Elements in Construction Grammar – serie

Construction Grammar (CxG) has developed into a broad and highly diverse family of approaches that have in common that they see constructions, i.e. form-meaning pairs at various levels of abstraction and complexity, as the basic units of language. This Element gives an overview of the origin and the current state of the art of constructionist approaches, focusing, on the one hand, on basic concepts like the notion of 'constructions', while at the same time offering an in-depth discussion of current research trends and open questions. It discusses the commonalities and differences between the major constructionist approaches, the organization of constructional networks as well as ongoing research on linguistic creativity, multimodality and individual differences. This title is also available as Open Access on Cambridge Core.

It is one of the central claims of construction grammar that constructions are organized in some kind of network, commonly referred to as the constructicon. In the classical model of construction grammar, developed by Berkeley linguists in the 1990s, the constructicon is an inheritance network of taxonomically related grammatical patterns. However, recent research in usage-based linguistics has expanded the classical inheritance model into a multidimensional network approach in which constructions are interrelated by multiple types of associations. The multidimensional network approach challenges longstanding assumptions of linguistic research and calls for a reorganization of the constructivist approach. This Element describes how the conception of the constructicon has changed in recent years and elaborates on some central claims of the multidimensional network approach.

Construction Grammar has gained prominence in linguistics, owing its popularity to its inclusive approach that considers language units of varying sizes and generality as potential constructions - mentally stored form-function units. This Element serves as a cautionary note against complacency and dogmatism. It emphasizes the enduring importance of falsifiability as a criterion for scientific hypotheses and theories. Can every postulated construction, in principle, be empirically demonstrated not to exist? As a case study, the author examines the schematic English transitive verb-particle construction, which defies experimental verification. He argues that we can still reject its non-existence using sound linguistic reasoning. But beyond individual constructions, what could be a crucial test for Construction Grammar itself, one that would falsify it as a theory? In making a proposal for such a test, designed to prove that speakers also exhibit pure-form knowledge, this Element contributes to ongoing discussions about Construction Grammar's theoretical foundations.

This Element in Construction Grammar addresses one of its hottest topics and asks: is the unimodal conception of Construction Grammar as a model of linguistic knowledge at odds with the usage-based thesis and the multimodality of language use? Are constructions verbal, i.e. unimodal form-meaning pairings, or are they, or at least are some of them, multimodal in nature? And, more fundamentally, how do we know? These questions have been debated quite controversially over the past few years. This Element presents the current state of research within the field, paying special attention to the arguments that are put forward in favour and against the uni-/multimodal nature of constructions and the various case studies that have been conducted. Although significant progress has been made over the years, the debate points towards a need for a diversification of the questions asked, the data studied, and the methods used to analyse these data.

This Element presents a computational theory of syntactic variation that brings together (i) models of individual differences across distinct speakers, (ii) models of dialectal differences across distinct populations, and (iii) models of register differences across distinct contexts. This computational theory is based in Construction Grammar (CxG) because its usage-based representations can capture differences in productivity across multiple levels of abstraction. Drawing on corpora representing over 300 local dialects across fourteen countries, this Element undertakes three data-driven case-studies to show how variation unfolds across the entire grammar. These case-studies are reproducible given supplementary material that accompanies the Element. Rather than focus on discrete variables in isolation, we view the grammar as a complex system. The essential advantage of this computational approach is scale: we can observe an entire grammar across many thousands of speakers representing dozens of local populations.

How do we understand any sentence, from the most ordinary to the most creative? The traditional assumption is that we rely on formal rules combining words (compositionality). However, psycho- and neuro-linguistic studies point to a linguistic representation model that aligns with the assumptions of Construction Grammar: there is no sharp boundary between stored sequences and productive patterns. Evidence suggests that interpretation alternates compositional (incremental) and noncompositional (global) strategies. Accordingly, systematic processes of language productivity are explainable by analogical inferences rather than compositional operations: novel expressions are understood 'on the fly' by analogy with familiar ones. This Element discusses compositionality, alternative mechanisms in language processing, and explains why Construction Grammar is the most suitable approach for formalizing language comprehension.

AI can assist the linguist in doing research on the structure of language. This Element illustrates this possibility by showing how a conversational AI based on a Large Language Model (AI LLM chatbot) can assist the Construction Grammarian, and especially the Frame Semanticist. An AI LLM chatbot is a text-generation system trained on vast amounts of text. To generate text, it must be able to find patterns in the data and mimic some linguistic capacity, at least in the eyes of a cooperative human user. The authors do not focus on whether AIs "understand" language. Rather, they investigate whether AI LLM chatbots are useful tools for linguists. They reframe the discussion from what AI LLM chatbots can do with language to what they can do for linguists. They find that a chatty LLM can labor usefully as an eliciting interlocutor, and present precise, scripted routines for prompting conversational LLMs.

AI can assist the linguist in doing research on the structure of language. This Element illustrates this possibility by showing how a conversational AI based on a Large Language Model (AI LLM chatbot) can assist the Construction Grammarian, and especially the Frame Semanticist. An AI LLM chatbot is a text-generation system trained on vast amounts of text. To generate text, it must be able to find patterns in the data and mimic some linguistic capacity, at least in the eyes of a cooperative human user. The authors do not focus on whether AIs "understand" language. Rather, they investigate whether AI LLM chatbots are useful tools for linguists. They reframe the discussion from what AI LLM chatbots can do with language to what they can do for linguists. They find that a chatty LLM can labor usefully as an eliciting interlocutor, and present precise, scripted routines for prompting conversational LLMs.

Construction Grammar has gained prominence in linguistics, owing its popularity to its inclusive approach that considers language units of varying sizes and generality as potential constructions - mentally stored form-function units. This Element serves as a cautionary note against complacency and dogmatism. It emphasizes the enduring importance of falsifiability as a criterion for scientific hypotheses and theories. Can every postulated construction, in principle, be empirically demonstrated not to exist? As a case study, the author examines the schematic English transitive verb-particle construction, which defies experimental verification. He argues that we can still reject its non-existence using sound linguistic reasoning. But beyond individual constructions, what could be a crucial test for Construction Grammar itself, one that would falsify it as a theory? In making a proposal for such a test, designed to prove that speakers also exhibit pure-form knowledge, this Element contributes to ongoing discussions about Construction Grammar's theoretical foundations.

This Element offers a primer for the study of meaning in a Construction Grammar approach. It reviews the main principles of meaning shared across constructionist frameworks, including its ubiquity in grammatical structure, its usage-based formation, and its nature as the output of cognitive representations. It also reviews the importance given to meaning in construction-based explanations of sentence composition, innovative language use, and language change. Paradoxically, the Element shows that there is no systematic framework delineating the rich structure of constructional meaning, which has led to theoretical disagreements and inconsistencies. It therefore proposes an operational model of meaning for practitioners of Construction Grammar. It details the characteristics of a complex interface of semantic, pragmatic, and social meaning, and shows how this framework sheds light on recent theoretical issues. The Element concludes by considering ways in which this framework can be used for future descriptive and theoretical research questions.

This Element offers a primer for the study of meaning in a Construction Grammar approach. It reviews the main principles of meaning shared across constructionist frameworks, including its ubiquity in grammatical structure, its usage-based formation, and its nature as the output of cognitive representations. It also reviews the importance given to meaning in construction-based explanations of sentence composition, innovative language use, and language change. Paradoxically, the Element shows that there is no systematic framework delineating the rich structure of constructional meaning, which has led to theoretical disagreements and inconsistencies. It therefore proposes an operational model of meaning for practitioners of Construction Grammar. It details the characteristics of a complex interface of semantic, pragmatic, and social meaning, and shows how this framework sheds light on recent theoretical issues. The Element concludes by considering ways in which this framework can be used for future descriptive and theoretical research questions.

How do we understand any sentence, from the most ordinary to the most creative? The traditional assumption is that we rely on formal rules combining words (compositionality). However, psycho- and neuro-linguistic studies point to a linguistic representation model that aligns with the assumptions of Construction Grammar: there is no sharp boundary between stored sequences and productive patterns. Evidence suggests that interpretation alternates compositional (incremental) and noncompositional (global) strategies. Accordingly, systematic processes of language productivity are explainable by analogical inferences rather than compositional operations: novel expressions are understood 'on the fly' by analogy with familiar ones. This Element discusses compositionality, alternative mechanisms in language processing, and explains why Construction Grammar is the most suitable approach for formalizing language comprehension.

The topic of linguistic networks unites different frameworks in cognitive linguistics. This Element explores two approaches to networks, specifically Construction Grammar of the Goldberg variety and Word Grammar as developed by Hudson, and how they inform work on language change. Both are usage-based theories, but while the basic units of Construction Grammar are conventionalized form-meaning pairings gathered in a construct-i-con, the basic units of Word Grammar are words in dependency and other relations. Construction Grammar allows for schematic, hierarchized abstract generalizations attributable to social groups, whereas Word Grammar focuses on relations at the micro-level and attributable primarily to individuals. Consequences of the differences are discussed with reference to perspectives on the diachronic development of causal connectives in English, especially because.

The topic of linguistic networks unites different frameworks in cognitive linguistics. This Element explores two approaches to networks, specifically Construction Grammar of the Goldberg variety and Word Grammar as developed by Hudson, and how they inform work on language change. Both are usage-based theories, but while the basic units of Construction Grammar are conventionalized form-meaning pairings gathered in a construct-i-con, the basic units of Word Grammar are words in dependency and other relations. Construction Grammar allows for schematic, hierarchized abstract generalizations attributable to social groups, whereas Word Grammar focuses on relations at the micro-level and attributable primarily to individuals. Consequences of the differences are discussed with reference to perspectives on the diachronic development of causal connectives in English, especially because.

This Element in Construction Grammar addresses one of its hottest topics and asks: is the unimodal conception of Construction Grammar as a model of linguistic knowledge at odds with the usage-based thesis and the multimodality of language use? Are constructions verbal, i.e. unimodal form-meaning pairings, or are they, or at least are some of them, multimodal in nature? And, more fundamentally, how do we know? These questions have been debated quite controversially over the past few years. This Element presents the current state of research within the field, paying special attention to the arguments that are put forward in favour and against the uni-/multimodal nature of constructions and the various case studies that have been conducted. Although significant progress has been made over the years, the debate points towards a need for a diversification of the questions asked, the data studied, and the methods used to analyse these data.

This Element presents a computational theory of syntactic variation that brings together (i) models of individual differences across distinct speakers, (ii) models of dialectal differences across distinct populations, and (iii) models of register differences across distinct contexts. This computational theory is based in Construction Grammar (CxG) because its usage-based representations can capture differences in productivity across multiple levels of abstraction. Drawing on corpora representing over 300 local dialects across fourteen countries, this Element undertakes three data-driven case-studies to show how variation unfolds across the entire grammar. These case-studies are reproducible given supplementary material that accompanies the Element. Rather than focus on discrete variables in isolation, we view the grammar as a complex system. The essential advantage of this computational approach is scale: we can observe an entire grammar across many thousands of speakers representing dozens of local populations.

In null instantiation (NI) an optionally unexpressed argument receives either anaphoric or existential interpretation. One cannot accurately predict a predicator's NI potential based either on semantic factors (e.g., Aktionsart class of the verb) or pragmatic factors (e.g., relative discourse prominence of arguments), but NI potential, while highly constrained, is not simply lexical idiosyncrasy. It is instead the product of both lexical and constructional licensing. In the latter case, a construction can endow a verb with NI potential that it would not otherwise have. Using representational tools of sign based construction grammar, this Element offers a lexical treatment of English null instantiation that covers both distinct patterns of construal of null-instantiated arguments and the difference between listeme-based and contextually licensed, thus construction-based, null complementation.