Morphological Paradigms

A key notion in understanding language is 'possible word (lexeme)'. While there are lexemes that are internally homogeneous and externally consistent, we find others with splits in their internal structure (morphological paradigm) and inconsistencies in their external behavior (syntactic requirements). I first explore the characteristics of the most straightforward lexemes, in order to establish a point in the theoretical space from which we can calibrate the real examples we find. I then schematize the interesting phenomena that deviate from this idealization, including suppletion, syncretism, deponency, and defectiveness. Next I analyze the different ways in which lexemes are 'split' by such phenomena. I set out a typology of possible splits, along four dimensions: splits that are (i) based on the composition/feature signature of the paradigm versus those based solely on morphological form; (ii) motivated (following a boundary motivated from outside the paradigm) versus purely morphology-internal ('morphomic'); (iii) regular (extending across the lexicon) versus irregular (lexically specified); (iv) externally relevant versus irrelevant: we expect splits to be internal to the lexeme, but some have external relevance (they require different syntactic behaviors). I identify instances of these four dimensions separately: they are orthogonal, and therefore not dependent on each other. Their interaction gives a substantial typology, and it proves to be surprisingly complete: the possibilities specified are all attested. The typology also allows for the unexpected patterns of behavior to overlap in particular lexemes, producing some remarkable examples. Such examples show that the notion 'possible word' is more challenging than many linguists have realized.*

In many languages of the world prominent (human/animate, definite/specific, topical) non-subject participants receive special marking by means of flagging (case or adpositions), indexing (verbal agreement/cross-reference or clitic-doubling), or both, i.e. double-marking. I investigate the distribution of double-marking of objects cross-linguistically and the factors that favour it on the basis of a representative sample of languages. I focus on the semantic (macro)roles that are subject to double-marking and their interplay with prominence hierarchies. My first case-study explores the considerable typological variation of double-marking patterns in monotransitive and ditransitive constructions and shows how the observed asymmetries in this domain are motivated by prominence-related diachronic trends. My second case-study focuses on the virtually unexplored phenomenon of double-marking of peripheral semantic roles instantiated through indexing of oblique objects. I show that this phenomenon, even if infrequent cross-linguistically, is robustly attested world-wide and is again functionally and diachronically related to referential and discourse prominence.

This paper analyzes variation in the marking of number on plural nouns in mesolectal Jamaican Patwa (JP) – one of only three variable features for which comparable quantitative data exist from Creole and African American English speech communities. Earlier theoretical claims for grammatical and functional principles to constrain variation in JP, and English-related Creoles generally, are tested and found wanting. Many previous empirical studies lacked a valid, sufficiently nuanced taxonomy of surface forms which can reliably map onto the level of reference, and permit reorganization at a more abstract level capable of allowing generalizations. This analysis considers the choice between plural –z and zero in regular nouns in light of the major known potential linguistic constraints – syntactic, semantic, pragmatic and phonological. Results are compared with other contemporary Creoles, African American Vernacular English (AAVE), and African American Diaspora varieties. Two corpora are...

Language contact phenomena have increasingly been researched from different historical linguistic, sociolinguistic and areal-typological perspectives. However, since most of this research is based on case studies, an assessment of contact phenomena from a worldwide comparative perspective has been missing in the literature. In this article, we draw inspiration from historical linguistics and language typology to present a new typological approach for evaluating evidence that given linguistic domains have been affected by language contact. This method has three parts: (1) a new approach to sampling, (2) the analysis of typological data, and (3) making probabilistic inferences about language contact. We argue that this is a parsimonious method for evaluating contact effects that can serve as a starting point for the further development of typological approaches to language contact.

THE GLIDE SHIFT IN RUSSIAN DEVERBAL DERIVATION* 0. INTRODUCTION ~ 0.1. One of the positive achievements to emerge from the recent trend toward increasingly abstract analysis of phonological systems has been the recognition of the complementary relationship ofj and v in the phonological system of Contemporary Standard Russian (CSR). Evidence that j and v are, respectively, the front and back glides in CSR was convincingly presented by Henning Andersen in 1969.9 Since that time several investigators have attempted to capitalize on this notion to provide new analyses of CSR morphological and derivational phenomena. 0.2. In 1970 I proposed a glide shift rule j> v (Flier, [1970] 1972), which predicted the appearance of v before the imperfectivizing suffix aj, e.g., UBIT' (u+b,#j)/UBIVAT" ((u+b,iv)+dj) 'to kill', SDUT' (s+daj)/ SDUVAT' ((s+duv)+dj) 'to blow away', ZAVLADET' (za+(vlad+dj))/ ZAVLADEVAT' ((za + (vlad+ev)) +dj) 'to take possession, seize'. Similar solutions were proposed independently by Merz (1970) and Coats (1970). The complementary relationship between the genitive plural desinences ov and ej mentioned by Worth (1970) was elaborated by Shapiro (1971). A similar case can probably be made for the verb suffix ova, which is preserved before consonantal desinences, but replaced by uj before vocalic ones. This awaits further investigation.Z 0.3. The present paper focusses attention on the glide shift as it pertains to CSR deverbal derivation. To my knowledge this topic has not been investigated previously in any depth. Specifically the discussion concentrates on the glide shift in derivatives based on verb roots ending in j, e.g., BIT' (b,#j) 'to beat': RAZBOfNIK 'robber', UBIJCA 'murderer', MASLO-BOJNJA 'churn', RAZBIVKA 'spacing out', OTBIV 'repelling', NA-BIVAT' 'to stuff'. Such verb roots are designated j/v roots to distinguish them from other verb roots ending inj which do not utilize the glide shift in derivation, e.g. BOJAT'SJA (boj+4 +...s,a) 'to fear': BOJAZLIVOST' 'timidity', BOJAZN' 'dread, fear', POBAIVAT'SJA 'to be rather afraid'; TAJAT' (t~j+a) 'to melt', STAIVAT" 'to melt (away)', TAJALKA 'melter',

Since Blevins (2006), there has been a shift in morphological frameworks away from what he called a constructive perspective towards an abstractive perspective based on data directly available to speakers (i.e whole words). This evolution towards word-based morphology is part of a more general anticonstructionist movement in social sciences characterised by the quote in (1) about constructive approaches cited by Blevins et al. (2016a):

This paper deals with development and application of Russian morphology software and resources. The approach is particularly dependent on advanced morphological analysis. The paper presents the structure, formats and content of Russian dictionaries and corpora. Relevant aspects of the UML data models, XML format and related technologies are surveyed. We introducee the system based on Java and Oracle 9i DBMS.

Among dD qrce persons, tlE 6rst clearly has special ilnpoltatrc€ to speak€rs Recall dtat Newms r€constructed distirct foims fot lhe Rob-Salish 6rst ad s€cond pe$on' singul,t alld plu.al sufixes. In Somish and Saanich Norh Straits Salish, how€v€r, only fi$t Frsotr suffixos now have distinct sirgular and plDnl fotms. Numb€r need not be distinguish€d fd se€ond p€rson (Montler 1986' G.Ioway 1986). A srparate post-predicate p{tigle day be used to indicae numb€t if dle speakor vistes: ?hele// 'secoDd person pluraliz€d indical€s that any r€cmd p€.son leference in the clause is to be consEu€d as a plrral reftrt[ce' (Mor t€f, 1986:218). The Eatrspeency of lhis form indicale's lhal it is r€c€n! although it does trot demons[ate &at mdking numb€a wilh s€codd p€rsons ic etr iDnovatiql. It ep€ars lhat rumb€r in first p€rsotr was suffrciertly important to pe.sist' but number in second D€asol le€s so.

This paper critically reviews S. J. Keyser and Paul Kiparsky's "Syllable Structure in Finnish Phonology" (1984). We also treat Finnish morphophonemics using a CV-tier analysis, but account for forms that their method fails on. We posit six stem types in the verbal lexicon and derive the forms of all inflections of all types with just twelve morphophonemic rules. After the stems are modified and suffixes (and person/number endings) are attached, a single, general, phonological rule of Consonant Gradation applies postlexically. To develop a general theory from our analysis, we consider restrictions on rules which may alter stems: lexical rules apply to stems before affixation and only to the final segments of stems; only post-lexical phonological rules apply to affixes. We eliminate all morphological conditions on individual rules and extrinsic ordering conditions.

This paper proposes a sketch of how /sk/, a Latin suffix of word formation, developed into an inflectional stem-extension in the French i-conjugation. I will try to show how an understanding of inflectional classes as sets of properties and implications in the line of Wurzel (1984) and Dressler et al. (2003) can be used to reconstruct this process. Moreover, a hypothesis will be proposed regarding the role of lexical storage in morphological change. The paper is organized as follows: In section 1, I shortly address the general assumptions I am making about inflectional morphology. The topics are: morphological vs. phonological variation of word-forms, the concept of inflectional class as applied to the French i-conjugation, the relative lexical weight of inflectional classes, multiple class membership, the assignment of inflectional class membership, and inflectional classes as features. In section 2, I will give an outline of the evolution of /sk/ during the Latin period. In section 3, I will first discuss the question of why /sk/ could survive after the loss of its meaning and then propose a sketch of the lexical evolution of i-verbs from Old to Modern French. * This work was financially supported by the University of Konstanz and the Cooperative Research Centre 471 "Variation and Evolution in the Lexicon", funded by the Deutsche Forschungsgemeinschaft (German Research Foundation). I want to thank Bernard Fradin and Sascha Gaglia for giving me most helpful comments on earlier versions of this paper. I am also indebted to Verena Rick, who helped me retrieve and organize the Old French data.

Although the Lithuanian and English languages are bound within the family of Indo-European languages, the typological differences between the two languages lie in the system of inflectional and derivational morphology. The paper analyses the concept of nominalization and discusses the deverbal process and result nominalizations in Lithuanian and English. For the comparative qualitative and quantitative analysis, 965 equivalents of deverbal nouns have been selected from the "Parallel Corpus". Out of them, 802 examples belong to the category of deverbal process nouns, whilst the category of deverbal result nouns includes 163 examples. From the point of view of morphology, in both languages nominalization is a word-formation process by which a noun is derived from a verb, adjective or another noun, or even other parts of speech, usually through suffixation and by adding the ending in the Lithuanian language. Two types of nominalization can be found across languages: lexical and syntactic. Lexical nominalization refers to the formation of deverbal nouns or nominal words derived from the verb or a nominal word, and syntactic nominalization refers to turning a clause into a noun phrase. In summary, the investigation of the derivational affixes of deverbal nouns in Lithuanian and their equivalents in English has revealed the following differences: in Lithuanian, the deverbal nominalizations-deverbal process nouns and deverbal result nouns-can be formed with 132 suffixes and 5 endings, whilst in English-with 10 suffixes and by employing the derivational strategy of conversion. Also, the analysis of the empirical material revealed that the suffix-imas/-ymas in Lithuanian prevails in forming deverbal process nouns (they make 73 per cent of all deverbal process nouns), while the suffix-inys is the most prolific in forming deverbal result nouns (they make 38 per cent of all deverbal result nouns). The English equivalents usually have the suffix-ion/-tion/-sion/-ation, quite many derivatives have the suffixing. It should be noted that deverbal nominalizations in the Lithuanian language often correlate with abstract and concrete nouns (non-derivatives) in the English language: 23 per cent of all derivatives in Lithuanian have more than one equivalent (derivative or non-derivative) in English.

In Urdu phrases, a very small feature [e] that marks certain nouns plays astonishingly complex and extensive role. Until now it has been just taken for granted. It used to be described merely as a link between a noun and its case marker called oblique marker. But when the researchers focused it closely, we came to know it is not simply a linking feature. It plays an important role in binding the constituents of a DP, and also in its recurrence. It also transforms different word classes into nominals, and often it functions as a postposition. Different researchers like Schmidt, Mohanan, Butt have described Urdu postpositions differently, but exploration of this particle has enabled the researchers to synthesize their theories, and differentiate the functions of case markers from postpositions.

Although the Lithuanian and English languages are bound within the family of IndoEuropean languages, the typological differences between the two languages lie in the system of inflectional and derivational morphology. The paper analyses the concept of nominalization and discusses the deverbal process and result nominalizations in Lithuanian and English. For the comparative qualitative and quantitative analysis, 965 equivalents of deverbal nouns have been selected from the “Parallel Corpus”. Out of them, 802 examples belong to the category of deverbal process nouns, whilst the category of deverbal result nouns includes 163 examples. From the point of view of morphology, in both languages nominalization is a word-formation process by which a noun is derived from a verb, adjective or another noun, or even other parts of speech, usually through suffixation and by adding the ending in the Lithuanian language. Two types of nominalization can be found across languages: lexical and syntacti...

Since Blevins (2006), there has been a shift in morphological frameworks away from what he called a constructive perspective towards an abstractive perspective based on data directly available to speakers (i.e whole words). This evolution towards word-based morphology is part of a more general anticonstructionist movement in social sciences characterised by the quote in (1) about constructive approaches cited by Blevins et al. (2016a):

The vocative is a largely understudied linguistic phenomenon. This text examines the implications of treating the vocative as a case value and of not treating it as such in languages with a neutral alignment of case marking of full noun phrases. Special attention is given to such languages lacking (overt) case (marking), barring, supposedly, (marking for) the vocative. The text argues in favour of employing a principled treatment of the vocative that affords sufficient room for typological generalisation while remaining faithful to language data.

Linguistics, and typology in particular, can have a bright future. We justify this optimism by discussing comparability from two angles. First, we take the opportunity presented by this special issue of Linguistic Typology to pause for a moment and make explicit some of the logical underpinnings of typological sciences, linguistics included, which we believe are worth reminding ourselves of. Second, we give a brief illustration of comparison, and particularly measurement, within modern typology.

Features are central to all major theories of syntax and morphology. Yet it can be a non-trivial task to determine the inventory of features and their values for a given language, and in particular to determine whether to postulate one feature or two in the same semantico-syntactic domain. We illustrate this from tense-aspectmood (TAM) in Kayardild, and adduce principles for deciding in general between one-feature and two-feature analyses, thereby contributing to the theory of feature systems and their typology. Kayardild shows striking inflectional complexities, investigated in two major studies (Evans 1995a; Round 2013), and it proves particularly revealing for our topic. Both Evans and Round claimed that clauses in Kayardild have not one but two concurrent TAM features. While it is perfectly possible for a language to have two features of the same type, it is unusual. Accordingly, we establish general arguments which would justify postulating two features rather than one; we then apply these specifically to Kayardild TAM. Our finding is at variance with both Evans and Round; on all counts, the evidence which would motivate an analysis in terms of one TAM feature or two is either approximately balanced, or clearly favours an analysis with just one. Thus even when faced with highly complex language facts, we can apply a principled approach to the question of whether we are dealing with one feature or two, and this is encouraging for the many of us seeking a rigorous science of typology. We also find that Kayardild, which in many ways is excitingly exotic, is in this one corner of its grammar quite ordinary.

0.1: In the preceding paper of this volume I have tried to show that not only the nominal (and adjectival) e-stems of Lithuanian, but also the verbal preterit foirmations in -e;ceui be derived from older formations in'/*-jaor *-i-a>) *-iJa.. The qiiestion arises whether the historical developments postulated in that paper have, as such historical processes often do, left traces in the synchronic structure which would permit (or force) the native speaker to infer underlying forms and phonological derivations identical with, or at least quite similar to, the reconstructed pre-forms and historical processes.

Some Portuguese verbs have two different past participles, such as, e.g., aceitar 'accept', with participles aceitado and aceito; and limpar 'clean', with limpado and limpo. The first one in each pair mentioned is thematic, whereas the second one is athematic. While regular thematic participles all bear stress on the theme vowel, these athematic participles all bear the primary stress on the athematic stem. As the morphosyntactic category first person singular present indicative (1spi) is realized by {-o}, it normally coincides the masculine form of this athematic participle, giving rise to a syncretism between 1spi and the participle. The aim of this paper is to track the appearance of this kind of participle and the resulting syncretism in Portuguese and the changes making it possible for new participles to be formed in this way in colloquial Brazilian Portuguese.

Carstairs-McCarthy (2001) proposes a theory of directionality in grammatically conditioned allomorphy based on the idea of ancestry in word structure and purporting to derive the following empirical claim: (1) Grammatical allomorphy may be inwardly sensitive to particular features of a node in a morphological structure, or to the category of the whole node; however, it cannot be outwardly sensitive to particular features of a node. This approach aims to improve on Bobaljik (2000), which differs empirically in two respects: in outwardly sensitive conditioning, it predicts no distinction

Although Qurʾānic Arabic (7 th century CE) is usually considered an exponent of pre-Classical Arabic, 2 it shows an agreement system rather distinct from the one displayed in other texts from this oldest stage of the language, namely in pre-Islamic Poetry (6 th century CE). In fact, Qurʾānic Arabic agreement must be considered revelative of a transitional stage between the oldest manifestations of pre-Classical Arabic on one hand, and Classical Arabic on the other hand, i.e., the increasingly standardized language found in official and literary texts from the mid-8 th century CE onward. While in pre-Islamic poetry, feminine plural agreement with non-human plurals is still very common (with controllers 3 of any gender or morphological type), it is an agreement type clearly on the decline in the Quran: 4 Here, the use of feminine plural with non-human plural controllers has become restricted to several specialized contexts 5-a fact which, at least quantitatively, foreshadows the further receding and semantic reanalysis of feminine plural agreement in Classical Arabic, 6 and finally its almost complete loss in Modern Standard Arabic. 7 As we look upon feminine plural agreement in a broad diachronic perspective, its monotonous decline along the succession of varieties and text groups belonging to the Old Arabic 1 The research reported in this article was supported by the "Bavarian Equal Opportunities Sponsorship-Realisierung von Chancengleichheit von Frauen in Forschung und Lehre (FFL)-Realization Equal Opportunities for Women in Research and Teaching." 2 See Fischer 1982: 38. 3 The term 'controller' commonly denotes "the element which determines the agreement" (Corbett 2006: 4).

0.1: In the preceding paper of this volume I have tried to show that not only the nominal (and adjectival) e-stems of Lithuanian, but also the verbal preterit foirmations in -e;ceui be derived from older formations in'/*-jaor *-i-a>) *-iJa.. The qiiestion arises whether the historical developments postulated in that paper have, as such historical processes often do, left traces in the synchronic structure which would permit (or force) the native speaker to infer underlying forms and phonological derivations identical with, or at least quite similar to, the reconstructed pre-forms and historical processes.

Features are central to all major theories of syntax and morphology. Yet it can be a non-trivial task to determine the inventory of features and their values for a given language, and in particular to determine whether to postulate one feature or two in the same semantico-syntactic domain. We illustrate this from tense-aspectmood (TAM) in Kayardild, and adduce principles for deciding in general between one-feature and two-feature analyses, thereby contributing to the theory of feature systems and their typology. Kayardild shows striking inflectional complexities, investigated in two major studies (Evans 1995a; Round 2013), and it proves particularly revealing for our topic. Both Evans and Round claimed that clauses in Kayardild have not one but two concurrent TAM features. While it is perfectly possible for a language to have two features of the same type, it is unusual. Accordingly, we establish general arguments which would justify postulating two features rather than one; we then apply these specifically to Kayardild TAM. Our finding is at variance with both Evans and Round; on all counts, the evidence which would motivate an analysis in terms of one TAM feature or two is either approximately balanced, or clearly favours an analysis with just one. Thus even when faced with highly complex language facts, we can apply a principled approach to the question of whether we are dealing with one feature or two, and this is encouraging for the many of us seeking a rigorous science of typology. We also find that Kayardild, which in many ways is excitingly exotic, is in this one corner of its grammar quite ordinary.

In this paper, I approach the agglutionation-fusion distinction from an empirical point of view. Although the well-known morphological typology of languages (isolating, agglutinating, flexive/fusional, incorporating) has often been criticized as empty, the old idea that there are (predominantly) agglutinating and (predominantly) fusional languages in fact makes two implicit predictions. First, agglutination/fusion is characteristic of whole languages rather than individual constructions; second, the various components of agglutination/fusion correlate with each other. The (unstated, but widely assumed) Agglutination Hypothesis can thus be formulated as follows: (i) First prediction: If a language is agglutinating/fusional in one area of its morphology (e.g. in nouns, or in the future tense), it shows the same type elsewhere. (ii) Second prediction: If a language is agglutinating/fusional with respect to one of the three agglutination parameters (a-c) (and perhaps others), it shows the same type with respect to the other two parameters: (a) separation/cumulation, (b) morpheme invariance/morpheme variability, (c) affix uniformity/affix suppletion. I report on a study of the nominal and verbal inflectional morphology of a reasonably balanced worldwide sample of 30 languages, applying a variety of measures for the agglutination parameters and determining whether they are cross-linguistically significant. The results do not confirm the validity of the Agglutination Hypothesis, and the current evidence suggests that "agglutination" is just one way of trying to capture the strangeness of non-Indo-European languages, which all look alike to Eurocentric eyes. 1 Earlier versions of this paper were presented at the 3rd conference of the Association for Linguistic Typology (Amsterdam 1999) and at the 9th International Morphology Meeting (Vienna 2000). I am grateful to the audiences at these occasions, as well as to two reviewers, for useful comments. (The core ideas of this paper were first presented at the DGfS Summer School on Language Typology in Mainz, September 1998.) 2 For the flexive type, other term variants such as (in)flexional/inflectional are also often used. As was often noted (e.g. Bazell 1958), the term inflectional is confusing because it also has a different sense: One also says that agglutinating languages have inflection (i.e. different word-forms belonging to a single lexeme), so Sapir's term fusional has tended to supplant it in the typological sense. Plank (1999) retains the term flexive (deliberately differentiating it from inflectional), presumably because he feels that there is much more to the agglutination/flexion distinction than what Sapir meant by fusion (cf. note 7). In this paper, I use fusion as the opposite of agglutination, simply because it seems that this term is now better known.

We examine the notion of '(inflectional) periphrasis' within the framework of Canonical Typology, and argue that the canonical approach allows us to define a logically coherent notion of periphrasis. We propose a set of canonical criteria for inflectional morphology and a set of canonical criteria for functional syntax, that is, syntactic constructions which include functional elements and which express grammatical features. We argue that canonical periphrasis is exemplified in our theoretical space of possibilities whenever a cell in a (canonically morphological) inflectional paradigm ('feature intersection') is expressed by a multiword construction which respects the canonical properties of functional syntax. We compare our canonically-based approach with the approach of other authors, notably, Ackerman & Stump (2004), who argue for three sufficient conditions for a construction to be regarded as periphrastic: feature intersection, non-compositionality and distributed exponence. We argue that non-compositionality and distributed exponence, while sometimes diagnostic of periphrasis on a language-particular basis, do not constitute canonical properties of periphrasis. We also examine crucial but neglected syntactic aspects of periphrastic constructions: recursion of periphrases and headedness of periphrastic constructions. The approach we propose allows us to distinguish between constructions in actual languages which approximate the ideal of canonical periphrasis to various degrees without committing us to a categorical distinction between periphrastic and non-periphrastic constructions. At the same time we can capture the intuition that there is in some languages a distinct identifiable set of multiword constructions whose principal role is to realize grammatical features.

Morphological systems often reuse the same forms in different functions, creating what is known as syncretism. While syncretism varies greatly, certain cross-linguistic tendencies are apparent. Patterns where all syncretic forms share a morphological feature value (e.g., first person, or plural number) are most common cross-linguistically, and this preference is mirrored in results from learning experiments. While this suggests a general bias towards natural (featurally homogeneous) over unnatural (featurally heterogeneous) patterns, little is yet known about gradients in learnability and distributions of different kinds of unnatural patterns. In this paper we assess apparent cross-linguistic asymmetries between different types of unnatural patterns in person-number verbal agreement paradigms and test their learnability in an artificial language learning experiment. We find that the cross-linguistic recurrence of unnatural patterns of syncretism in person-number paradigms is proport...

We provide lexical profiling for Arabic by covering two important linguistic aspects of Arabic lexical information, namely morphological inflectional paradigms and syntactic subcategorization frames, making our database a rich repository of Arabic lexicographic details. First, we provide a complete description of the inflectional behaviour of Arabic lemmas based on statistical distribution. We use a corpus of 1,089,111,204 words, a pre-annotation tool, knowledge-based rules, and machine learning techniques to ...

In this paper we extend our work described in by adding more conjugational rules to the labelling system introduced there, in an attempt to capture the entire dataset of Romanian verbs extracted from (Barbu, 2007), and we employ machine learning techniques to predict a verb's correct label (which says what conjugational pattern it follows) when only the infinitive form is given.

Like the numbers in a sudoku puzzle, a lexeme's principal parts provide enough information-but only enough-to deduce all of the remaining forms in its paradigm. Because principal parts are a distillation of the implicative relations that exist among the members of a lexeme's paradigm, they afford an important (but heretofore neglected) basis for typological classification. We recognize three logically distinct sorts of principal-part systems that might be postulated for a given language: static, adaptive, and dynamic. Focussing for present purposes on dynamic systems, we propose five crosscutting criteria for the typological classification of principal-part systems. These criteria relate to (i) how many principal parts are needed to determine a lexeme's paradigm; (ii) whether distinct lexemes possess parallel sets of principal parts; (iii) how many principal parts are needed to determine a given word in a lexeme's paradigm; (iv) what sort of morphological relation exists between a principal part and the forms that it is used to deduce; and (v) whether lexemes' nonprincipal parts are inferred from their principal parts in the same way from one inflection class to another. Drawing on these criteria, we propose a novel classification of a range of typologically diverse languages.

Nominal classification remains a fascinating topic but in order to make further progress we need greater clarity of definition and analysis. Taking a Canonical Typology approach, we use canonical gender as an ideal against which we can measure the actual gender systems we find in the languages of the world. Building on previous work on canonical morphosyntactic features, particularly on how they intersect with canonical parts of speech, we establish the distinctiveness of gender, reflected in the Canonical Gender Principle:In a canonical gender system, each noun has a single gender value. We develop three criteria associated with this principle, which together ensure that canonically a noun has exactly one gender value; we give examples of non-canonicity for each criterion, thus gradually building the typology. This is the essential groundwork for a comprehensive typology of nominal classification: the Canonical Typological approach allows us to tease apart clusterings of properties...

We examine the notion of '(inflectional) periphrasis' within the framework of Canonical Typology, and argue that the canonical approach allows us to define a logically coherent notion of periphrasis. We propose a set of canonical criteria for inflectional morphology and a set of canonical criteria for functional syntax, that is, syntactic constructions which include functional elements and which express grammatical features. We argue that canonical periphrasis is exemplified in our theoretical space of possibilities whenever a cell in a (canonically morphological) inflectional paradigm ('feature intersection') is expressed by a multiword construction which respects the canonical properties of functional syntax. We compare our canonically-based approach with the approach of other authors, notably, Ackerman & Stump (2004), who argue for three sufficient conditions for a construction to be regarded as periphrastic: feature intersection, non-compositionality and distributed exponence. We argue that non-compositionality and distributed exponence, while sometimes diagnostic of periphrasis on a language-particular basis, do not constitute canonical properties of periphrasis. We also examine crucial but neglected syntactic aspects of periphrastic constructions: recursion of periphrases and headedness of periphrastic constructions. The approach we propose allows us to distinguish between constructions in actual languages which approximate the ideal of canonical periphrasis to various degrees without committing us to a categorical distinction between periphrastic and non-periphrastic constructions. At the same time we can capture the intuition that there is in some languages a distinct identifiable set of multiword constructions whose principal role is to realize grammatical features.

Sometimes dismissed as linguistically epiphenomenal, infl ectional paradigms are, in reality, the interface of a language's morphology with its syntax and semantics. Drawing on abundant evidence from a wide range of languages (French, Hua, Hungarian, Kashmiri, Latin, Nepali, Noon, Old Norse, Sanskrit, Turkish, Twi, and others), Stump examines a variety of mismatches between words' content and form, including morphomic patterns, defectiveness, overabundance, syncretism, suppletion, deponency, and polyfunctionality. He demonstrates that such mismatches motivate a new grammatical architecture in which two kinds of paradigms are distinguished: content paradigms , which determine word forms' syntactic distribution and semantic interpretation, and form paradigms , which determine their infl ectional realization. In this framework, the often nontrivial linkage between a lexeme's content paradigm and its stems' form paradigm is the nexus at which incongruities of content and form are resolved. Stump presents clear and precise analyses of a range of morphological phenomena in support of this theoretical innovation. gregory stump is a Professor of linguistics at the University of Kentucky. His principal research area is the theory and typology of complex systems of infl ectional morphology.

Some Portuguese verbs have two different past participles, such as, e.g., aceitar 'accept', with participles aceitado and aceito; and limpar 'clean', with limpado and limpo. The first one in each pair mentioned is thematic, whereas the second one is athematic. While regular thematic participles all bear stress on the theme vowel, these athematic participles all bear the primary stress on the athematic stem. As the morphosyntactic category first person singular present indicative (1spi) is realized by {-o}, it normally coincides the masculine form of this athematic participle, giving rise to a syncretism between 1spi and the participle. The aim of this paper is to track the appearance of this kind of participle and the resulting syncretism in Portuguese and the changes making it possible for new participles to be formed in this way in colloquial Brazilian Portuguese.

Nen verbal morphology is remarkably complex; a transitive verb can take up to 1,740 unique forms. The combined effect of having a large combinatoric space and a low-resource setting amplifies the need for NLP tools. Nen morphology utilises distributed exponence - a non-trivial means of mapping form to meaning. In this paper, we attempt to model Nen verbal morphology using state-of-the-art machine learning models for morphological reinflection. We explore and categorise the types of errors these systems generate. Our results show sensitivity to training data composition; different distributions of verb type yield different accuracies (patterning with E-complexity). We also demonstrate the types of patterns that can be inferred from the training data through the case study of syncretism.

This paper sketches the current status of morphology and paradigms in linguistic<br> theorising. In particular, it is shown that from a constructionist as well as from<br> a compositional perspective, morphology including paradigms tends to dissolve.<br> The former might be less obvious; however the paper argues that a constructional<br> deconstruction of paradigms and morphology follows directly from Haspelmath's<br> (2011) take on Booij (2010) and related approaches in the realm of Construction<br> Morphology (CxM). The latter is more obvious; in particular, proponents of<br> Distributed Morphology (DM) regularly emphasise that morphology is but an<br> interface and paradigms are epiphenomenal. Throughout the paper I assume some<br> familiarity with Construction Morphology and Construction Grammar more generally<br> whereas I introduce DM specifically. However, the paper is not intended as a<br> thorough discuss...

We examine the notion of '(inflectional) periphrasis' within the framework of Canonical Typology, and argue that the canonical approach allows us to define a logically coherent notion of periphrasis. We propose a set of canonical criteria for inflectional morphology and a set of canonical criteria for functional syntax, that is, syntactic constructions which include functional elements and which express grammatical features. We argue that canonical periphrasis is exemplified in our theoretical space of possibilities whenever a cell in a (canonically morphological) inflectional paradigm ('feature intersection') is expressed by a multiword construction which respects the canonical properties of functional syntax. We compare our canonically-based approach with the approach of other authors, notably, , who argue for three sufficient conditions for a construction to be regarded as periphrastic: feature intersection, non-compositionality and distributed exponence. We argue that non-compositionality and distributed exponence, while sometimes diagnostic of periphrasis on a language-particular basis, do not constitute canonical properties of periphrasis. We also examine crucial but neglected syntactic aspects of periphrastic constructions: recursion of periphrases and headedness of periphrastic constructions. The approach we propose allows us to distinguish between constructions in actual languages which approximate the ideal of canonical periphrasis to various degrees without committing us to a categorical distinction between periphrastic and non-periphrastic constructions. At the same time we can capture the intuition that there is in some languages a distinct identifiable set of multiword constructions whose principal role is to realize grammatical features. Brown, Dunstan; Marina Chumakina; Greville Corbett; Gergana Popova & Andrew Spencer. Forthcoming. Defining 'periphrasis': key notions. To appear in Morphology. (2) kamlang2 1 son3-siaw3 PROG chat 'were chatting' (Enfield 2007:209) In (2) the preverbal modifier kamlang2 means 'to be in the process of (V)-ing'. The progressive modifier in Lao is not obligatory, it "does not occur often in texts, and is limited to situations in which the ongoing or extended nature of the action is critical to the current framing of discourse" (Enfield 2007:209). This type of expression must be accounted for by syntax. There is, however, a third possibility: a language can have inflectional ways of realizing some grammatical features, but there are also instances of syntactic structures fulfilling this function. Thus, present and past tense in Russian are realized within the verb, as ( ) and (4) show. (3) my bolta-em we chat-PRS.1PL 'We chat/we are chatting.' (4) my bolta-l-i we chat-PAST-PL 'We chatted/we were chatting.' However, for imperfective verbs the corresponding future in ( ) is realized as two words. (5) my bud-em bolta-t' we be.FUT-1PL chat-INF 'We will chat/ we will be chatting.' Given the expectation arising from (3) and ( ) that a grammatical distinction will be realized by a single inflected word, we have an instance of periphrasis in (5). Of course, this expectation can come about for a number of reasons, and periphrasis is a term which describes a space of related phenomena. Existing approaches to the problem typically try to address periphrasis in terms of the binary question of whether it is syntax or morphology. In this paper we argue that, logically, it can be both, and we take on the challenge of demonstrating that this represents a coherent claim. We do this by adopting a canonical approach and defining the dimensions relevant for periphrasis. A significant motivation for doing this is that it is a prerequisite for empirical research. This paper makes use of some of the distinctions of , but takes issue with the need for non-compositionality and distributed exponence as criterial, while suggesting that recursion and headedness are relevant. We claim that this follows naturally from basing our definitions on canonical syntax and canonical morphology together, because deviations in terms of non-compositionality and distributed exponence, are themselves deviations from canonical syntax and canonical morphology. Ackerman & Stump's (2004:111) stated aim is to develop, "… an explanatory account of the special characteristics of periphrastic expressions." They do this by applying an inferential-realizational framework. Our aim, in contrast, is to map out the logical space of 1 Numbers here denote tones

Latin has a large group of derivatives whose suffixes appear to attach to the stem of the verb’s passive participle. The agent noun caesor ‘cutter’ is an example: inside caesor there is a stem caesthat is identical to that of the participle caesus, -ī ‘which has been cut-masc.-sg’, from the verb caedere. The same stem caesappears in other derivatives: caesiō ‘an act of cutting’, caesūra ‘a cutting, separation’, caesus, -ūs ‘the fact of cutting’, caesim ‘by cutting’. Other deverbal forms add their suffixes to the root: e.g. caedēs ‘a cutting down, carnage’ and caeduus ‘fit for cutting’. What is striking about this data is that no syntactic property is shared exclusively by the forms using caes-, or by their complement set using caed-. What then is the rationale for the stem selection? Why did Latin choose caes-or over *caed-or, but caed-ēs over *caes-ēs? This is what we set out to explain. This Latin pattern is an instance of a general morphological phenomenon. I call it syntacticall...