This thesis deals with the nature and definition of phonological similarity and shows that, when similarity plays a role in the motivation of phonological processes, it is evaluated over abstract, phonological features and not purely phonetic properties. Empirical evidence for this position is drawn from the domain of consonant harmony. Typological studies of consonant harmony (Hansson 2001, Rose and Walker 2004) have argued that segments which interact in consonant harmony processes must be highly similar to one another. This thesis provides analyses of a range of consonant harmony processes and demonstrates that, in each case, the notion of similarity needed in order to determine participating segments is evaluated over contrastive feature specifications. Contrastive specifications are established according to language-specific feature hierarchies (Jackobson and Halle 1956, Dresher 2003, forthcoming) with some features taking scope over others. Languages analyzed in some detail include Bumo Izon, Kalabari Ijo, Hausa, Dholuo, Anywa, Tzutujil and Aymara. Two definitions of similarity are proposed in order to account for two sets of cases. In one set of consonant harmony processes, interacting segments are similar in the sense that they constitute the natural class of segments contrastively specified in the harmonic feature. In another set of cases, participating segments must be similar according to the following definition; they must differ in only a single marked and contrastive feature specification.

This thesis deals with the nature and definition of phonological similarity and shows that, when similarity plays a role in the motivation of phonological processes, it is evaluated over abstract, phonological features and not purely phonetic properties. Empirical evidence for this position is drawn from the domain of consonant harmony. Typological studies of consonant harmony (Hansson 2001, Rose and Walker 2004) have argued that segments which interact in consonant harmony processes must be highly similar to one another. This thesis provides analyses of a range of consonant harmony processes and demonstrates that, in each case, the notion of similarity needed in order to determine participating segments is evaluated over contrastive feature specifications. Contrastive specifications are established according to language-specific feature hierarchies (Jackobson and Halle 1956, Dresher 2003, forthcoming) with some features taking scope over others. Languages analyzed in some detail include Bumo Izon, Kalabari Ijo, Hausa, Dholuo, Anywa, Tzutujil and Aymara. Two definitions of similarity are proposed in order to account for two sets of cases. In one set of consonant harmony processes, interacting segments are similar in the sense that they constitute the natural class of segments contrastively specified in the harmonic feature. In another set of cases, participating segments must be similar according to the following definition; they must differ in only a single marked and contrastive feature specification.

This dissertation explores the nature and extent of retroflex consonant harmony in South Asia. Using statistics calculated over lexical databases from a broad sample of languages, the study demonstrates that retroflex consonant harmony is an areal trait affecting most languages in the northern half of the South Asian subcontinent, including languages from at least three of the four major families in the region: Dravidian, Indo-Aryan and Munda (but not Tibeto-Burman). Dravidian and Indo-Aryan languages in the southern half of the subcontinent do not exhibit retroflex consonant harmony. In South Asia, retroflex consonant harmony is manifested primarily as a static co-occurrence restriction on coronal consonants in roots/words. Historical-comparative evidence reveals that this pattern is the result of retroflex assimilation that is non-local, regressive and conditioned by the similarity of interacting segments. These typological properties stand in contrast to those of other retroflex assimilation patterns, which are local, primarily progressive, and not conditioned by similarity. This is argued to support the hypothesis that local feature spreading and long-distance feature agreement constitute two independent mechanisms of assimilation, each with its own set of typological properties, and that retroflex consonant harmony is the product of agreement, not spreading. Building on this hypothesis, the study offers a formal account of retroflex consonant harmony within the Agreement by Correspondence (ABC) model of Rose & Walker (2004) and Hansson (2001; 2010). Two Indo-Aryan languages, Kalasha and Indus Kohistani, figure prominently throughout the dissertation. These languages exhibit similarity effects that have not been clearly observed in other retroflex consonant harmony systems; retroflexion is contrastive in both non-sibilant (i.e., plosive) and sibilant obstruents (i.e., affricates and fricatives), but harmony applies only within each manner class, not between them. At the same time, harmony is not sensitive to laryngeal features. Theoretical implications of these and other similarity effects are discussed.

This dissertation explores the nature and extent of retroflex consonant harmony in South Asia. Using statistics calculated over lexical databases from a broad sample of languages, the study demonstrates that retroflex consonant harmony is an areal trait affecting most languages in the northern half of the South Asian subcontinent, including languages from at least three of the four major families in the region: Dravidian, Indo-Aryan and Munda (but not Tibeto-Burman). Dravidian and Indo-Aryan languages in the southern half of the subcontinent do not exhibit retroflex consonant harmony. In South Asia, retroflex consonant harmony is manifested primarily as a static co-occurrence restriction on coronal consonants in roots/words. Historical-comparative evidence reveals that this pattern is the result of retroflex assimilation that is non-local, regressive and conditioned by the similarity of interacting segments. These typological properties stand in contrast to those of other retroflex assimilation patterns, which are local, primarily progressive, and not conditioned by similarity. This is argued to support the hypothesis that local feature spreading and long-distance feature agreement constitute two independent mechanisms of assimilation, each with its own set of typological properties, and that retroflex consonant harmony is the product of agreement, not spreading. Building on this hypothesis, the study offers a formal account of retroflex consonant harmony within the Agreement by Correspondence (ABC) model of Rose & Walker (2004) and Hansson (2001; 2010). Two Indo-Aryan languages, Kalasha and Indus Kohistani, figure prominently throughout the dissertation. These languages exhibit similarity effects that have not been clearly observed in other retroflex consonant harmony systems; retroflexion is contrastive in both non-sibilant (i.e., plosive) and sibilant obstruents (i.e., affricates and fricatives), but harmony applies only within each manner class, not between them. At the same time, harmony is not sensitive to laryngeal features. Theoretical implications of these and other similarity effects are discussed.