This paper exploits recent work on the normative and constitutive roles of knowledge in practical rationality, to put pressure on the idea that speakers could communicate without exploiting linguistic knowledge. I defend cognitivism about meaning, the view that speakers have rationally accessible (i.e., implicit rather than tacit) knowledge of semantic facts and principles, and that this knowledge is constitutive of their linguistic competence.

Over the last decade, AI models of language and word meaning have been dominated by what we might call a statistics-of-occurrence, strategy: these models are deep neural net structures that have been trained on a large amount of unlabeled text with the aim of producing a model that exploits statistical information about word and phrase co-occurrence in order to generate behavior that is similar to what a human might produce, or representations that can be probed to exhibit behavior similar to (...) what a human might produce (meaningsemblant behavior). Examples of what we can call Statistics-of-Occurrence Models (SOMs) include: Word2Vec (CBOW and Skip-Gram), BERT, GPT-3, and, most recently, ChatGPT. Increasingly, there have been suggestions that such systems have semantic understanding, or at least a proto-version of it. This paper argues against such claims. I argue that a necessary condition for a system to possess semantic understanding is that it function in ways that are causally explainable by appeal to its semantic properties. I then argue that SOMs do not plausibly satisfy this Functioning Criterion. Rather, the best explanation of their meaning-semblant behavior is what I call the Statistical Hypothesis: SOMs do not themselves function to represent or produce meaningful text; they just reflect the semantic information that exists in the aggregate given strong correlations between word placement and meaningful use. I consider and rebut three main responses to the claim that SOMs fail to meet the Functioning Criterion. The result, I hope, is increased clarity about why and how one should make claims about AI systems having semantic understanding. (shrink)

Understanding Semantics offers an up-to-date, broad and thorough introduction to linguistic semantics, the field of linguistic meaning central to the understanding of language. The book takes a step-by-step approach, starting with the basic concepts and moving through central questions to an examination of the methods and results of the science of linguistic meaning.

This essay considers what it means to understand natural language and whether a computer running an artificial-intelligence program designed to understand natural language does in fact do so. It is argued that a certain kind of semantics is needed to understand natural language, that this kind of semantics is mere symbol manipulation (i.e., syntax), and that, hence, it is available to AI systems. Recent arguments by Searle and Dretske to the effect that computers cannot understand natural language are discussed, and (...) a prototype natural-language-understanding system is presented as an illustration. (shrink)

In his essay ‘“Conceptual Truth”’, Timothy Williamson (2006) argues that there are no truths or entailments that are constitutive of understanding the sentences involved. In this reply I provide several examples of entailment patterns that are intuitively constitutive of understanding in just the way that Williamson rejects, and I argue that Williamson’s argument does nothing to show otherwise. Williamson bolsters his conclusion by appeal to a certain theory about the nature of understanding. I argue that his theory (...) fails to consider the role that the structure of a sentence plays in determining its meaning. The cases I present suggest that this role imposes greater cognitive requirements on understanding than Williamson can acknowledge. (shrink)

John Searle once said: "The Chinese room shows what we knew all along: syntax by itself is not sufficient for semantics. (Does anyone actually deny this point, I mean straight out? Is anyone actually willing to say, straight out, that they think that syntax, in the sense of formal symbols, is really the same as semantic content, in the sense of meanings, thought contents, understanding, etc.?)." I say: "Yes". Stuart C. Shapiro has said: "Does that make any sense? (...) Yes: Everything makes sense. The question is: What sense does it make?" This essay explores what sense it makes to say that syntax by itself is sufficient for semantics. (shrink)

In his monograph Semantic Relationism, Kit Fine proposes two characterizations of coordination between proper names: an intuitive test; and a technical definition. The intuitive characterization is grounded in a notion of understanding distinct from the familiar notion of linguistic competence. Three prima facie appealing proposals to characterize this notion of understanding will be examined in the present paper and then dismissed as intrinsically implausible or as incompatible with Fine’s semantics. Not even his technical characterization of coordination, involving (...) the notion of semantic requirement, will enable us to escape the impasse. Ultimately, the question of what exactly coordination between names is will remain open. (shrink)

This essay is part of a larger project aimed at making sense of rational thought and agency as part of the natural world. It provides a semantic framework for thinking about the contents of: 1) descriptive thoughts and sentences having a representational or mind-to-world direction of fit, and which manifest our capacity for theoretical rationality; and 2) prescriptive and intentional sentences having an expressive or world-to-mind direction of fit, and which manifest our capacity for practical rationality. I use a (...) modified version of Allan Gibbard’s hyperstate semantics, employing both maximally determinate possible worlds and maximally determinate plans of action, as a basis for providing a unified understanding of moral judgments and expressions of individual and collective intentionality – they one and all give voice to our ability as rational agents to adopt the perspectives of various individuals within a community and consider how we would behave were we in their positions. In the course of spelling out the view I draw on and criticize ideas that Wilfrid Sellars advanced in the middle of the 20th century, while employing the tools of contemporary modal logic and model-theoretic semantics to give perspicuous formulation to his thought that the moral judgment ‘one ought to do A’ should be understood in terms of the collective intention ‘we shall do A’, where the pronoun denotes the unrestricted class of rational agents and the ‘ought’ is in some sense unconditionally binding. (shrink)

The topic of the present essay is the proper understanding of Donald Davidson’s version of the publicity requirement for the determinants of linguistic meaning. On the understanding I promote, the requirement is very strict indeed. My narrow aim is to show how such a strict conception of the publicity requirement can be maintained despite the evident need for interpreters to go beyond what is public on that conception in the process of constructing Davidsonian theories of meaning. Towards that (...) aim, I engage dialectically with treatments of Davidson’s principle of charity owing to Lepore and Ludwig and to Bar-On and Risjord, each of which, in different ways, recommend a more permissive approach to the publicity requirement than the one I recommended here. A broader aim is to shed some light on what would be required to take seriously the larger ambitions of Davidson’s semantic program. (shrink)

Understandably absorbed in technical details, discussion of the semantic paradoxes risks losing sight of broad methodological principles. This chapter sketches a general approach to the comparison of rival logics, and applies it to argue that revision of classical propositional logic has much higher costs than its proponents typically recognize.

Most theories of semantic competence in natural language implicitly assume the Language of Thought Hypothesis. According to this hypothesis, all human cognition consists in the deployment of a language of thought. This language of thought is supposed to be independent of natural language, yet at the same time, it is supposed to be semantically isomorphic with natural language. Given this assumption, it is easy to answer basic questions regarding semantic competence in natural language. What are semantic properties (...) of natural language, and how did they originate? Natural language inherits the semantic properties of the language of thought it is used to express. What does semantic competence consist in? It consists in knowing how to translate items in one's language of thought into a public medium. ;The starting point of this dissertation is the rejection, by some prominent paradigms in cognitive science, of the view that human cognition consists in the deployment of a language of thought that is, at the same time, independent of natural language, yet semantically isomorphic with it. The dissertation constitutes a fledgling attempt to explore an alternative approach to explaining semantic competence in natural language, an approach that is compatible with this rejection of the Language of Thought Hypothesis. ;Rather than conceive of natural language as a tool for expressing items in a pre-existing, semantically isomorphic, language of thought, I urge that language be conceived of as a norm-governed social practice. On this view, the semantic properties of natural language are determined by the inferential norms governing linguistic performances. In order to explain the origin of its semantic properties, one must explain the origin of the inferential norms that govern language. In order to explain what semantic competence consists in, one must explain how individuals learn to abide by these inferential norms. The bulk of the dissertation is devoted to showing that neither of these explanatory projects requires endorsing the Language of Thought Hypothesis. I do this by sketching the form that such explanations might take, and showing that they are consistent with much current research in ethology, comparative psychology, developmental psychology, and evolutionary biology. (shrink)

Elizabeth Fricker; IV*—Semantic Structure and Speakers' Understanding1, Proceedings of the Aristotelian Society, Volume 83, Issue 1, 1 June 1983, Pages 49–66, h.

People undergo many idiosyncratic experiences throughout their lives that may contribute to individual differences in the size and structure of their knowledge representations. Ultimately, these can have important implications for individuals' cognitive performance. We review evidence that suggests a relationship between individual experiences, the size and structure of semantic representations, as well as individual and age differences in cognitive performance. We conclude that the extent to which experience-dependent changes in semantic representations contribute to individual differences in cognitive aging (...) remains unclear. To help fill this gap, we outline an empirical agenda that utilizes network analysis and involves the concurrent assessment of large-scale semantic networks and cognitive performance in younger and older adults. We present preliminary data to establish the feasibility and limitations of such empirical, network-analytical approaches. (shrink)

Increasing computational power and improving deep learning methods have made computer vision technologies pervasively common in urban environments. Their applications in policing, traffic management, and documenting public spaces are increasingly common (Ridgeway 2018, Coifman et al. 1998, Sun et al. 2020). Despite the often-discussed biases in the algorithms' training and unequally borne benefits (Khosla et al. 2012), almost all applications similarly reduce urban experiences to simplistic, reductive, and mechanistic measures. There is a lack of context, depth, and specificity in these (...) practices that enables semantic knowledge or analysis within urban contexts, especially within the context of using and occupying urban space. This paper will critique existing uses of artificial intelligence and computer vision in urban practices to propose a new framework for understanding people, action, and public space. This paper revisits Geertz's (1973) use of thick descriptions in generating interpretive theories of culture and activity and uses this lens to establish a framework to approach evaluating the varied uses of computer vision technologies that weigh meaning. By discussing cases of implemented examples of urban computer vision—from LinkNYC and Numina's urban measurements to the Detroit Police's use of DataWorks Plus's facial recognition technology—it proposes a framework for evaluating the thickness of the algorithm's conclusions against the computational method's complexity required to produce that outcome. Further, we discuss how the framework's positioning may differ (and conflict) between different users of the technology, from engineer to urban planner and policymaker, to citizen. This paper also discusses how the current use and training of deep learning algorithms and how this process limits semantic learning and proposes three potential methodologies toward gaining a more contextually specific, urban-semantic, description of urban space relevant to urbanists. This paper contributes to the critical conversations regarding the proliferation of artificial intelligence by challenging the current applications of these technologies in the urban environment by highlighting their failures within this context while also proposing an evolution of these algorithms that may ultimately make them sensitive and useful within this spatial and cultural milieu. (shrink)

This paper critiques a number of standard ways of understanding the role of the metalanguage in a semantic theory for natural language, including the idea that disquotation plays a nontrivial role in any explanatory natural language semantics. It then proposes that the best way to understand the role of a semantic metalanguage involves recognizing that semantics is a model-based science. The metalanguage of semantics is language for articulating features of the theorist's model. Models are understood as mediating (...) instruments---idealized structures used to represent select aspects of the world, aspects the theorist is seeking some theoretical understanding of. The aspect of reality we are seeking some understanding of in semantics is a dimension of human linguistic competence---informally, knowledge of meaning. (shrink)

Presented in this book is a theory of concept formation and understanding that does not make use of a notion of an innate mental language as a means of concept representation. Instead, experimental concepts are treated semantically as stabilising structuring of growing sets of data, which are sets of experienced satisfaction situations for expressions, and theoretical concepts are based on coherent sets of general sentences held true. There are two kinds of structures to be established: general concepts by means (...) of similarity sets under perspectives and historical concepts. This gives rise to a theory of understanding new situations and expressions by integrating new data into established sets of data salva stability, or by extending the conceptual structure in a metaphorical or metonymical way. The theory provides a way to understand what identity between propositional attitudes amounts to, especially how people can have more or less the same belief. (shrink)

Kit Fine (2007) outlines an account of semantic coordination, an account motivated by the role of semantic coordination in cognition. Actually, Fine outlines two accounts of semantic coordination, one in terms of co-reference and another in terms of synonymy. I argue, first, that Fine's two accounts are not equivalent, with one being logically stronger than the other, but second and more importantly, that neither account is correct. I outline an alternative account of semantic coordination – the (...) epistemic conception of semantic coordination – that links semantic and epistemic aspects of cognition more directly. The most surprising result of my alternative account of semantic coordination is that semantic coordination is a relation that can hold between elements of thought and language of different semantic types. I close the paper by briefly outlining some applications of the idea that semantic coordination is a relation that can hold across semantic types. (shrink)

This meta-study of animal semantics is anchored in two claims, seemingly creating a fuzzy mismatch, that animal utterances generally appear to be simple in structure and content variation and that animals’ communicative understanding seems disproportionally more advanced. A set of excerpted, new studies is chosen as basis to discuss whether the semantics of animal uttering and understanding can be fused into one. Studies are prioritised due to their relatively complex designs, giving priority to dynamics between syntax, semantics, pragmatics, (...) and between utterers and receivers in context. A communicational framework based on utterance theory is applied as a lens for inspection of how these aspects relate to the assumed mismatch. Inspection and discussions of the studies bring several features to surface of which five are stressed in the following. Firstly, both syntactic structures and possible semantic content are seen as lean, although richer than earlier believed, and research continues to reveal new complexities in utterances. Secondly, there is a clear willingness to broaden the perception of animals’ semantic capacity to comprehend communication both by arguing theoretically and by generating empirical research in new contexts. Thirdly, the ambition to make sense of these tendencies is still often motivated by an evolutionary search for early building blocks for verbal language, with the pro et cons that such a position can have. Fourthly, the ‘allowed’ scientific frame for studying semantic capacity among animals is extended to new fields and contexts challenging the only-in-the-wild norm. Fifthly, the dilemma of integrating uttering and understanding as aspects of an after all functional communicational system, calls for new epistemological concepts to make sense of the claimed mismatch. Affordances, abduction, life-genre, and lifeworld are suggested. (shrink)

The book series Linguistische Arbeiten (LA) publishes high-quality work in linguistics that addresses current issues in synchrony and diachrony, theoretically or empirically oriented.

I advocate a theory of syntactic semantics as a way of understanding how computers can think (and how the Chinese-Room-Argument objection to the Turing Test can be overcome): (1) Semantics, considered as the study of relations between symbols and meanings, can be turned into syntax – a study of relations among symbols (including meanings) – and hence syntax (i.e., symbol manipulation) can suffice for the semantical enterprise (contra Searle). (2) Semantics, considered as the process of understanding one domain (...) (by modeling it) in terms of another, can be viewed recursively: The base case of semantic understanding –understanding a domain in terms of itself – is syntactic understanding. (3) An internal (or narrow ), first-person point of view makes an external (or wide ), third-person point of view otiose for purposes of understanding cognition. (shrink)

In this research, Description Logics (DLs) will be employed for logical description, logical characterisation, logical modelling and ontological description of concept understanding in terminological systems. It’s strongly believed that using a formal descriptive logic could support us in revealing logical assumptions whose discovery may lead us to a better understanding of ‘concept understanding’. The Structure of Observed Learning Outcomes (SOLO) model as an appropriate model of increasing complexity of humans’ understanding has supported the formal analysis.

Semantic priming has been a focus of research in the cognitive sciences for more than 30 years and is commonly used as a tool for investigating other aspects of perception and cognition, such as word recognition, language comprehension, and knowledge representations. Semantic Priming: Perspectives from Memory and Word Recognition examines empirical and theoretical advancements in the understanding of semantic priming, providing a succinct, in-depth review of this important phenomenon, framed in terms of models of memory and (...) models of word recognition. The first section examines models of semantic priming, including spreading activation models, the verification model, compound-cue models, distributed network models, and multistage activation models (e.g. interactive-activation model). The second section examines issues and findings that have played an especially important role in testing models of priming and includes chapters on the following topics: methodological issues (e.g. counterbalancing of materials, choice of priming baselines); automatic vs. strategic priming; associative vs. "pure" semantic priming; mediated priming; long-term semantic priming; backward priming; unconscious priming; the prime-task effect; list context effects; effects of word frequency, stimulus quality, and stimulus repetition; and the cognitive neuroscience of semantic priming. The book closes with a summary and a discussion of promising new research directions. The volume will be of interest to a wide range of researchers and students in the cognitive sciences and neurosciences. (shrink)

Semantic Leaps explores how people combine knowledge from different domains in order to understand and express new ideas. Concentrating on dynamic aspects of on-line meaning construction, Coulson identifies two related sets of processes: frame-shifting and conceptual blending. Frame-shifting is semantic reanalysis in which existing elements in the contextual representation are reorganized into a new frame. Conceptual blending is a set of cognitive operations for combining partial cognitive models. By addressing linguistic phenomena often ignored in traditional meaning research, Coulson (...) explains how processes of cross-domain mapping, frame-shifting, and conceptual blending enhance the explanatory adequacy of traditional frame-based systems for natural language processing. The focus is on how the constructive processes speakers use to assemble, link, and adapt simple cognitive models underlie a broad range of productive language behavior. (shrink)

This article aims to demonstrate the interdependence of semantics and noetics against the referentialist trend in Abelard studies conceiving semantics as confined to the truth/falsity function. The article takes as a turning point of the argument Abelard’s question “can a true proposition generate a false understanding?” which secondary literature does not take into account. Starting from the analysis of this question, the article aims to show the development of an enhanced notion of understanding compared to the Boethian one. (...) The core of this enhanced notion of understanding is the action of attentio, which is the condition for the elaboration of a coherent and global definition of signification. The analysis of the mental action of attentio entails the description of other kinds of mental actions, leading one to conceive understanding as a multi-layered complex of actions. (shrink)

Suppose, familiarly, that you and a friend have landed in an alien territory, amidst people who speak a language you do not know. Upon seeing you, one of them starts yelling, seemingly alarmed. You say to your friend, “She thinks we want to hurt her. She's scared. We must seem very strange to her.” Your friend, who is facing you, says, “No, I think she's actually trying to warn you: there's a snake right above your head, on that tree. You (...) see the sling in her hand? I think she's going to try to shoot it down.“On a prevalent view, much discussed in recent years, you and your friend have engaged in a mini-theoretical enterprise. Using certain observations of the alien's behaviour as your data, and deploying certain generalizations and principles concerning human behaviour, you advance hypotheses regarding the internal psychological states which issued in her observed behaviour. (shrink)

The article argues that cognitive linguistic theory may prove an alternative to the Montague paradigm for designing natural language understanding systems. Within this framework it describes a system which models language understanding as a dialogical process between user and computer. The system operates with natural language texts as input and represent language meaning as entity-relationship diagrams.

This major publication is a history of the semantic tradition in philosophy from the early nineteenth century through its incarnation in the work of the Vienna Circle, the group of logical positivists that emerged in the years 1925-1935 in Vienna who were characterised by a strong commitment to empiricism, a high regard for science, and a conviction that modern logic is the primary tool of analytic philosophy. In the first part of the book, Alberto Coffa traces the roots of (...) logical positivism in a semantic tradition that arose in opposition to Kant's theory that a priori knowledge is based on pure intuition and the constitutive powers of the mind. In Part II, Coffa chronicles the development of this tradition by members and associates of the Vienna Circle. Much of Coffa's analysis draws on the unpublished notes and correspondence of many philosophers. The book, however, is not merely a history of the semantic tradition from Kant 'to the Vienna Station'. Coffa also critically reassesses the role of semantic notions in understanding the ground of a priori knowledge and its relation to empirical knowledge and questions the turn the tradition has taken since Vienna. (shrink)

Semantic externalism is the view that the meanings of referring terms, and the contents of beliefs that are expressed by those terms, are not fully determined by factors internal to the speaker but are instead bound up with the environment. The debate about semantic externalism is one of the most important but difficult topics in philosophy of mind and language, and has consequences for our understanding of the role of social institutions and the physical environment in constituting (...) language and the mind. In this long-needed book, Jesper Kallestrup provides an invaluable map of the problem. Beginning with a thorough introduction to the theories of descriptivism and referentialism and the work of Frege and Kripke, Kallestrup moves on to analyse Putnam’s Twin Earth argument, Burge’s arthritis argument and Davidson’s Swampman argument. He also discusses how semantic externalism is at the heart of important topics such as indexical thoughts, epistemological skepticism, self-knowledge, and mental causation. Including chapter summaries, a glossary of terms, and an annotated guide to further reading, _Semantic Externalism_ an ideal guide for students studying philosophy of language and philosophy of mind. (shrink)

This book presents modern logic as the formalization of reasoning that needs and deserves a semantic foundation. Chapters on propositional logic; parsing propositions; and meaning, truth and reference give the reader a basis for establishing criteria that can be used to judge formalizations of ordinary language arguments. Over 120 worked examples illustrate the scope and limitations of modern logic, as analyzed in chapters on identity, quantifiers, descriptive names, and functions. The chapter on second-order logic shows how different conceptions of (...) predicates and propositions do not lead to a common basis for quantification over predicates, as they do for quantification over things. Notable for its clarity of presentation and supplemented by many exercises, this volume will be invaluable for philosophers, linguists, mathematicians, and computer scientists who wish to better understand the tools they use in formal reasoning. (shrink)

This major publication is a history of the semantic tradition in philosophy from the early nineteenth century through its incarnation in the work of the Vienna Circle, the group of logical positivists that emerged in the years 1925–1935 in Vienna who were characterised by a strong commitment to empiricism, a high regard for science, and a conviction that modern logic is the primary tool of analytic philosophy. In the first part of the book, Alberto Coffa traces the roots of (...) logical positivism in a semantic tradition that arose in opposition to Kant's theory that a priori knowledge is based on pure intuition and the constitutive powers of the mind. In Part II, Coffa chronicles the development of this tradition by members and associates of the Vienna Circle. Much of Coffa's analysis draws on the unpublished notes and correspondence of many philosophers. The book, however, is not merely a history of the semantic tradition from Kant 'to the Vienna Station'. Coffa also critically reassesses the role of semantic notions in understanding the ground of a priori knowledge and its relation to empirical knowledge and questions the turn the tradition has taken since Vienna. (shrink)

Evaluating the usefulness of Davidson's semantics to computational understanding of language requires an examination of the role of a theory of truth in characterizing sentence meaning and logical form, and in particular of the connection between meaning and belief. The suggested conclusion is that the relevance of Davidson's semantics for computational semantics lies not so much in its methods and particular proposals of logical form as in its general orientation towards "desubstantializing" meaning.

Evaluating the usefulness of Davidson's semantics to computational understanding of language requires an examination of the role of a theory of truth in characterizing sentence meaning and logical form, and in particular of the connection between meaning and belief. The suggested conclusion is that the relevance of Davidson's semantics for computational semantics lies not so much in its methods and particular proposals of logical form as in its general orientation towards "desubstantializing" meaning.

The contemporary popularity of semantic externalism has arisen from so-called Twin Earth thought experiments which suggest that the representational content of a natural kind term cannot be wholly determined by processes within a speaker's body. Such arguments depend on the intuition that the extensions of natural kind terms cannot have changed as the result of the scientific investigation of natural kinds' constitutions. I demonstrate that this externalist intuition depends on an assumption about the mentality of isomorphic doppelgangers which has (...) never been questioned but which is nonetheless arguably false. I develop an alternative view of the instantiation of mind which entails a revision of our understanding of the constitution of environmental objects. The picture seems to be fully coherent despite its oddity and I can find no good reason to reject it. The conclusion must be that the case for semantic externalism is thus less compelling than is often supposed. (shrink)

I explore a strand of reception history that follows Rudolf Carnap’s shift from a purely syntactical analysis of constructed languages to his conception of pure semantics. My exploration focuses on Gustav Bergmann’s and Everett Hall’s interpretation of pure semantics, their understanding of what constitutes a ’formal’ investigation of language, and their arguments concerning the relationship between expressions and their extra-linguistic referents. I argue that Bergmann and Hall strongly misread Carnap’s semantic project and, subsequently, their misunderstanding is passed down (...) through colleagues and students. (shrink)