Abstract This article explores the interplay between perception and conceptualization in human language, focusing on how linguistic constructions reveal underlying cognitive processes. It examines the debate between perceptual conceptualism and nonconceptualism, considering whether perceptual content is inherently conceptual or if judgments based on perception require active conceptualization. The study centers on English constructions using the verb see followed by direct objects and either bare or to -infinitives, arguing that the bare infinitive construction corresponds to direct sensory perception, while the to -infinitive signals inference or conceptual judgment based on perception. Through detailed analysis of the linguistic evidence, it demonstrates that even perceptual verb constructions commonly blend sensory and inferential content, highlighting how states and absences may be inferred from direct perception. Ultimately, the article contends that perception is inseparable from cognitive and conceptual processes, underscoring the active interpretive role language plays in perceptual phenomena.

In the Generative model of human language, the semantic component has the role of interpreting the syntax. This amounts to semantics being swallowed up by syntax and explains Chomsky’s reiterated statements that at bottom natural language has only syntax and pragmatics: “natural language has no semantics … it has syntax (symbol manipulation) and pragmatics (modes of use of language)” (Chomsky, Noam. 2013. Notes on denotation and denoting. In Ivano Caponigro & Carlo Conchetto (eds.), From grammar to meaning: the spontaneous logicality of language , 38–45. Cambridge: Cambridge University Press). It will be argued that this conclusion is predetermined by the view of language as a computationally perfect syntactic system. Language is thereby reduced to syntactic distribution and symbol manipulation, linguistic form separated from meaning, with syntax being investigated according to the goals of coherence and consistency and subsequently reconnected extrinsically to the world by means of pragmatics. It will be shown that a basic understanding of the nature of language, which is constituted by the fundamental correlation between linguistic signs and their meanings, reveals that one cannot isolate and study separately the physically observable side of language without tearing asunder the very object of one’s investigation. The causal conditioning of syntax by both semantics and pragmatics will be demonstrated via the examples of English causative verbs and the ‘degree adverb + proper name’ construction.

The purpose of this study is to assess from a corpus -based discourse -pragmatic perspective certain claims made in the literature concerning English wh- concessive conditional constructions (e.g. Whoever/No matter who comes to the party, it will be fun), namely that these utterance -types correlate with interrogative semantics, scalarity and potential modality. By means of an extensive investigation of corpus data these claims are shown to be largely unsupported by attested usage. Based on Dancygier and Sweetser's classification of conditional constructions, it is found that potential modality is paradigmatic only of content -level concessive conditionals, and not of the epistemic, speech -act or metalinguistic varieties. Contrary to claims in the literature, scalarity is demonstrated to not be typical of wh- concessive conditionals. The lack of scalarity in most wh- concessive conditionals is argued to cast into doubt the category label "concessive conditional" applied to these constructions in a substantial part of the literature and to favour an alternative designation such as "irrelevance conditional." The empirical data further reveals that wh- concessive conditionals practically never involve pure ignorance, and this is argued to be problematic on the discoursepragmatic level for the claim that they have interrogative semantics.

This study argues that many of the formalizations used in analyses employing the notion of logical scope fail to conform to natural language in important ways and lead to false predictions. This is due to the fact that they pursue the logic-driven goal of making the structure of logical arguments more transparent and mechanically calculable rather than the language-driven goal of accounting for how the linguistic signs used in an utterance and their configuration contribute to the conveying of the message being fashioned by the speaker. The focus of the study is on categories associated with the verb: tense, aspect, modality and negation. The conclusion suggests that very precise and rigid theories using logical scope relations may force the theorist to straitjacket the data so that they fit the theory, thereby obscuring rather than clarifying the nature of linguistic categories and their interactions. Informal analyses that hew closer to natural language's semantic reality can provide greater understanding of phenomena such as the purported non-negatability of must. Seeing this English modal's meaning as defined in opposition to real existence leads to the realization that it does not interact with negation the same way as the reality of the existence of the property of being necessary does.

Abstract The purpose of this study is to assess from a corpus-based discourse-pragmatic perspective certain claims made in the literature concerning English wh- concessive conditional constructions (e.g. Whoever/No matter who comes to the party, it will be fun), namely that these utterance-types correlate with interrogative semantics, scalarity and potential modality. By means of an extensive investigation of corpus data these claims are shown to be largely unsupported by attested usage. Based on Dancygier and Sweetser’s classification of conditional constructions, it is found that potential modality is paradigmatic only of content-level concessive conditionals, and not of the epistemic, speech-act or metalinguistic varieties. Contrary to claims in the literature, scalarity is demonstrated to not be typical of wh- concessive conditionals. The lack of scalarity in most wh- concessive conditionals is argued to cast into doubt the category label “concessive conditional” applied to these constructions in a substantial part of the literature and to favour an alternative designation such as “irrelevance conditional.” The empirical data further reveals that wh- concessive conditionals practically never involve pure ignorance, and this is argued to be problematic on the discourse-pragmatic level for the claim that they have interrogative semantics.

There are broad disagreements between existing models regarding the mental representations and processes involved in the "DEGREE ADVERB + PROPER NAME" construction, including divergences regarding the semantics of the degree device, the category status of the proper name, the construction's expressed meaning, its compositionality, and, crucially, the operation holding between the degree device and the proper name. Our corpus-based investigation of two competing models from Construction Grammar and Formal Semantics shows that while both make useful contributions to the scientific understanding of the construction, neither is empirically adequate. Most importantly, we find that the construction participates in several non- predicted expressed meanings; multivariate analyses show that the three meanings amenable to statistical analysis cluster with different semantic usage-features. We argue that the best way to account for the construction's semantics/pragmatics is via a previously-dismissed cognitive mechanism: an enrichment/strengthening-type operation whereby a pragmatically-supplied scale is added to the message.

An abstract is not available for this content so a preview has been provided. Please use the Get access link above for information on how to access this content.

This book steers a middle course between two opposing conceptions that currently dominate the field of semantics, the logical and cognitive approaches. It brings to light the inadequacies of both frameworks, and argues along with the Columbia School that linguistic semantics must be grounded on the linguistic sign itself and the meaning it conveys across the full range of its uses. The book offers 12 case studies demonstrating the explanatory power of a sign-based semantics, dealing with topics such as complementation with aspectual and causative verbs, control and raising, wh- words, full-verb inversion, and existential-there constructions. It calls for a radical revision of the semantics/pragmatics interface, proposing that the dividing-line be drawn between semiologically-signified notional content (i.e. what is linguistically encoded) and non-semiologically-signified notional content (i.e. what is not encoded but still communicated). This highlights a dimension of embodiment that concerns the basic design architecture of human language itself: the ineludable fact that the fundamental relation on which language is based is the association between a mind-engendered meaning and a bodily produced sign. It is argued that linguistic analysis often disregards this fact and treats meaning on the level of the sentence or the construction, rather than on that of the lower-level linguistic items where the linguistic sign is stored in a stable, permanent, and direct relation with its meaning outside of any particular context. Building linguistic analysis up from the ground level provides it with a more solid foundation and increases its explanatory power.

There are broad disagreements between existing models regarding the mental representations and processes involved in the "DEGREE ADVERB + PROPER NAME" construction, including disagreements regarding the semantics of the degree device, the category status of the proper name, the construction’s expressed meaning and its (non-)compositionality, and, crucially, the operation that holds between the degree device and the proper name. Our corpus-based investigation into two competing models from Construction Grammar and Formal Semantics shows that these models collectively make useful contributions to the scientific understanding of this construction, but neither is empirically adequate. Most importantly, we find that the construction participates in several non-predicted expressed meanings; multivariate analyses show that the three amenable to statistical analysis cluster with different semantic usage-features. We argue that the best way to account for the construction’s semantics-pragmatics is via a previously-dismissed cognitive mechanism: an enrichment-/strengthening-type operation whereby a pragmatically-supplied scale is added to the message.

Previous research on the Present Perfect suggests that the latter is acceptable in since-clauses only if the situation expressed by the subordinate clause extends in time up until the present, as in Tony has been happy since he has been taking Prozac, i.e. if the Present Perfect in the since-clause is a Universal Perfect whose predication is true at all subintervals in the time-interval delimited by since and not an Existential Perfect whose predication is true only at a certain moment therein, as in *Tony has been happy since he has visited Cape Cod. The present study puts the implications of this claim and others found in the literature to the empirical test through an examination of all of the 2,621 occurrences of the Present Perfect in since-clauses with pronominal subjects in the British National Corpus and the Corpus of Contemporary American English. A significant proportion of the data is shown to contradict the generalizations proposed in previous studies. An explanation capable of accounting for all of the uses of the Present Perfect in since-clauses is put forward based on the complex meaning of this verbal structure which involves the notion of present stative after-phase of a past situation. The Present Perfect is possible in a since-clause whenever the beginning of the after-phase serves to mark the beginning-point in the past from which the main-clause situation unfolds towards the present. The beginning of the after-phase is argued to be interpretable in two ways: either as the first moment after the end of the prior situation or as the first moment in the process of accruing the after-phase itself.

This article explores the tension between the antithetical philosophies of empiricism and innatism underlying Chomskyan linguistics. It first follows the trail of empiricism in North American linguistics, starting from the work of Leonard Bloomfield at the beginning of the Twentieth century, and its influence on the Chomskyan paradigm, after which the Kantian trail of innatism initiated by Chomsky himself is reconnoitered. It is argued that the Chomskyan approach to natural language represents a paradigmatic example of the unsavory consequences of the divorce between mind and matter instituted by Kant, in particular because human language involves an intimate relation between both types of reality. In Chomsky’s Generative Grammar, on the other hand, the material side of language is treated as completely autonomous from its mental correlate and analyzed in terms of a priori conceptual structures and computational operations; for its part, the mental side of language is treated as innate; the relation between the two is thus made utterly obscure and incomprehensible. The conclusion of the article argues in favour of a more balanced approach inspired by Aristotelianism and Paul Ricoeur’s hermeneutics.

In reply to the claim that syntax is not taken into account in Linguistic Meaning Meets Linguistic Form, I show that local syntactic analysis has been implemented in the treatment of aspectual verbs and verbs of positive and negative recall, where the syntactic function of the -ing form as direct object of the main verb is put into relation with the main verb's meaning as the basis for the inferences drawn concerning the temporal relation between the main verb's event and that expressed by the complement. I argue that I have also developed new tools of syntactic analysis for the to-infinitive, demonstrating that it is not the direct object of the main verb, but rather a goal-or result-specifier, and showing how this accounts for the fact that its event is always understood to be somehow subsequent to that of the main verb. Regarding the applicability of formal semantics to natural language, I argue that the absolute priority accorded to the truth-functional dimension of language by this type of semantics leads to the artificial separation of use-conditions from truth-conditions, with the former being treated as an additional interpretational function added on to the truth-functional one. Contra the autonomous syntax claim that our desire to express meaning is to a great extent independent of the means we use to express those meanings, it is argued that how we perceive the world in our experience is influenced by our system of linguistic representation.

Reply to John Collins A Plea for Explanation.

In mainstream generative grammar (MGG), Logical Form (LF) is at the same time a syntactic representation delivered by the grammar and a representation that can be truth-conditionally evaluated. Iacona (2018) argues however that formal logic raises a problem for MGG's intrinsicalist notion of LF: logic requires that the logical form assigned to a sentence should not depend solely on the sentence itself, but also on the logical relations the sentence bears to other propositions in the language, making logicians' logical form dependent on something outside the sentence. This paper argues that LF is also not linguistic enough for linguistics because it is modelled on formal logic rather than on careful analysis of natural language and the natural articulations of linguistic utterances. This leads to a misidentification of the basic linguistic unit, which is not the sentence, and a misanalysis of the sentence's basic structure. The conclusion argues that the only methodologically sound basis for linguistic analysis is not truth, nor the sentence that expresses it, but rather the means that allow the speaker to produce the utterance in the first place, the stable pairings of meaning and sound corresponding to the units that the speaker has stored in memory.

This chapter argues that the autonomous syntax postulate leads to a view of semantics as interfacing with syntax only once the latter has autonomously generated a sentential sequence, which is then sent to the semantic component for interpretation. Since only sentences are true or false, this situates semantics on the level of propositions and the truth-conditions associated with them. While this hypothesis has the advantage of providing linguists with a ready-made toolkit of logical concepts and formulae, it is argued to misrepresent the way meaning is related to form in natural language, as it is generally not the case that a stable meaning is paired with a stable linguistic form on the level of the sentence.

This chapter highlights a dimension of embodiment that is often overlooked and that concerns the basic design architecture of human language itself: the ineludable fact that the fundamental relation on which language is based is the association between a mind-engendered meaning and a bodily-produced sign. It is argued that this oversight is often due to treating meaning on the level of the sentence or the construction, rather than of the lower-level linguistic items where the linguistic sign is stored in a stable, permanent, and direct relation with its meaning outside of any particular context. Building linguistic analysis up from the ground level provides it with a solid foundation and increases its explanatory power.

This chapter demonstrates the explanatory power of an approach that grounds the analysis of natural-language meaning on the linguistic sign itself. Cases covered include the multifarious uses of the preposition for, verbal complementation with aspectual and causative verbs, the phenomena of control and raising in adjective + to-infinitive constructions, the use of wh- words with the bare and to-infinitives, the modal and non-modal uses of the verbs dare and need, and a meaning-based account of full-verb inversion and existential-there constructions.

This chapter lays down the gauntlet to challenge the autonomy of syntax, a basic principle of linguistic analysis proposed by Chomsky in 1982 and followed by most researchers working in the generative paradigm. It announces the book’s intent to argue that semantics plays a highly significant role in syntax, and that a properly articulated linguistic semantics, together with the requisite pragmatics, goes a long way towards explaining the relational processes involved in the building of syntactic sequences in natural language. This relates the approach taken in the book to those advocated by the Columbia School and by Cognitive Grammar.