Comment: on Sosa’s telic epistemology

Sosa’s response to the swamping problem improves upon traditional responses, even upon those within virtue epistemology itself, in at least two ways. It proposes a (still) novel view of epistemic normativity in the form of action/performance normativity. In addition, through this view, it conceives of beliefs as forms of attempting, consciously or not, to get it right about some particular fact or state of affairs. This way of viewing beliefs avoids the traditional reliabilist notion of a belief as a product of a well-functioning cognitive process, which seems to be easily affected by the challenges raised by critics of several externalist theories of justification.

The particular type of telic virtue epistemology that Sosa (2021) succinctly presents is the result of a multi-decade long effort he underwent to build an elegant epistemology that perhaps could be fit to face both more traditional justification-related tasks, like responding to Gettier-like cases, and more specific problems, such as the problem of explaining the value of the output of a well-oiled belief formation process over the equally accurate outputs stemming from luck (either good or bad luck). By uniting both the idea that our belief-forming processes are attempts to fulfil a particular telos (getting to the truth) and the idea that to do that in the best way possible we should embark in a second order reflexive endeavor, Sosa (2010) points to the path that could lead us to, what he calls elsewhere, “knowledge full well”.

The way we walk this path, he argues in this paper, must be the main object of epistemic evaluation. If beliefs are nothing more than attempts to get it right, we should, according to this view, evaluate the quality of such attempt, which, in turn, will reveal the quality of the resulting state (whatever this may be called). Thus, the main task of epistemic normativity is to assess the quality of a particular type of action or performance, the action of forming a belief that aims to truth and, hopefully, gets to it competently and consequently - it gets to the truth as a consequence of the competent performance. According to Sosa, this picture is enough to direct us away from the swamping problem and its question of the value of knowledge.

Sosa’s account in this paper, however, leaves us with a few knots to untie in terms of a complete understanding of the dynamic nature of belief-forming mechanisms. One can be completely sympathetic to the praxis turn in epistemology that Sosa promotes, via the view that agency is central to epistemic evaluation, and still be puzzled by how to exactly cash out some of the details.

If the currency that epistemic agency trades in is representation as an attempt or endeavor towards truth, either in the form of affirmations or judgements, how should we explain the nature of long-lasting cognitive states such as unrevised beliefs one holds for a long time? How should we see beliefs that are seared in our intellect, long after their acquisition, as something more than mere mental states without the dynamic nature of the relation an agent has to propositions that are current to her mind, and to which truth she attempts to get, accurately, adroitly and aptly?

It is one thing to argue that believing is endeavoring towards truth, and that epistemic normativity should focus only on the quality of such endeavor. It is another thing to cash out all our current and dispositional beliefs as exhibiting the same degree of transparency of the performance responsible for their formation. How can we evaluate long-held beliefs and other cognitive dispositions in terms of their performance towards truth when it seems that the only element we can grasp to assess is the final product of the epistemic praxis?

Take the case of a dispositional belief, such as my belief that there were four mango trees in my street when I was a kid. Suppose I do not currently entertain such belief. Is this belief one that can be evaluated in terms of the quality of the past performance towards truth? Or should this belief be entertained in order to be assessed? In this case, what are the conditions under which this belief should be evaluated? Should we count past perceptual mechanisms to assess the competence when counting the believing as adroit? Should we reserve this evaluation only to my memory and other inferential mechanisms that I currently make use of?

Maybe these are not so interesting questions when we are dealing with such a robust and neat proposal. But since Sosa’s theory want us to abandon the idea of beliefs as products in favor of the view of them as processes, as attempts to get it right, maybe we need a clearer account of the beliefs that don’t seem to fit perfectly to the performance model; beliefs we tend to see more as products of a cognitive process in the traditional reliabilist sense. If we want to move past the swamping problem, as we might think it could be done so via Sosa’s account, we might need to explain away any loose thread that could throw us back into its hold. Until then, though, let’s keep admiring the unique and elegant theorizing contained in Sosa’s writing and let’s get inspired by his notable performance in improving and advancing epistemology in its more interesting aspects.

References

• SOSA, Ernest. Knowing full well. Princeton: Princeton University Press, 2010.
• SOSA, Ernest. Representations, judgments, and the swamping problem for reliabilism: why the problem applies to process reliabilism, but not to virtue reliabilism. Trans/Form/Ação: Unesp journal of philosophy, vol. 44, Special issue in honor of Ernest Sosa, p. 19-24, 2021.

Publication Dates

• Publication in this collection
  28 June 2021
• Date of issue
  06 2021

History

• Received
  27 Apr 2021
• Accepted
  05 May 2021