Search Results

Derivation is the formation of lexemes by means of affixation, reduplication, conversion, or root-and-pattern morphology. The outputs of such processes often exhibit idiosyncratic properties. Morphological processes differ as to the number of constraints they are subject to and in their degree of productivity. Morphological analysis of complex words also comprises the investigation of constraints on affix ordering in multiply complex words.

This chapter introduces the contents of this book which consists of four parts. Firstly it looks at syntactic and semantic ordering. Then it considers honological and morphological ordering. Next, it looks at psycholinguistic and cognitive ordering. Finally, it moves on to description and acquisition of affix order. The volume discusses novel data from various typologically diverse languages, including languages whose affix ordering has not been examined so far, and the reader will find examples from Adyghe (Caucasian), Athpare and Camling (Kiranti), Bulgarian, Russian and Polish (Slavic), English and German (Germanic), Greek, French and Italian (Romance), Huave (language isolate, Huavean), Karachay-Balkar (Turkic), Kurmanji (Kurdish), and Lithuanian (Baltic). Unlike previous research on affix ordering, most of the contributions in this collection consider more than one language. The introductory chapter briefly introduces the approaches to affix order and provides descriptions of the chapters included in the volume, along with a discussion of the recent developments in the research on affix ordering.

This chapter is an application of the model introduced in Chapter 5 to much of the empirical data summarized in Chapter 3, illustrating the intermediate types of lexical relatedness, specifically transpositions, argument nominalizations (such as DRIVER), the various paradigmatically mixed categories such as nouns which have the morphology of adjectives, diminutives and augmentatives, and finally, derivational morphology which can be defined formally but which is not associated with any systematic semantics (‘meaningless derivation’), as illustrated by prefixed verbs such as under-stand. One of the aspects of PFM which is explored in Stump’s work on paradigm linkage and related issues is the question of how inflected forms are interpreted semantically. In this chapter I take up this aspect in more detail. I modify Booij’s distinction between contextual and inherent inflection, arguing that some types of inflection are best thought of as introducing an additional semantic predicate into the lexical representation, in addition to realizing the value of a morphosyntactic feature. Inherent inflection is defined as relatedness which is part of the inflectional paradigm of a lexeme, and which therefore does not alter the lexemic index, but which nonetheless introduces a non-trivial change to the SEM value. Examples are case-marked forms of nouns in which the case marker has exactly the same meaning/function as a spatial preposition in English.

This chapter analyzes the linear order of inflectional suffixes in the Kiranti language Athpare. It is argued that the order of those suffixes reflects a language-specific hierarchy of morphosyntactic feature classes. The analysis is couched in Optimality Theory and based on ALIGNMENT constraints and a markedness constraint that demands an unambiguous marking of the agent argument. This demand refers to the well-known finding that case-marking and (fixed) order of elements interact in a crucial way. Some closely related Eastern Kiranti languages show a slightly different strategy to mark agent arguments prominently and easily fall out under a slightly modified ranking of the same constraints. From a typological point of view, this study of affix ordering is interesting since the Eastern Kiranti languages do not obey the typological tendency of ordering between person and number agreement following the hierarchy Person >> Number.

This chapter investigates the structure of the Slavic word with data from Bulgarian, Russian, and Polish. Affix order is analyzed in terms of two-affix combinations, and it is argued that with respect to affixation the Slavic word consists of different domains and subdomains defined through the affix-order peculiarities they exhibit. Special attention is paid to the order of suffixes in the derivational domain where the lexical-category specification of a suffix, suffix-particular semantics, and the notion of default are the factors governing the combinability of suffixes and the majority of the suffix combinations are either fixed or predictable. The lexical- and semantic-category specifications of suffixes are seen as cognitively defined and it is demonstrated that entrenchment, in the sense of cognitive grammar, plays an important role in affixation. The structure of the Slavic word is also compared with that of the English word.

Originally Inuit languages had a temporal opposition between future and non-future. Over time, Iñupiaq has evolved an opposition between past, present, and future, while Inuktitut and West Greenlandic (WG) have retained the future/non-future opposition. This chapter will show that it is incorrect to describe WG as tenseless; WG has grammaticalized tense affixes. There are five such future-tense affixes: -ssa ‘future’, -niar ‘inevitable future’, -jumaar ‘vague future’, -ssamaar ‘planned future’, and -ler ‘near future’. The first three also have modal meanings: -ssa ‘should’, -niar ‘intention’, -jumaar ‘promise’. While, ssamaar has only one meaning, -ler has an additional three: two in the inner phasal aspect (‘begin’ and ‘be about to’), and one in outer phasal aspect (‘be about to’). This chapter shows that it is necessary to pay close attention to morphological structure, and to allow for the possibility that tense can be expressed by derivational affixes.

Huave, a language isolate of Oaxaca State, Mexico, has the phenomenon of mobile affixation, where the same affix may surface as a prefix or as a suffix depending at least partly on phonological context. In light of Paster’s (2009) claim that the modular architecture of grammar precludes phonological and morphological constraints from operating in tandem, I reassess Kim’s (2010) analysis of Huave mobile affixation in which phonological constraints outrank morphological ones. An empirically superior generalization in terms of phonological subcategorization frames is formulated, although the fine line between nonmodular and modular analyses of the same phenomenon serves to highlight the unstable relationship between typology and theory, and emphasizes the need to cast an inclusive net in the cross-linguistic study of phonological influences on affix ordering.

Combinations of affixes in Italian have been addressed by describing individual, affix-pertinent restrictions that rely on grammatical principles. The chapter examines whether constraints that rely on extra-grammatical principles may account for suffix combinations, testing the psycholinguistic hypothesis known as Complexity-Based Ordering (CBO; Hay 2002). This hypothesis suggests that affix ordering is based on processing complexity and aims to establish an affix hierarchy to predict all affix combinations. Compared to a previous paper on this subject (Gaeta 2005), a larger set of suffixes and new sources of linguistic data are employed. Moreover, suffix hierarchies are computationally established. Using data from works on the productivity of Italian affixes, this study also assesses whether any correlations between productivity and suffix position in the hierarchies exist, as the CBO hypothesis claims. Finally, an assessment of the role of selectional restrictions in suffix combinations and a description of such constraints are provided.

This chapter investigates how affixes are organized and specifically how negation morphology operates in Kurmanji Kurdish. In Kurmanji, the verb stem encodes tense information and has two different shapes: the present and past verb stem. Person agreement and tense/aspect/mood (henceforth TAM) information are carried out by the verb stem and certain affixes. It is argued that affix ordering in Kurmanji is templatic because some affixes occupy the same slot, but they cannot co-occur simultaneously, which implies the presence of blocking effects within the affix positions. Negation functions in a similar way; that is, it appears on verb stems as a prefix which has basically three morphological markers, n(a)-, n(e)-, and ni-, and the appearance of the negation prefix precludes the appearance of certain prefixes, such as progressive aspect prefix di- and subjunctive mood prefix bi -.

This chapter discusses cases in which morphological constructions are prevented from combining with certain bases, or from combining with each other in the expected manner, because of phonological conditions that the combinations would violate. These situations are described in Optimality Theoretic terms as instantiating the P » M ranking, in which a phonological consideration outranks a morphological consideration. Suppletive allomorphy, ineffability (blocking), haplology (e.g. the Repeated Morph Constraint), and phonologically conditioned affix ordering are explored, along with theoretical treatments of each. The chapter concludes that phonological interference with morphology, especially in the area of affix ordering, is not nearly as common as many theories predict that it should be.

This chapter deals with parasynthetic verbs in Modern Greek whose structure contains both prefixes and suffixes, such as apo-cefal-iz-o ‘to decapitate’ (DPREF-BASE-DSUFF-ISUFF). After a brief overview of the relevant literature it is shown that the linear ordering of the prefixes and suffixes reflects the information content of each affix and its contribution to the morphological structure of the parasynthetic verb. More specifically, it is suggested that prefixes in parasynthetic verbs behave like internal prefixes (Di Sciullo 1997) and determine the core semantics of the verb and that affixation is influenced by various factors, such as register and frequency. Finally, the structure of the Modern Greek parasynthetic verbs (MGPVs) is explained in terms of optimal shape, constructional iconicity, transparency, and biuniqueness. The findings of this study are exemplified by reference to 308 parasynthetic verbs, collected from the Reverse Dictionary of Modern Greek (2002).

Hyman (2000, 2002) and Kiparsky (2011) have noted that Mirror-Principle-violating morpheme orders often give rise to non-local morphophonological effects. Kiparsky (2011) explicitly argues that this generalization cannot be captured in syntactic approaches to morphology, such as Distributed Morphology. This chapter shows that the generalization can be explained via the combination of two pre-existing tenets of such theories. One is the idea that Vocabulary Insertion proceeds from the most deeply embedded constituent outwards (Bobaljik 2000; Halle and Marantz 1993). The other is the proposal that violations of the Mirror Principle are to be accounted for via phrasal movement of a category containing the lexical root ‘stranding’ one or more affixes (Koopman 2005; Buell 2005; i.a.). The possibility of non-local phonological effects arises because the movements involved in deriving Mirror-Principle-violating orders lead to a disconnect between linear distance from the root and temporal order of Vocabulary Insertion.

This chapter distinguishes among different types of competing motivations and discusses examples of competing motivations in morphology on its own (notably affix order and markedness vs. frequency) or in its interface either with the lexicon or with discourse (in compounding) or with phonology (in morphonotactics). It also discusses competing motivations in the context of investigations of first language acquisition and experiments on the acquisition and processing of actual vs. potential vs. illegal German plurals. It is concluded by discussing the implications of the nature of competing motivation in these different types of research for the notion of competing motivation in general and for the manner in which the resolution of conflict in language research can lead to the advancement of knowledge in linguistics and psycholinguistics.