In lieu of an abstract, here is a brief excerpt of the content:

A GRADUAL APPROACH TO SIBILANT + /J/ METATHESIS IN HISPANO-ROMANCE Kenneth Wireback Miami University 1. Introduction During the transition from the Latin of the Iberian Peninsula, at the time of the Roman Empire, to the dawn of Old Spanish and Old Portuguese, several consonant + palatal-glide sequences underwent metathesis to glide + consonant. In Spanish, this glide metathesis occurred regularly in the Latin /sj/ and /rj/ sequences, and additionally with Latin /s:j/ and in certain stem allomorphs of two verbs, caber and saber (indicative and subjunctive for caber, but only the subjunctive stem of saber). This is shown by the data in (1). (1) CASEU > *[kazju] > *[kajzo] > *[kejzo] > Sp. queso 'cheese' BASiu > *[bazju] > *[bajzo] > *[bejzo] > Sp. beso 'kiss' *quassiare > *[kasjare] > *[kajsar] > [kejjar] > Sp. quejar 'complain' area > *[arja] > *[ajra] > *[ejra] > Sp. era 'threshing floor' SAPiAT > *[sapja] > *[sajpa] > *[sejpa] > Sp. sepa 'that s/he know' CAPiO > *[kapjo] > *[kajpo] > *[kejpo] > Sp. quepo ? fit' The Portuguese data are similar to the Spanish, in the sense that the same sequences, /sj/, /s:j/, /rj/, and /pj/, also underwent metathesis in Portuguese, but there are nonetheless a few important differences. First, regarding the Latin /sj/ sequence, the sibilant consonant palatalized to /3/ in Portuguese, while apparently remaining alveolar in Spanish . Second, in the vast majority of cases, the off-glide subsequent to metathesis remained in Portuguese, but was eliminated in Spanish. La corónica 30.1 (Fall, 2001): 159-203 160Kenneth WirebackLa coránica 30.1, 2001 Third, in Portuguese, metathesis of labials and semivowel was regular, in contrast to its restriction to caber and saber in Spanish. These Portuguese data are presented below in (2). (2) CASEU > *[kazju] > *[kaJ3o] > *[keJ3o] > Ptg. queijo 'cheese' BASiu > *[bazju] > *[baJ3o] > *[beJ3o] > Ptg. beijo 'kiss' *quassiare > *[kasjare] > *[kajsar] > [kejjar] > Ptg. queixar 'complain' area > *[arja] > *[ajra] > *[ejra] > Ptg. eira 'threshing floor' SAPiAT > *[sabja] > *[sajba] > Ptg. saiba 'that s/he know' APiU > *[apju] > Ptg. aipo 'celery' rabia > *[rabja] > *[rajva] > Ptg. raiva 'anger, rage' RUBEU > *[rubju] > *[rujvo] > Ptg. ruivo 'red-haired' Noviu > Ptg. noivo 'fiancé' Blevins and Garrett (1998: 510-17) account for cases of consonant + vowel metathesis similar to those exemplified above in (1) and (2) with the concept of perceptual metathesis. From their perspective, consonant-vowel and consonant-glide metatheses originate when the consonant and the (semi)vowel contain an acoustic or perceptual feature with a rather long duration. For the Spanish and Portuguese cases of consonant + glide metathesis cited above, the perceptual/acoustic feature in question was palatalization. The cornerstone of perceptual metathesis is the implication that there must be an intermediate stage in which the consonant shifts to a palatalized or palatal point of articulation : "We view [high-vowel metathesis with adjacent consonants] as the resegmentation of elongated palatalization and labialization phases, an extreme version ofthe common type ofchange where highvowel loss results in distinctive secondary articulations on adjacent consonants" (Blevins and Garrett 1998: 514). Therefore, consonant + (semi)vowel metathesis involves at least two stages of misperception due to the relatively elongated perceptual/acoustic cue of the palatalization gesture. First, /Cj/ would be misperceived as a palatal (or palatalized ) consonant; then, subsequently, this palatalized consonant would in a similar fashion be misperceived as /jC/. Torreblanca (1988, 1992) rejects this gradual approach, especially the likelihood of depalatalization of the consonant in Spanish subsequent to the palatalized intermediate stage. In his view, had HispanoRomance /?/ become palatalized, evidence such as ec(c)lesia > OSp. egrija 'church' suggests that it would have remained so, in Spanish as well as Portuguese (1992: 309). Similarly, if /r/ had become partially A GradualApproach to Sibilant + IjI Metathesis161 palatalized preceding /j/, then /rj/ ought to have produced /j/ in Spanish as it has in Italian, e.g., area > Ital. aia 'threshing floor', as opposed to Hispano-Romance area > *[a.rja] > *[aj.ra] > *[ej.ra] > Ptg. eira, Sp. era (1992: 306-308). In response to these perceived shortcomings of the gradual approach, Torreblanca postulates an abrupt metathesis of consonant and yod. In this article, in support of the gradual approach, I undertake a detailed examination of both the articulatory and acoustic aspects of diachronic sibilant + palatal-glide metathesis. From this basis, I then explain why the Spanish and Catalan data are at odds with the corresponding Portuguese data...


Additional Information

Print ISSN
pp. 159-203
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.