Jussi Haukioja

The argument known as the ‘McKinsey Recipe’ tries to establish the incompatibility of semantic externalism (about natural kind concepts in particular) and a priori self‐knowledge about thoughts and concepts by deriving from the conjunction of these theses an absurd conclusion, such as that we could know a priori that water exists. One reply to this argument is to distinguish two different readings of ‘natural kind concept’: (i) a concept which in fact denotes a natural kind, and (ii) a concept which aims to denote a natural kind. Paul Boghossian has argued, using a Dry Earth scenario, that this response fails, claiming that the externalist cannot make sense of a concept aiming, but failing, to denote a natural kind. In this paper I argue that Boghossian’s argument is flawed. Borrowing machinery from two‐dimensional semantics, using the notion of ‘considering a possible world as actual’, I claim that we can give a determinate answer to Boghossian’s question: which concept would ‘water’ express on Dry Earth?1.

All the experimental work on the reference of proper names during the last 20 years has utilised setups modelled after Kripke’s Gödel and Jonah cases. Doubts remain, however, about the viability of the setup. This paper reports the results from a new experiment, using a different and novel kind of setup. The novel setup is simultaneously used on proper names, artefact terms, and definitional terms, with the last two categories serving as controls. Our vignettes describe cases of potential reference swaps, where two referring expressions begin, due to a mix-up, to be used with the intention of referring to each others’ original referents. In otherwise analogous cases, it was found that a majority of Western subjects understand proper names to continue to refer to their original referents, whereas this was not the case for the two other classes of expression. The results indicate, first, that this experimental setup, unlike earlier ones, succeeds in distinguishing between different mechanisms of reference determination, and second, that proper names indeed do refer differently from the other classes of expression that were used.

Metasemantics studies the foundations of meaning, asking what makes it the case that certain words have the meanings that they do. But what makes metasemantic theories true? This question has been all but ignored in philosophy of language. In this book, we address this issue and argue that just as in metasemantics, both internalist and externalist answers are available for this foundational question. In the book, we introduce and defend _meta-internalism_, arguing that the foundations of reference and meaning are anchored in the individual dispositions and psychological states of language users, offering an alternative to meta-externalist views that would appeal to broader community-based or otherwise external factors. Meta-internalism is, moreover, compatible with semantic externalism as usually understood, and provides an _explanation_ of why externalism is true. Through a critical examination of prominent theories and thought experiments, the book explores fundamental issues like reference failure, conceptual engineering, and the metaphysical implications of reference, as well as the methodology of theories of reference.

Two recent experimental studies, by Shaun Nichols et al. and by Michael Devitt & Brian Porter claim to find evidence for the view that both causal-historical factors and descriptive factors play a role in determining the extensions of natural kind terms. Both studies use versions of a vignette featuring the fictional natural kind term “Catoblepas”. We conducted an experiment where we used vignettes and corresponding tasks that were otherwise fully analogous, but featured terms which are not natural kind terms. We found that such terms received the same patterns of response which Nichols et al. and Devitt & Porter took to justify their conclusions. We conclude that either the kind of view they arrive at applies more widely than just for natural kind terms or, as we think is more likely, the original results do not support such a two-factor view in the first place.

Since Kripke’s and Putnam’s work in the 1970s, most philosophers have assumed that our natural kind concepts are externally individuated. However, both psychologists and philosophers have questioned this assumption, partly on empirical grounds. There is some evidence of systematic variation in how subjects apply natural kind terms; it has been argued that this shows that natural kind concepts are not as universally shared, or temporally stable, as many philosophers have been assuming. Yet, it is not clear exactly what kind of variation in subjects’ application of natural kind terms would seriously cast doubt on the view that natural kind concepts are externally individuated. Here, we take a detailed look at this question, and argue that the existing studies do not call for dramatic revisions to the externalist mainstream, and conclude by exploring some possible new directions for the experimental study of natural kind terms and concepts.

Ever since Machery et al. first decided to test whether non-philosophers assign reference in accordance with the causal-historical account, the reference of proper names has been tested by means of setups modelled on Kripke’s Gödel and Jonah cases. Over the years, the use of these setups as a means to test theories of reference has attracted much criticism. However, previous follow-up studies have supposedly accounted for these criticisms, for the most part without changing the original outcome. We conducted experiments suggesting that participants’ responses in these setups never tracked what they are supposed to track. In our study, we tested the setup itself by using analogues of different setups modelled on Kripke’s Gödel and Jonah cases. Instead of proper names, our analogues featured terms for which we have independent reason to believe that the causal-historical account is not true. The analogues elicited large proportions of supposed causal-historical responses.

Recent experimental studies have claimed to find evidence for the view that natural kind terms such as “water” are ambiguous: that they have two extensions, one determined by superficial properties, the other by underlying essence. In an online experiment, we presented to 600 participants scenarios describing discoveries of novel samples that differ in deep structure from samples of a familiar kind but are superficially identical, such as a water-like substance that is not composed of H2O. We used three different types of question sets to probe whether the participants considered the sample as a member of the kind or not. Our results did not confirm the predictions of the ambiguity view. They were, rather, consistent with views that take underlying essences to be the sole criterion for membership in a natural kind.

This chapter understands internalism and externalism as supervenience theses, or rejections thereof. It focuses on different arguments for various kinds of externalist theses, rather than on arguments for internalism. It also reviews the central thought experiments often considered as giving strong support to externalist theses, paying close attention to how internal duplicates figure in the experiments. The chapter looks at methodological and meta‐philosophical aspects of the internalism/externalism debate, and discusses what makes a particular kind of semantic externalist claim true, when it is true. It outlines the central arguments for externalism by Saul Kripke, Hilary Putnam, and Tyler Burge. The chapter focuses on the consequences that a meta‐internalist view would have on the methodological questions raised, and suggests that a certain kind of meta‐internalist view can make good sense of the use of thought experiments in arguing for and against externalism about extension.

This paper will look at three closely interrelated questions about necessary a posteriori identities, in particular concerning natural kinds. First-ly, what is the semantic phenomenon responsible for a posteriori necessities in general, and theoretical identity statements concerning natural kinds in particular? I will argue that rigidity, as it is usually defined, cannot do the job for theoretical identity statements. Rather, a posteriori necessities are grounded in a semantic phenomenon that I have in earlier work called actuality-dependence. Secondly, what is the basis for this semantic phenomenon? In other words, what makes a given expression actuality-dependent, and thereby fit for appearing in a posteriori necessities? I will argue that actuality-dependence is grounded in our linguistic patterns and dispositions of application and interpretation. Thirdly and finally, what does this tell us about what essences are, and how we possess knowledge of them? I will claim that the view I am proposing gives direct support to a conferralist view of essences, at least when it comes to natural kinds. Moreover, the explanation of a posteriori necessities does not rely on independent essentialist premises; however, it does rely on essentialist beliefs or expectations on the part of ordinary speakers.

Some experimental studies have recently claimed to undermine semantic externalism about natural kind terms. However, it is unclear how philosophical accounts of reference can be experimentally tested. We present two externalistic adaptations of psychological placeholder essentialism, a strict externalist and a hybrid externalist view, which are experimentally testable. We examine Braisby, Franks, and Hampton's (1996) study which claims to undermine externalism, and argue that the study fails in its aims. We conducted two experiments, the results of which undermine internalism and the hybrid theory, and support strict externalism. Our conclusion is that lay speakers’ natural kind concepts involve a belief in an external category essence, which determines reference.

Ted Warfield has presented a new version of the Putnamian argument for the conclusion that we are not brains in a vat. This version is intended to avoid reliance on some questionable background assumptions which other versions have made. It seems that Warfield's argument fails, for reasons pointed out by Anthony Brueckner. However, in this paper I present a new version of the argument -- my version relies on assumptions no more objectionable than Warfield's, yet it is immune to Brueckner's objection

The so-called missing explanation argument, put forward by Mark Johnston in the late 80’s purported to show that our ordinary concepts of secondary qualities such as the colours cannot be response-dependent. A number of flaws were soon found in the argument. Partly in response to the criticism directed at the original argument, Johnston presented a new version in 1998. In this paper I show that the new version fails, too, for a simple reason: the kind of explanation which Johnston claims to be incompatible with a response-dependent account of the relevant concept is not an empirical explanation at all, but merely looks like one because of certain factors in Johnston’s stage- setting for the argument.

Some experimental studies have recently claimed to undermine semantic externalism about natural kind terms. However, it is unclear how philosophical accounts of reference can be experimentally tested. We present two externalistic adaptations of psychological placeholder essentialism, a strict externalist and a hybrid externalist view, which are experimentally testable. We examine Braisby’s et al. (1996) study which claims to undermine externalism, and argue that the study fails in its aims. We conducted two experiments, the results of which undermine internalism and the hybrid theory, and support strict externalism. Ourconclusion is that lay speakers’ natural kind concepts involve a belief in an external category essence, which determines reference

ABSTRACT Temporal externalism is the view that the meanings and extensions of linguistic expressions can be partly determined by contingent linguistic and/or conceptual developments that take place after the time of utterance. In this paper, I first clarify what it would take for temporal externalism to be true, relying on the notion of burden-shifting dispositions. I then go on to argue that existing thought experiments give us reason to expect that temporal externalism can be true of some natural kind terms, and present a new thought experiment suggesting it can also be true of some proper names.

Following the influential thought experiments by Hilary Putnam and others, philosophers of language have for the most part adopted semantic externalism concerning natural kind terms. In this article, we present results from three experiments on the reference of natural kind terms. Our results confirm some standard externalist assumptions, but are in conflict with others: Ordinary speakers take both appearance and underlying nature to be central in their categorization judgments. Moreover, our results indicate that speakers’ categorization judgments are gradual, and proportional to the degree of similarity between new samples and familiar, “standard” samples. These findings pose problems for traditional theories, both externalist and internalist.

The argument known as the 'McKinsey Recipe' tries to establish the incompatibility of semantic externalism (about natural kind concepts in particular) and _a priori _self- knowledge about thoughts and concepts by deriving from the conjunction of these theses an absurd conclusion, such as that we could know _a priori _that water exists. One reply to this argument is to distinguish two different readings of 'natural kind concept': (i) a concept which _in fact _denotes a natural kind, and (ii) a concept which _aims_ to denote a natural kind. Paul Boghossian has argued, using a _Dry Earth _scenario, that this response fails, claiming that the externalist cannot make sense of a concept aiming, but failing, to denote a natural kind. In this paper I argue that Boghossian's argument is flawed. Borrowing machinery from two-dimensional semantics, using the notion of 'considering a possible world as actual', I claim that we can give a determinate answer to Boghossian's question: which concept would 'water' express on Dry Earth?

Kripke argued, famously, that proper names are rigid designators. It is often assumed that some kind terms (most prominently natural kind terms) are rigid designators as well. This is thought to have significant theoretical consequences, such as the necessity of certain a posteriori identities involving natural kind terms. However, there is no agreement on what it is for a kind term to be rigid. In this paper I will first take a detailed look at the most common view: that rigid kind terms are those which designate the same kind in all possible worlds. This view has been subjected to much recent criticism. I will argue that, while the proponents of the view do seem to have good answers to some of the arguments presented against it, it fails because this notion of rigidity cannot deliver aposteriori necessities. Time permitting, I will also sketch an alternative view which seems far more promising.

When a volume deals with the work of a philosopher such as David Lewis who has worked on a broad range of questions, it is helpful to restrict the topics in some manner. The editors of this collection of essays have chosen Humean supervenience as the unifying theme. However, only one of the papers is directly concerned with supervenience. The rest are on subjects such as modal realism, time travel, endurance vs. perdurance, causation, conditionals, and physicalism about the mind. In Lewis’s philosophy, supervenience does, of course, play a key role in the treatment of all these issues.

Causal theories of reference for natural kind terms are widely agreed to play a central role in arguments for the claim that theoretical identity statements such as “Water is H2O” are necessary, if true. However, there is also fairly wide-spread agreement, due to the arguments of Nathan Salmon, that causal theories of reference do not alone establish such essentialism about natural kinds: an independent, non-trivial essentialist premise is also needed. In this paper I will question this latter agreement. I will argue that there is an independently attractive explanation of why such identity statements are metaphysically necessary, if true: an explanation which relies on assumptions about the semantics of natural kind terms, general philosophical assumptions about reference, and straightforward empirical assumptions, but presupposes no non-trivial essentialist premises

It is often thought that Blackburn and Boghossian have provided an effective reply to the finiteness objection to dispositional theories of meaning, presented by Kripke's Wittgenstein. In this paper I distinguish two possible readings of the sceptical demand for meaning-constitutive facts. The demand can be formulated in one of two ways: an A-question or a B-question. Any theory of meaning will give one of these explanatory priority over the other. I will then argue that the standard reply only works if B-questions are seen as prior, while the dominant dispositionalist theories of meaning see A-questions as prior

What is it for a predicate or a general term to be a rigid designator? Two strategies for answering this question can be found in the literature, but both run into severe difficulties. In this paper, it is suggested that proper names and the usual examples of rigid predicates share a semantic feature which does the theoretical work usually attributed to rigidity. This feature cannot be equated with rigidity, but in the case of singular terms this feature entails their rigidity, as understood in the standard characterisation. Hence, it is appropriate to call this feature proto-rigidity.

It is generally assumed that rigidity plays a key role in explaining the necessary a posteriori status of identity statements, both between proper names and between natural kind terms. However, while the notion of rigid designation is well defined for singular terms, there is no generally accepted definition of what it is for a general term to be rigid. In this paper I argue that the most common view, according to which rigid general terms are the ones which designate the same kind in all possible worlds, fails to deliver a posteriori necessities. I also present an alternative view, on which the work of explaining a posteriori necessities is not done by rigidity, but by a related metasemantic notion, which I call actuality - dependence