Reformed Philosophy

In the paper ‘Warrant for Nothing (and Foundations for Free?)’, Crispin Wright takes the proposition ‘our perceptual faculties are generally accurate’ to be a type of proposition that can be deemed entitled; conversely, the proposition ‘there is an external world’ is a type of proposition that he thinks cannot be deemed entitled since it fails to meet his second criterion. The second criterion reads:

(ii)  ‘The attempt to justify P would involve further presuppositions in turn of no more secure a prior standing…and so on without limit: so that someone pursuing the relevant enquiry who accepted that there is nevertheless an onus to justify P would implicitly undertake a commitment to an infinite regress of justificatory projects, each concerned to vindicate the presuppositions of its predecessor.’

According to Wright, the proposition ‘our perceptual faculties are generally accurate’ leads to an ‘infinite regress of justificatory projects, each concerned to vindicate the presuppositions of its predecessor’ once we accept that ‘there is…an onus to justify’ it; the proposition ‘there is an external world’, according to Wright, does not do this. Now, I didn’t find Wright to be terribly clear on this point. What is the difference between the two propositions? How does this infinite regress go precisely? Both make a statement about what ontology is correct (which is important to note considering Wright’s broader project in that paper), and if impressed with an onus to justify either one, I feel the same sense bemusement as to how I might do that. In my reflections, the only pertinent distinction between the two propositions is that I feel a slightly greater initial sense of despair when trying to justify the latter than the former. I cannot figure out how trying to justify ‘our perceptual faculties are generally accurate’ leads to a type infinite regress, that trying to justify ‘there is an external world’ does not lead to too. I am quite frankly at a loss.

If anyone has any ideas, I would love to hear (read) them!

‘Entitlement of cognitive project does not…extend to matters of ontology.’

                                                -Crispin Wright, ‘Warrant for Nothing (and Foundations for Free?)’ page 197

In the paper ‘Warrant for Nothing (and Foundations for Free?)’ Crispin Wright takes steps toward developing a theory of claimable non-evidential warrant for certain epistemically foundational (cornerstone) propositions, called entitlement of cognitive project. In this paper, I argue that the proposition ‘God exists’ is entitled if given the right kind of ontology. I do not do this out of a religious zeal to make Wright’s theory theistic, but instead to exercise a recognized flaw in the theory, to exercise the need for some sort of ontology filter.[1] Wright recognized the ontological limitations of entitlement of cognitive project, and as such, I think the proposal of this paper is quite modest; by the end, however, I will gesture toward some more serious concerns using the groundwork I have covered.

I argue for this thesis in three sections: First, I elucidate Wright’s criteria for determining what propositions are entitled under entitlement of cognitive project (this section will be the bulk of my paper). Secondly, I show that the proposition “God exists” meets these criteria when considered from an ontology found within the theology of the protestant Reformation, which, depending on one’s religious and theological belief may or may not be problematic for Wright’s theory in and of itself. In the concluding section, I briefly note why I think the groundwork I cover in this paper might be reason for further concern.

As a point of reference, I use throughout this paper the following propositions to both clarify Wright’s position and put forward my own argument:

1.      The rolling of dice can accurately predict the future

2.      Our perceptual faculties are generally accurate

3.      God exists

I will take 1 as a token proposition that Wright does not want to be deemed entitled; it is especially helpful in that it seems to fail all of Wright’s criteria. Conversely, 2 is a token proposition that Wright does want to be deemed entitled. As I already mentioned, 3 is a hinge proposition for a species of ‘Reformed’ ontology.

I

The typical skeptical argument works by calling into question the justification of key cornerstone propositions, which then, in turn, affects the justification of the host of other propositions that epistemically rely on and presuppose those cornerstones. For example, if I find that I don’t have justification for by belief in an external world (the cornerstone proposition), then my beliefs in the existence of my wife, my hands, and my computer may not have nearly as much justification as I would have supposed.  The problem is that as we look for justification for these cornerstone propositions we end up digging through presupposition after presupposition, none of which are any more epistemically secure, ad infinitum; there seems to be a need for a type of justification that is not accrued by justification transfer, in other words, justification that is not based on evidence (some call it warrant). Wright considers entitlement to be just this kind of justification.

            For Wright, ‘entitlement is rational trust.’[2] That is to say, entitlement is a doxastic attitude that is warrant conferring for a given proposition that meets Wright’s criteria for entitlement eligibility, which allows us to, as it were, ‘help ourselves to’ that proposition.[3] What are the criteria for discerning whether a given proposition is entitled? For any cornerstone or ‘hinge’ proposition, P, Wright lists two explicit criteria:

(i)         ‘We have no sufficient reason to believe that P is untrue’

(ii)       ‘The attempt to justify P would involve further presuppositions in turn of no more secure a prior standing…and so on without limit: so that someone pursuing the relevant enquiry who accepted that there is nevertheless an onus to justify P would implicitly undertake a commitment to an infinite regress of justificatory projects, each concerned to vindicate the presuppositions of its predecessor.’[4]

Given these criteria, the problem, now, is knowing exactly how these criteria are suppose to work.

The first thing to get clear, however, is that there is in actuality another criterion in addition to the two above that Wright explicitly chronicles, namely, the qualification of being a cornerstone proposition. The above two criteria are for any cornerstone proposition P, and as such being a cornerstone is a precondition of entitlement eligibility. To be sure (and to play it safe by generously limiting the kinds of propositions under consideration), let us say that not just any cornerstone proposition will do, but only the most cornerstone of the cornerstone propositions; that is, the cornerstones of rationality and of our most fundamental cognitive projects, as Patrice Philie puts it, the cornerstone propositions we are looking for are ‘presupposed in the game of giving and asking for reasons’.[5] Take the criterion to be: P is entitled if…

Cornerstone Criterion: P is a cornerstone proposition of ‘the game of giving and asking for reasons’.

According to the Cornerstone Criterion, the proposition ‘the rolling of dice can accurately predict the future’ is not entitlement eligible because there is no reason to think that such a proposition is needed for rationality to function. The general accuracy of our perceptual faculties, on the other hand (no Moorean pun intended), does seem to be important for ‘the game of giving and asking for reasons’. How could we play this game (with anyone other than ourselves) if all empirical data is untrustworthy?

            Now that the Cornerstone Criterion is clear, let us move on to the first of the more explicit of Wright’s criteria. How do we determine what counts as a ‘sufficient reason’ for believing a proposition is untrue? How is a sufficient reason for believing that a proposition is untrue different from a sufficient reason for simply doubting a proposition? The difference between doubting a proposition and believing it untrue is sharp and worth noting; after all, skeptical arguments seem to provide a reason to doubt certain propositions, though, if Wright’s criterion is to get off the ground, presumably not a reason to believe them untrue.[6] To answer these questions we should examine the work Wright wants these criteria to do and then try to glean from that how they are doing it. For example, Wright does not want us to have a sufficient reason to believe that the proposition ‘our perceptual faculties are generally accurate’ is untrue; there may be doubts concerning the accuracy of our perceptual faculties, but presumably not a reason to believe them untrue. On the other hand, Wright does want there to be a sufficient reason to believe the proposition that ‘the rolling of dice is an accurate way to predict the future’ is untrue and not merely a sufficient reason to doubt it. What is doing the work in the criterion ‘We have no sufficient reason to believe that P is untrue’ that produces these results?

            Criterion 1 Proposal 1: Perhaps what determines whether we have a sufficient reason to believe a given proposition is untrue is how truth-conducing it seems to be.[7] The accuracy of our perceptual faculties do seem to be generally truth-conducing, while, in contrast, the accuracy of predicting the future with dice seems very far from truth-conducing. Unfortunately, this reading of Wright’s first (explicit) criterion is far from charitable; it seems to run into an immediate problem with arbitrariness. If our perceptual faculties have a truth-conduciveness level of 99 out of 100 and the truth-conduciveness of rolling dice to predict the future is 1 out of 100, then at what truth-conduciveness level do we cross over from sufficient reason to doubt to sufficient reason to believe untrue? There doesn’t seem to be an easy/non-arbitrary answer.

            Criterion 1 Proposal 2: Perhaps, instead, what determines whether we have a sufficient reason to believe a given proposition is untrue is what kind of defeaters, if any, are against it. There are many ways of discerning and labeling the different sorts of defeaters, but let us focus on those in the following two classes:

Counter Defeater: Provides justification/warrant contrary to a proposition.

Undermining Defeater: Blocks the justification/warrant in favor of a proposition.

For example, say that Jones has accused Smith of committing a nationally publicized murder of a high-profile celebrity in the center of an amusement park maze. A Counter Defeater for the proposition ‘Smith is the murderer’ would be to point out that Smith has a profoundly awful sense of direction, thus providing a reason against thinking Smith is the maze murderer. An Undermining Defeater for Jones’ accusation would be to point out that Jones is a schizophrenic and ‘Smith’ is an illusory manifestation of Jones’ cognitive malfunction. If Smith does not exist, that undermines any justification/warrant in favor of the accusation that he is the maze murderer.

Might it be that Counter Defeaters merely provide reasons to doubt propositions, but Undermining Defeaters provide reason to believe propositions untrue? I think something like this is at work in Wright’s first criterion. We have a sufficient reason to believe the dice proposition is untrue because we can undermine it by simply showing that it is empirically false (since the dice proposition would hypothetically be based on empiricism it is undermined by pointing out that there is no empirical support). On the other hand, there does not seem to be an Undermining Defeater against our perceptual faculties’ accuracy. As Wright points out early-on in ‘Warrant for Nothing (and Foundations for Free?)’, skepticism shows us we don’t have evidential-justification for our cornerstone propositions, but it does not show us that we don’t have any non-evidential justification (warrant), which is the lacuna/gap in which Wright finds room for his entitlement thesis to begin with. Skepticism, it seems, at best can only provide a Counter Defeater against the accuracy of our perceptual faculties. Not only does this conception of Wright’s first (explicit) criteria do what he needs it to do without being internally problematic (like my first proposal), but it seems to fit the structure of the rest of ‘Warrant for Nothing (and Foundations for Free?)’. To be clear, here is the revised version of Wright’s first criterion for determining entitlement eligibility: For any cornerstone proposition P, P is entitlement eligible if…

            Revised First Criterion: We have no Undermining Defeater for P

This revision of the first criterion, however, leads to the result that if a proposition, P, has a host of Counter Defeaters against it but no Undermining Defeaters, then it looks like we can only ever merely doubt P. While we may have intuitions that we should be able to believe propositions that seem terribly un-truth-conducive untrue and not merely doubt them, I don’t think we want such intuitions to be doing the main work in Wright’s first criteria lest we fall back into the problems illustrated in my first proposal.  

For reference, here once more is the second criterion:

(ii)        ‘The attempt to justify P would involve further presuppositions in turn of no more secure a prior standing…and so on without limit: so that someone pursuing the relevant enquiry who accepted that there is nevertheless an onus to justify P would implicitly undertake a commitment to an infinite regress of justificatory projects, each concerned to vindicate the presuppositions of its predecessor.’[8]

Again, it seems like the best way to understand how this criterion works is to plug in our test propositions, but first there is an ambiguity that needs to be cleared up. While the attempt to justify an entitlement eligible proposition will ‘involve further presuppositions… of no more secure a prior standing’, we need to know how the term ‘prior’ should be understood. This is a philosophically substantial question that immediately threatens to side-track my thesis, so without getting bogged down in detail I will simply gesture toward the appropriate clarification.

There seems to be two main ways to assess priority in this case: temporally or epistemically. Surely, however, Wright, with his talk of presuppositions and infinite regresses, must mean something like epistemically prior. Besides, would not a temporal understanding of ‘prior’ for this second criterion epistemically damn any unfortunate chap who happened to be raised by a couple of radical skeptics such that some of his very first beliefs were ‘there is no material world’ and so forth? Leaving aside issues of what the justifying relationship looks like or of what it means for one propositions to ‘rest’ on another, let us simply take it that Wright has some idea of epistemic priority in mind for this criterion.

With that in mind, if we plug in our test propositions, the rest of the criterion functions rather straightforwardly. [9] What happens when we attempt to justify the proposition ‘our perceptual faculties are generally accurate’? We tend to go around in circles; there doesn’t seem to be a good way to offer justification for it without begging the question. What happens, however, when we attempt to justify the proposition ‘the rolling of dice can accurately predict the future’? If someone were to try to justify such a proposition, they would likely turn to empirical observation and the uniformity of nature to vindicate the accuracy of the dice’s predictions, but propositions like ‘there is a uniformity to nature’ and ‘empirical observation is a generally good way of finding truths’ must surely be both epistemically prior and more secure than the dice proposition. So let us now note our (slightly) revised second criterion: For any cornerstone proposition P, P is entitled if…

(Slightly) Revised Second Criterion: The attempt to justify P would involve further presuppositions in turn of no more secure an epistemically prior standing…and so on without limit: so that someone pursuing the relevant enquiry who accepted that there is nevertheless an onus to justify P would implicitly undertake a commitment to an infinite regress of justificatory projects, each concerned to vindicate the presuppositions of its predecessor.

 II

In the sixth section of ‘Warrant for Nothing (and Foundations for Free?)’, Wright notes a concern that people could postulate different ‘tracts of reality’ such that would affect the results of his entitlement criteria. It is this concern, as Wright is aware, that leaves entitlement of cognitive project susceptible to some kinds of skepticism, and it is this exact concern that I now intend to exercise. 

Now that we have fleshed out what I take to be Wright’s criteria for determining what propositions are entitled under entitlement of cognitive project, I will put forward the proposition ‘God exists’ for consideration. This is not because I am trying to push a ‘fundamentalist’ agenda or even because I am trying to make Wright’s theory theistic. Instead, I am using the proposition ‘God exists’ because whether or not it is entitled will depend on the ontology, a procedure which highlights what I take to be the Achilles’ heel of Wright’s theory. To make this point, I will in this section very briefly look at the proposition ‘God exists’ from a Reformed ontology (more specifically, the conception of Reformed ontology according to the philosophical theologian Cornelius Van Til), and then move from there to briefly sketch more problematic cases in the final section.

To roughly describe the position, Van Til saw belief in God as epistemically basic. That is to say, Van Til saw the ontology of God’s existence as of primary epistemic importance for all facts, both a priori and empirical, such that theistic belief is needed if we are to prove anything.[10] Indeed, Van Til saw the connection between theistic belief and rationality to be so strong that he said:

‘Christianity can be shown to be, not ‘just as good as’ or even ‘better than’ the non-Christian position, but the only position that does not make nonsense of the human experience.’ [11]

This view of ontology, though rarely held, seems to be internally coherent, and the nature of logic, rationality, mathematics, etc. is not so established as to bar its consideration. Since entitlement of cognitive project does not pick out a specific ontology, I see no reason why this Reformed view could not be hypothetically considered using Wright’s theory.

Assuming Van Til’s ontology is at least internally viable, let us proceed and see if the proposition ‘God exists’ passes what I am taking as Wright’s criteria: P is entitled if…

Cornerstone Criterion: P is a cornerstone proposition of ‘the game of giving and asking for reasons’.

Revised First Criterion: We have no Undermining Defeater for P

(Slightly) Revised Second Criterion: The attempt to justify P would involve further presuppositions in turn of no more secure an epistemically prior standing…and so on without limit: so that someone pursuing the relevant enquiry who accepted that there is nevertheless an onus to justify P would implicitly undertake a commitment to an infinite regress of justificatory projects, each concerned to vindicate the presuppositions of its predecessor.

It is worth noting that from a more naturalistic ontology, the proposition “God exists” does not even get off the ground in terms of entitlement eligibility, since conceivably it would not be a cornerstone of rationality. From Van Til’s Reformed ontology, however, ‘God exists’ is indeed a cornerstone of rationality, a cornerstone of ‘the game of giving and asking for reasons’. In addition, there does not seem to be any Undermining Defeaters against it; how could there be if belief in God is epistemically foundational for Reformed epistemology? Finally, what would happen if we were to ‘attempt to justify’ the proposition ‘God exists’? From Van Til’s perspective, we would simply run in circles, because we presuppose the supremacy of the presupposition ‘God exists’ at the outset.[12] For Van Til, all proof for the existence of God (like the teleological argument) and all proof against the existence of God (the argument from the problem of evil), at base must presuppose the truth of the proposition ‘God exists’. Without an outside reason to reject Van Til’s ontology, the proposition ‘God exists’ is eligible for entitlement of cognitive project according to what I understand to be Wright’s theory.[13]

III

If my thinking thus far is correct, does this mean Wright’s theory supports Christianity or at least theistic belief? I don’t think so. Wright understood that his theory lacked a certain ontological filter; all I have done so far is exploit that weakness. The mere fact that the proposition ‘God exists’ is deemed entitled according to Wright’s theory when viewed from Van Til’s ontology is not necessary a strike against it. Some people would welcome this result. Unfortunately, however, at this point we may have reason to believe that the case of God’s existence being deemed entitled is simply highlighting the insufficiency of Wright’s criteria alone.

            The more basic concern would be something like this: Take the basic structure of Van Til’s ontology, if we substitute the terminology of ‘God’ with ‘The Flying Spaghetti Monster’ or ‘The Great Pumpkin’ or, even worse, ‘The Evil Demon (of the Cartesian sort)’, then might it be the case that such ‘x exists’ propositions are entitled according to Wright’s criteria if the right ontology is presupposed? Given, this move is not straightforward; there are metaphysical and epistemic intricacies that need to be worked out. However, people have indeed argued that Wright’s theory is still (despite what Wright thinks) susceptible to Cartesian skepticism and relativism, so this concern might very well be a legitimate one.[14] Without an expressed means of picking one ontology over another, there may not be any way to withhold entitlement status from such odd and wholly problematic propositions. As they stand, Wright’s criteria are at best helpful only for elucidating what propositions are entitled in each ontological theory, but this may have limited epistemic value if there are no means to champion an ontology that advocates the proposition ‘our perceptual faculties are generally accurate’ over an ontology that advocates the existence of evil demons.

***

            In this paper, I argued that the proposition ‘God exists’ is entitled according to Wright’s theory of entitlement of cognitive project, as long as the proposition is viewed from Van Til’s ontology. Since Wright’s theory fails to pick out a specific ontology, there is no (immediate) basis to reject Van Til’s as a viable option, assuming it is internally coherent. This, however, is of little ultimate affect since it is a mere exploitation of a known weakness of Wright’s theory; yet, in exercising this weakness we saw how further and more serious problems may arise.

References

Banhsen, Greg L. Van Til’s Apologetic. New Jersey, USA: Presbyterian and Reformed, 1998.

Davies, Martin. ‘Epistemic Entitlement, Warrant Transmission, and Easy Knowledge’.

Aristotelian Society Supplementary Volume. Vol. 78, 2004, pp. 213-245.

Jenkins, C.S. ‘Entitlement and Rationality’. Synthese. Vol. 157, 2007, pp. 25-45

Philie, Patrice. ‘Entitlement as a Response to I-II-III Scepticism’. Synthese. 19 February,

2008. 11 March, 2008 < http://www.springerlink.com/content/3660588122r6k054/>

Prichard, D. ‘Wittgenstein’s On Certainty and Contemporary Anti-Scepticism’. Readings of

Wittgenstein’s On Certainty. Eds. Moyal-Shamrock, D. and Brenner, W.H. London, UK: Palgrave Macmillion, 2005, 189-224.

Van Til, Cornelius. Christian Apologetics. New Jersey, USA: Presbyterian and Reformed,

2003.

Van Til, Cornelius. A Christian Theory of Knowledge. New Jersey, USA: Presbyterian and

Reformed, 1969.

Wright, Crispin. ‘Warrant for Nothing (and Foundations for Free?)’. Aristotelian Society

Supplementary Volume. Vol. 78, 2004, pp. 167-212.