1 Defining Case
The category of case belongs to the stock of traditional grammatical categories; morphological case is familiar from languages which informed linguistic theory (like Ancient Greek, illustrated in Table 1, and Latin, illustrated in (2)), and the definition of case seems uncontroversial. Blake (1994, 2001), on the first page of his classic textbook on case, defines case as “a system of marking dependent nouns for the type of relationship they bear to their heads” (Blake, 2001, p. 1). Similar definitions have been proposed by other authors. For example, Moravcsik (2009, p. 231) defines cases (case markers) as “a formal device associated with a noun phrase that signals the grammatical role of that noun phrase.”
Table 1. Declension of anthrōpos ‘man’ in Ancient Greek
Source: Blake (2009, p. 14).
In spite of the wide-ranging consensus, the analyses of individual languages differ widely with respect to the number of cases acknowledged; it may be even controversial whether language L has case value at all. For example, does Abkhaz have declension, if “predicative case” is the only overt case marker in a putative paradigm (Hewitt, 1979)? Do locative suffixes in Algonquian languages counts as cases, or are they rather derivational suffixes (Iggesen, 2005)? Does Bulgarian vocative vs. nominative opposition on nouns qualify as a case paradigm, and more generally is vocative a case at all (see Daniel and Spencer, 2009)? Is it possible for a language to have huge case systems numbering more than a hundred cases,1 as has been claimed for some Daghestanian languages, or are these cases misanalyzed (see Comrie & Polinsky, 1998, on “Daghestanian case hoax”; see also Section “Structural and Semantic Case”). Notably, similar problems arise for better described languages as well. Thus, views vary widely, ranging from 6 to 10 cases, as to how many cases there are in Russian (Corbett, 2008). And while most Hungarian grammars present a case paradigm including over a dozen cases, Spencer (2008) doubts that Hungarian has a case feature at all.
One obvious reason why case can be contested for individual languages is that not all languages have case as an inflectional category, so inflectional case is just a subset of the phenomena discussed by Blake and Moravcsik. The distinction may seem obvious, but in fact a distinction between case affixes and clitics (or particles) is problematic, when applied cross-linguistically (see Iggesen, 2005). For example, it is debatable whether case markers in Manchu should be regarded as suffixes or clitics (particles), as neither vowel harmony nor other diagnostics qualify them unambiguously as either bound or free forms, but rather suggest an intermediate status (Tsumagari, 2006). This is related to the more general problem of word segmentation cross-linguistically, and some authors like Haspelmath (2011) question the typological validity of the distinction between affixes and clitics altogether. However, even adopting a broad view on case, which is not restricted to inflectional case, does not resolve the controversies due to the fact that many notions that enter into a definition of case, once they are spelled out, allow for different interpretations. For example, Moravcsik (2009), after offering her definition of case, feels obliged to explain how “formal device,” “grammatical roles,” and “noun phrases” entering the definition are to be understood.
Therefore, many scholars advocate a multifactorial definition of case and other grammatical categories. For example, Moravcsik (2009) proposes a number of criteria that characterize distribution of prototypical case markers, including the following:
Other principles require that case markers be associated with only those noun phrases that have a grammatical role, and only with those NPs—that NPs with different roles be signaled by different markers and those with the same role be signaled by the same marker. Some further principles pertain more specifically to linear order, stating, in particular, that all cases have a single slot within the noun phrase and that case markers are adjacent to NPs they are associated with.
These properties are expected to hold for prototypical cases, but exceptions are not uncommon. The most common counterexample to (1) is the phenomenon of case concord, as illustrated by the familiar Latin examples, adopted from (Moravcsik, 2009, p. 238):
These counterexamples as such do not challenge the definition of case if they are understood as contributing the prototype definitions. And those that do show a deviation may receive an alternative interpretation. Indeed, some authors analyze concordial case as a category distinct from governed case (Mel’čuk, 1998).
Another, less common violation is when the same nominal takes multiple case markers. The term ‘multiple case marking’ may pertain to different phenomena (see Section “Structural and Semantic Case” on Daghestanian languages), but one relatively common pattern of double case marking involves cases when a noun bearing an adnominal genitive can take further case marking (Plank, 1995). The following example from Old Georgian illustrates this:
The pattern of double case marking clearly violates the principle in (1), if both categories involved are conceded to represent cases (recall that in Mel’čuk’s approach ‘concordial case’ is a distinct category); otherwise it can be explained as representing dependency with two different heads, the verb and the head noun in a possessive construction (Moravcsik, 2009).
Occasionally, one also finds instances where a single case appears to mark two different syntactic relations. This is arguably the case of designative case in Tungusic, which in addition to assigning the theme (patient, result) role to its host nominal (direct object), assigns a beneficiary interpretation to its possessor (Malchukov, 2009). Compare the following two constructions, with the object being marked with the accusative case in (4a) and designative case in (4b):
In the former example (4a) the possessor of the noun phrase is not associated with any particular role (it is more likely to be interpreted as a source argument ‘took the sledge from you’, but allows also for other interpretations), but if one replaces the accusative case with a designative, the possessor in the noun phrase is invariably interpreted as a beneficiary. Such instances when cases are associated with two noun phrases clearly deviate from the case prototype and, in fact, are also amenable for alternative interpretations (for example, a destinative marker in Samoyedic languages with a similar function is sometimes analyzed as a marker of future possession or even as a nominal tense marker; Nikolaeva, 2015).
Within his general program of canonical typology, Corbett (2008, 2012) provides a definition of a canonical case, which relies on two general principles defining the concept of a morphosyntactic feature. The principles require that “[f]eatures and their values are clearly distinguished by formal means” and that “[t]he use of canonical morphosyntactic features and their values is determined by simple syntactic rules” (Corbett, 2008, pp. 6, 10). Each of the principles has several corollaries providing diagnostics for canonical morphosyntactic features, including what counts as a ‘canonical case.’ On the morphological side, for example, it is required that a canonical case be consistently coded by a dedicated form, which is moreover consistently coded across lexical classes (for which case is licensed), as well as for the members of respective classes. But of course these properties do not always hold; thus, not all nominals can express the same case values (see discussion of split ergativity in Section “Morphological and Syntactic Case”), and many languages have at least a few indeclinables. Finally, in many languages of the inflectional type, not all cases have dedicated forms; in the literature these are referred to as “non-autonomous cases” (Corbett, 2008, following Zaliznjak, 1973; Mel’čuk, 1986, among others). The problem of non-autonomous cases is of theoretical importance, but also has practical repercussions for determining the number of case markers in individual languages, leading to sometimes long-lasting controversies. Corbett (2008) provides a detailed discussion of controversies concerning the number of cases in Russian, where some of the non-autonomous cases are contested. In fact, on a closer inspection, also bona fide cases may turn out as non-canonical. Only in the -a declension class do we find a dedicated form of the accusative (knig-u ‘book’ is accusative singular); in all other classes the accusative is expressed by forms syncretic with the nominative (stol ‘table’ is nominative or accusative singular) or genitive (student-a ‘student’ is genitive or accusative singular). For this reason, accusative is close to non-autonomous case (Zaliznjak, 1973, pp. 74–75) and does not fully comply with Corbett’s criteria for canonical case (Corbett, 2008, p. 7). And some other cases, like the “second genitive” (the partitive genitive) acknowledged by Jakobson (1958), are unmistakably non-autonomous, as their form is syncretic to dative (see stakan čaju ‘a glass of tea’; -u is a regular dative marker). Examples like this are part of a larger problem which will be discussed in the next section under the heading of distributional vs. form-based approaches to identification of cases.
2 Morphological and Syntactic Case
Another point of controversy is a conception of case as a morphological or a morphosyntactic feature. While the syntactic significance of case is generally beyond doubt, it has been realized only recently that describing case as a morphosyntactic feature requires special argumentation. One proposal of this sort is found in Beard (1995) and Spencer (2008), who argue that a [Case] attribute is sanctioned in a formal description if and only if that attribute is needed to generalize across cases independently of form—thus, a case feature as motivated either by declensional mismatches violating one-to-one correspondences between morphology and syntax or else by syntactic rules making reference to case, such as case agreement. Thus, it can be shown, in reference to ‘Beard’s Criterion’ (Spencer & Otoguro, 2005), that in Russian case is a morphosyntactic category. Consider the following examples that show cases of mismatch between an adjective and a noun when belonging to two different declensional classes:
Most masculine animate nouns (such as čelovek ‘man; human’ in (5a)) of the default masculine consonant-final declension class take an accusative which is identical to the genitive, and adjectives agreeing with these nouns show the same case syncretism (see (5a)). A few masculine animate nouns such as mužčina ‘man’ belong exceptionally to the -a declension class and have a dedicated accusative form; still the adjective shows the same genitive form as in the former pattern (see (5b)). Clearly the latter pattern defies generalization in purely morphological terms and needs to take recourse to a syntactic feature of (accusative) case, which may have different morphological manifestations depending on the type of the host nominal (see Spencer, 2009, for further discussion).
While this reasoning is rather straightforward, the offshoot of this approach is paradoxical: it implies that cases in agglutinating languages where case expression is regular (ignoring phonologically conditioned allomorphy) and case agreement is lacking do not qualify as having case. Indeed this is the conclusion Spencer (2008) reaches for Hungarian; he argues that the grammar of Hungarian does not need a case feature, since all the rules can be formulated in terms of morphological exponents. This conclusion is, of course, warranted only if case is understood strictly as a morphosyntactic feature (i.e., a feature relevant to both the morphological and the syntactic components of the language), rather than a morphological or morphosemantic feature (see Corbett, 2012, on the latter concept). It should also be noted that even subscribing to the concept of case as a morphosyntactic feature, authors can reach quite different conclusions about particular languages, depending on their understanding of syntactic relevance. Thus, generative grammarians allude to a wide array of syntactic facts far beyond the abovementioned agreement phenomena to motivate the concept of abstract (syntactic) case as distinct from morphological case.2
While most approaches agree that we should make a distinction between syntactic and morphological case (or s-case and m-case, in terms of Spencer, 2009),3 and the two may not match, views differ whether case variation in individual constructions should be regarded as a morphological or syntactic matter (variation of m-cases instantiating the same s-case, or variation of s-cases). For example, the much discussed phenomenon of differential object marking (DOM) has been offered both interpretations in the literature. A classic case of differential object marking is provided by languages such as Hindi or Turkish, where an object bears an overt case depending on features such as specificity, definiteness, animacy, and topicality (Aissen, 2003; Bossong, 1985). For example, in Hindi, an animate object should be marked (see (6a)), while marking of inanimate object depends on further properties such as definiteness (see (6b)):
Such patterns have been interpreted as either variation in case marking (ACC case being restricted to more prominent nouns) or else relegated to a morphological matter (see discussion in optimality theoretic literature between Aissen, 2003, and Woolford, 2006).
Similar debates have been waged in linguistic typology, particularly in the context of Australian noun/pronoun systems (see contributions to Plank, 1991). As is well known, in many of these languages the lexical nouns exhibit ergative-absolutive patterning, while the pronouns show nominative-accusative patterning. In one approach, adopted by Silverstein (1976) and much of the literature on aboriginal Australian languages (e.g., Dixon, 1994), such systems are described as split ergative: the noun forms have an ergative-absolutive distribution, while the pronominals have nominative-accusative distribution, and so these are the cases those word forms are ‘in.’ An alternative approach, first proposed in the context of Australian languages by Goddard (1982), but traditional for the description of European languages, treats a discrepancy between nouns and pronouns in terms of case syncretism/allomorphy. On Goddard’s approach we assume three cases: ergative, nominative-absolutive and accusative. In nouns the nom-abs case syncretizes with the accusative, while in pronouns the ergative and the nom-abs cases syncretize. Hence, there are three structural s-cases (defined distributionally), although for any given nominal there are only two m-case forms for the three grammatical functions (A, S, and P). In Goddard’s approach the relation between m-cases and s-cases is complicated. In the split-ergativity approach the relation between case marking and syntax is complicated.
This controversy, frequently referred to under the heading of distributional vs. form-based approaches to identification of cases, remains unresolved. Blake (2001, pp. 19–29) is noncommittal in that respect, noting that the two approaches derive partly from different traditions and partly from the differences in the data addressed. For inflecting languages (Latin, Russian) that have numerous paradigms with case syncretisms and where each case is used in a number of functions, we can only capture the distributional facts by appeal to s-cases (‘distributional’ account). On the other hand, if a language has case markers that are morphologically very regular, with no inflectional classes, then it will often be possible, in principle, at least, to state regularities over the form classes themselves (‘formal’ account).
The problem can be arguably elucidated from a diachronic perspective; indeed, what starts as a syntactic alternation may over time become morphologized. Baerman (2009, p. 229), who generally adopts a distributional approach, also concedes that “the most widespread type of case syncretism, that of the core cases, may in many instances represent the outcome of desyntacticization, that is, the morphologized relic of what was once an active syntactic rule.”
3 Structural and Semantic Case
A distinction is often made between more abstract cases expressing core syntactic relations, such as subject and object, and more concrete cases that express various specific semantic roles, especially spatial relationships (see Blake, 1994, pp. 32–34). These have been referred as grammatical vs. semantic cases, alternatively grammatical cases are referred to as syntactic or structural, and semantic cases as “concrete” or oblique cases (see Haspelmath, 2009, for discussion of terminology). A somewhat similar distinction is made in generative tradition under the rubric of structural vs. inherent case. While the distinction as such seems to be clear, there are reasons to view this opposition as scalar rather than dichotomous. First, the distinction between semantic and syntactic cases is not clear-cut since the same case may encode both sorts of information (Butt, 2006, pp. 142–147). For the same reason, Mel’čuk (1998, p. 328) talks about semantic vs. syntactic uses of cases rather than semantic vs. syntactic case dichotomy. Second, the distinction is problematic and is debated with respect to certain cases; in particular, the dative has been alternatively analyzed as either a structural or an oblique case (see, e.g., Vogel & Steinbach, 1998, for German). The same discussion pertains to the status of ergative cases, which some authors treat as a syntactic (structural) case (Baker, 2015; Bobaljik, 2008), while others treat it as an inherent case (Legate, 2014; Woolford, 2006). Third, to complicate the matter further, in some theories such as role and reference grammar (Van Valin, 2005, assignment of core cases (nominative, accusative, ergative, absolutive) is also handled through appeal to the semantic roles (admittedly, generalized ones, macro-roles) actor and undergoer.
The uncertainties concerning delimitation of structural and semantic cases are related to another point of controversy related to a general function of (core) case markers. In typology this debate has been waged since the 1980s under the heading of discriminating vs. indexing approach (Comrie, 1989; Kibrik, 1985; Mallinson & Blake, 1981).4 According to the indexing view, cases are used to encode a semantic role of an argument or signal its features (animacy, definiteness, etc.). According to the discriminatory view, case marking is rather motivated by the need to distinguish between the core arguments (subjects and objects). Although initially seen as alternative and even competing motivations, both are now seen as necessary ingredients in order to account for the cross-linguistic variation in case marking (see, e.g., Song, 2001). For example, the indexing approach provides a better account of case marking of oblique arguments and of the marking of core arguments in languages where case reflects semantic roles of arguments (‘role-dominated’ languages in terms of Van Valin & LaPolla, 1997). On the other hand, the indexing approach on its own fails to account for a well-known tendency, attested both in accusative and ergative languages, to leave the single argument of an intransitive verb (S) as well as one of the core arguments of the transitive verb (A or P) unmarked. To account for this pattern one usually invokes other functional factors, such as the need to differentiate between the two arguments (i.e., the discriminating function mentioned above), as well as economy, which disfavors overt marking of (core) arguments. Thus, both factors are needed to explain variation in case marking across languages, in terms of both possible alignment systems and argument alternations (Malchukov & de Swart, 2009).5
From the gradualist perspective the controversies concerning delimitation of structural and semantic cases are unproblematic. At one pole we have unmistakably structural cases (the nominative within the nominative-accusative system, and the absolutive within the ergative system), which are formally and functionally unmarked (see below). Typical examples of semantic cases are locative cases which represent a subsystem of their own, often quite an elaborate one as in Daghestanian languages. These languages have been claimed to have very large case inventories, but in fact they represent a combination of the two distinct categories of configuration (called localization by Kibrik et al., 1977, p. 51) and (directional) case.
Table 2. Agul: Two Configurations x Three Directions
behind the house
(to) behind the house
from behind the house
in front of the house
(to) in front of the house
from in front of the house
Source: Daniel and Ganenkov (2009, p. 675).
By combining directions and configurations with each other and adding further markers for deictic distinctions, over 100 spatial cases can be distinguished. But Comrie and Polinsky (1998) point out that these “cases” are not single inflectional categories but combinations of categories from at least two different inflectional category systems. As Haspelmath (2009, p. 517) notes, “on this view, a label such as super-elative would not stand for a single case, but for a localization-case combination (analogous to labels such as past perfect or pluperfect, which stands for a particular tense-aspect combination).”6 On the other hand, as pointed about by Blake (2009), over time configuration markers and (directional) case can fuse, and then each fused form should be regarded as a case marker in its own right. Blake illustrates this scenario with Finnish, where forms for ‘in’ and ‘on’ have partly fused with the locative (essive), partitive, and translative –ksi to yield nine local cases. Further discussion of spatial case systems from a typological perspective can be found in Creissels (2009) and Stolz (1992); for further discussion of spatial case systems from a formal semantic perspective, the reader is referred to de Hoop and Zwarts (2009) as well as the earlier treatment by Kracht (2002); see also Lestrade (2010) for an attempt to synthesize the typological and formal semantic approaches to spatial case marking.
Given that locative cases are prime examples of semantic cases, it is instructive that the distinction between locative and structural cases is not clear-cut. Thus, locative cases may develop non-spatial (syntactic) uses (this is even true for Daghestanian spatial cases, which appear in subcategorization frames of certain verbs). Indeed, on some accounts, appropriately called ‘localist,’ all non-spatial case functions are derived from spatial ones (see Anderson, 2009, for references; see also Anderson, 2006, for a fuller exposition, and Hjelmslev, 1935, for pioneering proposal). If taken as a synchronic claim, the localist accounts, relying on some theory-internal assumptions, may be challenged, but from a diachronic perspective, the claim is certainly valid, as the paths of reanalysis of spatial case markers into the non-local domain are well documented (see, e.g., Yamaguchi, 2004).
4 Case Paradigms: Conflation of Syntactic, Semantic, and Pragmatic Case
It is a common observation in the literature that cases mark different sorts of information: syntactic, semantic, and discourse-pragmatic (Blake, 2001; Givón, 1985, 2001; Kibrik, 1997). Conflation between syntactic/semantic and discourse-pragmatic information is less conspicuous in European languages, although it is common to regard subject case as a grammaticalized topic and correspondingly relate nominative case to topicality (Givón, 1984). Other languages conflate pragmatic functions with grammatical functions in a clearer fashion. Thus in Korean, both nominative and accusative case perform pragmatic (information structural) functions as can be seen in contexts of case stacking and case spreading (Van Valin, 2009). Another well-known example is Japanese, where the topic marker wa is to some extent in a paradigmatic distribution with respect to other cases: it is excludes nominative (ga), accusative (o), and dative (ni), although it may co-occur with oblique cases such as de ‘in’ and locative uses of ni (Ogawa, 2009). While, some nominative markers, if overt, are linked to topicality, ergative marking in some languages is demonstrably restricted to focal/contrastive contexts. This phenomenon is well attested in Australian languages, where it has been studied under the heading of ‘optional ergativity’ (McGregor & Verstraete, 2010), but is also found elsewhere (Malchukov, 2008). This is exemplified below for Newari, where As take the ergative marker when rhematic (Givón, 1984, p. 154). Thus, (7a) would be appropriate as an answer to “Who is breaking the window?,” (7b) as an answer to “What is the man doing?”
As far as semantic and syntactic (‘structural’) cases are concerned, it is more common cross-linguistically to have both types within the same paradigm. A possible exception in this respect are languages with a minimal (two-term) case system where cases cannot be associated with a particular semantic function (see Arkadiev, 2009). Equally rare are systems where all cases qualify as semantic. Possible candidates are languages with ‘active’ alignment (also known as ‘semantic alignment,’ in terms of Donohue & Wichmann, 2008), although it manifests itself more often in agreement than case, and some authors describe ergative languages in a similar way. For example, according to Kibrik (1997), Daghestanian languages are semantically based and distinguish semantic hyperroles of Agentive and Absolutive. This is also consistent with the view of Blake (2001), who characterizes ergative as a basically semantic case, nominative as a syntactic case with a pragmatic bases (topic), while accusative and absolutive are syntactic cases with Patient as a prototype. More generally, it is probably true that languages with “mixed” paradigms conflating different dimensions are more widespread than “pure” languages, which are either role-oriented or discourse-oriented (see Kibrik, 1997).
Thus, conflation of different types of information within a single case system is more a rule than an exception. This has a diachronic explanation; it is well known that semantic cases develop into syntactic cases in the course of grammaticalization (thus, ergative can develop from instrumental or locative markers, and accusative from dative markers of allative origin; Lehmann, 1995). Similarly, conflation of pragmatic and syntactic cases is due to reanalysis of discourse markers into syntactic markers. Thus, the marked nominative case frequently originates from definiteness/topicality markers (König, 2009), and the ergative case arises from contrastive topicalization (Malchukov, 2008; McGregor, 1998, 2008).
5 Case Hierarchy
Heterogeneity of case paradigms, which reflects the complex histories of paradigms in individual languages, obviously presents a challenge in defining the meaning of individual cases within a paradigm. It also makes all the more challenging the task of making generalizations about the composition of case paradigms across languages. The sizes of case systems vary dramatically (Iggesen, 2005), from minimal (two-case) systems (see Arkadiev, 2009) to the large inventories exemplified by Daghestanian (see Daniel & Ganenkov, 2009). The task of comparing case systems (and also individual cases) across languages is further aggravated by discrepancies in terminology, where, for example, similar functions can be referred to as dative, directive, or allative (see Haspelmath, 2009, on challenges presented by case terminology). It is all the more remarkable that in spite of these challenges typologists made much progress in discovering constraints on the types of possible case systems. The first important advance was made by Greenberg (1966), who proposed a number of markedness hierarchies for grammatical categories, including markedness hierarchy of case as shown in (8):
This hierarchy aims to capture the markedness pattern among cases, as reflected in overt marking, case frequency, but also implicational relations within case systems. It is expected that the case higher in the hierarchy will display less overt marking (note the frequent zero marking of the nominative case) and is more frequent in texts as compared to lower cases. Most important in the present context is the typological generalization to the effect that availability of a lower case implies presence of a higher case in a case system. Greenberg’s markedness hierarchy has been extensively discussed and confirmed by other authors (Blake, 1994; Hawkins, 2004; Primus, 1999; Silverstein, 1976).
In particular, Blake (1994) took Greenberg’s proposals, further suggesting the most elaborate version of case hierarchy to date. Blake (1994, 2001), following earlier insights by Greenberg and Silverstein, proposed the case hierarchy in (9):
The hierarchy in (9) is intended as an implicational hierarchy: presence of a case to the right entails availability of case(s) to the left. Blake (2001, pp. 155–160) provides cross-linguistic evidence to support this claim. Thus two-case languages typically have cases covering the core functions (NOM/DIR7 vs. ACC/OBL), with other functions being expressed, say, by adpositions. Other languages have three cases (NOM-ACC-GEN/OBL as in, e.g., Semitic), four cases (NOM-ACC-GEN-DAT/OBL, as in German or Icelandic), five cases (NOM-ACC-GEN-DAT-ABL/OBL, as in Latin), or six cases (NOM-ACC-GEN-DAT-LOC-INS as in many Slavic languages, or NOM-ACC-GEN-DAT-LOC-ABL, as in Turkish). In these examples the expansion of the inflectional case system proceeds in accordance with Blake’s hierarchy. Another observation made by Blake is that the case lowest on the hierarchy is a kind of elsewhere case which subsumes a variety of lower ‘functions.’ This is most obvious for the minimal case systems (see Arkadiev, 2009).
Blake is quick to make qualifications to cover apparent counterexamples. First, a language need not have an overt case marker for some cases. This is obviously true for nominative case, which is usually unmarked, but it may be hold in other cases, when the corresponding function is expressed by alternative means. Second, presence of an overt marker for a core argument is frequently obviated through the use of agreement and/or (strict) word order. Another qualification that may account for another group of counterexamples is that a separate case lower on the hierarchy may be missing due to the fact that a higher case has taken over its duty. This is the reason why Blake did not introduce a separate allative case into the hierarchy, as its function is frequently encoded by the (higher) dative or locative cases.
With these qualifications in mind, Blake’s hierarchy can be shown to hold for a wide variety of cases systems, at least as a tendency (see Malchukov & Spencer, 2009). It is also consistent with the findings of grammaticalization research with respect to the order in which cases grammaticalize (Heine, 2009; Lehmann, 1995). As already anticipated by Blake, who intended his hierarchy to show that “morphological case systems grow and decay in a certain order” (Blake, 2001, p. 155), the hierarchy also accounts for the order in which individual cases are lost. Thus, in South Slavic dialects (Sobolev, 2009) and Ancient Greek (Luraghi, 2003), loss of case proceeds roughly in accordance with the hierarchy, and the same seems to be true of mergers of individual cases in different branches of Indo-European (see Barðdal & Kulikov, 2009).
Ultimately case hierarchy may be motivated by the frequency of individual cases, as already anticipated by Greenberg (1966; see also Hawkins, 2004, pp. 64–68). Thus, it has been observed for Germanic languages that loss of particular uses of individual cases (“case constructions”) depends on their (token) frequency (Barðdal & Kulikov, 2009). The effects of case hierarchy—or parts of it—have also been confirmed in psycholinguistic literature. Thus, cases higher on the hierarchy are acquired earlier as compared to the lower ones (e.g. accusative before dative in German; see Eisenbeiss et al., 2009). And in language comprehension, violations resulting from a substitution of a higher case for a lower one are felt as less severe as compared to violations of the opposite kind (Bader & Lamers, 2009). Thus, on the whole, Blake’s hierarchy seems well established, even though a large-scale empirical validation is still outstanding. In fact, proposals have been made in the generative literature to view it as part of universal grammar (Caha, 2008).
6 Case Polysemies and Semantic Maps
Defining meaning of cases, whether for particular languages or cross-linguistically, presents a separate problem. Traditional grammars associated cases with individual grammatical or semantic roles or described the meaning of individual cases by listing individual uses. In the structuralist period attempts were made (most famously by Hjelmslev, 1935, for Eskimo, and Jakobson, 1936, 1958, for Russian) to describe the meanings of individual cases in terms of features. These attempts have been influential and provided important insights (for example, introducing the concept of markedness into morphology) but on the whole were not fully successful. Consider Jakobson’s well-known characterization of cases in Russian in terms of three binary features (±Marginal; ±Quantifying; ±Ascriptive), depicted in Table 3.
Table 3. Russian Case System in Terms of Jakobson’s Features
Source: Adapted from Blake (2001, p. 40).
Commenting on Jakobson’s characterization of cases in Russian, Blake (2001, p. 40) notes lack of independent motivation for the features proposed, and Corbett (2012, p. 19) expressed similar concerns, characterizing Jakobson’s (1958) feature analysis of Russian case system as a “heroic failure.”8 Yet, in spite of its shortcomings, feature analysis has been put to good use in subsequent work in explaining the patterns of case syncretism (see Blake, 2001, p. 43, for an analysis of syncretism in Latin case systems along these lines). The analyses in terms of shared semantic components or feature underspecification are most appealing in case of systematic syncretisms; class-specific syncretisms have to be captured by other means, such as Zwicky’s “rules of referral” (Blevins, 2009).
Another approach to the study of meaning of cases and other grammatical categories is the semantic map approach. This approach, pioneered by Anderson (1982) and followed up in the work by Croft (2001) and Haspelmath (2003), among others, starts from the iconicity assumption that recurrent similarities in form reflect regularities in meaning (see Haiman, 1985, on iconicity in grammar). The basic methodology behind the semantic map approach as fleshed out by Croft (2001), and Haspelmath (2003) relies on the establishment of cross-linguistic regular polysemies, which on iconicity assumptions reflect conceptual similarities. The functions covered by the same form in (some) languages are considered to be semantically close, which is represented by putting them adjacently in the semantic space. Although the semantic map is established empirically through the study of polysemy patterns across languages, the established semantic connections are conceived as universal and give rise to predictions concerning possible and impossible polysemy patterns across languages.
The semantic map approach has been successfully applied to different domains of grammatical structure (see Cysouw, Haspelmath, & Malchukov, 2010, for references), including case. This research generally produced only partial maps, as it takes specific functions (semantic roles or “deep cases” in terms of Fillmore, 1968) as point of departure and explores frequent polysemies. Thus, partial maps have been proposed for the dative domain (Haspelmath, 2003), instrumental domain (Narrog & Ito, 2007), and allative domain (Rice & Kabata, 2007, among others).9 Malchukov and Narrog (2010), building on the earlier literature, present the general map for the major functions, represented in Figure 1. The map introduces case functions (semantic roles), mostly from a non-spatial domain: Agent (A), Patient (P), Theme (T),10 Recipient (R), Instrument (INS; see He hit the dog with a stick), Material/Means (MAT/MEANS; see He built the house with bricks), Beneficiary (BEN: see I built him a house), Possessor (POS, see I saw his house), Source (SO: see He returned from the store), Goal (G; see He returned to the store), and Comitative (COM; see He returned with a friend).
On the one hand, the semantic map in Figure 1 has semantic plausibility insofar as the related (connected) functions have been shown to share semantic features (see Grimm, 2011; Næss, 2007; Reinhart, 2002; Rozwadowska, 1988; for a sample of partially converging proposals; Levin & Rappaport Hovav, 2005, for general discussion). Thus Agents and Possessors are controlling entities; Agents, Possessors, Beneficiaries, and Recipients are typically animate (sentient); Patients and Themes are affected entities; Agents and Instruments are instigating entities; Instruments, Means, and Themes are moving entities; and finally, Material, Themes, and Patients share the feature of incrementality.
On the other hand, the proposed map finds empirical support in the most frequent polysemy patterns, as described in the typological literature (see, in particular, contributions to Malchukov & Spencer, 2009, dealing with individual cases for discussion and further references): the dative-allative polysemy, as familiar from English (recall the functions of to), is common across languages (Creissels, 2009; Næss, 2009); the dative-genitive polysemy is attested in many Australian and Austronesian languages but is also found elsewhere (Lander, 2009; Næss, 2009); the genitive-ablative polysemy is especially common in languages using adpositions for these functions (Heine, 2009; Lander, 2009); the dative-accusative polysemy is familiar from languages with differential object marking (Malchukov & de Swart, 2009; Næss, 2009); instrumental-accusative polysemy is typical for languages with secundative alignment (Kittilä & Malchukov, 2009); the instrumental-comitative polysemy is the most frequent polysemy pattern involving both cases (Narrog, 2009; Stolz et al., 2009); the ergative-instrumental and ergative-genitive polysemies are identified as the two most frequent polysemy patterns involving ergative case (Palancar, 2009).11
Semantic maps can also be given a diachronic interpretation, as they most frequently reveal the regular patterns of meaning extensions as described in grammaticalization studies by Lehmann (1995), Heine et al., 1991, and others. Taking into account recent work in the semantic map tradition, such as Yamaguchi (2004) on the extensions of the locative markers to non-local domain and Narrog (2010) on evolution of the dative markers, as well as grammaticalization literature (as summarized in Heine & Kuteva, 2002), the diachronic dimension can be represented by the semantic map in Figure 2.
7 Case and Alternative Strategies
It is conventional to regard case on a par with alternative strategies such as agreement and word order (see Kiparsky’s  view of case, agreement, and word order as alternative linkers; see Butt, 2006). If case marking is not possible, or is not unambiguous, for particular words or constructions, we may find the phenomenon of word order freezing (or conversely we may find that case marking allows scrambling; see Neeleman & Weerman, 2009). A similar point can be made with respect to agreement, which is often seen as an alternative way of marking syntactic relations. This view is implicit in Nichols’s (1992) influential typological study, which opposes case and agreement as dependent vs. head-marking of grammatical relations. This view is also echoed in generative grammar, which also frequently views case and agreement as instantiations of the same fundamental grammatical relationship (Bobaljik & Wurmbrand, 2009). Yet, as noted by Bobaljik and Wurmbrand (2009), this conclusion, if taken at face value, is unmotivated, given cross-linguistically well-attested mismatches between case and agreement (see, e.g., languages like Warlpiri that combine nominative-style agreement with ergative-style case marking). Neither is there observed a consistent complementarity of case and agreement (Siewierska & Bakker, 2009); rather, this option depends on the argument type, as the A-arguments trigger agreement in preference to P-arguments. Interestingly, Siewierska and Bakker (2009) find more support for an inverse correlation between availability of dependent marking (accusative or ergative) and strict SVO (subject verb object) word order. A recent typological study by Sinnemäki (2011) confirmed that the absence of a correlation between case and agreement and a presence of an inverse correlation between case and word order, suggesting that there is indeed a trade-off between the two mechanisms in distinguishing between the core arguments.
But again, this latter parallel is obviously incomplete, as the primary function of word order is discourse pragmatic. As emphasized by Malchukov and Spencer (2009), case, agreement, and word order have different domains of application; there may be spillovers to other domains when certain strategies are used in extended functions, but in this case the strategy would be used in its central functions as well. Indeed, marking semantic functions is a universal property of cases and adpositions, even though some languages can extend their use to pragmatic functions as well (e.g., marked nominative languages). Conversely, word order is universally used for encoding discourse-pragmatic notions, but some languages (especially SVO languages) may extend its use for identifying subjects and objects; yet, word order is hardly ever used to encode semantic roles. Finally, agreement has typically a syntactic and/or pragmatic function in cases when just one argument is cross-referenced but may perform semantic functions as well, in particular, in languages where verbs show multiple agreement paradigms. More discussion of cases of complementarity and overlap between case and alternative strategies can be found in Malchukov and Spencer (2009) and Siewierska and Bakker (2009).
A transitive subject/agent; ABL ablative; ACC accusative; ANTE localization in front of the landmark; AOR aorist; AUX auxiliary; BEN beneficiary; COM comitative; DAT dative; DES designative case; DIR direct (case); EL elative; ERG ergative; Go, G goal; GEN genitive; INS instrumental; LAT lative; LOC locative; MAT material; MS masculine; NOM nominative; OBL oblique; P transitive object/patient; PERF perfect; POS possessive; possessor; POST localization behind the landmark; PST past; R recipient; S intransitive subject; So source; T ditransitive object theme.
The literature on case is vast and cannot be done justice in this brief review. An (already somewhat obsolete) bibliography on case and semantic roles and grammatical relations (Campe, 1994) numbers over 600 pages. The present overview draws heavily on the contributions to the Oxford Handbook of Case (Malchukov & Spencer, 2009), which should be consulted for further references. A classic introduction to case from a typological perspective is Blake (1994, 2001); it also reviews the history of research on case. Anderson (2006) provides detailed discussion of the localist grammar approaches and its predecessors, in particular, Fillmore’s (1968) case grammar. A comprehensive overview of modern approaches to case is presented in Butt (2006); for a recent in-depth discussion of case from the generative perspective, see Baker (2015).
In addition, a number of recent publications feature case as one major research topic. Morphological aspects of case have been investigated in the work by Corbett and his associates in the Surrey Morphology Group (see, e.g., Corbett, 2012, as well as a monograph by Baerman et al., 2005, dealing, in particular, with case syncretism). Two early classics explicating a procedure of identifying (morphological) cases are Comrie (1986) and Mel’čuk (1986). Syntactic aspects of case marking from cross-linguistic and cross-theoretical perspectives are addressed in such collections as Kulikov, Malchukov, and De Swart (2006), Amberber and de Hoop (2005), von Heusinger and de Hoop (2011), and Donohue and Barðdal (2011), among others; the papers in Kittilä et al. (2011) focus on semantic aspects of case marking. The role of discourse factors in case marking has attracted much attention recently; see, e.g., contributions to Barðdal and Chelliah (2009), as well as contributions to McGregor and Verstraete (2010), for discussion of discourse factors behind “optional ergativity,” and Dalrymple and Nikolaeva (2011) on the role of information structure with regard to differential object marking (“optional accusativity”). The functions of individual cases have been addressed (predominantly from a cognitive linguistics perspective) in a multivolume series Case and Grammatical Relations Across Languages edited by a team of linguists based at the University of Leuven (see, e.g., Carlier & Verstraete, 2013, on the genitive case). Two other earlier studies representative of cognitive grammar approaches to case are a classic study by Wierzbicka (1980) and Laura Janda’s (1993) Geography of Case Semantics.
Aissen, J. (2003). Differential object marking: Iconicity vs. economy. Natural Language and Linguistic Theory, 21, 435–483.Find this resource:
Amberber, M., & H. de Hoop (Eds.). (2005). Competition and variation in natural languages: The case for case. Oxford: Elsevier.Find this resource:
Anderson, J. M. (1977). On case grammar: Prolegomena to a theory of grammatical relations. London: Croom Helm.Find this resource:
Anderson, J. M. (2006). Modern grammars of case: A retrospective. Oxford: Oxford University Press.Find this resource:
Anderson, J. (2009). Case in localist case grammar. In A. Malchukov & A. Spencer (Eds.), pp. 121–135.Find this resource:
Anderson, L. B. (1982). The “perfect” as a universal and as a language specific category. In P. J. Hopper (Ed.), Tense–aspect: Between semantics and pragmatics (pp. 227–264). Amsterdam: John Benjamins.Find this resource:
Arkadiev, P. (2009). Poor (two-term) case systems: Limits of neutralization. In A. Malchukov & A. Spencer (Eds.), pp. 686–700.Find this resource:
Bader, M., & Lamers, M. (2009). Case in language comprehension. In A. Malchukov & A. Spencer (Eds.), pp. 402–419.Find this resource:
Baerman, M. (2009). Case syncretism. In A. Malchukov & A. Spencer (Eds.), pp. 219–231.Find this resource:
Baerman, M., Brown, D., & Corbett, G. (2005). The syntax-morphology interface: A study of syncretism. Cambridge, U.K.: Cambridge University Press.Find this resource:
Barðdal, J., & Kulikov, L. (2009). Case in decline. In A. Malchukov & A. Spencer (Eds.), pp. 470–479.Find this resource:
Barðdal, J., & Chelliah, S. (2009). The role of semantic, pragmatic and discourse factors in the development of case. Amsterdam: John Benjamins.Find this resource:
Baker, M. (2015). Case: Its principles and its parameters. Cambridge, U.K.: Cambridge University Press.Find this resource:
Beard, R. (1995). Lexeme morpheme base morphology. Stony Brook, NY: SUNY Press.Find this resource:
Bickel, B., Zakharko, T., Bierkandt, L., & Witzlack-Makarevich, A. (2014). Semantic role clustering: An empirical assessment of semantic role types in non-default case assignment. Studies in Language, 38(3), 485–511.Find this resource:
Blake, B. J. (1994). Case. Cambridge, U.K.: Cambridge University Press.Find this resource:
Blake, B. J. (2001). Case (2d ed.). Cambridge, U.K.: Cambridge University Press.Find this resource:
Blake, B. J. (2009). History of the research on case. In A. Malchukov & A. Spencer (Eds.), pp. 13–27.Find this resource:
Blevins, J. (2009). Case and declensional paradigms. In A. Malchukov & A. Spencer (Eds.), pp. 200–219.Find this resource:
Bobaljik, J., & Wurmbrand, S. (2009). Case in GB/minimalism. In A. Malchukov & A. Spencer (Eds.), pp. 44–59.Find this resource:
Bobaljik, J. (2008). Where’s phi? Agreement as a post-syntactic operation. In D. Adger, D. Harbour, & S. Béjar (Eds.), Phi theory: Phi features across interfaces and modules (pp. 295–328). Oxford: Oxford University Press.Find this resource:
Bossong, G. (1985). Differentielle Objektmarkierung in den neuiranischen Sprachen. Tübingen: Narr.Find this resource:
Butt, M. (2006). Theories of case. Cambridge, U.K.: Cambridge University Press.Find this resource:
Caha, P. (2008). The case hierarchy as functional sequence. In M. Richards & A. L. Malchukov (Eds.), Scales (pp. 247–276). Leipzig: Linguistische Arbeitsberichte.Find this resource:
Campe, P. (1994). Case, semantic roles, and grammatical relations: A comprehensive bibliography. Amsterdam: John Benjamins.Find this resource:
Carlier, A., & Verstraete, J.-C. (Eds.). (2013). The genitive. Amsterdam: John Benjamins.Find this resource:
Comrie, B. (1986). On delimiting cases. In R. D. Brecht & J. S. Levine (Eds.), Case in Slavic (pp. 86–106). Columbus, OH: Slavica.Find this resource:
Comrie, B. (1989). Language universals and linguistic typology. Chicago: University of Chicago Press.Find this resource:
Comrie, B., & Polinsky, M. (1998). The great Daghestanian case hoax. In A. Siewierska & J. J. Song (Eds.), Case, typology and grammar (pp. 95–114). Amsterdam: John Benjamins.Find this resource:
Corbett, G. (2008). Determining morphosyntactic feature values: The case of case. In G. Corbett & M. Noonan (Eds.), Case and grammatical relations: Studies in honor of Bernard Comrie (pp. 1–34). Amsterdam: John Benjamins.Find this resource:
Corbett, G. (2012). Features. Cambridge, U.K.: Cambridge University PressFind this resource:
Creissels, D. (2009). Spatial cases. In A. Malchukov & A. Spencer (Eds.), pp. 609–626.Find this resource:
Croft, W. (1991). Syntactic categories and grammatical relations. Chicago: The University of Chicago Press.Find this resource:
Croft, W. (2001). Radical construction grammar. Oxford: Oxford University Press.Find this resource:
Croft, W. (2003). Typology and universals (2d ed.). Cambridge, U.K.: Cambridge University Press.Find this resource:
Cysouw, M., Haspelmath, M., & Malchukov, A. (Eds.). (2010). Semantic maps: Theory and applications. Linguistic Discovery, 8(1).Find this resource:
Daniel, M., & Spencer, A. (2009). The vocative—an outlier case. In A. Malchukov & A. Spencer (Eds.), pp. 626–635.Find this resource:
Daniel, M., & Ganenkov, D. (2009). Case marking in Daghestanian: Limits of elaboration. In A. Malchukov & A. Spencer (Eds.), pp. 668–686.Find this resource:
Dalrymple, M., & Nikolaeva, I. (2011). Objects and information structure. Cambridge, U.K.: Cambridge University Press.Find this resource:
Dixon, R. M. W. (1994). Ergativity. Cambridge, U.K.: Cambridge University Press.Find this resource:
Donohue, C., & J. Barðdal (2011). A special guest-edited issue on Empirical Approaches to Morphological Case. Morphology,21(3–4).Find this resource:
Donohue, M. (2009). Case in an Austronesian language: Distinguishing case functions in Tukang Besi. In A. Malchukov & A. Spencer (Eds.), pp. 770–779.Find this resource:
Donohue, M., & Wichmann, S. (Eds.). (2008). Typology of languages with semantic alignment. Oxford: Oxford University Press.Find this resource:
Dryer, M. (1986). Primary objects, secondary objects, and antidative. Language, 62, 808–845.Find this resource:
Eisenbeiss, S., Narasimhan, B., & Voeikova, M. (2009). The acquisition of case. In A. Malchukov & A. Spencer (Eds.), pp. 369–384.Find this resource:
Fillmore, C. J. (1968). The case for case. In E. Bach & R. T. Harms (Eds.), Universals of linguistic theory (pp. 1–90). New York: Holt, Rinehart and Winston.Find this resource:
Givón, T. (1984). Syntax: A functional-typological introduction (Vol. 1). Amsterdam: Benjamins.Find this resource:
Givón, T. (1985). Ergative morphology and transitivity gradients in Newari. In F. Plank (Ed.), Relational typology (pp. 89–107). Berlin: Mouton.Find this resource:
Givón, T. (2001). Syntax. A functional-typological introduction (Vol. 2). Amsterdam: John Benjamins.Find this resource:
Goddard, C. (1982). Case systems and case marking in Australian languages: A new interpretation. Australian Journal of Linguistics, 2, 167–196.Find this resource:
Greenberg, J. (1966). Language universals, with particular reference to feature hierarchies. The Hague: Mouton.Find this resource:
Grimm, S. (2011). Semantics of case. Morphology, 21, 515–544.Find this resource:
Haiman, J. (1985). Natural syntax. Iconicity and erosion. Cambridge, U.K.: Cambridge University Press.Find this resource:
Hartmann, I., Haspelmath, M., & Cysouw, M. (2014). Identifying semantic role clusters and alignment types via microrole coexpression tendencies. Studies in Language, 38(3), 463–484.Find this resource:
Haspelmath, M. (2003). The geometry of grammatical meaning: semantic maps and cross-linguistic comparison. In M. Tomasello (Ed.), The new psychology of language (Vol. 2, pp. 211–243). New York: Lawrence Erlbaum.Find this resource:
Haspelmath, M. (2005). Argument marking in ditransitive alignment types.Linguistic Discovery, 3(1), 1–21.Find this resource:
Haspelmath, M. (2009). Terminology of case. In A. Malchukov & A. Spencer (Eds.), pp. 505–518.Find this resource:
Haspelmath, M. (2011). The indeterminacy of word segmentation and the nature of morphology and syntax. Folia Linguistica, 45(1), 31–80.Find this resource:
Hawkins, J. A. (2004). Efficiency and complexity in grammars. Oxford: Oxford University Press.Find this resource:
Heine, B. (2003). Grammaticalization. In B. D. Joseph & R. D. Janda (Eds.), The handbook of historical linguistics (pp. 575–601). Oxford: Blackwell.Find this resource:
Heine, B., Claudi, U., & Hünnemeyer, F. (1991). Grammaticalization. A conceptual framework. Chicago: Chicago University Press.Find this resource:
Heine, B., & Kuteva, T. (2002). World lexicon of grammaticalization. Cambridge, U.K.: Cambridge University Press.Find this resource:
Heine, B. (2009). Grammaticalization of cases. In Malchukov & Spencer (Eds.), pp. 458–470.Find this resource:
von Heusinger, K., & de Hoop, H. (Eds.). (2011). Semantic aspects of case variation [Special issue]. Lingua, 121(1).Find this resource:
Hewitt, B. G. (1979). Abkhaz. Amsterdam: North Holland.Find this resource:
Hjelmslev, L. (1935). La catégorie des cas. Etude de grammaire générale. Aarhus, Denmark: Acta Jutlandica.Find this resource:
de Hoop, H., & Malchukov, A. (2008). Case-marking strategies. Linguistic Inquiry, 39, 565–587.Find this resource:
de Hoop, H., & de Swart, P. (Eds.). (2009). Differential subject marking. Dordrecht, The Netherlands: Springer.Find this resource:
de Hoop, H., & Zwarts, J. (2009). Case in formal semantics. In Malchukov & Spencer (Eds.), pp. 170–185.Find this resource:
Iggesen, O. (2005). Case-asymmetry. Munich: Lincom Europa.Find this resource:
Iggesen, O. (2009). Asymmetry in case marking: Nominal vs. pronominal systems. In A. Malchukov & A. Spencer (Eds.), pp. 246–260.Find this resource:
Jackendoff, R. (1972). Semantic Interpretation in Generative Grammar. Cambridge: The MIT Press.Find this resource:
Jakobson, R. (1936). Beiträge zur Allgemeinen Kasuslehre: Gesamtbedeutungen der russischen Kasus. Travaux du Cercle Linguistique de Prague, 6, 240–288.Find this resource:
Jakobson, R. (1958/1971). Morphological inquiry into Slavic declension: Structure of Russian case forms [text in Russian, summary in English in 1971 version]. In Selected writings II: Words and language (pp. 154–183). The Hague: Mouton.Find this resource:
Janda, L. (1993). A geography of case semantics: The Czech dative and the Russian instrumental. Berlin: Mouton de Gruyter.Find this resource:
Kibrik A. E., Kodzasov S. V., Olovjannikova I. P., & Samedov D. S. (1977) Arčinskij jazyk. Teksty i slovari. Moscow: Moscow State University.Find this resource:
Kibrik, A. E. (1985). Toward a typology of ergativity. In J. Nichols & A. Woodbury (Eds.), Grammar inside and outside the clause (pp. 268–324). Cambridge, U.K.: Cambridge University Press.Find this resource:
Kibrik, A. E. (1997). Beyond subject and object: Towards a comprehensive relational typology. Linguistic Typology, 1(3), 279–346.Find this resource:
Kiparsky, P. (1997). The rise of positional licensing. In A. van Kemenade & N. Vincent (Eds.), Parameters of morphosyntactic change (pp. 460–494). Cambridge, U.K.: Cambridge University Press.Find this resource:
Kittilä, S. & Malchukov, A. (2009). Varieties of accusative. In: A. Malchukov & A. Spencer (Eds.), pp. 549–561.Find this resource:
Kittilä, S., Västi, K., & Ylikovski, J. (2011). Case, animacy and semantic roles. Amsterdam: Benjamins.Find this resource:
Kracht, M. (2002). On the semantics of locatives. Linguistics and Philosophy, 25, 157–232.Find this resource:
König, C. (2009). Marked nominatives. In A. Malchukov & A. Spencer (Eds.), pp. 535–549.Find this resource:
Kulikov, L., Malchukov, A., & de Swart, P. (Eds.). (2006). Case, valency and transitivity. Amsterdam: Benjamins.Find this resource:
Lander, Y. (2009). Varieties of genitive. In A. Malchukov & A. Spencer (Eds.), pp. 581–593.Find this resource:
Legate, J. (2014). Split ergativity based on nominal type. Lingua, 148, 183–212.Find this resource:
Lehmann, Ch. (1995). Thoughts on grammaticalization. Munich: Lincom Europa.Find this resource:
Lestrade, S. (2010). The space of case (PhD diss.). Radboud University, Nijmegen, The Netherlands.Find this resource:
Levin, B., & Rappaport Hovav, M. (2005). Argument realization. Cambridge, U.K.: Cambridge University Press.Find this resource:
Levinson, S. C., & Meira, S. (2003). “Natural concepts” in the spatial topological domain—adpositional meanings in crosslinguistic perspective: An exercise in semantic typology. Language, 79, 45–516.Find this resource:
Luraghi, S. (2001). Some remarks on instrument, comitative, and agent in Indo-European. Sprachtypologie und Universalienforschung, 54(4), 385–401.Find this resource:
Luraghi, S. (2003). On the meaning of prepositions and cases. A study of the expression of semantic roles in ancient Greek. Amsterdam: Benjamins.Find this resource:
Mallinson, G., & Blake, B. (1981). Language typology: Cross-linguistic studies in syntax. Amsterdam: North Holland.Find this resource:
McGregor, W. (1998). Optional case-marking in Gooniyandi revisited: Implications to a theory of marking. Leuvense Bijdragen, 87, 491–534.Find this resource:
McGregor, W., & Verstraete, J. (Eds.). (2010). Optional ergative marking [Special issue] Lingua.Find this resource:
Malchukov, A., & Spencer, A. (Eds.). (2009). The Oxford handbook of case. Oxford: Oxford University Press.Find this resource:
Malchukov, A., & Spencer, A. (2009). Typology of case systems. In A. Malchukov & A. Spencer (Eds.), pp. 651–668.Find this resource:
Malchukov, A., & Narrog, H. (2009). Case polysemy. In A. Malchukov & A. Spencer (Eds.), pp. 518–535.Find this resource:
Malchukov, A., & de Swart, P. (2009). Differential case marking and actancy variations. In A. Malchukov & A. Spencer (Eds.), 339–356.Find this resource:
Malchukov, A. (2008). Animacy and asymmetries in differential case marking. Lingua, 118, 203–221.Find this resource:
Malchukov, A. (2009). Rare and exotic cases. In A. Malchukov & A. Spencer (Eds.), pp. 635–651.Find this resource:
Malchukov, A., Haspelmath, M., & Comrie, B. (2010). Ditransitive construction: A typological overview. In A. L. Malchukov, M. Haspelmath, & B. Comrie (Eds.), Studies in ditransitive constructions: A comparative handbook (pp. 1–64). Berlin: Mouton De Gruyter.Find this resource:
McGregor, W. (2008). Indexicals as sources of case markers in Australian languages. In Folke Josephson & Ingmar Söhrman (Eds.), Interdependence of diachronic and synchronic analyses (pp. 299–321). Amsterdam/Philadelphia: John Benjamins.Find this resource:
Melčuk, I. (1986). Toward a definition of case. In R. D. Brecht & J. S. Levine (Eds.), Case in Slavic (pp. 35–85). Columbus, OH: Slavica.Find this resource:
Mel’čuk, I. A. (1998). Kurs obščej morfologii (Vol. 2). Moscow: Jazyki russkoj kul’tury.Find this resource:
Mohanan, T. (1990). Arguments in Hindi (PhD diss.). Stanford University, Stanford.Find this resource:
Mohanan, T. (1994). Argument structure in Hindi. Stanford, CA: CSLI.Find this resource:
Moravcsik, E. (2009). The distribution of case. In A. Malchukov & A. Spencer (Eds.), pp. 231–246.Find this resource:
Neeleman, A., & Weerman, F. (2009). Syntactic effects of morphological case. In A. Malchukov & A. Spencer (Eds.), pp. 276–290.Find this resource:
Nichols, J. (1992). Linguistic diversity in space and time. Chicago: University of Chicago Press.Find this resource:
Nikolaeva, I. (2015). On the expression of TAM on nouns: Evidence from Tundra Nenets. Lingua, 166, 99–126.Find this resource:
Narrog, H., & Ito, S. (2007). Reconstructing semantic maps. The comitative-instrumental area. Sprachtypologie und Universalienforschung, 60(4), 273–292.Find this resource:
Palancar, E. L. (2002). The origin of agent markers (Studia Typologica 5). Berlin: Akademie Verlag.Find this resource:
Narrog, H. (2009). Varieties of instrumental. In A. Malchukov & A. Spencer (Eds.), pp. 593–601.Find this resource:
Narrog, H. (2010). A diachronic dimension in maps of case functions. In M. Cysouw, M. Haspelmath, & A. Malchukov (Eds.), Semantic maps: theory and applications [Special issue]. Linguistic Discovery.Find this resource:
Næss, Å. (2007). Prototypical transitivity. Amsterdam: Benjamins.Find this resource:
Næss, Å. (2009). Varieties of dative. In A. Malchukov & A. Spencer (Eds.), pp. 572–581.Find this resource:
Ogawa, A. (2009). Case in a topic-prominent language: Pragmatic and syntactic functions of cases in Japanese. In A. Malchukov & A. Spencer (Eds.), pp. 779–789.Find this resource:
Palancar, E. (2009). Varieties of ergative. In A. Malchukov & A. Spencer (Eds.), pp. 562–572.Find this resource:
Pesetsky, D., & Torrego, E. (2010). Case. In Cedric Boeckx (Ed.), The oxford handbook of linguistic minimalism (pp. 52–72). Oxford: Oxford University Press.Find this resource:
Plank, F. (Ed.). (1991). Paradigms: The economy of inflection. Berlin: Mouton.Find this resource:
Plank, F. (Ed.). (1995). Double case. Agreement by Suffixaufnahme. New York: Oxford University Press.Find this resource:
Primus, B. (1999). Cases and thematic roles—ergative, accusative and active. Tübingen: Niemeyer.Find this resource:
Reinhart, T. (2002). The theta system: An overview. Theoretical Linguistics, 28(3), 229–90.Find this resource:
Rice, S., & Kabata, K. (2007). Crosslinguistic grammaticalization patterns of the allative. Linguistic Typology, 11, 451–514.Find this resource:
Rozwadowska, B. (1988). Thematic restrictions on derived nominals. In W. Wilkins (Ed.), Thematic relations (pp. 147–166). San Diego, CA: Academic Press.Find this resource:
Siewierska, A. (2004). Person. Cambridge, U.K.: Cambridge University Press.Find this resource:
Siewierska, A., & Bakker, D. (2009). Case and alternative strategies: Word order and agreement marking. In A. Malchukov & A. Spencer (Eds.), pp. 290–304.Find this resource:
Silverstein, M. (1976). Hierarchy of features and ergativity. In R. M. W. Dixon (Ed.), Grammatical categories in Australian languages (pp. 112–171). Canberra: Australian Institute of Aboriginal Studies.Find this resource:
Sinnemäki, K. (2011). Language universals and linguistic complexity: Three case studies in core argument marking (PhD diss.). University of Helsinki, Helsinki.Find this resource:
Sobolev, A. (2009). From synthetic to analytic case: Variation in South Slavic dialects In A. Malchukov & A. Spencer (Eds.), pp. 716–730.Find this resource:
Song, J. J. (2001). Linguistic typology: Morphology and syntax. London: Longman.Find this resource:
Spencer, A., & Otoguro, R. (2005). Limits to case—a critical survey of the notion. In M. Amberber & H. de Hoop (Eds.), Competition and variation in natural languages. The case for case (pp. 119–145). Oxford: Elsevier.Find this resource:
Spencer, A. (2006). Syntactic vs. morphological case: Implications for morphosyntax. In L. Kulikov, A. Malchukov, & P. de Swart (Eds.), Case, valency and transitivity: A cross-linguistic perspective. Amsterdam: Benjamins.Find this resource:
Spencer, A. (2008). Does Hungarian have a case system? In G. Corbett & M. Noonan (Eds.), Case and grammatical relations: Studies in honor of Bernard Comrie (pp. 35–56). Amsterdam: John Benjamins.Find this resource:
Spencer, A. (2009). Case as a morphological phenomenon. In A. Malchukov & A. Spencer (Eds.), pp. 185–200.Find this resource:
Stolz, T. (1992). Lokalkasussysteme—Aspekte einer strukturellen Dynamik. Wilhelmsfeld, Germany: Gottfried Egert.Find this resource:
Stolz, T., Stroh, C., & Urdze, A. (2009). Varieties of comitative. In A. Malchukov & A. Spencer (Eds.), pp. 601–609.Find this resource:
Tsumagari, T. (2006). Morphological status of the Manchu case markers: Particle or suffix? In A. Pozzi, J. Janhunen, & M. Weiers (Eds.), Manchu studies in honor of Giovanni Stary (pp. 225–235). Wiesbaden, Germany: Harrassowitz.Find this resource:
Van Valin, R., & Lapolla, R. (1997). Syntax: Structure, meaning and function. Cambridge, U.K.: Cambridge University Press.Find this resource:
Van Valin, R. D., Jr. (2005). Exploring the syntax-semantics interface. Cambridge, U.K.: Cambridge University Press.Find this resource:
Van Valin, R. D., Jr. (2009). Case in role and reference grammar. In A. Malchukov & A. Spencer (Eds.), pp. 102–121.Find this resource:
Vogel, R., & Steinbach, M. (1998). The dative—an oblique case. Linguistische Berichte,173, 65–90.Find this resource:
Wierzbicka, A. (1980). The case for surface case. Ann Arbor, MI: Karoma.Find this resource:
Woolford, E. (2001). Case patterns. In G. Legendre, J. Grimshaw, & S. Vikner (Eds.), Optimality-theoretic syntax (pp. 509–545). Cambridge, MA: MIT Press.Find this resource:
Woolford, E. (2006). Lexical case, inherent case, and argument structure. Linguistic Inquiry,37(1), 111–130.Find this resource:
Yamaguchi, K. (2004). A typological, historical, and functional study of adpositions in the languages of the world (PhD thesis). University of New Mexico.Find this resource:
Zaliznjak, А. А. (1973). O ponimanii termina ‘padež’ v lingvističeskix opisanijax. In A. A. Zaliznjak (Ed.), Problemy grammatičeskogo modelirovanija (pp. 53–87). Moscow: Nauka.Find this resource:
(2.) In Chomskyan generative grammar it is assumed that every (overt) NP should be assigned an abstract case in compliance with a “Case Filter,” but this abstract case needs not be spelled out morphologically. The empirical evidence for abstract case and Case Filter is complex and appeals to a wide array of syntactic phenomena, such as movement of NPs/DPs into the “subject position” to satisfy the Case Filter, or distribution of zero anaphora in infinitive construction (see Bobaljik & Wurmbrand, 2009, for discussion of the evidence for Case Theory in different versions of generative grammar). Baker (2015) proposes that Case Filter is better viewed as a parameter (rather than as a universal) and relies on sophisticated analysis to ascertain its relevance for individual languages.
(3.) In fact, some authors posit further distinctions, which go beyond the m-case vs. s-case dichotomy; thus, Mel’čuk’s (1998) multilevel framework distinguishes between surface morphological vs. deep-morphological, and surface-syntactic vs. deep syntactic levels.
(4.) This discussion has been echoed in other research traditions as well. For example, in in optimality theory, there were similar controversies as to whether case variation should be accounted for in terms of faithfulness constraints (see Woolford, 2006) or markedness constraints (see Aissen, 2003). And in mainstream generative grammar (GB/Minimalism), there are two opposing views on case; an agreement-based approach, on the one hand, and a dependent case theory, on the other hand (see Pesetsky & Torrego, 2010; Bobaljik & Wurmbrand, 2009, for different takes on this issue). The former approach is more in line with the indexing perspective, while the latter approach is more in line with a markedness/distinguishability account.
(5.) In particular, combination of the two factors (distinguishability and Indexing constraints) correctly predicts that differential object marking is cross-linguistically more consistent than differential subject marking (see de Hoop & Malchukov, 2008; Malchukov, 2008).
(6.) On Kracht’s (2002) account, spatial constructions have the underlying semantic structure [M [L [DP]]], where M stands for direction (“Mode,” in Kracht’s terms) and L stands for localization/configuration.
(8.) As Corbett (2012, p. 19) observes in this context, “there are 6720 possible ways to describe eight values using three binary features. In view of this, unless there are principled reasons for postulating particular binary features from the outset, it should not be taken as significant if there is an analysis using binary features which is partially successful.”
(9.) An original approach to the study of locative markers (adpositions) has been developed by S. Levinson and his associates at MPI for Psycholinguistics in Nijmegen (see, e.g., Levinson & Meira,, 2003). This approach is similar in spirit to the semantic map approach but differs in methods of data collection, informed by psycholinguistic research
(10.) The term “theme” refers to the moved object of a ditransitive verb (as in Haspelmath, 2005; Malchukov et al., 2010; see John gave Mary a book), as well as to a moved object of a monotransitive verb (see Jackendoff ,1972; see John moved a chair).
(11.) Full-scale statistical evaluation of this and similar maps is still outstanding. Two recent attempts at clustering verb specific roles (for a sample of verbs meanings) with respect to their coding across a range of diverse languages are Hartmann et al. (2014) and Bickel et al. (2014). These cluster analyses may be seen as statistically informed versions of the semantic map method operating in a strictly bottom-up fashion.