Is a subpersonal virtue epistemology possible?

ABSTRACT

Virtue reliabilists argue that an agent can only gain knowledge if she responsibly employs a reliable belief-forming process. This in turn demands that she is either aware that her process is reliable or is sensitive to her process’s reliability in some other way. According to a recent argument in the philosophy of mind, sometimes a cognitive mechanism (i.e. precision estimation) can ensure that a belief-forming process is only employed when it’s reliable. If this is correct, epistemic responsibility can sometimes be explained entirely on the subpersonal level. In this paper, I argue that the mechanism of precision estimation–the alleged new variety of epistemic responsibility–is a more ubiquitous phenomenon than epistemic responsibility. I show that precision estimation operates at levels that are not always concerned with the epistemic domain. Lastly, I broaden this argument to explain how all subpersonal epistemologies are likely to fall prey to the problem of demarcating cognitive agency and the problem of attributing beliefs.

KEYWORDS:

• Extended cognition
• subpersonal epistemology
• epistemic responsibility
• virtue epistemology
• cognitive integration
• precision estimation

Acknowledgements

I would like to thank Duncan Pritchard, Mog Stapleton, Orestis Palermos, Kathryn Nave, and two anonymous referees for their helpful comments on different drafts of this paper. I would also like to thank Julian Hauser for their invaluable feedback.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Virtue reliabilists claim that cognitive ability, which is to responsibly employ one’s reliable belief-forming processes, is necessary for knowledge. See Sosa (1988); Sosa and Plantinga (1993); and Greco (2007) for more on cognitive ability.

2 See Greco (2010) for a discussion on our need for a perspective on the reliability of our belief-forming process. He discusses how philosophers have tried to debate the skeptics who argue that without a perspective on how one’s forming true belief knowledge is not possible. Demanding a perspective also leads to infinite regress, circularity or arbitrary assumption.

3 The word weak only refers to the conditions required for this kind of integration. Integration isn’t weak in any way, but the conditions for it are not very demanding.

4 Note that Clark does not take issue with the strong and weak cognitive integration accounts as such. He agrees that stringent epistemic hygiene practices are necessary in some cases. But, for some cases, like employing ones trusted internal cognitive faculties (and extended cognitive faculties), Clark argues that a much weaker notion of epistemic responsibility can be sufficient (Clark 2015).

5 The Temp case is an adaptation of Lehrer (1990)’s Mr Truetemp.

6 See Pritchard (2018b) for a discussion on evidential environmental luck that is compatible with knowledge.

7 This paper is specifically concerned with the virtue reliabilist notion of epistemic responsibility. That is, how an epistemic agent can responsibly employ her reliable belief-forming to generate knowledge. Following are some other ways the concept of epistemic responsibility can be found in epistemology literature. Virtue responsibilists argue that epistemic responsibility is cultivated by virtuous character traits. Some epistemologists investigate what epistemically responsible beliefs are and how we form them (Peels 2017). Others inquire if and how justification is related to forming responsible beliefs (Kornblith 1983).

8 The word sensitivity here is not meant to indicate the sensitivity principle which claims that S knows p only if S would not believe p if it was false (Nozick 1981). I only mean to indicate a kind of responsiveness to one’s reliability.

9 Pritchard (2012) cautions us to use knowledge in this case. Temp does not necessarily meet conditions for knowledge by simply meeting the cognitive ability condition. There is another intuition on knowledge–the anti-luck intuition. See (Pritchard 2014, 2017; and Pritchard 2018b) for more on anti-luck virtue epistemology.

10 For arguments in favour of a strong epistemic responsibility criteria, see Chisholm (1977), BonJour (1985), and Steup (1999). Also, see Greco and Sosa (1999) on why such a strong concept of epistemic responsibility is vulnerable to skepticism.

11 This variation of the Temp case is borrowed from Pritchard (2010). He asks to imagine ‘Tempo’, someone who is fitted with a smart device that records ambient temperatures. He is brought up in a culture in which relying on such a device is common. We wouldn’t demand that Tempo be aware of the reliability of his process in this case, just like we wouldn’t demand that children have meta-beliefs about the reliability of their memory and perceptual faculties. And yet, we do credit children with knowledge of most of the things they see or the things they remember. This indicates that there is a weaker requirement that may capture epistemic responsibility (Pritchard 2010, 146).

12 According to Greco and Sosa (1999, 287), our cognitive characters are not only made of our internal neural faculties that make our central nervous system. Skills of perception, methods of inquiry, advanced technologies, etc. can also become a part of our cognitive characters by integrating properly.

13 Palermos (2021) mentions how these meta-cognitive cues are produced when parts of our cognitive system interact in a way that acts a specific kind of defeater. The latter alerts the agent that there might be something wrong with the way they have formed their belief. Also, see Palermos (2021) to understand epistemic responsibility in terms of epistemic defeaters and how our cognitive systems react to them.

14 See Audi (2015) to understand different kinds of believing: occurrent and dispositional.

15 In Carter et al. (2018), we learn how perhaps epistemic integration is different from metaphysical integration. The latter allows an external vehicle to become a part of our cognitive structure such that it extends our cognition into the environment. This kind of integration can be an entirely subpersonal phenomenon. Epistemic integration, however, requires more agential involvement from the agent.

16 In this paper, Stich (1978) is arguing against Gilbert Harman’s claim that retinal simulations provide inference to our perceptual beliefs. Harman also claims that inference is a relation exclusively amongst beliefs, and so he regards subdoxastic states as kinds of beliefs. Stich (1978) distinguishes between beliefs and subdoxastic states and claims that inference is not a relation exclusively amongst beliefs.

17 In response to these arguments, it would help to note that necessary subpersonal mechanisms may only be enabling conditions and not constitutive conditions for epistemic responsibility. While enabling conditions make-possible a certain personal level phenomenon, constitutive conditions are needed to explain what the phenomenon is. See McDowell (1994), Hornsby (2000), and Colombo (2013) for the debate on subpersonal states as enabling and constitutive conditions. Generally put, this debate discusses if subpersonal mechanisms only enable personal level phenomenoa (like epistemic responsibility). My worry here is that Clark perhaps doesn’t believe that epistemic responsibility is a personal level phenomenon. I am thankful to an anonymous referee for pointing me in the direction of this debate.

18 There is another worry that is related to subpersonal epistemology, but I will explore it in more detail in another paper. Clark writes, ‘I believe that a piece of new cognitive technology could be so well-designed as to be immediately assimilated into our daily routines, requiring no reflective window before properly being counted as delivering knowledge’ (2015, 3764). On Clark’s account, a belief-forming process can develop into an epistemic process immediately. Whereas, the cognitive integrationists, Greco and Pritchard, have argued that an agent ought to employ their belief-forming process frequently over a period for it to integrate into their cognitive system.

19 In conference talk, Hernández-Orallo and Vold (2019) propose that AI extenders, or external AI resources that extend our cognition outside our bodies ought to be treated differently from regular extenders. There is no doubt that with artificial intelligence extending our cognition, the complications will be tenfold. Also, see Hetherington (2012) for a discussion on how knowers or epistemic agents might extend into the environment.

Additional information

Notes on contributors

Hadeel Naeem

Hadeel Naeem is a visiting researcher at Bogaziçi University, Istanbul. She works in epistemology and philosophy of mind and is currently investigating how beliefs are generated through AI extension.