Fredrik Haraldsen

It is often thought that some general terms or kind terms, in particular natural kind terms, are rigid designators, and that a properly extended notion of singular-term rigidity can help explain the behaviour of such general terms. In this article, I argue that the only legitimate notion of general-term rigidity is a trivial one and identify some crucial asymmetries between a posteriori necessary truths involving names and a posteriori necessary truths involving general terms. If we pay attention to these asymmetries, it becomes clear that attempts to draw a non-trivial rigid/non-rigid distinction for general terms are bound to fail to offer the explanations we expected. As opposed to singular terms, what is interesting about the behaviour of natural kind terms, even in modal contexts, cannot be accounted for by modal distinctions like rigidity/non-rigidity but must rather track non-modal features.

Non-actualist theories promise straightforward accounts of meaning, truth and reference of fictional discourse but are ostensibly saddled with a Selection Problem, that multiple possible candidates satisfy the role-descriptions associated with names used in fictions and no principled way to distinguish between them; yet if names are referential, there can only be one referent. The problem is often taken to be a serious—even decisive—obstacle for non-actualism, and the aim of this article is to show that the challenge can be met. I suggest that storytellers and authors fix the referents of referential terms they use by arbitrary selecting referents through simple acts of stipulation, and then determine which of the worlds containing the relevant individuals serve as truth-makers for the story by adding properties to the characters. The resulting view of reference is consistent with reasonable, foundational requirements on reference, such as a causal-historical theory, and can moreover be used to support plausible theories of truth in fiction and facilitate elegant models of fiction as a type of discourse.

According to possibilism, or non-actualism, fictional characters are possible individuals. Possibilist accounts of fiction do not only assign the intuitively correct truth-conditions to sentences in a fiction, but has the potential to provide powerful explanatory models for a wide range of phenomena associated with fiction (though these two aspects of possibilism are, I argue, crucially distinct). Apart from the classic defense by David Lewis the idea of modeling fiction in terms of possible worlds have been widely criticized. In this article, I provide a defense of a possibilist account against some lines of criticism. To do so, I assume that names for fictional characters are directly referential and a possible-worlds model that accommodates transworld identity. On this background, I argue, it is possible to construct an elegant model of fictional discourse using familiar models of information exchange in ordinary discourse, and I sketch how this model can be used to i) make a natural distinction between fictional and counterfactual discourse, ii) account for creativity, and iii) sustain a natural definition of truth-in-fiction that avoids certain familiar objections to possibilism. Though I set aside questions about the metaphysical commitments of a possible-world interpretation here, there is accordingly reason to think that the battle over possibilist treatments of fiction will have to be fought over metaphysical foundations rather than technical shortcomings.

Though it is often claimed that some general terms are rigid designators, it has turned out to be difficult to give a satisfying definition that avoids making all general terms rigid, and even if a non-rigid reading is available, makes that non-rigid reading matter. Several authors have attempted to develop examples that meet the trivialization challenge, with Martí and Martínez-Fernández providing what is, perhaps, the most convincing strategy. I show that the type of example Martí and Martínez-Fernández offer nevertheless fails to meet the trivialization challenge and, accordingly, that we should still have serious doubts about whether continuing the search for a non-trivial definition of general term rigidity would be fruitful.

It is often thought that a notion of general term rigidity could help explain the particular behavior of natural kind terms in modal contexts. An influential strategy for developing a non-trivial account of general term rigidity appeals to essential properties of the things to which such terms apply. I show that essentialism cannot underpin a notion of rigidity that can play the expected explanatory roles. Essentialists are committed to presuppositions that themselves play those roles without implying essentialism.

Although it is common to claim that certain general terms or kind terms are rigid designators and that their rigidity helps explain their behavior in modal contexts, it has turned out to be surprisingly difficult to define an adequate notion of rigidity for general terms. Such definitions tend, as argued in particular by Scott Soames, to lead to a type of overgeneralization that leaves the purported rigidity of general terms explanatorily inert. In recent years, several attempts have been made to circumvent the problem, and the present article focuses on a particular and potentially powerful strategy developed by Joseph LaPorte in his recent book Rigid Designation and Theoretical Identities for blocking one of the core inferences in Soames’s case against general term rigidity. I argue that the type of response LaPorte promotes is bound to fail; though it might initially appear to circumvent the threat of trivialization described by Soames, it is susceptible to a different and arguably even worse kind of trivialization challenge. In conclusion, Soames’s arguments remain a significant obstacle to identifying a non-trivializing definition of kind term rigidity, and we have good reasons to think that an explanation for the modal status of theoretical identity statements must be sought elsewhere.

The characteristic property of definite descriptions in natural language is commonly assumed to be their uniqueness requirement, although there is disagreement with respect to how occurrences should be interpreted, for instance with regard to the well-known restriction problem. I offer a novel argument against characterizing definite expressions in terms of uniqueness. If a singular definite description ?the F? implies that its denotation is the unique satisfier of ?F? (relative to a context) then there are real-life states of affairs that can be described in simple first-order languages, but which we are simply unable to describe accurately in natural language. I argue, first, that there is no way to describe these states of affairs properly without using definite descriptions. Second, if definite descriptions imply uniqueness we will systematically get the wrong truth conditions, regardless of whatever semantic or pragmatic resources the defender of the uniqueness-implying approach invokes. Hence, the Russellian idea of characterizing definite descriptions in terms of uniqueness must be given up. In the final section I explain what an adequate account of definite expressions must achieve?primarily coordinating elements brought to salience by context or previous discourse?and sketch a way of accomplishing this by using a liberal version of familiarity theory. Although there may be other accounts that can do the job, the account I sketch (based on an approach that has largely been overlooked) shows promise and can easily deal with the inexpressibility problem

The actually-operator, understood as a rigidifier, has been employed for a range of purposes in natural language semantics. In this article I argue that the properties of the operator do not correspond to any feature of natural language or feature natural language users have access to. Nor is it needed to provide a formal representation of natural language sentences—the examples usually provided to illustrate the indispensability of the operator are much more plausibly interpreted using plural quantifiers. This lack of connection to natural language is a serious worry for accounts that appeal to rigidifying operations to explain natural language phenomena, as well as a challenge to theories that appeal to the operator to capture the difference between different kinds of necessity expressed in natural language