A phrasal explanation of Spanish secondary stress

Recent theories of metrical structure differ starkly in whether feet can be aligned gradiently Particularly challenging for categorical theories are apparent cases of the initial-dactyl effect, where all feet align rightward except for the leftmost, combined with End Rule Right (1-3). In long words, medial feet have to be forced rightward, but they are not adjacent to an edge and categorical alignment cannot determine their placement. If the leftmost foot is the main stress (End Rule Left), then this fact can be used to force a lapse adjacent to the initial main foot, generating the effect without gradient directionality (Kager 2001). But certain languages show an initial-dactyl effect with a final main foot, in which case the positioning of the lapse cannot be controlled in this way. In Spanish, the medial secondary stress in (gràma)ti(càli)(dád) in (1) can be described by a traditional cycle or by output-output faithfulness (Benua 1997), based on (gràma)ti(cál) 'grammatical'. But Hyde & McCord (2012) show that forms like (màte)ma(tìci)(dád) in (2) cannot be handled by faithfulness to stress on a previous cycle (or to a morphologically contained surface form); for example, (màte)(máti)co 'mathematical' wrongly predicts *(màte)(màti)ci(dád), and loanwords have no internal cycle. Hyde & McCord claim that these data show the need for gradient directional alignment, but I show in this paper that a more complete account of Spanish stress makes such gradience unnecessary. More detailed descriptions of Spanish secondary stress (Harris 1983, Roca 1986) report two variants: one, more colloquial, that follows the initial-dactyl pattern (3), and another, more formal, in which feet are fully aligned to the right like gra(màti)(càli)(dád) in (4). Harris reports this as one of several " firmly established " generalizations regarding Spanish stress. (Hyde & McCord mention the second pattern, but do not analyze it.) In addition, secondary stress interacts with phrasal context in a way that requires a further dimension to the analysis. In (5) the initial dactyl is found across a phrase (Navarro Tomás 1977); and in (6), word-internal footing will pattern like (4) if a syllable is added to the left within the phrase (Roca 1986). A fuller analysis taking these facts into account is more empirically adequate, and also eliminates the argument for gradient directionality in (2). The essential insight is that the formal pattern in (4), with right-alignment of all feet, characterizes the lexical derivation. At the phrasal level, a new trochaic foot is …