It is well known that stress assignment in English noun–noun compounds is non-uniform (compare e.g. left-prominent ópera glasses and right-prominent steel brídge), and recent corpus-based studies (e.g. Plag et al. 2007, 2008) have shown that categorical, rule-based approaches that make use of argument structure (e.g. Giegerich 2004) or semantics (e.g. Fudge 1984) are not able to account satisfactorily for the existing variability. Using data from the corpus studies by Plag and collegues, I argue in this paper that an exemplar-based approach is better-suited to accounting for stress assignment in English noun–noun compounds than a traditional, rule-based paradigm. Specifically, it is shown that two current implementations of exemplar-based algorithms, TiMBL (Daelemans et al. 2007) and AM::Parallel (Skousen & Stanford 2007), clearly outperform comparable rule models in terms of how well they predict stress assignment in the corpora. Furthermore, systematic testing reveals that the reasons for the differences between exemplar and rule models mainly lie in their ability to incorporate detailed, non-abstract information (specifically, constituent family information). The present study therefore adds to the growing evidence in favour of the importance of constituent family information in compounding (e.g. Gagné 2001, Krott, Schreuder & Baayen 2002).

In this article, I argue that at least in some subsets of grammar, non-standard dialects are indeed more natural than their standard counterparts. I present data from the new Freiburg English Dialect corpus FRED, for the first time comparing and quantifying traditional dialect data from across the whole of Great Britain. The most frequent non-standard verb forms cluster around forms like drink – drunk – drunk and sing – sung – sung. The framework of Natural Morphology (Wurzel 1984, 1987) in combination with Bybee's Network Model (Bybee 1985, 1995) is employed to define the notion of naturalness and to explain why this verb class has been strengthened historically, and is still attracting new members today.

The idea that certain morphological and phonological irregularities are due to speakers' desire to avoid homophony is widely invoked, yet has also come under strong criticism as an explanation which is neither necessary nor sufficient. In most cases there is no way to resolve the question, since the assumption that something is being avoided is itself a theoretical construct. In this article I attempt to address this last difficulty by looking at gaps in inflectional paradigms – where it is clear that something is being avoided – that plausibly correlate with potential homophony. These fall into two types: (i) lexical, where portions of the paradigms of two lexeme would be homophonous, and (ii) paradigmatic (i.e. syncretism), where forms within the paradigm of a single lexeme would be homophonous. Case studies of Tuvaluan, Russian, Mazatec, Tamashek and Icelandic confirm the effects of homophony avoidance as a genuine, if non-deterministic, principle.

Phonological chain shifts have been the focus of many theoretical, developmental, and clinical concerns. This paper considers an overlooked property of the problem by focusing on the typological properties of the widely attested ‘s>θ>f’ chain shift involving the processes of Labialization and Dentalization in early phonological development. Findings are reported from a cross-sectional study of 234 children (ages 3 years; 0 months–7;9) with functional (nonorganic) phonological delays. The results reveal some unexpected gaps in the predicted interactions of these processes and are brought to bear on the evaluation of recent optimality theoretic proposals for the characterization of phonological interactions. A developmental modification to the theory is proposed that has the desired effect of precluding certain early-stage grammars. The proposal is further evaluated against the facts of another widely cited developmental chain shift known as the ‘puzzle>puddle>pickle’ problem (Smith 1973).

Unlike most other grammatical domains, that of personal pronouns is clearly under-researched in works on grammaticalization. One reason can be seen in the fact that personal pronouns differ in their diachronic behavior from most other grammatical categories to the extent that they present a challenge to grammaticalization theory. In the present paper it is argued that in order to account for this behavior, an extended understanding of grammaticalization is needed.

Allomorphy is a reaction of the morphological system to problems that the unrestrained application of inflectional and other rules creates at the phonological level. These problems are dealt with in some cases but left unattended in others. A diachronic analysis of English reveals that phonemically conditioned allomorphy originates from gradual sound change, with the old and the new variant forming a morphophonological paradigm. The historical stability as well as the synchronic motivation of allomorphy are claimed to be frequency-based. The higher the frequency, the longer the life expectancy. Synchronically, there is a (language- and domain-specific) frequency threshold above which morphophonological variation occurs and below which it fails to occur. The underlying logic of the model is that frequency encourages lexicalization at all linguistic levels. The relative ease with which high-frequency items are accessed enhances the tolerance towards formal variation, hence the emergence of natural phonological rules. As the application of these rules is context-dependent, allomorphy arises as a context-sensitive process. Repair strategies are also argued to be under the sway of frequency. Epenthesis is found in highly frequent structures while coalescence is reserved for less frequent ones. Frequency also determines the scope and the optionality of morphophonological rules. Phonemically governed allomorphy is shown to be a member of the larger family of variationist phenomena which are bound together by their sensitivity to frequency.

This paper argues that neutralizing phonological alternations are sensitive to how much homophony they create among distinct lexical items: neutralizing rules create fewer homophones than expected. Building on a case study of Korean by Silverman (2010), I compare the neutralizing rules of Korean to a large number of hypothetical alternatives generated by Monte Carlo simulations. The simulations reveal that the actual rules of Korean frequently create far fewer homophones than similar (but unattested) rules, even when the rules that are compared are controlled for the number of phonemic contrasts they eliminate. These results suggest that phonological patterns are sensitive not only to high-level contrasts among phonemes but also to contrasts among individual lexical items. The effect is most pronounced when homophones are not weighted by frequency, a result that adds to evidence in the literature that the relevant measure of lexical frequency for many lexicon-sensitive phonological patterns is type frequency, not token frequency.

In this paper it is argued that there existed, in a certain period of the history of Latin, a floating C-Place node in some lexical items in word- and stem-initial position. Notably, this was involved in the phonological representation of the words written – in an archaising fashion – with initial 〈gn〉. Based on a thorough analysis of the Brepols Corpus (CLCLT-5) it is demonstrated that the diachronic distribution of the prefixed forms of 〈gn〉-initial stems shows restrictions that can only be explained if one assumes a geometric representation involving a floating C-Place node that remained in the place of the original velar stop inherited from Proto-Indo-European. By classical times this floating node was also lost, and thus the possibilities for combining prefixes with original 〈gn〉-initial stems expanded.

Since early studies in language development, scholars have noticed that function words, in particular auxiliaries, often appear to be missing in early speech, with the result that child utterances sometimes exhibit verbs with non-finite morphology in seemingly matrix clauses. This has led to the idea of a ‘deficit’ in the child's syntactic representations. In contrast with previous studies, this article explores the possibility that the child's phonology may considerably impact her overt realization of auxiliaries. Specifically, it examines the hypothesis that non-finite verbs in early speech are in fact attempted periphrastics (i.e. auxiliary/modal+non-finite verb) in which the auxiliaries are just reduced phonetically, often to the point where they remain unpronounced. We studied 28 normally developing French-speaking children aged between 23 and 37 months. New observational data uncovered a continuum in a given child's phonetic realizations of auxiliaries. Children showed various levels of auxiliary reduction, suggesting that their non-finite verbs are best analyzed as being part of periphrastics involving an auxiliary form that represents the endpoint on this continuum, i.e. is (completely) deleted. Further examination of these verbs revealed that their semantics corresponds to the semantics of adult periphrastics. Additionally, the results of an experiment where children imitated sentences with either periphrastic or synthetic verbs showed that responses with non-finite verbs were predominantly produced when the target sentence involved a periphrastic, rather than a synthetic verb.

This paper presents an analysis of bare nominals unmarked for number (BNs) occurring in object position in Spanish and Catalan, on which the BN is a syntactic complement to the verb, but not a semantic argument. After describing the properties that distinguish BNs from other indefinite expressions (bare plurals, indefinite singulars preceded by un ‘a’, and bare mass terms), we argue that these BNs occur in a monadic syntactic configuration in the sense of Hale & Keyser (1998),that they denote first-order properties, and that they are combined with the verb via a modified version of Dayal's (2003) semantics for pseudo-incorporation. Specifically, the proposal consists of a lexical rule that generates the class of verbs that productively accept BN objects, plus a composition rule that treats the BN as modifier of the verb. We point out the advantages of this analysis over three other well-known semantic analyses for combining verbs with property-type nominals. Finally, we show how the analysis can be naturally extended to existential sentences, which combine with BNs although, prima facie, they do not appear to meet the lexical conditions for doing so.

I investigate the syntax and semantics of a previously unexamined English adjective construction, exemplified by sentences like Middlemarch is a long book to assign. The construction, which I call the nominal attributive-with-infinitive construction (nominal AIC), is of interest for the semantics of gradability and modality. I argue that the major interpretive characteristic of the nominal AIC – the interpretation of inappropriateness associated with it – arises from the interaction between the positive degree operator associated with the gradable adjective and the modality of the infinitival relative clause, which contributes to the computation of the standard of comparison. Nominal AICs are compared and contrasted with a surface-identical construction I call the clausal AIC, with attributive too, and with attributive comparatives; they are shown to exhibit major syntactic and semantic differences from all of these. The paper serves both as a contribution to the semantic literature on gradability and as a contribution to the descriptive grammar of English, as it is, to the best of my knowledge, the first systematic description and analysis of the nominal AIC.

Blocking in inflection occurs when a morphological exponent prevents the application of another exponent expressing the same feature value, thus barring the occurrence of multiple exponents of a single morphosyntactic feature value. In instances of extended exponence, more than one exponent in the same word realizes the same feature value. We provide a unified account of blocking and extended exponence that combines a realizational approach to inflection with Optimality Theory (Realization Optimality Theory), encoding morphological realization rules as ranked violable constraints. The markedness constraint *Feature Split bars the realization of any morphosyntactic feature value by more than one exponent. If *Feature Split ranks lower than two or more realization constraints expressing the same feature value, then we observe extended exponence. Otherwise, we find blocking of lower-ranked exponents. We show that Realization Optimality Theory is superior to various alternative approaches to blocking and extended morphological exponence.

It is argued that in Francilian French, the dialect of French spoken in the Paris metropolitan area, in-situ and fronted wh-questions have the same answerhood conditions but vary with respect to their respective focus-set (Reinhart 2006). The difference between the two types of questions lies in the discourse status of their non-wh portion. Whereas the wh-phrase is never discourse-given, the non-wh portion may or may not be, depending on the discourse context. In Francilian French in-situ wh-questions, the non-wh portion must be discourse-given. As this language exhibits a strong requirement on sentence stress to be kept rightmost it cannot, in contrast with English, assign sentence stress to a fronted wh-phrase when the non-wh portion is discourse-given and needs to be destressed. The only way to simultaneously destress discourse-given items and keep sentence stress rightmost is by aligning the wh-phrase with the right edge of the clause. Whereas in Hungarian prosody triggers movement (Szendrői 2003), in Francilian French, prosody prevents it from occurring. An Optimal Theoretic analysis in the spirit of much recent work on focus and givenness in declaratives (Samek-Lodovici 2005, Féry & Samek-Lodovici 2006) captures this phenomenon.

This paper presents a description and a theoretical analysis of agreement within coordination in Lamaholot (Austronesian), where the first conjunct agrees with the conjunction ‘and’. Conjunctive agreement is obligatory in the subject position but is optional in the object position. The analysis is couched in terms of markedness of case and proposes that the case of the subject (nominative) is unmarked compared to the case of the object (accusative), and that only unmarked case enables phi-features of the agreement control to be copied onto the agreement host. Apparent optionality is accounted for by manners of case spreading in coordination. Conjunctive agreement is also reported in the genetically unrelated but areally related language of Walman. Although conjunctive agreement originates in verbal agreement with the comitative function in both languages, it is shown that grammaticalization from the comitative to the conjunction is more advanced in Lamaholot, at least in terms of syntax and morphology.

This paper is an investigation of freezing effects, that is, cases where an element (e.g. an object or a subject, or an element within it) is unable to move from a certain structural position. An account of the most prominent properties of freezing in Norwegian is followed by a comparative study of primarily English and Norwegian indirect objects, with important consequences for the general approach to indirect objects. Although recent analyses capture central properties of indirect objects, they fall short of accounting for freezing properties, seen here in terms of agreement properties, most notably Case agreement. It is shown that both subjects and indirect objects disallow sub-extraction in both English and Norwegian; however, unlike English, Norwegian allows the indirect object to A-bar move. This relates to the question of whether Case is structural or inherent. As such, this paper offers a new argument in favor of Case as a central ingredient in deriving freezing effects.

Lightfoot (2002) argues that syntactic reconstruction is rendered impossible by the lack of any analogue in syntax to the traditional notion of the phonological ‘correspondence set’ of the Comparative Method and by the radical discontinuity caused by reanalysis between successive grammars. Alice Harris and Lyle Campbell, in various works, have defended the notion of ‘syntactic pattern’ as the analogue of the correspondence set, arguing that patterns can be compared across languages, with innovations being stripped away to reveal aspects of the protolanguage. In this article, I argue that syntactic reconstruction can be carried out while maintaining and indeed utilizing core notions in generative approaches to syntactic change such as the central role of reanalysis and child language acquisition and the distinction between the abstract grammatical system and the surface output of that system. Reanalysis itself is constrained by the fact that both pre- and post-reanalysis grammars must be acquirable on the basis of the same primary linguistic data. This imposes limits on the possible hypotheses that can be entertained (‘local directionality’) even in the absence of any crosslinguistic generalizations about patterns of change (‘universal directionality’). This approach is then applied to aspects of the syntax of free relative clauses and negation in the early Brythonic Celtic languages (Welsh, Breton and Cornish), showing that non-trivial reconstructions can be achieved even where the daughter languages manifest significant differences.