The paper reviews current understandings of the Minimalist enterprise and defends one version thereof. Our main departure from the norm involves substituting ‘cognitive generality’ (viz. non-linguistic-parochialism) for ‘computational efficiency’ as a minimalist regulative ideal. We argue that the former is important to realize minimalist ambitions. The latter, though welcome, is less vital.
The aim of this paper is to evaluate the theoretical motivations given for feature inheritance, and the workings and distribution thereof. The standard motivations for feature inheritance in the literature are shown not to be tenable. The rationale for feature inheritance given in the literature is based on the requirement that Value and Transfer happen at the same time. This requirement falls through, however; hence, feature ineritance cannot be derived on that basis. Feature inheritance can instead be enforced as the only way to meet a constraint to the effect that the EPP property of a phase head must be satisfied within the minimal structure created by Merge of the phase head with its complement. Syntax then requires feature inheritance as long as ‘EPP’ is necessary and is defined as a Spec–Head relation. Both of these premises are subjected to close scrutiny in the paper, as is the question of whether the predicates ‘(un)interpretable’ and ‘(un)valued’ both need to be recognised by syntactic theory.
In this chapter we pose the non-trivial question regarding the status of functional features (f-features) in minimalist inquiry. This investigation explores whether or not f-features can be considered an essential part of narrow syntactic operations if they can be relegated to some sort of Late Insertion rule at PF. We advance the argument here that at least some f-features must be present in the narrow syntax to ensure derivational compliance.
This paper examines a surprising correlation between adjunct control and wh-movement in Portuguese: object control into an adjunct clause may be allowed in addition to subject control if the matrix object undergoes wh-movement. Assuming Hornstein’s (2001) account of adjunct control within the Movement Theory of Control and making an amendment to Bošković’s (2007) parameterization of edge features, I argue that the unexpected cases of object control arise in Portuguese when Merge-over-Move is inapplicable due to the presence of edge features on wh-elements.
In this paper I present several arguments that argue against the assumption in current generative syntactic theory that certain formal features are semantically active as well (so-called interpretable formal features). Instead, I propose that the set of formal features and the set of semantic features (to the extent that they are featural in the first place) are fully independent. An acquisitional and diachronic theory further constrains the possible combinations of syntactic and semantic features that can be lexically stored, which results in the apparent overlap in the distribution of particular syntactic and semantic features (which has originally been the cause of taking them on a par).
This paper proposes that morphological selection and subcategorization are derived from conditions on Merge, specifically the claim that Merge is only possible when it leads to feature valuation, which I argue takes place under Reverse Agree. The Merge Condition, together with a Reverse Agree mechanism, allows us to unify different types of selection and provides a strictly local and derivational mechanism for structure building which does not require recourse to special selector features or separate notions of (lexical) selection. I provide an explicit feature system encoding the selectional properties of verbs and a detailed account of clausal complementation structures in English and German. The system offers a new way of deriving verb second configurations, doubly filled Comp effects, the distribution of T–to–C movement, as well as the syntactically restricted behavior of embedded root clauses.
This paper considers how the system identifies multiple occurrences of a syntactic object α as a chain, a set of copies. For Chomsky (1995, 2000, 2001), copies can arise only by movement (internal merge); lexical items introduced by external merge are stipulated to be distinct tokens, coded by indexation, or by introducing some other special concept of numeration, which violates the inclusiveness condition. We argue that this type/token dichotomy is unnecessary and that copies can be distinguished from repetitions in terms of syntactic context alone. This yields interesting situations where two arguments introduced by external merge may also be recognized as a chain, and we propose that obligatory control and parasitic gaps should be analyzed in exactly these terms.
This paper addresses a set of puzzles associated with Spell-Out. Of primary concern is the pronunciation and interpretation of hypothetical intermediate copies of moved constituents. I show that LF wh-movement never exhibits any intermediate effects and argue that intermediate effects are best accommodated by rejecting successive cyclic movement in favor of a “one fell swoop” feature-driven approach. I regard “movement” as feature-driven multiattachment, rather than re-merge of actual copies. This relation is unbounded in the syntax; wh-movement intermediate effects arise through the attempt to form a chain for Spell-Out purposes. Since these can only involve C (there being no intermediate SpecCPs), all such effects are head effects.
In this paper we revisit a distinction that has been core to GB and Minimalist approaches to the property of displacement in natural languages: the opposition between A(rgumental) and A′ (non-argumental) positions, as sources and targets of the general operation Move-α. Minimalism brought an additional requirement for the establishment of chains generated via movement, the Chain Uniformity Principle, requiring that all members of a chain be in uniform positions with respect to a certain property. We will argue that such chains, even if possible in principle, are not desirable for both theoretical and empirical reasons, with Radical Minimalism as our theoretical framework. The rejection of the Chain Uniformity Principle will ultimately lead us to revisit the A/A′ distinction as a real and relevant theoretical concept.
I address the question of how morphology can be approached within the Minimalist program, focusing on the notion of asymmetry which has been shown to contribute to the understanding of language and other complex systems. I distinguish the Internalist from the Externalist approach to morphology. I discuss the properties of the operations deriving morphological expressions, including the structure building operation and the operation governing the relations between features. Lastly, I raise the question whether morphological and syntactic complexity is limited by the same kind of conditions stemming from other sub-systems of the mind/brain than the language faculty.
The necessity for roots to be categorized in syntax is recast as an interface condition, resulting from the SEM-deficient character of free acategorial roots. The question of how much descriptive content roots (may) bear is linked to the idiomatic, non-compositional interpretation of the First Phase. The consequences of such a version of syntactic decomposition of words for the morphological realization of roots are outlined, as well as this account’s compatibility with conceptual atomism.
The paper investigates the operations regulating grammar on the basis of experimental child data, and suggests that there are significant computational differences in the acquisition of single interface phenomena and multiple interface phenomena. This finding has direct implications for the precise mechanisms and nature of a learnability account grounded in the Minimalist Program model.
Intensionality, the apparent failure of a normal referential interpretation of nominals in embedded positions, is a phenomenon that is pervasive in human language. It has been a foundational problem for semantics, defining a significant part of its agenda. Here we address the explanatory question of why it exists. Distinguishing lexical aspects of meaning from those that depend on grammatical patterning, we argue that intensionality is mainly grammatical in nature and origin: intensionality is an architectural consequence of the design of human grammar, although, in language use, lexical and pragmatic factors also play a role in the genesis of intuitions of non-substitutability salva veritate. Over the course of this paper, we offer a sequence of ten empirical arguments for this conclusion. A particular account of recursive structure-building in grammar is also offered, which predicts intensionality effects from constraints that govern how nominals of different grammatical types are embedded as arguments in larger units. Crucially, our account requires no appeal to a traditionally postulated semantic ontology of ‘senses’ or ‘thoughts’ as entities ‘denoted’ by embedded clauses, which, we argue, are explanatorily inert. It also covers intensionality characteristics in apparently non-sentential complements of verbs, which we further argue, against the claims of the recent ‘Sententialist Hypothesis’, not to be sentential complements in disguise.
This paper focuses on two important discrepancies between the T-model of the grammar and performance systems responsible for production and comprehension. It argues that independently from the assumed perspective on the competence-performance distinction, one of them is not problematic and the other is. There is no real contradiction in directionality conflicts, i.e. in the fact that the grammar works strictly bottom-up, while performance systems involve many top-down processes. However, the fact that the computational system takes only lexical items and their features as its input presents a real problem, which manifests itself in the domains of scope and Information Structure. This problem can be solved in the grammar architecture where the C-I interface can be used during the derivation.
“3rd factor” considerations are argued here to be a consequence of “dynamical frustration”. This process is seen as the irreconcilable tension between opposing tendencies that gives rise to a form of dynamical stability. Such tendencies are argued to be orthogonal computations: the left-to-right PF and a bottom-up computation involving conceptual relations, which organize into a model specifying Conditions Liberating a Simple Hiatus — or CLASH in acronym format. It is suggested that the CLASH model has a natural account of cyclic conditions within derivations, also predicting the existence of Fibonacci patterns within the linguistic system.
The problem pursued in this paper concerns the role of Arbitrariness in linguistic expressions. Saussurean Arbitrariness is due to the conventional sound-meaning-relation of linguistic expressions. It is systematically projected (but not reduced) from Lexical Items to linguistic expressions of unrestricted complexity by the operation of Merge. This raises the question, whether this arbitrariness is an incidental byproduct of the systems complexity, which could be avoided under conditions of optimal design. With this perspective, language is compared to mental systems of comparable complexity, but without arbitrariness. Obviously, neither the visual system nor the system of music (which consists like language in auditory signals with combinatorial structure) involves arbitrariness in any sense akin to language. The by no means trivial conclusion is, that due to the conventional nature of symbolic signs, linguistic expressions can correspond to structures of any possible domain, differing thereby especially from iconic signs. The upshot of this conclusion: arbitrariness provides the space, by means of which language allows to talk about anything that can be subject to mental awareness. Arbitrariness does not fall short of optimal design, but rather allows language to be the organ of thought and its expression, which it is.
The paper reviews current understandings of the Minimalist enterprise and defends one version thereof. Our main departure from the norm involves substituting ‘cognitive generality’ (viz. non-linguistic-parochialism) for ‘computational efficiency’ as a minimalist regulative ideal. We argue that the former is important to realize minimalist ambitions. The latter, though welcome, is less vital.
The aim of this paper is to evaluate the theoretical motivations given for feature inheritance, and the workings and distribution thereof. The standard motivations for feature inheritance in the literature are shown not to be tenable. The rationale for feature inheritance given in the literature is based on the requirement that Value and Transfer happen at the same time. This requirement falls through, however; hence, feature ineritance cannot be derived on that basis. Feature inheritance can instead be enforced as the only way to meet a constraint to the effect that the EPP property of a phase head must be satisfied within the minimal structure created by Merge of the phase head with its complement. Syntax then requires feature inheritance as long as ‘EPP’ is necessary and is defined as a Spec–Head relation. Both of these premises are subjected to close scrutiny in the paper, as is the question of whether the predicates ‘(un)interpretable’ and ‘(un)valued’ both need to be recognised by syntactic theory.
In this chapter we pose the non-trivial question regarding the status of functional features (f-features) in minimalist inquiry. This investigation explores whether or not f-features can be considered an essential part of narrow syntactic operations if they can be relegated to some sort of Late Insertion rule at PF. We advance the argument here that at least some f-features must be present in the narrow syntax to ensure derivational compliance.
This paper examines a surprising correlation between adjunct control and wh-movement in Portuguese: object control into an adjunct clause may be allowed in addition to subject control if the matrix object undergoes wh-movement. Assuming Hornstein’s (2001) account of adjunct control within the Movement Theory of Control and making an amendment to Bošković’s (2007) parameterization of edge features, I argue that the unexpected cases of object control arise in Portuguese when Merge-over-Move is inapplicable due to the presence of edge features on wh-elements.
In this paper I present several arguments that argue against the assumption in current generative syntactic theory that certain formal features are semantically active as well (so-called interpretable formal features). Instead, I propose that the set of formal features and the set of semantic features (to the extent that they are featural in the first place) are fully independent. An acquisitional and diachronic theory further constrains the possible combinations of syntactic and semantic features that can be lexically stored, which results in the apparent overlap in the distribution of particular syntactic and semantic features (which has originally been the cause of taking them on a par).
This paper proposes that morphological selection and subcategorization are derived from conditions on Merge, specifically the claim that Merge is only possible when it leads to feature valuation, which I argue takes place under Reverse Agree. The Merge Condition, together with a Reverse Agree mechanism, allows us to unify different types of selection and provides a strictly local and derivational mechanism for structure building which does not require recourse to special selector features or separate notions of (lexical) selection. I provide an explicit feature system encoding the selectional properties of verbs and a detailed account of clausal complementation structures in English and German. The system offers a new way of deriving verb second configurations, doubly filled Comp effects, the distribution of T–to–C movement, as well as the syntactically restricted behavior of embedded root clauses.
This paper considers how the system identifies multiple occurrences of a syntactic object α as a chain, a set of copies. For Chomsky (1995, 2000, 2001), copies can arise only by movement (internal merge); lexical items introduced by external merge are stipulated to be distinct tokens, coded by indexation, or by introducing some other special concept of numeration, which violates the inclusiveness condition. We argue that this type/token dichotomy is unnecessary and that copies can be distinguished from repetitions in terms of syntactic context alone. This yields interesting situations where two arguments introduced by external merge may also be recognized as a chain, and we propose that obligatory control and parasitic gaps should be analyzed in exactly these terms.
This paper addresses a set of puzzles associated with Spell-Out. Of primary concern is the pronunciation and interpretation of hypothetical intermediate copies of moved constituents. I show that LF wh-movement never exhibits any intermediate effects and argue that intermediate effects are best accommodated by rejecting successive cyclic movement in favor of a “one fell swoop” feature-driven approach. I regard “movement” as feature-driven multiattachment, rather than re-merge of actual copies. This relation is unbounded in the syntax; wh-movement intermediate effects arise through the attempt to form a chain for Spell-Out purposes. Since these can only involve C (there being no intermediate SpecCPs), all such effects are head effects.
In this paper we revisit a distinction that has been core to GB and Minimalist approaches to the property of displacement in natural languages: the opposition between A(rgumental) and A′ (non-argumental) positions, as sources and targets of the general operation Move-α. Minimalism brought an additional requirement for the establishment of chains generated via movement, the Chain Uniformity Principle, requiring that all members of a chain be in uniform positions with respect to a certain property. We will argue that such chains, even if possible in principle, are not desirable for both theoretical and empirical reasons, with Radical Minimalism as our theoretical framework. The rejection of the Chain Uniformity Principle will ultimately lead us to revisit the A/A′ distinction as a real and relevant theoretical concept.
I address the question of how morphology can be approached within the Minimalist program, focusing on the notion of asymmetry which has been shown to contribute to the understanding of language and other complex systems. I distinguish the Internalist from the Externalist approach to morphology. I discuss the properties of the operations deriving morphological expressions, including the structure building operation and the operation governing the relations between features. Lastly, I raise the question whether morphological and syntactic complexity is limited by the same kind of conditions stemming from other sub-systems of the mind/brain than the language faculty.
The necessity for roots to be categorized in syntax is recast as an interface condition, resulting from the SEM-deficient character of free acategorial roots. The question of how much descriptive content roots (may) bear is linked to the idiomatic, non-compositional interpretation of the First Phase. The consequences of such a version of syntactic decomposition of words for the morphological realization of roots are outlined, as well as this account’s compatibility with conceptual atomism.
The paper investigates the operations regulating grammar on the basis of experimental child data, and suggests that there are significant computational differences in the acquisition of single interface phenomena and multiple interface phenomena. This finding has direct implications for the precise mechanisms and nature of a learnability account grounded in the Minimalist Program model.
Intensionality, the apparent failure of a normal referential interpretation of nominals in embedded positions, is a phenomenon that is pervasive in human language. It has been a foundational problem for semantics, defining a significant part of its agenda. Here we address the explanatory question of why it exists. Distinguishing lexical aspects of meaning from those that depend on grammatical patterning, we argue that intensionality is mainly grammatical in nature and origin: intensionality is an architectural consequence of the design of human grammar, although, in language use, lexical and pragmatic factors also play a role in the genesis of intuitions of non-substitutability salva veritate. Over the course of this paper, we offer a sequence of ten empirical arguments for this conclusion. A particular account of recursive structure-building in grammar is also offered, which predicts intensionality effects from constraints that govern how nominals of different grammatical types are embedded as arguments in larger units. Crucially, our account requires no appeal to a traditionally postulated semantic ontology of ‘senses’ or ‘thoughts’ as entities ‘denoted’ by embedded clauses, which, we argue, are explanatorily inert. It also covers intensionality characteristics in apparently non-sentential complements of verbs, which we further argue, against the claims of the recent ‘Sententialist Hypothesis’, not to be sentential complements in disguise.
This paper focuses on two important discrepancies between the T-model of the grammar and performance systems responsible for production and comprehension. It argues that independently from the assumed perspective on the competence-performance distinction, one of them is not problematic and the other is. There is no real contradiction in directionality conflicts, i.e. in the fact that the grammar works strictly bottom-up, while performance systems involve many top-down processes. However, the fact that the computational system takes only lexical items and their features as its input presents a real problem, which manifests itself in the domains of scope and Information Structure. This problem can be solved in the grammar architecture where the C-I interface can be used during the derivation.
“3rd factor” considerations are argued here to be a consequence of “dynamical frustration”. This process is seen as the irreconcilable tension between opposing tendencies that gives rise to a form of dynamical stability. Such tendencies are argued to be orthogonal computations: the left-to-right PF and a bottom-up computation involving conceptual relations, which organize into a model specifying Conditions Liberating a Simple Hiatus — or CLASH in acronym format. It is suggested that the CLASH model has a natural account of cyclic conditions within derivations, also predicting the existence of Fibonacci patterns within the linguistic system.
The problem pursued in this paper concerns the role of Arbitrariness in linguistic expressions. Saussurean Arbitrariness is due to the conventional sound-meaning-relation of linguistic expressions. It is systematically projected (but not reduced) from Lexical Items to linguistic expressions of unrestricted complexity by the operation of Merge. This raises the question, whether this arbitrariness is an incidental byproduct of the systems complexity, which could be avoided under conditions of optimal design. With this perspective, language is compared to mental systems of comparable complexity, but without arbitrariness. Obviously, neither the visual system nor the system of music (which consists like language in auditory signals with combinatorial structure) involves arbitrariness in any sense akin to language. The by no means trivial conclusion is, that due to the conventional nature of symbolic signs, linguistic expressions can correspond to structures of any possible domain, differing thereby especially from iconic signs. The upshot of this conclusion: arbitrariness provides the space, by means of which language allows to talk about anything that can be subject to mental awareness. Arbitrariness does not fall short of optimal design, but rather allows language to be the organ of thought and its expression, which it is.