![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
Wittgenstein distinguished between two uses of “I”, one “as object” and the other “as subject”, a distinction that Shoemaker elucidated in terms of a notion of immunity to error through misidentification (“IEM”); first-personal claims are IEM in the use “as subject”, but not in the other use. Shoemaker argued that memory judgments based on “personal”, episodic memory are not strictly speaking IEM; Gareth Evans disputed this. Similar issues have been debated regarding self-ascriptions of conscious thoughts based on first-personal awareness, in the light of claims of “thought insertion” in schizophrenic patients. The paper aims to defend a Shoemaker-like line by critically engaging with some compelling recent contributions. Methodologically, the paper argues that to properly address these issues the all-inclusive term “thought” should be avoided, and specific types of thoughts countenanced.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1. Wiseman (Citation2019) raises interpretive issues about Shoemaker’s “elucidation”; Palmira (Citation2022) compellingly addresses the philosophical implications she derives from them.
2. Putnam held that being H2O defines the real essence of the kind designated by “water” in its “predominant sense” (Putnam, Citation1975, p. 239); he explicitly allows that “water” has another sense defined by superficial traits, cf. Tobia et al. (Citation2020) for X-φ validation. Like me, Campbell (Citation2002, p. 35) distinguishes IEM as a pre-theoretical datum, from the philosophical theories that aim to account for it; cf. also Recanati (this volume, Section 1–2). Campbell’s indications on what he takes to be the datum (Campbell, Citation1999, p. 89) agree with the proposal I’ll go on to make in the main text.
3. This slightly modifies Seeger’s (Citation2015b, p. 2) account, aiming to cover cases of Pryor’s (Citation1999) “wh-immunity”, see below. McGlynn (Citation2021, p. 2305) offers a definition that I take to develop this core idea; for my purposes a simpler one is good enough. Recanati (this volume, Section 1) also offers a good, intuitive similar characterization. Cf. Morgan (Citation2019, pp. 446–447) and Palmira (Citation2020, Section 2) for related notions and McGlynn (Citation2021, pp. 2309–2310) for a defense of the specifics of his own.
4. Wright (Citation1998, p. 19), Campbell (Citation1999, p. 89) and Prosser (Citation2012, pp. 161–162) support this core idea; cf. Coliva & Palmira (this volume-a, Section 2) for more details on different views on the phenomenon. Pryor’s “package deal” metaphor parallels Evans’s (Citation1982, p. 182) “no gap” one.
5. Cf. Coliva and Palmira (Citationthis volume-b, Section 1) and Recanati (Citation2024, Section 2) for elaboration.
6. To elaborate: “this keyboard is black” appears to be IEM because, if the existential “part” of the judgment is true and justified (i.e., something in the immediate vicinity of the speaker, the source, is black), it doesn’t seem to be possible that it is not the referent of “this keyboard”. The intuitive reason – which the Metasemantic account I outline below articulates – is that the referent of the demonstrative “this keyboard” is in part perceptually fixed in the relevant context as the black object before the speaker. Formally and philosophically well-developed versions of the no-reference view have been advanced, cf. Ninan (Citation2013), Bochner (Citation2023), p. I don’t think these models vindicate the Simple Account, because they still presuppose forms of “identification” consistent with the Metasemantic view; but I cannot develop this point here.
7. Cf. Coliva and Palmira (Citationthis volume-a, Section 1) for elaboration on the debate.
8. Campbell (Citation1997, p. 70) argues against a view like the one just outlined with the following example: “if you judge, ‘that chair is yellow’, it may be that you thereby know of something that it is yellow, but that thing is not the chair, if, for instance, the chair is transparent and set against a yellow background. If your judgment is mistaken, you can rectify it by retreating to the more cautious, ‘At any rate, something is yellow’.” García-Carpintero (Citation2018) retorts that this just shows that the IEM in these cases is merely de facto.
9. Perhaps they are not that weird. Folk viewers of Blade Runner (Scott, 1982) don’t manifest any difficulty in accepting that this admission by Rachael presents a coherent possibility: “I remember [music] lessons, but I don’t know if it is me or Tyrell’s niece” – Deckard has just told her that her private memories were “implants” from those of Tyrell’s niece.
10. “IEM arises only if the original justification cannot be cited as support for ‘Someone is F’ when the assertion is doubted for any reason at all” (Hamilton Citation2007, p. 411), my emphasis. This is the desideratum that Palmira (Citation2020, p. 3840) calls “Preservation of Grounds”; cf. also Coliva and Palmira (Citationthis volume-a, Section 2).
11. This is Shoemaker’s (Citation1968, 563f.) “package deal” intuitively correct point, quoted in Section 1 “[…] in being aware that one feels pain one is, tautologically, aware, not simply that the attribute feel(s) pain is instantiated, but that it is instantiated in oneself”.
12. I’ll avoid henceforth the circumlocutions required by the factivity of “perception” and “memory” by using those terms for perceptual and mnemic experiences or seemings. I hope no misunderstanding ensues. I’ll come back to the importance of the distinction in Section 4.
13. Cf. Perrin and McCarroll (Citation2023, p. 306) for a characterization congenial with this view.
14. I mostly agree with Teroni’s (Citation2024) account of how memories (mnemic experiences really) are identified by the subject, which assumes as I do that this is a “personal-level” (as opposed to subpersonal) matter. It allows for mistakes both of omission (failing to realize that an attitude is a mnemic experience) and commission (mistaking as one something that is not, perhaps an imagining), but it assumes that typically we are reliable. I also agree with him that, given sensible constraints he sets up, accounts like Fernández’s (Citation2024) in terms of the contents of memories are inadequate; adequate accounts should rely instead on traits of the attitude type. Teroni offers a compelling one on which it features a “feeling of familiarity”.
15. Hamilton argues that story cannot be taken as I do, by invoking Evans’s consideration outlined above to dismiss my Shoemakerian conclusion: without smuggling information into the defeater, making it additive instead of undercutting, “the idea that, if it was not myself, then it must have been someone else who made the objection, would just be a stab in the dark” (Hamilton Citation2007, p. 414). García-Carpintero (Citationin press) argues that his considerations assume preservationist intuitions and disregard thereby observer memories.
16. The same may apply to bodily self-ascriptions, including pain if we are prepared to question the logical status of Shoemaker’s “package deal” view about them, fn. 11.
17. Even if the story as I imagined it was not just a coherent counterfactual possibility, but an actual process, this wouldn’t establish that memory is not de facto IEM. This depends on the proper way of characterizing that notion, and the epistemically relevant facts. There might be, as envisaged, good reasons for the view that, in clear-cut situations constituting an epistemic default, memories (and bodily proprioception-based self-ascriptions) must be reliable. Recanati (Citation2024, Section 3–4) develops an account I sympathize with.
18. The full report was “One evening the thought was given to me electrically that I should murder Lissi” (Mullins & Spence Citation2003, p. 295). Partly in response to sensible complaints by Gunn (Citation2016) and Henriksen et al. (Citation2019) that philosophers repeatedly recycle the same examples, I’ll offer new ones, see the Kip Kinkel case. I am afraid that I don’t comply with their advice that to address these issues we obtain “a solid, clinical grasp of the phenomenon of thought insertion” (Henriksen et al. Citation2019, p. 3). I agree that ideally one should; I’ll try to limit my claims to what can be justified by standard philosophical methodology and theories of contentful attitudes.
19. Cf. Roessler’s (Citation2013, pp. 666–668) nuanced discussion. Verdejo (Citation2023) argues for the bold view that “in the light of their experiences, people undergoing thought insertion express, or potentially express, fully rational judgments concerning the ownership of their thoughts” (ibid., Section 1). Even if relativized to subjects’ experiences, the “fully rational” appraisal sounds a bit overblown given how confusing patients’ reports are, and the little we can in fact know about their experiences.
20. Cf. Rothenfluch (Citation2021, p. 166 fn). Defenders of this sort of account might instead appeal to “additions” to the experience of subjectivity, as opposed to deficits; cf. Parrott (Citation2017, Section 2). Mathieson (Citation2023, Section 5) doubts that this might be the whole story.
21. Gunn (Citation2016, p. 572) argues that “[t]he subjective ‘feel’ of the experience of one’s own mental activity is how one knows that one is the agent, author and owner of a thought. If one does not recognize that a thought is one’s own (for whatever reason) then personal ownership is lacking”. By “personal ownership” she is clear that she means subjectivity (ibid.). The argument begs the question. As we have seen, subjects typically acknowledge a sense in which the thoughts they disown are still theirs (“the thought was given to me electrically … ”, fn. 18, my emphasis). It is a theoretical question to articulate in what sense subjects deny ownership; we cannot just declare that patients’ denials already establish that subjectivity is missing or at least diminished (Sousa & Swiney, Citation2013, p. 643). Seeger (Citation2013, p. 262) points out the same questionable move in Martin and Pacherie (Citation2013, p. 113).
22. Rothenfluch (Citation2021) offers a recent variation on this sort of account, on which “the experience that an introspectively accessed thought is not one’s own consists in the feeling of rational indifference toward the thought. Patients have this experience because they lack an expectation of rational authority” (ibid., 168) that they should have, given that “inserted thoughts are of the type that we would ordinarily expect to be rationally justified” (ibid., 169).
23. Rothenfluch (Citation2021, p. 169, fn.) should grant this, for her account of spontaneous thoughts as nonetheless expected to be under our rational control (in that we might discard them) doesn’t apply to orders felt as inserted. The problem they pose for her account is that patients’ “rational indifference” toward them is in such cases entirely apt. Also, patients accept the commands as authoritative and are thereby committed to the ensuing normative judgments, I should kill Lissi/I should kill Dad. They thus fully meet any “expectation of rational authority” vis-à-vis them; but they still disclaim them as not their own.
24. Proponents of commitment accounts are indeed right that there are many cases in which what patients report is lack of commitment, cf. Gunn (Citation2016, p. 563) for illustration.
25. Roessler (Citation2016, Section 2) has an illuminating discussion of the debate confronting this view and its competitor, that inner speech is just the imaginative representation of speech.
26. Roessler (Citation2013, p. 661) claims that “in the familiar sense, the term [authorship] applies to types, not tokens. It is books that have authors, not individual copies of books”. But artists also author paintings, sculptures and performances, in addition to books and symphonies.
27. Currie’s account has it that subjects miscategorize their imaginings as beliefs because imaginings are actions and the subpersonal mechanism underlying action awareness (which, like Campbell, Sugden (Citation1999, pp. 611–614), he identifies with Frith’s forward-efference/comparator model) fails in them. If they don’t endorse those beliefs, they explain their presence as having been produced by others. This is on his view how the delusions can be seen as “broadly rational responses to highly unusual experiences” (Campbell, Citation2002, p. 39). Langland-Hassan (Citation2016) discusses objections to Frith’s model and offers alternative accounts that might underwrite similar explanations.
28. Patel (Citation2023) defends the prima facie hugely implausible “antirealist” view that “sufferers of thought insertion do not have a thought, but, as with dreams, merely simulate having a thought inserted into their minds” (ibid., Section 1). But dreams are categorized as either imaginings or hallucinations (Gregory Citation2023), and hence thoughts in a generic sense; and accounts of pretense and simulation explain them as resulting from imaginings (Picciuto & Carruthers, Citation2016). Patel in fact argues for the more nuanced claim “that thoughts corresponding to first-person descriptions of thought insertion are not parts of thought insertion episodes. … the thought corresponding to the description is the thought Kill god, and so the anti-realist is only denying that this thought is a part of the episode”, (ibid., Section 3). To the extent that I understand it, this more nuanced claim still seems wrong to me. Their reports offer good reasons, I take it, to think that patients carry out acts of propositional imagining that someone tells them to kill God (Mom, Lissi …); that they sometimes experience themselves as receiving such orders from a source invested with the required authority, judge that they should act on them, and, tragically, do so.
29. On Seeger’s (Citation2015b, p. 847) account of agency, subjects would be the producers also in this case, but I take them to be assuming a concept of agency on which they are not the (ultimate, truly significant) agents – cf. Vosgerau and Voss (Citation2014, pp. 539–541), who offer a nuanced account of what I am calling agency in terms of causal and control features; cp. Seeger’s (Citation2015b, pp. 849–850).
30. This condition is meant to block the objection that the possibility I am envisaging inaptly characterizes an additive defeater (see the discussion above in §2), along the lines of Pryor’s (1998, 296) response to Evans. The defeater I consider doesn’t bring with it the notion that someone else authored the command; it is a general background condition that supports it. It is thus as with standard cases establishing that a self-ascription is VEM, for instance the Perky (Citation1910) inspired cases I describe immediately below.
31. I assume that the envisaged situations are remote possibilities, leaving open the view that self-ascriptions of intentions or self-commands are de facto IEM. A reviewer worries that if they are possible an excessive skepticism about our own self-knowledge follows, threatening our rationality. I addressed this worry in fn. 17. Consider the Perky-inspired more ordinary cases mentioned in response to Roessler’s Anscombian argument. My own experience shows them to be real. However, I think a sensible, fallibilist epistemology would nonetheless allow that we can typically tell by introspection whether we are perceptually experiencing or imagining.
32. Verdejo (Citation2023) invokes his “full rationality” claim (fn. 19) to argue for the philosophical significance of the agent-and-author based self-reference rule, not just the experiencer rule.