Manifestations of oblique case in Serbian/Croatian (specifically, the dative and the instrumental... more Manifestations of oblique case in Serbian/Croatian (specifically, the dative and the instrumental) present a puzzling pattern of distribution. The paper proposes an analysis for the source of the relevant case phenomena that can resolve the puzzle, rendering it a direct consequence of an independently attested property, namely, the Case Realization Condition of oblique case in the language. It also provides striking evidence regarding the nature and assigner of oblique case in general. The phenomena and their account will lead to the conclusion that contrary to appearances, and in contrast to the standard assumption, lexically governed cases can only be assigned by (overt or null) instances of the category P.
Cross-language studies of linguistic phenomena often reveal facts which are unknown to the lingui... more Cross-language studies of linguistic phenomena often reveal facts which are unknown to the linguist who is concerned with the study of one particular language or language family. Certain rules or processes repeat themselves in unrelated languages and this is expected, since all mankind is subject to the same neurological, physiological and psychological constraints upon his language. In phonology, for example, velar palatalization before high front vowels, word-final devoicing of obstruents and the assimilation of nasal stops in point of articulation to following consonants are commonly found and, if these processes vary from language to language, it is probably a matter of degree only. One such process will be examined here in some detail: voicing assimilation (VA). The starting point is the seemingly recalcitrant behavior of the voiced labio-dental fricative [v] in Russian. Evidence brought from a Finno-Ugric language, Hungarian, and from a Semitic language, Hebrew, shows that the intransigence of [v] in the Slavic language is also found in these two genetically unrelated languages. Implications will be discussed and an explanation, based on a sonority hierarchy, will be offered.
Based on data from different types of parasitic gap (PG) constructions drawn primarily from Hunga... more Based on data from different types of parasitic gap (PG) constructions drawn primarily from Hungarian, the present study investigates and reassesses the longstanding issue of the ränge and possible source(s) of the so-called 'anti-ccommand' effects in PG constructions. Special attention will be paid to the facts of PG licensing by subject variables, and to the way particular theories of PG licensing asymmetries bear on hypotheses of nonconfigurationality of clausestructure, äs in the allegedly nonconfigurational Hungarian. We will address the question of whether the füll ränge of subject/object asymmetries observable in PG licensing are to be derived (a) based on notions of hierarchical structure — such äs c-command — (see e.g. Chomsky 1982; Kayne 1983; Browning 1987), (b) based on some nonstructural difference between subjects vs. objects, such äs, e.g., the particular Case they bear (äs argued in Kiss 1985), or (c) jointly by (a) and (b). First we will argue that the theory of parasitic chain licensing proposed in Kiss (1985) which crucially excludes the relevance of anti-c-command, and replaces it by a 'Case-compatibility' requirement fails to provide an adequate alternative (nonconfigurational) account for the pattern of subject/object asymmetries involved both on empirical and on conceptual grounds. It will be shown that the Case-compatibility requirement is only a necessary, but not a sufficient condition for deriving the subject/object asymmetries observable with respect to the licensing of parasitic chains, and furthermore that the neutralization of Condition C within PG constructions attempted in Kiss (1985) leads to a paradox and hence her framework would end up retaining anti-c-command effects in any case. Given these points, the cases to be discussed not only indicate the relevance of hierarchical structure in the licensing of PGs (contrary to purely Case-based accounts, such äs in Kiss 1985), they also constitute an
This article appeared in a journal published by Elsevier. The attached copy is furnished to the a... more This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier's archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright
This paper reassesses the genetic classification customarily associated with what we refer to as ... more This paper reassesses the genetic classification customarily associated with what we refer to as "unspoken languages," that is, language forms that are, or were at some earlier stage, fulfilling exclusively nonspoken functions (e.g. Medieval and Modern Hebrew, Medieval Latin). We argue that if claims of genetic affiliation are to have any empirical linguistic content, and not simply sociopolitical significance, then genetic descendance from an earlier linguistic form must require the existence of uninterrupted normal transmission. Since the conditions of transmission attested in the case of "unspoken languages" arguably fail to qualify as such, we conclude that these cannot possibly be genetic descendants of the older spoken languages whose names they usually bear. We present our own hypothesis as to the nature and genesis of "unspoken languages," with particular emphasis on the process of relexification of the user's native language with lexical items of the "ancestor" language. As a revealing test case, we discuss Modern Hebrew, claiming that its apparent affinity to Biblical and Mishnaic Hebrew is indeed only a reflection of relexification, not of descendance. Finally, we point out some general methodological implications for genetic and historical linguistics.
Pre-theoretically, wh-interrogatives are commonly assumed to divide languages into three basic ty... more Pre-theoretically, wh-interrogatives are commonly assumed to divide languages into three basic types: (i) wh-movement (of the English-type), (ii) wh-in-situ, and (iii) multiple wh-fronting languages. Although the proposed accounts vary, advancing different hypotheses regarding the structure of wh-phrases, the formal features and interpretive mechanisms that trigger the relevant movements and provide the appropriate interpretations, yet they share the assumption that if an interrogative wh-construction requires movement of a wh-interrogative element, this movement targets the edge of CP (Spec, CP position) (see in particular Cheng's (1991) clause-typing hypothesis, and subsequent work such as Tsai (1994), Watanabe (1992), Richards (1997), Hagstrom (1998), Cable (2010)). As is also widely-known, though Hungarian wh-interrogatives require overt movement of a wh-phrase, the landing site of these wh-phrases is clearly not the Spec, CP position. The prevailing view has been that interrogative wh-phrases in Hungarian land in the position that is known as the designated structural Focus position, namely a position leftadjacent to V in Hungarian clause structure (hierarchically clearly lower than material occurring to its left, namely C and a variety of topic and quantifier positions). Based on this observation it is commonly concluded that what looks like wh-movement in fact is an instance of "Focusmovement"; accordingly, the wh-interrogatives (at least one, if there are multiple) are claimed to bear a [Focus] feature, the same feature that was assumed to drive the movement of non-wh phrases to the designated Focus position in the language (Brody 1990, Horvath 1981, 1986, Lipták 2001). Moreover, this is far from being an isolated property of Hungarian. As is well-known, a wide range of languages exhibit syntactic and/or morphological parallelism between their whinterrogative and (non-wh) Focus constructions. An early proposal to formally capture this crosslinguistic observation Horvath's (1986, sect. 2.3), hypothesis according to which the interpretation of a wh-operator can be a non-echo interrogative only if it comes to bear the syntactic feature [Focus]. In more recent work (Horvath 2000, 2007), I argued that the widely-assumed "Focus movement" to a designated structural position in Hungarian is in fact not movement of Focus at all, and subsequently advanced the proposal (Horvath 2010) that contrary to common assumptions, no formal feature [Focus], or any other non-truth conditional, "discourse", feature, is encoded and active in the computational system. The Hungarian "Focus-movement" construction was instead argued to involve (overt) syntactic movement driven by a quantificational feature. The proposal postulated a (phonologically null) Exhaustive Identification (EI) operator, performing "exclusion by identification" (in the sense of Kenesei 1986) on a set of pragmatically given alternatives for which the predicate can potentially hold. Specifically I claimed that the EI-operator gets merged as Spec of DP, drawing a parallel with the case of the Q-operator of Japanese interrogative whphrases (under Tsai's (1994) analysis). The EI-operator was assumed to associate with Focus-a constituent bearing main stress within its c-command domain-on a par with other focus-sensitive operators (such as ONLY or EVEN). A corresponding functional head EI 0 in the clausal projection served as probe for the EI operator feature, establishing an Agree relation with the EI-operator phrase, and triggering overt phrasal movement to Spec, EIP. Taking the above results as a point of departure, the present paper will examine and reevaluate the status, driving mechanism and landing site(s) of interrogative wh-movements, with the aim to (a) further test and elaborate the EI-operator movement proposal in relation to this challenging new domain, and (b) assess potential implications with respect to the minimalist framework's rationale for and alternative implementations of syntactic displacements. Specifically, parallelisms and discrepancies in distribution and interpretation displayed by interrogative wh and (non-wh) EI-operator movements will be examined within Hungarian and cross-linguistically, with regard to the position of focus (pitch-accent-bearing constituent) and wh-elements within their respective phrases ("pied-piping" options), with regard to the range and limitations of multiple and "mixed" occurrences of interrogative-wh and Focus phrases in a clause, as well as manifestations of locality, "scope-marking", and the potential role of PF constraints.
Die Deutsche Bibliothek-CIP-Einheitsaufnahme Relexiflcation in Creole and Non-Creole languages: w... more Die Deutsche Bibliothek-CIP-Einheitsaufnahme Relexiflcation in Creole and Non-Creole languages: with special attention to Haitian Creole, modern Hebrew, Romani, and Rumanian/ ed. by Julia Horvath and Paul Wexler.-Wiesbaden: Harrassowitz, 1997 ( ...
Natural Language and Linguistic Theory, May 1, 1987
... Note in particular the verb a termina 'to finish', which occurs in (5a), (12a) and (1... more ... Note in particular the verb a termina 'to finish', which occurs in (5a), (12a) and (13a), and which was shown in (19) not to ... Although the grammatical realization of this semantic property in languages often does involve non-finite forms (eg, in English ORIOD constructions), it can ...
This paper addresses the issue of "COMP Accessibility" in free relative constructions, ... more This paper addresses the issue of "COMP Accessibility" in free relative constructions, initially raised by Groos and van Riemsdijk in a 1979 paper, and proposes a refinement of it in terms of the notion "Pseudo Head" (PH). The proposed theoretical refinement is motivated through a detailed discussion of a hitherto unanalyzed type of free relative construction found in Rumanian (and referred to in the text as the CEL-ΌΈΕ. construction), with respect to which earlier COMP Accessibility proposals make incorrect predictions. The usefulness of the notion PH is further demonstrated with respect to constructions other than free relatives; in particular, it is argued that this notion plays a role in deciding the distribution of interrogative complement clauses within the framework of the strict version of X-bar theory proposed in Chomsky (1986, forthcoming).
History happens only once. This seems to set up an impenetrable barrier for social sciences, like... more History happens only once. This seems to set up an impenetrable barrier for social sciences, like historical linguistics, that concern themselves with change over time. We have the historical record to go on with no convincing way to generate alternative histories that could be used for hypothesis testing. Nevertheless, it is of some interest to ask whether what we see in the historical record is due to particular forces or whether the time series we see could be the result of random drift. In this paper, I will spell out some simple principles of random drift that can be used to construct null hypotheses against which we can study particular cases of language change. The study of random drift allows us to sharpen our analyses of language change and develop more constrained theories of language variation and change.
Hungarian is widely considered as a paradigm case of a language exhibiting a designated "Foc... more Hungarian is widely considered as a paradigm case of a language exhibiting a designated "Focus position " in its clause structure. The existence of the structural position and of overt syntactic movement targeting it has been firmly established in the extensive literature on the topic, and is by now not subject to controversy. Similarly
Introduction Being phrases with internal syntactic structure but manifesting idiosyncrasy of mean... more Introduction Being phrases with internal syntactic structure but manifesting idiosyncrasy of meaning, idioms involve an inherent duality. As such, they provide a unique window into how modules of the language faculty and other aspects of cognition mature and interact. Previous studies have addressed the acquisition of idioms mainly with regard to effects of context and degree of semantic opacity and the developmental differences between idiom comprehension and production based on English, French and Italian ([1], [4, 5], [6], [9]). Little is known about (i) the potential implications of the idioms' internal structure for their acquisition; (ii) idiom acquisition in other languages. We present an experimental study of the acquisition of Hebrew verb phrase idioms, investigating (i) stages of acquisition; (ii) effects of internal structure; (iii) potential consequences regarding idiom storage in the lexicon. Two variables are manipulated: (1) full/gapped idiom: whether an idiom forms a full constituent (e.g. shoot the breeze) or includes a gap (e.g. cut x some slack) (2) decomposable/nondecomposable idiom: whether the idiom is isomorphic with its meaning, i.e. each of its components corresponds to an element of its meaning ([8]) (e.g. [kill] [two birds] [with one stone] '[achieve] [two goals] [with one action]'), or not (e.g. shoot the breeze 'chat'). Experiment 1: comprehension Participants: 90 Hebrew speaking children; 30 first graders (age 6-7), 30 second graders (7-8) and 30 third graders (8-9). Stimuli: 20 Hebrew phrasal idioms having also a licit literal meaning, composed of frequent lexical items, including transitive V NP PP: 5 full and decomposable, 5 full and non-decomposable, 5 gapped and decomposable, and 5 gapped and non-decomposable. Phrase length, structure and frequency were controlled. Design: Illustrated short stories providing the appropriate setting without revealing the meaning of idioms were individually read to participants. Each story ended with an idiom, and was followed by a multiple-choice question. Stories and tasks were pseudorandomly ordered. Subjects were asked to choose the correct interpretation among (i) the correct idiomatic meaning (ii) a literal, contextually inappropriate meaning (iii) a contextually appropriate invented idiomatic meaning. Results: See figure 1. Experiment 2: production (a completion task) Participants: 90 Hebrew speaking children who did not take part in experiment 1, divided into 3 groups as above. Stimuli: the same as in experiment 1. Design: 20 pseudo-randomly ordered illustrated short stories and tasks were individually read to each participant. The target idiom appeared at the end incomplete, allowing recognition of the idiom but not a correct guess. Subjects had to complete idioms. Results: See figure 2. Discussion The ability to comprehend and produce idioms develops gradually in Hebrew from age 6-7, with a gradual decrease in literal responses (figure 3).This is consistent with the findings of previous studies of idiom comprehension targeting children who acquire Italian, French and English ([1],[4],[6],[7]). Internal structure had an effect on acquisition in production (completion) only, and while (non)decomposability played a role across age groups, gaps only affected 2nd graders. We suggest the effect of gaps is related to idiom length, and explain why the effect is limited. (Non)decomposability, in contrast, played a role in production: Across all age groups, nondecomposable idioms scored significantly better in the completion task. We first explain why proposals assigning different storage to the two idiom types [8] cannot explain the data. We then argue that during acquisition children, unlike adults, store nondecomposable verb phrase idioms (in contrast to decomposable ones) as independent 'big-unit' entries, which facilitates idiom completion. This predicts that adults will not exhibit the non-decomposability effect found in children's production. An experiment conducted on 45 adult native Hebrew speakers to test this prediction shows that the adult participants demonstrated the opposite pattern and performed better with decomposable
The paper presents results of our investigation of the distribution of idioms across diatheses (v... more The paper presents results of our investigation of the distribution of idioms across diatheses (voice alternations) in English and Hebrew. We propose an account and discuss its consequences for idiom storage and its implications for alternative architectures of grammar. We provide evidence that idioms split into two distinct subtypes, which we label "phrasal" versus "clausal" idioms. Based on idiom surveys, we observe that phrasal idioms can be specific to the transitive, the unaccusative or the adjectival passive diathesis, but cannot be specific to the verbal passive. Clausal idioms, in contrast, do not discriminate between diatheses: they tend to be specific to a single diathesis. These findings, we argue, cannot be accommodated by a Construction Grammar approach, such as Goldberg (2006), which assumes knowledge of language consists merely of an inventory of stored 'constructions', and does not distinguish between a storage module versus a computationa...
Manifestations of oblique case in Serbian/Croatian (specifically, the dative and the instrumental... more Manifestations of oblique case in Serbian/Croatian (specifically, the dative and the instrumental) present a puzzling pattern of distribution. The paper proposes an analysis for the source of the relevant case phenomena that can resolve the puzzle, rendering it a direct consequence of an independently attested property, namely, the Case Realization Condition of oblique case in the language. It also provides striking evidence regarding the nature and assigner of oblique case in general. The phenomena and their account will lead to the conclusion that contrary to appearances, and in contrast to the standard assumption, lexically governed cases can only be assigned by (overt or null) instances of the category P.
Cross-language studies of linguistic phenomena often reveal facts which are unknown to the lingui... more Cross-language studies of linguistic phenomena often reveal facts which are unknown to the linguist who is concerned with the study of one particular language or language family. Certain rules or processes repeat themselves in unrelated languages and this is expected, since all mankind is subject to the same neurological, physiological and psychological constraints upon his language. In phonology, for example, velar palatalization before high front vowels, word-final devoicing of obstruents and the assimilation of nasal stops in point of articulation to following consonants are commonly found and, if these processes vary from language to language, it is probably a matter of degree only. One such process will be examined here in some detail: voicing assimilation (VA). The starting point is the seemingly recalcitrant behavior of the voiced labio-dental fricative [v] in Russian. Evidence brought from a Finno-Ugric language, Hungarian, and from a Semitic language, Hebrew, shows that the intransigence of [v] in the Slavic language is also found in these two genetically unrelated languages. Implications will be discussed and an explanation, based on a sonority hierarchy, will be offered.
Based on data from different types of parasitic gap (PG) constructions drawn primarily from Hunga... more Based on data from different types of parasitic gap (PG) constructions drawn primarily from Hungarian, the present study investigates and reassesses the longstanding issue of the ränge and possible source(s) of the so-called 'anti-ccommand' effects in PG constructions. Special attention will be paid to the facts of PG licensing by subject variables, and to the way particular theories of PG licensing asymmetries bear on hypotheses of nonconfigurationality of clausestructure, äs in the allegedly nonconfigurational Hungarian. We will address the question of whether the füll ränge of subject/object asymmetries observable in PG licensing are to be derived (a) based on notions of hierarchical structure — such äs c-command — (see e.g. Chomsky 1982; Kayne 1983; Browning 1987), (b) based on some nonstructural difference between subjects vs. objects, such äs, e.g., the particular Case they bear (äs argued in Kiss 1985), or (c) jointly by (a) and (b). First we will argue that the theory of parasitic chain licensing proposed in Kiss (1985) which crucially excludes the relevance of anti-c-command, and replaces it by a 'Case-compatibility' requirement fails to provide an adequate alternative (nonconfigurational) account for the pattern of subject/object asymmetries involved both on empirical and on conceptual grounds. It will be shown that the Case-compatibility requirement is only a necessary, but not a sufficient condition for deriving the subject/object asymmetries observable with respect to the licensing of parasitic chains, and furthermore that the neutralization of Condition C within PG constructions attempted in Kiss (1985) leads to a paradox and hence her framework would end up retaining anti-c-command effects in any case. Given these points, the cases to be discussed not only indicate the relevance of hierarchical structure in the licensing of PGs (contrary to purely Case-based accounts, such äs in Kiss 1985), they also constitute an
This article appeared in a journal published by Elsevier. The attached copy is furnished to the a... more This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier's archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright
This paper reassesses the genetic classification customarily associated with what we refer to as ... more This paper reassesses the genetic classification customarily associated with what we refer to as "unspoken languages," that is, language forms that are, or were at some earlier stage, fulfilling exclusively nonspoken functions (e.g. Medieval and Modern Hebrew, Medieval Latin). We argue that if claims of genetic affiliation are to have any empirical linguistic content, and not simply sociopolitical significance, then genetic descendance from an earlier linguistic form must require the existence of uninterrupted normal transmission. Since the conditions of transmission attested in the case of "unspoken languages" arguably fail to qualify as such, we conclude that these cannot possibly be genetic descendants of the older spoken languages whose names they usually bear. We present our own hypothesis as to the nature and genesis of "unspoken languages," with particular emphasis on the process of relexification of the user's native language with lexical items of the "ancestor" language. As a revealing test case, we discuss Modern Hebrew, claiming that its apparent affinity to Biblical and Mishnaic Hebrew is indeed only a reflection of relexification, not of descendance. Finally, we point out some general methodological implications for genetic and historical linguistics.
Pre-theoretically, wh-interrogatives are commonly assumed to divide languages into three basic ty... more Pre-theoretically, wh-interrogatives are commonly assumed to divide languages into three basic types: (i) wh-movement (of the English-type), (ii) wh-in-situ, and (iii) multiple wh-fronting languages. Although the proposed accounts vary, advancing different hypotheses regarding the structure of wh-phrases, the formal features and interpretive mechanisms that trigger the relevant movements and provide the appropriate interpretations, yet they share the assumption that if an interrogative wh-construction requires movement of a wh-interrogative element, this movement targets the edge of CP (Spec, CP position) (see in particular Cheng's (1991) clause-typing hypothesis, and subsequent work such as Tsai (1994), Watanabe (1992), Richards (1997), Hagstrom (1998), Cable (2010)). As is also widely-known, though Hungarian wh-interrogatives require overt movement of a wh-phrase, the landing site of these wh-phrases is clearly not the Spec, CP position. The prevailing view has been that interrogative wh-phrases in Hungarian land in the position that is known as the designated structural Focus position, namely a position leftadjacent to V in Hungarian clause structure (hierarchically clearly lower than material occurring to its left, namely C and a variety of topic and quantifier positions). Based on this observation it is commonly concluded that what looks like wh-movement in fact is an instance of "Focusmovement"; accordingly, the wh-interrogatives (at least one, if there are multiple) are claimed to bear a [Focus] feature, the same feature that was assumed to drive the movement of non-wh phrases to the designated Focus position in the language (Brody 1990, Horvath 1981, 1986, Lipták 2001). Moreover, this is far from being an isolated property of Hungarian. As is well-known, a wide range of languages exhibit syntactic and/or morphological parallelism between their whinterrogative and (non-wh) Focus constructions. An early proposal to formally capture this crosslinguistic observation Horvath's (1986, sect. 2.3), hypothesis according to which the interpretation of a wh-operator can be a non-echo interrogative only if it comes to bear the syntactic feature [Focus]. In more recent work (Horvath 2000, 2007), I argued that the widely-assumed "Focus movement" to a designated structural position in Hungarian is in fact not movement of Focus at all, and subsequently advanced the proposal (Horvath 2010) that contrary to common assumptions, no formal feature [Focus], or any other non-truth conditional, "discourse", feature, is encoded and active in the computational system. The Hungarian "Focus-movement" construction was instead argued to involve (overt) syntactic movement driven by a quantificational feature. The proposal postulated a (phonologically null) Exhaustive Identification (EI) operator, performing "exclusion by identification" (in the sense of Kenesei 1986) on a set of pragmatically given alternatives for which the predicate can potentially hold. Specifically I claimed that the EI-operator gets merged as Spec of DP, drawing a parallel with the case of the Q-operator of Japanese interrogative whphrases (under Tsai's (1994) analysis). The EI-operator was assumed to associate with Focus-a constituent bearing main stress within its c-command domain-on a par with other focus-sensitive operators (such as ONLY or EVEN). A corresponding functional head EI 0 in the clausal projection served as probe for the EI operator feature, establishing an Agree relation with the EI-operator phrase, and triggering overt phrasal movement to Spec, EIP. Taking the above results as a point of departure, the present paper will examine and reevaluate the status, driving mechanism and landing site(s) of interrogative wh-movements, with the aim to (a) further test and elaborate the EI-operator movement proposal in relation to this challenging new domain, and (b) assess potential implications with respect to the minimalist framework's rationale for and alternative implementations of syntactic displacements. Specifically, parallelisms and discrepancies in distribution and interpretation displayed by interrogative wh and (non-wh) EI-operator movements will be examined within Hungarian and cross-linguistically, with regard to the position of focus (pitch-accent-bearing constituent) and wh-elements within their respective phrases ("pied-piping" options), with regard to the range and limitations of multiple and "mixed" occurrences of interrogative-wh and Focus phrases in a clause, as well as manifestations of locality, "scope-marking", and the potential role of PF constraints.
Die Deutsche Bibliothek-CIP-Einheitsaufnahme Relexiflcation in Creole and Non-Creole languages: w... more Die Deutsche Bibliothek-CIP-Einheitsaufnahme Relexiflcation in Creole and Non-Creole languages: with special attention to Haitian Creole, modern Hebrew, Romani, and Rumanian/ ed. by Julia Horvath and Paul Wexler.-Wiesbaden: Harrassowitz, 1997 ( ...
Natural Language and Linguistic Theory, May 1, 1987
... Note in particular the verb a termina 'to finish', which occurs in (5a), (12a) and (1... more ... Note in particular the verb a termina 'to finish', which occurs in (5a), (12a) and (13a), and which was shown in (19) not to ... Although the grammatical realization of this semantic property in languages often does involve non-finite forms (eg, in English ORIOD constructions), it can ...
This paper addresses the issue of "COMP Accessibility" in free relative constructions, ... more This paper addresses the issue of "COMP Accessibility" in free relative constructions, initially raised by Groos and van Riemsdijk in a 1979 paper, and proposes a refinement of it in terms of the notion "Pseudo Head" (PH). The proposed theoretical refinement is motivated through a detailed discussion of a hitherto unanalyzed type of free relative construction found in Rumanian (and referred to in the text as the CEL-ΌΈΕ. construction), with respect to which earlier COMP Accessibility proposals make incorrect predictions. The usefulness of the notion PH is further demonstrated with respect to constructions other than free relatives; in particular, it is argued that this notion plays a role in deciding the distribution of interrogative complement clauses within the framework of the strict version of X-bar theory proposed in Chomsky (1986, forthcoming).
History happens only once. This seems to set up an impenetrable barrier for social sciences, like... more History happens only once. This seems to set up an impenetrable barrier for social sciences, like historical linguistics, that concern themselves with change over time. We have the historical record to go on with no convincing way to generate alternative histories that could be used for hypothesis testing. Nevertheless, it is of some interest to ask whether what we see in the historical record is due to particular forces or whether the time series we see could be the result of random drift. In this paper, I will spell out some simple principles of random drift that can be used to construct null hypotheses against which we can study particular cases of language change. The study of random drift allows us to sharpen our analyses of language change and develop more constrained theories of language variation and change.
Hungarian is widely considered as a paradigm case of a language exhibiting a designated "Foc... more Hungarian is widely considered as a paradigm case of a language exhibiting a designated "Focus position " in its clause structure. The existence of the structural position and of overt syntactic movement targeting it has been firmly established in the extensive literature on the topic, and is by now not subject to controversy. Similarly
Introduction Being phrases with internal syntactic structure but manifesting idiosyncrasy of mean... more Introduction Being phrases with internal syntactic structure but manifesting idiosyncrasy of meaning, idioms involve an inherent duality. As such, they provide a unique window into how modules of the language faculty and other aspects of cognition mature and interact. Previous studies have addressed the acquisition of idioms mainly with regard to effects of context and degree of semantic opacity and the developmental differences between idiom comprehension and production based on English, French and Italian ([1], [4, 5], [6], [9]). Little is known about (i) the potential implications of the idioms' internal structure for their acquisition; (ii) idiom acquisition in other languages. We present an experimental study of the acquisition of Hebrew verb phrase idioms, investigating (i) stages of acquisition; (ii) effects of internal structure; (iii) potential consequences regarding idiom storage in the lexicon. Two variables are manipulated: (1) full/gapped idiom: whether an idiom forms a full constituent (e.g. shoot the breeze) or includes a gap (e.g. cut x some slack) (2) decomposable/nondecomposable idiom: whether the idiom is isomorphic with its meaning, i.e. each of its components corresponds to an element of its meaning ([8]) (e.g. [kill] [two birds] [with one stone] '[achieve] [two goals] [with one action]'), or not (e.g. shoot the breeze 'chat'). Experiment 1: comprehension Participants: 90 Hebrew speaking children; 30 first graders (age 6-7), 30 second graders (7-8) and 30 third graders (8-9). Stimuli: 20 Hebrew phrasal idioms having also a licit literal meaning, composed of frequent lexical items, including transitive V NP PP: 5 full and decomposable, 5 full and non-decomposable, 5 gapped and decomposable, and 5 gapped and non-decomposable. Phrase length, structure and frequency were controlled. Design: Illustrated short stories providing the appropriate setting without revealing the meaning of idioms were individually read to participants. Each story ended with an idiom, and was followed by a multiple-choice question. Stories and tasks were pseudorandomly ordered. Subjects were asked to choose the correct interpretation among (i) the correct idiomatic meaning (ii) a literal, contextually inappropriate meaning (iii) a contextually appropriate invented idiomatic meaning. Results: See figure 1. Experiment 2: production (a completion task) Participants: 90 Hebrew speaking children who did not take part in experiment 1, divided into 3 groups as above. Stimuli: the same as in experiment 1. Design: 20 pseudo-randomly ordered illustrated short stories and tasks were individually read to each participant. The target idiom appeared at the end incomplete, allowing recognition of the idiom but not a correct guess. Subjects had to complete idioms. Results: See figure 2. Discussion The ability to comprehend and produce idioms develops gradually in Hebrew from age 6-7, with a gradual decrease in literal responses (figure 3).This is consistent with the findings of previous studies of idiom comprehension targeting children who acquire Italian, French and English ([1],[4],[6],[7]). Internal structure had an effect on acquisition in production (completion) only, and while (non)decomposability played a role across age groups, gaps only affected 2nd graders. We suggest the effect of gaps is related to idiom length, and explain why the effect is limited. (Non)decomposability, in contrast, played a role in production: Across all age groups, nondecomposable idioms scored significantly better in the completion task. We first explain why proposals assigning different storage to the two idiom types [8] cannot explain the data. We then argue that during acquisition children, unlike adults, store nondecomposable verb phrase idioms (in contrast to decomposable ones) as independent 'big-unit' entries, which facilitates idiom completion. This predicts that adults will not exhibit the non-decomposability effect found in children's production. An experiment conducted on 45 adult native Hebrew speakers to test this prediction shows that the adult participants demonstrated the opposite pattern and performed better with decomposable
The paper presents results of our investigation of the distribution of idioms across diatheses (v... more The paper presents results of our investigation of the distribution of idioms across diatheses (voice alternations) in English and Hebrew. We propose an account and discuss its consequences for idiom storage and its implications for alternative architectures of grammar. We provide evidence that idioms split into two distinct subtypes, which we label "phrasal" versus "clausal" idioms. Based on idiom surveys, we observe that phrasal idioms can be specific to the transitive, the unaccusative or the adjectival passive diathesis, but cannot be specific to the verbal passive. Clausal idioms, in contrast, do not discriminate between diatheses: they tend to be specific to a single diathesis. These findings, we argue, cannot be accommodated by a Construction Grammar approach, such as Goldberg (2006), which assumes knowledge of language consists merely of an inventory of stored 'constructions', and does not distinguish between a storage module versus a computationa...
Uploads
Papers by Julia Horvath