Near-mergers and the suspension of phonemic contrast

Abstract

In 1972, Labov, Yaeger, and Steiner reported a series of "near-mergers" that have since proved to be difficult to assimilate to the standard conception of the phoneme and that challenged our current understanding of how language production is related to perception and learning (Labov, Yaeger, & Steiner, 1972). In these situations, speakers consistently reported that two classes of sounds were "the same," yet consistently differentiated them in production. Labov (1975a) suggested that this phenomenon was the explanation for two "falsely reported mergers" in the history of English, where word classes were said to have merged and afterward separated. It appears that sound change may bring two phonemes into such close approximation that semantic contrast between them is suspended for native speakers of the dialect, without necessarily leading to merger. This article reports on further observations of near-mergers, which confirm their implications for both synchronic and diachronic issues, and presents the results of experiments that show how phonemic contrast is suspended for an entire community.

Perception of suprasegmental structure in a non-native dialect

“…Users of one dialectal variety of a language can, accordingly, fail to distinguish contrasts made by speakers of another variety. Labov, Karen & Miller (1991), for instance, report that users of one dialect of American English fail to discriminate vowel contrasts made in another American English variety when asked to choose between alternative words, even though separate tests show that they can perform an accurate psycho-acoustic discrimination. Similarly, speakers of American English can use the form of an intervocalic consonant as a perceptual cue to the presence or absence of a syntactic phrase boundary between two words; the "nal sound of visit is more likely to be expressed as a #ap rather than as a [t] in &&Try to visit India in the spring'' than in &&For a visit, India is great''.…”

“…The vowel contrasts studied by Labov et al (1991) are e!ectively the hardest type, namely the single-category assimilation: the distinctions made in some American varieties assimilate to one vowel category in Philadelphia English, and thus such distinctions are particularly hard to make in speech categorization tasks. The stop consonant allophones studied by Scott & Cutler (1984) are instances of the Best et al category-goodness di!erences, since the distinction is between one American version which maps well onto British /t/ and another which does not; such di!erences should be learnable, according to Best et al,and Scott & Cutler indeed found 250 ¹.…”

The role of duration in the perception of vowel merger

“…For instance, when diagnosing near-merger, sociolinguists have often relied on minimal pairs tasks, which are based on participants' self-judgments regarding whether they believe a minimal pair differing only by the vowel contrast of interest should be pronounced the same or different (e.g., Di Paolo & Faber, 1990;Labov et al, 1991Labov et al, , 1972. The advantage of such tasks is that they get at speakers' conscious knowledge of vowel productions.…”

“…Perceptual abilities are generally determined with commutation tests, which typically involve auditory presentation of isolated words, often from participants' own speech, which they are then asked to identify, usually in a multiple forced-choice format. Studies that use this method to diagnose near-merger often report that speakers do not 'pass' commutation tests unless they achieve 100% accuracy rates; however, many of these 'failures' are still at above-chance (e.g., 80%) levels (e.g., Bowie, 2000;Labov et al, 1991Labov et al, , 1972. With 100% accuracy as the cut-off point for a perceptual distinction, it is unclear whether participants who 'fail' commutation tests are still picking up on some sort of perceptual cues-perhaps cues beyond vowel quality.…”

“…One problem with having various measures to choose from is empirical: Measures of production and perception-or even different measures of either production or perception-have been observed not to align (e.g., Herold, 1990;Labov et al, 1991Labov et al, , 1972Nycz, 2013). For instance, individuals may perceive vowel contrasts that they do not produce; such findings have often been attributed to merged speakers' contact with unmerged speakers (Hay et al, 2013;Thomas & Hay, 2005) or the influence of orthography (Faber & Di Paolo, 1995;Herold, 1990).…”