![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
ABSTRACT
Drawing on the relevant alternatives framework and Mary Kate McGowan’s work on conversational scorekeeping, I argue that usage of the term ‘conspiracy theory’ in ordinary language and public discourse typically entails the performance of what I call a dismissive conversational exercitive, a kind of speech act that functions to exclude certain propositions from (or prevent their inclusion in) the set of alternatives considered relevant in a given conversational context. While it can be legitimate to perform dismissive conversational exercitives, excluding alternatives that deserve to be taken seriously can be highly problematic for a variety of reasons. For one, it can give rise to what I call the problem of premature conclusions when subjects illegitimately dismiss certain propositions as irrelevant and, as a result, prematurely take certain conclusions or claims to be warranted. Depending on the kind of conclusion or claim, the problem can come in different variants, three of which I shall examine in more detail: the problem of premature knowledge claims, the problem of premature causal claims, and the problem of premature generic generalizations.
Acknowledgments
Many thanks to M R. X. Dentith, Melina Tsapos, and Will Mittendorf for helpful comments on an earlier draft of this paper.
Disclosure Statement
No potential conflict of interest was reported by the author.
Notes
1. That S must be in a position to rule out all the relevant alternatives to p is a necessary condition for S to know that p. Is it also a sufficient condition? Lewis (Citation1996) held that it is, but like many other authors I think that for a complete analysis of ‘S knows that p’ further conditions should be added, in particular that p is true and S believes that p.
2. It might not be clear at first sight that an alternative can be both irrelevant and evidentially excluded at the same time. But here is an example. Consider the possibility that your cat has shrunk to one centimetre and is sitting right in front of you on your desk. This is certainly an irrelevant alternative in the everyday life context of you looking for your cat. Nevertheless, you also have the evidence to exclude it: if it were there, you would easily see it (despite its size). So, the fact that this is an irrelevant alternative does not necessarily imply that it cannot be falsified by the available evidence. It implies that no evidence is needed to rule it out.
3. Note that the relevant alternatives framework as such does not entail a commitment to epistemic contextualism and can be used by invariantists as well (see, e.g. Rysiew Citation2006) (invariantists may assume, for example, that the set of alternatives to be ruled out does not vary across contexts). As for my argument about premature knowledge claims in section 6.1, I wish to emphasize that it does not in principle presuppose a commitment to contextualism and should also be acceptable to invariantists.
4. Surprisingly, conspiracy theorists are often accused of endorsing contradictory conspiratorial hypotheses, such as Princess Diana having been killed by MI6 and her faking her own death (an oft-cited reference for this charge is Wood, Douglas, and Sutton Citation2012). However, it seems reasonable to interpret the empirical data as suggesting not that conspiracy theorists irrationally believe contradictory propositions (as Wood et al. do), but that they consider them to be relevant alternatives to certain official accounts, which need not be irrational at all. Basham (Citation2018) and Hagen (Citation2018) have already raised similar objections to the Wood et al. study.
5. What about solitary subjects not currently involved in a conversation? As Blome-Tillmann (Citation2009, 256) has argued, the conversational model can also be applied to this case by conceiving these subjects to be individuals who are engaged in a sort of silent conversation with themselves. Furthermore, it is also possible to conceive of each temporally extended public debate as a conversational context.
6. Lewis calls this the ‘rule of attention’.
7. For the notion of a conversational scoreboard, see Lewis (Citation1979). For an influential argument to the effect that the mere mention of an alternative does not necessarily make it relevant, see Blome-Tillmann (Citation2009).
8. For example, an alternative’s relevance may depend on whether the conversation takes place in an everyday setting or a scientific one, and there may be further specific differences depending on the scientific discipline. Hawke makes an interesting point about how Kuhnian disciplinary matrices, paradigms, and scientific revolutions can be conceived in terms of relevant alternatives: ‘for an RA theorist, a change in paradigm, it might be suggested, involves a major shift in the space of the relevant hypotheses that a normal scientist must seek to select between’ (Hawke Citation2016, 218).
9. For example, one might argue that an alternative is relevant in a context if it would be taken seriously by an epistemically virtuous person in that context (see Tuckwell Citationforthcoming for a related proposal).
10. One thing to note is that the ‘question under discussion’ (Roberts Citation2012) in this example is a normative one (what Maureen should do), and the set of alternatives under conversational consideration in this case includes possible courses of action (in contrast to the cat search example, where the set includes possibilities of where the cat might be or what might have happened to it). This example illustrates not only how SScore can expand, but also the breadth of conversational moves that can be involved in and affected by such expansions.
11. Exercising one’s conversational veto power is also akin to what Langton (Citation2018) calls ‘blocking’. See also McGowan (Citation2019, 47ff.) for her (somewhat different) take on blocking. When an alternative has been added to the conversational scoreboard and is later removed through a dismissive conversational exercitive, it might also be natural to interpret this process as a special case of what Caponetto (Citation2020) calls ‘undoing things with words’.
12. There are similarities between dismissive conversational exercitives and a phenomenon referred to by Cull (Citation2019) as ‘dismissive incomprehension’, which occurs when ‘a speaker purports ignorance of the meaning of another speaker’s speech in order to undermine that other speaker’ (262). While this phenomenon can also be realized through utterances such as ‘Nonsense!’ or ‘That’s just crazy talk!’, dismissive incomprehension involves speakers who do not even linguistically understand (or pretend not to understand) the meaning of an interlocutor’s utterances. (As Pigden Citation2010 observes, this phenomenon also abounds among philosophers in the history of philosophy.) By contrast, I am concerned primarily with situations in which speakers do understand one another’s utterances linguistically, but do not consider (or pretend not to consider) it worth taking the expressed proposition seriously (e.g. because they regard the probability of its being true as too low).
13. See Langton (Citation2018) for an analysis of ‘back-door’ speech acts.
14. See also Uscinski (Citation2019, 20), who observes that ‘if conspiracy theorists investigate a theory that eventually turns out to be true, that theory stops being labeled conspiracy theory’.
15. Of course, the cover-up of an accident can assume a conspiratorial character.
16. There have also been attempts to show that the conspiracy-theory label is less pejorative than typically assumed, most notably in the study by Wood (Citation2016). However, as Husting, Orr, and Dentith (Citation2022) have argued, this study has serious limitations and fails to demonstrate conclusively a ‘rhetorical innocence’ of the label.
17. Note, however, that even in these cases the dismissive character of the label is visible. For example, if someone says, ‘I wish it were just a conspiracy theory’, we may well understand the person to be implicitly expressing that if the proposition in question were a conspiracy theory, it could be dismissed out of hand – but unfortunately it is not.
18. While I have argued in section 2 that SScore can have members not included in SRel, some contextualists may reject this assumption, arguing that every act of adding an alternative to SScore automatically adds it to SRel as well, creating a new context in which that alternative deserves to be taken seriously. But in response one could ask a question similar to the one just raised: Can it be legitimate to perform dismissive conversational exercitives in order to prevent the creation of contexts in which the dismissed alternatives would be relevant? I would answer this question positively as well (see Ichikawa Citation2020 for an argument to the effect that the creation of certain contexts can be problematic and should be prevented).
19. One might argue, for example, that ‘conspiracy theory’ is a particularly insidious term, not because of its pejorative nature per se, but because of the mismatch between its surface grammar, which suggests that it refers to conspiratorial hypotheses in general, and its depth grammar, i.e. its exclusive use to dismiss such hypotheses as unreasonable.
20. Conspiracy theories have occasionally been used in the relevant-alternatives literature as examples of propositions that can properly be ignored (e.g. by McKenna Citation2014). To my knowledge, however, the potential use of the conspiracy theory label to dismiss alternatives that should not be ignored has not yet been studied.
21. On conspiracy theories and democratic deliberation, see also Mittendorf (Citationforthcoming).
22. This can be regarded as a form of ‘context control’, which Cappelen and Dever examine in their discussion of what they call ‘devious scoreboards’: ‘If you are in a position to manipulate what the possibilities on the scoreboard are, you can indirectly control what is mutually understood to be possible and impossible’. (Cappelen and Dever Citation2019, 28) As an illustration, consider the following variation of McGowan’s example from section 3. Suppose that Maureen is confronted by her husband with the accusation of adultery and firmly rejects it, remarking that this is ‘nonsense’. In this case, she herself does not consider her husband’s assumption absurd (in fact, she knows it is true). Rather, she merely wants to create this impression; she wants to exclude this option from the set of alternatives he considers relevant, thus ensuring that he no longer seriously pursues it, collects evidence for it, etc.
23. The ‘conspiracy-theory’-as-propaganda view is defended by DeHaven-Smith (Citation2013), who argues that the term was popularized by the CIA in the course of what has turned out to be ‘one of the most successful propaganda initiatives of all time’ (DeHaven-Smith Citation2013, 25; see also Coady Citation2018a). For a detailed examination of the potential strategic motives behind ‘conspiracy theory attributions’, and how one may deal with them, see Martin (Citation2020).
24. Even the mere anticipation of a dismissive conversational exercitive can lead to epistemic injustice. As an example, consider a person who refrains from putting forward an investigative hypothesis worth considering out of fear that it will be stigmatized as a ‘conspiracy theory’ (see Lee Citation2021 for a discussion of the concept of anticipatory epistemic injustice).
25. The individual(s) ascribing the knowledge can, but need not be identical to the subject(s) to whom the knowledge is ascribed. The epistemic contextualism literature commonly distinguishes between the subject’s context (i.e. the context of the would-be knower(s)) and the ascriber’s context (i.e. the context of the person(s) who attribute – or deny attribution of – knowledge to the subject). (Relativists emphasize that there can be a third context, that of an assessor, who need not be identical with either of the other two (Wright Citation2017)). The performance of a dismissive conversational exercitive can have different effects depending on the context in which it is made. For example, we can imagine a group of ascribers who illegitimately dismiss p as irrelevant and, as a result, prematurely attribute knowledge that q to an absent subject who is unaware both of this attribution and the alternative’s dismissal. On the other hand, if an alternative is dismissed within the subject’s context (e.g. by the subject themselves), this potentially has an additional effect: it discourages the subject from gathering evidence to test that alternative (assuming that gathering evidence to test an irrelevant alternative would be a waste of epistemic resources).
26. The media then began publishing more and more articles like ‘Covid origin: Why the Wuhan lab-leak theory is being taken seriously’ (BBC, 27 May 2021) or ‘Covid “lab leak” theory is no longer just a conspiracy theory’ (The Herald, 12 June 2021).
27. For an application of the underlying idea to conspiracy theories, see also Gardiner (Citation2021).
28. There are certain differences between Blake-Turner’s epistemological framework and mine. For example, I don’t think that encountering and taking seriously a fake news story or conspiracy theory automatically shifts the subject into a context in which that story or theory is relevant. But I agree that it is possible for such context changes to occur. Moreover, regardless of whether or not a story or theory is really a relevant alternative, if the subjects in the context consider it relevant, they will only then attribute the knowledge in question to themselves if they have the evidence to falsify it. Blake-Turner’s argument – or a version of it – can thus be made to work independently of his particular epistemological assumptions.
29. See for example Dentith (Citation2016); Cappelen and Dever (Citation2019, 56ff.); Habgood-Coote (Citation2019); van der Linden, Panagopoulos, and Roozenbeek (Citation2020); Coady (Citation2021); Jahng, Stoycheff, and Rochadiat (Citation2021).
30. On inference to the best explanation, see for example the essays in McCain and Poston (Citation2017). For an interesting application of the relevant alternatives framework to causal claims from a contextualist perspective, see Maslen (Citation2013). Note, however, that while my argument in this section is compatible with contextualism about causal claims, it does not presuppose any commitment to it.
31. There are different types of generics, some of which can be justified on purely ‘inductivist’ grounds, while others require a combination of inductivist and modal elements for their justification, i.e. they require that there be a statistically salient pattern generated by some counterfactual-supporting mechanism. Still others cannot be inductively justified at all, e.g. generics like ‘Bishops move diagonally’, which express norms, definitions, or conventions. For a discussion of the scope and limitations of an inductivist approach to justifying generics, see Cohen (Citation2001, 193ff.); a typology of generics is provided by Prasada et al. (Citation2013); for a general overview see also Cappelen and Dever (Citation2019, Ch. 8).
32. In Hauswald (Citation2023) I discuss some of these terms in context.
Additional information
Notes on contributors
Rico Hauswald
Rico Hauswald is a Privatdozent at the Department of Philosophy at Technische Universität Dresden. His work focuses on social epistemology, philosophy of science, social ontology, philosophy of medicine, and philosophy of artificial intelligence.