A prosodic analysis of the Head Final Filter effects and beyond

In the Principles and Parameters framework of Generative Grammar, the various positions occupied by the verb have been identified as functional heads hosting inflectional material (affixes or features), which may or may not attract the verb. This gave rise to a hypothesis, the Rich Agreement Hypothesis (RAH), according to which the verb has to move to the relevant functional head when the corresponding inflectional paradigm counts as “rich.” The RAH is motivated by synchronic and diachronic variation among closely related languages (mostly of the Germanic family) suggesting a correspondence between verb movement and rich agreement. Research into this correspondence was initially marred by the absence of a fundamental definition of “richness” and by the observation of counterexamples, both synchronically (dialects not conforming to the pattern) and diachronically (a significant time gap between the erosion of verbal inflection and the disappearance of verb movement). Also, the research was based on a limited group of related languages and dialects. This led to the conclusion that there was at best a weak correlation between verb movement and richness of morphology. Recently, the RAH has been revived in its strong form, proposing a fundamental definition of richness and testing the RAH against a typologically more diverse sample of the languages of the world. While this represents significant progress, several problems remain, with certain (current and past) varieties of North Germanic not conforming to the expected pattern, and the typological survey yielding mixed or unclear results. A further problem is that other Germanic languages (Dutch, German, Frisian) vary as to the richness of their morphology, but show identical verb placement patterns. This state of affairs, especially in light of recent minimalist proposals relocating both inflectional morphology and verb movement outside syntax proper (to a component in the model of grammar interfacing between narrow syntax and phonetic realization), suggests that we need a more fundamental understanding of the relation between morphology and syntax before any relation between head movement and morphological strength can be reliably ascertained.

This chapter examines new data from Older and Modern Faroese, Older Scots, and Shetland Dialect that bring into question the nature of the relationship between the richness of morphology and (the loss of) V-to-I. It begins with an overview of V-to-I in Modern Icelandic and contrasts the Icelandic facts with those of Mainland Scandinavian, showing how Icelandic and Mainland Scandinavian differ with respect to V-to-I. It then presents an overview of V-to-I in Faroese from a diachronic and synchronic perspective, arguing that Older Faroese had V-to-I, whereas there is dialect split with respect to V-to-I in Modern Faroese. The chapter discusses residual V-to-I in Shetland Dialect both from a synchronic and diachronic perspective. Finally, it discusses syntactic triggers for V-to-I in light of the fact that Older Scots and Shetland Dialect had V-to-I in the absence of rich agreement morphology for a long period in their history.

This paper deals with the phenomenon of V-to-T movement, which is one of the major parameters differentiating Romance from the majority of modern Germanic languages, and it defends the idea that rich morphology is the cause or trigger of V-to-T: in Romance, in a modern Germanic language like Icelandic, and very particularly in Old English, the precursor of the modern English language. More generally, the discussion endorses the idea that all Germanic languages used to be V-to-T languages in their old periods. I begin by arguing that verbal forms in Spanish contain a specific kind of segment, namely the stem or thematic vowel, which gives rise to morphological variations or asymmetries across tenses in the language. Such a productive system of stem verb classes is also shown to be the case in Icelandic, though not in German (which is therefore rendered as non-V-to-T), and ultimately it is acknowledged for a language like OE. The hypothesis is that the syntactic computation of (OE) verbal forms demands it that the speaker first identifies the verb class that the form in question belongs to before tackling the processing of tense morphology and agreement morphology. In pure syntactic terms, the stem or thematic vowel segment is identified in the present account with a v-feature that T must value, which valuation is realised by means of the displacement of the verb to the T head, that is, by means of V-to-T movement. After the valuation of T’s v-feature comes the valuation of τ-features and φ-features, respectively.

Agreement attraction in Spanish comprehension

Spanish exhibits a determiner phrase (DP)-internal phenomenon (noun-drop or N-drop) closely analogous to subject pro-drop. Where English has the near-vacuous nominal one in the DP the blue one, for example, Spanish lacks any overt nominal: el azul, literally 'the blue'. The availability of N-drop in a language has been linked by some authors to richness of the overt agreement morphology on adjectives, determiners, or both. The evidence from child language acquisition, however, runs counter to this view. In particular, a detailed case study of the longitudinal corpus of child Spanish data from Montes (1987) revealed that the child acquired the full Spanish system of DP-internal agreement morphology significantly earlier than she acquired N-drop. This finding indicates that rich agreement morphology is not in itself a sufficient condition for N-drop.

This paper examines and evaluates what may be called the "Rich Agreement Hypothesis" (RAH) in the domain of verb movement asymmetries in Germanic. The most prominent current accounts (e.g., Rohrbacher's 1999 Morphology-Driven Syntax) require inspection of the internal make-up of paradigms and take overt morphological variation to be the cause of syntactic variation. A survey of the literature shows that these proposals are empirically untenable in their strong (bi-conditional) form; there are numerous cases of syntactic variation attested in the absence of corresponding morphological variation. The strongest sustainable descriptive generalization is a one-way implication from rich morphology to verb movement. Though this has been noted before, its implications have not been adequately discussed. While morphology-driven approaches could have explained a strong RAH; when faced with the weaker, one-way implication they can provide no account of why that correlation should hold, and are thus at best incomplete. That is, they provide no insight as to why there are no languages with rich morphology, but in which the finite verb remains in the VP. The particular correlations that are attested, and in particular the absence of a certain class of languages, do however follow from a theory which takes morphology to be not the cause, but rather a reflection, of syntactic structure, in line with common theorizing in morphology. The inflection-movement correlations that do exist therefore challenge rather than supporting morphology-driven approaches to morphosyntax.

The indigenous languages of the Americas exemplify a number of uncommon typological patterns, especially in their morphology. Here, that rich morphology is illustrated via the Muskogean languages of the southeastern United States. Muskogean languages are agglutinative, but even more interesting and uncommon patterns emerge in an analysis of their morphology. These include subtractive morphology, suppletion, infixation, ablaut, and the use of suprasegmentals. These morphological patterns present considerable complexity. Inflected verbs in narratives and conversation often reflect more than one of the morphological processes. This morphological complexity also demonstrates characteristics of being nonlinear, of being prosodic yet not aligning with neat prosodic boundaries, of not having direct correspondence between grammatical categories and surface segments or suprasegmentals, or having more than one of those characteristics. Six of the seven Muskogean languages are still currently spoken by fluent first language speakers, and many of the tribal nations who represent these languages are involved in ongoing documentation and revitalization efforts, often in partnership with linguists. Thus, despite their highly endangered status, excellent existing documentation and new questions in research create an opportunity to collect even more intricate inflected forms that will enrich models of morphology and morphological theory while having broader impacts, like supporting tribal language revitalization.

In the literature on Null Subject Languages (NSLs) since Rizzi (1982), the three properties that are commonly thought to be connected are (i) the richness of inflectional morphology, (ii) free subject inversion, and (iii) the COMP-trace effect. The connection between them is that if a language (e.g., Italian) has the option of having a null subject (NS) pronominalproin subject position (i.e., SpecIP) as a result of having rich inflectional (i.e., verbal agreement) morphology, as represented in (1a), then the same element may appear in the same position in cases of subjects occurring postverbally as in (1b) and in cases of long-distance subject extraction where the postverbal trace is properly governed by the verb, as in (1c):

In this entry we provide an overview of the alleged correlation between the finite verb and the properties of the inflection of the verb. Specifically, it has been attested that in languages where the verb exhibits little or no inflectional morphology, no elements can intervene between the verb and the direct object (in discourse‐neutral contexts). In contrast, such intervention effects have been abundantly attested in languages where the verb exhibits a sufficiently ‘rich’ morphology. This led to the hypothesis that verb movement is triggered by rich inflectional morphology. Thus far this relation between the syntax and the morphology of the finite verb has been addressed in great detail primarily for Germanic and related languages, but more recently the research has extended beyond them. We will review the empirical claims that led to the hypothesis, as well as their theoretical implications. In addition, we will evaluate the empirical facts in a number of non‐Germanic and non‐Indo‐European languages and address the issues that arise in them. We will see that the hypothesis has not been fully confirmed or rejected yet, partly because, even though it is superficially far‐reaching, there often are particular, language‐specific properties and confounding effects that make its straightforward evaluation impossible.

We assessed whether and under what conditions noncanonical agreement patterns occur in Russian, with the goal of understanding the factors involved in normal agreement. Russian is a morphosyntactically rich language in which agreement involves features for number, gender, and case. If consistent, overt specification of number and gender agreement features supports agreement processes in language production, agreement should be less vulnerable to number and gender attraction than in languages with sparse agreement morphology. A related question was the degree to which notional number influences agreement patterns in morphologically rich languages. We varied the grammatical and notional number properties of sentence subjects and examined the effects on the predicate in sentence completion tasks using native Russian speakers. Noncanonical agreement occurred, but at rates lower than those observed in English and other languages without rich number morphology. Noncanonically plural predicates occurred more often after notionally plural subjects, suggesting notional number agreement, but the incidence was also lower than in languages with sparser agreement morphology. Gender attraction was almost nonexistent. The results suggest that morphology arbitrates grammatical agreement processes and reduces the impact of variations in notional number.

In a number of languages, agreement in specificational copular sentences can or must be with the second of the two nominals, even when it is the first that occupies the canonical subject position. Béjar & Kahnemuyipour (2017) show that Persian and Eastern Armenian are two such languages. They then argue that ‘NP2 agreement’ occurs because the nominal in subject position (NP1) is not accessible to an external probe. It follows that actual agreement with NP1 should never be possible: the alternative to NP2 agreement should be ‘default’ agreement. We show that this prediction is false. In addition to showing that English has NP1, not default, agreement, we present new data from Icelandic, a language with rich agreement morphology, including cases that involve ‘plurale tantum’ nominals as NP1. These allow us to control for any confound from the fact that typically in a specificational sentence with two nominals differing in number, it is NP2 that is plural. We show that even in this case, the alternative to agreement with NP2 is agreement with NP1, not a default. Hence, we conclude that whatever the correct analysis of specificational sentences turns out to be, it must not predict obligatory failure of NP1 agreement.

The purpose of this paper is to analyze the distribution and characteristics of null arguments in Korean Sign Language (KSL). Languages such as Spanish and Italian, which have a rich system of overt morphological agreement, allow subject-drop but not object-drop. For this reason, it has been said that the null arguments are licensed by rich verbal morphology. On the other hand, languages such as Korean and Chinese do not have a rich agreement system like Spanish and Italian. Thus, the verbs are used with the same ending regardless of person, number, and gender features. In this paper, we show that KSL is different from those languages like Korean and Chinese in that arguments can be dropped with both agreeing and non-agreeing verbs. In some contexts, in addition, null arguments of non-agreeing verbs are not allowed, whereas null arguments of agreeing verbs are totally acceptable. Therefore, to solve these problems, in this paper, we suggest hybrid analysis in licensing null arguments in KSL.

Some languages allow subjects and even other arguments to be null, while others do not. This entry explores the role morphology plays in the licensing of pro‐drop. There are two ways in which morphology has been implicated. First, Taraldsen's generalization stated that languages that allow null subjects have rich verbal agreement morphology. Although this generalization seems to have stood the test of time, it is less clear what richness of agreement exactly means. In addition, there is also no consensus on what the exact grammatical mechanisms are that account for this generalization. We explore these issues by looking at languages with somewhat syncretic paradigms that nevertheless allow null subjects, languages with rich paradigms that do not, partial pro‐drop languages and the loss of null subject over time in languages like French. On the other hand, Neeleman and Szendrői proposed that in radical pro‐drop languages the personal pronominal paradigm must be agglutinating for case, number or some other nominal feature. Languages with non‐transparent, fusional pronominal paradigms do not allow radical pro‐drop. We discuss this proposal in detail as well as some challenges it faces, including a potential counter‐example. The final part of the entry explores the blocking effects that arise from languages having different pronominal inventories and how these contribute to the availability of null subjects. We conclude that overall morphology plays a substantial role in the licensing of null subjects and other arguments and that our understanding of null subject licensing is enhanced by exploring the role of morphology in this domain.

The languages of Borneo display remarkable typological diversity, both in their phonology and in their morphosyntax. Phonologically we identify two main types: (a) languages whose phonological patterns resemble those found in mainland SE Asia, including such features as word-final stress and complex syllable structure; and (b) languages whose phonological patterns are similar to those found in other parts of island SE Asia, including such features as trochaic stress patterns and less syllabic complexity. In terms of grammatical features, we identify three main types: (a) the Philippine-type, found in NE Borneo; (b) the Indonesian-type, found throughout the remainder of Borneo; and (c) the Central Bornean grammatical type, which, like the MSEA phonological type, occurs mostly in the central latitudes. Characteristic features of the Central Bornean grammatical type include: a lack of applicative suffixes and pronominal prefixes; simple two-way voice distinctions often involving a periphrastic, rather than morphological, passive or Undergoer Voice; and complex pronominal number systems, with as many as five number distinctions in some languages. The other two types have been widely discussed by other authors. The Philippine-type is characterized by verb-initial word order, rich verbal morphology, and complex voice systems often involving four or more voices. Indonesian-type morphosyntax, characterized by the presence of both pronominal prefixes and applicative suffixes, is described here only briefly.

Grapheme-To-Phoneme (G2P) converts an input letter sequence into its corresponding pronunciation. Several rule-based G2P models are available for Korean, but their performance in view of individual phonological rules in Korean is not well-investigated. We establish phonological rules to be reflected in a G2P model for Korean, and test the model performance of two most popular G2P models of Korean, i.e., g2pk and KoG2P, regarding their performance on each of these rules. We created a golden corpus to evaluate the performance of the current G2P models based on manual phonetic transcription. We then measured the performance of two G2P models, and identified the phonological rules in which the models show relative success or failure in deriving the correct output. We implemented additional phonological rules such as h-deletion, ui-variation, consonant place assimilation, and restructured the ordering of rules such as h/th-neutralization in the model. We show that our revised model makes a substantial improvement on model performance. Further, we argue that the major limitation of the current rule-based approaches to G2P is in its binary approach to phonological rules and lack of information about prosodic boundaries. We propose that a rule-based G2P system should reflect the stochastic nature of phonological processes informed by existing research on the gradient nature of phonological rule application as a function of factors such as lexical frequency and speech register.