Abstract
In this paper I consider metaphysical positions which I label as ‘perspectival’. A perspectivalist believes that some portion of reality cannot extend beyond what an appropriately characterised investigator or investigators can (in some sense) reveal about it. So a perspectivalist will be drawn to claim that a portion of reality is, in some sense, knowable. Many such positions appear to founder on the paradox of knowability. I aim to offer a solution to that paradox which can be adopted by any perspectivalist, which involves no restriction on the claim of knowability and which allows certain sentences to be unknowable. The solution hinges on recognising that what is meant by ‘knowable’ will vary from one type of proposition to another and thus that characterising the modality involved in the notion in terms of possible worlds will be impossible. I thus offer a subjunctive conditional reading of that modality, a reading which, I claim, has the virtues just recounted.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Here, for the sake of the example, I talk of decision procedures, which I take to be a subclass of investigative procedures; but, in many cases investigations will fail to be decision procedures: though nothing is an investigative procedure which cannot, in principle, be carried out, there may be no effective means for finding the appropriate investigative procedure in a given case. One may, for instance, claim that if Goldbach’s conjecture is true there is a proof of it, i.e., a suitable investigation which will reveal this truth, but there may be no effective means of finding this proof. Just so in empirical cases too.
Take it that a is the only exemplar of the property Fx ∧ Gx and that there is no means of referring to a other than by ‘a’, or a synonym (or that ‘a’ is a canonical name for a).
The additional premises shouldn’t worry us; too often we aim at refutation through absurdity, where implausibility will do.
Cf. Restall’s (2009, sect. 4) argument that without factivity the conjunctivist’s claim is trivial.
See Rosenkranz (2004) for similar concerns.
For more detailed discussion of the issues raised in this section see my 2007.
It might then simply be claimed that the fault in Fitch is thus that it makes use of an uninterpreted and perhaps uninterpretable notion of possibility. Though this may be a fair response it then becomes incumbent on the perspectivalist to explain the conception of possibility which figures in her favoured notion of knowability. The rest of this paper attends to that problem. (My thanks to a reviewer for this point.)
Cozzo (1994) chooses to understand truth in terms of the existence of an ideal argument. I am sympathetic; but prefer the subjunctive construal as being closer to one’s pre-theoretic intuitions and less committed on their theoretical development. My sense too is that if we are to understand the character of existence that applies to these abstract entities it will be via the truth of the relevant subjunctive conditionals.
The issue is too large to deal with adequately here but it is worth noting that (SK) does not make all the assumptions that spawn Plantinga’s objections to a subjunctive conditional analysis of truth (see Plantinga 1982, 64–66). There it is assumed that we have a claim of the form ‘p is true iff if Q conditions were to hold then p would be known’. Where ‘Q’ represents conditions of epistemic idealisation; crucially the argument assumes Q conditions are uniform for all p. For the argument works by substituting the proposition ‘We are in Q conditions’ in the clause governing truth. On my account this makes no definite sense until we’ve specified which proposition’s Q conditions are relevant, when we do so we fail to generate the required reflexivity. If we try to get round this by contemplating Q conditions relevant to all propositions then I would be happy to admit that these cannot be realised. So I and (SK) make no assumption about uniformity.
There’s certainly no implication that the account then rules out a possible worlds analysis of subjunctive conditionals.
Obviously the situations we contemplate will vary as we move from considering one proposition to another (see note 9, above).
Possible objection: take r = ¬q, then it seems that r is false just in case it cannot be investigated. But here we don’t have K(r) → ⊥. What needs to be shown is that the biconditional holds when the proposition cannot be known.
A minor problem is that one or other conjunct may itself be a problematic conjunct. But clearly we can run the same form of argument for any finite string of conjunctions.
The point is of no minor relevance. Restall may well have overlooked the sort of example we are considering because he supposes that the unknowablity of a true proposition is a matter of logic (and a priori reasoning): the conditional ‘if the proposition were true then it would be unknowable’ is to be established on a priori grounds. But I see no reason to accept this: contingencies may give rise to unknowably true propositions. Since, as remarked, we are considering possible investigations of the world as it is constituted, we cannot simply consider possible circumstances in which those contingencies fail in order to show the knowability of our proposition.
See Hand (2009) for discussion exploitation of this same distinction.
References
Cozzo C (1994) What can we learn from the paradox of knowability? Topoi 13:71–78
Dummett M (2001) Victor’s error. Analysis 61:1–2
Hand M (2009) Performance and paradox. Salerno 283–301
Plantinga A (1982) How to be an Anti-realist. Proc Addresses Am Philos Assoc 56(1):47–70
Restall G (2009) Not every truth can be known (as least, not all at once). Salerno 339–354
Rosenkranz S (2004) Fitch back in action again? Analysis 64:67–71
Salerno J (ed) (2009) New essays on the knowability paradox. Oxford University Press, Oxford
Tennant N (2002) Victor vanquished. Analysis 62:135–142
Weiss B (2007) Truth and the enigma of knowability. Dialectica 61(4):521–537
Acknowledgments
My thanks to audiences at the Universities of Witwatersrand and Cape Town and to the participants in the workshop on “Anti-realistic Notions of Truth”, Certosa di Pontignano, University of Siena, September 2010. Many thanks to Gabriele Usberti for organising and inviting me to the latter.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Weiss, B. Perspectives and the World. Topoi 31, 27–35 (2012). https://doi.org/10.1007/s11245-011-9112-9
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11245-011-9112-9