Introduction: The Scope of the Handbook
Abstract and Keywords
This chapter defines the term ‘derivation’ and sets out the scope of the phenomena to be covered in the handbook. It gives a brief overview of the formal and semantic categories that fall under derivation and of the theoretical treatment of derivation over the last fifty years. It also sets out the organization of the handbook.
1.1 Why Derivation on its Own?
This handbook is intended as a companion to our earlier Oxford Handbook of Compounding (2009), as well as to the Oxford Handbook of Inflection (Baerman in press), and the Oxford Handbook of Morphological Theory (Audring and Masini forthcoming). We might justify it simply on the basis of symmetry, as part of an effort to cover all areas of the study of morphology thoroughly in this series. Nevertheless, we ought to have a better reason in mind for compiling a book of this sort. In this Handbook we hope to look at derivational morphology on its own terms to see what is distinctive about it, what defines it as a phenomenon, and how it is manifested in the languages of the world.
What do we mean by “derivation on its own terms”? To determine this, we must start first with defining what we mean by word formation. The term “word formation” refers to the creation of new lexemes in a language and is generally said to be composed of compounding and derivation. By “derivation” we therefore mean to refer to those parts of word formation other than compounding, a definition that is also used by Aikhenvald (2007: 1). Although the definition of “compounding” is by no means straightforward, we have dealt with it extensively in our Introduction to the Oxford Handbook of Compounding. For our purposes here, it is sufficient to make use of Bauer’s (2003: 40) definition of a compound as “the formation of a new lexeme by adjoining two or more lexemes.”1 What we are left with when we subtract compounding from word formation are ways of creating new lexemes other than putting two or more lexemes together. In formal terms, this encompasses various kinds of affixation (prefixation, suffixation, infixation, circumfixation), but also (p. 4) reduplication, templatic or root and pattern word formation, subtractive word formation, conversion, and miscellaneous tone and stress changing operations, specifically when they are not used for the purposes of inflection.
Approached from the perspective of function or semantics, we might define the core of derivation as including, but not limited to the creation of:
• event, process, and result lexemes;
• personals, including agent and patient;
• lexemes expressing non-inflectional gender (e.g. actress);
• lexemes expressing location in time and space;
• collectives and abstracts;
• evaluatives (including both size and attitude);
• negatives and privatives;
• lexemes relating to non-evaluative size and quantity;
• causatives, anti-causatives, applicatives, factitives, inchoatives, duratives, and the like.
Derivation may be either category-changing, or non-category-changing; for example, personal nouns may be formed from verbs (writer, accountant) but also from other nouns (Londoner, pianist). Verbs can be created from nouns or adjectives (unionize, civilize), or can be formed from other verbs, such as the causatives and applicatives that are typical of the Niger-Congo languages (Creissels, this volume). There are no doubt many other semantic categories into which derivation can fall, especially if we take into account the sort of lexical derivation that is to be found in polysynthetic languages, such as those of the Athabaskan (Rice, this volume) or Eskimo-Aleut languages (Johns, this volume). Indeed, some semantic categories can be quite idiosyncratic, as is the case with the suffix -ier in French, which creates names of trees from names of the respective fruit (poire ‘pear’ ~ poirier ‘pear tree’).
It would be convenient, of course, if we could take the intersection of these formal and functional categories and be left with a clearly delineated domain of derivation as the subject of this handbook. But language is rarely so obliging and we must acknowledge that on all sides we are faced with fuzzy boundaries. In some cases there is difficulty separating derivation from compounding. As Olsen (this volume) points out, identifying the point at which an independent lexeme becomes an affix is almost impossible to do. Or consider the case of reduplication. Some authors (e.g. Štekauer et al. 2012) treat full reduplication as a form of compounding, apart from partial reduplication; there is something to be said for this choice, as full reduplication certainly does fulfill the main criterion of compounding as being the composition of two lexemes. Still, others (Inkelas, this volume) find the most salient characteristic of reduplication—repetition—sufficient to treat full reduplication as a phenomenon distinct from compounding. On the other side, there are cases where the boundaries between derivation and inflection are indistinct, as with evaluatives in languages that have extensive noun class systems, (p. 5) with certain classes being reserved for diminutives or augmentatives (see Creissels, this volume). Indeed, the puzzling nature of evaluatives has led some researchers to treat it as distinct from either derivation or inflection (see Körtvélyessy, this volume).
In spite of difficulties of this sort, the present volume is predicated on the assumption that there is something in the intersection between the formal means and the functional/semantic territory covered by derivation that defines a coherent field of study. Is this the case? Oddly, this is a question that does not seem to have been asked. One reason for this is that derivation has only rarely been treated apart from other sorts of morphology—compounding on the one hand and inflection on the other.
1.2 A Brief Foray into History
We do not mean to dwell on the historical development of the field of morphology, as this is a subject that has already been covered in our Handbook of Word Formation (2005) and is to be the subject of The Oxford Handbook of Morphological Theory (Audring and Masini forthcoming). But at least a brief mention of the treatment of derivation in morphological theory seems justified here. Seminal works in the American structuralist tradition, such as Harris (1946) or Hockett (1947, 1954) were preoccupied with methods of analyzing morphemes, and do not seem to provide separate treatments of inflection and derivation.2 Nor do some of the key works in morphology from the middle of the 20th century single out derivation as a distinct matter for study. Lees’ The Grammar of English Nominalizations (1960) represents early work characteristic of the generative tradition in North America. Lees focuses primarily on noun-noun compounds, but also assumes that transformations of various sorts can introduce category-changing derivational morphology, in particular affixes that nominalize verbs in English. Marchand’s The Categories and Types of Present-day English Word-Formation (1960/9) is representative of the mid-century view on word formation in Western Europe. The scope of Marchand’s work, drawing mainly on the structuralist tradition of the Geneva School and the ideas of Coseriu (1952), is much broader, covering a wide range of word-formation processes in English derivation and compounding. Dokulil’s Tvoření slov v češtině I. Teorie odvozování slov [Word-Formation in Czech. A Theory of Word Derivation] (1962) is representative of the field in Central Europe. His is the most comprehensive theory from among the authors of the 1960s.3 Dokulil discusses and foreshadows a number of topics which have become central to the field of derivational morphology, including a general onomasiological theory of word formation, individual word-formation processes and cognitive foundations of these processes, productivity, (p. 6) derivational paradigms, and lexicalization, among others. His work continues to be of influence among morphologists in Central Europe.
Subsequent work has only rarely singled out derivation from compounding and inflection. Indeed, Aronoff’s Word Formation in Generative Grammar (1976) seems to be the lone example.4 Aronoff is careful to distinguish derivation from inflection, the latter being a matter of syntax: he mentions in passing that unlike derivational morphemes, inflectional morphemes may be attached higher in a tree than the X0 node (1976: 2). He does not treat compounding, but interestingly does not comment on the decision to exclude compounding from the scope of his monograph. In other words, Aronoff’s decision to discuss derivation apart from inflection and compounding does not seem to be a principled one or to have any particular theoretical significance.
Subsequent work on morphology has generally been inclusive, encompassing derivation and either compounding or inflection or both. Important dissertations such as Siegel’s (1974) Topics in English Morphology, Allen’s (1978) Morphological Investigations, and Lieber’s (1980) On the Organization of the Lexicon all cover parts of the territory of morphology beyond derivation, as does subsequent influential work in word structure (Williams 1981b, Selkirk 1982, Lieber 1992), in Lexical Phonology and Morphology (Kiparsky 1982b, Halle and Mohanan 1985, Giegerich 1999), in realizational frameworks (Anderson 1992, Stump 2001), in Lexeme Morpheme Base Morphology (Beard 1995), in the onomasiological tradition (Štekauer 1998, 2005), or in the framework of lexical semantics (Lieber 2004).
Those works over the last thirty or so years that have treated derivation have tended to be focused on specific theoretical issues, for example the formal nature of rules (Aronoff 1976, Lieber 1980, 1992, Selkirk 1982, Beard 1995, Booij 2010, to name just a few), productivity (Aronoff 1976, van Marle 1985, Baayen 1989, Plag 1999, Bauer 2001), affix ordering (Fabb 1988, Hay 2000, Plag and Baayen 2009), lexicalization (Kastovsky 1982, Bauer 1983, Lipka et al. 2004), the nature of evaluative affixation (Scalise 1984, Stump 1993, Bauer 1996, 1997a, Jurafsky 1996, Grandi and Körtvelyessy forthcoming), the analysis of root-and-pattern word formation (McCarthy 1979), reduplication (Moravcsik 1978, Marantz 1982, Hurch 2005, Inkelas and Zoll 2005), and infixation (Ultan 1975, Yu 2007a). But no one seems to have taken a broad view of the subject.
1.3 A Comprehensive Overview
The chapters of this handbook thus give us a chance to ask what is distinctive about derivation. Our idea is to fill in a picture that is fragmented and currently missing key pieces. Although we have theoretical treatments of derivation, we lack a comprehensive overview that encompasses both concatenative and non-concatenative formal processes on the one (p. 7) hand, and various semantic categories of derivation on the other. Further, there are surprisingly few substantial descriptive accounts of derivation in the languages of the world that allow us to make cross-linguistic comparisons; grammars of specific languages often do not have more than a few pages on derivation, and language families are almost never treated as a whole. Štekauer et al. (2012) is a step in the direction of filling in descriptive gaps, but they present isolated facts about many languages rather than focused snapshots of languages and language families. The present handbook seeks to fill this descriptive gap.
We also believe that a cross-linguistic perspective on derivation has been hampered by a view that might be too heavily Eurocentric. We give two examples. Consider the term “conversion.” This term for category change with no concomitant change in form makes sense in the context of languages like English; but it becomes increasingly problematic when we consider languages that are heavily inflected and even more so with languages that do not exhibilt clear distinctions between syntactic categories (see Valera, this volume). A second example of a Eurocentric perspective might be the common notion that the formation of ideophones is not to be treated as part of derivation; current English-language textbooks on morphology (Spencer 1991, Haspelmath 2002, Booij 2007, Lieber 2010a, Aronoff and Fudeman 2011) do not even mention ideophones in the context of derivation. But the chapters in this volume on derivation in Uralic, Niger-Congo, Nilo-Saharan, and Sino-Tibetan all suggest that our view has been too narrow. In each of these families ideophones have a role to play in derivation.
Interestingly, one thing that has emerged from Štekauer et al.’s (2012) recent typological work is that it seems to be an absolute universal that languages have some sort of derivation, and this alone would justify our focus on this phenomenon. Štekauer et al. cite one language (Vietnamese) in their sample of fifty-five languages that lacks affixation, but significantly Vietnamese does not lack derivation entirely, as new lexemes in that language may be formed by various sorts of reduplication (see also Inkelas, this volume). In contrast, they cite five languages that lack compounding (Dangaléat, Diola-Fogny, Karao, Kwakw’ala, and West Greenlandic (Kalaallisut)), but that do have various formal mechanisms of derivation. The literature also suggests that some languages (Thai, Burmese, Yoruba, Vietnamese) lack inflection (Lehmann and Moravcsik 2000: 745), which would leave derivation as the only sort of morphology that all languages may be said to have.5 Of course this makes sense from a functional perspective: all languages need to add to their lexical stock somehow, and relying exclusively on coinage and borrowing to increase lexical stock seems implausible at best.6
Looking more closely at derivation, several researchers have concluded that suffixation is the most common means of derivation in the languages of the world (Hawkins and Gilligan 1988, Štekauer et al. 2012); only one affixing language in the Štekauer et al. (p. 8) sample, Yoruba, lacks suffixation as a derivational device. Prefixation is somewhat less well-attested, although still widespread (70.91% in the Štekauer et al. sample), as are reduplication (80% in Štekauer et al., but closer to 75% in the WALS sample) and conversion (61.82% of languages in Štekauer et al.). Other forms of derivation are not nearly so widespread: Štekauer et al. say that 25.45% of the languages in their sample exhibit infixation, 21.82% circumfixation, and 23.64% stem vowel alternation (which for them includes both ablaut and root and pattern derivation). Other sorts of derivation appear in an even smaller percentage of the languages they sampled.
We therefore have some very basic knowledge of the formal, functional, and typological characteristics of derivation, but this is a bare skeleton. We intend with this Handbook to begin to fill in details in all these areas. It is our intention that the chapters gathered in this volume will be of use not only to morphologists, but also to psycholinguists, historical linguists, syntacticians, and phonologists, as well as to students and scholars in related fields that need to know about how languages add to their lexical stocks.
1.4 The Organization of the Handbook
In the first part of this Handbook, we look at derivation from several perspectives. We begin with boundary issues—where to draw the line between derivation and inflection (Chapter 2) and between derivation and compounding (Chapter 3). Not surprisingly, this brings to the fore the difficulty of delineating our subject matter with perfect clarity. We next take up several “big-picture” issues including the theoretical treatment of derivation (Chapter 4), the issue of productivity and lexicalization (Chapter 5), methodologies used in obtaining data on derivation (Chapter 6), and experimental and psycholinguistic approaches to derivation (Chapter 7). Chapters 8–12 look at particular formal means of derivation (affixation, infixation, conversion, reduplication, and other non-concatenative processes). Chapter 13 looks at issues concerning allomorphy in derivation. Next, we take up derivation of nouns (Chapter 14), verbs (Chapter 15), adjectives and adverbs (Chapter 16), evaluative derivation (Chapter 17), and derivation of functional categories (Chapter 18). We also consider a number of themes that are particularly salient in the study of derivation: homophony versus polysemy in affixes (Chapter 19), paradigmatic organization in derivation (Chapter 20), and the ordering of derivational affixes (Chapter 21). Part I ends with three chapters situating derivation with respect to the wider fields of sociolinguistics, language change, and child language acquisition (Chapters 22–4).
In the second part of this volume (Chapters 25–39) we have made an attempt to fill a descriptive gap in the literature by looking at derivation across a wide range of languages. Instead of focusing on individual languages as we did in the Oxford Handbook of Compounding, however, we decided here to look more broadly at language families with the aim of exploring the extent of variation both within and across families. As is usually (p. 9) the case in surveys of this kind, we aimed for a broad distribution of families in terms of areal and typological characteristics. Inevitably, of course, we were limited to families for which we could find willing authors. We were extraordinarily fortunate, however, in finding authors able to cover fifteen language families, ranging geographically across Europe, Eurasia, East and South Asia, Australia, the Pacific, Africa, North and South America. The reader will note that these chapters are not uniform in composition; this was inevitable, given a very wide range in the size of the language families and in the availability of data. Some chapters range broadly over many languages in the family; others give a brief overview of the family and then concentrate on one or two specific languages in the family. Chapter 34 is unique in that we could find no single author to take on all of Sino-Tibetan; this chapter is therefore divided into three sections, each covering a major branch of Sino-Tibetan. We hope that in spite of their differences in composition, these chapters nevertheless give a usefully broad overview of the range of derivation that occurs in the languages of the world.
In the last two chapters we return to broader themes. The penultimate chapter of the handbook takes an areal rather than genetic view of derivation, looking both at the mechanisms of areal spread and specific examples of areal tendencies in derivation. And in the final chapter we return to the theme of universals, assessing what the chapters of Part II of this volume can tell us about various cross-linguistic generalizations that have appeared in the literature.
We close with a word on what we have not provided in this Handbook, namely a comprehensive overview of the theoretical frameworks in which derivation has been treated. This omission was a deliberate decision on our part. On the one hand, we have already published a Handbook of Word Formation (2005) that covers a number of theoretical approaches to word formation. On the other, the Oxford Handbook of Morphological Theory (Audring and Masini forthcoming) will cover recent theoretical developments. What we hope to provide in what follows is a rich picture of how word formation works, what sorts of meanings it tends to express, how it may be studied, and how it is manifested in the languages of the world. Inevitably there will be many facets of derivation we have failed to cover adequately. Nevertheless, we hope to have provided a broad enough overview of the state-of-the-art to aid further research in the field.
(1) We remain neutral on whether noun-incorporation is to be treated as a sort of compounding or as a matter of syntax. We assume, however, that it is not to be included as a part of derivation.
(3) Unfortunately, his publications were not written in English, so they have had limited influence in North America or Western Europe.
(5) Greenberg (1963a) proposed the universal that “If a language has inflection, it always has derivation” (Universals Archive U506). It appears that this universal can be strengthened in light of the results we cite here: if we are correct, all languages have some sort of derivation whether or not they have inflection.
(6) Adding to the lexical stock exclusively by borrowing may be a feature of dying languages, but is not a feature of any living language to our knowledge.