Jean-Yves Pollock Cyclicity, and Move NP in ... The rule of Stylistic Inversion (Styl-Inv) in Fre... more Jean-Yves Pollock Cyclicity, and Move NP in ... The rule of Stylistic Inversion (Styl-Inv) in French moves a subject NP to the right. It plays a role in the derivation of sentences such as those in (1): ... (1) a. Quand partira ton ami? when will leave your friend 'When will your friend ...
Phenomena traditionally thought of as morphological can be accounted for in terms of syntactic op... more Phenomena traditionally thought of as morphological can be accounted for in terms of syntactic operations and principles, hence bringing forth questions that traditional morphology fails to ask (for instance, concerning the licensing of empty morphemes). The language faculty contains no specific morphological component, nor any post-syntactic morphological operations.
French hypercomplex inversion can be used as a probe into the question whether or not the languag... more French hypercomplex inversion can be used as a probe into the question whether or not the language faculty countenances the existence of true expletive pronouns, i.e. pronouns that make no interpretive contribution of any kind. The conclusion is that there are no expletive pronouns (in any language) in the strict sense of the term.
There are no mirror-image pairs of languages. This restriction on the otherwise vast set of possi... more There are no mirror-image pairs of languages. This restriction on the otherwise vast set of possible languages must be accounted for, and puts boundary conditions on any theory of the human language faculty. There are implications for externalization and in the longer run for the evolution of the language faculty. Antisymmetric linear/temporal order is part of core syntax. Temporal order is partly (though not fully) integrated into core syntax via Merge itself. When two elements X and Y are merged, a relative linear/temporal order is assigned to them. At the same time, that instance of Merge assigns no relative order to the subparts of X and Y. Core syntax can explicitly have X precede Y without having any subpart of X precede any subpart of Y.
Relative clauses can be found that contain a relative pronoun whose antecedent is not the head of... more Relative clauses can be found that contain a relative pronoun whose antecedent is not the head of the relative. The familiar relation between the head of a relative and the relative pronoun can thus be seen as a special case of a more general relation between a relative pronoun (a stranded determiner) and its antecedent (whose movement has stranded that determiner). The piece of relative clause syntax that is the antecedent-relative pronoun relation is less specific to relative clauses that it might have seemed.
In the course of the past fifty or sixty years, our knowledge and understanding of human language... more In the course of the past fifty or sixty years, our knowledge and understanding of human language syntax has become qualitatively better. Part of that qualitative improvement has come from advances in the subfield of syntax called comparative syntax. It is useful to think in terms of the notions of observational, descriptive and explanatory adequacy. 1 Observational adequacy can be said to be achieved when one has gotten the facts right concerning acceptability judgments and judgments of interpretation. ‘The facts ’ are the facts that one is concerned with, and not, of course, ‘all the facts ’ of syntax or comparative syntax, whose order of magnitude lies, as in other sciences, far beyond our reach. This is true even for a single language. 2 Syntacticians take as a primary object of study the set of possible human languages. The entire set is again far beyond our reach. To one degree or another, we have access to those languages currently spoken and to an exceedingly small percentag...
Continuity and Variation in Germanic and Romance, 2021
Many Germanic languages have a finite-clause complementizer that resembles a demonstrative, e.g. ... more Many Germanic languages have a finite-clause complementizer that resembles a demonstrative, e.g. English that, Dutch dat, German dass. No Romance language does. The traditional view of complementizers as simplex projecting heads that take IP or some comparable category as a complement has no way of accounting for this difference between Germanic and Romance. In this chapter, I will attempt to make progress toward an account, in part by reinterpreting finite-clause complementizers as relative pronouns.
Jean-Yves Pollock Cyclicity, and Move NP in ... The rule of Stylistic Inversion (Styl-Inv) in Fre... more Jean-Yves Pollock Cyclicity, and Move NP in ... The rule of Stylistic Inversion (Styl-Inv) in French moves a subject NP to the right. It plays a role in the derivation of sentences such as those in (1): ... (1) a. Quand partira ton ami? when will leave your friend 'When will your friend ...
Phenomena traditionally thought of as morphological can be accounted for in terms of syntactic op... more Phenomena traditionally thought of as morphological can be accounted for in terms of syntactic operations and principles, hence bringing forth questions that traditional morphology fails to ask (for instance, concerning the licensing of empty morphemes). The language faculty contains no specific morphological component, nor any post-syntactic morphological operations.
French hypercomplex inversion can be used as a probe into the question whether or not the languag... more French hypercomplex inversion can be used as a probe into the question whether or not the language faculty countenances the existence of true expletive pronouns, i.e. pronouns that make no interpretive contribution of any kind. The conclusion is that there are no expletive pronouns (in any language) in the strict sense of the term.
There are no mirror-image pairs of languages. This restriction on the otherwise vast set of possi... more There are no mirror-image pairs of languages. This restriction on the otherwise vast set of possible languages must be accounted for, and puts boundary conditions on any theory of the human language faculty. There are implications for externalization and in the longer run for the evolution of the language faculty. Antisymmetric linear/temporal order is part of core syntax. Temporal order is partly (though not fully) integrated into core syntax via Merge itself. When two elements X and Y are merged, a relative linear/temporal order is assigned to them. At the same time, that instance of Merge assigns no relative order to the subparts of X and Y. Core syntax can explicitly have X precede Y without having any subpart of X precede any subpart of Y.
Relative clauses can be found that contain a relative pronoun whose antecedent is not the head of... more Relative clauses can be found that contain a relative pronoun whose antecedent is not the head of the relative. The familiar relation between the head of a relative and the relative pronoun can thus be seen as a special case of a more general relation between a relative pronoun (a stranded determiner) and its antecedent (whose movement has stranded that determiner). The piece of relative clause syntax that is the antecedent-relative pronoun relation is less specific to relative clauses that it might have seemed.
In the course of the past fifty or sixty years, our knowledge and understanding of human language... more In the course of the past fifty or sixty years, our knowledge and understanding of human language syntax has become qualitatively better. Part of that qualitative improvement has come from advances in the subfield of syntax called comparative syntax. It is useful to think in terms of the notions of observational, descriptive and explanatory adequacy. 1 Observational adequacy can be said to be achieved when one has gotten the facts right concerning acceptability judgments and judgments of interpretation. ‘The facts ’ are the facts that one is concerned with, and not, of course, ‘all the facts ’ of syntax or comparative syntax, whose order of magnitude lies, as in other sciences, far beyond our reach. This is true even for a single language. 2 Syntacticians take as a primary object of study the set of possible human languages. The entire set is again far beyond our reach. To one degree or another, we have access to those languages currently spoken and to an exceedingly small percentag...
Continuity and Variation in Germanic and Romance, 2021
Many Germanic languages have a finite-clause complementizer that resembles a demonstrative, e.g. ... more Many Germanic languages have a finite-clause complementizer that resembles a demonstrative, e.g. English that, Dutch dat, German dass. No Romance language does. The traditional view of complementizers as simplex projecting heads that take IP or some comparable category as a complement has no way of accounting for this difference between Germanic and Romance. In this chapter, I will attempt to make progress toward an account, in part by reinterpreting finite-clause complementizers as relative pronouns.
To say that all dative arguments originate DP-internally is appreciably less implausible that it ... more To say that all dative arguments originate DP-internally is appreciably less implausible that it might sound.
There is a point of similarity between English suffixal superlatives and English past tense. Ther... more There is a point of similarity between English suffixal superlatives and English past tense. There is the question whether or not comparative -er should be decomposed in parallel fashion to superlative -est, and how that fits in with FOFC effects. And then there is the importance of all the contrasts between 'most' and 'least'.
Understanding the syntactic behavior of 'half', 'all', 'both', 'each' and 'of' requires postulati... more Understanding the syntactic behavior of 'half', 'all', 'both', 'each' and 'of' requires postulating in a range of cases silent elements, in particular, ALL, THE, A, OF, ONE, TIME(S), OUT, and THING.
The language faculty has met the challenge of externalization by partially
integrating temporal o... more The language faculty has met the challenge of externalization by partially integrating temporal order into its core.
The language faculty has chosen to have 'zero' accompanied by NOT. Why did it choose to do so? Th... more The language faculty has chosen to have 'zero' accompanied by NOT. Why did it choose to do so? The most straightforward answer would seem to be that it had no choice. The only way to express what 'zero' expresses involves negation.
The interpretation of certain modal sentences rests on the presence within them of certain silent... more The interpretation of certain modal sentences rests on the presence within them of certain silent modal elements. The language faculty imposes the presence of these silent elements via a Principle of Decompositionality that prohibits multiple interpretable syntactic features on a single node.
Against the background of certain similarities between 'half' and 'all', a
certain difference bet... more Against the background of certain similarities between 'half' and 'all', a certain difference between them having to do with stranding can be understood in terms of a silent OF needed with 'half' (in the absence of an overt 'of'), but not with 'all'. The account appears to fit well with the stranding analysis of Sportiche (1988).
'Have' + infinitival ECM is specific to English because only English can, in that context, provid... more 'Have' + infinitival ECM is specific to English because only English can, in that context, provide 'have' with the preposition that it needs (there and elsewhere), with that preposition being in this case FOR, the silent counterpart of complementizer 'for', which is itself, in ECM contexts, highly specific to English, as is its silent counterpart.
The comparative study of Hypercomplex Inversion (HCI) in French and in the North Italian dialect ... more The comparative study of Hypercomplex Inversion (HCI) in French and in the North Italian dialect of Carmignano di Brenta, in addition to being of interest in its own right, provides evidence suggesting that there are no expletive pronouns (in any language) in the strict sense of the term.
Work on cartographic hierarchies can in some cases be affected in important ways by work on silen... more Work on cartographic hierarchies can in some cases be affected in important ways by work on silent elements. In what follows, I will try to show in a range of cases what I mean by this.
The two readings of 'other' that Cinque (2015) has individuated need not fall under a specific po... more The two readings of 'other' that Cinque (2015) has individuated need not fall under a specific positional statement of the cartographic type. Rather, the two readings correspond to 'other' merging with either TOKEN or TYPE.
All instances of English 'else' must be accompanied by a noun. This leads to the postulation of s... more All instances of English 'else' must be accompanied by a noun. This leads to the postulation of silent THING (or PLACE or REASON) in a range of cases. The noun requirement on 'else' may be a subcase of a more general requirement that all arguments must be built on a noun, whether pronounced or unpronounced.
In all cases of suppletion one needs to ask why the unavailable expected form is not well-formed,... more In all cases of suppletion one needs to ask why the unavailable expected form is not well-formed, and one needs to look for an answer that is independent of the existence of the unexpected form. In a number of cases, or in perhaps all cases, suppletion can be seen to involve, not direct competition between two forms, but instead two structures that differ in that (at least) one contains a silent element lacking in the other.
Antisymmetric linear order is part of core syntax. There are implications for externalizati... more Antisymmetric linear order is part of core syntax. There are implications for externalization and in the longer run for the evolution of the language faculty.
Spanish 'ir' (‘go’) has unexpected f-initial forms in the preterite and in the imperfect subjunct... more Spanish 'ir' (‘go’) has unexpected f-initial forms in the preterite and in the imperfect subjunctive. Rather than being suppletive forms of 'ir' they are forms of 'ser' (‘be’). The interpretative contribution of ‘go’ is provided by the presence of a silent GO. The syntax of silent verbs turns out to be of importance to what appear to be traditional questions of morphology.
There is an answer to the question why bare 'long' and 'a long time' differ in the way they do, w... more There is an answer to the question why bare 'long' and 'a long time' differ in the way they do, with 'a long time' showing no need for negation at all. The reason is that 'a long time' contains no silent TIME in need of licensing. The apparent negative polarity status of bare 'long' is, indirectly, just a side effect of its containing TIME.
Many Germanic languages have a finite-clause complementizer that resembles a demonstrative, e.g. ... more Many Germanic languages have a finite-clause complementizer that resembles a demonstrative, e.g. English that, Dutch dat, German dass. No Romance language does. The traditional view of complementizers as simplex projecting heads that take IP or some comparable category as complement has no way of accounting for this difference between Germanic and Romance. In this talk, I will attempt to make progress toward an account, in part by reinterpreting finite-clause complementizers as relative pronouns.
The (Italian) pair '*il luo libro' vs. 'il suo libro' (‘the his/her book’) that is typical of Rom... more The (Italian) pair '*il luo libro' vs. 'il suo libro' (‘the his/her book’) that is typical of Romance lends itself to an account of the first in terms of constraints also seen in the syntax of compounding, and to an account of the second that links its 's-', despite initial appearances, to what we think of as Romance reflexive 's-'. We might informally call this pair an instance of suppletion, as long as we recognize that calling it suppletion falls short of an account.
Significant evidence is produced for a proposed linguistic universal, the transformational cycle ... more Significant evidence is produced for a proposed linguistic universal, the transformational cycle (cf. phases), as well as for those aspects of the grammar which crucially interact with it, in this case the transformations (movement operations) developed in different chapters of this work, and, finally, for the linguistic theory that provided the framework in which we could formulate that analysis, that is, for the theory of transformational grammar (generative syntax).
Collection of articles, including "Connectedness" and "Unambiguous Paths", which contains the pro... more Collection of articles, including "Connectedness" and "Unambiguous Paths", which contains the proposal that all branching is binary.
Each paper in this volume is, for reasons of space, presented in excerpted form, and is preceded ... more Each paper in this volume is, for reasons of space, presented in excerpted form, and is preceded by an introduction that provides some background and a small set of relevant references. Each paper is also followed by a set of questions intended to encourage the student to explore new lines of thought.
To say that all dative arguments originate DP-internally is appreciably less
implausible that it ... more To say that all dative arguments originate DP-internally is appreciably less implausible that it might sound.
Antisymmetric linear order is arguably part of core syntax. The implications for externalization ... more Antisymmetric linear order is arguably part of core syntax. The implications for externalization and for the evolution of the language faculty need to be looked into further.
The question what is primitive/undecomposable and what is not is a pervasive one. I argue that m... more The question what is primitive/undecomposable and what is not is a pervasive one. I argue that modal 'need' is decomposable into nominal 'need' plus silent MUST, with which nominal 'need' is in an adjunct-like relation that may be mediated by silent OF.
Core syntax is open to much that is not of any obvious semantic importance, both to semantically ... more Core syntax is open to much that is not of any obvious semantic importance, both to semantically neutral movement operations, and, in line with the early years of generative grammar, to linear/temporal order itself.
Uploads
integrating temporal order into its core.
certain difference between them having to do with stranding can be understood in terms of a silent OF needed with 'half' (in the absence of an overt 'of'), but not with 'all'. The account appears to fit well with the stranding analysis of Sportiche (1988).
direct competition between two forms, but instead two structures that differ in that (at least) one contains a silent element lacking in the other.
implausible that it might sound.