Doctoral Seminar

Speaker:
Filip Buekens (KU Leuven)

Abstract

Philosophers tend to identify truth and accuracy: a belief or assertion is accurate if and only if what is believed or asserted – its content, with a propositional format -- is true (THISHOR 2017). After discussing some obvious counterexamples where both notions come apart and do not even entail each other (accurate representations can be false, true statements can be inaccurate) I’ll make a case for separating both notions on more principled grounds. I propose a suitable precisification of the concept of an accurate (re)presentation of a subject matter or target, how accuracy conditions of a representation differ from truth conditions of propositions, and why the distinction is important for understanding acts of representing a subject matter or target system. The background theory of truth I employ is a minimalist conception of truth (which may itself be considered a precisified concept derived from our ordinary concept of truth). The distinction I propose was already hinted at  in Frege’s Der Gedanke. While truth is a thin inferential property ascribed to propositions, accuracy is a thick, evaluative property of representations of a subject matter or target system and, indirectly, of acts of representing and of agents who represent a subject matter. A minimalist conception of truth allows us to separate the non-normative (i.e. descriptive), non-evaluative and non-relative concept of truth from the standard-sensitive, evaluative and assessment-relative notion of accuracy. Accuracy pertains to ways things are represented, truth is what you have when a possibility obtains or is the case When evaluating assertions, our eye can be on the truth or truth/falsity of the proposition expressed, or its (in)accuracy, i.e. a quality of its way of presenting a subject matter. Unlike (the concept of) truth, the concept of an accurate representation is refinable in many directions and for a wide range of purposes.

Two upshots: first, a technical point. Our distinction allows us to explain away intuitions about the alleged gradability of the truth-predicate (Henderson 2021, Egré 2021, but see Mankowitz 2023 for critical remarks) as genuine features of what it takes for a (re)presentation to be (more, or less, or completely) accurate.  Secondly, and more generally: propositions (the primary bearers of truth and falsity) don’t have a representational nature. They are possibilities (Hunter 2022). Representations, on the other hand, are epistemic tools designed to engender beliefs about their subject matter. The distinction between truth and accuracy helps us separating the deeply relative, purpose-dependent and perspective-sensitive character of the accuracy of a representation of a subject matter from truths and falsehoods that can be gained by interpreting representations.

References

Buekens, F., & Truyen, F. (2014). The Truth about Accuracy. In C. Martini e. a. (Eds.), Experts And Consensus In Social Science, pp. 213–229). Springer; Heidelberg New York Dordrecht London.

Egré, P. (2021). Half-truths and the liar. In Modes of truth (pp. 18-40). London:  Routledge.

Henderson, J. (2021). Truth and gradability. Journal of Philosophical Logic, 50(4), 755-779.

Hunter, David (2022). On Believing: being right in a world of possibilities. Oxford: OUP

Mankowitz, Poppy (2023). "Not half true." Mind 132, 84-112.