How much homophony is normal?

Abstract

This paper argues that neutralizing phonological alternations are sensitive to how much homophony they create among distinct lexical items: neutralizing rules create fewer homophones than expected. Building on a case study of Korean by Silverman (2010), I compare the neutralizing rules of Korean to a large number of hypothetical alternatives generated by Monte Carlo simulations. The simulations reveal that the actual rules of Korean frequently create far fewer homophones than similar (but unattested) rules, even when the rules that are compared are controlled for the number of phonemic contrasts they eliminate. These results suggest that phonological patterns are sensitive not only to high-level contrasts among phonemes but also to contrasts among individual lexical items. The effect is most pronounced when homophones are not weighted by frequency, a result that adds to evidence in the literature that the relevant measure of lexical frequency for many lexicon-sensitive phonological patterns is type frequency, not token frequency.

Aspects of Japanese loanword devoicing

“…Ichimura (2006) argues that (Japanese) speakers avoid such mergers of two existing lexical items. More generally, such anti-merger constraints have been proposed by many researchers for various languages (Blevins, 2005;Crosswhite, 1999;Itô and Mester, 2004;Kaplan, 2010;Lubowicz, 2007;Padgett, 2002;Urbanczyk, 2005), especially within the context of Optimality-Theoretic (Prince andSmolensky, 1993/2004) Dispersion Theory (Flemming, 1995). Kawahara (in press) found a non-significant tendency toward the expected direction, and a follow-up study with more relevant lexical items was claimed to be necessary.…”

Homophony and Contrast Neutralization in Southern Min Tone Sandhi Circle

“…By preserving phonological contrasts that are often used to distinguish minimal pairs, the attested neutralization patterns create less homophony than the unattested ones, and therefore avoid creating confusion in the listeners. Two recent studies by Silverman (2010) and Kaplan (2011) focused on the consonant neutralization patterns in Korean, and suggested that the attested neutralization patterns create less homophony than other possible-yet-unattested patterns. For example, the coda neutralization eliminates the distinction between aspirated and plain stops.…”

“…number of homophonous word tokens, and 3.) information content, in terms of Shannon's entropy in bits (Kaplan 2011). The calculation of homophony is exemplified in the toy example (15) Phonological neutralization decreases the number of word types, while increasing the number of homophonous word tokens.…”

“…Using Korean corpora, Silverman (2010) and Kaplan (2011) simulated the attested consonant neutralization pattern (such as (13)), as well as other possible-yet-unattested patterns (such as (14)). They compared the amount of homophony before and after the attested and unattested patterns, and found that the attested patterns produced less homophony than the unattested ones.…”

“…It was more optimal to preserve certain tonal contrasts over others, and thus chain shifts were formed to avoid neutralization of particular tonal contrasts. If we assume that the reason for certain phonological contrast to be preserved over others is to avoid homophony (Silverman 2010;Kaplan 2011;Wedel et al 2013b), and that the Southern Min tone sandhi circle is motivated by contrast preservation, we should expect that the attested tone sandhi circle generates less homophony than other possible-yet-unattested patterns. The next section describes three simulations, in which we compared the amount of homophony generated by attested and unattested tone sandhi patterns.…”