Abstract
Mark Nelson argues that we have no positive epistemic duties. His case rests on the evidential inexhaustibility of sensory and propositional evidence—what he calls their ‘infinite justificational fecundity’. It is argued here that Nelson’s reflections on the richness of sensory and propositional evidence do make it doubtful that we ever have an epistemic duty to add any particular beliefs to our belief set, but that they fail to establish that we have no positive epistemic duties whatsoever. A theory of epistemic obligation based on Kant’s idea of an imperfect duty is outlined. It is suggested that such a theory is consistent with the inexhaustibility of sensory and propositional evidence. Finally, one feature of our epistemic practice suggestive of the existence of imperfect epistemic duties is identified and promoted.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
This reading is corroborated on pp. 92 and 93 where Nelson replaces talk of what we ought or ought not believe given our ‘epistemic circumstances’ with talk of what we ought or ought not believe ‘given the evidence’. He confirms it expressly on p. 96.
While the language of duty is arguably too strong here, there may be a weaker sense of ‘ought’—call it a deflated epistemic ought—in which I ought to believe every proposition that my evidence supports. If we deny this, then we deny the normativity of logic. See Stapleford (2012) for discussion.
For a quite different argument that we never have a positive epistemic duty to adopt any particular doxastic attitude (including disbelief and suspension of judgement), see my (2012).
Perhaps Nelson could argue that the negative construal is preferable because the positive policy requirement of proportioning one’s beliefs to one’s evidence is extremely vague, while the injunction ‘Do not believe anything on insufficient evidence’ is much more determinate. The case for positive epistemic policy obligations would certainly be strengthened if what they require of us could be stated in very definite terms. A referee for Synthese suggested this point to me. A second referee conceives a different argument for Nelson. If I understand correctly, it goes like this: In ethics, it is often suggested that the seemingly positive duty to tell the truth is best understand negatively as the duty not to tell lies. The reason given is that it’s not obvious which truths I should tell, to whom and when. Analogously, one might argue that at least some of the things that are regularly construed as positive epistemic duties are better interpreted negatively. An alleged duty to believe the truth, for instance, doesn’t tell me which truths to believe, in which contexts and when. Perhaps our inability to answer such questions indicates a need to reformulate the duty in negative terms: ‘Avoid believing falsehoods.’ I admit that Nelson could mount such an argument (though he does not in fact do so). My response is that one could just as easily turn this around and say that it’s not clear which falsehoods one should avoid believing, in which contexts and when. And it doesn’t wash to say that one can avoid believing falsehoods by doing nothing, since beliefs arise (unlike lies) without needing to be willed. Even if I do nothing about my evidence, I will inevitably believe many falsehoods, and I shouldn’t. By contrast, I can avoid telling lies by choosing to say nothing. So the analogy between ethics and epistemology is not tight enough to give us a clear argument in favour of the negative construal of epistemic policy obligations, at least not on the grounds stated.
See Stapleford (2012) for such an argument.
Literally, this means ‘the broadly determining laws’, but Kant uses the phrase interchangeably with ‘imperfect duties’.
There is no decision procedure, in other words, for determining when I should reflect, how much I should reflect and on what.
Some of these points are adapted from Hill’s (1992) account of imperfect moral duties.
This way of formulating the point is taken from Susan Hale’s argument against morally supererogatory actions (Hale 1991, p. 273). Though Hale writes on ethics rather than epistemology, her comments on moral excuse making helped cement my own views on the significance of excuse making in an epistemic context.
‘I shouldn’t have drawn such a hasty conclusion’ might seem to suggest that there are negative epistemic policy obligations such as ‘Don’t draw hasty conclusions’. But at least some excuses are offered for failing to do something that it would have been epistemically appropriate to do. My sense is that most or all epistemic excuses are offered for failing to deliberate on one’s evidence. But I cannot support the claim here.
An anonymous referee suggests that I am not really excusing myself for failing to help the poor people in Africa since I have not violated any duty—I did what I could by giving to the Red Cross—and so I am not in any respect subject to blame. I agree that no duty has been violated in the imagined case, since I have already given amply to the Red Cross, but this does not imply that there is no moral shortfall. The reason that imperfect duties can never be fully discharged is that there is always some outstanding claim upon us to do more, though the claim is not strong enough to impose an additional duty to perform any particular action. Excuses are offered because we feel guilty for not doing more—we recognize the moral deficit—but they are accepted when it is believed that we have at least done enough to make the minimum moral payment. But I still feel guilty about the people in Africa. It would be hardhearted to say that there is no sense of ‘ought’ in which I ought to help the poor in Africa simply because I have done well by helping others in America. I have argued previously for a corresponding distinction in epistemology. There is a deflated sense of ‘ought’ in which I ought to believe every proposition that is supported by my evidence, but I still have no duty to believe any particular propositions. See note 2 above.
References
Feldman, R. (2004). Having evidence. In E. Conee & R. Feldman (Eds.), Evidentialism: Essays in epistemology (pp. 219–241). Oxford: Oxford University Press.
Hale, S. (1991). Against supererogation. American Philosophical Quarterly, 28, 273–285.
Hill, T. (1992). Dignity and practical reason in Kant’s moral theory. Ithaca: Cornell University Press.
Kant, I. (1996). In M. J. Gregor (Ed.), Practical philosophy. New York: Cambridge University Press.
Kant, I. (1997). In P. Heath & J. B. Schneewind (Eds.), Lectures on ethics. New York: Cambridge University Press.
Kim, K. (1994). The deontological conception of epistemic justification and doxastic voluntarism. Analysis, 54, 282–284.
Nelson, M. (2010). We have no positive epistemic duties. Mind, 119, 83–102.
Stapleford, S. (2012). Epistemic duties and failure to understand one’s evidence. Principia, 16, 147–177.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Stapleford, S. Imperfect epistemic duties and the justificational fecundity of evidence. Synthese 190, 4065–4075 (2013). https://doi.org/10.1007/s11229-013-0249-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11229-013-0249-5