The purpose of this thesis is to attempt to answer the question why chatbot computer programs are often not very good at communicating in human natural language. It is argued that one possible reason why chatbots are often not good conversationalists is because they model communication in terms of only encoding and decoding processes. Human communication, however, involves making inferences about the mental states of others. Chapter one begins by exploring a popular theory about how communication works called the code model of communication. The code model describes human communication as having to do with speakers encoding thoughts into utterances and listeners decoding utterances to recover a representation of the thought that the speaker wanted to communicate. A variation on the code model is also explained; this is referred to as the information-theoretic code model. Two arguments against the code model will then be presented. Finally, an alternative to the code model is considered, called the ostensive-inferential model of communication. Chapter two begins with an explanation of how chatbots work. Chatbots are made up of several different components. The language model component provides chatbots with the ability to produce and interpret utterances. Next, an explanation of how language models work, and how chatbots can represent the meanings of words is provided. The chapter concludes by documenting the fact that chatbots communicate using only encoding and decoding processes – that is, that chatbots communicate within the paradigm of the code model of communication. Chapter three explains how the fact that chatbots communicate using only encoding and decoding processes can help explain why chatbots often cannot communicate effectively in human natural language. The poor conversational abilities of chatbots are a result of the fact that chatbots only access linguistic context, whereas listeners need access to non-linguistic context to be able to grasp utterance meaning. The question of whether chatbots are able to make inferences about non-linguistic properties of context at all is also considered. It is argued that they cannot, precisely because the neural language models that they rely upon for their linguistic competence are natural codes that merely associate percepts with output behaviours using encoding and decoding processes. / Thesis / Master of Philosophy (MA) / The purpose of this thesis is to explore the question why chatbot computer programs are often not very good at communicating in human natural language. It is argued that one possible reason why chatbots are often not good conversationalists is because they model communication in terms of only encoding and decoding processes. Human communication, however, involves making inferences about the mental states of others.

No description available.

This thesis calls for a reconfiguration of hate speech as a primarily normative phenomenon. All hate speech strives to weaken the social-moral normative status of its targets and in doing, justifies violence against its target. In light of this normative function, the harm of hate speech is reconsidered. Against traditional defenders of hate speech regulation, I claim that individual and collective harm is a highly likely, but not a necessary consequence of hate speech, while intrinsic harm and reckless risk necessarily follow from hate speech’s normative capacity. In light of the normative origin of such harms, a societal response with normative clout is required. However, while individual responses are insufficient to block the normativity of hate speech, I suggest that the legal system is characteristically well-suited to do so. / Thesis / Master of Arts (MA)

The problem of induction is most famously discussed by David Hume, though he himself has never resorted to the notion of “induction.” As part of Hume’s overall project of identifying the logical fallacy of causal relations, the problem of induction is identified as the problem of “the unobservable matters of fact.“ Hume argues that, as empirical beings, we can believe in two types of knowledge with confidence: relations of ideas and observable matters of facts. We can be certain about the relations of ideas, since by analyzing the relation itself we can come to necessarily true conclusions about such relations. On the other hand, we do not need to question observable matters of fact, since such are the content of our experience, and empirical matters such as “tables” and “chairs” are not subject to truth or falsity: it would be quite meaningless to argue that “it is true that table.” Though our perceptive experience easily invites another interesting philosophical discussion - most famously discusses by Descartes - that is, the external world scepticism, that is not something that this paper would address. Hume’s point, in the end, is to show that when we infer any unobservable matters of fact - that is, inductive inferences - from observable matters of facts, we are committing to a logical fallacy where the true premises of the inference does not guarantee the truth of the conclusion.

The origin of this thesis can be traced back to a deceptively simple question that struck me when reading Hegel for the first time: What, if anything, can be made of human freedom when we live in a world that has a profound impact on who we are and what we do? Unhappy with the way existentialist reactions to Hegel characterized freedom as our ability to step out of our world and determine our identity through our own decisions and will, but nevertheless inspired by Heidegger’s depiction of human agents as always already in the world, this thesis answers the aforementioned question by turning the existentialist conception of freedom on its head: that is, instead of characterizing freedom as detached decisionism, I argue that freedom is a function of our ability to recognize and respond to the disparate demands our world places upon us. Specifically, and unlike Heidegger’s existentialist interpreters, I read Heidegger’s account of authenticity as a case of engaged-agency in which we clarify the possibilities others make available and then act accordingly. There is, however, a certain limitation to this interpretation of human agency: namely, that treating freedom as an active response to the wants and needs of others binds the agent to possibilities present in her current situation and therefore fails to capture the kind of freedom we associate with cultural transformation or artistic creativity. Hence, this thesis addresses a second set of questions: What conditions make historical change possible? And how is it that artists are able to alter the world? In response to the first query, I turn to Heidegger’s claim that we are in truth and in untruth as well as his discussion of Gelassenheit to argue that the play between the possibilities present in a particular culture and those that are excluded by it along with a release from our present activities create the conditions for cultural transformation. In reply to the second question, I examine Heidegger’s account of the happening of truth and show how thinkers and artists are able to reveal the possibilities concealed in their culture through the creative use of language. Finally, I contend that the freedom associated with cultural transformation and artistic creativity is also a form of responsibility insofar as the success of a given transformation depends on others recognizing that transformation as valuable and thus worthy of their support.

This thesis presents an account of the nature of structured propositions and addresses a series of questions that arise from that proposal. Chapter 1 presents the account and explains how it meets standard objections to such views. Chapter 2 responds to the objection that this version of propositionalism is really a form of sententialism by arguing for the distinct advantages of the propositionalist view. Chapter 3 argues against a closely related view of propositions by way of general principles about how to construct such theories. Chapter 4 illustrates how a theory of propositions of the sort proposed can be defended against a recent argument that propositions need not play a central role in linguistic theory.

Is truth valuable? In addressing this question, one must parse it into questions that are more manageable. Is the property of truth only instrumentally valuable, or is it both instrumentally valuable and noninstrumentally valuable? Is the normativity of the concept of truth an intrinsic or extrinsic property of the concept? In addressing the first of these questions, I show that certain arguments are flawed, arguments that purport to show that truth is not valuable in any kind of way. After establishing that it is reasonable to think that the property of truth is valuable, I show how inflationists and deflationists can agree that the property of truth is noninstrumentally valuable. In addressing the second question, I rely on the distinction between semantics and pragmatics and the resources of moral semantics to claim that the normativity of the concept of truth is an extrinsic feature of the concept. I conclude that the property of truth is both instrumentally and noninstrumentally valuable and that the normativity associated with the concept of truth is an extrinsic property of the concept. In doing so, I suggest that beginning with an investigation about the value and normativity of truth has important ramifications for theories of truth in general.

Introduction: I lay out the broad contours of my thesis: a defence of mathematical nominalism, and nominalism more generally. I discuss the possibility of metaphysics, and the relationship of nominalism to naturalism and pragmatism. Chapter 2: I delineate an account of abstractness. I then provide counter-arguments to claims that mathematical objects make a di erence to the concrete world, and claim that mathematical objects are abstract in the sense delineated. Chapter 3: I argue that the epistemological problem with abstract objects is not best understood as an incompatibility with a causal theory of knowledge, or as an inability to explain the reliability of our mathematical beliefs, but resides in the epistemic luck that would infect any belief about abstract objects. To this end, I develop an account of epistemic luck that can account for cases of belief in necessary truths and apply it to the mathematical case. Chapter 4: I consider objections, based on (meta)metaphysical considerations and linguistic data, to the view that the existential quantifier expresses existence. I argue that these considerations can be accommodated by an existentially committing quantifier when the pragmatics of quantified sentences are properly understood. I develop a semi-formal framework within which we can define a notion of nominalistic adequacy. I show how our notion of nominalistic adequacy can show why it is legitimate for the nominalist to make use of platonistic “assumptions” in inference-making. Chapter 5: I turn to the application of mathematics in science, including explanatory applications, and its relation to a number of indispensability arguments. I consider also issues of realism and anti-realism, and their relation to these arguments. I argue that abstraction away from pragmatic considerations has acted to skew the debate, and has obscured possibilities for a nominalistic understanding of mathematical practices. I end by explaining the notion of a pragmatic meta-vocabulary, and argue that this notion can be used to carve out a new way of locating our ontological commitments. Chapter 6: I show how the apparatus developed in earlier chapters can be utilised to roll out the nominalist project to other domains of discourse. In particular, I consider propositions and types. I claim that a unified account of nominalism across these domains is available. Conclusion: I recapitulate the claims of my thesis. I suggest that the goal of mathematical enquiry is not descriptive knowledge, but understanding.

This thesis defends Semantic Pluralism, the view that sentences express sets of propositions in context. It puts forward two arguments against Contextualism, the main opposing view, on which each sentence expresses exactly one proposition in context. It spells out two versions of Pluralism: Flexible Pluralism, which takes most sentences to be context-sensitive, and Strong Pluralism, which denies that context-sensitivity is widespread. And it defends Flexible Pluralism and Strong Pluralism from a number of objections.

In contemporary formal theory of truth, model-theoretic and non-classical approaches have been dominant. I rather pursue the so-called classical axiomatic approaches toward truth and my dissertation begins by arguing for the classical axiomatic approach and against the others. The classical axiomatic approach inevitably leads to abandonment of the nave conception of truth and revision of the basic principles of truth derived from that nave conception such as the full T-schema. In the absence of the general guiding principles based on that nave conception, we need to conduct tedious but down-to-earth eld works' of various theories of truth by examining and comparing them from various points of view in searching for satisfactory theories of truth. As such attempt, I raise two new criteria for comparison of truth theories, make a proof-theoretic study of them in connection to the foundation of mathematics.