Phonotactic learning has been a fertile ground for research in the field of phonology. However, the challenge of lexical exceptions in phonotactic learning remains largely unexplored. Traditional learning models, which typically assume all observed input data to be grammatical, often blur the distinction between lexical exceptions and grammatical words, consequently skewing the learning results. To address this issue, this article innovates a categorical-grammar-plus-exception-filtering approach that harnesses the discrete nature of categorical grammars to filter out lexical exceptions using statistical criteria adapted from probabilistic models. Applied to naturalistic corpora from English, Polish and Turkish, the learnt grammars showed a high correlation with the acceptability judgements in behavioural experiments. Compared to benchmark models, the model performs increasingly better with data that contain a higher proportion of lexical exceptions, reaching its peak in learning Turkish non-local vowel phonotactics, highlighting its ability to handle lexical exceptions.

This study tests the hypothesis that alternation patterns with strong lexical support are more robust than those with no, or weak, lexical support. Focusing on three alternation patterns in Korean with varying productivity and generality, we measured lexical support in two ways. First, we conducted an acceptability-rating experiment investigating Korean speakers’ judgements on non-words with and without violations of the phonotactic constraints motivating the alternations. In addition, we performed a simulation of learning a maximum entropy (MaxEnt) Harmonic Grammar from a dictionary corpus. The results of the experiment and computational modelling confirmed the hypothesis by showing that if an alternation is robust, its associated phonotactic constraint is learned with a high weight from the MaxEnt simulation, and it affects the participants’ well-formedness ratings for non-words. Consequently, the results of this research support the claim of a tight link between alternations and phonotactics.

It is well-attested that floating tones can associate across a word boundary, but it is typologically unusual for floating weight units to do so. The Nuer language presents a floating suprasegmental component (FSC), which is part of lexical morphemes, and includes a unit of quantity and a High tone. This component is located at the left edge of nouns and is realised primarily across a word boundary on a preceding vowel. This article examines the FSC through a phonological analysis and a production study with eight speakers. These investigations reveal how the FSC interacts with the specifications for vowel length and tone of the adjacent context. Specifically, the weight unit of the FSC lengthens a preceding word-final short vowel, and its High tone combines in a compositional manner with tone of this preceding context. Comparisons with related languages suggest that the FSC developed out of a word-initial vowel /a/.

It is well-known that English variable word-final coronal stop deletion (CSD) is less likely to occur when the final coronal stop instantiates the inflectional suffix -ed. It is sometimes hypothesised that the reason for this effect is to avoid the homophony between past and present tenses that would result from the suffix -ed being deleted. This reasoning suggests another hypothesis: that CSD should also be disfavoured when it would create homophony between two distinct lexical items, such as bald and ball. In this squib, we test that hypothesis on data from a corpus of Philadelphia English. We find no evidence that probability of CSD is affected by homophony avoidance between lexical items. This weakens the case that homophony avoidance is at play in disfavouring CSD in the -ed case, and may have implications for the theory of homophony avoidance in phonology in general.

Mismatches in weight criteria across weight-sensitive processes within individual languages present difficulties for theories of moraic structure, particularly regarding coda weight. Previous accounts, which stipulate that codas are variably moraic to account for the typological variation in the weight status of CVC for primary stress, make incorrect predictions for the status of CVC in other weight-sensitive phenomena, including tone, word minimality and secondary stress, among others. This article proposes a theory of Uniform Moraic Quantity coupled with a new syllable weight metric as a solution, which captures CVC’s flexible weight status while maintaining the cross-linguistic moraicity of codas and avoiding the incorrect predictions that frustrate the standard variable-weight approach.

A key goal in phonology is to understand the factors that affect phonological learning. This article addresses the issue by examining how paradigms are reanalysed over time. Malagasy has a class of stems called weak stems, whose final consonants alternate under suffixation. Comparison of historical and modern Malagasy shows that weak stem paradigms have undergone extensive reanalysis in a way that cannot be predicted by the probabilistic distribution of alternants. This poses a problem for existing quantitative models of reanalysis, where reanalysis is always towards the most probable alternant. I argue instead that reanalysis in Malagasy is driven by both distributional factors and a markedness bias. To capture the Malagasy pattern, I propose a maximum entropy learning model, with a markedness bias implemented via the model’s prior probability distribution. This biased model successfully predicts the direction of reanalysis in Malagasy, outperforming purely distributional models.

Tima has a typologically unusual 12-vowel advanced tongue root (ATR) harmony system, contrasting six [+ATR] vowels /i e ɨ ʌ o u/ with six [−ATR] vowels /ɪ ɛ ɘ a ɔ ʊ/. This harmony system provides a test case for generalisations about ATR systems: for example, that [−ATR] is less compatible with higher vowels; that [+ATR] is less compatible with lower vowels and that central vowels are incompatible with [ATR] systems. After showing that all vowels participate fully in ATR harmony, this article presents an acoustic study of the Tima ATR contrast. We show that /ʌ/, the [+ATR] counterpart of /a/, patterns as a mid vowel, and that duration and voice quality differences characterise Tima’s crowded vowel inventory. Though F1 is the primary individual correlate of the ATR contrast, as is true cross-linguistically, a number of measures support voice quality differences as well, as predicted by the Laryngeal Articulator Model account of ATR systems.

The ambition of this article is to provide a phonological account of an intricate pattern of lenition and gemination in Campidanese Sardinian. The data show two things: that a model of phonology needs some way of showing strength and weakness as positional effects and that neither can be reliably understood in phonetic terms. In this analysis, the discovery procedure does not depend on raw phonetic facts, but rather on a rich model of abstract phonological representations. These representations are of two kinds, melodic and prosodic, and they allow for a substance-free phonological analysis of lenition and fortition in Campidanese that is not confronted by the difficulties inherent in surface-oriented approaches.

Do inputs need to be restricted on a language-specific basis? Classic Optimality Theory claims that they do not: the rich base is filtered by constraints that yield full contrast, complementary distributions or positional neutralisation depending on the ranking. The problem arises when positional neutralisation affects a gappy contrast. In Russian, voicing neutralisation works on all obstruents alike, including non-contrastively voiceless ones – but it creates voiced allophones that are otherwise disallowed. In the popular OT account of positional neutralisation, analysing these cases requires handling voicing twice: once for all segments, then again for gaps. I argue that the solution is to relax the rich base assumption by ruling gaps out at the UR level through morpheme structure constraints (Halle 1959 et seq.).

Phonological processes tend to involve local dependencies, an observation that has been expressed explicitly or implicitly in many phonological theories, such as the use of minimal symbols in SPE and the inclusion of primarily strictly local constraints in Optimality Theory. I propose a learning-based account of local phonological processes, providing an explicit computational model. The model is grounded in experimental results that suggest children are initially insensitive to long-distance dependencies and that as their ability to track non-adjacent dependencies grows, learners still prefer local generalisations to non-local ones. The model encodes these results by constructing phonological processes starting around an alternating segment and expanding outward to incorporate more phonological context only when surface forms cannot be predicted with sufficient accuracy. The model successfully constructs local phonological generalisations and exhibits the same preference for local patterns that humans do, suggesting that locality can emerge as a computational consequence of a simple learning procedure.

Feature-based models of sign language use distinctive features to describe the phonological structure of signs. We use near-minimal pairs and phonological phenomena like productivity and neutralisation in French Sign Language to show that the feature [web], which refers to the webbing part of the fingers, should be (re)introduced into the list of phonologically active features. In discussing potential cases of [web] in other sign languages and the impact on the shape of phonological inventories, we first offer an account of [web] in terms of a location feature in line with most traditional feature-geometry models. We then offer some speculations on why a more uniform characterisation of [web] and the features in the same subclass in terms of the orientation type results in more economical models.

We review and elaborate an account of consonantal strength that is founded on the model of speech as a modulated carrier signal. The stronger the consonant, the greater the modulation. Unlike approaches based on sonority or articulatory aperture, the account offers a uniform definition of the phonetic effect lenition has on consonants: All types of lenition (such as debuccalisation, spirantisation and vocalisation) reduce the extent to which a consonant modulates the carrier. To demonstrate the quantifiability of this account, we present an analysis of Ibibio, in which we investigate the effects of lenition on the amplitude, periodicity and temporal properties of consonants. We propose a method for integrating these different acoustic dimensions within an overall measure of modulation size. Not only does the modulated-carrier account cover all the classically recognised lenition types, but it also encompasses loss of plosive release in final stops – which, although not traditionally classed as lenition, is clearly related to processes that are.

Chong (2017) claims that a derived environment process is productive to the extent that it is supported by phonotactics. The present study tests this claim by comparing variable patterns observed in Korean vowel harmony of suffix alternation with vowel co-occurrence restrictions in the lexicon. A corpus study replicated findings of previous studies that the harmony in suffix alternation is losing productivity, conditioned by the quality of the stem vowel, the number of intervening consonants between the vowels and the stem class. Phonotactic generalisations in vowel sequences matched such tendencies in the alternation in that harmony was feeble in phonotactics and that some of the factors that modulate the harmony in alternation were found to affect the harmony in phonotactics as well. The findings generally support Chong’s claim that lack of phonotactic support for an alternation makes it harder to learn.

Southern Pomo (Pomoan, California) displays a process of rhythmic vowel deletion (syncope) reflecting two mutually incompatible metrical structures. This phenomenon, called metrical incoherence, can be derived by an ordered sequence of independent subgrammars, that is, strata. Metrical incoherence is under-attested crosslinguistically, and the stratal models of phonology necessary to generate it have been criticised for predicting counter-typological phenomena. Nevertheless, the Southern Pomo data cannot be generated in more restrictive frameworks. This article argues that overgeneration is a necessary property of the phonological component, and that metrical incoherence is rare because it is difficult to learn. In Southern Pomo, this difficulty appears to have caused grammatical competition and restructuring: a second pattern of syncope, occurring in only a limited context, suggests that learners have reanalysed the grammar as having consistent metrical structure across the derivation. This work thus supports the proposal that diachronic change – and therefore typology – is constrained by extragrammatical factors.

This study examines phonological and phonetic properties of ATR contrasts in the vowel system of Akebu (Kwa). The sum of descriptive evidence, including vowel harmony, vowel distribution in non-harmonising contexts, vowel reduction and typological and etymological considerations, indicates a rare vowel inventory with an ATR contrast in front/back vowels but a height contrast in the three redundantly [−ATR] central vowels /ᵻ, ə, a/. This analysis was checked against four common acoustic metrics of ATR: F1 and F2 frequencies, spectral slope and F1 bandwidth size (B1). As expected, the results for the last three metrics were variable across speakers and vowel types, and are therefore inconclusive. The results for F1 were consistent but do not distinguish between ATR and vowel height. Two results nonetheless suggest the [−ATR] status of central vowels: they occupy the same belt of F1 frequencies and show the same position of observed-over-predicted B1 values as front and back [−ATR] vowels.

The phenomenon of incomplete neutralisation describes a situation where a putative case of categorical phonological neutralisation is observed to be phonetically non-neutralising. This has been argued to be a problem for phonological theories that employ categorical features. Here, we use two distinct feeding orders of tone sandhi processes from Huai’an Mandarin to show that incomplete phonetic neutralisation is compatible with categorical phonological phenomena. Therefore, incomplete phonetic neutralisation does not automatically inform us of gradient phonological representations. We further show that incomplete phonetic neutralisation can in fact have a large effect size. Such results are not surprising from a classic generative view of phonology where linguistic performance is argued to be a multi-factorial problem, and linguistic knowledge (i.e., competence) is only one of the many factors involved. Furthermore, our results suggest that the observed incompleteness or gradience may have a source outside phonological knowledge.

Tone is distinct from other phonological phenomena both qualitatively and quantitatively (Hyman 2011), and has been instrumental in shaping phonological theory in many ways. However, the contributions to current linguistic theory of “grammatical tone’ (GT) – a type of non-concatenative morphology where a morpheme is expressed at least in part by tone and/or tone changes – have been less apparent. In this paper, we take stock of the types of GT patterns attested in the literature and the different theoretical treatments of GT that have been proposed to date. We show that GT is still to a large extent underexplored, and highlight the immense potential of the study of GT for improving our understanding of phonology and its outer limits. This paper serves as an introduction to the high-quality research articles collected in this special issue, which directly address how GT critically informs phonological theory and its current developments.

Dagaare is a language of northern Ghana and adjoining areas of Burkina Faso. There are two tones, H and L, and contrastive downstep H!H that involves a non-automatic pitch drop between two H tones. The challenge is to explain the extensive morphological conditioning of tonal processes, including dissimilation, downstep and spreading. Our solution involves level ordering: tones are introduced at different morphological levels (stems, words and phrases) and later processes can make earlier processes opaque. Tonal differences between nouns (spreading) versus verbs (no spreading) and stems (dissimilation) versus words (downstep) arise from constraint ranking differences within and across levels. There are two kinds of downsteps: stem-level downsteps are underlying L tones affiliated with some morpheme; word-level downsteps are L tones inserted by a general process of word-final lowering. Only one downstep per word is allowed. If more would arise, the morphologically inner downstep blocks the morphologically outer downstep.

Grammatical tone (GT) can be the sole exponent or a co-exponent of grammatical meaning (Hyman 2012; Rolle 2018), but there has been little discussion of how they distribute within a single language. In this article, I explore the relationship between tonal and segmental materials in Gyeli (Bantu A801, Cameroon), adopting a property-driven approach to phonological typology (Plank 2001; Hyman 2009). Gyeli has eight GTs in simple predicates, which serve as sole exponents of tense, aspect, mood and polarity distinctions and object-marking. When GT is a co-exponent accompanied by segmental material, for example, in auxiliary constructions, the information that the tonal component contributes to the meaning is insufficient to distinguish between grammatical categories: its functional load is weak. The decrease in functional load is correlated with an increase in length of a segmental co-exponent. This can be explained by the tonal cophonologies of segmental morphemes and their different GT dominance types.

Metathesis poses challenges for a typologically constrained theory of phonology: despite being simple to describe, its distribution is highly restricted, making it difficult to create analyses that make predictions while not overgenerating. Here, I investigate metathesis in Uab Meto (Austronesian; Indonesia), an understudied language with CV metathesis that is synchronic and productive. Drawing on original fieldwork, I argue that metathesis is not transposition, but instead a serial delete-and-spread mechanism (cf. Takahashi 2018, 2019). To support this, I present a deep case study into the language’s phonology, showing that metathesis arises from spreading, deletion and epenthesis patterns. I propose that synchronic metathesis systems like Uab Meto’s can only emerge from the successive application of these mechanisms, and hypothesise that true transposition, if it exists, only arises through morpheme-specific operations. This study thus presents a new look onto the typology of synchronic metathesis, and offers an explanatory account of its typological rarity.