Academic journal article Journal of Slavic Linguistics

Palatalized Labials in Polish Dialects: An Evolutionary Perspective

Academic journal article Journal of Slavic Linguistics

Palatalized Labials in Polish Dialects: An Evolutionary Perspective

Article excerpt

Abstract: Two types of explanations for typological asymmetries are in current use: synchronic, which rely on phonological filters that make learners more receptive to some patterns than others (e.g., markedness), and diachronic, which appeal to phonetically systematic errors that arise in the transmission of the speech signal. This paper provides a diachronic account of palatalized labials in standard and dialectal Polish. It is shown that the weak perceptibility of the palatal element in a specific phonetic context is a good predictor of depalatalization and that dissimilation arises whenever a phonetic signal can be interpreted in a non-unique manner. The Polish data exemplify three sources of natural sound change: (i) neutralization of perceptually weak contrasts, (ii) phonological reanalysis of ambiguous signals, and (iii) change in the frequency of phonetic variants. Sound change is shown to be non-deterministic and non-optimizing. There is no role for markedness in this account.

1. Introduction

Two questions about which phonologists disagree are whether the explanation for typological patterns is synchronic or diachronic and whether linguistic systems are goal-oriented or not. The line of research represented by Chomsky and Halle (1968), Archangeli and Pulleybank (1994), Flemming (1995), Steriade (2001), and Hayes and Steriade (2004) assumes that typology finds an explanation in synchronic biases. These are either innate and make up Universal Grammar (Chomsky and Halle 1968) or emerge primarily from the phonetic input the learner is exposed to (Hayes and Steriade 2004). Adopting the framework of Optimality Theory (Prince and Smolensky 2004), Hayes and Steriade (2004) propose that phonetic knowledge informs the ranked constraints that make up the phonological component. In other words, synchronic grammars, which comprise hierarchically ranked constraints, are induced from phonetic substance. The emergent rankings of these phonetically grounded constraints account for attested patterns, while other patterns are predicted not to exist because the constraints or the rankings that could generate them cannot be induced from phonetic input.

Synchronic accounts typically invoke markedness to explain typological asymmetries (Hayes and Steriade 2004). It is argued that crosslinguistic high frequency of certain patterns correlates with their unmarked status and the rarity of others with their marked status. In Optimality Theory (OT) markedness laws assume the form of violable markedness constraints which penalize particular structures in surface forms. Faithfulness constraints provide a counterbalance by favoring similarity between input and output forms. Markedness constraints ensure that grammars are inherently optimizing (goal-oriented).

Representing a different approach, Evolutionary Phonology (EP), Blevins (2004), following Ohala (1981), proposes that explanations for recurrent sound patterns in the world's languages are historical and not goal-oriented. Natural sound change, which gives rise to linguistic patterns, is phonetically based and stems from systematic errors that occur during language transmission between the speaker and the listener. Ohala's (1981) model relies on "innocent misapprehensions", because the basic mechanism of innovation involves mishearing a structure and assigning it an interpretation that differs from that assigned by the previous generation. Blevins relieves synchronic grammars of the task of providing explanations and argues that they are primarily descriptive. It follows that in EP finding motivation in phonetics is central to diachronic accounts. There is no role for teleology or markedness in this model.

Proponents of EP raise two arguments against synchronic generative models such as OT. One comes from parsimony and the other from typology. Insofar as OT constraints are derived from phonetic input, parsimony dictates against the need for a phonological component that copies phonetic knowledge only to translate it into constraints. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.