We provide a new text corpus from the social medium Telegram, which is rich in indirect forms of divisive speech. We scraped all messages from one channel of supporters of Donald Trump, covering a large part of his presidency from late 2016 until January 2021. The discussion among the group members over this long time period includes the spread of disinformation, disparaging of out-group members, and other forms of offensive speech. To encourage research into such practices of poisoning public (...) political discourse, we added automatic annotations of offensive language to all messages. We further added manual annotations of harmful language to a portion of the posts in order to enable the analysis of more implicit forms of online harm. (shrink)

Harmful and dangerous language is frequent in social media, in particular in spaces which are considered anonymous and/or allow free participation. In this paper, we analyse the language in a Telegram channel populated by followers of Donald Trump, in order to identify the ways in which harmful language is used to create a specific narrative in a group of mostly like-minded discussants. Our research has several aims. First, we create an extended taxonomy of potentially harmful language (...) that includes not only hate speech and direct insults, but also more indirect ways of poisoning online discourse, such as divisive speech and the glorification of violence. We apply this taxonomy to a large portion of the corpus. Our data gives empirical evidence for harmful speech such as in/out-group divisive language and the use of codes within certain communities which have not often been investigated before. Second, we compare our manual annotations to several automatic methods of classifying hate speech and offensive language, namely list based and machine learning based approaches. We find that the Telegram data set still poses particular challenges for these automatic methods. Finally, we argue for the value of studying such naturally occurring, coherent data sets for research on online harm and how to address it in linguistics and philosophy. (shrink)

In her analysis of pejoratives, Eva Picardi rejects a too sharp separation between descriptive and expressive content. I reconstruct some of her arguments, endorsing Eva’s criticism of Williamson’s analysis of Dummett and developing a suggestion by Manuel Garcia Carpintero on a speech act analysis of pejoratives. Eva’s main concern is accounting for our instinctive refusal to endorse an assertion containing pejoratives because it suggests a picture of reality we do not share. Her stance might be further developed claiming that uses (...) of pejoratives not only suggest, but also promote a wrong picture of reality. Our refusal to endorse implies rejecting not only a wrong picture of reality but also a call for participation to what that picture promotes. (shrink)

In this paper, I highlight some important implications of a non-individualistic account of derogatory words. I do so by critically examining an intriguing claim of Jennifer Hornsby‘s: that derogatory words – words that, as she puts it, ―apply to people, and that are commonly understood to convey hatred and contempt‖ – are useless for us. In their stead, she maintains, we employ neutral counterparts: words ―that apply to the same people, but whose uses do not convey these things. I argue (...) that Hornsby‘s distinctions – between derogatory words and neutral counterparts, and them (speakers who have use for the former) and us (who do not) – is not sustainable. I begin by considering examples that suggest that some of the words that some of us have use for are indeed derogatory. I then offer reasons for thinking that words that would presumably be identified as acceptable counterparts to derogatory words are not, in general, neutral. (shrink)

The use of language in hate speech is understandably offensive. Though words do not kill, they convey an alarming message that can harm the victim. To understand how words can harm, it is necessary to understand the nature of the meaning of pejoratives or slurs that are used in hate speech. Pejoratives are undeniably offensive. However, they are puzzling as they can be used in two directions, namely, the offensive power preservation and the offensive power (...) destruction. This paper proposes that the direct reference theory of pejoratives can solve the puzzle. A characterization of pejoratives is that it has the property of immediacy. They refer directly to the object of speech. Grounding on a shared context, any descriptions are unnecessary for understanding the offensive message of pejoratives. In this sense, pejoratives have indexical content as it is context-sensitive. The kind of indexical content that pertains to pejoratives is action-oriented. However, its object of reference is empty. In discussing pejoratives that are used in hate speech, some examples of Thai slurs are shown. (shrink)

I argue that the offense generation pattern of slurring terms parallels that of impoliteness behaviors, and is best explained by appeal to similar purely pragmatic mechanisms. In choosing to use a slurring term rather than its neutral counterpart, the speaker signals that she endorses the term. Such an endorsement warrants offense, and consequently slurs generate offense whenever a speaker's use demonstrates a contrastive preference for the slurring term. Since this explanation comes at low theoretical cost and imposes few constraints on (...) an account of the semantics of slurs, this suggests that we should not require semantic accounts to provide an independent explanation of the offense profile. (shrink)

Slurs such as spic, slut, wetback, and whore are linguistic expressions that are primarily understood to derogate certain group members on the basis of their descriptive attributes and expressions of this kind have been considered to pack some of the nastiest punches natural language affords. Although prior scholarship on slurs has uncovered several important facts concerning their meaning and use –including that slurs are potentially offensive, are felicitously applied towards some targets yet not others, and are often flexibly (...) used not only derogatorily to convey offense towards out-group members but also non-derogatorily to convey affiliation with in-group members– the literature remains largely focused on slurs that typically target African Americans, male homosexuals, and sexually active females. Since no account of slurs that typically target Hispanics or Mexican-Americans has so far been proposed, here I offer the first systematic and empirically informed analysis of these that accounts for both their derogatory and appropriative use. Importantly, this article reviews over a dozen Spanish stereotypes and slurs and explains how the descriptive attributes involved in a stereotype associated with a slur can contribute to the predication of certain content in the application of that slur toward its target in context. This article further explains how the psychological effects of stereotype threat and stereotype lift can be initiated through the application of a relevant slur towards its target in context as well. -/- ----- -/- Las expresiones peyorativas tales como spic (‘spic’), slut (‘zorra’), wetback (‘espalda mojada’) y whore (‘puta’) son expresiones lingüísticas que se entienden principalmente para minusvalorar ciertos miembros de un grupo sobre la base de sus atributos descriptivos (como la raza o el sexo). Se ha considerado que las expresiones de este tipo conllevan algunos de los puñetazos más desagradables que el lenguaje natural puede proporcionar. Aunque la literatura especializada sobre expresiones peyorativas ha descubierto varios hechos importantes en cuanto a significado y uso –entre los que se incluyen que tales expresiones son potencialmente ofensivas, apuntan efectivamente hacia unos objetivos pero no hacia otros, y con frecuencia se utilizan con flexibilidad no sólo despectivamente para ofender a miembros por fuera de un grupo, sino que también de forma no despectiva para afiliar con miembros dentro de un mismo grupo–, tal literatura sigue centrada en gran medida en las expresiones peyorativas que típicamente apuntan contra los afroamericanos (nigger ‘negro’), los homosexuales varones (fagot ‘maricón’), y las mujeres sexualmente activas (slut ‘zorra’). En tanto que no se ha propuesto al momento dar cuenta de expresiones peyorativas dirigidas contra hispanos o mexicano-americanos, en este trabajo se ofrece el primer análisis sistemático y empíricamente informado de tales expresiones, tanto en sus usos despectivos y de apropiación. Es importante destacar que en este artículo se revisan más de una docena de estereotipos y expresiones peyorativas en español, además de explicar cómo los atributos descriptivos que participan de un estereotipo asociado con una difamación pueden contribuir a la predicación de determinados contenidos en la aplicación de esa expresión hacia su objetivo en contexto. Asimismo, en este artículo se explica cómo comienzan los efectos psicológicos de la amenaza estereotipada y el realce estereotipado cuando se emplea una expresión peyorativa relevante contra un objetivo en contexto. (shrink)

Text-guided image generation models can be prompted to generate images using nonce words adversarially designed to robustly evoke specific visual concepts. Two approaches for such generation are introduced: macaronic prompting, which involves designing cryptic hybrid words by concatenating subword units from different languages; and evocative prompting, which involves designing nonce words whose broad morphological features are similar enough to that of existing words to trigger robust visual associations. The two methods can also be combined to generate images associated with more (...) specific visual concepts. The implications of these techniques for the circumvention of existing approaches to content moderation, and particularly the generation of offensive or harmful images, are discussed. (shrink)

In this paper I adopt Kaplan’s framework for distinguishing between descriptive and expressive content. Racial slurs are an especially difficult challenge for truth-conditional semantics because of their projection behaviors. That is to say, the offensive content of slurs “scopes out” of logical operators. I argue that racial slurs express contempt and lack descriptive content, so that many sentences containing slurs are not truth apt. My theory accounts for the intuition of the ordinary speaker who refuses to assent to the (...) truth of a sentence containing a slur, but accepts the same statement made using a neutral counterpart of that slur. Weaknesses of rival theories are briefly discussed. (shrink)

In several recent contributions to the growing literature on slurs, Hedger draws upon Kaplan’s distinction between descriptive and expressive content to argue that slurs are expressions with purely expressive content. The distinction between descriptive and expressive content and the view that slurs are expressions with purely expressive content has been widely acknowledged in prior work, and Hedger aims to contribute to this tradition of scholarship by offering novel arguments in support of his ‘‘pure expressivist’’ account of slurs. But the account (...) that PE offers is explanatorily inadequate, resting on suspect a priori intuitions which also commit one to denying many basic facts about slurs, such as that slurs largely display systematic differential application and that slurs can be used non-offensively between in-group speakers. In this article I provide clear reasons for rejecting PE, arguing particularly against Hedger as one of PE’s most explicit and recent proponents. In showing that PE is inadequate in at least 11 ways, I argue in favor of a mixed or hybrid approach. (shrink)

Slurring expressions display puzzling behaviour when embedded, such as under negation and in attitude and speech reports. They frequently appear to retain their characteristic qualities, like offensiveness and propensity to derogate. Yet it is sometimes possible to understand them as lacking these qualities. A theory of slurring expressions should explain this variability. We develop an explanation that deploys the linguistic notion of focus. Our proposal is that a speaker can conversationally implicate metalinguistic claims about the aptness of a focused slurring (...) expression. This explanation of variability relies on independently motivated mechanisms and is compatible with any theory of slurring expressions. (shrink)

The aim of natural language ontology is to uncover the ontological categories and structures that are implicit in the use of natural language, that is, that a speaker accepts when using a language. This article aims to clarify what exactly the subject matter of natural language ontology is, what sorts of linguistic data it should take into account, how natural language ontology relates to other branches of metaphysics, in what ways natural language ontology is (...) important, and what may be distinctive of the ontological categories and structures reflected in natural language. (shrink)

Recently, an increasing number of scholars have been showing interest in old-fashioned racism again. While recent studies on old-fashioned racism apparently increase our knowledge of this psychological theory of racism, the studies actually shed light on a different type of racism, namely offensive racism. The aim of this text is to argue that psychological theories of racism, like old-fashioned racism and modern racism, depend on societies’ social norms. I will show that questionnaires are highly sensitive to social norms, and (...) if there is change in social norms, the original measurement object alters. The theory of old-fashioned racism implicitly assumes agents that follow (or conform to) the social norm to behave prejudiced. Today, however, this social norm does not exist in Western societies. If agents express the same prejudices today, they indicate their willingness to breach the social norm to behave unprejudiced. Thereby, classic old-fashioned racism measures reflect a new kind of racism today. (shrink)

van Els, Paul. "Tegen aanvalsoorlogen" (Against Offensive Warfare). Dutch translation of Mozi chapters 17, 18, and 19. Het Trage Vuur 40 (December 2007): 8–19.

The semantics of racial slurs has recently become a locus of debate amongst philosophers. While everyone agrees that slurs are offensive, there is disagreement about the linguistic mechanism responsible for this offensiveness. This paper places the debate about racial slurs into the context of a larger issue concerning the interface between semantics and pragmatics, and argues that even on minimalist assumptions, the offensiveness of slur words is more plausibly due to their semantic content rather than any pragmatic mechanism. Finally, (...) I note that slurs make a good test case for expanding our semantic theories beyond the truth conditional tradition of Frege, which will be necessary in order to broaden the types of expressions handled by semantic theories. (shrink)

Since the Supreme Court upheld the partial birth abortion ban in 2007, more U.S. abortion providers have begun performing intraamniotic digoxin injections prior to uterine dilation and evacuations. These injections can cause medical harm to abortion patients. Our objective is to perform an in-depth bioethical analysis of this procedure, which is performed mainly for the provider’s legal benefit despite potential medical consequences for the patient.

Current work in language models (LMs) helps us speed up or even skip thinking by accelerating and automating cognitive work. But can LMs help us with critical thinking -- thinking in deeper, more reflective ways which challenge assumptions, clarify ideas, and engineer new concepts? We treat philosophy as a case study in critical thinking, and interview 21 professional philosophers about how they engage in critical thinking and on their experiences with LMs. We find that philosophers do not find LMs (...) to be useful because they lack a sense of selfhood (memory, beliefs, consistency) and initiative (curiosity, proactivity). We propose the selfhood-initiative model for critical thinking tools to characterize this gap. Using the model, we formulate three roles LMs could play as critical thinking tools: the Interlocutor, the Monitor, and the Respondent. We hope that our work inspires LM researchers to further develop LMs as critical thinking tools and philosophers and other 'critical thinkers' to imagine intellectually substantive uses of LMs. (shrink)

A comprehensive introduction to the Language of Though Hypothesis (LOTH) accessible to general audiences. LOTH is an empirical thesis about thought and thinking. For their explication, it postulates a physically realized system of representations that have a combinatorial syntax (and semantics) such that operations on representations are causally sensitive only to the syntactic properties of representations. According to LOTH, thought is, roughly, the tokening of a representation that has a syntactic (constituent) structure with an appropriate semantics. Thinking thus consists (...) in syntactic operations defined over representations. Most of the arguments for LOTH derive their strength from their ability to explain certain empirical phenomena like productivity, systematicity of thought and thinking. (shrink)

Chinese Chan or Zen Buddhism is renowned for its improvisational, atypical, and perplexing use of words. In particular, the tradition’s encounter dialogues, which took place between Chan masters and their interlocutors, abound in puzzling, astonishing, and paradoxical ways of speaking. In this chapter, we are concerned with Chan’s use of paradoxical language. In philosophical parlance, a linguistic paradox comprises the confluence of opposite or incongruent concepts in a way that runs counter to our common sense and ordinary rational thinking. (...) One naturally wonders about Chan masters’ rationales for their use of paradox. There are also concerns about whether the use violates the logical principle of noncontradiction to the effect that nothing can be both P and not-P all over in the same way at the same time. Chan became a viable Chinese Buddhist tradition during the Tang dynasty (618−907) and continued to develop for several centuries. The tradition had produced a huge literature; consequently, our investigation of its use of paradox cannot but be limited and selective. In the second section, I first sketch key ideas of Chan that are pertinent to our investigation and then examine the use of paradox in the sermons associated with certain Tang masters of the southern Chan. In the third section, I analyze the presence of paradoxical language in post-Tang encounter dialogues. The fourth section concludes. (shrink)

The essential role of language in rational cognition is analysed. The approach is functional: only the results of the connection between language, reality, and thinking are considered. Scientific language is analysed as an extension and improvement of everyday language. The analysis gives a uniform view of language and rational cognition. The consequences for the nature of ontology, truth, logic, thinking, scientific theories, and mathematics are derived.

Along with the promulgation of 2008 Law on Anti-Corruption, thanking to the comprehensive solutions and determinations of the whole governmental apparatus, there are signals of a positive change in the fight against corruption in Vietnam’s market economy. However, compared to other countries around the world, the corruption in Vietnam is still a national problem. The number of corruption cases may decline, but the scale and severity has been increasing. Many cases has involved high-ranking officials in the government with more than (...) 20 general officers in the armed forces to be sentenced. On the basis of analysing the current situation of corruption in recent years, the author hereby recommends some synchronous solutions to improve the effectiveness of anti-corruption. (shrink)

Some language encourages essentialist thinking. While philosophers have largely focused on generics and essentialism, I argue that nouns as a category are poised to refer to kinds and to promote representational essentializing. Our psychological propensity to essentialize when nouns are used reveals a limitation for anti-essentialist ameliorative projects. Even ameliorated nouns can continue to underpin essentialist thinking. I conclude by arguing that representational essentialism does not doom anti-essentialist ameliorative projects. Rather it reveals that would-be ameliorators ought to attend to (...) the propensities for our representational devices to essentialize and to the complex relationship between essentialism and prejudice. (shrink)

Fodor and Pylyshyn's critique of connectionism has posed a challenge to connectionists: Adequately explain such nomological regularities as systematicity and productivity without postulating a "language of thought" (LOT). Some connectionists like Smolensky took the challenge very seriously, and attempted to meet it by developing models that were supposed to be non-classical. At the core of these attempts lies the claim that connectionist models can provide a representational system with a combinatorial syntax and processes sensitive to syntactic structure. They are (...) not implementation models because, it is claimed, the way they obtain syntax and structure sensitivity is not "concatenative," hence "radically different" from the way classicists handle them. In this paper, I offer an analysis of what it is to physically satisfy/realize a formal system. In this context, I examine the minimal truth-conditions of LOT Hypothesis. From my analysis it will follow that concatenative realization of formal systems is irrelevant to LOTH since the very notion of LOT is indifferent to such an implementation level issue as concatenation. I will conclude that to the extent to which they can explain the law-like cognitive regularities, a certain class of connectionist models proposed as radical alternatives to the classical LOT paradigm will in fact turn out to be LOT models, even though new and potentially very exciting ones. (shrink)

Scholarly opinions vary on what language is, how it evolved, and from where or what it evolved. Long considered uniquely human, today scholars argue for evolutionary continuity between human language and animal communication systems. But while it is generally recognized that language is an evolving communication system, scholars continue to debate from which species language evolved, and what behavioral and cognitive features are the precursors to human language. To understand the nature of protolanguage, some look (...) for homologs in gene functionality, brain areas, or anatomical structures such as the supralaryngeal vocal tract; others point toward primates, their gestural, vocal, multimodal, and in later evolving hominins also their pantomimic communication systems; and still others draw parallels between the musicality that characterizes language and the pitch found in the numerous sounds produced by animals. Accordingly, protolanguage theories today are multiple and diverse, and protolanguages might have also been diverse. This special issue on Evolving (Proto)Language/s for Lingua bundles several of the protolanguage theories that were put forward at the sixth edition of the Ways to Protolanguage conference series, held at the Calouste Gulbenkian Foundation in Lisbon, in 2019. Not aimed at surveying all the different ways there are to conceptualize, study, and model protolanguage/s, this issue provides interested readers with good overviews on the role played in current theorizing on protolanguage/s by (paleo)anthropology, genetics, physiology, developmental, evolutionary, ecological, and pragmatic research lines. (shrink)

We discuss potential biorisks from large language models (LLMs). AI assistants based on LLMs such as ChatGPT have been shown to significantly reduce barriers to entry for actors wishing to synthesize dangerous, potentially novel pathogens and chemical weapons. The harms from deploying such bioagents could be further magnified by AI-assisted misinformation. We endorse several policy responses to these dangers, including prerelease evaluations of biomedical AIs by subject-matter experts, enhanced surveillance and lab screening procedures, restrictions on AI training data, and (...) access controls on biomedical AI. We conclude with a suggestion about future research directions in bioethics. (shrink)

3rd ed, 2021. A circumscription of the classical theory of computation building up from the Chomsky hierarchy. With the usual topics in formal language and automata theory.

Abstract The purpose of this paper is twofold: (i) we will argue that formal semantics might have faltered due to its failure in distinguishing between two fundamentally very different types of concepts, namely ontological concepts, that should be types in a strongly-typed ontology, and logical concepts, that are predicates corresponding to properties of, and relations between, objects of various ontological types; and (ii) we show that accounting for these differences amounts to a new formal semantics; one that integrates lexical and (...) compositional semantics in one coherent framework and one where formal semantics is embedded with a strongly typed ontology; an ontology that reflects our commonsense knowledge of the world and the way we talk about it in ordinary language. We will show how in such a framework a number of challenges in the semantics of natural language are adequately and systematically treated. (shrink)

Politics abounds with bad language: lying and bullshitting, grandstanding and virtue signaling, code words and dogwhistles, and more. But why is there so much bad language in politics? And what, if anything, can we do about it? In this paper I show how these two questions are connected. Politics is full of bad language because existing social and political institutions are structured in such a way that the production of bad language becomes rational. In principle, by (...) modifying these institutions we can reduce the prevalence of bad language. However, as I show, such practical efforts are fraught with difficulties. After first outlining an account of bad language (Section 1), I examine the rationality of three different types of bad language: inaccurate language, insincere language, and unclear language (Section 2). Next, I discuss the possibility of implementing institutional reforms to improve the quality of political discourse (Section 3). However, I then outline and discuss two serious complications for institutional reforms—namely, they create risk of abuse, and they could preclude instances of seemingly bad language that, in fact, are socially beneficial (Section 4). I conclude with some thoughts about how to pursue institutional reform in an appropriately circumspect manner (Section 5). (shrink)

According to ordinary language philosophy (OLP), philosophical problems can be solved by investigating ordinary language, often because the problems stem from its misuse. According to ideal language philosophy (ILP), on the other hand, philosophical problems exist because ordinary language is flawed and has to be improved or replaced by constructed languages that do not exhibit these flaws. OLP and ILP together make up linguistic philosophy, the view that philosophical problems are problems of language. Linguistic philosophy (...) is opposed to what may be called, for lack of a better word, ‘traditional philosophy’ (TP), the view that philosophical problems can be solved by discovering non-linguistic facts. In the following, OLP, ILP, and TP are taken to be methodologies, that is, frameworks in which to interpret and evaluate different philosophical methods (i.e., argumentative strategies). The two linguistic methodologies are discussed separately with TP as a foil, and then used to interpret the status of different philosophical methods. While each of the methods discussed here finds a plausible interpretation in each methodology, there are other arguments for and against linguistic philosophy in general, and for and against ILP and OLP in particular. As none of these arguments is decisive, I conclude with a superficial moral about peaceful co-existence. (shrink)

I show that there are good arguments and evidence to boot that support the language as an instrument of thought hypothesis. The underlying mechanisms of language, comprising of expressions structured hierarchically and recursively, provide a perspective (in the form of a conceptual structure) on the world, for it is only via language that certain perspectives are avail- able to us and to our thought processes. These mechanisms provide us with a uniquely human way of thinking and talking (...) about the world that is different to the sort of thinking we share with other animals. If the primary function of language were communication then one would expect that the underlying mechanisms of language will be structured in a way that favours successful communication. I show that not only is this not the case, but that the underlying mechanisms of language are in fact structured in a way to maximise computational efficiency, even if it means causing communicative problems. Moreover, I discuss evidence from comparative, neuropatho- logical, developmental, and neuroscientific evidence that supports the claim that language is an instrument of thought. (shrink)

[This is an edited version of a keynote talk at the conference on Neural Information Processing Systems (NeurIPS) on November 28, 2022, with some minor additions and subtractions.] -/- There has recently been widespread discussion of whether large language models might be sentient or conscious. Should we take this idea seriously? I will break down the strongest reasons for and against. Given mainstream assumptions in the science of consciousness, there are significant obstacles to consciousness in current models: for example, (...) their lack of recurrent processing, a global workspace, and unified agency. At the same time, it is quite possible that these obstacles will be overcome in the next decade or so. I conclude that while it is somewhat unlikely that current large language models are conscious, we should take seriously the possibility that successors to large language models may be conscious in the not-too-distant future. (shrink)

This paper analyzes debates on animal language in eighteenth-century German philosophy and science. Adopting a history of ideas approach, I explain how the study of animal language became tied to the investigation into the origin and development of language towards the end of the eighteenth century. I argue that for large parts of the eighteenth century, the question of the existence of animal languages was studied within the context of the philosophical question of whether animals possess reason. (...) In Germany, the debate concerning animal reason was influenced by Christian Wolff and was taken up by diverse thinkers such as Winkler, Meier, and Reimarus. I argue that in the second half of the eighteenth century the study of animal language became more loosely related to the question of whether animals possess reason: animal language was studied not only in light of the debate on animal reason but also because it sheds light on the nature of animals, on the differences and similarities between animals and humans, and on the origin and development of language. This systematic study of animal language coincided with the rise of linguistics, anthropology, and biology as independent sciences. (shrink)

The potential for Large Language Models (LLMs) to attain technological singularity—the point at which artificial intelligence (AI) surpasses human intellect and autonomously improves itself—is a critical concern in AI research. This paper explores the feasibility of current LLMs achieving singularity by examining the philosophical and practical requirements for such a development. We begin with a historical overview of AI and intelligence amplification, tracing the evolution of LLMs from their origins to state-of-the-art models. We then proposes a theoretical framework to (...) assess whether existing LLM technologies could satisfy the conditions for singularity, with a focus on Recursive Self-Improvement (RSI) and autonomous code generation. We integrate key component technologies, such as Reinforcement Learning from Human Feedback (RLHF) and Direct Preference Optimization (DPO), into our analysis, illustrating how these could enable LLMs to independently enhance their reasoning and problem-solving capabilities. By mapping out a potential singularity model lifecycle and examining the dynamics of exponential growth models, we elucidate the conditions under which LLMs might self-replicate and rapidly escalate their intelligence. We conclude with a discussion of the ethical and safety implications of such developments, underscoring the need for responsible and controlled advancement in AI research to mitigate existential risks. Our work aims to contribute to the ongoing dialogue on the future of AI and the critical importance of proactive measures to ensure its beneficial development. (shrink)

[This is an earlier (1997), much longer and more detailed version of my entry on LOTH in the _Stanford Encyclopedia of Philosophy_] The Language of Thought Hypothesis (LOTH) is an empirical thesis about thought and thinking. For their explication, it postulates a physically realized system of representations that have a combinatorial syntax (and semantics) such that operations on representations are causally sensitive only to the syntactic properties of representations. According to LOTH, thought is, roughly, the tokening of a representation (...) that has a syntactic (constituent) structure with an appropriate semantics. Thinking thus consists in syntactic operations defined over representations. Most of the arguments for LOTH derive their strength from their ability to explain certain empirical phenomena like productivity, systematicity of thought and thinking. (shrink)

The fact that Gilbert Ryle and J.L. Austin seem to disagree about the ordinary use of words such as ‘voluntary’, ‘involuntary’, ‘voluntarily’, and ‘involuntarily’ has been taken to cast doubt on the methods of ordinary language philosophy. As Benson Mates puts the worry, ‘if agreement about usage cannot be reached within so restricted a sample as the class of Oxford Professors of Philosophy, what are the prospects when the sample is enlarged?’ (Mates, Inquiry 1:161–171, 1958, p. 165). In this (...) chapter, we evaluate Mates’s criticism alongside Ryle’s and Austin’s specific claims about the ordinary use of these words, assessing these claims against actual examples of ordinary use drawn from the British National Corpus (BNC). Our evaluation consists in applying a combination of methods: first aggregating judgments about a large set of samples drawn from the corpus, and then using a clustering algorithm to uncover connections between different types of use. In applying these methods, we show where and to what extent Ryle’s and Austin’s accounts of the use of the target terms are accurate as well as where they miss important aspects of ordinary use, and we demonstrate the usefulness of this new combination of methods. At the heart of our approach is a commitment to the idea that systematically looking at actual uses of expressions is an essential component of any approach to ordinary language philosophy. (shrink)

Ludwig deals with the relations between language, thought, and rationality, and, especially, the role and status of assumptions about rationality in interpreting another’s speech and assigning contents to her psychological attitudes—her beliefs, desires, intentions, and so on. The chapter is organized around three questions: What is the relation between rationality and thought? What is the relation between rationality and language? What is the relation between thought and language? Ludwig argues that some large degree of rationality is required (...) for thought and consequently that same degree of rationality at least is required for language since language requires thought. Thought, however, does not require language. In answering questions and, Ludwig gives particular attention to Davidson’s arguments for the Principle of Charity, according to which it is constitutive of speakers that they are largely rational and largely right about the world, and to Davidson’s arguments for the thesis that without the power of speech we lack the power of thought. (shrink)

There are several linguistic phenomena that, when examined closely, give evidence that people speak through characters, much like authors of literary works do, in everyday discourse. However, most approaches in linguistics and in the philosophy of language leave little theoretical room for the appearance of characters in discourse. In particular, there is no linguistic criterion found to date, which can mark precisely what stretch of discourse within an utterance belongs to a character, and to which character. And yet, without (...) at least tentatively marking the division of labor between the different characters in an utterance, it is absolutely impossible to arrive at an acceptable interpretation of it. As an alternative, I propose to take character use seriously, as an essential feature of discourse in general, a feature speakers and listeners actively seek out in utterances. I offer a simple typology of actions in discourse that draws on this understanding, and demonstrate its usefulness for the analysis of a conversation transcript. (shrink)

The state has always authoritatively used criminal law to give effect to its policy of condemning acts either antisocial or unacceptable to the conscience of the law and society. The existence of criminal law is well justified on grounds of ‘social welfare’ or “reinforcement of those values most basic to proper social functioning”. This initiates or sustains the process of criminalization. The relativity of ‘social welfare’ makes law ‘dynamic’ as well as ‘varying’, vis-à-vis its ambit and scope. Current scholarship is (...) critical of what is referred to as the trend of overcriminalization or rapid increase in criminalizing of acts, as leading to ‘uncertainty’ in criminal law. The rationale for such a critique is that there are activities that need not be labeled as offences if they do not possess the potential to cause damage that criminal law seeks to protect. In light of the overcriminalization critique, this paper examines the criminalization of certain offences labeled as public welfare offences. (shrink)

This paper gives an outline of natural language ontology as a subdiscipline of both linguistics and philosophy. It argues that part of the constructional ontology reflected in natural language is in significant respects on a par with syntax (on the generative view).

A. J. Ayer’s Language, Truth, and Logic had been responsible for introducing the Vienna Circle’s ideas, developed within a Germanophone framework, to an Anglophone readership. Inevitably, this migration from one context to another resulted in the alteration of some of the concepts being transmitted. Such alterations have served to facilitate a number of false impressions of Logical Empiricism from which recent scholarship still tries to recover. In this paper, I will attempt to point to the ways in which LTL (...) has helped to foster the various mistaken stereotypes about Logical Empiricism which were combined into the received view. I will begin by examining Ayer’s all too brief presentation of an Anglocentric lineage for his ideas. This lineage, as we shall see, simply omits the major 19th century Germanophone influences on the rise of analytic philosophy. The Germanophone ideas he presents are selectively introduced into an Anglophone context, and directed towards various concerns that arose within that context. I will focus on the differences between Carnap’s version of the overcoming of metaphysics, and Ayer’s reconfiguration into what he calls the elimination of metaphysics. Having discussed the above, I will very briefly outline the consequences that Ayer’s radicalisation of the Vienna Circle’s doctrines had on the subsequent Anglophone reception of Logical Empiricism. (shrink)

I demonstrate that a "speech act" theory of meaning for imperatives is—contra a dominant position in philosophy and linguistics—theoretically desirable. A speech act-theoretic account of the meaning of an imperative !φ is characterized, broadly, by the following claims. -/- LINGUISTIC MEANING AS USE !φ’s meaning is a matter of the speech act an utterance of it conventionally functions to express—what a speaker conventionally uses it to do (its conventional discourse function, CDF). -/- IMPERATIVE USE AS PRACTICAL !φ's CDF is to (...) express a practical (non-representational) state of mind—one concerning an agent's preferences and plans, rather than her beliefs. -/- Opposed to speech act accounts is a preponderance of views which deny that a sentence's linguistic meaning is a matter of what speech act it is used to perform, or its CDF. On such accounts, meaning is, instead, a matter of "static" properties of the sentence—e.g., how it depicts the world as being (or, more neutrally, the properties of a model-theoretic object with which the semantic value of the sentence co-varies). On one version of a static account, an imperative 'shut the window!' might, for instance, depict the world as being such that the window must be shut. -/- Static accounts are traditionally motivated against speech act-theoretic accounts by appeal to supposedly irremediable explanatory deficiencies in the latter. Whatever a static account loses in saying (prima facie counterintuitively) that an imperative conventionally represents, or expresses a picture of the world, is said to be offset by its ability to explain a variety of phenomena for which speech act-theoretic accounts are said to lack good explanations (even, in many cases, the bare ability to offer something that might meet basic criteria on what a good explanation should be like). -/- I aim to turn the tables on static accounts. I do this by showing that speech act accounts are capable of giving explanations of phenomena which fans of static accounts have alleged them unable to give. Indeed, for a variety of absolutely fundamental phenomena having to do with the conventional meaning of imperatives (and other types of practical language), speech act accounts provide natural and theoretically satisfying explanations, where a representational account provides none. (shrink)

This paper presents a simple model to estimate the number of languages that existed throughout history, and considers philosophical and linguistic implications of the findings. The estimated number is 150,000 plus or minus 50,000. Because only few of those remain, and there is no reason to believe that that remainder is a statistically representative sample, we should be very cautious about universalistic claims based on existing linguistic variation.

The paper explores Hegel's theory of language, from the Subjective Spirit book of his Encyclopedia. Hegel distinguishes between linguistic signs, as arbitrary signifiers and words, which occur when the signs are filled with thought or meaning. Words have greater objectivity than signs. The words of the positive, empirical sciences are taken up into Hegelian Science (system), affording it greater objectivity, which it, reciprocally re-confers on its linguistic contents.

Philosophical enquiry stands to benefit from the inclusion of methods from the digital humanities to study language use. Empirical studies using the methods of the digital humanities have the potential to contribute to both conceptual analysis and intuition-based enquiry, two important approaches in contemporary philosophy. Empirical studies using the methods of the digital humanities can also provide valuable metaphilosophical insights into the nature of philosophical methods themselves. The use of methods from the digital humanities in philosophy should be expected (...) to follow a similar trajectory to another attempt to introduce empirical methods in philosophy. Experimental philosophers use empirical methods from the cognitive sciences to bear on philosophical questions. At first, we should expect the main contribution of methods from the digital humanities in philosophy to be at the metaphilosophical level. Then, given the right impetus and support, we are optimistic that methods from the digital humanities can make important and sustained contributions to first-order philosophical enquiry. This chapter gives two case studies of recent studies in which methods from the digital humanities are used to address metaphilosophical questions about the use of the word ‘intuition’ and the methods of philosophy and physics. It gives three examples of contemporary first-order philosophical debates to which methods fromthe digital humanities could make an important contribution in epistemology, metaphysics and philosophy of language. The chapter discusses some methodological challenges and limitations which are of particular importance when considering the application of methods from the digital humanities in philosophy, including concerns about the demographic representativeness of widely available corpora. (shrink)

Simon Blackburn objects that Wittgenstein's private language argument overlooks the possibility that a private linguist can equip himself with a criterion of correctness by confirming generalizations about the patterns in which his private sensations occur. Crispin Wright responds that appropriate generalizations would be too few to be interesting. But I show that Wright's calculations are upset by his failure to appreciate both the richness of the data and the range of theories that would be available to the private linguist.

Language, one of humanity’s major transformative innovations, is foundational for many cultural, artistic, scientific, and economic advancements, including the creation of artificial intelligence (AI). However, in the fight against climate change, the power of such innovation is constrained due to the boring language of climate science and science communication. In this essay, we encapsulated the situation and risks of boring language in communicating climate information to the public and countering climate denialism and disinformation. Based on the Serendipity-Mindsponge-3D (...) knowledge management framework, we recommend several strategies for climate scientists and science communicators to be more creative and make their communication more interesting, including collaboration with other cultural sectors (e.g., stand-up comedians, climate fiction, etc.) and AI. (shrink)

Recent advances in natural language processing have given rise to a new kind of AI architecture: the language agent. By repeatedly calling an LLM to perform a variety of cognitive tasks, language agents are able to function autonomously to pursue goals specified in natural language and stored in a human-readable format. Because of their architecture, language agents exhibit behavior that is predictable according to the laws of folk psychology: they function as though they have desires (...) and beliefs, and then make and update plans to pursue their desires given their beliefs. We argue that the rise of language agents significantly reduces the probability of an existential catastrophe due to loss of control over an AGI. This is because the probability of such an existential catastrophe is proportional to the difficulty of aligning AGI systems, and language agents significantly reduce that difficulty. In particular, language agents help to resolve three important issues related to aligning AIs: reward misspecification, goal misgeneralization, and uninterpretability. (shrink)

English is essential as an effective communication tool in both local and international contexts. In addition to being used in schools, it is also a teaching tool in colleges and universities. ESL (English as a Second Language) classes are now required in all educational institutions and can't be skipped. When learning a second language, anyone must be physically, mentally, and emotionally involved to communicate and understand what is being said. This systematic review employed qualitative documentary research and adopted (...) content analysis. Following a random selection of reliable and legitimate sources, purposive sampling was used to conduct a content analysis of the selected papers. For this systematic review, sources from EBSCO, Google Scholar, Scopus, Web of Science, and ScienceDirect were used. The findings reveal that English-language obstacles in this article include English learners' writing, speaking, listening, reading, translation, and interpretation skills. The conclusion focuses on English learning obstacles and their improvement through teaching methods and educational technology adoption. The recommendation is to look into quantitative and qualitative research for further study. (shrink)

This paper surveys some philosophical issues with the language surrounding mental illness, but is especially focused on pejoratives relating to mental illness. I argue that though 'crazy' and similar mental illness-based epithets (MI-epithets) are not best understood as slurs, they do function to isolate, exclude, and marginalize members of the targeted group in ways similar to the harmfulness of slurs more generally. While they do not generally express the hate/contempt characteristic of weaponized uses of slurs, MI-epithets perpetuate epistemic injustice (...) by portraying sufferers of mental illness as deserving minimal credibility. After outlining the ways in which these epithets can cause harm, I examine available legal and social remedies, and suggest that the best path going forward is to pursue a reclamation project rather than aiming to censure the use of MI-epithets. (shrink)