The Phonetics and Phonology of Contrast

Margaret E. L. Renwick

• Verlag: De Gruyter Mouton
• Jahr: 2014
• ISBN:
  9783110394764
• Auflage: 1
• Thema: Sprachen & Linguistik
• Thema: Sprachwissenschaft

This book proposes that phonological contrast, in particular the robustness of a phonemic contrast, does not depend solely on the presence of minimal pairs, but is instead affected by a set of phonetic, usage-based, and systemic factors. This perspective opens phonology to a more direct interpretation through phonetic analysis, undertaken in a series of case studies on the Romanian vowel system. Both the synchronic phonetics and morpho-phonological alternations are studied, to understand the forces that have historically shaped and now maintain the phonemic system of Romanian. A corpus study of phoneme type frequency in Romanian reveals marginal contrasts among vowels, in which a sharp distinction between allophones and phonemes fails to capture relationships among sounds. An investigation of Romanian /?/ provides insight into the historical roots of marginal contrast, and a large acoustic study of Romanian vowels and diphthongs is a backdrop for evaluating the phonetic and perceptual realization of marginal contrast. The results provide impetus for a model in which phonology, phonetics, morphology and perception interact in a multidimensional way.

1 Introduction

1.1 Goals of this book

This book investigates the system of phonemic contrasts among Romanian vowels, centering on examples of marginal contrast, in which a traditional sharp distinction between allophones and phonemes is insufficient to capture the relationships among sounds. These relationships are made clear in a series of case studies from historical, phonological and phonetic perspectives. In several experiments speech production and perception are explored in Romanian, a language that presents a prime opportunity for empirically examining the interaction of phonology and morphology, and is under-studied with respect to other Romance languages.

These studies are motivated by the question of how a language’s phonological inventory shapes the acoustic realization of its members, and how the language’s phonological processes interact with the synchronic phonetics to affect production and perception. The book focuses on three vowels that are innovations to Romanian among Romance languages, and investigates their role in the vowel inventory first with a synchronic study of phoneme type frequency. These are the high central vowel /ɨ/, and the diphthongs

and

. The initial study reveals that type frequency is unevenly distributed across the inventory, and in particular, the roles of these innovative vowels are very small. The remaining chapters examine these vowels’ characteristics from several points of view. It will be shown that the historical dimension is crucial: many facts about the relative frequencies of Romanian phonemes, and their synchronic distribution, can be traced to historical processes and the influences of borrowings throughout the language’s history. The distributional facts of the Romanian vowel system motivate not only text-based corpus studies, but also phonetic studies that survey the acoustics of the entire inventory, and initial investigations of the perception of vowels in Romanian.

Together, these studies demonstrate that Romanian has a set of marginally contrastive vowels, whose characteristics motivate a series of questions: How did these vowels enter the language, and what factors determined their distribution? What are the consequences of marginal contrastiveness for these sounds’ phonetics and perception? Where phonological alternations are active, do we find parallels in the phonetics, and is that variability systematically related to the size and makeup of a language’s vowel inventory? The results of these studies have implications for our expectations of how phonological contrast is realized.

This book addresses questions that highlight the interface of phonetics and phonology, focusing in particular on factors that trigger change over time, and have the potential to reshape a phonemic inventory. The data show a variety of forces at play in the Romanian vowel system. This chapter provides background on the topics investigated in the rest of the book. First, theoretical motivation is provided for analyses of marginally contrastive phonemic relationships, in section 1.2. This is followed in section 1.3 by an overview of previous linguistic treatises on Romanian, and in 1.4 a brief history of Romania and its linguistic situation, including general observations on systematic changes to the Latin vowel inventory that resulted in the Romanian system. The phonemic inventory of Romanian is described in section 1.5, and in section 1.6 I lay out further typological motivations for studying the language’s system of contrasts, based on the relative frequencies of its vowels across the world’s languages. Finally, section 1.7 outlines the remaining eight chapters.

1.2 Phonological contrast: Categorical or gradient?

Traditionally, a sound is considered to be either phonemic, in which case it can be underlyingly specified in a language’s lexicon to distinguish one word from another; or a sound is in an allophonic relationship with another sound or sounds when its appearance is predictable based on phonological context or other linguistic factors. If two sounds are in an allophonic relationship, the presence of one vs. the other never triggers a change in lexical meaning; in other words, they are not responsible for lexical contrast. This is the case for English [l] and [ɫ], both of which native speakers associate with the phoneme /l/; but the former appears in syllable-initial position, as in lock [lɑk], while the latter always falls in syllable-final position, in e.g. call [kɑɫ]. Since a sound cannot (typically) be simultaneously syllable-initial and syllable-final, these sounds have separate distributions, and cannot form a lexical contrast. In a derivational approach to phonology, words with either [l] or [ɫ] on the surface contain /l/ in their underlying representations, and the appropriate allophone is selected for the phonetic form. Two sounds are considered contrastive when they can appear in the same environment, and the appearance of one vs. the other does give rise to a change in lexical meaning; this results in minimal pairs, such as English heed [hid] vs. had [hæd], in which the substitution of a single sound correlates with different meanings. In many cases, this contrastiveness is reinforced by an abundance of minimal pairs: in English, the contrast between /i/ and /æ/ can be attested also by bead [bid] vs. bad [bæd]; beat [bit] vs. bat [bæt]; read [rid] vs. rad [ræd], seat [sit] vs. sat [sæt], and many others.

What happens, however, when very few minimal pairs separate two phonemes? This is the case in Romanian, whose vowels /ɨ/ and /ʌ/ contrast in only six clear minimal pairs (see Chapter 5). Additionally we find that where /ɨ/ and /ʌ/ are concerned, it is easy to predict where one phoneme is preferred over the other: /ɨ/ tends to occur in stressed syllables and especially before /n/, while /ʌ/ is typically unstressed. This means that their distributions, while overlapping, are largely separate. With these two pieces of evidence, we see that the contrast between /ɨ/ and /ʌ/ is not robust. I refer to such a contrast as marginal.

The idea that phonemes can fall into a relationship of marginal contrastiveness is not a new one; simply put, this captures cases of contrast which are phonemic according to a narrow definition (for example, the presence of at least one minimal pair), but in which the members of the pair fail to freely occur in all the same contexts, and thus are not robustly contrastive. This can be due, for example, to the presence of phonological restrictions or large gaps in the occurrence of each phoneme. When this lack of contrastiveness results in few minimal pairs, as in the case of /ɨ/ vs. /ʌ/, the contrast is described as having a low functional load (Hockett 1966). Phonemic contrasts have been described as falling along a continuum, since sounds may exhibit varying degrees of contextual neutralization. One way to characterize marginally contrastive relationships is in subcategories of phonemic contrast; this is eloquently done by Goldsmith (1995) in his introduction to the Handbook of Phonological Theory, using the following descriptions:

• (1) Contrastive segments: “[I]n every phonetic context in which x may be found, y may also be found, but in a fashion that produces a word that is grammatically or lexically distinct […].” (Goldsmith 1995:10)
• (2) Modest asymmetry case: A pair of sounds, x and y, “are uncontroversially distinct, contrastive elements in the underlying inventory, but […] in at least one context there seems to be a striking asymmetry in the distribution of segments, judging by the relative number of words with the one and words with the other, or by some other criterion.” (Goldsmith 1995:11)
• (3) Not-yet-integrated semi-contrasts: “[A] language has two sounds, x and y, which may well be in contrast in some environments, but which in a particular environment show a sharp asymmetry, in that x appears in large numbers, while y appears in small numbers in words that are recent and transparent borrowings.” (Goldsmith 1995:11)
• (4) Just barely contrastive sounds: “x and y are phonetically similar, and in complementary distribution over a wide range of the language, but there is a phonological context in which the two sounds are distinct and may express a contrast. […] Thus while a contrast exists, a stubborn one which will not disappear under scrutiny, the contrast occurs in an extremely small part of the range of contexts in which the sound is found. The contrast is a lexical one, but only just barely.” (Goldsmith 1995:11)
• (5) Allophones in complementary distribution or free variation: “In the former case, there is no phonetic environment in which both x and y appear, while in the latter x and y may occur freely with no lexical or grammatical difference involved.” (Goldsmith 1995:10)

Goldsmith’s view of phonological contrast is clearly more fully articulated than a binary division between allophones and phonemes; however, we must consider the possibility that contrast is even more gradient than he describes, or that some contrasts result from combinations of factors not captured by Goldsmith’s categories. This is the case in Romanian, where the nature of the contrast between /ɨ/ and /ʌ/ depends on the portion of the lexicon we examine: among native words, Goldsmith’s category of “just barely contrastive sounds” is appropriate; but when we take into account the influence of borrowings to Romanian, the relationship between /ɨ/ and /ʌ/ seems to be a “not-yet-integrated semi-contrast.” When the acoustics are taken into account, however, we see that the two sounds are distinct, and they seem to be best classified as a “modest...