Year: 2005
Number 75: 1-36
Jóhanna Barðdal & Thórhallur Eythórsson
Case and Control Constructions in German, Faroese and Icelandic: Or How to Evaluate Marginally-Acceptable Data?
An examination of German language use reveals that subject-like obliques of impersonal predicates and dative passives can be left unexpressed in control infinitives, exactly as in Icelandic and Faroese, contra claims in the literature that there are no oblique subjects in German. Native-speaker judgments on these attested examples are subject to some controversy, bringing to the fore the issue of how to evaluate marginally-accepted data. We argue that this must be addressed in relation to the fact that there are also examples of control infinitives in Faroese and Icelandic which are judged ill-formed or ungrammatical by native speakers, again contra the established view in the literature that Icelandicand Faroese have oblique subjects. The distribution of the acceptability judgments correlates with the fact that the control infinitives under investigation are low-frequency constructions in all the Modern Germanic languages, including Modern Icelandic. The scarcity of such control infinitives in the modern languages prognosticates that only very few such instantiations should be found in earlier stages of Germanic, as is indeed borne out.
Number 75: 1-36
Jóhanna Barðdal & Thórhallur Eythórsson
Case and Control Constructions in German, Faroese and Icelandic: Or How to Evaluate Marginally-Acceptable Data?
An examination of German language use reveals that subject-like obliques of impersonal predicates and dative passives can be left unexpressed in control infinitives, exactly as in Icelandic and Faroese, contra claims in the literature that there are no oblique subjects in German. Native-speaker judgments on these attested examples are subject to some controversy, bringing to the fore the issue of how to evaluate marginally-accepted data. We argue that this must be addressed in relation to the fact that there are also examples of control infinitives in Faroese and Icelandic which are judged ill-formed or ungrammatical by native speakers, again contra the established view in the literature that Icelandicand Faroese have oblique subjects. The distribution of the acceptability judgments correlates with the fact that the control infinitives under investigation are low-frequency constructions in all the Modern Germanic languages, including Modern Icelandic. The scarcity of such control infinitives in the modern languages prognosticates that only very few such instantiations should be found in earlier stages of Germanic, as is indeed borne out.
Year: 2005
Number 75: 37-54
Fredrik Heinat
Reflexives in a phase based syntax
The syntax of anaphors and pronouns has been in the centre of syntactic research for a long time. The present paper argues that the binding principles A and B from GB-theory can be accommodated in recent developments of syntactic theory, namely the minimalist program. The main claim is that since there is no distinction between phrases and heads in the building of the syntactic structure (Chomsky 1995, Collins 2002, Starke 2001) it is reasonable to assume that all unvalued features, be they on heads or phrases, probe their domain. The result is that there can, and sometimes must, be a syntactic agree-relation between DPs. In other cases such a relation cannot be formed. The consequence is that in the latter case the spell-out of a root pronoun will be a reflexive pronoun, whereas in the latter case it will be a pronoun. The conclusion is that this analysis achieves almost the same result as previous analyses, but without the use of filters on representations. Instead, the distribution of pronouns and reflexives is a result of the way the syntactic derivation proceeds.
Number 75: 37-54
Fredrik Heinat
Reflexives in a phase based syntax
The syntax of anaphors and pronouns has been in the centre of syntactic research for a long time. The present paper argues that the binding principles A and B from GB-theory can be accommodated in recent developments of syntactic theory, namely the minimalist program. The main claim is that since there is no distinction between phrases and heads in the building of the syntactic structure (Chomsky 1995, Collins 2002, Starke 2001) it is reasonable to assume that all unvalued features, be they on heads or phrases, probe their domain. The result is that there can, and sometimes must, be a syntactic agree-relation between DPs. In other cases such a relation cannot be formed. The consequence is that in the latter case the spell-out of a root pronoun will be a reflexive pronoun, whereas in the latter case it will be a pronoun. The conclusion is that this analysis achieves almost the same result as previous analyses, but without the use of filters on representations. Instead, the distribution of pronouns and reflexives is a result of the way the syntactic derivation proceeds.
Year: 2005
Number 75: 55-96
Gunlög Josefsson
How could Merge be free and word formation restricted: The case of compounding in Romance and Germanic
Focussing on compounds with an element that corresponds to a noun as non-head, e.g. flick+skola (girl + school) 'school for girls', this article addresses the question of how word formation can be syntactic – hence principally free – and yet at the same time constrained by language-specific rules. The answer will be that the basic structure of the lexicon differs between languages, according to what is described as lexical macroparameters, i.e. whether or not the lexicon contains elements devoid of checkable features that are available for productive compounding. The lexicon of non-English Germanic languages contains such elements, viz. pure roots (like dag- in dag+bok (day+book) 'dairy') or roots + a linking element (like dag + -s, in dag-s+ljus (day-s+light) 'day light'). The modern Romance languages do not, hence compounding of the described type is unavailable. Free Merge may thus very well be reconciled with the idea of constraints on word formation. Three language groups are discussed: non-English Germanic languages, Romance, and English. A diachronic overview shows how changes in the phonology may cause a transition from one system (represented by Latin and Old English), where compounding involving roots was generally OK, to another system (represented by Modern Romance and Modern English), where (root) compounding is not available. Productive compounding in the latter languages is presumably phrasal in nature.
Number 75: 55-96
Gunlög Josefsson
How could Merge be free and word formation restricted: The case of compounding in Romance and Germanic
Focussing on compounds with an element that corresponds to a noun as non-head, e.g. flick+skola (girl + school) 'school for girls', this article addresses the question of how word formation can be syntactic – hence principally free – and yet at the same time constrained by language-specific rules. The answer will be that the basic structure of the lexicon differs between languages, according to what is described as lexical macroparameters, i.e. whether or not the lexicon contains elements devoid of checkable features that are available for productive compounding. The lexicon of non-English Germanic languages contains such elements, viz. pure roots (like dag- in dag+bok (day+book) 'dairy') or roots + a linking element (like dag + -s, in dag-s+ljus (day-s+light) 'day light'). The modern Romance languages do not, hence compounding of the described type is unavailable. Free Merge may thus very well be reconciled with the idea of constraints on word formation. Three language groups are discussed: non-English Germanic languages, Romance, and English. A diachronic overview shows how changes in the phonology may cause a transition from one system (represented by Latin and Old English), where compounding involving roots was generally OK, to another system (represented by Modern Romance and Modern English), where (root) compounding is not available. Productive compounding in the latter languages is presumably phrasal in nature.
Year: 2005
Number 75: 97-128
Christer Platzack
Uninterpretable features and EPP: a minimalist account of language build up and break down
Like most L2 speakers, speakers of early L1, children with SLI, and Broca's aphasics lack the ordinary speaker's ability to produce utterances on line with almost no errors. Often they produce well-formed utterances, but quite frequently they produce errors that are almost non-existent in the speech of normal grown-up speakers. In my paper I will suggest, based on empirical material from Swedish, that this behavior can be understood as a performance problem: these speakers have the same knowledge of the target language as ordinary speakers, but cannot automatically adjust their production to the language specific distribution of EPP, i.e. the demand to express a particular grammatical relation overtly. To obtain this I adopt an idea put forward by Pesetsky & Torrego (2001) that EPP is to be distinguished from the grammatical relation itself: EPP is connected to an uninterpretable feature, forcing Agree and Move to apply to avoid a violation of the interface condition. The distribution of EPP is language specific, hence my account correctly predicts cross-language variation for the four groups of speakers discussed here.
Number 75: 97-128
Christer Platzack
Uninterpretable features and EPP: a minimalist account of language build up and break down
Like most L2 speakers, speakers of early L1, children with SLI, and Broca's aphasics lack the ordinary speaker's ability to produce utterances on line with almost no errors. Often they produce well-formed utterances, but quite frequently they produce errors that are almost non-existent in the speech of normal grown-up speakers. In my paper I will suggest, based on empirical material from Swedish, that this behavior can be understood as a performance problem: these speakers have the same knowledge of the target language as ordinary speakers, but cannot automatically adjust their production to the language specific distribution of EPP, i.e. the demand to express a particular grammatical relation overtly. To obtain this I adopt an idea put forward by Pesetsky & Torrego (2001) that EPP is to be distinguished from the grammatical relation itself: EPP is connected to an uninterpretable feature, forcing Agree and Move to apply to avoid a violation of the interface condition. The distribution of EPP is language specific, hence my account correctly predicts cross-language variation for the four groups of speakers discussed here.