AUTHORS: Andrew Spencer and Ana R. Luís TITLE: Clitics SUBTITLE: An Introduction SERIES TITLE: Cambridge Textbook in Linguistics PUBLISHER: Cambridge University Press YEAR: 2012
REVIEWER: Marios Mavrogiorgos, Special Scientist, University of Cyprus
SUMMARY “Clitics, An Introduction” is a textbook on clitic systems within and across a large number of languages. The authors, Andrew Spencer and Ana R. Luís, offer a detailed presentation and discussion of the various clitic types, functions, and properties, of the various interactions of clitic properties at and across phonology, morphology, and syntax, illustrating why clitics are important for the deeper understanding of grammatical systems and of the models used to deal with such systems. They also present cross-linguistic similarities and differences among clitics, words and affixes, and the conclusions one may draw about their ontology as a separate grammatical category. Finally, they review the main theoretical approaches that have been proposed for the analysis of clitic systems. The major achievement of this quite impressive textbook is that, after so many years of copious research on clitics by numerous researchers (from various theoretical perspectives and on various languages), it presents the linguistic community with the first detailed and thorough introduction into this exciting but baffling topic, and argues for its importance for linguistic theory. It will be of use to students, lecturers, and researchers in theoretical and (possibly) applied linguistics.
Chapter 1 introduces the notion of clitic, and briefly introduces several issues that are further discussed later, such as how their special phonological, syntactic, and morphological properties pose important questions regarding their position in the grammar (words vs. affixes, but also atypical in various respects), or how distinct categorization criteria may lead to contradictory results. Their main goal, then, is to describe and evaluate the criteria proposed in the literature on various issues regarding clitics cross-linguistically, and to offer a clearer picture of what clitics are (or are not). Chapter 1 also includes information on the book, a short summary of each chapter, and a non-exhaustive list of valuable or influential works on clitics.
Chapter 2 surveys the various functions clitics may have, drawing examples from different languages. Typically, these functions are associated with functional content (on a par with function words and affixes) and they may be verbal, clausal, nominal, argument-related; and agreement marking-related. This obvious similarity with affixes and function words reflects the diachronic processes through which affixes arise from independent words via an intermediate clitic stage. Less frequently, clitics may have functions of conjunctions/prepositions, adverbs (especially locative and temporal ones), and discourse particles. The authors also discuss a number of clitic properties (e.g. weak phonology, or special syntax), whose status as reliable diagnostics for clitichood has been debated, and which will be a focus later.
Chapter 3 investigates different types of clitic systems. In the first sub-section, three influential studies of clitic systems are briefly presented (Wackernagel’s (1892) seminal monograph on Second Position (2P) clitics in Indo-European, Zwicky’s (1977) typology of clitics, with its two way distinction between simple clitics and special clitics, and Klavans’ (1985) clitic typology, which highlights the notion of clitics as phrasal affixes), along with their shortcomings and the important issues they raise. The second sub-section discusses clitic clusters and the highly idiosyncratic morphological templates they build, as well as the various patterns of special clitic placement that may be found cross-linguistically (including, among others, 2P, Tobler-Mussafia, verb oriented, and ditropic clitics). In the last sub-section, the authors look into domains of clitic placement, such as the clause as a whole, or the noun phrase.
Chapter 4 discusses interactions of clitics with the phonological component of the grammar. First, the authors argue that clitics do not have inherent prosodic prominence (a notion independent of stress). They also discuss cases where phonology seems to determine clitic cluster placement (e.g. Pashto), or the internal organization within a clitic cluster (e.g. Bulgarian li and ne). Section 4.4 illustrates interactions between clitics and the stress patterns of their host. Weak function words (instances of simple clitics) are discussed in 4.5, the main point being that many cases of simple clitics cannot be viewed just as weak variants of corresponding strong forms, on phonological, syntactic and morphological grounds. Finally, in 4.6, the authors briefly introduce basic components of Prosodic Structure theories, the problems clitics raise for such theories, and how phonologists have attempted to address these issues.
Chapter 5 deals with the morphological aspects of clitics. Sections 5.2 and 5.3 present the ‘famous’ criteria proposed by Zwicky and Pullum (1983) and their application on English ‘nt. These are heuristics on the assumption that morphology and syntax are separate components, and which may be used to categorize a formative as an affix or a clitic. Typically, use of these criteria only is inconclusive, as many instances of clitics behave more like affixes, and affixes may also behave like clitics. Section 5.4 deals with templatic morphology of clitic clusters. After discussing a fairly representative case of an inflectional cluster (using data from Nahuatl), they illustrate how more well known cases of clitic clusters (e.g. Italian, or Luiseño) display similar, typically affix-associated, properties (e.g. fixed order, or idiosyncratic allomorphy). Section 5.5 presents a further similarity between clitics and affixes, in that both may mark (via phrasal affixation or edge inflection respectively) a grammatical function that normally belongs to the head of the phrase they are attached to. In connection to this, the definitive accent in Tongan is also introduced, which involves the marking of a noun phrase as definite via stress shift (a non-affixal processual morphological operation) at the right edge of the definite phrase.
Chapter 6 treats the interaction of clitics with syntactic structure. First, syntactically manipulated (phonologically) weak function words may exhibit both clitic-like and affix-like properties, which illustrates the inherent difficulties when dealing with weak elements. Then, the morphosyntax of agreement is surveyed, with an emphasis on non-canonical agreement systems, which share properties with clitics (including, agreement markers in complementary distribution with certain noun phrases, and long distance agreement (e.g. Chukchee)). Section 6.4 critically examines three topics in the syntax of pronominal clitics: Northern Italian subject clitics, (optional or obligatory) clitic doubling, and clitic climbing. Clitics in these constructions (if not better analyzed as discourse function words -- e.g. vocalic subject clitics) typically share a similar morphosyntax with affixal systems (even though it may be atypical in certain respects), which undermines their analysis as syntactic clitics. Ethical datives are also briefly discussed, especially their idiosyncratic behavior in clusters. Section 6.5 demonstrates how complex the interaction between syntactic structures and clitics can be, using various Tagalog discourse particles, how such a degree of complexity and variability does not seem to be amenable to a small set of abstract syntactic principles, and how typical this is of clitic systems more generally.
Chapter 7 investigates whether one can define clitics, affixes, and words unambiguously. After presenting three cases of affixes which have erroneously been analyzed as clitics, i.e. German zu, Finnish possessor agreement markers, and Modern Greek verbal clitics (and which illustrate how misleading traditional or promising analyses may be in this respect), they point out, using examples from various languages, that it is not feasible to draw a clear line, since there are affixes with properties typical of clitics (e.g. stress neutrality, or variable ordering), and clitics with properties typical of lexemes (e.g. inflection, as in Czech pronominal and auxiliary clitics). The former raise problems for syntactic approaches, while the latter do the same for morphological approaches (although a plausible solution is offered). Still, words may also exhibit affixal properties (e.g. phonologically conditioned idiosyncratic allomorphy, as in the case of the French definite article in certain environments), in which case the relevant constructions could be analyzed as morphological idioms, or, alternatively, clitic-like properties (e.g. the Russian conditional marker by).
Chapter 8, the longest, surveys a number of selected theoretical approaches from different models, the issues they address, and the ways they deal with those issues. Data are drawn from typologically distinct languages. The authors first present four main clitic typologies that have been proposed in the literature: the tripartite pronominal typology by Cardinaletti & Starke (1999); the four-way word typology by Toivonen (2003); the bipartite typology of clitics as phrasal affixes by Anderson (2005); and the Canonical Typology of clitics by Spencer and Luís (2012). The authors’ general position is that clitics do not actually exist, although one may talk about clitic properties, and that the criteria used to differentiate between clitics and affixes, on the one hand, and simple vs. special clitics on the other, are not reliable. Sections 8.3-8.8 present the main theoretical clitic models, and some critical evaluation. These are: (a) phonological/prosodic; (b) morphological (including a brief introduction to realizational morphological models, and OT approaches, and a number of morphological analyses of clitic systems in various languages and morphological models (e.g. the Paradigm Function Morphology framework, or Distributed Morphology); (c) syntactic (divided into LFG/HPSG and PPT approaches). Sections 8.9-8.12 examine four issues regarding clitic morphosyntax that are particularly troublesome for PPT approaches: (a) 2P clitics, whose placement appears to be linked to both syntactic and prosodic factors; (b) Clitic Doubling in languages like Modern Greek, Macedonian, and Romanian, where doubling clitics are agreement-like elements; (c) Clitic Climbing, which has been used as evidence in favor of a movement approach to clitics, although climbing is also found with agreement systems, and although numerous analyses, both in LFG and HPSG, dispense with clitic movement (through complex predicate formation); Clitic Clusters, probably the most difficult problem PPT analyses have to face, as the mapping of complex, and mostly arbitrary ordering facts to general syntactic principles (even if a template, or some other post-syntactic morphological devise, is used) is extremely difficult, and possibly counter-intuitive. The authors argue that morphological approaches are not less valid in terms of explanatory adequacy, and that they are not refuted by any typological systematicity within the data.
Chapter 9 summarizes the main issues.
EVALUATION This textbook is very interesting and quite successful in the goals set out in Chapter 1, and in certain respects a noticeable achievement. First of all, in my knowledge it constitutes the first textbook which focuses on clitics as a natural language phenomenon viewed from different empirical, typological, and theoretical perspectives. Therefore, it offers a useful introduction into the notion and topic of clitics, and into the immense complexity this involves. Given the huge literature on clitics, and considerable confusion (due to reasons mentioned by the authors) regarding various facets of the phenomenon, a book that clarifies this interesting, but still unclear, picture is more than welcome. This, in turn, points to the challenges clitics still pose for grammatical theories and descriptions, and might indicate why they would be of special interest to any theory that looks into the interaction of grammatical levels, but also into the interaction of inherent language properties with more general cognitive/performance based procedures. This is mainly because clitics are atypical, seeming to straddle distinct, traditionally defined, grammatical levels, often giving the impression that in their case the mapping between these different levels is unpredictable or incongruent (in comparison to affixes or words). In this sense, it is interesting to see how or why the language faculty (and/or hypothetically some other related cognitive component, if relevant) categorizes a clitic in relation to syntactic, morphological, phonological and semantic properties, possibly in some cases even at a micro-construction level, and what this might tell us about language.
More particularly, the book is well organized and comprehensive in scope. It includes data from over 100 languages and quite many details (with a slight emphasis on the morphology of clitic systems, and its relation to other morphological systems), along with relevant sources, which could be extremely useful to linguists, especially those interested in the empirical and phenomenological aspects of clitics. The discussion of the theoretical approaches is also very useful, offering a detailed overview of the diverse ways cliticization has been analyzed within distinct theoretical frameworks as a general summary overview of a huge literature, and of the different ways one may look into this topic.
As a PPT linguist who advocates the Minimalist Program (Chomsky 1995), I was intrigued by the special emphasis on the morphological properties of clitics, and the impression that for the authors a morphological approach is probably more adequate in certain respects, mainly because the morphology cannot be explained by the syntax, but also because the syntactic evidence for cliticization is much less robust, if viewed within the more general picture. Of course, this position is understandable and possibly correct, given the authors’ theoretical stance and research, but also given the typologically distinct languages being discussed. Like much else, though, this may be a two-edged sword. In my view, the positive side lies in the observation that at least a good knowledge of other aspects of clitics (besides their syntactic and/or most common morphological/phonological ones) is valuable even to PPT/Minimalist researchers, as it would help them understand which aspects could be attributable to syntax, which to other levels, and which to the interface of syntax with these levels. Given the current major interest in interface phenomena (and more generally into the question of what counts as ‘grammar’), such a perspective could be useful even to students and/or researchers that do not view morphology as a separate grammatical component, and in particular to syntacticians and other linguists who might not be aware of this side to the story. This is a positive contribution to linguistics (including morphology), as it might promote a more balanced or integrative approach to the phenomenon.
Another positive aspect of this textbook is that it includes insights and observations that may have quite interesting implications for theory, and which possibly are not widely known to the general community. For example, the authors put forward the idea that sometimes the same clitic form, in the same position within a cluster, may have distinct properties in different constructions, which is a potentially important observation. Another point worth mentioning is that both clitic and affixal systems seem to share a number of properties (besides their morphological similarities): for instance, they may both realize or not realize certain (interpretable/inherent) semantic or pragmatic features, like definiteness or topic/focus. This empirical observation, which might be linked to other phenomena shared among affixes and clitics (e.g. the Person Case Constraint), raises equally important questions for both morphological and syntactic approaches.
There are also some shortcomings that should be mentioned. At certain points the discussion of the data includes too many details (especially morphological ones), some of which are not always directly relevant to the point under consideration. In other cases too little information or discussion is offered. This may lead to confusion, in that the reader may miss the thread of the argument, or may not understand what is argued for. Of course, this might be more of a problem for someone reading a whole chapter, or more than one chapter, but less so if one is only using the book as a reference source for certain phenomena.
A second issue regards the degree to which this survey is, can be, or should be theory-neutral (and to some extent, the meta-language around ‘theory neutral’). Although it is true that every theory will eventually have to account in one way or another for all (if possible) aspects of clitics, even though it may choose to focus first on certain issues only, it also remains a fact that various aspects of data handling (including collection, categorization, description, etc.) presume basic (theoretical and/or empirical) premises or assumptions, which may lead us to see certain patterns (but not others) in the data, and hence to various types of problems and solutions (according to standard assumptions regarding the interaction between a theoretical model and hypothesis formulation and testing). In the present case, the theoretical assumption (based on a theoretical model of some kind, which is also influenced by empirical observations) is that word formation and phrase formation are essentially distinct, in that despite their similarities and differences, they are handled by separate grammatical components, which interact with each other in certain ways and order (depending on the framework). This basic premise unavoidably influences the way one looks at the data (namely, clitics lie somewhere between words and affixes), and this is evident also in the textbook: some of the cases discussed, as noted by the authors, would not even be a problem if a different premise was assumed, or alternatively a different solution might be possible (of course, this does not imply that this solution would necessarily be better, or that the opposite does not hold). If this is made explicit to the reader in a clear way (and I am not convinced that this is the case here, at least to some extent), it would be a perfectly valid choice, even for an introductory textbook.
A further related issue is the impression one gets while reading the book that the authors consistently try to ‘promote’ the morphological analysis of clitics. This may take different, though very subtle forms, as for example when syntactic analyses are typically discussed in terms of problems (as opposed to morphological ones, which have both problems and merits), when syntactic approaches are considered in a more positive light if they analyze clitics as agreement markers (for instance), or when certain syntactic properties are not fully integrated into the picture. More generally, although, on a macro-level, both syntactic and morphological approaches are described in both positive and negative light, on the micro-level the approach is less balanced. This creates a sort of confusion, especially because at least in some cases other solutions are at least equally adequate. Still, this is again understandable to an important extent, as mirroring the authors’ stance. For me, one striking example was the analysis of Greek verbal clitics in chapter 7. There, the authors present only certain morphological aspects of the phenomenon (some of which are definitely important, although others do not tell us much, or may even be irrelevant or marginal), whereas there is a dearth of evidence in the literature that Greek clitics also have important syntactic effects (e.g. Anagnostopoulou 2003, Mavrogiorgos 2010, and references therein). Even if these could be accounted for in a purely morphological approach, one’s choices would have important repercussions on the rest of the system (e.g. the nature of agreement in syntax and morphology, given the properties of clitic doubling/resumption in Modern Greek). Although such discussion may not be relevant for a textbook, the impression that may follow from the discussion (and of the evaluative approach entertained) in this particular context may nevertheless be important.
Despite all these issues, I firmly believe that this is a very good and important textbook, which will prove useful and challenging to students and researchers alike who are interested in clitic systems (and related affixal systems) across languages.
REFERENCES Anagnostopoulou, Elena. 2003. The syntax of ditransitives. Evidence from Clitics. Berlin: Mouton de Gruyter.
Anderson, Stephen R. 2005. Aspects of the Theory of Clitics. Oxford: Oxford University Press.
Cardinaletti, Anna & Michal Starke. 1999. The typology of structural deficiency: a case study of the three classes of pronouns. In van Riemsdijk, Hendrik C. (ed.), Clitics and the Languages of Europe. Berlin: Mouton de Gruyter, pp. 145-233.
Chomsky, Noam. 1995. The Minimalist Program. Cambridge, MA: The MIT Press.
Klavans, Judith L. 1985. The independence of syntax and phonology in cliticization. Language 61. 95-120.
Mavrogiorgos, Marios. 2010. Clitics in Greek. A Minimalist Account of Proclisis and Enclisis. Amsterdam: John Benjamins.
Spencer, Andrew & Ana Luís. 2012. The canonical clitic. In Brown, Dunstan P., Chumakina, Marina, and Corbett, Greville C. (eds.), Canonical Morphology and Syntax. Oxford: Oxford University Press.
Toivonen, Ida. 2003. Non-Projecting Words: A Case Study of Swedish Particles. Dordrecht: Kluwer.
Wackernagel, Jakob. 1982. Über ein Gesetz der indogermanischen Wortstellung. Indogermanische Forschungen 1. 333-346.
Zwicky, Arnold M. 1977. On Clitics. Bloomington, IN: Indiana University Linguistics Club.
Zwicky, Arnonld M. & Geoffrey K. Pullum. 1983. Cliticization vs. inflection: English n’t. Language 59. 502-513.
ABOUT THE REVIEWER:
Dr. Marios Mavrogiorgos works in English Studies at the University of Cyprus, specializing in clitics and other related morphosyntactic phenomena in Greek and other languages within the Minimalist Framework. He has published a number of papers (alone and in collaboration), and a monograph (with John Benjamins Publishing).