Volume 38, Issue 3 - Special Issue: Phonological solutions to morphological problems; Guest Editors: Heather Newell and Shanti Ulfsbjorninn

There is a tendency for syncretism between future and infinitive stems in Modern Hebrew. Verbs with final orthographic gutturals do not follow this trend in one verbal type. In another, they do follow it, but their exponent is different from that of regular verbs. Previous studies have claimed that (i) gutturals are represented in Modern Hebrew as a vowel /a/ (Faust, Noam. 2005. The fate of gutturals in Modern Hebrew . Tel Aviv: Tel Aviv University MA Thesis); (ii) Infinitives are derived in two cycles (Faust, Noam & Vered Silber-Varod. 2014. Distributed Morphology and prosody: The case of prepositions. In Burit Melnik (ed.), Proceedings of IATL29 (MITWPL 72) , 71–92. Cambridge (MA): MIT Press); and (iii) stems seek to be no shorter than two syllables (e.g. Bat-El, Outi. 2003. The fate of the consonantal root and the binyan in Optimality Theory. Recherches Linguistiques de Vincennes 32. 31–60.). Relying on these claims, an analysis is proposed involving two allomorphs with a priority relation. Phonological considerations of multiple correspondence, word size and cyclicity may nevertheless override the effect of priority, leading to the selection of the non-default allomorph. In the last section I briefly discuss two alternatives to the priority relation: the autosegmental alternative and the gradient alternative.

Athapaskan verbal morphology appears to violate the Mirror Principle in multiple ways and, thus, the ordering of affixes in these languages has resisted a straightforward analysis. We adopt a new morphological tool of Iterative Root Prefixation, which allows for a more direct mapping from syntax to morphology in languages of this profile. Apparent violations of affix ordering that remain, namely the puzzling placement of the transitive and causative morphemes, are argued to be explained by overriding phonological constraints.

Minimalist Morphology predicts that allomorphy is conditioned inward and locally, and that the domains of morphosyntactically and phonologically conditioned allomorphy selection are identical. Amy Rose Deal and Matthew Wolf have put forward two cases of allomorphy in Nez Perce that appear to be conditioned by an outward phonological context. I present an analysis of Nez Perce morphology and phonology which supports the conclusion that the first case is not outward-conditioned, and the second case is not allomorphy but phonology.

Bracketing Paradoxes (BPs) have been the subject of many different analyses since the 1970s. Each of these analyses have included BP-specific machinery to account for the apparent mismatch between the syntactico-semantic and morpho-phonological structures argued to be necessary for a complete analysis of this phenomenon. This article proposes that independently necessary operations and structures in the morpho-syntactic and phonological modules allow for an analysis of BPs that avoids postulating ad-hoc tools. Specifically, a system that includes cyclic (phasal) interpretation of the morpho-syntax in combination with a flat (CVCV) phonological framework avoids the emergence of paradoxical structures altogether. The discussion therefore includes both current morpho-syntactic and phonological analyses of each construction proposed to give rise to a BP; comparatives ( unhappier ), Level-ordering BPs ( ungrammaticality ), Phrasal BPs ( modular grammarian ), Compound BPs ( particle physicist ), Particle-verbs ( podžëg ‘set fire’ [Russian]), and Reduplicated BPs ( kwíita-kwíita ‘to pour a bit’ [Kihehe]). The proposal that a flat phonological framework is key in avoiding the paradoxical nature of BPs has implications for the correct structure of phonological representations generally.

Ik presents a widespread pattern of allomorphy characterised by morpheme-specific segment-zero alternations (deletions) and de-affrication. Part of the process is clearly phonological because it applies to every item in the language. Final vowels are devoiced into oblivion, though they are always recoverable in Non-Domain-Final context. Case allomorphy shows various item-specific patterns, some affixes overwrite the final-vowels of roots, others do not, and others yet exhibit consonant-zero alternations. Meanwhile, interestingly, there is (ostensibly) morphologically-sensitive consonant deletion that is fully complementary with what is clearly phonological vowel deletion. In fact, an entirely phonological analysis is possible if the consonant-zero alternations are modelled by floating consonants belonging to the exponents of certain morphemes. Therefore, the item/morpheme-specific aspect of the process is actually specific to the exponent of that morpheme. This locates the whole analysis in the phonology and provides a straightforward account of the complementarity of the two deletion processes. The formal analysis is very close to what has been proposed for French, with its floating consonants and its irregular hiatus-driven vowel deletion. The analysis will then be extended to the item-specific idiosyncratic process of de-affrication that is blocked, not by coincidence, precisely in morphologically overwriting environments. In the end, all the phenomena can be accounted for in a modular analysis, provided one gets the representations right.

The paper discusses a three-way allomorphic pattern of neuter-gender stems in Czech. We argue that there are two surface alternations involved in the three-stem pattern, i.e. a-e and t -zero, and that the two alternations are driven by two distinct mechanisms, i.e. suppletion and regular phonological computation respectively. We postulate two suppletive stem-building suffixes that are lexically connected via the Nanosyntactic tool of pointers. We argue that the default a -initial suffix spells out the neuter gender and its suppletive e -initial variant is a portmanteau morpheme of the (neuter) gender and (singular) number. We further propose that the suppletive suffix is stored with a floating consonant that does not surface word-finally due to regular phonological computation.

A purely phonological account of reduplication based on the affixation of empty prosodic nodes predicts the attested typology of multiple reduplication. Languages that can combine more than one reduplication-triggering morpheme in a word differ in (1) whether all reduplicants surface faithfully, (2) whether they systematically avoid adjacent multiple reduplicants, or (3) whether one of the reduplicants is smaller than expected if another reduplicant is adjacent in multiple reduplication contexts. Morphological accounts of reduplication not only violate the modularity between phonology and morphology, they also fail to predict this attested typology.