Competing motivations are often at work in the choice of form and meaning of possessive and associative noun phrases. The article offers a broad typological review of the ways of expressing possession at the NP-internal level. In particular, it discusses how iconicity and economic motivations interact in determining the shape of NP-internal possessive expressions, also addressing the topics of how socio-cultural factors affect the grammar of possession and how societal changes are reflected in language variation.
This study introduces the possessive schemas of revived Cornish. By means of a survey and interviews with 25 fluent speakers and a corpus study I identified several possessive schemas being employed for different purposes due to semantic and structural reasons. Possession splits, which are also attested for the other Brythonic languages (Stolz et al. 2008), occur for the distinction of temporary possession and the possession of illnesses. Furthermore, the speakers may be developing a new form to express the possession of abstract nouns by employing the short form of bos ‘to be’ and the preposition dhe ‘to’. For the development of this form the history of interrupted language transmission and the use of Cornish by New Speakers are relevant factors.
Turkic canonical possessive noun phrases consist of the possessor marked with the genitive, and the possessee with a possessive suffix. This study deals with Turkic non-canonical constructions, where a genitive marker attaches to the possessor, but the possessee remains unmarked. It is argued that in the non-canonical construction, the frame of reference is delimited to the world known/familiar to the interlocutors through shared knowledge of the world or shared discourse. Thus, these constructions express the concept of “familiarity” with possible overtones of empathy or endearment, or with negative connotations. As the speaker and hearer are typical participants in the deictic system, non-canonical possessives are most frequently used with first- and second-person possessors.
In this paper, we analyse and describe the HAVE-constructions in South Saami (Saamic, Uralic), from a comparative perspective with other Saamic and Uralic languages. The Saamic languages can be divided into three subgroups: in the first, HAVE is expressed with a verb meaning ‘to be’; the second has a HAVE-construction based on a verb ‘to have’ and one on the verb ‘to be’. The third subgroup comprises only of South Saami, which has three HAVE-constructions, one based on the verb ‘to have’ and two on the verb ‘to be’. South Saami is also unique among Saamic languages in that it has a HAVE-construction based on the verb ‘to be’ where the possessor in the genitive.
This article studies syntactic, semantic and discursive properties of non-subordinate (main) clauses conveying possession in Oghuz and Kipchak Turkic languages. In Turkic, the concept of possession is typically encoded by clauses based on existential predicates. The language-specific and crosslinguistic properties of two predicate types, {bar} and {bol}, will be contrastively surveyed. As for the marking of possessor in clauses containing {bar}, three patterns will be described, one of which is a contact-induced structure restricted to Turkic varieties in Iran. As a multifunctional verb, {bol} can convey, among other things, dynamic or static possession. The results indicate that the clauses based on the static possession marker {bol} are more operative in Kipchak languages and in Turkmen (East Oghuz), than in West Oghuz languages. It will further be shown that the structures based on {bar} or the static marker {bol} typically exhibit discourse-related distribution in the respective languages.
This article analyses two types of possessive constructions in Czech: pronominal attributive possession and adjectival attributive possession. By means of a usage-based approach to language, the aim of the present corpus investigation is to reveal relevant patterns of grammar and usage in order to explain the evolution of the concerned constructions. The main focus is put on spoken language. I analyse the given data from the Czech corpora of both spoken (ORAL2013) and written language (SYN2010). I test the hypotheses concerning different frequency distributions, and the position of the possessor on the prominence hierarchies. The results support functional explanations of emergence of the pronominal and adjectival constructions, based on the prominent status of the possessor.
This paper describes the strategies employed by North Saami (Uralic, Saamic) and Norwegian (Indo-European, North Germanic) to express predicative possession. It shows that both North Saami and Norwegian do not provide evidence of splits in their possession systems; rather, they use the same constructions to express both ownership as well as less prototypical possessive notions, displaying a typical European behaviour. The expression of some notions contiguous to possession, namely experience, location and attribution is also analyzed: these are rarely coded by means of possessive constructions in Norwegian, but more often in North Saami.
The article deals with the linguistic expression of possession and ownership in Modern Uyghur, a Turkic language spoken in Northern China. It describes the structure and semantics of Uyghur attributive and predicative possessive constructions and their interaction with the categories of inalienability, tense, aspect and mood. The paper distinguishes canonical and non-canonical possessive constructions, and shows their semantic and structural features. Special attention is paid to the structure of complex constructions with possessors and possessees as subjects of dependent clauses with the predicate in the -(i)p converb form. The relations of inalienability between the subjects of the dependent and main clauses license this use of the -(i)p converb.
The article describes the superlative, hyperlative and elative use of formally possessive constructions in a number of Turkic languages from a comparative perspective, analyzing their structural and semantic types and pragmatic properties. Similar possessive superlative constructions are found all over Eurasia in languages belonging to various language families: they mostly express abstract (absolute) comparison of referred entities where their quantities are compared with a certain norm. One of the most unclear issues of possessive superlative constructions is their origin. They could have emerged as structural copies of corresponding Semitic Biblical expressions or as effect of language internal developments. The material of corresponding English and German non-canonical possessive constructions allows assuming that both factors have contributed to their emergence.
The article describes the functions and the usage of the genitive in Khinalug, a Nakh-Dagestanian language spoken in the North-East of Azerbaijan. Khinalug stands out for its tripartite genitive system with two subsystems: (a) a possessee-based subsystem, which distinguishes alienability versus certain types of inalienability according to the degree of bondedness towards the possessee; only animate possessors can take an inalienably marked possessee. (b) a possessor-based subsystem, which distinguishes alienability versus family-relatedness according to the bondedness among the members of the possessor group towards each other: They have to be in a family relationship to justify the use of this genitive. Moreover, independently from animacy, the alienable and the inalienable genitive are governed by a range of postpositions.
Icelandic makes use of different morphosyntactic patterns for different kinds of possessive relations both in predicative and attributive possession. The fact that this differentiation is at least partly motivated by the semantics of the possessum suggests an alienability split in the Icelandic possession system (Stolz 2008). There are various approaches to explain the formal distinctions in the system, one of them depending on grammaticalization processes. Following this approach, a diachronic perspective needs to be taken in order to understand how splits arise. Accordingly, this paper presents the results of a corpus-linguistic and frequency-based analysis of the attributive pronominal possessive system of Icelandic. The most significant finding to emerge is that splits in the possessive system do not only manifest in a different synchronic behaviour but also in different degrees of susceptibility to language-internal and/or contact-induced changes.
The paper is focused on the functions of possessive suffixes in the Beserman dialect of Udmurt. Considering the data from the Beserman corpus of oral texts we find the parameters influencing the presence/omission of the possessive suffixes in the contexts of possessive (alienable and inalienable) and non-possessive contexts. We review the claim about the grammaticalization of the Beserman possessives into markers of definiteness, and suggest that they are subject to pragmaticization in terms of Fried (2009).
This study aims at pinpointing certain properties of belong-constructions which might prove helpful for a future cross-linguistic inquiry into the grammar of those categories which associate with the notion belong in particular and with predicative possession in general. To this end, languages from two different regions – Mexico and Europe – are reviewed as to the evidence they provide of belong-constructions and their relation to other predicative-possessive categories. Cases of areal and genetic micro-variation in the domain of belong are discussed. The methodological issue of collecting data which are usually not featured in the descriptive-linguistic literature is raised as well.
This paper illustrates the grammaticalisation of the preposition de as a genitive marker through the analysis of Early Medieval notary deeds written in northern and central Italy (Codice diplomatico longobardo, CDL). In Classical Latin, de is used mainly as a verbal adjunct; its usage at the noun phrase level is sporadic and semantically determined, conveying mainly the meaning of ‘source’. By the time of the CDL documents, de has a more grammaticalised status and a higher frequency, but has not yet affected the expression of kinship and ownership. Semantic factors such as the prototypicality of the possessive relation and the degree of animacy of the modifier can motivate this concurrence between the synthetic and the analytical pattern of adnominal possession.
Competing motivations are often at work in the choice of form and meaning of possessive and associative noun phrases. The article offers a broad typological review of the ways of expressing possession at the NP-internal level. In particular, it discusses how iconicity and economic motivations interact in determining the shape of NP-internal possessive expressions, also addressing the topics of how socio-cultural factors affect the grammar of possession and how societal changes are reflected in language variation.
This study introduces the possessive schemas of revived Cornish. By means of a survey and interviews with 25 fluent speakers and a corpus study I identified several possessive schemas being employed for different purposes due to semantic and structural reasons. Possession splits, which are also attested for the other Brythonic languages (Stolz et al. 2008), occur for the distinction of temporary possession and the possession of illnesses. Furthermore, the speakers may be developing a new form to express the possession of abstract nouns by employing the short form of bos ‘to be’ and the preposition dhe ‘to’. For the development of this form the history of interrupted language transmission and the use of Cornish by New Speakers are relevant factors.
Turkic canonical possessive noun phrases consist of the possessor marked with the genitive, and the possessee with a possessive suffix. This study deals with Turkic non-canonical constructions, where a genitive marker attaches to the possessor, but the possessee remains unmarked. It is argued that in the non-canonical construction, the frame of reference is delimited to the world known/familiar to the interlocutors through shared knowledge of the world or shared discourse. Thus, these constructions express the concept of “familiarity” with possible overtones of empathy or endearment, or with negative connotations. As the speaker and hearer are typical participants in the deictic system, non-canonical possessives are most frequently used with first- and second-person possessors.
In this paper, we analyse and describe the HAVE-constructions in South Saami (Saamic, Uralic), from a comparative perspective with other Saamic and Uralic languages. The Saamic languages can be divided into three subgroups: in the first, HAVE is expressed with a verb meaning ‘to be’; the second has a HAVE-construction based on a verb ‘to have’ and one on the verb ‘to be’. The third subgroup comprises only of South Saami, which has three HAVE-constructions, one based on the verb ‘to have’ and two on the verb ‘to be’. South Saami is also unique among Saamic languages in that it has a HAVE-construction based on the verb ‘to be’ where the possessor in the genitive.
This article studies syntactic, semantic and discursive properties of non-subordinate (main) clauses conveying possession in Oghuz and Kipchak Turkic languages. In Turkic, the concept of possession is typically encoded by clauses based on existential predicates. The language-specific and crosslinguistic properties of two predicate types, {bar} and {bol}, will be contrastively surveyed. As for the marking of possessor in clauses containing {bar}, three patterns will be described, one of which is a contact-induced structure restricted to Turkic varieties in Iran. As a multifunctional verb, {bol} can convey, among other things, dynamic or static possession. The results indicate that the clauses based on the static possession marker {bol} are more operative in Kipchak languages and in Turkmen (East Oghuz), than in West Oghuz languages. It will further be shown that the structures based on {bar} or the static marker {bol} typically exhibit discourse-related distribution in the respective languages.
This article analyses two types of possessive constructions in Czech: pronominal attributive possession and adjectival attributive possession. By means of a usage-based approach to language, the aim of the present corpus investigation is to reveal relevant patterns of grammar and usage in order to explain the evolution of the concerned constructions. The main focus is put on spoken language. I analyse the given data from the Czech corpora of both spoken (ORAL2013) and written language (SYN2010). I test the hypotheses concerning different frequency distributions, and the position of the possessor on the prominence hierarchies. The results support functional explanations of emergence of the pronominal and adjectival constructions, based on the prominent status of the possessor.
This paper describes the strategies employed by North Saami (Uralic, Saamic) and Norwegian (Indo-European, North Germanic) to express predicative possession. It shows that both North Saami and Norwegian do not provide evidence of splits in their possession systems; rather, they use the same constructions to express both ownership as well as less prototypical possessive notions, displaying a typical European behaviour. The expression of some notions contiguous to possession, namely experience, location and attribution is also analyzed: these are rarely coded by means of possessive constructions in Norwegian, but more often in North Saami.
The article deals with the linguistic expression of possession and ownership in Modern Uyghur, a Turkic language spoken in Northern China. It describes the structure and semantics of Uyghur attributive and predicative possessive constructions and their interaction with the categories of inalienability, tense, aspect and mood. The paper distinguishes canonical and non-canonical possessive constructions, and shows their semantic and structural features. Special attention is paid to the structure of complex constructions with possessors and possessees as subjects of dependent clauses with the predicate in the -(i)p converb form. The relations of inalienability between the subjects of the dependent and main clauses license this use of the -(i)p converb.
The article describes the superlative, hyperlative and elative use of formally possessive constructions in a number of Turkic languages from a comparative perspective, analyzing their structural and semantic types and pragmatic properties. Similar possessive superlative constructions are found all over Eurasia in languages belonging to various language families: they mostly express abstract (absolute) comparison of referred entities where their quantities are compared with a certain norm. One of the most unclear issues of possessive superlative constructions is their origin. They could have emerged as structural copies of corresponding Semitic Biblical expressions or as effect of language internal developments. The material of corresponding English and German non-canonical possessive constructions allows assuming that both factors have contributed to their emergence.
The article describes the functions and the usage of the genitive in Khinalug, a Nakh-Dagestanian language spoken in the North-East of Azerbaijan. Khinalug stands out for its tripartite genitive system with two subsystems: (a) a possessee-based subsystem, which distinguishes alienability versus certain types of inalienability according to the degree of bondedness towards the possessee; only animate possessors can take an inalienably marked possessee. (b) a possessor-based subsystem, which distinguishes alienability versus family-relatedness according to the bondedness among the members of the possessor group towards each other: They have to be in a family relationship to justify the use of this genitive. Moreover, independently from animacy, the alienable and the inalienable genitive are governed by a range of postpositions.
Icelandic makes use of different morphosyntactic patterns for different kinds of possessive relations both in predicative and attributive possession. The fact that this differentiation is at least partly motivated by the semantics of the possessum suggests an alienability split in the Icelandic possession system (Stolz 2008). There are various approaches to explain the formal distinctions in the system, one of them depending on grammaticalization processes. Following this approach, a diachronic perspective needs to be taken in order to understand how splits arise. Accordingly, this paper presents the results of a corpus-linguistic and frequency-based analysis of the attributive pronominal possessive system of Icelandic. The most significant finding to emerge is that splits in the possessive system do not only manifest in a different synchronic behaviour but also in different degrees of susceptibility to language-internal and/or contact-induced changes.
The paper is focused on the functions of possessive suffixes in the Beserman dialect of Udmurt. Considering the data from the Beserman corpus of oral texts we find the parameters influencing the presence/omission of the possessive suffixes in the contexts of possessive (alienable and inalienable) and non-possessive contexts. We review the claim about the grammaticalization of the Beserman possessives into markers of definiteness, and suggest that they are subject to pragmaticization in terms of Fried (2009).
This study aims at pinpointing certain properties of belong-constructions which might prove helpful for a future cross-linguistic inquiry into the grammar of those categories which associate with the notion belong in particular and with predicative possession in general. To this end, languages from two different regions – Mexico and Europe – are reviewed as to the evidence they provide of belong-constructions and their relation to other predicative-possessive categories. Cases of areal and genetic micro-variation in the domain of belong are discussed. The methodological issue of collecting data which are usually not featured in the descriptive-linguistic literature is raised as well.
This paper illustrates the grammaticalisation of the preposition de as a genitive marker through the analysis of Early Medieval notary deeds written in northern and central Italy (Codice diplomatico longobardo, CDL). In Classical Latin, de is used mainly as a verbal adjunct; its usage at the noun phrase level is sporadic and semantically determined, conveying mainly the meaning of ‘source’. By the time of the CDL documents, de has a more grammaticalised status and a higher frequency, but has not yet affected the expression of kinship and ownership. Semantic factors such as the prototypicality of the possessive relation and the degree of animacy of the modifier can motivate this concurrence between the synthetic and the analytical pattern of adnominal possession.