| s-1
| Distant Rhythm: Automatic Enjambment Detection on Four Centuries of Spanish Sonnets |
| s-2
| Pablo Ruiz Fabo |
| s-3
| pabloruizfabo@gmail.com |
| s-4
| Lattice Lab, CNRS, France |
| s-5
| Clara Martínez Cantón |
| s-6
| cimartinez@flog.uned.es |
| s-7
| Universidad Nacional de Educación a Distancia, Spain |
| s-8
| Thierry Poibeau |
| s-9
| thierry.poibeau@ens.fr |
| s-10
| Lattice Lab, CNRS, France |
| s-11
| Introduction |
| s-12
| Enjambment takes place when a syntactic unit is broken up across two lines of poetry (Domínguez Caparrós, 2000: 103), giving rise to different stylistic effects (e.g. increased emphasis on elements of the broken-up phrase, or contrast between those elements), or creating double interpretations for the enjambed lines (García-Paje, 1991). |
| s-13
| In Spanish poetry, the syntactic configurations under which enjambment takes place have been described extensively, and detailed studies on the use of enjambment by individual authors exist (see Martínez Cantón, 2011 for an overview) including, among others Quilis (1964), Domínguez Caparrós, (2000), Paraíso, (2000), Spang (1983) for a description of enjambment, and Alarcos (1966), Senabre (1982), Luján (2006), Martínez Fernández (2010) for case-studies on a single author. |
| s-14
| However, a larger-scale study to identify enjambment across hundreds of authors spanning several centuries, enabling distant reading (Moretti, 2013), was not previously available. |
| s-15
| Given that need, we have developed software, based on Natural Language Processing, that automatically identifies enjambment in Spanish, and applied it to a corpus of approx. 3750 sonnets by ca. 1000 authors, from the 15th to the 19th century. |
| s-16
| What is the interest of such large-scale automatic analyses of enjambment? |
| s-17
| First, the literature shows a debate about which specific syntactic units can be considered to trigger enjambment, if split across two lines, and whether lexical and syntactic criteria are sufficient to identify enjambment. |
| s-18
| Second, the stylistic effects that enjambment permits are also an object of current research (Martínez Fernández, 2010). |
| s-19
| Systematically collecting large amounts of enjambment examples provides helpful evidence to assess scholars’ current claims, and may stimulate novel analyses. |
| s-20
| Finally, our study complements Navarro’s (2016) automatic metrical analyses of Spanish Golden Age sonnets, by covering a wider period and focusing on enjambment. |
| s-21
| The abstract is structured thus: |
| s-22
| First we provide the definition of enjambment adopted. |
| s-23
| Then, our corpus and system are described, followed by an evaluation of the system. |
| s-24
| Finally, findings on enjambment in our diachronic sonnet corpus are discussed. |
| s-25
| The project’s website provides details omitted here for space reasons, including samples for the corpus, results, and other details. |
| s-26
| Enjambment in Spanish |
| s-27
| Syntactic and metrical units often match in poetry. |
| s-28
| However, this trend has been broken since antiquity for various reasons (Parry (1929) on Homer, or Flores Gómez (1988) on early classical poetry). |
| s-29
| In Spanish tradition, enjambment (in Spanish, 'encabalgamiento') is considered to take place when a pause suggested by poetic form (e.g. at the end of a line or across hemistichs) occurs between strongly connected lexical or syntactic units, triggering an unnatural cut between those units. |
| s-30
| Quilis (1964) performed poetry reading experiments, proposing that the following strongly connected elements give rise to enjambment, should a poetic-form pause break them up: |
| s-31
| Lexical enjambment: Breaking up a word. |
| s-32
| We translated 'lexical enjambment' from Quilis’s terms 'encabalgamiento léxico' or 'tmesis'. |
| s-33
| Phrase-bounded enjambment: Within a phrase, breaking up sequences like 'noun + adjective', 'verb + adverb', 'auxiliary verb + main verb', among others. |
| s-34
| We translated 'phrase-bounded enjambment' from 'encabalgamiento sirremático'. |
| s-35
| Cross-clause enjambment: Between a noun antecedent and the pronoun heading the relative clause that complements the antecedent. |
| s-36
| We translated 'cross-clause enjambment' from Quilis’s 'encabalgamiento oracional'. |
| s-37
| The project site includes Quilis’s complete list of syntactic environments that can trigger enjambment, as well as the types identified by our system. |
| s-38
| Besides the enjambment types above, Spang (1983) noted that if a subject or direct object and their related verbs occur in two different lines of poetry, this can also feel unusual for a reader, even if the effect is less pronounced than in the environments identified by Quilis. |
| s-39
| To differentiate these cases from enjambment proper, Spang calls these cases 'enlace', translated here as 'expansion'. |
| s-40
| Quilis (1964) was the only author so far to gather recitation-based experimental evidence on enjambment. |
| s-41
| His typology is still considered current, and was adopted by later authors, although complementary enjambment typologies have been proposed, as Martínez Cantón (2011) reviews. |
| s-42
| Our system identifies Quilis’ types, besides Spang’s expansion cases. |