The Bilingual Education Program in the Northern Territory was established in 1973 to deliver an effective education to students in remote Indigenous schools. In 2008, the NT Government closed the Program, citing low student achievement in English literacy as its rationale. This chapter evaluates the program according to a wider set of criteria, with particular focus on the Warlpiri schools. It considers the training of local staff, community involvement in schools, the development of curriculum, and extensive literature and linguistic documentation. These criteria are not prominent in government evaluations, yet are identified in national and international literature as key to achieving student outcomes (UNESCO 2008a, 2008b; Silburn et al. 2011) and echo priorities expressed by Indigenous adults involved in education.
Baby Talk (BT), also known as child-directed speech, is a non-standard form of speech used by adults when talking to infants. In Arandic languages BT involves the use of a small set of unique but widely known words, onomatopoeic-derived words, as well as phonological modifications to standard vocabulary. As in neighboring Warlpiri (Laughren 1984), Arandic BT contains a simplified phonology that conflates coronal contrasts and avoids rhotics and consonant clusters. Whilst standard Arandic words are mostly vowel-initial, this weak initial syllable is omitted in BT and the preferred CVCV(C) word structure is achieved through patterns of reduplication and truncation. The BT phonology becomes more complex with the perceived development of the child’s phonological competence, a case of fine-tuning over time.
Pre-stopped nasals and laterals occur in several Australian languages, but why? Nasal pre-stopping likely enhances cues to place of articulation contrasts (Butcher 1999, 2006). Though recent work proposes that lateral pre-stopping does likewise (Keyser & Stevens 2006: footnote 7; Loakes et al. 2008), the supporting argumentation appears unconvincing upon scrutiny. With reference to enhancement theory, articulatory phonology, and articulatory and acoustic phonetics, I speculate on: plausible diachronic sources of pre-stopped nasals via stochastic and functional processes; the extension of pre-stopping to laterals through a drive for regularity in the organisation of gesturally complex segments (Browman & Goldstein 1995; inter alia); and explanations for the asymmetrical prevalence of pre-stopping across places of articulation (Hercus 1972). Both language use and innate universals are crucial.
Pronoun systems are often described as a single linguistic category consisting of a paradigm of forms that serve as referential indices, or as “dual” systems consisting of more than one paradigm. In dual systems, a paradigm of “bound” pronouns serves the basic referential functions usually for core grammatical functions, with a paradigm of free pronouns used in contexts of emphasis or contrast, and in oblique grammatical roles (Bresnan 2001; Mushin & Simpson 2008). The same kinds of pragmatic splits can be systematically signalled in single pronoun system for example by means of prosody. Bresnan (2001: 4) claims that a system is only dual when there is a divergence in both form and syntactic distribution. In this chapter I examine the usefulness of this claim in defining pronoun systems as dual (or not dual) using evidence from Garrwa (Northern Australia). Here I argue that Garrwa pronouns function as a liminal system, neither single nor dual. This analysis raises questions about the degree to which categorical distinctions can be made in language description. I also consider implications this has for understanding the typology of pronouns more generally.
The Australian language Jingulu makes use of a small set of optional deictic markers on nominals which are homophonous with some of the verbal tense markers in the language. It is a fairly straightforward matter to argue that the former use is derived from the latter via re-analysis of clauses. However, the choice of theoretical framework can lead to very different synchronic analyses: the standard lexicalist approach listing the verbal and nominal uses as separate homophonous morphemes; while a late-insertion approach opens up the possibility that a single vocabulary item can appear in either the verbal or the nominal context, with its interpretation subject to morphosyntactic context alone.
This paper surveys the approaches that have been taken to the synchronic description of the inflectional classes of verbs of the Pama-Nyungan family of Australian languages, highlighting problems with the application of the concept of the morpheme, and the notion of “conjugation markers”. It then summarizes and critically assesses attempts to reconstruct the prehistory of Pama-Nyungan verb inflection, considering primarily the contrasting approaches of Dixon (1980, 2002) and Alpher (1990). The methodological requirements for such a reconstruction are then discussed. Finally, the proposed principles are applied first to the internal reconstruction of verbs in Walmajarri, a language of the Ngumpin-Yapa subgroup, then to the comparative reconstruction of some Pama-Nyungan (monosyllabic) verbs that display heterogeneous patterns of inflection.
Definiteness and Specificity are assumed to be universal semantic categories, but they are not marked in all languages. Languages with only two articles mark either Definiteness or Specificity, not both (Ionin 2003). I apply Chomsky’s theory of Derivation by Phase (2001a, 2001b) to the analysis of the specificity marker la in Mauritian Creole to argue that this morpheme must surface as a “last resort” to license the null definite article in some syntactic environments. Building on Chierchia’s (1998) Nominal Mapping Parameter – according to which nouns vary with respect to the features “argumental” or “predicative” – I propose that languages whose nouns are argumental lack a definite article and mark the specific vs. non-specific contrast. Languages whose nouns are predicative require an overt definite article and mark the definite vs. indefinite contrast.
This chapter examines the syntax and semantics of negative elements across Romance languages and proposes that there are three basic parameters that are encoded in the morphemes used to express sentential negation and negative determiners. These parameters define whether simple negation occurs pre- or post-inflection, whether the negation morpheme has a substantive meaning (e.g. Spanish no, Italian non, French pas) or expletive (e.g. French ne) value or else is ambiguous between the two (e.g. Catalan no), and finally whether negative phrases are ambiguous between “negative operator” and “polarity item” status. The proposal stresses typological implications between having post-inflection negation and post-inflection negative phrases that do not require co-occurrence with a sentential negation. It also tries to explain long puzzling cross-linguistic differences in locality constraints on negative dependencies. The theoretical focus of the work is on exploring how minimalist research on syntactic diversity should be conducted and in which formats its results could be formulated in a rigorous explanatory way.
Australian languages are traditionally not thought of as having serial verb constructions (although cf. Goddard 1988; Wilkins 1988), and are therefore rarely discussed in the extensive typological and theoretical literature on verb serialization. However, in recent work Laughren (2009, 2012) has reported on the existence of serial verb constructions in Waanyi, a non–Pama-Nyungan language of northern Australia. In this paper I show that serial verb constructions are also present in Wambaya, another non–Pama-Nyungan language which shares some areal and lexical similarities with Waanyi. I show that the serial verbs in Wambaya exhibit many of the key morphosyntactic and semantic properties described as characteristic for serial verb constructions in the literature (e.g. Sebba 1987; Durie 1997; Aikhenvald 2006a). A particularly interesting property of Wambaya serial verb constructions, and one that is highly unusual cross-linguistically, is that the verbs need not be ordered iconically. I suggest that this anti-iconic ordering may be related to Wambaya’s nonconfigurational clausal structure and its free word order possibilities, highlighting the need for consideration of the full typological range of language structures in the analysis of verb serialization cross-linguistically.
Generativists have argued that nominals in non-configurational languages such as Warlpiri do not have the status of arguments. This paper provides new evidence for this claim from an unlikely source: code-switching between Kriol, an English-based creole, and Gurindji, a Ngumpin-Yapa language closely related to Warlpiri. In Gurindji-Kriol code-switching, case-marked nominals are non-obligatory and, where they do occur, they are cross-referenced with a pronoun. This pattern is found even where Kriol sets the morpho-syntactic frame and provides the pronouns. The structure reflects that of monolingual Gurindji where bound pronouns are obligatory and nominals are optional. Given the resistance of inflectional morphology to switching, it is unexpected to find Gurindji case-marked nominals present in an otherwise Kriol morpho-syntactic frame. Nonetheless, structural material can enter into a code-switched clause via structural islands which do not participate in the predicate argument structure of the code-switched clause. This paper argues that case-marked nominals are structural islands and this adjunct-like structure must have been available in the source language, Gurindji, thereby providing further evidence for the non-argument status of Gurindji nominals.
Three somewhat conflicting semantic theories of the Futur Simple (FS) in spoken European French are discussed here: Fleischman’s (1982) “diachronic model”, which portrays the current semantics of the FS as largely modal; Schrott’s (1997) “conditioning” model, which views both the FS and Futur Composé as temporal and modal exponents selected in terms of context of use; Sokol’s (1999) “aspectual model” which sees the FS not as a tense but as a perfective aspect marker interacting with different verb action-classes. Evidence from empirical work relating to these theories does not support the modal view of the FS, and raises some questions for the aspectual model. Robust support emerges, however, for the FS as a future tense.
As a deadjectival verb, English clean should be a clear-cut result verb, yet it often shows hallmarks of a manner verb. This paper investigates this dual behavior in light of manner/result complementarity: the proposal that verbs lexicalize either manner or result meaning components, but not both. We demonstrate that once lexicalized and contextually determined meaning components are distinguished, clean conforms to manner/result complementarity. It can be a result verb, not entailing a particular manner. However, given its strong association with cleaning routines, some uses simply lexicalize manner. Crucially, in manner uses the result drops out, consistent with manner/result complementarity. The manner-only and result-only uses of clean, then, instantiate related senses, each conforming to manner/result complementarity.
This chapter examines the connections between shared cultural knowledge about kinship structure and the pragmatic inferences that enable interlocutors to assess each other’s (multiple) perspectives. By drawing on Bininj Gunwok conversational data this chapter shows how linguistic choices are influenced by the dynamics of social relationships, particularly by context-specific speaker goals and stance-taking that focuses on intersubjectivity. The choice of kinterm is an essential component of stance-taking. A switch in kinterm shifts the indexes of various aspects of speaker agency (e.g. effecting solidarity, ratifying relationships with addressee and referent, justifying joint action). Kinterms in four different contexts are discussed: basic kin terms for affines; basic kin terms for skewed relationships; kin term shifts within unnamed superclasses; and triadic terms.
The Bilingual Education Program in the Northern Territory was established in 1973 to deliver an effective education to students in remote Indigenous schools. In 2008, the NT Government closed the Program, citing low student achievement in English literacy as its rationale. This chapter evaluates the program according to a wider set of criteria, with particular focus on the Warlpiri schools. It considers the training of local staff, community involvement in schools, the development of curriculum, and extensive literature and linguistic documentation. These criteria are not prominent in government evaluations, yet are identified in national and international literature as key to achieving student outcomes (UNESCO 2008a, 2008b; Silburn et al. 2011) and echo priorities expressed by Indigenous adults involved in education.
Baby Talk (BT), also known as child-directed speech, is a non-standard form of speech used by adults when talking to infants. In Arandic languages BT involves the use of a small set of unique but widely known words, onomatopoeic-derived words, as well as phonological modifications to standard vocabulary. As in neighboring Warlpiri (Laughren 1984), Arandic BT contains a simplified phonology that conflates coronal contrasts and avoids rhotics and consonant clusters. Whilst standard Arandic words are mostly vowel-initial, this weak initial syllable is omitted in BT and the preferred CVCV(C) word structure is achieved through patterns of reduplication and truncation. The BT phonology becomes more complex with the perceived development of the child’s phonological competence, a case of fine-tuning over time.
Pre-stopped nasals and laterals occur in several Australian languages, but why? Nasal pre-stopping likely enhances cues to place of articulation contrasts (Butcher 1999, 2006). Though recent work proposes that lateral pre-stopping does likewise (Keyser & Stevens 2006: footnote 7; Loakes et al. 2008), the supporting argumentation appears unconvincing upon scrutiny. With reference to enhancement theory, articulatory phonology, and articulatory and acoustic phonetics, I speculate on: plausible diachronic sources of pre-stopped nasals via stochastic and functional processes; the extension of pre-stopping to laterals through a drive for regularity in the organisation of gesturally complex segments (Browman & Goldstein 1995; inter alia); and explanations for the asymmetrical prevalence of pre-stopping across places of articulation (Hercus 1972). Both language use and innate universals are crucial.
Pronoun systems are often described as a single linguistic category consisting of a paradigm of forms that serve as referential indices, or as “dual” systems consisting of more than one paradigm. In dual systems, a paradigm of “bound” pronouns serves the basic referential functions usually for core grammatical functions, with a paradigm of free pronouns used in contexts of emphasis or contrast, and in oblique grammatical roles (Bresnan 2001; Mushin & Simpson 2008). The same kinds of pragmatic splits can be systematically signalled in single pronoun system for example by means of prosody. Bresnan (2001: 4) claims that a system is only dual when there is a divergence in both form and syntactic distribution. In this chapter I examine the usefulness of this claim in defining pronoun systems as dual (or not dual) using evidence from Garrwa (Northern Australia). Here I argue that Garrwa pronouns function as a liminal system, neither single nor dual. This analysis raises questions about the degree to which categorical distinctions can be made in language description. I also consider implications this has for understanding the typology of pronouns more generally.
The Australian language Jingulu makes use of a small set of optional deictic markers on nominals which are homophonous with some of the verbal tense markers in the language. It is a fairly straightforward matter to argue that the former use is derived from the latter via re-analysis of clauses. However, the choice of theoretical framework can lead to very different synchronic analyses: the standard lexicalist approach listing the verbal and nominal uses as separate homophonous morphemes; while a late-insertion approach opens up the possibility that a single vocabulary item can appear in either the verbal or the nominal context, with its interpretation subject to morphosyntactic context alone.
This paper surveys the approaches that have been taken to the synchronic description of the inflectional classes of verbs of the Pama-Nyungan family of Australian languages, highlighting problems with the application of the concept of the morpheme, and the notion of “conjugation markers”. It then summarizes and critically assesses attempts to reconstruct the prehistory of Pama-Nyungan verb inflection, considering primarily the contrasting approaches of Dixon (1980, 2002) and Alpher (1990). The methodological requirements for such a reconstruction are then discussed. Finally, the proposed principles are applied first to the internal reconstruction of verbs in Walmajarri, a language of the Ngumpin-Yapa subgroup, then to the comparative reconstruction of some Pama-Nyungan (monosyllabic) verbs that display heterogeneous patterns of inflection.
Definiteness and Specificity are assumed to be universal semantic categories, but they are not marked in all languages. Languages with only two articles mark either Definiteness or Specificity, not both (Ionin 2003). I apply Chomsky’s theory of Derivation by Phase (2001a, 2001b) to the analysis of the specificity marker la in Mauritian Creole to argue that this morpheme must surface as a “last resort” to license the null definite article in some syntactic environments. Building on Chierchia’s (1998) Nominal Mapping Parameter – according to which nouns vary with respect to the features “argumental” or “predicative” – I propose that languages whose nouns are argumental lack a definite article and mark the specific vs. non-specific contrast. Languages whose nouns are predicative require an overt definite article and mark the definite vs. indefinite contrast.
This chapter examines the syntax and semantics of negative elements across Romance languages and proposes that there are three basic parameters that are encoded in the morphemes used to express sentential negation and negative determiners. These parameters define whether simple negation occurs pre- or post-inflection, whether the negation morpheme has a substantive meaning (e.g. Spanish no, Italian non, French pas) or expletive (e.g. French ne) value or else is ambiguous between the two (e.g. Catalan no), and finally whether negative phrases are ambiguous between “negative operator” and “polarity item” status. The proposal stresses typological implications between having post-inflection negation and post-inflection negative phrases that do not require co-occurrence with a sentential negation. It also tries to explain long puzzling cross-linguistic differences in locality constraints on negative dependencies. The theoretical focus of the work is on exploring how minimalist research on syntactic diversity should be conducted and in which formats its results could be formulated in a rigorous explanatory way.
Australian languages are traditionally not thought of as having serial verb constructions (although cf. Goddard 1988; Wilkins 1988), and are therefore rarely discussed in the extensive typological and theoretical literature on verb serialization. However, in recent work Laughren (2009, 2012) has reported on the existence of serial verb constructions in Waanyi, a non–Pama-Nyungan language of northern Australia. In this paper I show that serial verb constructions are also present in Wambaya, another non–Pama-Nyungan language which shares some areal and lexical similarities with Waanyi. I show that the serial verbs in Wambaya exhibit many of the key morphosyntactic and semantic properties described as characteristic for serial verb constructions in the literature (e.g. Sebba 1987; Durie 1997; Aikhenvald 2006a). A particularly interesting property of Wambaya serial verb constructions, and one that is highly unusual cross-linguistically, is that the verbs need not be ordered iconically. I suggest that this anti-iconic ordering may be related to Wambaya’s nonconfigurational clausal structure and its free word order possibilities, highlighting the need for consideration of the full typological range of language structures in the analysis of verb serialization cross-linguistically.
Generativists have argued that nominals in non-configurational languages such as Warlpiri do not have the status of arguments. This paper provides new evidence for this claim from an unlikely source: code-switching between Kriol, an English-based creole, and Gurindji, a Ngumpin-Yapa language closely related to Warlpiri. In Gurindji-Kriol code-switching, case-marked nominals are non-obligatory and, where they do occur, they are cross-referenced with a pronoun. This pattern is found even where Kriol sets the morpho-syntactic frame and provides the pronouns. The structure reflects that of monolingual Gurindji where bound pronouns are obligatory and nominals are optional. Given the resistance of inflectional morphology to switching, it is unexpected to find Gurindji case-marked nominals present in an otherwise Kriol morpho-syntactic frame. Nonetheless, structural material can enter into a code-switched clause via structural islands which do not participate in the predicate argument structure of the code-switched clause. This paper argues that case-marked nominals are structural islands and this adjunct-like structure must have been available in the source language, Gurindji, thereby providing further evidence for the non-argument status of Gurindji nominals.
Three somewhat conflicting semantic theories of the Futur Simple (FS) in spoken European French are discussed here: Fleischman’s (1982) “diachronic model”, which portrays the current semantics of the FS as largely modal; Schrott’s (1997) “conditioning” model, which views both the FS and Futur Composé as temporal and modal exponents selected in terms of context of use; Sokol’s (1999) “aspectual model” which sees the FS not as a tense but as a perfective aspect marker interacting with different verb action-classes. Evidence from empirical work relating to these theories does not support the modal view of the FS, and raises some questions for the aspectual model. Robust support emerges, however, for the FS as a future tense.
As a deadjectival verb, English clean should be a clear-cut result verb, yet it often shows hallmarks of a manner verb. This paper investigates this dual behavior in light of manner/result complementarity: the proposal that verbs lexicalize either manner or result meaning components, but not both. We demonstrate that once lexicalized and contextually determined meaning components are distinguished, clean conforms to manner/result complementarity. It can be a result verb, not entailing a particular manner. However, given its strong association with cleaning routines, some uses simply lexicalize manner. Crucially, in manner uses the result drops out, consistent with manner/result complementarity. The manner-only and result-only uses of clean, then, instantiate related senses, each conforming to manner/result complementarity.
This chapter examines the connections between shared cultural knowledge about kinship structure and the pragmatic inferences that enable interlocutors to assess each other’s (multiple) perspectives. By drawing on Bininj Gunwok conversational data this chapter shows how linguistic choices are influenced by the dynamics of social relationships, particularly by context-specific speaker goals and stance-taking that focuses on intersubjectivity. The choice of kinterm is an essential component of stance-taking. A switch in kinterm shifts the indexes of various aspects of speaker agency (e.g. effecting solidarity, ratifying relationships with addressee and referent, justifying joint action). Kinterms in four different contexts are discussed: basic kin terms for affines; basic kin terms for skewed relationships; kin term shifts within unnamed superclasses; and triadic terms.