Símbolo y signo: vínculo y diferencia con el mito

In the remainder of this chapter, I shall offer an alternative analysis in which output representations contain only privative primes, are redundancy-free, and are fully interpretable. For this purpose, as already discussed in §3.2, I adopt the version of element theory assumed in, for example, Harris (1990, 1994) and Harris & Lindsey (1995).

In this theory, the melodic prime most similar to the privative feature [voice] is the element pertaining to slack vocal cords, labelled [L]. This element is present in fully voiced obstruents. The acoustic characteristic of [L] is lowered fundamental frequency. The other element relevant to the present discussion is the nasality element [N], which is similar to privative [nasal] employed in distinctive feature models. This element manifests itself in the signal as a broad resonant peak at the lower end of the frequency range. ^

As discussed in §4.1, nasals and [L] show a strong correlation, as demonstrated by the phenomenon of postnasal assimilation found in many languages (Herbert 1986). In other words, while it is [N] that determines the nasality of a segment, it can be said that the elements [N] and [L] are clearly related. This is supported by cross-linguistic observation: voicing assimilation in an NC-cluster is relatively unmarked (Lass 1984, Itô, et ai. 1995).

In order to capture the [N]-[L] relation, 1 assume that these elements are in fact two instantiations of the same object, [N] (cf. Harris & Lindsey 1995). This object manifests its properties without recourse to redundancy rules or markedness conventions

and without reference to the major-class feature [sonorant]. The idea of the [N]-[L] merger originates in the element-oriented speech perception programme based in London in early 1990’s. The first formal evidence that [N] and [L] are two instantiations of the same object is provided in Nasukawa (1995a, 1997), which present an integrated approach to the paradoxical behaviour of nasal and voice (postnasal voicing assimilation, Lyman’s Law and verb-stem-final b-n alternation in Japanese).

Ploch (1999) provides further arguments to support the [N]-[L] merger. First, in languages displaying nasal harmony, nasal and voiced stops exhibit a correlation: they alternate with each other, depending on the nasality/orality of the adjacent vowel. Second, unlike the other elements, both [N] and [L] (interpreted as voicing in non­ nuclear positions and low tone in nuclear positions) are incompatible with [H] (interpreted as voiceless aspirated and high tone in non-nuclear and nuclear positions respectively): voiced stops block high tone spreading and voiceless obstruents block nasal spreading. Third, two dissimilation laws, Dahl’s Law and Meinhof s Law, are captured by a unified account if we assume the merger of the two elements.

Now let us consider how we arrive at the dual manifestation of this object, as nasality on the one hand, and as voicing on the other. The ET literature puts forward two different representational approaches to the issue of a single element having a dual phonetic interpretation: one is the notion of melodic headedness; the other is the notion of complement tier. The former is the more widely accepted means of expressing melodic dependency relations in the literature. This dates back to the earliest models of Dependency Phonology (DP: Anderson & Jones 1974,1977). In the standard version of the theory (Harris & Lindsey 1995), for example, the contrast between ATR and non- ATR is, as already discussed in §3.3, captured via headship distinctions.^ It is now widely held that, for example, ATR u and non-ATR u are distinguished phonologically by the headship status of the prime [U]. If [U] is headed, it is interpreted as tense w; on the other

^This structural representation was originally introduced to eliminate the independent prime contributing ATR-ness which belonged to the element inventory in the original version of ET (Kaye, Lowenstamm & Vergnaud 1985). The ATR element was ultimately rejected because of its questionable phonetic interpretability and its less active participation in phonological processes compared with other elements.

Chapter 4. An integrated approach to nasality and long-lead voicing 91 hand, if [U] is headless, it is interpreted as its lax counterpart. In this treatment, the headedness of a given element is viewed as an intrinsic property which enhances the acoustic image of the element (Harris 1994, Harris & Lindsey 1995).

(12)

Element Phonetic manifestation

[U] (non-headed [U]) non-ATR u [U] (headed [U]) ATR u

On the other hand, the notion of complement tier has been introduced in the geometry-based element theory developed in Backley (1998) and Backley & Takahashi (1998), which has arisen in recourse to problematic aspects of melodic headedness (which will be discussed in the next subsection). Underlying this approach is the assumption that a full set of resonance elements is present within each position, permitting all the melodic contrasts of a language to be expressed using the same structural configuration. According to this view, melodic oppositions are stated not in terms of the presence or absence of particular elements, but via the active or inactive status of elements latently present in the structure. Under this approach, two distinct phonetic instantiations of a single object are expressed, not by the notion of headedness, but by the active or inactive state of a complement tier, which may be licensed by a single element within a given expression and has the effect of enhancing the acoustic image of the overall expression. In the context of vowels, the complement tier contributes ATRness to an expression in which it is active. For example, ATR u and non- ATR u are distinguished by the active status of the complement tier licensed by [U]. This is represented in (13), where [comp] denotes the complement tier.

(13) Element

u;

u

As depicted in (13), if [U] licenses its [comp] to be active, an expression is interpreted as tense «, on the other hand, if the [comp] is inactive, it is interpreted as its lax counterpart.

Both representational approaches share some common ground, in that the structural distinctions are in both cases manipulated via parametric setting: the majority of languages show no ATR contrast, and so exclude the structural distinction based on headedness or the complement tier (Off setting); only languages exhibiting ATR-ness contrasts display the necessary distinctions (On setting: e.g. English, French,

Wolof, Akan).

In this work, I adopt the representation in (13), since the alternative approach utilising melodic headship presents certain problems in relation to phonologically dynamic processes, as I explain below.