1. The argument of this paper may also be seen as a response to Sudduth, Michael ‘Reformed epistemology and Christian apologetics’, Religious Studies, 39 (2003), 299–321CrossRefGoogle Scholar.

2. The legendary exchange between Barth and Brunner is paradigmatic; Karl Barth Nein! Antwort an Emil Brunner (Munich: Chr. Kaiser Verlag, 1934).

3. See Alvin Plantinga Warranted Christian Belief (Oxford: Oxford University Press, 2000), [hereafter WCB], 67–107, 167–198; William Alston Perceiving God: The Epistemology of Religious Experience (Ithaca NY: Cornell University Press, 1991); and, Wolterstorff, Nicholas ‘Can belief in God be rational if it has no foundations?’, in Alvin Plantinga & Nicholas Wolterstorff (eds) Faith and Rationality: Reason and Belief in God (Notre Dame IN: University of Notre Dame Press, 1983), 135–186Google Scholar.

4. This does not imply that Christian faith is reducible to belief, nor am I suggesting that a minimum threshold of cognitive function is required for Christian faith to take root in a person. I employ the term ‘Christian faith’ very specifically to indicate the gift of our human reception of the knowledge of God which comes as relationship with God is established by God. Faith in this context is not a synonym for ‘trust’ or ‘psychological conviction’.

5. Though we can say that an explicit commitment to a doctrine of the Trinity, for example, would not be a claim specified by essential Christian faith, since essential Christian faith was possible prior to explicit expression of such a doctrine.

6. Barth argued, furthermore, that recognizing human dependency on divine self-revealing for the formation of Christian belief is itself an affirmation that arises retrospectively out of properly formed Christian belief. In Barth's view, there are no first principles one must establish or grasp. God has taken the initiating action, such that ‘already on the way, we give an account of the way which we tread’; Geoffrey Bromiley & T. F. Torrance (eds) Church Dogmatics (Edinburgh: T & T Clark, 1975), I, 1, 43.

7. Christian theology confesses that this is the gift of the triune God – by the Spirit, ministering God the Son, who through the Incarnation reveals God the Father. See Ephesians, 2.18, which informs the following statement from the 1992 Orthodox-Reformed Agreed Statement on the Holy Trinity: ‘The priority of the Father does not imply that there is something more in him compared to the Son, for all that the Father is the Son is apart from “Fatherhood”, and likewise all that the Son is the Spirit is apart from “Sonship”. Thus, the order inherent in the Trinitarian relations is grounded on the fact that the Son is begotten of the Father and the Spirit proceeds from the Father. This applies also to the unique revelation of the Father through the Incarnation of his only begotten Son and the sending of the Holy Spirit by the Father in the name of the Son’; T. F. Torrance (ed.) Theological Dialogue Between Orthodox and Reformed Churches, II (Edinburgh: Scottish Academic Press, 1993), 222.

8. Similarly, from Plantinga: ‘Only God bestows saving faith, of course, but his way of doing so can certainly involve cooperation with his children, as in preaching and even argumentation’; Plantinga, Alvin ‘Christian philosophy at the end of the twentieth century’, in Griffioen, Sander & Balk, Bert M. (eds) Christian Philosophy at the Close of the Twentieth Century: Assessment and Perspective (Kampen: Uitgeverij Kok, 1995), 40Google Scholar.

9. We could extend this claim to affirm, following Alan Torrance, that: ‘it is a fundamental claim of Christian orthodoxy that the conditions for the recognition and affirmation of Christian truth are intrinsic to the event of revelation’; Torrance, Alan ‘Auditus fidei: Where and how does God speak?’, in Griffiths, Paul J. & Hütter, Reinhard (eds) Reason and the Reasons of Faith (London: T & T Clark, 2005), 28.Google Scholar

10. James, 2.19 seems to indicate that even demons have some true beliefs about God.

11. This would not be the case on the view that demonic corruption is isolated to the will.

12. Knowledge of the real referent, God, is enabled not by an impersonal supernatural infusion of information, but is a function of our being brought into communion with God in Christ, the ‘one true human who is both the incarnate Logos and the eschatos Adam’. By the Spirit, humans may participate in Jesus Christ's perfect human knowing of the Father. See Torrance ‘Auditus fidei’, 50–51.

13. Some would refer to the perfected condition not as faith but as sight (see Aquinas Summa Theologica, I.12 and III.9). As I note above, I employ the term ‘Christian faith’ as the gift of our human reception of the knowledge of God. Both the knowledge and the reception are a gift that grows with the transformation that comes as communion with God increases. On this view, the perfected condition is a matter of degree. Fullness in faith is fullness in sight.

14. Just as we cannot countenance a docetic Christology with its denial of the reality and significance of Christ's material body in which the mediation of our salvation is accomplished, so also we cannot allow a trivialization of the creatureliness of those witnesses by which the mediation of revelation is accomplished.

15. On Alvin Plantinga's model, for instance, both human external and internal rationality are involved in the formation of Christian belief; WCB, 110–112, 255–257.

16. This requires no subjugation of revelation to human standards for knowledge. It is instead an expression of the conviction that God's design for human knowledge includes His design for His own self-disclosure to human beings.

17. ‘Warrant’, as Plantinga indicates, is that which distinguishes knowledge from mere true belief. Warrant is that which provides the truth-motivated connection between reality and belief, giving the belief what Chisholm calls ‘positive epistemic status’; see Plantinga, AlvinWarrant: The Current Debate (Oxford: Oxford University Press, 1993), 43–46CrossRefGoogle Scholar; and Chisholm, Roderick M.The Foundations of Knowing (Brighton: Harvester Press, 1982), 94Google Scholar.

18. Consider Duncan Pritchard's formulation of a principle of epistemic safety intended to rule out luck: ‘For all agents … if an agent knows a contingent proposition ϕ, then, in most nearby possible worlds in which she forms her belief about ϕ in the same way as she forms her belief in the actual world, that agent only believes that ϕ when ϕ is true’; Pritchard, DuncanEpistemic Luck (Oxford: Oxford University Press, 2005), 156CrossRefGoogle Scholar.

19. Some construals of justification fall short of ensuring that belief is truth-grounded, as Gettier famously shows; Gettier, Edmund ‘Is justified true belief knowledge?’ Analysis, 23 (1963), 121–123CrossRefGoogle Scholar.

20. Ideally Christian belief would be held with maximal psychological and epistemic confidence. For more on the distinction between psychological and epistemic confidence, see Yandell, KeithThe Epistemology of Religious Experience (Cambridge: Cambridge University Press, 1993), 120Google Scholar, 169.

21. See Alvin Plantinga ‘Reason and belief in God’, in Plantinga & Wolterstorff Faith and Rationality, 50.

22. Many other beliefs may be equally as basic as Christian belief, but Christian belief has a uniquely maximal impact on the conceptuality informing all other beliefs.

23. A complete defence of this point will need to wait for another time. James Beilby argues, unconvincingly I think, that ‘a belief can be based at least partially on propositional evidence and arguments, and thereby fail to be properly basic, but still have a maximal depth of ingression’: James Beilby Epistemology as Theology: An Evaluation of Alvin Plantinga's Religious Epistemology (Aldershot & Burlington VT: Ashgate, 2005), 197. It seems impossible that ulterior inferential support could enhance the strength of a belief with these maximal credentials. If this were so, then non-basic beliefs would be stronger than the basic beliefs from which they derive their strength by inference. In fact, in the process of inference, to the degree that some probability calculus is in operation, non-basic beliefs will be weaker than the beliefs supporting them. Plantinga takes this further in noting that arguments from historical evidence depend on a chain of ‘dwindling’ probabilities (WCB, 271–280). Plantinga also argues that if Christian beliefs can be properly basic, then the ‘most satisfactory way to hold them will not be as the conclusions of argument’ (WCB, 210). Plantinga seems to agree with Calvin's view that, as Plantinga puts it, ‘the Christian ought not to believe on the basis of argument; if he does, his faith is likely to be unstable and wavering’; Alvin Plantinga ‘The Reformed objection to natural theology’, in Philosophical Knowledge (Washington DC: American Catholic Philosophical Assoc., 1980), 53.

24. A great deal could be said on the subject of defeasibility which is not required for our general endorsement of a commonly held no-defeater condition. See Bergmann, Michael ‘Internalism, externalism and the no-defeater condition’, Synthèse, 110 (1997), 399–417, 407CrossRefGoogle Scholar; idem ‘Defeaters and higher-level requirements’, The Philosophical Quarterly, 55 (2005), 419–436CrossRefGoogle Scholar, 420; and Sudduth ‘Reformed epistemology and Christian apologetics’, 309.

25. Plantinga locates the concern for defeaters in the domain of internal rationality (WCB, 112). Plantinga also distinguishes rationality defeat from warrant defeat. Warrant defeat involves a warrant-impeding failure of some kind in one's noetic processes, and may have nothing to do with beliefs in conflict (ibid., 359–360).

26. Bergmann ‘Defeaters and higher-level requirements’, 422.

27. A certain kind of experience might also qualify as a defeater for a particular belief, even if it fails to issue in a defeater belief. Our concern here is more narrowly with defeater beliefs. See John Pollock ‘The structure of epistemic justification’, American Philosophical Quarterly monograph series 4 (1970); and John L. Pollock & Joseph Cruz Contemporary Theories of Knowledge, 2nd edn, Studies in Epistemology and Cognitive Theory (Lanham MD: Rowman & Littlefield, 1999), 195–202. See also Plantinga Warrant: The Current Debate, 216–221.

28. Of course, one always has the option of attempting to maintain two beliefs in conflict by introducing a third belief – that the conflict between the first two beliefs must only be apparent and not real. Maintaining this belief, however, is only possible if one does not acquire a defeater for that belief (e.g. the belief that the conflict is indeed real).

29. We state the no-defeater condition formally as a condition for knowledge, not merely rationality, on the assumption that rationality is a necessary condition for knowledge.

30. Plantinga, Alvin ‘The foundations of theism: a reply’, Faith and Philosophy, 3 (1986), 298–313CrossRefGoogle Scholar, 311–312. Philip Quinn, to whom Plantinga is replying, accepts the possibility of intrinsic defeater-defeaters, but doubts that typical believers have enough warrant for their theistic beliefs to operate as intrinsic defeater-defeaters for all the potential defeaters typically encountered; see Quinn, Philip ‘The foundations of theism again: a rejoinder to Plantinga’, in Zagzebski, Linda (ed.) Rational Faith: Catholic Responses to Reformed Epistemology (Notre Dame IN: University of Notre Dame Press, 1993), 39Google Scholar. Similarly, see Sudduth, Michael ‘The internalist character and evidentialist implications of Plantingian defeaters’, International Journal for Philosophy of Religion, 45 (1999), 167–187CrossRefGoogle Scholar, 181–182. For cases where a defeater is never actually acquired, only repelled, Plantinga has introduced the term defeater-deflector as a more felicitous descriptor; Plantinga, Alvin ‘Reply to Beilby's cohorts’, in Beilby, James (ed.) Naturalism Defeated? Essays on Plantinga's Evolutionary Argument Against Naturalism (Ithaca NY: Cornell University Press, 2002), 224Google Scholar.

31. In G. E. Moore's response to Hume on scepticism about the existence of material objects, the move Moore makes has become known as Moore's shift. Rather than taking on the validity of the argument or the truth of the premises, one may simply have a more firmly held belief that the conclusion is not true. In this case then the refutation of the objection is just to restate the objection as a reductio ad absurdum. See G. E. Moore's lecture delivered in 1911 at Morley College in London, ‘Hume's theory examined’, in idem Some Main Problems of Philosophy (New York NY: Humanities Press, 1953), 108–126. Or, as Plantinga suggests, if your belief that God exists is firmer than your belief in the premises of an argument against that belief, ‘you would go modus tollens and take it that you had an argument against the premises’; from an unpublished letter to Clark, Kelly J., printed in his ‘Plantinga vs. Oliphint: and the winner is …’, Calvin Theological Journal, 33 (1998), 160–169, 164nGoogle Scholar.

32. James Beilby seems to assume that any defeater-defeater constitutes additional ‘propositional evidence’ which then becomes part of the basis for continuing in the original belief; Beilby Epistemology as Theology, 197.

33. There is of course the possibility that an undercutting or rebutting defeater-defeater belief might both defeat a defeater and contribute to the warrant for x. One example might be an argument in the form of disjunctive syllogism, where the defeater-defeater ‘not-n’ serves as a premise along with ‘x or n’ in an argument concluding ‘therefore, x’. This does not, however, change the point that defeater-defeater beliefs need make no positive contribution to warrant in order to restore the warranted status of x.

34. ‘[F]for the beliefs of faith: you don‘t have either sensuous imagery or evidence from other things you believe to go on; the beliefs are none the worse, epistemically speaking, for that. In fact (on the model) they are all the better for that; they have (or can have) much more firmness and stability than they could sensibly have if accepted on the basis of rational argument or, as in this case, historical investigation; they can also have much more warrant. These beliefs (on the model) are not accepted on the basis of other beliefs; in fact, other beliefs are accepted on the basis of them’ (WCB, 264).

35. Despite this, one commonly recurring critique of Plantinga's work is that he unduly minimizes the importance of argument and evidence for Christian belief. Three collections of essays are worth highlighting, not least because they each contain helpful responses from Plantinga himself. The first collection addresses Plantinga's general theory of warrant, pre-dating WCB: Jonathan L. Kvanvig (ed.) Warrant in Contemporary Epistemology: Essays in Honor of Plantinga's Theory of Knowledge (Lanham MD: Rowman & Littlefield, 1996). The other two collections provide interaction with Plantinga's proposals in WCB: ‘Book symposium on Warranted Christian Belief’, Philosophia Christi, 3 (2001), 327–400Google Scholar; and, ‘Book symposium: Warranted Christian Belief,’ Philosophical Books, 43 (2002), 81–135CrossRefGoogle Scholar.

36. Unlike perception in its phenomenal imagery, but analogically related to perception in being a direct apprehension (WCB, 181, 286–289).

37. Ibid., 259, 262.

38. This position is strengthened further by Plantinga's treatment of the affective aspect of the gift of faith. Here it is not merely the truth of the proposition that is made evident, but also its loveliness, beauty, and desirability. This affective component is also more satisfactorily arrived at in the immediate, basic way, rather than on the basis of an argument. See ibid., 304–306.

39. As Sudduth notes, even Reformed epistemology with its aversion to natural theology has a high regard for the contribution of good arguments to the Christian apologetical task; Sudduth ‘Reformed epistemology and Christian apologetics’, 299–321.

40. Plantinga ‘Christian philosophy at the end of the twentieth century’, 39–40; idem ‘Internalism, externalism, defeaters and arguments for Christian belief,’ 385; and, idem ‘Reply’, Philosophical Books, 43 (2002), 127Google Scholar, n. 4. In Plantinga's view, however, while such arguments could play a supporting and secondary role, they would in no way be sufficient on their own to deliver the kind of warrant necessary for knowledge and ‘paradigmatic’ Christian belief. An argument from public evidence might show that Christian belief is somewhat more probable than not, but this ‘is insufficient for its being warrantedly believed with any degree of firmness’; Plantinga, Alvin ‘Rationality and public evidence: a reply to Richard Swinburne’, Religious Studies, 37 (2001), 215–222CrossRefGoogle Scholar, 220; see also WCB, 271, 274, 280, 379. Plantinga's argument is that Christian belief can be properly held on grounds that are in fact firmer than the insufficient, though not insignificant, warrant delivered by propositional or historical evidence.

41. Plantinga ‘Reply’, 128. Plantinga also suggests that if one already, by means of the IIHS, has warrant sufficient for knowledge, upon learning about the historical evidence for the resurrection, it is possible to ‘gain an excess of warrant’ for Christian belief; ibid., 127.

42. It should be clear that the analysis to follow is not a theological argument; instead, I will be adopting an inductive approach from instances of ordinary belief in support of a constructive proposal.

43. Note that I take the faculty of sense perception to be more than the presentation of strictly empirical data. Sense perception also includes the learned apprehension of the forms of that data.

44. Perhaps a Gettier-type objection could be raised to the CW₂ construal, if it is possible, as I have suggested, that a mistaken assumption could motivate a fortuitous result. This would only be an effective objection, however, if the belief in question continued to depend on the mistaken assumption to deliver warrant sufficient for knowledge. On the CW₂ account above, however, warrant eventually swings free of the mistake and depends on the faculty of perception alone.

45. If we follow Thomas Reid's thought on testimony and credulity, believing testimony involves no inference. Credibility is assumed, though it can be defeated by discrediting the witness. Credulity is an innate gift and a natural faculty – ‘a disposition to confide in the veracity of others and to believe what they tell us’; Dugald Stewart & G. N. Wright (eds) Essays on the Active Powers of the Human Mind: An Inquiry into the Human Mind on the Principles of Common Sense (London: T. Tegg, 1843), 567.

46. Why would it matter on this view that the argument involved is a good argument – especially if the argument need not finally contribute to the warrant for belief? We need to make the same qualification here that we made earlier about testimony. This is not an occasionalist view. The usefulness of an argument as a catalyst in the service of the Spirit's work is not arbitrary; it depends on the strength of the argument. This is not to say that God cannot use weak arguments to lead someone to the truth, but the positive participation of a weak argument in that work would be proportionally diminished.

47. While Plantinga is open to historical evidence supplementing warrant from the IIHS to boost the warrant for belief in the resurrection of Jesus, he is clear that properly judging the probabilities involved is not possible on the basis of the data alone. See WCB, 271–280, and, ‘Ad Wykstra’, in ‘Reply’, 124–128. Keith Mascord incorrectly suggests that Plantinga allows no role whatsoever for arguments from historical evidence in support of Christian belief; Mascord, KeithAlvin Plantinga and Christian Apologetics (Milton Keynes: Paternoster, 2006), 168ff, 207–209Google Scholar.

48. I deeply appreciate comments received on an earlier draft, particularly from Robert Audi, John Hare, Joseph Jedwab, Nathan King, Matthew Lee, Mark Murphy, Alvin Plantinga, Michael Rea, Alan Rhoda, and Scott Williams.