is a study of some metrical peculiarities of a corpus of 159 Egyptian-Arabic quatrains. The light CV-syllable, accepted in all metrical positions, is the exclusive type in three of the eleven positions of the standard line, but it figures together with heavy syllables in the eight remaining positions. Does quantity, therefore, matter in a restricted number of cases only? The riddle is solved by distinguishing between two classes of CV-syllables, depending on whether the vowel is lax or tense: CVlax is always light, CVtense is generally heavy, but light in a weak metrical position. Other findings concern CVCV, which may fill several strong positions in the metre, and CVVC, which may be either heavy or superheavy. Superheavy CVVC can fill a strong + weak position, or, together with CVlax, two strong positions. Nothing of the sort occurs in classical Arabic; still, classical and vernacular Arabic metrics are related to a point where the existing differences are easily overlooked. Only a thorough investigation of the vernacular metres on their merits can establish the exact relationship between classical and vernacular prosodic systems.

This article traces the origins of the Italian endecasillabo, the earliest surviving examples of which were composed in the thirteenth century, soon after the Albigensian Crusades destroyed the Occitan-speaking culture of the Southern French courts. It begins by discussing the achievements of that culture and, in particular, the ways in which the troubadours delighted in formal innovation and experiment, constantly recombining the structural elements of their verse in new ways. It then proceeds to analyse all the structural variants found in the earliest endecasillabi and demonstrates that each had antecedents in Occitan verse. This analysis supports Beltrami (1986) and Billy (2000), who argue that the endecasillabo evolved from the vers de dix largely as a result of the distinctive lexicon and phonology of the Italian language. It then examines the alternative hypothesis, that the endecasillabo evolved independently of its French and Occitan cognate metre, and finds all the arguments advanced in its favour unsatisfactory: the forged evidence of Baruffaldi (1713), the Latin origin proposed by Gasparov (1996), and the hypothesis of Meillet (1923) that all Indo-European verse has syllabic origins and a tendency to revert to them. This article concludes that the endecasillabo is a lasting legacy of the Occitan poets.

We propose a new classification of the Portuguese decasyllable into periods, as well as an overview of the specific features which have, over the centuries, marked the variety of this verse form. We thus distinguish between: the decassílabo trovadoresco (Middle Ages); the decassílabo quatrocentista (15th century); the decassílabo clássico (16th century); the decassílabo romântico (19th century); the decassílabo decadente e simbolista (late 19th and early 20th century). Whether in medieval or modern poetry, the Portuguese decasyllable exhibits an extreme variety of forms, rhythms and scansion patterns, all equally possible and codified in the poetic idiom; so that the only constant distinctive feature of the verse appears to be the compulsory accent on the 10th syllable. Moreover, the massive recourse to hiatus and dieresis, as well as to synaloepha and syneresis, always allows the Portuguese poets to attain the required number of syllables.

The Russian pentameter is historically associated with the English and German traditions, but typologically it has with some justice been compared to the French decasyllable. The present article analyzes the structure and cultural context of Russian pentameter and examines in detail the use of caesura in a small corpus of iambic pentameter poems by Afanasy Fet. It is shown that the use of caesura correlates with patterns of word stress. In particular, the appearance of caesuraed lines in poems in which caesura is relatively weak correlates with the stress patterns of the lines in question: caesuraed lines are less heavily stressed than uncaesuraed ones, a correlation that theoretically should promote equalization of line length across the text. Russian poetry has a general tendency to promote equality of line length, and the intrusion of occasional I6 lines into I5 texts, a phenomenon known in many Russian I5 poems, can be viewed as a related strategy for handling ragged I5 lines.

Current theories of meter tend to rely exclusively on stress-based properties. This paper argues that certain phenomena in Spanish metrics — some well-known, some neglected — should be accounted for in terms of intonational primitives. Any standard stress-based metrical pattern will be compatible with several Pierrehumbert-style intonational configurations, but there can also be alignment conflicts between stress and tone which cause metrical tension and/or unmetricality. As a case in point, the Spanish (hen)decasyllable, which has two possible stress patterns, discourages lines in which the intonational structure corresponds to the pattern that is not actually instantiated by them. Intonation-based treatments are also proposed for assonance and for the caesural features of the Spanish alexandrine.