Non-ideal Epistemology

In this post, Robin McKenna presents a new book, Non-Ideal Epistemology (Oxford University Press 2023).

I started thinking about the ideas that became Non-Ideal Epistemology when I was teaching social epistemology for the first time. I wanted to cover more than just the epistemologies of testimony and disagreement. I also wanted to do more than simply finish with a unit on epistemic injustice. 

I wanted to cover the bits of social epistemology that overlap with philosophy of science and political philosophy, the bits that engaged with social psychology, the bits that asked political questions about knowledge and knowledge production. I wanted to help students identify and question the assumptions underlying the kind of approach to social epistemology that is more interested in how things might work than in how they actually work.

But I didn’t want to do too much. I didn’t want to lose the students. A grand narrative was needed. But what would this grand narrative look like? What unifies the diverse and disparate questions, approaches, and ideas that constitute “social epistemology”?

The answer I came up with was: nothing, really. There is a sharp divide between idealised and non-idealised approaches to social epistemology. Idealised approaches work with idealised models of human beings, the social interactions between them, and the social spaces in which they interact. Non-idealised approaches work with more realistic models—models that recognise the ways in which human beings depart from these idealised models and are interested in the implications this has for social epistemology.

One of the central claims of my book is that, in many ways, this distinction—the distinction between ideal and non-ideal epistemology—is more useful than the distinction between “traditional” and social epistemology. The ideal epistemologist usually works within the same “problem space” as the traditional epistemologist but has a few more tools. The non-ideal epistemologist is interested in a different problem space and develops new tools with which to tackle it.

Robin McKenna

The first part of the book (Chapters 1-3) develops the distinction between ideal and non-ideal epistemology using the debate between ideal and non-ideal theory in political epistemology. Ideal epistemology is (roughly) analogous to Rawlsian ideal theory—it is interested in ideals (e.g. knowledge) and its normative claims are conditional on idealistic assumptions (e.g. of universal norm-compliance). Non-ideal epistemology is (equally roughly) analogous to non-ideal theory—it is less interested in ideals and its normative claims are based on more realistic assumptions.

The second part (Chapters 4-8) has two aims. The first is to demonstrate the value of doing non-ideal epistemology. I take up several important questions, such as: How should we respond to science denialism? Should we try to be intellectually autonomous? When do we need to engage with challenges to our views? Are we responsible for our intellectual characters? How many of us have justified beliefs about controversial matters? While I think there is something to be said for the answers I give to these questions, my overarching aim is to demonstrate the role that non-ideal epistemology can play in addressing them. 

The second aim is to highlight some problems with ideal epistemology that (roughly) parallel the problems Charles Mills identified for ideal theory in political philosophy. Ideal theory only has implications for what we, in a non-ideal world, should do if the assumptions on which it is based aren’t too distorting or idealistic. While it is perfectly possible to do ideal theory while recognising that the underlying assumptions are unrealistic, there is an unfortunate tendency to downplay how unrealistic the underlying assumptions really are. As a result, ideal theorists, whether in epistemology or politics, tend to exaggerate the extent to which they provide any sort of normative guidance for humans.