Academic journal article Bulletin of Hispanic Studies

The Rise and Fall of Proclisis in Old Spanish Postprepositional Infinitival Clauses: A Quantitative Approach

Academic journal article Bulletin of Hispanic Studies

The Rise and Fall of Proclisis in Old Spanish Postprepositional Infinitival Clauses: A Quantitative Approach

Article excerpt

The placement of weak object pronouns, or 'clitic linearization', in relation to the verb is an important variable in the history of Spanish. Initially, Old Spanish main clauses can be grouped with infinitival clauses, in that both proclisis (immediate preverbal placement) and enclisis (immediate postverbal placement) may be found in the two contexts.1 This is illustrated in the medieval examples (1)to (4) below.


(1) Et por ende le dixieron este sobre nombre don Sancho el gordo.

[And so they gave him this sobriquet don Sancho the Fat.]

(Estoria de España II; Escorial X-I-4, copied 1284-1347)

(2) quisol toller ell Jmperio por lo dar a Eucherio su fijo

[He tried to take the empire from him to give it to Eucherius his son]

(Cánones de Albateni; Paris: Arsenal 8322, copied 1290-1300)


(3) Et dieron les a partir en vna heredat.

[And they allowed them to share in an inheritance.]

(Libro de los fueros de Castilla; BNE MSS/431, copied 1301-1400)

(4) o usen en dar les las moscas e las abeias

[or the practice should be followed of giving them flies and bees]

(Libro de las animalias que cazan; BNE RES/270, copied 1250)

In contrast, finite subordinate clauses have near obligatory proclisis from the earliest documented period. As Bouzouita (2009: 6) observes, 'preverbal placement is found almost without exception in the presence of a preceding complementizer, relative pronoun, or subordinating conjunction' (see also Fontana 1993; Castillo Lluch 1996; Granberg 1988; and Nieuwenhuijsen 2006). An illustration is given in (5) below.

(5) Et enuiaron le pedir que les diesse cada anno.L. donzellas de las mas fijas dalgo

[And envoys were sent to request that he give them each year 50 young women of the highest birth]

(Estoria de España II; Escorial X-I-4, copied 1284-1347)

Modern Spanish has categorical proclisis in all finite clauses, other than positive imperatives, and categorical enclisis in non-finite clauses. This situation results in part from a progressive increase in proclisis in finite main clauses (for details of which, see the authors listed above). According to Bouzouita (2008, sections 5 and 6), this increase initially affected imperatives as well, but it was subsequently reversed in the case of positive imperatives. Somewhat analogously, proclisis in an important subset of infinitival clauses, viz. those that are introduced by a preposition,2 increased markedly in the late medieval period before declining and then disappearing. It is this up-and-down event, or 'failed change' (Postma 2010), which constitutes the focus of the present paper.

Davies (1997), Castillo Lluch (2001), Nieuwenhuijsen (2006) and Wanner (2006) have adduced some relatively coarse-grained quantitative data in relation to the event in question. However, the data points in those surveys are too widely spaced (usually at intervals of a century or more) to permit detailed analysis. The present paper reports findings based on a very large corpus (over 17 million words, drawn from 194 prose texts), with data points located at decadal intervals. The volume of tokens collected (nearly 25,000 for the affirmative context alone), together with their arrangement by decade, enables a fine-grained quantitative picture to be constructed, on the basis of which a more far-reaching analysis can be made of this event than has previously been possible.

Among other things, the findings presented here constitute a robust test of the theory of failed changes advanced by Postma (2010). According to that author, a failed change can be modelled mathematically as a rising logistic curve S multiplied by the falling curve 1 - S, the product being the symmetric Hubbert curve. The present paper gives reliable estimates of the slope parameters for the rise and for the decline of proclisis in the context examined. These estimates suggest that the factors involved in a failed change need not interact in the manner proposed by Postma. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.