Читать книгу Stress Variation in English - Alexander Tokar - Страница 12

The importance of distinguishing between accented and non-accented positions is also exemplified by the (absence of the) phonetic contrast between primary- and secondary-stressed syllables (in English). According to PlagPlag et al. (2011: 362),

There is a large body of literature available on the acoustic correlates of stress in English (probably starting with FryFry, 1955, 1958), and there is a host of parameters that have been suggested to be acoustic correlates of stress. However, previous research has almost entirely focused on the question of how stressed and unstressed syllables differ from each other in terms of their acoustic properties, whereas the distinction between primary and secondary stress has hardly received any attention by phoneticians. (PlagPlag et al. 2011: 362)

To fill in the research gap, PlagPlag et al. (2011) have studied the phonetic properties of primary and secondary stress in both left- and right-prominent English words. Left-prominent words are words such as, for example, illustrate, in which primary stress precedes secondary stress, i.e., /ˈɪl·əˌstreɪt/ (Cambridge Dictionaries Online, henceforth CDO). In right-prominent words, such as, for example, illustration, the order of stresses is reversed: Primary stress is followed by secondary stress, i.e., /ˌɪl·əˈstreɪ·ʃən/ (CDO). The most interesting findings of Plag et al.’s (2011) study are as follows.

In accented positions, in which words bear phrasal stress, which in English is phonetically realized as pitchpitch accent, the differences between left-prominent words such as illustrate and right-prominent words such as illustration resemble the above mentioned difference between ˈworking theory and working ˈtheory (PlagPlag et al. 2011: 372). That is, in left-prominent words occurring in accented positions there is only one pitch accent, which falls on the primary-stressed syllable: Illustrate is pronounced ˈillustrate, with pitch accent being placed only on the primary-stressed initial syllable /ˈɪ/, but not on the secondary-stressed final syllable /ˌstreɪt/. (Note also that both the OD and LDOCE give the transcription /ˈɪləstreɪt/, where there is only the primary stress symbol (ˈ). A well-known fact is that -ate-words in English sometimes occur not only as verbs but also as adjectives. E.g., what distinguishes to subordinate from the adjective subordinate is that the former has a diphthong in the ult—/səˈbɔːdɪneɪt/ (OD)—whereas the latter is pronounced with a reduced vowelreduced vowel: /səˈbɔːdɪnət/ (OD). Given that both the transcription /səˈbɔːdɪneɪt/ and the transcription /səˈbɔːdɪnət/ contain only the primary stress symbol (ˈ), the present monograph argues that with regard to stress, to subordinate is not different from the adjective subordinate, i.e., both the ult /neɪt/ of the former and the ult /nət/ of the latter are unstressed (but because the ult /nət/ of the adjective subordinate contains a reduced vowel, it is of course less prominent than the ult /neɪt/ of to subordinate, which contains a diphthong). Similarly, a graduate and to graduate both have antepenultimate stress—/ˈɡradʒʊət/ and /ˈɡradʒʊeɪt/ (OD)—but while the former is pronounced with a reduced vowel in the ult, the latter pronunciation contains a diphthong.)

In right-prominent words, by contrast, there are two (equal) pitchpitch accents. That is, illustration is pronounced ˈilluˈstration, with pitch accent being placed both upon the secondary-stressed first syllable /ˌɪ/ and upon the primary-stressed third syllable /ˈstreɪ/. (In agreement with this finding, the OD gives the transcriptions /ᵻˈməʊʃn̩(ə)lʌɪˈzeɪʃn/ for emotionalization, /ˈdʒɒlɪfɪˈkeɪʃ(ə)n/ for jollification, /fəˈrɪŋɡəlʌɪˈzeɪʃn/ for pharyngalization, etc.; similarly, agreeability is /əˈɡriːəˈbɪlᵻti/, allowability is /əˈlaʊᵻˈbɪlᵻti/, materiality is /məˈtɪərɪˈalɪti/, etc., with more than one syllable in these transcriptions counting, according to the OD, as syllables bearing primary stress).

With regard to non-accented positions, in which words do not receive phrasal pitchpitch accent, the finding of PlagPlag et al.’s (2011) investigation is that the phonetic contrast between primary- and secondary-stressed syllables is almost completely neutralized. As the authors report:

[…] primary and secondary stress syllables are stressed syllables that are different from unstressed syllables, but not from each other, unless the word is accented. In this case the target of a nuclear accent corresponds to what is usually labeled the primary stress syllable. If the word occurs in an environment in which no accents are present (e.g. in post-nuclear positionpost-nuclear position), there is no phonological difference between the first and third syllable in words such as i.so.late and i.so.la.tion. These syllables are simply stressed (or strong) syllables (which, of course, differentiates them from the second syllable .so., which is unstressed, or weak). (PlagPlag et al. 2011: 373)

Accordingly, “if we disregard accentuation, there is no difference between secondary and primary stress” (PlagPlag et al. 2011: 373). (This conclusion poses a major empirical challenge to Chomsky & HalleChomsky & Halle’s (1968) highly counterintuitive claim that non-primary stresses in English are not only secondary, but also tertiary, quaternary, quinary, senary, etc., i.e., as Halle & VergnaudHalle & Vergnaud (1987: 37–38) explain, since especially the derivation of a phrase might, depending upon its syntactic complexity, involve a fairly large number of cycles, there is no maximum number of degrees of stress in English.)

The finding that (if accentuation is left out of consideration) primary stress is phonetically not different from secondary stress also has implications for the study presented in this monograph. Consider, for instance, the adjective meningococcic, for which the OED gives the American English transcription /məˈˌnɪŋɡoʊˈkɑk(s)ɪk/. This transcription stands for the variation between the pronunciation /məˈnɪŋɡoʊˈkɑk(s)ɪk/, where there are two primary stresses, and /məˌnɪŋɡoʊˈkɑk(s)ɪk/, where there is only one primary stress. Given, however, PlagPlag et al.’s (2011) findings, we can argue that these pronunciations are identical. Since in non-accented positions the contrast between primary stress and secondary stress is neutralized, the secondary-stressed syllable /ˌnɪŋ/ of the pronunciation /məˌnɪŋɡoʊˈkɑk(s)ɪk/ is not different from the primary-stressed syllable /ˈnɪŋ/ of the pronunciation /məˈnɪŋɡoʊˈkɑk(s)ɪk/. Likewise, also in accented positions the two pronunciations cannot be different from each other because in right-prominent words such as /məˌnɪŋɡoʊˈkɑk(s)ɪk/ both primary- and secondary-stressed syllables are pitchpitch-accented. That is, the acoustic form that the listener will perceive as meningoˈcoccic will actually be pronounced meˈningoˈcoccic by the speaker, with two pitch accents being assigned to both the secondary-stressed syllable /ˌnɪŋ/ and the primary-stressed syllable /ˈkɑk/. The only possible variation that meningococcic can exhibit in an accented position is that between the left-prominent pronunciation meˈningococcic and the right-prominent alternative meningoˈcoccic (which, as just said, is phonetically equivalent to the double-prominentdouble-prominent meˈningoˈcoccic), but this type of variation is not stipulated by the OED’s transcription /məˈˌnɪŋɡoʊˈkɑk(s)ɪk/, where the stressed syllable /ˈkɑk/ is said to be always pronounced with primary stress, i.e., it is only the stressed syllable /ˈˌnɪŋ/ where, according to the OED, the level of stress vacillates between primary and secondary. In summary, the adjective meningococcic, which, according to the OED, vacillates between the pronunciations /məˈnɪŋɡoʊˈkɑk(s)ɪk/ and /məˌnɪŋɡoʊˈkɑk(s)ɪk/ (as far as the American variety is concerned), cannot exhibit this variation because the pronunciation /məˈnɪŋɡoʊˈkɑk(s)ɪk/, which involves two primary stresses, is in no way different from the pronunciation /məˌnɪŋɡoʊˈkɑk(s)ɪk/, where the primary-stressed syllable /ˈkɑk/ is preceded by the secondary-stressed syllable /ˌnɪŋ/.