Theistic Belief - Hinges and Proper Function


















John Dalsenni Olsen

Årskort: 19970826



Afleveret: 19/09-2003


Vejleder: Lars Bo Gundersen

Chapter 1 – Introduction: I-II-III skepticism 1

1.1 What is justification/warrant? 2

1.2 The I-II-III argument 9

1.2.1 The analogy 9

1.3 The Skeptical argument 13

Chapter 2 – Crispinsteinian Answer: Internalism and Hinge Certainties 18

2.1 Crispinstein’s suggestion 18

2.1.1Hinges 20

2.1.2 Hinges, Certainty of methodology 23 Norms in context 25 Context specific hinges 26

2.2 Wrapping up 31

Chapter 3 – Plantingian Answer: externalism and Proper Function 34

3.1 Resisting the inferential structure: Plantinga and the Skeptic, initial considerations 34

3.2 Plantinga’s Epistemic Response: Warrant and Proper Function 38

Proper function 39

3.3 Closing stage 43

3.3.1 Plantinga a disjunctivist? 43

3.3.2 Doxastic Experience 46

3.3.3 Uniform attitude 47

Chapter 4 – Religious belief: Crispinsteinian and Plantingian 50

4.1 The Crispinsteinian Stance 51

4.2 The Plantingian stance – The Aquinas/Calvin Model 55

4.2.1 Basicality with respect to Justification and Warrant 58

4.2.2 Perceptual or Experiental Knowledge? 59

4.2.3 Natural Knowledge of God and Sin 61

4.3 Is Belief in God Warrant-Basic? 63

4.4 Conclusion 66

Chapter 5 – Closing Study, Objections and Replies and Conclusion 68

5.1 Externalism, Transmission, Closure (and knowledge ascription) 68

5.1.1Transmission – Closure 71

5.2 The Anything Goes Objection 74

5.2.1 The SABS objection 75

5.3 Circularity 78

5.3.1 The argument from evil – an example 80

5.4 Conclusion 83

Bibliography 89





In this thesis, I’ll look at the epistemology behind religious belief – especially theistic belief. By ‘theistic’, I have in mind what is common in the great monotheistic religions: Islam, Judaism and Christianity: belief in one God who is the creator and sustainer of the universe. Some forms of Hinduism and Buddhism are also monotheistic; as far as these subscribe to these two characteristics, they also are considered as generic theisms.


Epistemological considerations of theistic belief need to comply with general epistemological standards. That is to say, the standards and requirements must be no higher or lower in mundane than in religious beliefs and vise versa. A substantial part of this thesis will be devoted to general epistemological considerations. This is because the crux of the epistemological problem identified in relation to religious epistemology is not isolated to theistic beliefs, but is equally pertaining to ordinary beliefs.


The first chapter identifies the problem for both religious belief and ordinary beliefs. The second and third chapters provide two epistemological answers to this problem. The forth chapter applies these two epistemological considerations to theistic belief. The fifth chapter singles out the view that complies most with what seems to be at stake in theistic belief formation and raises some objections.

Chapter 1 ­– Introduction: I-II-III skepticism


The Problem: Theistic belief, it is commonly thought, suffers from the significant disadvantage of being irrational, unreasonable, epistemically below par, ungrounded, unjustified or unwarranted.


In this thesis, I’ll set fort, and defend, the possibility of holding theistic belief their evidential predicament non-withstanding – holding theistic belief without resting it upon other beliefs. Defending this thesis strikes at the hearth of epistemological considerations of theistic belief as it has traditionally been conceived. On the one hand, there are those that say that theistic belief is only rational if there are good arguments for them; on the other, there are those that say that arguments are not necessary for theistic belief being rational (even if there are good arguments in the neighborhood).


As by way of entry, consider e.g., W. Clifford and Freud. The former says, “it is wrong always everywhere, and for anyone, to believe anything upon insufficient evidence”[1]. The thought seems to be that the theist holds theistic beliefs upon insufficient evidence. The latter says, “These [religious beliefs], which are given out as teachings, are not precipitates of experience or end-results of thinking: they are illusions, fulfillments of the oldest, strongest and most urgent wishes of mankind. Their secret strength lies in the strength of those wishes. As we already know, the terrifying impressions of helplessness in childhood aroused the need for protection – for protection through love – which was provided by the father; and the recognition that this helplessness lasts throughout life made it necessary to cling to the existence of a father, but this time a more powerful one. Thus the benevolent rule of a divined Providence allays our fear of the dangers of life; the establishment of a moral world-order ensures the fulfillment of the demand to justice, which have so often remained unfulfilled in human civilization; and the prolongation of earthly existence in a future life provides the local temporal framework in which these wish-fulfillments shall take place.”[2]


Now, these two objections come from wholly different perspectives. For Clifford it is wrong to form theistic beliefs because they are not justified in that they lack the evidence needed. Freud thinks that religious beliefs are not truth aimed, instead they arise by way of, so to speak, a process of wish fulfillment.


In this chapter, I’ll do two things. First, I’ll look at focal epistemological issues and relate the discussion to the two preliminary objections (Clifford’s and Freud’s). After that, I’ll present what I take to be a philosophically interesting and significant objection to theistic belief – I’ll present the I-II-III argument developed by Crispin Wright. According to this objection, not only theistic belief, but also some of our everyday beliefs seem to lack positive epistemic status. In the two following chapters, (chapters 2 and 3) I’ll present two epistemic responses to the I-II-III argument. These two epistemic responses represent two paradigms on what it is for a belief to be epistemically on par. These two responses serve as platform for the discussion of specifically theistic beliefs in the fourth chapter.


1.1 What is justification/warrant?


Epistemology is a highly complex field of philosophical study. Here, as elsewhere in philosophy there are few things that are universally agreed on. Nevertheless, for the purposes of this thesis I’ll take it that there is something which is added to mere true belief which makes it knowledge. Knowledge, of course, is the core term in epistemology, but nonetheless, I’ll in this thesis talk about concepts historically tied to the oft-used ‘justification’. Justification belongs to a cluster of normative terms that has nearby relatives such as ‘rational’ and ‘warranted’. Let us look closer at these concepts in order to sharpen our understanding of what the religious skeptic has in mind when he objects to theistic belief. 


One of the major happenings in the 20th century epistemology was when Edmund Gettier[3] published a set of counterexamples to the view that justification was that which made the difference between mere true belief and knowledge. The heart of this view is the so-called tri-partite definition of knowledge:


1.      S believes that p

2.      S is justified in believing in p

3.      p is true


This set of conditions was though to be both necessary and sufficient for knowledge. It is, of course, highly relevant to get clear about what ‘justification’ is. There are two broad views on the matter, both of which contain numerous, both major and minor divergences from paradigm cases. These are internalism and externalism on justification. Internalism is the view that only what the subject can be aware of can have a bearing on justification. Externalism is simply the denial of this claim. Within internalism, for example, there two major schools of thought: foundationalism and coherentism.[4]


Foundationalism is the view that epistemic justification has a hierarchical structure: some beliefs are justified not by relations to other beliefs. The foundationalist accepts two doctrines: The foundation doctrine: If any beliefs are justified, then some of the justified beliefs (i.e., the foundational beliefs/basic beliefs) are not justified in virtue of their logical or evidential relations to other beliefs; and the superstructure doctrine: Any justified non-foundational belief is justified, at least in part, in virtue of its logical or evidential relations to foundational beliefs/basic beliefs.


According to coherentism, epistemic justification is a holistic notion rather than a hierarchical one. Coherentist think it is incoherent to view some beliefs as being justified according to the foundational doctrine, and there from transmitting justification to the rest of the subject’s beliefs. More positively, coherentism is the view that epistemic justification arises when ones belief system hangs together or coheres. One fatal problem – I dare say, even without having come closer to what exactly justification is – is that two coherent systems can lead to contradictory beliefs. This is clear when we have in mind that justification is that which makes the difference between mere true belief and knowledge.


Justification of the internalist variety comes in a number of outfits, even when coherentism is set aside. William P. Alston[5] thinks that the natural understanding of ‘justification’ is deontological. Deontology has to do with duty, permission, requirement, and blame. In this sense, for example, when S looks at his watch (which reads: 12 o’clock) and by way of perceptual experience forms the justified belief that it is 12 o’clock, then S has preformed his epistemic duty, has epistemic permission, is within his epistemic right, and is without epistemic blame in forming this belief. If, on the other hand, S formed the belief that it is 02 o’clock, after that S has gone through the same perceptual experience as in the case where he forms the true belief, then S is floats his epistemic duty; S has not epistemic permission to form this belief, is not within is epistemic right, and is epistemically blamable for forming this belief.


On the tri-partite scheme, the above scene would be like this:


1*.  S believes that p

2*.  S is permitted to believe in p

3*.  p is true


If, as we are supposing, justification (in this particular case being within one’s epistemic right) is that which makes the difference between mere true belief and knowledge, then S knows that p. According to this conception of justification, there must be something S can be blamed for when he forms the wrong belief. If S did not have access to the ground of the belief – the introspective state on which the belief was formed then he would not, in any plausible sense of the above deontological terms, be blamable for forming the wrong belief. Therefore, the deontological view of justification needs to be backed up by a variety of internalism according to which the subject must have cognitive access to that which bears on to the justification of the belief. Furthermore, this view not only requires that S has access to the justifier – the perceptual experience: that it seems to S that the watch displays 12 o’clock – but actually has access to the “epistemic efficacy of the justifier”. The efficiency of the justifier, in the watch case, would be that S must know that his perceptual experience suffices to justify the belief.


Another major problem with the deontological conception of justification is that it assumes, and actually presupposes (in order to make sense) that belief formation is under voluntary control. I do not really know how to argue for the view that belief formation is not under voluntary control. It just seems clear to me that it is not. This goes, at least, for formation of beliefs about the immediate environment.


Let us revise the clock example a bit, by adding that as a matter of fact S’s watch stopped 24 hours earlier i.e., it showed the right time when S looked at it. Here we have the earlier advertised counterexample to the tri-partite view on knowledge. In this case, as in the above, S believed truly that p, was deontologically justified in his belief that p. But does this fit with what we like to think of knowledge? Certainly not! This is especially clear: it just was a matter of coincidence, chance or luck that S’s justified belief also is true.


Beside prompting philosophical work on a fourth condition, this counterexample has prompted an externalist turn, which as already mentioned, is simply a rejection of the constraint on introspective access, and accordingly, does not hold any particular view on that which makes the difference between mere true belief and knowledge. Externalism, as does Internalism, comes in several different versions. A paradigm case, though, is reliabilism, according to which that which makes the difference between mere true belief and knowledge is reliability of the belief producing, and preserving mechanisms. Accordingly a belief is warranted (the specific exteternalist word for that which makes the difference) if it is produced by a reliable belief forming process. Whether or not the belief forming process is reliable is not, of course, something that the subject can be aware of just by reflection. In schematic form:


1**.          S believes that p

2**.          S’s belief that p was produced reliably

3**.          p is true


Let us look at how reliabilism relates to the watch example. S fulfills conditions (1** - 3**), we may suppose. As far as the example goes S was not under the influence of anything that made his belief forming processes unreliable.[6] Then, in this version, externalism is not much of advancement from the internalist view. It still seems that reliabilism in the above formulation, does not suffice to block the luck scenario. 


Without going into the question on how to amend this view – a reliabilist orientated view – to take care of the watch example, we can, in order to bring in the concept rationality, (which I mentioned belongs among a cluster of concepts, two of which, we have glanced at) on the scene, ask the following question: how does S fare as relates to rationality in forming the belief as he does? As the example stands, S is rational to form the belief as he does. After all, the belief is justified deontologically in the internalist construal, and the belief is produced reliably according to the externalist construal. So according to both of these paradigm cases S’s can be said to be rational when he believes that p.


We can specify rationality to this overall distinction i.e. we can talk of internal and external rationality. In Warranted Christian Belief Plantinga talks[7] of internal rationality as “proper function downstream from experience”. (WCB p. 110) This metaphor is supposed to capture S’s belief being formed on the basis of or in response to the sensuous imagery – his perceptual awareness of the watch displaying 12 o’clock – and the psychological inclination to believe what the sensory imagery shows. Internal rationality is forming beliefs appropriate to both kinds of experience, and being sensitive to his other beliefs. Furthermore, S must also prefer to believe what is true as opposed to what is false. External rationality is proper function with respect to the formation of the sensuous imagery on which perceptual belief is formed, and formation of the right sort of psychological inclination.


In this section, we have looked at two paradigm cases of what makes the difference between mere true belief and knowledge: justification and warrant. According to these two views, S can still be rational in believing p even though S does not know that p. We have seen that these two views match the internalist and externalist views. According to the former view, it is only that which is cognitively accessible to the subject which can have a bearing on justification. In connection to this view, I made the distinction between access to the justifier (the perceptual experience e.g.) and access to the epistemically efficiency of the justifier. The externalist, on the other hand, claims that it is not necessary for a belief to be warranted that there is access to the ground of the belief; and even less so to the epistemic efficiency of the ground of the belief.


How does this stand with respect to Clifford’s and Freud’s complaints? As far as Clifford goes, his complaint pivots on the issue whether we have voluntary control over our beliefs. If we do not have voluntary control over our beliefs, which seems to me to be the case, then it would seem to be very difficult to maintain his objection that it is wrong to believe on sufficient evidence. Furthermore, this objection seems to require the implausible in relation to the access requirement. If it requires that, the subject in question must, not only have access to the experience in question, but also must know that the experience is epistemically efficient to justify the belief, then a whole panoply of beliefs which we take ourselves to be justified in enjoy a similar lot. The conception of justification underlying an objection to theistic belief must be such that every day beliefs can be justified according to that conception of justification.


What about Freud’s complaint? Freud seems to presuppose that the purpose of religious belief is not to depict or represent as is the case in other cognitive engagements, e.g. as is a function of scientific belief. Religious belief, then, does not have any epistemological value, albeit it has other values according to Freud. This complaint, then, does not amount to much of an epistemological objection to religious belief. Instead, it is simply a dismissal of the epistemological value of religious belief, according to which an essential part of the epistemological value of forming beliefs is to form true, in contrast to, false beliefs. However, according to Freud it is not such that religious belief is without merit. In fact, it seems that Freud thinks it has a function, albeit not an epistemological function. In that sense, we can say that for Freud it is still rational to form religious belief.


These two objections do not really have any force: the former presupposes a capacity that epistemic agents do not have. The latter presupposes that the process engaged in religious belief formation aims at something other than truth. So, we have seen that criticizing religious belief in the sense that it is wrong to form such belief does not amount to much of an interesting objection – in fact it is not really an objection, as much as it is a misunderstanding of our belief forming capacities. Freud on the other hand thinks that religious belief formation has a function – perhaps he is inclined to think that religious belief formation, just as any other of our belief formations enhances our survival fitness – but it has the significant disadvantage epistemologically, that it is not aimed at forming true belief, but something else.


So far, we have seen that criticism of religious belief must not rest upon an implausible epistemological view concerning belief voluntarism, if it is to of interest. Furthermore, interesting critique of theistic belief is not captured by pointing out that theistic belief is irrational – we have seen that, according to Freud at least, theistic belief can be rational, even though it is false. We have made a bit of progress in terms of what warrant and justification is by identifying two overall epistemological positions: internalism and externalism.


How then can we construct an objection to theistic belief that does not fall into the pitfall of Clifford and Freud? Crispin Wright has set up a general scheme, which, I think, will prove useful and illuminating for formulating the skeptic’s objection.


1.2 The I-II-III argument

In the first section of “Wittgensteinian Certainties”[8] Wright, from which I draw the I-II-III argument, establish an analogy between everyday epistemic situations and the notorious Moorean proof of the external world. In typical everyday situations epistemic agents take some things for granted; similarly in Moore’s case. We’ll look at some of his examples and see how they relate to Moore’s proof.

1.2.1 The analogy

One famous story wherein religious belief is formed is the one were Moses is spoken to through a burning bush. Moses, according to the story, hears a voice and sees a burning bush. He forms the belief that God speaks to him. So, there is some auditory and visual experience associated with this belief formation. We can represent this belief-forming situation in this manner (MOSES):


(e) It seems to Moses that God speaks to him through a burning bush

Moses forms the belief that

(P) God speaks to Moses

from (P) it follows that

(I) God exists


Even though this is somewhat an extraordinary story, there is a plausible structure that emerges from the above which can easily be used in more ordinary religious experience. The structure cannot only be used to religious experience. Here’s an example of an everyday epistemic situation (SOCCER):

(e)          Jones has just headed the ball into the net, he is being congratulated by team-mates and the crowd has gone wild.

Which provides defeasible warrant for

(P)          Jones has just scored a goal

which entails that

(I)           A game of soccer is taking place (WC p. 5)


In ordinary situations would this argument be fully convincing, assuming that it is only in a soccer game that a goal can be scored. However, imagine that you are in the vicinity of a studio which specializes in making soccer movies, and that the scene that you just witnessed most likely is part of a film take. Equipped with this ‘background’ information, there would not be much by way of warrant for P; e would not be able to warrant P. What I need if e is to provide any warrant for P is “precisely some independent corroboration of the context – that is, of I.” (WC p. 5)


This independent corroboration could be provided e.g. by asking a bystander, whether this is a genuine game or whether it is a film take. If the former you would acquire warrant for P, but that would, says Wright

“be absurd to regard that warrant transmissible across the entailment form P to I. You don’t get any additional reason for thinking that a game is in process by having warrant for P. It remains that your only ground for I is the bystander’s testimony and it is only because you have that ground that witnessing the scene provides a warrant for P at all.” (WC p. 5) 


So much for the first part of the analogy; now we turn to Moore’s notorious proof of the external world. Here is Wright’s formulation of the version that he detects in Moore’s paper Proof of an External World:[9]


Premise            Here is a hand

Conclusion       There is a material world (since any hand is a material object existing in space)


This argument clearly begs the question. It is not such that the skeptic is denying that Moore sees a hand when he lifts it up in front of his eyes, but rather that the argument does not enable the skeptic to overcome the rational doubt of the skeptical possibility. It would be just as plausible to say, “God speaks to me and therefore God exists”, which obviously would not be much by way of overcoming rational doubt theism. A version of the above argument that follows the (SOCCER) and (MOSES) template is construed in terms of his subjective hand experience, since the skeptic will insist that Moore’s argument “begins at the wrong place, since his premise is something which rests on more basic evidence […]”(WC p. 6) (MOORE):


(e)        My current state of consciousness seems in all respects like being aware of a hand held in front of my face

(P)       Here is a hand


(I)        There is a material world


(MOORE), as did (SOCCER) and (MOSES), lacks independent corroboration of I in order for e to provide warrant for P. So, when I already presuppose I, in my argument it would not be much of an forceful argument for the one who doubts I in the first place – i.e. the skeptic.


Here is one religiously inspired argument that follows the above template which is stated in a more common way not involving burning bushes (THEIST):

(e)        I am appeared to in a certain (spiritual) way

(P)       God is appearing to me

(I)        God exists


There are no logical mistakes or fallacies in (THEIST) and (MOORE) but the skeptic, will not be persuaded for that reason – he’ll point out that the arguments do not provide rational compel – they presuppose what they argue for. Unless that I believe, or presuppose, that God exists, I would not be particularly persuaded by (THEIST); even if I presupposed I would I not reckon the argument of any value, because of circularity. It isn’t the case that (MOORE) and (THEIST)[10] work with an outlandish concept of proof. Rather the focal issue is “under what circumstances a valid argument is indeed at the service of proof – i.e. the generation of a rational acceptance of the truth of its conclusion – or rational overcoming of doubt about it.” (WC p. 2) In other words, the crux is “when a particular epistemic warrant for its premises transmits across an entailment.”


There are a number of integrated concepts that are important in these matters. Wright says of Transmission of Warrant (TW):

A particular warrant for the premises of an entailment is transmitted to its conclusion only when one’s path to that warrant does not require picking up knowledge of the conclusion en route, or depend on some form of prior entailment to it. (WC p. 3)


All of the above cases fail the (TW) requirement for transmission. The cases pick up knowledge of I en route and thereby fail to transmit, whatever epistemic warrant there is for the premises, across to I. (TW) needs to be distinguished from the ‘weaker’ Closure Principle, which can find application even when transmission fails (CL):


Knowledge always transmits across known entailment: if A knows p, and A knows that q follows from p, then A knows q.


Perhaps the best way to see how (TW) is stronger than (CL) is in terms of what Wright calls a “Cogent Argument” (CA):


A cogent argument is one whereby someone could be moved to rational conviction of the truth of its conclusion. (C&QB p. 140)[11]


(TW) then consists of (CL) and (CA). Transmission takes place when (CL) and (CA) hold of the argument; only then there is transmission of warrant.


As is the case with (SOCCER), and other empirical cases, background information heavily influences the persuasiveness of the argument. So, normal empirical cases are information dependent. Wright says of information dependence of warrant (IDW): A body of evidence, e, is an information-dependent warrant for a particular proposition P if whether e is correctly regarded as warranting P depends on what one has by way of collateral information. (WC p. 3)


It is difficult to see how (MOORE) and (THEIST) could side step the information-dependence charge, such that the warrant (possibly) enjoyed by the premises would transmit across to the conclusion. However, (SOCCER) is more easily dealt with. In the case when I get independent information from my bystander the warrant that I have for (e) extends to (P) and is transmitted to (I). But, in the case where there is no such independent information the argument will not exhibit transmission.


1.3 The Skeptical argument

Substitute in the above arguments e with I, P with II, and I with III, and call them type-I, type-II, and type-III propositions.  (MOORE) then becomes (in a condensed version):

              I.      I seem to see a hand

           II.      I see a hand

         III.      The external world exists

The form of the arguments presupposes a noetic structure wherein experiential propositions (type-I propositions) only function as ground for type-II proposition if type-III propositions are presupposed.


Even if independent support for (SOCCER) III were not available, it would not thereby imply skepticism about the existence of soccer games. Similarly, for (MOORE) and (THEIST), even though they fail the transmission of warrant requirement it does not force skepticism about the external world nor the existence of God. But, the skeptical argument – the I-II-III argument – becomes overt, when we ask what (MOORE) and (THEIST) need in order for type-I to be suitable or eligible to justify type-II and there from to transmit the warrant across the entailment to type-III. “What – if Moore’s warrant for his original premise is information-dependent ­– could put the needed collateral information […] in place?” (WC p. 8) The skeptical argument has it that it cannot be anything other than inference, but that is exactly what the problem consists in – we cannot infer type-III propositions from any propositions about it – that would beg the question, just as Moore did. “How on earth else? The emergent sceptical challenge denies that there is any other way.” (WC p. 8) It involves these five claims:


                                        i.            Type-II propositions can only be justified on the evidence of type-I propositions.


                                       ii.            The evidence provided by type-I propositions for type-II propositions is information-dependent, requiring inter alia collateral warrant for a type-III proposition.


                                     iii.            Type-III propositions cannot be warranted by transmission of evidence provided by type-I propositions for type-II-to-type-III entailment


                                     iv.            Type-III propositions cannot be warranted any other way


                                      v.            Type-III propositions could be false (WC p. 9-10)


If we accept all of (i) – (v) we have a severe problem with skepticism; not only skepticism of the existence of the external world and God, but equally of other minds, of the past, of perceptual reliability, and of inductive inferences. These examples have a sufficiently broad kind of generality for the I-II-III argument to be applicable to it. Consider (MOORE); the type-III proposition – the external world exists – is very extensive for our epistemic situation. If the denial of (MOORE)’s type-III proposition, actually is the case: then the knowledge that I take myself to have concerning myself working on my master thesis, in a real world that exists independently of my cognitive activity, would be false.


An argument, or as Wright’s says, “a paradox” of this kind arises whenever we are inclined to think that the ultimate justification for a claim, type-II propositions, say, rest upon inference form information of another kind, type-I propositions. Wright says: “In any such case, the warrantability of the inference will arguably depend upon the presupposition that there is indeed a domain of fact apt to confer truth on type-II propositions in the first place, a domain whose details are broadly reflected in type-I information. So it will depend, a fortiori, on the first component of that: that a domain of fact which type-II propositions are distinctively apt to describe as such exist. Let this suppositions be the relevant type-III proposition.” (WC p. 9)


Other examples that fall to the I-II-III argument: (PER):

              I.      It seems to me that this perceptual belief (a cursor is blinking on the screen) is reliable

           II.      In most situations when I form beliefs by way of perception they are reliable

         III.      Perception is a reliable belief generator


and (MIND):

              I.      My son is bleeding.

           II.      My son is in pain

         III.      There are other minds


These arguments fall also to the I-II-III argument. The problem, if we agree to the I-II-III argument, is that the only way that I can justify II is by way of I. But, that move can only be made if I’m antecedently warranted in III. If PER type-I is to provide justification for PER type-II I must already rely upon my perceptual reliability if it is to justify II. But how could I do that, when the whole enterprise is to provide reasons – persuade the skeptic of perceptual reliability of perceptual reliability? (PER), (MIND), (MOORE), (MOSES) and (THEIST) have a sufficiently broad range of application to fall for the I-II-III argument – it is very difficult to justify perceptual reliability without already taking for granted the reliability of perception; it is very difficult to justify the existence of the external world without already presupposing its existence, it is very difficult to envisage an argument for the validity of (THEIST) III without already, in the premises of that argument already to presuppose his existence. This is not the case for (SOCCER). In that case it will be sufficient to get information from the person standing next to you about the genuineness of the game i.e. to reject (iv) in the I-II-III argument – the type-III proposition (a game of soccer is taking place) can enjoy warrant transmission, when independent information is provided, for example by the bystander.


The seriousness of the whole issue is obvious. Consider (PER). If we cannot argue by way of the I-II-III structure type argument that perceptual beliefs are reliable, then we seem to have a major problem – most of our beliefs, I take it, are the result of workings by our perceptual organs. It is vital, it seems, that we be able to produce good reasons – arguments – for our beliefs. One important belief is the belief that we have in relation to perception. But if we are not in a position to adduce non-question begging grounds for our belief, then it seems that we cannot continue to trust our perceptual beliefs. According to the I-II-III argument this is precisely what we’re not in a position to do.


To recapitulate, objections to theistic beliefs are familiar. Precisely what the objections amount to is not always clear, though. In this thesis we have made the objection in terms of the I-II-III argument according to which beliefs belonging to a certain domain (Religious, external world) cannot be warranted by information of another sort of that domain (experiential propositions) because the former beliefs depend upon the fact that there is such a domain that they are propositions about. More simply stated, experiential propositions cannot be adduced to justify type-II propositions unless we already are presupposing the relevant type-III proposition. When that is the case, there will not be any transmission of warrant from type-II to type-III, because the justification of type-II by type-I presupposes that the relevant type-III is in place.


In this way, we can say that unless we can produce a cogent argument – a warrant transmitting argument – from premises describing the occurrence of such experience to the belief about God appearing to a person, then that person is not entitled to consider him being justified in the type-II proposition in question. In other words, the ability to adduce non-question begging grounds is required for a belief to be justified. This has indeed proved difficult; it seems that there is no way to go from type-I propositions to type-III propositions via type-II without picking up, or relying upon, type-III if we accept all of the five points in the I-II-III argument. An argument that picks up or relies upon type-III cannot be among the class of arguments that we call cogent.


The I-II-III argument invites a number of replies. I’ll focus on two uniform attitudes, according to which there is a meta-story to be told as far as responding to the skeptic. Now, it is not necessary with a uniform attitude. Exempla gratia, one response could be targeted sidestepping the skeptic with regards to (PER), and another for (MIND). These two responses correspond to the two paradigms in epistemology identified earlier in this chapter: internalism and externalism.


The next two chapters will be devoted to a general response to the above argument. In chapter 2, we will look at how Crispin Wright, even though he accepts all of the five points, thinks that there is an intellectually satisfactory answer to this paradox. In chapter 3, I’ll present what I take to be Alvin Plantinga response to this argument. These two chapters lay the epistemological foundation for chapter 4 where specifically theistic belief will be the center of our attention. The first section of chapter 4 will be devoted to a possible construal of theistic belief built upon Wright’s general epistemological response to the I-II-III argument. This possibility hinges on the plausibility of setting on par theistic type-III and ordinary type-III, as I have done in this chapter. The second section of chapter 4 will follow the line suggested by Plantinga in chapter 3.


Chapter 2 – Crispinsteinian Answer: Internalism and Hinge Certainties


In this chapter, I’ll look at Wright’s Wittgensteinian inspired suggestion to handle the I-II-III argument. This response differs from the view we will look at in chapter 3, by accepting all of (i) – (v) (see p.7) i.e., accepting the inferential architecture of the argument, but nonetheless avoiding ending in a paralyzed epistemic situation. First, we will look at the general notion of a hinge, hereafter we will see how hinges are made up of two classes of certainties. In chapter 4, I’ll apply this suggestion to theistic belief.


2.1 Crispinstein’s suggestion


The plan is to offer a “uniform response” or “uniform attitude”, as he prefers to call it, to I-II-III skepticism. “The crux” of this suggestion “will be to point to a possible case that we are within our epistemic rights, as it were, in accepting the type-III proposition that we do their evidential predicament notwithstanding.” (WC p. 13) The idea is Wittgensteinian in nature; even though Wittgenstein did not explicitly develop it him-self[12], the resources are though to be found in his writings.


A distinctive theme in Wittgenstein later works is the distinction between knowledge properly called, and “another much wider class” of certainties: “propositions that ‘stand fast’ for us not because they have won through under scrutiny of relevant evidence but because, so he [Wittgenstein] suggests, they are presuppositional and basic in the very process of gathering and assessing evidence or within our more general ‘world picture’.” (WC p. 13)  The propositions that stand fast are not, in other words, standing fast because they are well established; on the contrary. These propositions are not the result of “cognitive achievement” they are, as Wittgenstein says in On Certainty[13], not explicitly learned.


OC 151.   I do not explicitly learn the proposition that stand fast for me. I can discover them subsequently like the axis around which a body rotates. This axis is not fixed in the sense that anything holds it fast, but the movement around it determines its immobility.


The crucial thing, says Wright, is the imagery of the axis, which suggest that there is noting outside – external – that holds the propositions in place. They are not basic certainties that are firmly grounded, nor are they solid foundations, in the sense that they cannot be shaken or moved.


The focal point revolves around the notion of a hinge; which we find in these well-known paragraphs:


OC 341.   That is to say, the questions that we raise and the doubts depend on the fact that some propositions are exempt from doubt, are as it were like hinges on which those turn.

OC 342.   That is to say, it belongs to the logic of our scientific investigations that certain things are in deed not doubted.

OC 343.   But it isn’t that the situation is like this: We just cant investigate everything, and for that reason we are forced to rest content with the assumption. If I want the door to turn, the hinges must stay put.

OC 344.   My life consists in my being content to accept many things.


A hinge proposition is – to give a first approximation – a proposition which is presupposed in the game of giving and asking for reasons.[14]  It is an “inherited background” against which true and false are distinguished. There are quite a few notions which cast some light on the hinge notion. Just to give a few:


Instead of extending the list let’s look at Crispin Wright’s elaborations of the ‘hinge-thought’.



Of special interest are issues drawn to our attention in a cluster of renowned aphorism in On Certainty. Here Wittgenstein writes:


OC 94.       But I did not get my picture of the world by satisfying myself of its correctness; nor do I have it because I am satisfied of its correctness. No: it is the inherited background against which I distinguish between true and false.

OC 95.       The propositions describing this world-picture might be part of a kind of mythology. And their role is like that of rules of a game; and the game can be learned purely practically, without learning any explicit rules.

OC 96.       It might be imagined that some propositions, of the form of empirical propositions, were hardened and functioned as channels for such empirical propositions as were not hardened but fluid; and that this relation altered with time, in that fluid propositions hardened, and hard one became fluid.

OC 97.       The mythology may change back into a state of flux, the river-bed of thoughts may shift. But I distinguish between the movement of the waters on the river-bed and the shift of the bed itself; though there is no sharp division of the one from the other.

OC 98.       But if someone were to say “So logic too is an empirical science” he would be wrong. Yet this is right: the same proposition may get treated at one time as something to test by experience, at another as a rule of testing. (Underline, my emphasis.)



There are a number of issues tied to this passage. But, let’s focus on the status of the “inherited background” against which the empirical enquiry is conducted – it is the background that allows the enquirer to judge a proposition true or false. It is appropriate here to distinguish between ‘cultural’ and ‘hard wired’ inheritance. I can for example inherit some physical, as well as psychological traits of my parents – these traits are hard wired. It could for example have the ear of my father, and the temper of my mother. On the other side there is cultural inheritance were for example such aspects of life as language and aesthetic sense are tied. Now, it is immediately reckoned that even though the hard-wired examples is a clear case (of me having the same ears as my father) it also has an instances that is not as clear (I have the temper of my mother.) Similarly in the cultural case. In spite of this, I think it appropriate to make this distinction. Now, the background is something which goes beyond the immediate experience, in that the background at the moment of experience is taken for granted. The background can be compared to a picture by which experiences are discriminated i.e. the background is that which allows the enquirer to ‘tell apart’ the different aspects of the visual experience, say. The picture is, at the moment of experience, not subject to empirical inspection; in this way it can form, or be a “part of a kind of mythology”. But, Wittgenstein immediately reminds us, myths, like the role of rules of a game, do not necessarily need to be learned by theory – a priori story telling, or indoctrination, if you prefer. They can be learned by practical engagement: learned in practice.


But, as Wittgeinstein says, even if the background can be compared with logic and a priori propositions it must not be thought of as an empirical science. The analogy to logic and a priori propositions is only applicable when in context. It is only in the process (when you are in the game) of forming-beliefs, arguing, philosophizing and practical interaction that the hinges have this peculiar status. The status of the hinges, then, is of a special kind. Special because, in context they play a certain presuppositional role, they are not doubted, they are held firm as rules of logic; but on the other hand, are they subject to empirical import – subject to change by way of new evidence. 


As by way of looking closer to the status question we might ask whether hinges are truth applicable i.e. whether they can be true or false. It might perhaps be tempting to view them as a certain special kind of ‘objects’ which cognitive agents stand in relation. The suggestion related to ‘special kind of objects’ is in contrast to ‘normal’ truth applicable propositions e.g. my car is a Peugeot 306, which, I take it, is clearly truth applicable. The proposition that my car of a certain type is either true or false; in fact I own a car that is of the mentioned type, therefore the proposition is true. This seems clear. But perhaps it is not the same with type-III propositions (e.g. “the external world exists”). In the car case there are certain a background frames in operation when I judge my car being a Peugeot 306; certain norms as to how to speak of cars. There are certain classifications as to what sort a car is – certain rules. And more to the crux: I presuppose that the world exists, in my judgment of the car. If I was convinced of the negation of (MOORE) type-III proposition in question, then my judgment of the car would not be eligible for the truth and false predicate, simply due to the fact that my background conviction undermined the possibility of ascribing properties to objects because there is no world where objects exist, let alone have properties. In normal circumstances we do not doubt the type-III proposition in question; but if these are some rules of evaluating types of cars then my car judgment finds a background frame wherein it can be truth-evaluated.


Can we construe a similar scenario wherein a type-III proposition finds a background frame wherein it can be truth-evaluated? That seems very difficult – what could be the background upon which the type-III proposition could be evaluated? This leads to a blind end: there does not seem to be propositions upon which type-III proposition can be truth-evaluated. If the background thought is to be the criteria for whether a proposition is truth-applicable, then we must set aside type-III proposition as a special kind of propositions that are not truth-applicable. But that goes against intuition. That ‘the external world exists’ seems, (even though there is difficulty to find a type-III proposition to evaluate it against) to be truth applicable seems intuitively clear. To provide an ‘argument’ to that effect is, perhaps, not done in any better way than Wright does it in relation to another type-III proposition – the world has existed more that 4 minutes. He simply asks the following question: “How can there be real facts about the winners of the FA Cup in 1930’s but no real fact that the world did not come into being 5 seconds ago replete with apparent traces of much extended past?” (WC p. 33) In relation to our type-III proposition we simply ask, how can there be real facts about threes, coffee cups and cars, and no fact of the matter concerning the existence of the external world?


Another aspect that pertains immediately to the status question is whether type-III propositions have any meaning – whether they have any content or cognitive meaning as opposed to emotive meaning. Given that type-III propositions are truth applicable, it is impossible to see them as void of content or cognitive meaning. How could they be truth applicable if they do not have meaning? From the above elucidations, it does not come as a surprise that type-III propositions do indeed have meaning. The contrary viewpoint might be defended along the lines of the ‘beyond supportive evidence framework’. But this would only be possible if the further requisite is the verificationist theory of meaning which seems to be amiss.

2.1.2 Hinges, Certainty of methodology

As already hinted the hinges can both have empirical as well as a priori content. Wright reflects on the proposition 13+7=20. This proposition has the status “that nothing is allowed to falsify it.” (WC p. 17) As an illustration, consider the following: My older son has just put 13 apples into a basket, and my younger son has put 7 bananas into the same basket. Let’s imagine further that there are no other fruits in the basket. As a demonstration that he is good at counting, my older son starts counting the fruits in the basket. But, he counts 21 fruits! If we agree, as we have, that there were in fact 20 are fruits in the basket, as well as that my son did not make a mistake counting them, and no generous fruit-providing-angel has supplied the last apple, then the picture we get seems to be awkward. There must be something wrong somewhere! The nature of 13+7=20 teaches that there must be a mistake in the counting process, provided there are no fruit-providing-angels around.


The dominating idea in On Certainty and Remarks on the Foundation of Mathematics says Wright,


“is that (some of) the needed principles of appraisal are provided in the guise of certainties of elementary logic and mathematics; and that the unwavering – dogmatic – confidence we repose in these propositions, rather than being the product of a different kind of – superlatively sure, a priori – form of cognition, attaches to them in their role as in effect rules: norm of enquiry whose job it is to regulate the appraisal of empirical evidence […]” (WC p. 18-19)


This seems to be the case when my son is counting. If it was true that my two sons lay 20 fruits in the basket, and that there were no apples in the basket before etc. then there must be a mistake in the counting process – due to the special status of the hinge – certainty guised in mathematical terms.


Not only do a priori propositions as 13+7=20 function as a regulative role; propositions – empirical seeming – such as “I have two hands” may function in the same way as the mathematical propositions. They too functions as norms of inquiry – this seems to be the new aspect in On Certainty: the extension of this certainty thought to empirical-seeming propositions. They are not doubted in normal cases, and the evidence that I could produce on its behalf, would in no instance be greater than the first-hand acceptance; in fact it would it be awkward if I tried to produce evidence for it.


But propositions of this kind – empirical seeming – would be taken as background or as a hinge only if the context was normal: if e.g. I’m on a fishing vessel in rough seas. On the deck where I’m located there is a large object that has torn itself loose. The object has smashed into the ship side just bedside where I’m standing. I see that a sharp edge just hit around the area where (I thought) my hand is located. In this case, we may suppose, the context is such that “I have two hands” may be the object of empirical investigation, were the proposition in fact is doubted. Usually, though, it functions as a norm, but in cases where there is ‘external’ reason to suggest that it might be otherwise, then the selfsame proposition that serves as background, and usually functions as a hinge “becomes a straightforwardly empirical one.” (WC p.20)[15]


The sense of the double status of the hinges – they are taken for granted at the moment of experience, but they are also subject to change – is latent here. The mythology may change back into a state of flux, as Wittgenstein says. The background is subject to change, the hinges are sometimes worn out – in need of repair, so to say. If the hinges were incorrigible, ultima facie justified, then, no matter how much experience that did go against them it would and could not be repaired. Think of the background frame as the Ptolemaic world picture, exempla gratia, which in ways was consistent with (ordinary) experience. But when new (specialized) experience came in, which did not fit the background or hinge, the time had come for repair – the hinge was worn out. The door, if we imagine it as hanging on the hinge, did not swing smoothly any longer; the game of giving and asking for reasons did not swing on the hinge, due to too much resistance. Norms in context

Wright talks about “certainty of methodology” (WC p.24) of which if we entertain doubt would have catastrophic epistemic consequences; it would amount to doubting the cognitive ‘hook up’ so that no matter how much evidence that is for a particular proposition, the ‘methodological doubt’ would undermine the epistemic value of the particular proposition. The idea of ‘certainty of methodology’ contains two different distinctions: On the one hand, that between rules governing enquiry, and the propositional objects of the inquiry. And on the other hand, that between beliefs that would consist with our normal method of appraising beliefs, and beliefs which we jettison would lead to potentially catastrophic epistemic effects.


Among propositions that doubt about, or rejection of, would lead to catastrophic epistemic effects are (of course) type-III propositions:

The earth exists.

Perception is a reliable belief generator.


If we entertain doubt concerning these kinds of propositions then it (potentially) will lead to catastrophic epistemic results. But, the question is whether these propositions are, in the first place, open to doubt? Wright writes: “[…] what might we ever have counted as potential evidence against the existence of the earth?” The suggestion is that “Evidence cannot count against a type-III proposition – if it could, it could count for the hypothesis that there is indeed no material world (or that all other humans are zombies, that the world is indeed no more than five seconds old and that there are no inductive regularities.)” (WC p. 25)


Now, three types of ‘norms-in-context’ propositions are individuated by Wright. Common for them is that the rejecting them would rationally facilitate a wide re-organization of our perspective on what is to count as evidence for what kind of proposition. 

A: Propositions (simple arithmetical equalities, “I have two hands”) which it is our practice, always or normally, to insulate from disconfirming evidence, and which thereby serve as, in effect rules for the evaluation – re-direction – of the significance of such evidence.


B: Propositions (“My name is CW”, “This calculation is correct”) which are supported by – by normal standards – an overwhelming body of evidence, whose significance would have to be overridden if they were doubted.


C: Propositions of type-III (“The earth exists”, “Physical objects continue to exist when unperceived”, “The earth has existed for many years past”) to doubt which would have the effect of undermining our confidence in a whole species of propositions, by calling into question the bearing of our most basic kinds of evidence for propositions of that kind. (WC p. 27)


Wright suggests that A – C draw attention to one important theme: what kind evidential support, if any, their members in principle do allow. A – C reflect our tacit acceptance of various kinds of rules of evidence. (…) This theme is what Wright thinks that Moore neglected when he “took himself to know” e.g. ‘that the world exists’. Moore cannot know this, because type-III propositions are not known, they are certainty propositions that “allow of no defense in terms of the idea of knowledge”. (WC p. 27) Moore’s mistake was that of confusing the ‘grammar’ of knowledge entertaining and the ‘grammar’ of certainty propositions which cannot be objects of knowledge, Wright opines. Context specific hinges

Just accepting type-III propositions does not give a satisfactory answer to the skeptical possibility: just acknowledging the hinge does not possess the resources to set aside the intuitive pull we have for (v) (see. p. 7) – that III could just as well be false!


 Fortunately, Wright suggests that there is an intuitive accommodation with skepticism, a response that is “prefigured by one tendency in Wittgenstein’s remarks” (WC p. 35). Wright asks us to compare:


OC 163.   ………. We check the story of Napoleon, but not whether all the reports about him are based on sense-deception, forgery and the like. For whenever we test anything, we are already presupposing something that is not tested. Now am I to say that the experiment which perhaps I make in order to test the truth of a proposition presupposes the truth of the proposition that the apparatus I believe I see is really there (and the like)?


OC 337.     One cannot make experiments if there are not some things that one does not doubt. But that does not mean that one takes certain presuppositions on trust. When I write a letter and post it, I take it for granted that it will arrive – I expect this. If I make an experiment I do not doubt the existence of the apparatus before my eyes. I have plenty of doubts, but not that. If I do a calculation I believe, without any doubts, that the figures on the paper aren’t switching of their own accord, and I also trust my memory the whole time, and trust it without reservation. The certainty here is the same as that of my never having been on the moon.


Implicit in these aphorisms, is the familiar issue of warrant transfer; transmission does not take place, we remember, when the alleged warrant in the conclusion is already presupposed in the premises. This, says Wright, is also the case where the warrant for the premises is non-inferential; because it “always involves various kinds of presupposition.” (WC p. 35)Among these presuppositions, Wright suggests, the following, which we can call ‘context specific propositions’:

1.           Proper function of the relevant capacities.

2.           The suitability of the occasion and circumstances for their effective function

3.           The integrity of the very concepts involved in the formulation of the proposition in question.[16]


Wittgenstein takes these propositions for granted – they are simply presupposed. If we’d go into a justificatory context of 1-3 then regress would lurk heavily. So, whatever warrant a proposition has, it must either be by way of warrant transfer or by way of presupposing propositions similar to those of 1-3. This leaves us in a pretty dim situation with regard transmission of warrant across an entailment wherein type-III propositions are presupposed, and where 1-3 also are presupposed. 


“The general source of the limitations on warrant transmission in these cases is thus a consideration about the essential limitations of any particular cognitive achievement: wherever I achieve warrant for a proposition, I do so courtesy of specific presuppositions […] for which I’ll have no specific, earned warrant. This is a necessary truth. […] These propositions are not just one more kind of ‘hinge’ as we have understood that term. Hinges, so far, are standing certainties, exportable from context to context (subject perhaps to certain restrictions on the receiving context.) Whereas the present range of cases are particular to the investigative occasion: they are propositions like that my eyes are functioning properly now, that the things that I am currently perceiving have not been extensively disguised so as to conceal their true nature, etc.” (WC p.36)


1-3 are comparable, though, to the three types of hinges already identified – A: a priory (simple arithmetic) propositions, B introspective propositions my name is CW, and C: type-III propositions. But, even though, the parallels between context-specific (standing certainties) and norms-in-context propositions are noticeable e.g., that both will lack warrant when a justificatory activity is engaged on their behalf, and they are propositions that the epistemic agent must take for granted; but  there is also differences: Norms-in-context[17] propositions or type-III hinges, license defeasible inferential warrants whereas Context-specific propositions (of type 1-3) license inferential and non-inferential warrant.


We are advised, then, not to think that context-specific presuppositions will help us provide warrant for an III-type proposition in pain of circularity. On top of that, the skeptical voice is also, present for 1-3 propositions, in such a way that whatever warrant I think provided by them is just as strong as the warrant that I have for context specific hinges hold at the occasion – which is none at all! How then is this suggestion helping? Wright is not satisfied with Wittgenstein’s ‘arbitrarily breaking off’ and taking for granted 1-3, and therefore suggests the following line of thought.


First an negative aspect: there is no such thing as a process “of warrant acquisition each of whose specific presuppositions warrant has already been earned”  (WC p. 37) which per se is reliable in such a way, that whatever that process licenses automatically gets warrant. Wright suggests that our ‘ordinary concept’ of an acquired warrant, suggests that the reliabilist or naturalized ideal is “incoherent”. (See WC p. 37) Instead, he suggests that


“we should view each and every cognitive project as irreducibly involving elements of adventure – I take a risk on the reliability of my senses, the amenability of the circumstances, etc., much as I take a risk on the continuing reliability of the steering, and the stability of the road surface every time I ride my bicycle.” (WC p. 37-38)


This peculiar metaphor that Wright uses might raise some eyebrows. If I understand this passage right, then it seems that Wright envisages every cognitive projects epistemic result to be ultimately depended on chance or luck. That contention, though, is a bit harsh, on first, but perhaps is, when reflected upon, nothing else than a neutral description of our epistemic situation.[18] When an agent engages epistemically (be that within the agent, or in the agents environment) that involves making presuppositions that at the occasion, at least, are not warranted – “are not themselves the fruits of such enquiry and therefore not known.” (WC p. 42) For every epistemic project we engage in, there is always an element of adventure, an element wherein the epistemic agents is entrusted the capacities to carry on in the respective environments.


The way to get going in the epistemic project is simply to take a risk; if the risk is not taken, the practical, as well as the cognitive, result would be devastating. So, fortunately reasons have to come to an end, and do come to an end – if I was to make sure that the bike was ok when I got on it, as well as I was to make sure that the surface was ok, when I had moved 1 cm, than after that I needed again to check the bike… then I would never be on route. Similarly for the epistemic project; I must stop somewhere, if not, I’ll never get started.


There are two frames drawn form the above. If the epistemic engagement is in the agents own environment, then the credibility of the beliefs formed, will depend upon “presuppositions about [the agent’s] proper functioning, and the suitability of the prevailing conditions, etc.” (WC p. 42) If the engagement is of states of affairs outside, then the credibility of the beliefs formed, will depend upon “presupposition of the augmented type-III propositions which condition my conception of how the locally accessible my provide indications of what lies beyond.” (WC p. 42)


The vigor of the risk is that rational agents are allowed to continue in an epistemic project save if there is information available indicating that something is amiss. The suggestion is to trust the reliability of our cognitive hookup unless there is evidence to the contrary. This suggestion stands in opposition to the traditional empiricist dictum, (the evidentialist dictum) that rational agency is displayed (only) when the agent believes what there is evidence for. So unless there is evidence for the reliability of our cognitive faculties, and evidence that the environment is suitable, we would not be entitled to trust the output of those organs functioning in that environment. According to Wright the right picture is that whenever an investigation is undertaken in an “epistemically responsible” manner (epistemic responsibility cannot involve an investigation of every presupposition whose falsity would defeat claim to warrant) the acquired warrant is “no stronger than the warrant for any of the presuppositions about which there is some specific reason to entertain a misgiving.” (WC p. 37)


The risk – that we believe the deliverances of our cognitive faculties, unless there is some indication to the effect that there is something wrong with this presupposition, not because there is positive evidence for their reliability – needs “its own version of the Serenity Prayer: [...] we must hope to be granted the discipline to take responsibility for what we can be responsible, the trust to accept what we must merely presuppose, and the wisdom to know the difference.” (WC p. 43) [19]

2.2 Wrapping up

We remember that one of the key issues that generated the I-II-III argument was: A body of evidence e is an information dependent warrant for a proposition P if whether e is rightly considered warranting P depends upon what one has by way of collateral information. How does this view on information dependence relate to the ‘defense’ against the I-II-III argument we just have looked at? Essential, of course, is the suggestion that there are two kinds of hinges: 


  1. Those that take care of the information-dependency problem: norms in context hinges.


  1. Proper function and environment: context specific hinges.


Remember that the certainty beliefs or propositions are not doubted. As Wittgenstein says “The reasonable man does not have certain doubts.” (OC 220) They consist of propositions that make up the background or framework within which our thinking and acting goes on – the game of giving and asking for reasons, searching for evidence, asking questions, running tests etc. The information dependency of warrant acquisition is ‘put in place’, as we might say, by the class of certainty beliefs that are taken for granted. They are taken for granted i.e. they are not the result of a cognitive achievement – and fail to be defended in terms of the idea of knowledge – even though they contain a class much wider than that of knowledge. The class of knowledge is a subclass of the certainty beliefs. Propositions of which we can say that a particular person knows, are propositions that are based on a “state of cognitive achievement”, and are “based upon completed enquiry” (WC p. 13)


Moreover there is also the class of certainty beliefs that specifically relate to our cognitive organs and suitability of belief forming environment. Propositions relating to these context specific hinges also “unavoidable lack earned warrant at the point which they need to be made” (WC p. 36) But nonetheless both hinges need to be in place if any genuine warrant is to be achieved when the epistemic engagement is about states of affairs outside the epistemic agent. Unless type-III propositions are presupposed, what is locally accessible cannot provide indications of what lies beyond.


The overall picture is that whenever epistemic activity is undertaken in a responsible manner the context specific hinges will provide warrant acquisition for propositions that are “locally accessible” i.e., type-I propositions, experiential propositions, as they also are known. These type-I propositions will then contain, or have, the positive epistemic status sufficient to justify type-II, if the appropriate type-III proposition is presupposed. The underlying inferential structure, wherein type-I propositions can only be justified by type-II propositions is accepted, of course, by both Crispinstein and the skeptic. This noetic structure does not only display a general foundational structure, but specifically demands that the foundational belief(s) are of a propositional form, according to which the agent in question inferentially justifies type-II propositions by way of type-I propositions.


Let me relate this to a distinction, that I made in chapter 1 (p. 5), between access to the justifier (type-I) and access the justificatory efficiency of the justifier. According to the view under consideration it is not only required for the epistemic agent to have access to the justifier, but rather the agent must have access to the epistemic efficiency of the justifier. This access is not attainable when, for example, the agent in question doubts the relevant type-III proposition. For example, when S formed the belief that it is 12 o’clock after looking at his watch there would be some perceptual experiences that is (for S) accessible. But if the agent in question doubted MOORE type-III, e.g., then that experiential proposition would not be epistemically efficient to justify S’s belief that it is 12 o’clock, since, unless ‘the external world exists’ was presupposed, there would not be any clear idea about what time it could be. Wright, then, seems to be suggesting that S needs to be in a position to know the epistemic efficiency or justificatory efficiency of the perceptual experience before it can justify type-II propositions – the relevant type-III proposition must be taken for granted.


Remember also that we are entitled to trust our belief forming processes unless we have indications to the contrary. So, this stance is rejecting what seems initially plausible, i.e., the dictum that unless we can produce good arguments, or have good evidence for a proposition, we cannot trust it. But it nonetheless holds onto the internalist view on warrant acquisition, according to which a belief does only have warrant if it is suitably grounded in experiential propositions of type-I variety. It is, of course, precisely that move by the skeptic, that type-II must be grounded in type-I, that generates the skeptical problem. Wright does not undermine the workings of the skeptical argument; rather he accommodates or, as he says, suggests an intellectually satisfactory answer to the paradox.  


In the next chapter we’ll look at another response to the I-II-III argument – the response by Alvin Plantinga.


Chapter 3 – Plantingian Answer: externalism and Proper Function


In chapter 1, we asked the following question before the five points in the skeptical argument were elucidated: “What – if Moore’s warrant for his original premise is information-dependent – could put the needed collateral information […] in place?”  (WC p. 8) In the previous chapter, we saw, what I called, the Crispinsteinian answer to that question. That answer was a uniform response – or uniform attitude, as Wright calls it – to the I-II-III argument were a special kind of certainties beliefs, or hinges, can put the needed information in place. The answer was uniform even though, there is nothing in the I-II-III argument that requires a uniform response. For example, a direct perceptual theory could be used to facilitate an answer to the external world skepticism, and another strategy for skepticism about other minds. In this chapter, we’ll look at another, in fact uniform, response to the skeptical argument.


This chapter proceeds in this way. In the first section, we’ll look at a way to question the inferential architecture which the I-II-III argument presupposes, specifically as relates to perceptual beliefs. This will give an opportunity, in the following section, to look at Plantinga’s Warrant and Proper Function. In the final section, I’ll ask whether Plantinga is a direct perceptual theory or an indirect. I’ll also look at what he calls doxastic experience and finally point out the way that his response to the I-II-III argument is a uniform response.


3.1 Resisting the inferential structure: Plantinga and the Skeptic, initial considerations

It will prove an elegant entry to the Plantingian theory to look at a discussion that Plantinga and Quinn had some years ago.[20] But first let’s look at what Plantinga thinks is typical in belief forming situations. In (RBG) Plantinga says:


“It is possible that I believe that proposition [I see a tree] on the basis of the proposition that I seem to see a tree; in the typical case, however, I’ll not believe the former on the basis of the latter because in the typical case I’ll not believe the latter at all.” (RBG p. 49)



So, it is possible that I believe type-II propositions on the basis of type-I propositions – it is abstractly possible, and might even be the case, at times. Just as it is possible, that agent A only holds three beliefs B, C, and D, and believes each of them on the basis of the other two. Furthermore, Plantinga says, it is possible that all of the agent’s beliefs are held in a way that does not rest upon other beliefs. But usually, an agent’s noetic structure will be as the doctrine of foundation and superstructure says – some beliefs are justified by their relation to other beliefs, and others are not so justified, the latter of course being the basic beliefs. So, in the typical case, forming beliefs about chairs, trees, hands and computers is not done upon the basis of other beliefs – experiential or otherwise.


These observations do clearly not entail a total elimination of type-I propositions from our noetic structures; ‘in the typical case’ is vital here. It is sometimes the case, that I base the proposition in question (I see a tree) on the basis of the ‘seem’ proposition – if for example confronted with the skeptic, but usually there is no propositional attitude towards type-I.


Talking about basing a propositions like I see a hand in front of me (type-II) on a proposition like it seems to me that I see a hand in front of me (type-I) Quinn makes this suggestion:

“If the proposition expressed by [type-II] were indirectly justified by being properly based on the proposition expressed by [type-I], it would be no less well justified than if it were directly justified by being directly grounded in visual experience. Since, by hypothesis, my visual experience in those conditions suffices to confer a certain degree of justification on the proposition expressed by [type-II], the amount of justification that reaches the proposition expressed by [type-II] from that experience will not be less in those conditions if it passes by way of the proposition expressed by [type-I] than if it is transmitted directly without intermediary.” (OFT p. 478) [21]


The idea seems to be that even though a proposition like I see a hand (type-II) can be basic, not believed on the basis of other propositions like it seems to me that I see a hand (type-I), so can the former without loss of justification be believed upon the basis of the latter. This suggestion is only true if type-I propositions are in fact good evidence for type-II propositions; this is only true if the former constitute good non-circular evidence. Plantinga writes:

“It is exceedingly hard to see how to construct a cogent argument – deductive, inductive, abductive or whatever – from experiential beliefs […] to propositions which, like I see a hand [type-II], entail the existence of such material objects as tables, houses, and horses. But if experiential propositions do not furnish much by way of evidence for such propositions as [type-II], then if such a proposition is believed on the basis of such experiential propositions (and has no other source of warrant or positive epistemic status) it will have little if any warrant” (FT p. 305)[22]


Wright, we have seen, does not think it possible to construe a cogent argument from type-I to type-II across to type-III. Plantinga agrees. If type-II is formed upon the basis of experiential propositions, were the former do not have any other source of warrant then it will be poorly situated as far as positive epistemic status goes – “it will have little if any warrant.” These considerations are analogous to the considerations that underlie the I-II-III argument.


However, it does not follow that type-II have little by way of warrant if taken as basic. Why, we might ask, “should we suppose that experiential propositions alone are a source of warrant […]” (FT p. 305) Plantinga points out that Reid claims “that there is nothing but arbitrary partiality in holding […] that only self-evident and experiential propositions are the sources of warrant.” (FT p. 305) Call this the Reidian move[23].


How exactly does this influence our thinking about the skeptical argument? Well, if in a belief forming situation I do not rely, directly or indirectly, on type-I then I cannot face the circularity problem. This means, in relation to the I-II-III scheme, that (i) is rejected; rejected in the sense that when the skeptic says that type-II can only be justified on the evidence type-I propositions his suggestion is thought to be false. However, this does not entail that type-I cannot be among the reasons that I give, if the skeptic asked why I believe that there is a hand. It only means that if I invoke type-I, then it does not provide any further warrant for my belief that I see a hand – invoking them does not play a “warrant giving” role.


Let us step back for a moment. In Warrant and Proper Function Plantinga suggest that the philosophical tradition has two quite different suggestions as to how beliefs of type-II can receive warrant. Plantinga writes:

“But, as the history of modern philosophy up through Thomas Reid makes abundantly plain, it is at best extremely unlikely that there are any decent (noncircular) arguments – inductive, deductive, abductive – whose premises are self-evident propositions together with the appropriate experiential propositions, and whose conclusions are propositions entailing the existence of such things as tables, chairs, trees, and houses. Reid was correct, I take it, in agreeing with Hume (as he understood Hume) that such beliefs as [type-I] do not in fact constitute (noncircular) evidence for such propositions as [type-II]; if the latter get the warrant they have by virtue of being believed on the evidential basis of the former, then they have no warrant.” (WPF p. 97)


On the one hand, there are those who think that such beliefs get warrant only if it is accepted on the evidential basis of other beliefs, in particular beliefs about immediate experience – like type-I propositions. On the other hand, there is the view that type-II beliefs “get warrant for just by virtue of being formed in the right circumstances […].” (WPF p. 96)


There seems to be much by way of agreement between Wright and Plantinga as pertains to the issue of circularity or question begging. But, whereas Plantinga thinks type-II can get warrant in a basic way if “formed in the right circumstances”, Wright, we have seen, thinks that “type-II propositions can only be justified on the evidence of type-I propositions.” Both agree on the dim prospects for providing a non-circular argument for the existence of the external world. Whereas Wright insists that type-III of e.g. (PER) and (MOORE) functions as hinges, and therefore are not objects of knowledge, Plantinga thinks that these propositions can be objects of knowledge. For example, it self evidentially follows, form the warranted belief that I see a hand, that the external world exists. The vital question then becomes, what are the right circumstances?


By asking this question: ‘what are these circumstances, wherein the Plantingian epistemic agent can hold type-II propositions, in such a way that it enjoys more by way of warrant than it does if it is based upon evidence of type-I propositions’, we basically ask the question of how to think of warrant. In fact, we ask the question of how to elucidate what is contained in the 1-3 hinges:[24]

1.           Proper function of the relevant capacities.

2.           The suitability of the occasion and circumstances for their effective function


These two issues, among a few others, are central in Alvin Plantinga’s theory of warrant which I’ll look at in the following section.


3.2 Plantinga’s Epistemic Response: Warrant and Proper Function


Alvin Plantinga suggest the term warrant for that which distinguishes knowledge from mere true belief and sees it as an “elusive quality or quantity enough of which, together with truth and belief, [that] is sufficient for knowledge.” (WPF p. v) In this way, Plantinga continues the search for whatever precisely it is that makes the difference between mere true belief and knowledge. That the search though, bears more resemblance to externalism than to than to internalism (see p. 3) is evident when we see that he links warrant with proper function PF; the suggestion is that warrant is closely tied to the  proper functionality of cognitive organs. If warrant is coupled with the proper workings of the organs that ‘supply’ the cognitive information, and there is no telling from the inside whether those organs are working as they should, then the theory is appropriately called an externalist theory.

Proper function

Central is the idea of proper function. A first approximation: cognitive organs function properly when they are working as they are supposed to do according to a certain blueprint or design plan. The blueprint may be either evolution or God, or the conjunction of both.[25] Plantinga being a theist, would either be inclined to think of the design plan as God’s work; perhaps God’s work in conjunction with evolution wherein God is envisaged to orchestrate the evolutionary process.


It is not easy to give a fully satisfying explication of what proper function really is[26], but this does not pose a severe problem for the theory. The central idea is clear enough i.e. the idea as it is commonly used. We have an idea of a television functioning as it is supposed to, as well as we have an idea of the machine that these letters are typed with can function properly or improperly. Similarly, for biological beings, like ourselves; a normally functioning fifth grader is supposed to be able to read at a certain level. Similarly, medical science tells us that certain functions in biological beings are supposed to work in a certain ways. E.g. if there is a genetic variation it can results in malfunction of certain functions. Glycogonosis Type-III is such a genetically determined malfunction – malfunction of a particular function that is supposed to ‘release’ stored glycogen when the blood sugar level falls to a certain level, resulting in malfunction of the specific function.


The notion of warrant, then, is closely tied to issue whether it the belief is produced by cognitive faculties that are functioning as they are supposed to do. But is this enough? Clearly, it is not. Consider: if the cognitive process that leads to belief is functioning as it should then the belief is knowledge. If a scientist in his laboratory is not under the influence of drugs, say, and his cognitive faculties are producing beliefs by way of a proper function then the beliefs will be knowledge. But, the question is whether proper function is sufficient, or enough, for knowledge. Is it possible to envisage a situation wherein a cognitive agent producing beliefs by way of proper function, but that the beliefs are in fact knowledge? Consider, e.g., a case where a subject S after he has got his annual cognitive checkup, S passes the test with flying colors and is in splendid epistemic condition. Suddenly, and without S’s knowledge, S is transported to an environment wholly different from earth; S finds himself on a planet revolving around Alpha Centauri. There conditions are quite different; elephants, we may suppose, are invisible to human beings, but emit a sort of radiation unknown on earth, a sort of radiation that causes human being to form the belief that a trumpet is sounding nearby. An Alpha Centaurian elephant wanders by; S is subjected to the radiation, and forms the belief that a trumpet is sounding nearby. But, in fact there is no trumpet in the nearby vicinity, let alone does it sound. There is nothing wrong with S’s belief forming mechanism, but the belief cannot be knowledge, even if in fact the belief is true. This suggests that proper function must be supplemented by other constraints. (The Alpha Centaurian elephant example is Plantinga’s WPF p. 6) 


The supplement is to integrate some clause about the environment. It wasn’t S’s cognitive faculties (CF) that were to blame; rather it was the relation between the environment and the ‘tuning’ of S’s CF. S’s CF were not tuned for the environment, resulting in belief that was formed by way of PF but missed the mark – the belief was false. Integrating these considerations we have, according to Plantinga the “zeroeth approximation”. Let (WPE) stand for Warrant, Proper function of cognitive organs in an appropriate Environment


(WPE) A belief has warrant for S iff the cognitive faculties responsible for the belief formation are functioning properly in an environment suitable for the exercise of those cognitive faculties.


Initially this seems to be to be a fair approximation. But, again, it is easy (at least if you are familiar with some modern objections to religious belief by Freud (which we glanced at in chapter 0) and Marx) to envisage situations were WPE is satisfied but the belief produced is not eligible for the predicate knowledge. Plantinga ask us to consider Freud. Freud thinks that it is not that there is anything wrong with subjects forming religious beliefs – it not the case that the cognitive organs of those subjects are malfunctioning, nor that they are functioning in an environment that isn’t suitable for the exercise of those faculties. Instead of aiming at forming true beliefs, these beliefs are form in order to enable the subject to cope, or otherwise get by, in this ‘present evil’ I’m tempted to say. In other words – the beliefs are simply not aimed at truth, and therefore do not qualify for the favorable status of being knowledge. They are simply the result of mechanisms that are aimed, by way of wishful thinking, at something else than true beliefs.


Again, we need a further constraint on WPE – i.e. a truth constraint. This is no surprise, though – most epistemological theories, those worth the name, I think, make it a requirement that a belief needs to be true if eligible for the predicate knowledge. Here the repair: call this approximation (WPET) for warrant, proper function, environment, and truth:


(WPET) A belief has warrant for S iff the cognitive faculties responsible for the belief formation are functioning properly in an environment suitable for the exercise of those cognitive faculties, and that the exercise is aimed at true beliefs.


Is this sufficient as an account of warrant – enough of which makes the difference between mere true belief and knowledge? The answer is no, it is not sufficient. How is the insufficiency brought to light? Here is how Plantinga brings it about.

 “For suppose a well meaning but incompetent angel – one of Hume’s infant deities, say – sets out to design a variety of rational persons, persons capable of thought, belief, and knowledge. As it turns out, the design is a real failure; the resulting beings hold belief, all right, but most of them are absurdly false. Here all three of our conditions are met: the beliefs of those beings are formed by their cognitive faculties functioning properly in the cognitive environment for which they were designed, and furthermore the relevant modules of the design plan are aimed at truth […]. But the beliefs of these pitifully deceived beings do not have warrant.”(WPF p. 17)


He continues to suggest a solution by way of repair, by asking what should be added?

“That the design plan is a good one – more exactly, that the design governing the production of the belief in question is a good one; still more exactly, that the objective probability of a beliefs being true, given that it is produced by cognitive faculties functioning in accord with the relevant module of the design plan, is high. Even more exactly, the module of the design plan governing its production must be such that it is objectively highly probable that a belief produced by cognitive faculties functioning properly according to that module (in a congenial environment) will be true or versisimilitudinous. This is the reliabilist constraint on warrant, and the important truth contained in reliabilist accounts of warrant.”(WPF p. 17)


To repair WPET we require of the design plan that the exercise of the cognitive faculties is in agreement with a good design plan; not only that the exercise is suitable according to the design plan (blueprint, if you prefer) but that it is suitable according to a good design plan. In Warranted Christian Belief we get this formulation, call it (WPETG):


“a belief has warrant for a person S only if that belief is produced in S by cognitive faculties functioning properly (subject to no dysfunction) in a cognitive environment that is appropriate for S’s kind of cognitive faculties, according to a design plan that is successfully aimed at truth.” (WCB p. 156)


As is clear from these elucidations, the 1-2 hinges (p. 27) bear particular resemblance with Plantinga’s notion of what makes the difference between true belief and knowledge. Now that we have the underlying notion of warrant in place, we see that type-II propositions can be basic in the sense that they do not rest upon type-I propositions. Here is one of Plantinga’s favorite examples. When he is appeared to in a characteristic and familiar way, he says, he simply forms the belief, that he sees a tiger lily. However, he is quick to add, “of course I don’t ordinary infer that I see a tiger lily from the belief that I am being appeared to in that way (together with other beliefs that I hold); nor do I hold the former belief on the evidential basis of the latter. I do not first note that I am being thus appeared to, and then reason to the belief that I see a tiger lily.” (WPF p. 93, underline: my emphasis) In fact, he supposes it “wholly obvious” (WPF p. 93), that in perceptual situations we do not form beliefs about the experiences in question. That they can be basic, obviously, does not imply that they are incorrigible, infallible, or certain or anything else of that sort.


Most importantly, however, is the fact that according to this construal basic perceptual belief P can be warranted for S (and thereby eligible for the predicate knowledge), if P is produced in S by cognitive equipment that functions properly in an environment that is appropriate for S’s cognitive equipment, according to a design plant that is successfully aimed at truth.

3.3 Closing stage

3.3.1 Plantinga a disjunctivist?

Why does Wright disagree about the noetic structure, according to which type-II can only be justified by type-I propositions? One possible candidate is the so-called argument from hallucination. Suppose that I see a real world computer in front of me – I veridically perceive a computer – by way of experiential evidence P. Furthermore, imagine that, instead of veridically perceiving the computer, I have a hallucinatory experience of the computer, by way of experiential evidence P (it needs to be reminded). In the latter case, it is by definition nothing that I perceive – there is in fact no computer in the vicinity – in the external world; but still I have precisely the same kind of experience P as when I veridically perceive the computer. Because I see something, when I have the hallucinatory experience, and we know that it is not the computer, there must be something else that I perceive, and this something else is some kind of intermediate object of perception – a sense-datum, say.  Furthermore, it is argued that because the two situations are experientially indistinguishable, it is therefore concluded that in veridical perception I’m also aware of an intermediate object, instead of the computer. The argument from hallucination, then, is thought to give us reasons to postulate an intermediate perceptual object, instead of aspects or properties of the object of perception – the computer. The conclusion drawn is that when S perceives X it must be some intermediate object X* which is the direct object of perception – a sense datum.


There are basically two overall ways to respond to the above argument. The first is sympathetic to the above argument, in that theories of perception need take into account the fact of hallucinatory experience and integrate it in the theory of perception. The second overall line is that of the naïve perceptual theories, which say that nothing follows from the argument from hallucination. This view is sometimes called the disjunctive view of perception, in that it claims that in veridical perception there is one structure to the experience, whereas in hallucinatory experiences it is another kind of structure: there is no single experiential structure – the highest common factor[27] – in   these two cases.


I said that the argument from hallucination is a possible candidate for a philosophical motivation for the I-II-III structure, but it is not a good candidate. It would be a good candidate if any position that allowed type-II to be warranted otherwise than by the evidence of type-I adopted a disjunctive position. In this connection, it is appropriate to ask whether Plantinga adopts the disjunctive point of view. The short answer is no. Therefore the highest common factor view cannot be Wright’s only reason, even though, it could be among a set of reasons for thinking that “type-II propositions can only be justified on the evidence of type-I proposition” is true.


What Plantinga says about beliefs being warranted as conditioned in (WPETG) and where the belief in question does not rest upon other beliefs, does not help us to locate his favored perceptual theory. Prima facie, it might seem as if he is a disjunctivist. But that does not fit with what he says towards the end of chapter 10 of WPF where he deals with, what he thinks is the evidentialist’s parochial view on evidential basing, wherein beliefs must always be based upon evidence in order to be warranted – propositional evidence.


According to proper functionalism, the proper relation between perceptual experience and the belief formed will be that of the well functioning agent. For example when I am appeared to computerly I do not form the belief that there is a cello in front of me; if I formed this latter belief it would be a token of something pathological. But can we be a bit more precise? We have seen that perceptual beliefs are formed in response to experience and not propositional evidence. Are these beliefs then not formed on the basis of evidence? Take e.g. the belief that I see a squirrel. Plantinga points out that in these kinds of beliefs – beliefs of the form, I see, I believe, I judge – there does not seem to be any evidence of any straightforward kind. These beliefs do not seem to be “formed on the basis of evidence” in the way that ordinary perceptual beliefs are; instead they “resemble memory and a priori beliefs.” But, he adds, in parenthesis: “Of course, we can perfectly properly add that beliefs of this sort are nonetheless formed in response to experience.” (WPF p. 190) He also considers such beliefs about himself as “I am now being appeared to in that tiger-lily way” and “I am not identical with that tiger-lily experience” and says of them, that they have, or can have, a high degree warrant.


We have two cases here: 1. The case where the belief is not formed on the basis of evidence, but still can be formed in response to experience. 2. The case where “I am not identical with the tiger-lily experience” the proposition is not believed on the basis of a perceptual experience.


Plantinga says that in neither of these cases is he “of course” “directly” aware of the object in question. (WPF p. 189) The sense of being “directly aware” is given in (WPF p. 53):

(DA): I am directly aware of something if I am aware of it and am not aware of it by way of being aware of something else.


He takes it that when he is appeared to such and such, then there is something that is the object of his awareness – “a thus and so appearance.” (WPF p. 53) “When I believe that I am not identical with the tiger-lily experience but am instead the subject of that experience, the one who is aware of it, I am not directly aware of myself. But of course the same goes for tiger-lily appearances; the latter, not the former, is what I am directly aware of. (Of course, in another and perfectly good sense of ‘directly aware of’, I am directly aware of the tiger lily.)” (WPF p. 189., underline my emphasis)


Now, these remarks do not fit very well with being a disjunctivist. There is this something else that one is aware of when perceiving a tiger-lily way. This seems to complicate matters. How can he say that type-II propositions can be warranted, and typically are warranted, in a basic way, and simultaneously hold, that when he is appeared to in that tiger-lily-way, that the object of the immediate object of perception is not the tiger-lily? Does this not simply imply that there must be some kind of inference from the object of perception to the perceptual belief?


Plantinga would answer in the negative: there is no inference; at least usually, there is no inference. When I am appeared to in a certain way then it is not as if I infer (however quickly) to the belief that so and so. That I am appeared to in a certain way does, as we saw above, not give me particularly good resources to infer – deductively, inductively or abuductively – the existence of the object of my perception. Rather, according to proper functionalism, a well functioning epistemic agent will simply hold the belief in response to the perceptual experience – hold it in a basic way.


Plantinga’s way then, to deal with the I-II-III argument is not facilitated by a certain perceptual theory – except, of course, that his wider epistemology allows epistemic subjects to be directly warranted in (as opposed to directly aware of), for example, the belief that there is a computer in front of me. That is, his answer to the I-II-III is independent of his perceptual theory.


3.3.2 Doxastic Experience

In belief forming situations there is a ‘special’ kind of evidence associated with the formation – not only the “sensuous imagery” that is associated with perceptual belief, but the “felt attractiveness” of the belief, or the psychological inclination to believe certain propositions, that I mentioned in chapter 1 p. 6-7 For example, when I hold the belief that it is I who see a tiger lily, instead of, say, you or Socrates, there is the felt attractiveness present. It is wholly obvious that it was I who held the belief and not you.


Consider also memory belief. You remember meeting Mark in Copenhagen last week and not in Madrid. You remember that Mark was there in Copenhagen. This proposition, says Plantinga, “has about it a sense of rightness, or fittingness, or appropriateness – as opposed, say, to the proposition that it was Tom there on the occasion.” (WPF p. 190-91) Correspondingly, for a priori beliefs, there is the felt attractiveness that is attached to this simple logical law: If Sam is happy, then someone is happy. The inclination to accept this a priori belief is not grounded in the fragmentary imagery that might, or might not go along with entertaining the belief. Instead, there is the felt attractiveness – this impulsional evidence or doxastic evidence, which he calls it in Warranted Christian Belief. (WCB pp. 110-111, 183, 264, 333) In belief forming situations then, “there are at least two kinds of phenomenology: there is the sensuous being-appeared-to type phenomenology, but also the perceived or felt inclination to believe.” (WPF p. 191)


These two epistemological responses – the Crispinsteinian and Plantingian – to the I-II-III argument correspond to two different views on warrant acquisition. On Crispinstein’s view, warrant acquisition arises, or gets generated, only when the subject in question has introspective access to the ground of the belief and the justificatory efficiency of the belief. Even though the first constraint can be fulfilled without a hinge being in place (the skeptic which doubts type-III propositions, can have access to the experiential proposition) the latter constraint does not obtain unless the hinges are in place (if doubt is entertained about the type-III in question, then pointing to a experiential propositions won’t do much by way of justifying type-II.) It is the information dependency of type-I (in order for it to have efficient justificatory power) on type-III that puts in place the justificatory efficiency of type-I.


According to Plantinga, beliefs produced by proper functioning cognitive faculties, where they function in a suitable environment according to a good design plan wherein the belief formation is aimed at truth acquire warrant by being so produced.


The core divergence among these two views, then, is that Wright’s view requires that the epistemic subject has access to the justificatory efficiency of the belief, whereas Plantinga rejects this requirement – Wright is an internalist on warrant acquisition whereas Plantinga is an externalist.

3.3.3 Uniform attitude

So far, I have not done much to the effect to show how Plantinga’s response to the I-II-III argument hinges on a uniform attitude. Let me get at it in this way. We remember that for Crispinstein, cases that are in danger of generating transmission failure are cases were type-I propositions depend upon type-III propositions for their eligibility to justify type-II propositions. This is particularly true of some of the cases that we looked at in chapter 1 i.e. (MOORE), (PER) and (MIND). The answer in chapter 2 was that the respective type-III propositions, of these arguments, function as hinges.


Plantinga, on the other hand, has a similar response. It is a sign of a proper functioning agent that the respective type-III propositions will be part of his noetic structure. Furthermore, if these type-III propositions are produced according (WPETG) then they are knowledge. (It is not normal to meet agents who go around trying to convince their next-door neighbor that there is an external world or that there are other minds. The normal epistemic situation will be that type-III beliefs, will self evidentially follow from their respective type-II beliefs that usually are warranted in a basic way.) This, of course, stands in sharp contrast to the positions advanced in chapter 2, according to which, type-III propositions are not to be evaluated in terms of the idea of knowledge. The principle underlying this view on knowledge is that if some proposition is to be evaluated in terms of knowledge, then it must be the result of a cognitive achievement – completed enquiry. I think it fair to say, that it must be the conclusion of a cogent argument in order to be knowledge; perhaps this is exactly what a cognitive achievement or completed enquiry is thought to capture. In chapter 1, I said “it is vital, it seems, that we be able to produce good reasons – arguments – for our beliefs”.


What is the thought here? Something along these lines, I think. Philosophers are in the business of arguing for their views; they are not like the dogmatist, who just proclaim that so and so. Perhaps, a good example would be, theologians who proclaim that so and so, according to so and so. Philosophers in contrast, need to argue for their positions. Now, arguments need some premises and rules of inference. Unless, the thought might go, we can produce valid arguments, from premises using deduction, induction or abduction as rules of inference, then we’re not in a position to philosophically embrace the truth of the conclusion. 


But, according to Plantinga, this is not true. Instead he would say that these beliefs are not ordinary the result of arguments. The proper functioning agent would not (ordinarily) believe them on the basis of arguments, or inferences form experiential propositions. Instead, Plantinga thinks that it is part of the design plan to ascribe mental states to persons in certain circumstances, to affirm the existence of the external world, to warrantedly believe that the world did not pop up 5 minutes ago with all its traces of older civilizations etc. In certain experiential circumstances, e.g., when I see a computer in front of me, is it not as by of some inferential step, that I conclude that there is a computer in front of me. Rules of inference do not dictate much by way of what sort of belief I form to what sort of experience. As, Plantinga says: “Proper function, of course, dictates much more. If (in my cognitive circumstances) I form the belief that I see a tiger in my backyard in response to being appeared to in that tiger-lilyish way, I display cognitive dysfunction.” (WPF p. 93)


So, Plantinga’s proper function view on warrant – that which distinguishes mere true belief from knowledge – has it that type-II propositions are not usually arrived at by arguments or inferential warrant. Rather they are warranted in a basic sense; similarly for a host of other beliefs e.g., those singled out on page 26 – A and B hinges. In the Plantingian optic, these beliefs can be warranted, and therefore can be defended in the idea of knowledge. This stands in contrast to Wright’s view on knowledge, but as a meta-epistemological response to the I-II-III argument both stand on par as far as a uniform response concerns. According to Crispinstein A – C hinges, and 1-2 hinges are taken for granted, the former are characterized as certainty beliefs, the latter presuppositions that must inevitably be taken for granted in any epistemic project. Similarly for Plantinga, that our cognitive faculties are reliable we cannot but take for granted, unless there are indications that our cognitive faculties are impeded by an unusual arrangement of the environment, say. And, as far as A – C hinges are concerned, these are beliefs that the proper functioning agent will have as part of his noetic structure, beliefs not arrived at by way of argument or inference, but nonetheless qualified for the predicate knowledge.


I now turn to look at the prospects for construing a religious epistemology according to these two views.


Chapter 4 – Religious belief: Crispinsteinian and Plantingian


In chapter 1, the I-II-III argument was presented. Arguments of the I-II-III structure were presented as responses to the skeptic, but they did not answer to the skeptic’s satisfaction is doubt about the type-III propositions on question. I suggested that the (MOSES) and (THEIST) could be set fort as answers to the religious skeptic. Both of these presuppose what they intend to prove, as do (MOORE), (MIND) and (PER). In chapter 2, the Wittgensteinian hinge response was presented, followed by, in chapter 3, the Plantingian response. Both of these responses take it that arguments that attempt to prove the existence of the external world, or arguments for the believability or rationality of belief in God, fail in the light of the skeptical argument. The Crispinsteinian accepted all of the five points in the skeptical argument. (see page ?) But this does not lead to skepticism because if the doubt is to be entertained then there needs to be something which the doubt must be revolving around; i.e. the special class of hinge propositions. The Plantingian, on the other hand, blocked the skeptical suggestion by pointing out that type-II propositions do not rest upon type-I propositions, for their epistemic status. The crux of this response is that there is a way that cognitive agents are supposed to function. The proper functioning agent will not believe type-II propositions on the evidential basis of type-I propositions; instead, he will usually believe type-II propositions in a basic way.


In the first part of this chapter, we’ll look at how theistic belief can be held in accord with the Crispinsteinian stance. I’ll present this possibility by way of analogy between theistic belief and type-III propositions. If the analogy between theistic type-III and other type-III propositions is sufficiently strong, then we can set theistic type-III on par with other type-III proposition. The underlying idea is that we should not demand more or less, by way of epistemic requirement for beliefs in one area than in another. But, as we’ll see, even though the analogy is strong enough in some respects, is it not sufficiently strong in other vital respects. In the second part, I’ll present the Plantingian view as he sets it forth in his Warranted Christian Belief, according to which theistic belief can be warranted – warranted in an basic way – if produced by properly functioning faculties.


4.1 The Crispinsteinian Stance

In this section we’ll look at the prospects for taking theistic belief to be on par with Crispinsteinian certainty beliefs – hinges. Theistic arguments (MOSES) and (THEIST), we have seen, share the non question begging difficulty similar to (MOORE); they fail to transmit warrant across to type-III. Prima facie then, I seem to be able to provide as my rationale for believing in God, by way of a hinge response analogue to the hinge response given by Wright in chapter 1. I could simply say: even though I cannot produce a non-question begging argument for (THEIST) III, I can still have it in my noetic structure – as a hinge belief. On the ground of the general epistemological response to the I-II-III argument, presented in chapter 2, we can say of (THEIST) III that it could, be held to be epistemically on par, its evidential predicament non-withstanding – as long as it is analogue to (MOORE) III.


The Crispinsteinian suggestion, then, is to take (THEIST) III as a hinge belief. If something like (THEIST) III was not taken for granted then whole idea of playing the theist game (in Wittgenstein’s peculiar sense) would not be played; it is a presupposition for that particular way of life. It is furthermore of interest that just as (MOORE) III is not held, according to the Crispinsteinian point of view, as a proposition that has gone through some ‘cognitive achievement’ commonly thought necessary for empirical propositions, it nonetheless carries significant epistemic relevance – significant relevance for the religious belief formation. Just as non-religious type-III propositions are hinges upon which the whole game of giving and asking for reasons turns, so the religious type-III functions as hinges upon which the game of giving and asking for religious reasons turns.


If the theist gave up the religious type-III proposition this would enforce a massive regulation in his belief system. This is the case with Paul e.g., which on a regular basis, let us suppose, due to his theological convictions, has helped a certain person with his car, twice a year in the ’tire change season’. Paul does this because he thinks that it is the right thing to do from the point of view of his theological convictions, say. Now, Paul comes to struggle emotionally with the argument form evil, perhaps he hears some persuasive presentation of the argument. (Alternatively he is informed through the media of the suffering of the Iraqi people through the war that the allies fought against Irak.) The argument, in its deductive version, has it that evil in the world is incompatible with the existence of an omnipotent and wholly good being. Paul’s noetic structure goes through a change – he simply negates his belief in (THEIST) III. Perhaps he replaces it with Naturalism, according to which the evolution of life is anchored in random genetic mutation and natural selection. The change that Paul’s cognitive frame goes through needs a massive regulation of his beliefs. First off all, he cannot continue helping this person on his theological grounds – some serious revision seems to be required. If, as I have suggested, he replaces his (THEIST) III with Naturalism[28] then his view on the significance of the living sphere would be considered as a cosmic coincidence, instead of being considered as orchestrated by God. But this cannot be because he as reasons or evidence for doing so – that would not fit with the Crispinsteinian picture which we are considering. (It is a characteristic of type-III propositions that it is difficult to see what might count as evidence against it. (See WC p. 25 and chapter 2 p. 25)


Suppose we look a bit closer at the thought. If I hold theistic belief in a hinge kind of way, then it certainly cannot be said to rest upon other of my beliefs. Wright, we remember, suggested that the external world proposition is “beyond supportive evidence” (See ch. 2 p. 23 and WC p. 25) – we cannot invoke any evidence for it. We cannot imagine a scene wherein evidence is provided against the inclination that we have for the existence of the external world, say. This is a peculiar quality of type-III propositions. Furthermore doubt about them would involve extensive undermining of investigative procedures and norms of assessment. This stands in clear opposition to the other type of hinges (A and B, see p. 26): doubt about which would go against overwhelming body of evidence. (WC p. 27)


If what Wright claims is the case with hinge beliefs also is the case with regard to (THEIST) III, it is hard to make sense of why there are some that think that, indeed, there is evidence for and against (THEIST) III – why down through the history of western civilization the issue of religion has had proponents and opponents, which have set forth arguments, the premises of which have been supported by evidence. What about the great philosophers[29] that have thought that by way of considering the argument form evil, both in the logical and evidential version, we are rationally justified in positively affirming the non-existence God? The hinge suggestion does not seem to fit well in relation to e.g. the arguments set forth in relation to the problem of evil; nor does it fit well in relation to what (perhaps) what (some) scientific inspired philosophers would conclude from recent advance in science.[30] How can we e.g. think of these arguments, if there is no evidence in the neighborhood?


So, as far as the idea of ‘beyond supportive evidence’ is concerned, it is clear that (THEIST) III does not stand on par with type-III hinges.


Furthermore, (MOORE) III has evidently a more direct influence on our current epistemic situation – as regards our epistemic interaction in a complex world, than does (THEIST) III. Consider Paul who is genuinely convinced of the negation of (MOORE) III – there is no external world. Perhaps, he is persuaded of Wright’s argument, but thinks that the hinge answer bluntly unbelievable. ‘How might Paul fare in everyday interaction in the word’ we can ask. For example, if he is about to cross the road in some big city – Chicago say. The ‘walk now’ sign is on. He starts walking across. Suddenly he hears the noise of a steaming trucker; as he looks to the left he sees the truck is coming towards him, with a speed, as far as he can tell, that indicates that its driver is not able to stop the vehicle before it hits Paul. Paul does what is in his capacity of avoid the steaming truck. Paul is timely – he barely escapes, only with smallish bruises on his knees, as he throws himself away from the truck. If Paul were thoroughly convinced of the non-existence of the external world, then he would not have acted as promptly as he did.[31] Just by way of being appeared to in a certain way, Paul acts.


The hinge response encapsulates a philosophical response to the non-question begging demand for (MOORE) III – it is ultimately a philosophical way to say that we can still go on epistemically engage in the world, even though we cannot claim any of (i) to (v) false. Is there any way we can say the same of (THEIST) III as pertains to practical indispensability? Well of course, if you engaged in activity wherein belief in the existence of God was a prerequisite – in worshiping, say – then it seems likely that, we needed to have that proposition in place; in that sense it could have some practical influence. We could, that is, if we engaged in such activity, use the hinge response, to rationalize our activity – and unless we had it in place, it would seem that the practical activity (worshiping) would loose its value.


This last observation is not strictly to the point though. It seems that if we sincerely were convinced of the negation of (MOORE) III, then it would have fatal consequences – fatal consequences that do not seem to affect theistic belief. Exempla gratia is it clear that Paul, if he was genuinely convinced of (MOORE) III, would face quick extinction in a modern city. This does not seem to be the case for theistic belief. Also at this juncture it seems that we cannot draw the analogue as tight as we, perhaps, wanted.


Another vital issue is that type-III hinges are exempted from doubt – they are not doubted by the reasonable person. Perhaps religious type-III are only doubted by non-believers, and never doubted by believers? I think it no exaggeration to say that, most theists if not all, undergo doubt about theistic convictions at certain moments in their life. This, for instance, was the case with the disciples, even after that they had witnessed the resurrection of the Second Person in the Trinity. It is said of them that even when they were in joy, they still disbelieved and were wondering; particularly famous, of course, is the story of Thomas who said, after the other disciples had informed him of the risen Christ (he was not present when the others first met him), “unless I see in his hands the print of the nails, and put my hand into his side, I’ll not believe”.[32] After his wish has been fulfilled, let us suppose, the confidence was reinstated. In any case, doubts seem often to be involved in theistic beliefs; this is not the case with type-III propositions. Furthermore, it cannot be said of theistic type-III that they are not doubted by the reasonable man, or that the reasonable man will not doubt (THEIST) III. (See OC 220) In fact I know quite a number of people that doubt that proposition, people that I take to be reasonable.


The major dis-analogies between type-III propositions and theistic type-III propositions: 1. Regular type-III hinges are beyond supportive evidence, whereas theistic type-III are not. 2. The consequences of neglecting type-III hinges, it seems, far exceeds neglecting theistic type-III. 3. Type-III propositions clearly involve empirical matters whereas theistic type-III do not. Doubting (MOORE) III would disable all empirical enquiries, as Wright says. Doubting (THEIST) III would not disable all empirical enquire – it would undermine all theistic enquiry, so it is restricted to matter that are clearly not empirical in nature. The enquiry involved seems rather to be of a spiritual nature. 4. The reasonable man does not doubt (MOORE) type-III, whereas quite a number of, what I take to be reasonable men, doubt (THEIST) type-III.


In summary, we can say that even though there are some similarities between regular type-III hinges and theistic type-III propositions, there are also major dis-analogies. This does not give us reason to set theistic beliefs on par with Crispinsteinian certainty beliefs. Let us turn to an exposition of Plantinga’s suggestion for how theistic belief can be warranted. 


4.2 The Plantingian stance – The Aquinas/Calvin Model

Before I turn to chapter 6 of Warranted Christian Belief (WCB) where Plantinga presents a model for how theistic belief can be Plantinga warranted, is it appropriate to give a rough outline of the project in WCB. The book contains four parts. The first part of the book asks whether there is a question at all, as pertains philosophical dealing with theism: is there reason to think that the concept God does apply at all? This is a de facto question as opposed to a de jure question, the former relates to the truth, the latter to the rationality or believability of theism. The second part asks what the jure question is; is it that theistic beliefs lack justification, are irrational, unbelievable, lack coherence etc. It comes as no surprise that Plantinga pictures the de jure question as being whether theistic belief is warranted or not. The third part presents two models wherein theistic belief can be warranted. The fist model pertains to theism generally, as we know them in the theistic traditions. The latter model is a specific Christian model. The forth and last part looks at various defeaters that are thought to undermine theistic beliefs.


The core project of WCB is that of establishing the reasonableness of Christian belief, this, Plantinga suggests, is best done by a model. How are we to think of a model? Plantinga gives this rough idea: “to give a model of a proposition or state of affairs S is to show how it could be that S is true or actual. The model itself will be another proposition (or state of affairs), one such that it is clear (1) that it is possible and (2) that if it is true, then so is the target proposition.” (WCB, p. 168) Call this the Plantingian defense strategy.


Plantinga draws from Thomas Aquinas and John Calvin in his initial[33] characterization of the model; (call it the A/C model) even though he uses Calvin mostly, is it not because he thinks that Calvin is a “cynosure of all eyes theological, but because he presents an interesting development of the particular notion.” (WCB. p. 170) It is pointed out that both Calvin and Aquinas concur on the claim that there is a kind of natural knowledge of God. According to Thomas Aquinas: “To know in a general and confused way that God exists is implanted in us by nature.”[34]


Paul the Apostle writes to the Romans “For what can be known about God is plain to them, because God has shown it to them. Ever since the creation of the world his invisible nature, namely, his eternal power and deity has been clearly perceived in the things that have been made.”[35] This thought is elaborated in both Calvin’s commentary on the epistle and further in his Institutes of the Christian Religion; I’ll quote:


“There is within the human mind, and indeed by natural instinct, an awareness of divinity. This we take to be beyond controversy […] God himself has implanted in all men a certain understanding of his divine majesty […] Since, therefore, men one and all perceive that there is a God and that he is their maker. […](T)here is, as the eminent pagan says [Cicero], no nation so barbarous, no people so savage, that they have not a deep seated conviction that there is a God. […] Therefore, since from the beginning of the world there has been no region, no city, in short, no household, that could do without religion, there lies in this a tacit confession of a sense of deity inscribed in the hearts of all.”[36]


The basic idea is that there is a kind of faculty or cognitive mechanism, which Calvin calls sensus divinitatis (SD), or sense of divinity. This faculty produces, in a wide variety of circumstances[37], beliefs about God. These circumstances, Plantinga suggests, “trigger the disposition to form the beliefs in question; they form the occasion on which those beliefs arise.” (WBC p. 172) In this respect the beliefs formed in this way resemble perceptual and memory beliefs. Similarly, the model has it, we form theistic beliefs, in these circumstances, without making an inference; we simply form, or perhaps better, these beliefs are formed in us. Another facet, which Plantinga suggests, to be drawn form the above, is that the “awareness of God is natural, widespread, and not easy to forget, ignore, or destroy.”(WBC p. 173) Now, this is proposed as a model; the model is not intended to be accepted by everyone; nonetheless, it is central to the model, that humans are God’s creation.

4.2.1 Basicality with respect to Justification and Warrant

According to the A/C model, the beliefs triggered by the SD are not arrived at by inference or argument. They are simply formed in an immediate or basic way. As Plantinga says: “upon the perception of the night sky or the mountain vista or the tiny flower, these beliefs arise within us. They are occasioned by the circumstances; they are not conclusions from them.” (WCB. p. 175) The upshot of these considerations is that the output of the SD is basic, in the sense that it is not accepted on the evidential basis of other propositions.


Are theistic beliefs produced by SD also basic in respect to justification? Justification considered here specifically in deontological terms, i.e. can S be within his epistemic right, or is S irresponsible, or is violating no epistemic or other duties in holding these beliefs? For a philosopher who accepts justification as that which transforms true belief into knowledge, it is clear that he can be so justified in holding the output occasioned by this faculty. The question to ask here is whether the beliefs formed are justified in the deontological way. Unquestionably this can be. Consider S who forms theistic beliefs in response to some circumstances. Now, S finds himself with this belief. He then reflects critically upon the objections made against theistic belief. He reads his Freud, Marx, Dawking, and Dennet as well as, say, and some proponents of higher criticism; but he finds these objections missing the mark. How can we sensibly claim that he is irresponsible with respect to some epistemic duty?


In line with Plantinga’s general epistemology he suggests that there is another sense in which a belief can be properly basic. p is properly basic for S in this sense “if and only if S accepts p in the basic way, and furthermore p has warrant for S, accepted in that way.” (WBC p. 178) This of course suggests that if the belief is warranted and held in such a way that it is not held upon evidential basis other beliefs i.e., it is not inferred form other propositions, then it is properly basic, (produced by cognitive faculties functioning properly in a congenial epistemic environment according to a design plan successfully aimed at truth.) If we imagine a person S who holds theistic belief in the basic way, triggered in certain circumstances by SD, but lacks warrant, then this is due to some cognitive malfunction, or to a “cognitive faculty’s being impeded by such conditions as rage, lust, ambition, grief, and the like; it can also be because the bit of the design plan governing the production of the belief is aimed not at truth but at something else (survival e.g.), or because something in the testimonial chain has gone wrong (one of your friends has lied to you), or for still other reasons.” (WCB p. 178)


If we grant that epistemic subjects do form beliefs by way of SD, and suffice that the cognitive organs have evolved according to a design plan wherein the production of true beliefs is among the primary functions, then if the faculty is functioning properly, the output is warranted. (Provided that the environment is suitable for the production of such beliefs) Sure, these considerations might not be much of an argument for the skeptic. The skeptic will point out that there are several difficult notions on the scene that he finds highly suspicious. E.g. the notions of ‘design plan’ and ‘truth’ do not, in a naturalistic setting, get as far as on the scene. The naturalist – perhaps with a capital N – will, perhaps, insist that the primary function of the cognitive organs is in line with the four F’s: feeding, fleeing, fighting, and reproduction rather than ‘truth-seeking’.[38] It is only intentional agents, the suggestion might be, which can predicate design or teleology upon nature. The naturalist (of course?), denies any intentional agents standing outside the evolutionary process – in nature there is no teleology.

4.2.2 Perceptual or Experiental Knowledge?

If we grant that the beliefs formed by way of SD are such as the model suggests, then the question arises whether the warrant enjoyed, (if enjoyed at all, that is) is by way of perceptual experience. Plantinga suggests “that the way to think of perception strictly so-called is such that it essentially involves specifically sensuous imagery.” (WCB p. 181) William P. Alston notes in his Perceiving God, that sensuous imagery does not typically accompany putative perception of God. In this case, surely, we could not say that warranted out-put of SD is by way of perception, strictly speaking that is. But, the suggestion goes (given the existence of God) there could certainly be something very much like perception of God.


“This something, therefore, can properly be called ‘perception’ in an analogically extended sense of that term. To the believer, the presence of God is often palpable. A surprising number of people report that at one time or another, they feel the presence of God – where the ‘feeling’ also doesn’t seem to go by way of sensuous imagery. Many others (by no means for the most part spiritual heroes or even serious believers) report of Hearing God speak to them. And among these cases, cases where it seems right or nearly right to speak of perceiving God (feeling his presence, perhaps hearing his voice), there is great variation.”(WCB p. 181)


Plantinga wants to make it clear that he in no way doubts that perception of God is possible, or indeed actual. But, the question whether the knowledge of God gained by SD is warranted by way of perception is not answered even this granted. If we grant that there is such a faculty as SD then it is fairly obvious that the operation of the faculty will always involve the presence of experience of some kind or other, even if sensuous imagery is not present.


All right, what we have so far is that the out-put, so to say, by SD involves, obviously, experience of some sort; but that it is not qualified as perception, because it does not necessarily involve sensuous imagery. Now, if the experience is not some kind of sensuous imagery what is it then? What is this experience to which Plantinga draws our attention? The suggestion is that there is a particular sort of experience that always is present in the operation of SD. He draws our attention toward what it feels like entertaining a true proposition in opposition to a false proposition. (See p. 46) The feeling connected when entertaining these propositions is not accompanied by sensuous imagery, as in, say, when I am appeared to redly, or when something red is presented to me. The experience associated with ‘the feeling right’ or ‘feeling wrong’ Plantinga, we remember, calls doxastic experience. So, doxastic experience is always involved in the operation of SD, “because it is always connected with the formation and sustenance of any belief.” (WCB p. 183)


The upshot of these considerations is that the warrant giving role of SD, is not necessarily by way of sensuous imagery. Thereby it is not qualified as perception, strictly construed. On the other hand there is always doxastic experience associated with the operation of SD; but whether this is enough for qualifying the warrant enjoyed by doxastic experience, as perceptual experience, is not easily answered. It depends upon the favored perceptual theory.

4.2.3 Natural Knowledge of God and Sin

It is obvious that the SD is, according to Plantinga’s model, part of the original cognitive equipment, part of as he says “the fundamental epistemic establishment with which we have been created by God.” (WCB p. 180) But, it is part of central doctrines of Christianity that this does not meet the requirements for being in a right relationship with God. In this paper as already mentioned, this special concern of Christian belief will be left out. But I think it is important to point out that Plantinga’s concern is obviously to supply a model for especially Christian belief. In this regard, he develops an extended model wherein the central goal is to provide a specific Christian model for theistic beliefs. This involves, obviously, central doctrines such as, the sacrificial death of the second person in the Godhead, reconciliation, justification and the Holy Spirit’s special work in application of this work; as well as several other topics. In this way it is reasonable to say that the A/C model suffices only to provide a general route on which theistic beliefs can have warrant. The extended A/C model, on the other hand, is intended to provide for specific Christian beliefs. The contrast is that between the activity of the Holy Spirit (specific Christian) and SD (theistic). The former is, says Plantinga, “part of a special response to the fallen condition into which humankind has precipitated itself; the latter is part of our original epistemic endowment. The former is a special divine response to sin; presumably, there would be no such activity had there been no sin. The latter would no doubt have been part of our epistemic establishment even if humanity had not fallen into sin.” (WCB p. 180)


The last facet, therefore to note – in relation to the model – is that this natural knowledge of God is somehow weakened, “reduced, smothered, overlaid, or impeded by sin and its consequences.” (WCB p. 184) It is only by the entering of faith that SD will be restored to its proper function; an aspect accounted for by the previously mentioned extended A/C model. As “for now” Plantinga remarks, “we note only that the knowledge of God provided by the SD, prior to faith and regeneration, is both narrowed in scope and partially suppressed. Due to one cause or another, the faculty itself may be diseased and thus partly or wholly disabled.” (WCB p. 184)


If SD is not functioning properly, how then can it produce warranted beliefs? This question, I think, is not easily dealt with. Even for (some) atheists, J. J. Smart[39] for instance, there are emotional grounds for belief in a higher meaning – in God – but the problem is of reconciling the emotional side and the rational side. Smart gives more weight the rational side, saying that were it not for everything else he knows, in science etc. then it would be easier for him to believe.[40] It seems, at least in the case just described, that the ‘rational’ side decides for the withholding of belief; even though there is an emotional tendency to form beliefs to the effect that God somehow exists, the ‘God-belief’ it withheld.


In some cases withholding displays a sort of analogue of warrant. But, withholdings are not always the rational thing to do. Consider upon reading Russell that I see it is logically possible that the world popped into existence five minutes ago, complete with all those apparent memories, crumbling mountains, and dusty books; as a result, I withhold the belief that I am more than five minutes old, or perhaps I also just be have been persuaded by Wright’s I-II-III argument, but reject the hinge answer. I find myself fully convinced the world popped into existence five minutes ago. Now, would this belief enjoy the analogue of warrant? Surely not; such withholdings are not exemplary of epistemic caution, but of cognitive malfunction, Plantinga opines.


There are situations where withholdings are the rational thing to do; in other it is not. According to the model, failure to form theistic beliefs by way of SD, are irrational and lack the analogue of warrant. The issue between natural theology and revealed theology seems latent here; if SD were enough to ground, say, the whole panoply of e.g. Christian belief, then Plantinga would side with the theologia naturalis project, proclaiming that general considerations on human reason (e.g.) is enough to warrant belief in the Christian God.


What we have so far is a model and some specifications of it. We do not yet know whether or not theistic beliefs can in fact enjoy warrant sufficient for knowledge. That question we’ll look at next.


4.3 Is Belief in God Warrant-Basic?

The first thing to note is that if theism is false, then the out-put of SD – belief in God – does not enjoy warrant sufficient for knowledge. No false belief can have enough of warrant such that it meets the requirements for being knowledge; “therefore, if theistic belief is false, it doesn’t have that degree of warrant.” (WCB. p. 186) What degree of warrant can it then enjoy?


A false belief can enjoy some warrant where the faculty that produces the belief is working at the limit of its capability. An ornithologist sees a bird in the distance and mistakenly think that it is a Fraticula Artica; as a matter of fact, it is just too far away for him to see clearly that it is a Fraticula Artica instead of, say, Fraticula Bartica – a bird very similar to Fraticula Artica, not normally seen in that area where the ornithologist finds himself. Perhaps this is even clearer if we imagine a particle physicist who mistakenly believes that a certain subatomic model is close to the truth; but working on the outer limit of the cognitive domain for which our faculties are designed, his belief is false but not without warrant. The ornithologist will not, sensitive as he is, insist that it was a Fraticula Artica, just as the particle physicist will not insist that his model is correct – but nonetheless these beliefs can be warranted, even if false.


But if there is no such person as God, then there is no such mechanism as SD, and what “truth-aimed faculty would be such that it is working at the limit of its ability in producing the belief that there is such a person as God, if that latter belief is false? A decent candidate is hard to find. But, on the other hand, if the cognitive faculties are functioning as they should, Plantinga claims, even if they are working at the limit of their capacity theistic belief are often believed with a degree of firmness much higher than the degree of firmness held by the two cases above. Such considerations, suggest, Plantinga opines, “that if theistic belief is false, it is not produced by cognitive processes successfully aimed at the truth, and hence do not have warrant.” (WCB p. 187)


Another important consideration: Might it not be that belief in God is produced by cognitive processes successfully aimed at the truth, even if that belief is, as a matter of fact, false? Reflect on the case where a reliable barometer may give a false reading because of an unusual and improbable confluence of circumstances; in this case even though the barometer is reliable, it issues a false reading – due to the extremely improbable, but still real possibility that all the air molecules in the room should congregate in the region of space where the barometer is not located – because there hasn’t been a long enough time for it to react to the chance. Could something similar not be the case for a cognitive process that produces belief in God?


Plantinga answers this question negatively. He does this, by considering the epistemic conditional probability in a possible word framework, by way of the following principle:

“A proposition is objectively probable, with respect to some condition C, only if that proposition is true in most of the nearby possible worlds that display C.” (WCB p. 188)


Consider the process that produces theistic belief: if it is successfully aimed at truth, then in most nearby possible worlds it produces belief in God. If this is the case, it follows that in most nearby possible worlds belief in God is in fact true: in most nearby possible worlds there is such a person as God. But, that can’t be, if, in fact there is no such person as God. If in fact (in the actual world) there is no God, then a world where God exists would be enormously, unimaginably different from the actual world. If we grant that there exists no God, it is probably not the case that the process that produces theistic beliefs produces a true belief in most nearby possible worlds. Therefore, it is unlikely that belief in God is produced by a process that is functioning properly in a congenial epistemic environment according to a design plan successfully aimed at the production of true belief. By way of conclusion then, if theistic belief is false, it probably doesn’t have warrant.


If on the other hand, God exists, and humans are created in his image, in order to be able to have knowledge of him and; if this is so, it is natural to think that the “cognitive processes that do produce belief in God are aimed by their designer at producing that belief. But then the belief in question will be produced by cognitive faculties functioning properly according to a design plan successfully aimed at truth: it will therefore have warrant.” (WCB p. 189)


Sure, it is possible, that even if God exists and has created us humans with a certain faculty f such that it always malfunctions; and also created some other faculty f* to produce some other beliefs, often malfunctions in such a way that it produces belief in God. Given this scenario, belief in God wouldn’t have warrant; despite the fact that theism is true.[41] This is possible; but it is more likely that if theism is true, then the epistemic probability of theistic beliefs being warranted is very high. “But is it true?”[42] Here I’ll quote the last paragraph in Warranted Christian Belief:


“This is the really important question. And here we pass beyond the competence of philosophy, whose main competence, in this area, is to clear away objections, impedances, and obstacles to Christian belief. Speaking for myself and not of course in the name of philosophy, I can say only that it does, indeed, seem to be to be true, and to be the maximally important truth.” (WCB p. 499)


This he says, even though he “doesn’t claim as part of his philosophical position that belief in God […] do have warrant. (WCB p. 347)


Summing up: in the first part of this chapter we considered how religious beliefs, as Crispinsteinian hinge beliefs, comply with intuitive understanding of religious epistemology. In this part we looked at Plantinga’s model for religious belief; it was suggested that the model is possible, and, if true, then very likely are the theistic belief in question is warranted. According to the model, religious beliefs are not usually arrived at by way of arguments, that they aren’t strictly speaking warranted by way of perception, that if the object of religious belief – God – does not exist, then religious beliefs will not have enough warrant to be knowledge, and lastly, that according to the model, this ‘inert’ propensity to form religious beliefs is weakened due to the effects that sin has on our noetic equipment. In other words if my cognitive faculties are functioning properly, in a suitable environment, according to the design plan, and are successfully aimed at truth, then religious beliefs, formed in those circumstances, will most likely have enough warrant for knowledge.


How does this model of religious belief stand in relation the hinge suggestion we encountered in the first part of this chapter? Considering that question will close this chapter. In the fifth and final chapter, we’ll look at critical issues pertaining to theistic beliefs.


4.4 Conclusion

In this chapter I have presented the two suggestions for how theistic belief can be on par epistemologically according to the two epistemologies in chapter 2 and 3. In the first section of this chapter, I looked at how theistic belief could be viewed on the Crispinsteinian view: if the analogy between (MOORE) III and (THEIST) III could be established, then we would be, at least prima facie, entitled to view (THEIST) III as a certainty belief. But, as we saw, even though there are certain similarities, there are also major dis-anologies. The dis-analogy was, I take it, reason enough to disqualify that attempt of answering the skeptical argument as pertains to e.g. (THEIST).


This leaves us with the Plantingian epistemological response to the I-II-III argument; according to which the inferential structure presupposed in the I-II-III argument is rejected. In the second section of this chapter I presented Plantinga’s model for how theistic belief can be warranted. According to the model, there is an inert propensity to form theistic belief in certain circumstances. It is central to the model, that humans are God’s creation. It is on this basis that it becomes plausible that theistic belief can be warranted. The question whether it is warranted, is not argued for by Plantinga, that is because, it is most likely warranted if theism is true, and most likely not warranted if is false.


In the fifth and final chapter, I’ll take up a number of issues that are latent in the above. Most importantly, of course, whether Plantinga gets around the problem of circularity; whether he is in a position to claim that theistic belief is warranted

Chapter 5 – Closing Study, Objections and Replies and Conclusion


We have seen that theistic beliefs do not fit well with the hinge response; the analogy is not sufficiently strong to set theistic belief on par with type-III positions. The Plantingian suggestion has it that the proper functioning agent will in certain circumstances form theistic belief; these beliefs are not the result of arguments, but rather are triggered by certain dispositions in certain circumstances. The Plantingian suggestion, furthermore, has it that these beliefs do not rest upon other beliefs of the agent, and thereby sidestepping the question-begging difficulty at center stage in this thesis.


This chapter falls into 4 sections. In section 5.1 I’ll argue for the primacy of the externalist answer to the I-II-III argument. This I’ll do by pointing out that the externalist ideal of what makes the difference between mere true belief and knowledge seems to be more in line with usual knowledge ascription. Hereafter I’ll look at how Plantinga’s epistemology stands in relation to transmission of warrant and closure. In section 5.2, I’ll raise an objection to Plantinga’s properly basic theistic belief viz. the anything goes objection which will lead in 5.2.1 to the sufficiently analogue beliefs system goes objection. In section 5.3 where the issue of circularity again will be in focus, in particular the question is whether Plantinga is in a position to claim that theistic do have warrant. In section 5.3.1, I’ll look at how the properly basic theistic belief stance can be facilitated in relation to the argument form evil, and see how this relates to the question of argument for theism and claim of warranted theistic belief. In section 5.4, I’ll draw this thesis to a conclusion, by highlighting the most important results of this thesis.


5.1 Externalism, Transmission, Closure (and knowledge ascription)

In this section, I’ll discuss and argue for the primacy of the Plantingian general epistemological view. My argument here concerns two issues. First, I’ll point out that there is reason to consider the inferential architecture (underlying the I-II-III skeptical argument) not to be equivalent to that of the normal belief forming situation. Thereafter I’ll discuss whether the Plantingian answer is capable of blocking any skeptical argument; this pertains to the issue of whether or not the Plantingian accepts that there is such thing as a process ‘of warrant acquisition each of whose specific presuppositions warrant has already been earned.’ If not, the externalist epistemologist (of the Plantingian stripe, at least) is not so much in a better situation, than the epistemologist who accepts the I-II-III structure, as he is in a different situation.  He is not so much in a better position because the experience could be indistinguishable form a situation where his belief is indeed false. He would be in a different situation because he is still allowed to ascribe knowledge to some of his beliefs according to proper functionalism.


In chapter 3, I presented and discussed aspects of the Quinn/Plantinga debate. That discussion pertained specifically to perceptual beliefs. Quinn argued that type-II propositions would have more by way of positive epistemic status if they were justified by type-I propositions. Plantinga disagreed. Type-II propositions can have, indeed, do have, according to Plantinga, more by way of positive epistemic status if accepted in a basic way, than by way of type-I. This was because Plantinga agreed with Reid, type-I propositions do not constitute non-circular evidence for type-II propositions.


The conception of warrant tied to this view was set forth in 3.2 wherein the notion of proper function plays a major role. According to this stance, (WPETG): “a belief has warrant for a person S only if that belief is produced in S by cognitive faculties functioning properly (subject to no dysfunction) in a cognitive environment that is appropriate for S’s kind of cognitive faculties, according to a design plan that is successfully aimed at truth.” (WCB p. 156)


Equipped with this notion of warrant, the externalist is allowed to ascribe warrant to a belief, even though it is not justified by type-I positions. The description of what usually is at stake in belief forming situations, is not necessary for elucidating the externalist notion of warrant, in that the externalist on warrant simply rejects the internalist ideal that whatever has a bearing on justification must be accessible to the epistemic agent. (see p. ) But, it gives us reason to accept an externalist notion, because, it seems to be more in line with our experience. For example, it seems that we ascribe knowledge to beings that seem not to have the reflective capacities that internalism requires. Think of young children. We are often in a position to say of them, that they know so and so, even though these agents have no substantive reflective awareness of what factors support that knowledge. It also is clear that our intuition that an agent being in a state of knowledge essentially is not something that is determined by the subject himself – whether the subject thinks, feels, is inclined or believes to a significantly strong degree etc..


The proper functioning stance gives us reason to be optimistic in relation to answering the I-II-III argument. But, does this purge the externalist of any liability of doubt concerning skeptical possibilities? In particular, does the proper functionalist think that there is such thing as a process “of warrant acquisition each of whose specific presuppositions warrant has already been earned”? If the answer is negative, is the proper functionalist then in the same camp as far as liability to a skeptical voice? This would be the claim of the skeptic, precisely because the he will insist, that for any belief generated and allegedly warranted by the proper functionality of those processes, will only enjoy as much warrant as there is warrant for the presuppositions of that process. That is, none at all.


The externalist does not think that he has earned warrant for each of the epistemological presuppositions in a given epistemic situation. However, this does not worry the externalist, because he does not think that it is necessary for a belief to be warranted, that the epistemic agent must be warranted in the epistemic presuppositions involved. If the belief is formed in line with the proper functional view of warrant i.e., if the requirements in (WPETG) are satisfied then the belief is warranted, whether or not the agent is warranted in the epistemological presuppositions. As long as there are no indications to the effect that there is something amiss, the subject is entitled to continue trusting those faculties and the amenability of the environment.


Now, is Plantinga’s theory purified form all liability of doubt or skeptical possibilities? A lurking skeptic might want to ask: Isn’t it all possible that we should all be deceived here? Isn’t it possible that I should have the very experience I do have and there be no state of affairs corresponding to those experiences? Couldn’t I be a brain in a vat, the only survivor of an Alpha Centaurian intergalactic attack, kept alive by my captors for scientific purposes, and completely deceived about what sorts of state of affairs there are? Is it not possible that I am the victim of a Cartesian evil demon who delights in deceiving people? From the Plantingian perspective, or course, there is no need to declare these things impossible; in fact he thinks that they seem perfectly possible (in the broadly logical sense), even though they seem a bit farfetched.


From the Plantingian perspective, there simply does not much follow from the skeptical possibility. He simply says that “nothing much follows. […] All that follows is that I don’t have the sort of certainty that Descartes sought. I don’t have that sort of certainty (if indeed there could be any such thing) in which I can simply see (in a way that is somehow beyond the possibility of doubt or mistake) that things could no be otherwise than thus and so. In the present case [beliefs in other minds], of course, my beliefs can be false, despite my best effort. But that doesn’t mean I don’t have knowledge of them; for while knowledge requires psychological certainty – or at any rate high degree of belief – it is not the case that knowledge requires Cartesian certainty.” (WPF p. 76-77)


5.1.1Transmission – Closure

How does the Plantingian stance stand in relation to the principles presented in the first chapter viz, the transmission principle and closure? We remember that Wright says of Transmission of Warrant (TW):

A particular warrant for the premises of an entailment is transmitted to its conclusion only when one’s path to that warrant does not require picking up knowledge of the conclusion en route, or depend on some form of prior entailment to it. (WC p. 3, see chapter 1 p. 12)


Let us take a Plantingian example; he forms the belief that there is a tiger-lily – call it p. Now this belief, we suppose, is formed in good epistemic circumstances – the cognitive faculties are functioning, as they should, in a congenial environment suitable for the exercise of those faculties aiming at true beliefs. This belief is properly basic for Plantinga, in that he accepts p in the basic way, and furthermore p has warrant for Plantinga, accepted in that way. Furthermore, if Plantinga is warranted in the basic way, in his belief that p, then what ever warrant that belief enjoys (that is typically very much) will transmit across to the belief which self-evidentially follows form p, i.e. that the external world exists. If Plantinga wanted to make an inference form his warranted belief that p to the external world (from type-II to type-III propositions) that warrant would be transmitted to the latter.


This is clear, in that the belief in question, p, does not pick up knowledge of the conclusion (it is a properly basic belief), and does not depend upon some form of prior entailment to it; whatever warrant the former has, transmits across to the type-III proposition. That the Plantingian conception of warrant is wholly different than Crispinsteins is obvious; the former an externalist and the latter an internalist. That this application would not make a good argument, or would not have enough force to convince the skeptic, I think, Plantinga would wholeheartedly agree to; but, nevertheless there would be transmission of warrant. So, the out-come is that even though Plantinga would agree that a cogent argument is one whereby one could be moved to rational conviction of the truth of its conclusion, he would not think that transmission of warrant from type-II to type-III (just illustrated) would fail, i.e. he is not inclined to think that transmission of warrant is depended upon whether there is a cogent argument in the vicinity.


Given that there are no immediate problems with transmission of warrant in the Plantingian optic, we would not expect there to be any problems with the weaker principle - the principle of closure:

Knowledge always transmits across known entailment: if A knows p, and A knows that q follows from p, then A knows q.


For the sake of explicitness, let us though, take (THEIST) as an example. According to Plantinga, type-II of (THEIST) can be properly basic for him. In order to make things clearer, let’s proceed as if Plantinga’s belief is properly basic for him. Plantinga, then knows that God is appearing to him – i.e. Plantinga knows that p. Furthermore, Plantinga knows that if p is the case, then he is not the subject of Freudian illusions or wish-fulfillments – i.e. Plantinga knows that q follows from p. So, Plantinga knows that q simply because he knows that p.


Plantinga lets his externalist theory of warrant do what it suffices to do, and points out that we do have knowledge if the requirements of warrant are fulfilled, even though the agent is not in a position to adduce non-question begging grounds for the belief. This was precisely what other (early) externalists, notably Dretske and Nozick, failed to do, when they rejected closure.[43]


Vital in the optic of these theorists is the necessary requirement on knowledge that the agent in question must be sensitive to a wide range of error possibilities. More specifically, in the nearest possible worlds where the proposition in question is false, the agent would not believe it. So e.g. in the nearest possible world where it is false that one is not a brain in vat (BIV) one still continues to believe that one is not a BIV. In this way, one cannot know that one is not a BIV. In contrast, if the actual world is pretty much as we take it to be, then it will follow that our beliefs will be sensitive in a sufficiently range of nearby possible worlds. Thus, it is thought that knowledge of everyday propositions are retained, as long as the belief is sensitive, and the failure to know far-fetched skeptical scenarios are accounted for. But, it seem to be a peculiarity with this anti-skeptical move that it calls into question the very ideal of knowledge that lies at the heart of the externalist move: that it is not necessary for a belief to be warranted that there is reflective access to the ground of the belief. Rather, if the belief satisfies the factual requirements, then the belief is warranted.  


Instead of focusing on that they do have knowledge of everyday propositions, if their beliefs are sensitive, they focused on why closure failed. Plantinga, instead, keeps his focus on what his theory of proper functions says of warranted beliefs i.e. that there are certain beliefs that are properly basic and therefore qualify as knowledge. Two questions pertain here: isn’t Plantinga theory of basic beliefs allowing far too many beliefs in that camp, e.g., basic theistic beliefs; and is Plantinga in a position to claim that he knows the denial of skeptical scenarios? The first question will be considered in section 5.2 whereas we get at the latter in section 5.3.


5.2 The Anything Goes Objection

Generally stated the objection has it that if belief in God can be basic, then any belief can be basic. This was first raised by Plantinga in Reason and Belief in God. Here is how Plantinga raised the objection:


“It is tempting to raise the following sort of question. If belief in God is properly basic, why cannot just any belief be properly basic? Could we not say the same for any bizarre aberration we can think of? What about voodoo or astrology? What about the belief that the Great Pumpkin returns every Halloween? Could I properly take that as basic? Suppose I believe that if I flap my arms with sufficient vigor, I can take off and fly about the room; could I defend myself against the charge of irrationalism by claiming that this belief is basic? If we say that belief in God is properly basic, will we not be committed to holding that just anything, or nearly anything, can be properly basic, thus throwing wide the gates to irrationalism and superstition?” (RBG p. 74)


DeRose identifies two questions – one though and one easy – in the RBG passage, of which Plantinga only answered the easy. The though question is “whether, if Plantinga held one of the ‘bizarre’ beliefs, he could ‘defend [him]self against the charge of irrationality by claiming this belief is properly basic” and the easy question is “whether, in accepting belief in God as properly basic, Plantinga was thereby committing himself to the ‘bizarre’ beliefs being properly basic as well”. (Voodoo Epistemology p. 2).


The latter easy question is answered in this way. Why think that when some beliefs are properly basic, then any beliefs can be in that that camp no matter how bizarre? From the fact there are some beliefs that qualify for being of a class it does not follow that any belief is of that class, is obvious, I think. But this commits the defender of theistic beliefs to suppose that there is a relevant difference between belief in God and belief in the Great Pumpkin, which, as a matter of fact Plantinga does – he holds that the former not the latter is properly basic. For example, Plantinga thinks that there is such a mechanism as Sensus Divinitatis (see p. 57) with a natural tendency to produce belief in God. “[…]the same cannot be said for the Great Pumpkin, there being no Great Pumpkin, and no natural tendency to accept beliefs about the Great Pumpkin. (RBG p. 78)


This answer is as good as it goes. No doubt the Great Pumpkinintes will object to what seems to them implausible and parochial. The Great Pumpkinites will think that there being no God, and therefore no natural tendency to form such belief. They too, will be committed to the view that there being a relevant difference between belief in God and belief in the Great Pumpkin, which as we may suppose they do – there is such  mechanism as sensus Great Pumpkinites (SGP) with a natural tendency to produce belief in the Great Pumpkin they. The same, they will continue, cannot be said for belief in God, there being no God, and no natural tendency to accept belief in God.


As far as I can see, there is nothing that a priori excludes the Great Pumpkinities to produce an analogous story, or a model, to Plantinga’s SD-model. According to this model, the Great Pumpkin orchestrated the evolutionary process in such a way, that we have a natural tendency to form belief in the Great Pumpkin – by way of SGP. In the right circumstances then, belief in the Great Pumpkin will be warranted. So far, we have a model, according to which not any belief can be warrant in the basic way (the model does not allow any belief to have the privilege to make use of the Plantingian defense strategy, see p. 56) but sufficiently analogue belief systems will be able to tell a similar story, as for them being epistemically on par. So, the anything goes objection is reduced to this objection: some sufficiently analogue belief systems can be Plantinga warranted. (SABS objection)


5.2.1 The SABS objection

How then is this objection thought to influence the Plantingian defense strategy? Some sufficiently analogue belief systems – Hinduism and Buddhism, theistic versions thereof – can successfully be defended along the Plantingian line. On schematic form SABS:


(1)   There are some sufficiently analogue beliefs systems that are defendable according to Plantinga’s defense strategy.


(2)   Plantinga’s defense strategy could not be used to successfully to exclude these beliefs systems.


(3)   Therefore, Plantinga’s defense strategy does not provide a successful defense of theistic belief against the charge of irrationality.



I see no reason to doubt (1). The Hindu and Buddhist epistemologist can defend the epistemological status of their religious belief along the Plantingian way: the respective epistemologist of these two belief systems can use a similar model to defend their beliefs against the charge of irrationalism. As far as I have set the scene, I have no ground to reject (2): Plantinga’s defense strategy could not be used to successfully exclude these belief systems. If Plantinga can defend his religious belief against the charge of irrationalism, then the Hindu and Buddhist also can defend their belief against the charge of irrationalism.


But is this really a substantive critique of Plangtinga’s defense of the rationality of theistic belief? Well, this might be of more interest, if the strategy used, would allow any belief system to be defended against the charge of irrationality. However, this is not the case – not just any belief can be defended according to this model. It would not be the case e.g. for voodooism, Great Pumpkinism or philosophical naturalism.


Take philosophical naturalism – the view that there is no God or anyone like him. Naturalism resembles some of the aspects of theistic belief: it tells us where we come from, where we are going etc. In this way I think it is a sufficiently analogue belief system. Among the central tenets of naturalism is the view that biological beings, such as Homo sapiens, are the result of evolutionary processes: random genetic mutation and natural selection. Not only species, but also survival enhancing systems in the species, such as the digestive system and the cognitive faculties are the result of evolutionary processes. In the 12th chapter of Warrant and Proper Function Plantinga set forth the much discussed and controversial “evolutionary argument against naturalism”. He argued that this system of belief (naturalism) is not such that if it is true then very likely it has warrant. Perhaps thoughts alone similar lines lay behind Wright’s comments towards the end of Wittgensteinain Certainties, where he gives expression to the thought that an epistemological position’s answer to the skeptic – that we are entitled to trust our belief forming faculties unless there is information to the contrary – needs its own version of the serenity prayer. It will take me a field to discuss this argument; suffice to quote this short paragraph: “For what the argument shows is that if these beliefs [naturalism and current evolutionary theory] are true, then it is not likely that our belief-producing processes and mechanisms are, in fact, reliable, in which case the beliefs that they produce, including the belief that naturalism is true, do not have warrant.” (WCB p. 351)


Now, the defense of theistic belief at center stage here is not able to set aside Hindu and Buddhist beliefs. In particular, the defense strategy undertaken here suffices to defend theistic belief systems. If the argument (1) – (3) is to be of interest then we need to make plausible that some of the belief systems under discussion here are irrational. Some of them are false, I am inclined to think, but form that fact, it doesn’t follow that some are irrational.


These considerations show that the defense undertaken here suffices to show how generic theistic (Islam, Judaism, Christianity, some versions of Hinduism, and Buddhism) belief can be defended against the charge of irrationality along the Plantingian line. The epistemologists, theologian and philosophers of religion of these respective belief systems need further elucidations and considerations to defend their more specific beliefs. The Hindu will argue for, and from certain points of view, not accepted by the Christian. The Buddhist will similarly argue for and from points of view that the Judaist will not accept, etc.


5.3 Circularity

But, is a real killer argument lurking in the neighborhood? Recall the Plantingian defense strategy for theistic belief: (a) the A/C model is possible both logically and epistemically; (b) given the truth of theistic belief, there is no philosophical objection to this model’s also being not merely possible but true; and (c) if theistic belief is indeed true, then very probably it does have warrant. Isn’t there a circularity problem here?  After all, central theistic beliefs are included in the A/C model. Plantinga is only warranted in thinking the A/C model true if he (already) is warranted in accepting theistic belief. But this is going in circle – how can the argument then be cogent? A cogent argument, we remember, being one whereby one could be moved to rational conviction of the truth of the conclusion. This certainly precludes going in circle.


On schematic form, the circularity objection would be like this:


              I.      Plantinga is warranted in holding his theistic belief

only if

           II.      he is warranted in believing that his A/C model is true.


The Plantingian answer is that the warrant that theistic belief enjoys does not arise or gets generated by this argument – the argument for the A/C model. If this was how the model was construed, then the project would suffer from vicious circularity. Theistic beliefs are not arrived at by way of arguments, but by way of this inert propensity accounted for by the SD mechanism. This inclination is God given and is perhaps best thought of as analogue to logical intuition. Just as logical intuition endows the enquirer to ‘see’ certain logical steps as valid, so the theist upon contemplating certain theistic propositions simply ‘sees’ them as true – and if the conditions of warrant are met then the belief is warranted.


“So if the source of the warrant of my theistic belief were this argument, then indeed the project would suffer from vicious circularity. But it isn’t, and it doesn’t. The source of warrant for theistic belief, according to the model, is not an argument of any sort; in particular, its warrant does not arise from some argument about how theistic belief can have warrant” (WCB p. 352, italics mine.)


Furthermore, a few pages back I quoted Plantinga saying that this epistemologist “doesn’t claim as part of his philosophical position that belief in God […] do have warrant.” (WCB p. 347) This is because in all likelihood (See chapter 3 p. 65) these beliefs do only have warrant if they are true. But nonetheless the reformed epistemologist “does believe that they are true and is prepared to claim that they are, even if he doesn’t propose to argue for that claim.” (WCB p. 347) Now, the reformed epistemologist is not prepared to argue for the claim of the truth of theism as part of his model for how theistic belief can enjoy warrant. What is the thought here? Is it because he thinks that arguments are irrelevant? Or is it perhaps that theistic belief goes against reason? That they go against reason in similar vain as the Russian theologian Shestof, held that one can only attain religious truth by rejecting the proposition that 2+2=4 and accept instead 2+2=5?


No, the thought is rather that, even though there are no good arguments – non-question begging – for theistic belief being warranted, it does not imply that theistic belief cannot be warranted. Similarly for belief in the existence of other minds, the external world etc., epistemic agents can be warranted in these beliefs even though there are no non-question begging arguments for them. Even if the beliefs that ‘there are other minds’ and ‘there is an external world’ are not conclusions of cogent arguments can we be warranted in holding them – it would be a sign of dis-function not to believe them because there are no non-question begging argument for them.


Another vital aspect needs to be pointed out. Philosophers have often the tendency to thinks of theistic belief as an explanatory hypothesis. According to the Plantingian stance this is not at all how theists think about their cognitive attitude towards theism. Theists do not believe in God because theism is the best explanation of certain experiences. Theists find themselves convinced of the proposition in question, not because it is a conclusion of a deductive, inductive or abductive argument. Similarly, we do not believe in the existence of the external world, because it best explains the experiences we have or anything of that sort; nor do we believe in the existence of other minds because it best explains the different behaviors of other homo sapiens. The situation is rather that we simply find ourselves believing these things, not by way of arguments, but simply in a way that does not rest upon other of our beliefs.


If something along the above lines is right, are we not then poorly situated in furthering the debate? Doesn’t this stance simply block any further deliberations and considerations – isn’t this a stand off? The theist will claim that theistic beliefs are true and therefore that theistic beliefs most likely are warranted. The atheist, on the other hand, naturally enough, will claim that theistic belief are not true, and therefore most likely do not enjoy warrant – unquestionably not enough of warrant to be knowledge. What we really want to know is whether theistic belief is true or not. At this juncture, it is vital to ask what the Plantingian stance relates to the project of natural theology, in the sense of providing arguments for the truth of theism.


Let us get at this question and the question whether it is a stand of or not by looking at one of the classical arguments against the rationality of theistic belief – the argument from evil.

5.3.1 The argument from evil – an example

The argument form evil comes in two overall versions: the deductive and the inductive. The former has the non-existence of God as its conclusion, the latter is intended to give us reasons to dismiss or withhold theistic belief. It is the latter that we’ll look at. As relates to the theist, the evidential argument from evil – which the latter sometimes is called – is intended to give him defeaters for his beliefs. Suppose that the objector – the atheologian – sets forth this consideration:


(r) God exists and is omniscient, omnipotent, and wholly good

Is extremely improbable given

(t) There are 1013 turps of evil (where one turp is a basic unit of evil)


Could this consideration not defeat the properly basic belief theistic belief? Well, in fact Plantinga thinks that this is perhaps a defeater. But, according to Plantinga this is not a good defeater for theistic belief. The theist, he says, need not even know or have a very good reason to think that it is false that (r) is improbable given (t). All the theist needs to do is to refute this argument – there is “no obligation to go further and produce an argument for the denial of its conclusion.” (FT p. 309) How does the theist refute this argument? The idea is very simple – too simple, perhaps. The theist according to Plantinga, has much by way of warrant concerning the basic theistic belief. This warranted belief is an intrinsic defeater-defeater. An intrinsic defeater-defeater is a basic belief that has more by way of warrant than some of its potential defeaters. It would be appropriate with an example. Here is Plantinga’s example.


A letter that could embarrass me disappears from my department chair’s office under mysterious circumstances. I had motive, means, and opportunity to steal it, and a reliable member of the department testifies to having seen me furtively entering the office around the time the letter must have disappeared. I have been known to steal things in the past. This circumstantial evidence persuades my colleagues (who are fair-minded people) that I am guilty. Yet the fact of the matter is that I spent the whole afternoon in question on a solitary walk in the woods – I clearly remember having done so. It is one of my basic beliefs that I was alone in the woods all that afternoon, and I did not steal the letter. The evidence I share with my colleagues (that I had motive, means and opportunity) gives me a substantial reason to believe a defeater for my belief that I spent the afternoon in question alone in the wood. However, the belief that I spent the afternoon in the woods has more by way of warrant for me – due to memory – than the potential defeater. The belief that I spent the afternoon in the woods is an intrinsic defeater-defeater.


Theistic belief can be defeated by (r) being very improbable given (t). This defeater could defeat theistic belief, if the defeater belief has more by way of warrant than the theistic belief – even if it is basic. If the basic theistic belief is warranted to a lower degree than the potential defeater, then the defeater would not only be potential but actual. Now, in the above example, the person in question has much by way of warrant for his basic belief – that he was in the wood the afternoon the letter disappeared. The crucial question then becomes: do basic theistic beliefs have this much by way of warrant as does the memory belief, or more specifically, does it have more by way of warrant, than does the warrant attached to the defeater? This depends upon the noetic state of the person in question. As far as the A/C model goes, the proper functioning epistemic agent does have a great degree of warrant for his basic theistic beliefs. Others whose cognitive organs do not functioning properly, according to the A/C model, will, perhaps not have that much by way of warrant that is required for the basic theistic belief being a defeater for the belief that (r) is improbable given (t); for this defeater defeats their theistic belief.


The defeater can of course be defeated in other ways than by an intrinsic defeater-defeater. This could for example be if the agent considered some of the classical arguments for the existence of God and found them convincing. He could thereby have a defeater for the belief (the original defeater) that (r) is improbable given (t). This would be the case if the defeater-defeater has more by way of warrant than does the original defeater. There is then, a place for theistic argument in the Plantingan scheme. However, as far as the project outlined in chapter 4 and discussed in this chapter is concerned, he does not argue for the truth of theism, nor for the conclusion that theistic beliefs do have warrant. As we have seen, this would saddle his project with vicious circularity.


Theistic arguments then do have a place in the Plantingian optic. This is so, in spite of the fact that the defense of theistic belief in this thesis has not facilitated its defense by arguing for the truth of theism. The considerations here have been to the effect that theistic belief do not need arguments in order to be warranted – they can be properly basic. But, as the considerations in relation to the argument from evil show, if the warrant theistic belief has is defeated by this argument, then the agent in question can acquire a defeater for the original defeater by theistic arguments.


Is Plantinga then in a position to claim to know that the Freudian skeptical scenario – that theistic beliefs are the result of wish-fulfillment – fails to obtain? In fact, it is not part of the official Plantinga’s theory to claim that any belief is warranted. That is because he is not in a position to know whether the conditions for warrant are satisfied in a given situation. But, given that the conditions of warrant are satisfied in theistic belief, then it is clear that he is warranted in his belief that the skeptical scenario does not obtain. Plantinga, then does not argue for the claim (it is not part of his theory to say that certain beliefs do have warrant) that theistic belief do have warrant, and therefore does not argue for the claim that he knows the denials of the skeptical hypothesis. Nevertheless, he is prepared to claim that if theistic belief is warrant (that he does indeed think) then he is in a position to claim that the denials of the skeptical scenario fails to obtain.


Isn’t this then a stand off? This epistemological stance implies that the question whether theistic belief do indeed enjoy warrant – enough for knowledge that is – is not only an epistemological question, but ultimately a question that depends upon what there is. In the end, it is a question whether theism is true – whether there is such a being as God. The theist thinks that there is such a being, the atheist disagrees; further elucidations must concern this question – the metaphysical question.


5.4 Conclusion

The contention that theistic belief can be basic, set forth and defended in this thesis, has been revolving around two epistemological responses to the skeptical argument presented in the first chapter. I initiated the discussion by presenting two preliminary objections to theistic belief. The first objection stated that it is wrong to engage in theistic belief formation because it lacks justification. The second objection had it that theistic beliefs are illusions or wish fulfillments.


Hereafter I discussed a complex of concepts were warrant and justification were at center stage. Internalism and externalism, it was pointed out, is the overall distinguishing mark in contemporary epistemology. According to the former, only what the subject can be aware of can have a bearing on justification. The latter rejected any such demand – whether a belief has warrant or not, is not something that is introspectively accessible. Of special interest for my thesis, was the distinction between access to the justifier and access to the epistemic efficiency of the justifier. The objection that it is wrong to hold theistic belief because it lacks justification, presupposes implausible assumptions about voluntary control over belief formation. Furthermore, it requires that we have access to the epistemic efficiency of the justifier.


The objection to the effect that theistic belief does not aim at truth simply dismisses the epistemic value in religious belief. It is simply a non starter – an epistemically interesting objection to theistic belief needs to be conjoined with the truth of the belief in question. So, these objections do not cause much concern for the defender of theistic belief. 


On this background the I-II-III argument was presented as being of central epistemological value in evaluating the epistemology of theistic belief. The I-II-III argument did not only apply to theistic belief, but functioned as a general epistemological problem, according to which type-I propositions must be the evidential base for type-II propositions. Furthermore it was argued that type-I proposition cannot justify type-II proposition unless type-III propositions are presupposed. The general epistemological problem provided the setting for the following two chapters; both of which were presented as uniform attitudes towards the I-II-III argument.


In chapter 2, I presented and discussed aspects of, what I called the Crispinsteinian view. According to this stance (which accepted all of the five skeptical points, see p.14) there are certain epistemological presuppositions – hinges – that need to be in place before type-I propositions can justify type-II propositions. The reasonable man does not doubt these hinges. Even though the epistemological presuppositions allow no defense in terms of knowledge (because they are not the result of a cognitive achievement) are they not to be criticized in terms of the failure of knowledge either. In this sense, they are a peculiar set of certainty beliefs – or certainty propositions.


The underlying workings of the skeptical argument i.e. that unless there is an ability to adduce non-question begging grounds for a belief, then that proposition can not be evaluated in terms of the idea of knowledge or warranted belief was accepted. In the case with type-III propositions, it is not possible to adduce grounds that escape the question-begging problem, given the inferential structure – the I-II-III structure. As a response to this devastating skeptical problem where (MOORE) type-III is doubted and therefore all empirical propositions as well, the Crispinsteinian argued to the effect that certain propositions, there among (MOORE) type-III function as epistemological presuppositions. Furthermore, it is vital for the Crispinsteinian response, that not only A-C hinges are presupposed, but equally that the 1-2 presuppositions are in place, which entitled the epistemic agent to continue to trust the functionality of the cognitive organs, and the suitability of the environment.


In chapter 3 the Plantingian answer to the I-II-III argument was at center stage. The I-II-III skeptical argument (which presupposed an internalist view on warrant acquisition) according to which experiential propositions are the only source for justification for type-II propositions. Just as Wright thought of his response as a uniform attitude, it was pointed out that the Plantingian response could also be seen as a uniform answer to the skeptical argument. In the first section, though, I discussed an issue that could just as well be part of a non-uniform answer to the skeptical argument i.e. the rejection of the first point of the skeptical argument. (Type-II propositions can only be justified on the evidence of type-I propositions.) According to Plantinga there are not only certain circumstances when type-II propositions can be justified otherwise then by the evidential relation to type-I; in fact it seem that that is precisely how type-II propositions do get warrant – not by justificatory relations to experiential propositions. 


On this background, I presented Plantinga’s theory of what makes the difference between mere true belief and knowledge – the theory of warrant. According to this theory, a belief has warrant for a person S only if the belief is produced in S by cognitive faculties functioning properly (subject to no dysfunction) in a cognitive environment that is appropriate for S’s kind of cognitive faculties, according to a design plan that is successfully aimed at truth. The uniform attitude to the I-II-III argument presented by Plantinga, suggests that in a wide variety of circumstances type-II propositions can be warranted without resting, or being inferred from type-I. More specifically, the Plantingian externalist construal of warrant rejected the view that an inability of adduce non-question-begging grounds entails an inability to know or warrantedly believe the proposition in question. So, in contrast to the view in chapter 2, type-III propositions can be warranted and evaluated in terms of knowledge. Right, it is not usual that epistemic agents form beliefs of the type-III variety – the external world exists, e.g., – but they can nonetheless be warranted because their respective type-II propositions are typically warranted in a basic way, from which type-III propositions self evidentially follow.


With these two general epistemological positions in place, I turned to theistic beliefs in chapter 4. I argued that if we could set (THEIST) type-III on par with other type-III propositions, then we would have a prima facie case that ‘God exists’ to function as a epistemological presupposition for religious belief formation. However, the analogy was not sufficiently strong. In particular, there were 4 major dis-analogies between type-III propositions and theistic type-III propositions: 1. Regular type-III hinges are beyond supportive evidence, whereas theistic type-III are not. 2. The consequences of neglecting type-III hinges, it seems, far exceeds neglecting theistic type-III. 3. Type-III propositions clearly involve empirical matters whereas theistic type-III do not. 4. The reasonable man does not doubt (MOORE) type-III, whereas quite a number of, what I take to be reasonable men, doubt (THEIST) type-III. Then I turned to the Plantingian suggestion for how theistic belief can be warranted.


According to the Plantingian model there is a sense of divinity, a faculty called sensus divinitatis, which in certain circumstances triggers theistic belief, e.g. when contemplating the sun set over the Rocky Mountains. (See note 37 p. 57) It is central to the model that humans are the creation of God, and that the proper functioning agent in certain circumstances will form theistic beliefs. The crux of my presentation of the Plantingian suggestion was the suggestion that according to the A/C model theistic belief can be basic with respect to both the internalist and externalist ideal of what makes the difference between mere true belief and knowledge. Most importantly however was the suggestion that theistic belief can be properly basic. These beliefs, if otherwise complying with the conditions of warrant, will therefore be warranted even if they do not rest upon other beliefs of the agent.


The most important question whether nor not theistic belief do enjoy warrant is not argued by Plantinga. This is because, as he says, theistic beliefs do most likely enjoy warrant if theism is true, and most likely do not enjoy warrant if theism is false. And, as we have seen in this chapter, Plantinga is not in a position to argue for the truth of theism as part of his argument to the effect that theistic belief can be properly basic – and he does not so argue. So his conditional conclusion: if theism is true, then most likely it enjoys warrant. If theism is false, then it most likely does not enjoy warrant.


In this fifth and final chapter, I also looked at how the Plantingian stance relates to warrant transmission and closure. Given that Plantinga is warranted – in the basic way – in type-II propositions, then the warrant would transfer to the respective type-III proposition. So, after all we can be warranted in the denial of skeptical scenarios – e.g. the Freudian skeptical scenario. It is not argued in this thesis, though, that theists are so warranted – but only can, and most likely are warranted if theism is true.


In this chapter I also defended the Plantingian stance from two objections – the anything goes, and the sufficiently analogue belief system objection. These two objections do not suffice to undermine the Plantingian stance. In particular, the former objection does not have enough force to show that if theistic belief can be basic then any belief can be basic. The second objection – does not have enough force to make any concern for the proper functionalist. In particular, this is because, the defense taken up in this thesis is not intended to defend full-blown theistic belief, such as they are set forth in Islam, Judaism, Christianity, Buddhism and Hinduism, but rather simply as a general defense of the possibility of holding generic theistic belief in a basic way. This I think is accomplished, on general considerations of the belief forming mechanisms, the belief forming situation – a description thereof – and objections raised and answered.


The crux is of course whether God exists or not. This is not for epistemology to decide. It is rather an metaphysical question; a question that requires careful considerations that need to take several other areas of human intellectual activity into consideration, such as what we know in the so-called hard sciences (physics, chemistry and perhaps biology) and soft sciences (socio biology, sociology of religion, theology, philosophy of religion etc.). This is no easy task, but vital and essential for the consideration of the question whether theistic beliefs not only can be properly basic, but actually are warranted in a basic way – whether the proper functioning agent not only can but is warranted in theistic type-II propositions.





Alston, William. P. “Epistemic Circularity” Philosophical and Phenomenological Research, 47, 1 1986., reprinted in his Epistemic Justification pp. 319-349. Cornell University Press, 1989.


Alston, William. P. The Reliability of Sense Perception. Cornell University Press, 1993.


Beilby, James. e.d. Naturalism Defeated? Cornell University Press. 2002.


Clifford, W.K. The ethics of belief and other essays / Ed. by Leslie Stephen and Frederick Pollock. London 1947


Dawkins, Richard. The Blind Watchmaker. Penguin Books. 1991. First published by Longman. 1986.


Dennet, Daniel. Darwin’s Dangerous Idea, Touchstone, NY. 1996.


Depaul, Micheal R. (ed.) Resurrecting Old-Fashioned Foundationalism, Rowman & Littlefield Publishers Inc. New York. 2001.


DeRose, Keith. “Voodoo-Epistemology” (unpublished talk manuscript) Available at: articles/derose_keith/voodoo_epistemology.pdf


Dretske, Fred. “Epistemic Operators”, Journal of Philosophy 67: 1007-23. 1970.


Dretske, Fred. “Conclusive Reasons” Australian Journal of Philosophy 49: 1-22. 1971.


Flew, Antony. The presumption of atheism: and other philosophical essays on God, freedom and immortality. London 1976.


Fumerton, Richard. “Classical Foundationalism” in Resurrecting Old-Fashioned Foundationalism e.d. Depaul, Micheal R., Rowman & Littlefield Publishers Inc. New York. 2001.


Fumerton, Richard. “Internalism/Externalism Controversy” Philosophical Perspectives vol. 2, 443-459, especially p. 449-451, 1988


Freud, Sigmund. The Future of an Illusion, translated and edited by James Atrachey. New York: W. W. Norton. 1961.


Edmund Gettier, “Is Justified True Belief Knowledge?” Analysis, Vol. 23, 1963.


Harris. J. James. Analytic Philosophy of ReligionHandbook in Comtemporary Philosophy of Religion. Kluwer Academic Publishers. 2002.


Howard-Snyder, Daniel. ed. The Evidential Argument from Evil. Indiana University Press. 1996.


Hume, David. A Treatise of Human Nature. (First published 1739 and 1740) Penguin London. 1985.



Klein, Peter. “Skepticism” in Stanford Encyclopedia of Philosophy 2001.


Locke, John. An Essay Concerning Human Understanding. (First published (fifth edition) 1706). Penguin London. 1997.


Luper, Steven. “Epistemic Closure Princple” in Stanford Encyclopedia of Philosophy 2001.


Mackie, John. The Miracle of Theism Oxford: Clarendon University Press. 1982.


Moore, G.E. Proof of the External World” Proceedings of the British Academy XV: 273-300. 1939.


Michael Williams. Skepticism” in Epistemology – the blackewell guide to, ed. Greco, John and Sosa, Ernest. Blackwell Publishers 1999.


Nozick, N. Philosophical Explanations, Oxford: Oxford University Press. 1981.


Phillips, D.Z. “’What God Himself Cannot Tell Us’: Realism Versus Metaphysical Realism” Faith and Philosophy 18: 4. pp. 483-500. 2001.


Plantinga, Alvin. Warrant and Proper Function. Oxford University Press. 1993.


Plantinga, Alvin. Warranted Christian Belief Oxford University Press. 2000.


Plantinga, Alvin. “Reason and Belief in God” in Faith and Rationality – Reason and Belief in God, (pp. 16 – 93).ed. Alvin Plantinga and Nicholas Wolterstorff, Notre Dame University Press. 1983.


Plantinga, Alvin. The Foundations of Theism: A Reply,” Faith and Philosophy 3 no. 3 298-313. 1986.


Pritchard, Duncan. “Is ‘God exists’ a ‘hinge proposition’ of religious belief?” International Journal for Philosophy of Religion: 47: 129-140. 2000.


Pritchard, Duncan. “Resurrecting the Moorean Response to the Sceptic” Internaltional Journal of Philosophical Studies Vol. 10 (3), 283-307. 2002.


Pryor, James. “The Skeptic and the Dogmatist” NOÛS 34:4, 517-549. 2000.


Quinn, Phillip. “On Finding the Foundations of Theism,” Faith and Philosophy 2, no. 4  469-486. 1985.


Quinn, Phillip. “The Foundations of Theism Again” in Rational Faith – Catholic Responses to Reformed Epistemology, ed. Linda Zagzebski, University of Notre Dame Press. 1993.


Ramachandran, V. S. Phantoms in the Brain. London: Forth Estate. 1998.


Reid, Thomas. Inquiry Into the Human Mind. Chicago University Press. 1970.


Smart, J. J. and Haldane. J. J. Atheism and Theism. Oxford: Blackwell Publishers. 1996.


Strong, A. H. Systematic Theology. Judson Press. 1907.


Wittgenstein, Ludwig. On Certainty e.d. by G.E.M. Anscombe and G.H. von Wright., New York: Harper Torchbooks 1972. first printed 1969.


Wolterstorff, N. and Plantinga, A. ed. Faith and Rationality. Notre Dame: University of Notre dame Press. 1983.


Wright, Crispin. “Wittgensteinian Certainties” yet unpublished, but forthcoming in Wittgenstein and Scepticism e.d. Dennis McManus, Routledge. Availiable at


Wright, Crispin. “Cogency and Question-Begging: Some Reflections on McKinsey’s Paradox and Putnam’s Proof” Philosophical Issues 10. pp. 140 – 163. 2000.


Zagzebski, Linda. Rational Faith – Catholic Responses to Reformed Epistemology, ed., University of Notre Dame Press. 1993.

[1] Clifford, W.K. The ethics of belief and other essays / Ed. by Leslie Stephen and Frederick Pollock. London 1947

[2] The Future of an Illusion, translated and edited by James Atrachey. New York: W. W. Norton. 1961. p. 30.

[3] Edmund Gettier, “Is Justified True Belief Knowledge?” Analysis, Vol. 23, 1963.

[4] Within internalism, there are also mixtures of foundationalism and coherentism: foundherentism.

[5] See his “Concepts of Epistemic Justification” The Monist, 68, no. 1. 1985. Reprinted in Epistemic JustificationEssays in the Theory of Knowledge. Cornell University Press. 1989.

[6] But it depends upon how we individuate the method. If we individuate the method as ‘looking at a stopped watch’ then the method is unreliable.

[7] Warranted Christian Belief, Oxford University Press. 2000. Plantinga also talks about other kinds of rationality, which I ignore here.

[8] Wright, Crispin., “Wittgensteinian Certainties” yet unpublished, but forthcoming in Wittgenstein and Scepticism e.d. Dennis McManus, Routledge. Availiable at Referred to as (WC) in the text.

[9] Originally printed in Proceeding of the British Academy, Vol. XV, 1939. Reprinted in Moore, G. E., Philosophical Papers. George Allen and Unwin Ltd. 1959 p. 126-148.

[10] At least not for (MOORE); it is perhaps more controversial for (THEIST)

[11] Wright. C., “Cogency and Question-Begging: Some Reflections on McKinsey’s Paradox and Putnam’s Proof” Philosophical Issues 10. pp. 140 – 163. 2000.

[12] Wright says that it even may be at odds with Wittgenstein’s later philosophy of language.

[13] Ludvig, Wittgenstein On Certainty Harper Torchbooks, London: 1969. Numbering (OC) refer to paragraphs in that text.

[14] Even though the words in italics eco a familiar Wittgensteinian notion, are they neither Wright’s nor Wittgenstein’s, as far as I know - I think that I caught them when listening to a lecture by R. Brandom in Århus, a few years ago.

[15] “But I do not think that Wittgenstein needs to be read as saying anything antithetical to the idea that one’s confidence e.g. that one has two hands is ultimately empirically based. The thrust is rather that if your certainty that you have two hands would dominate a sensory impression that represented that as missing, then you are implicitly prioritising one kind of evidence – something like: your lifelong experience of yourself as handed, together with the absence from your experience of any worrying tendency of material objects abruptly and inexplicable to go missing – over another. And that priority is not itself justified by experience.” (WC p. 20-21)

[16] Questions concerning 3 will be bracketed in this thesis.

[17] See WC p. 21

[18] See e.g. Sosa’s  and Van Cleve’s papers in Naturalism Defeated? e.d. James Beilby. Cornell University Press: 2002 for more on the reawakened debate, about the origin of our reliable cognitive faculties, which Descartes thinks pivots on the existence of God. The debate was reawakened, as far as I know, by Plantinga’s so-called evolutionary argument against naturalism. In the last two chapters of Warrant and Proper Function Oxford University Press, 1993.

[19]The original version goes:


God, grant me

the Serenity to accept the things that I cannot change

the Courage to change the things I can

and the Wisdom to know the difference (R. Niebuhr, 1932)

[20] This discussion, then, is inspired by this debate between Phillip Quinn’s and Alvin Plantinga’s interaction in a number of articles concerning basic religious beliefs. The original article in the debate is: Plantinga, “Reason and Belief in God” in Faith and Rationality – Reason and Belief in God, ed. Alvin Plantinga and Nicholas Wolterstorff, Notre Dame University Press. 1983. (pp. 16 – 93). Referred to as (RBG). This book has become a classic in the literature in religious epistemology. For the subsequent articles: Phillip Quinn “On Finding the Foundations of Theism,” Faith and Philosophy 2, no. 4. 469-486, referred to as (OFT) and Plantinga’s reply “The Foundations of Theism: A Reply,” Faith and Philosophy 3 no. 3 298-313, referred to as (FT) in the text, and the response to Plantinga’s reply by Quinn “The Foundations of Theism Again” in Rational Faith – Catholic Responses to Reformed Epistemology, ed. Linda Zagzebski, University of Notre Dame Press. 1993. See also: pp. 95-98 in Warrant and Proper Function.

[21] I have inserted the terminology [type-I] and [type-II] of this thesis in the citation, and in the subsequent material.

[22] His actual examples: (8) I see a hand in front of me (type-II); (9) It seems to me that I see a hand in front of me (type-I).

[23] Here is what Reid says in a discussion of Hume’s skepticism:

“The skeptic asks me, Why do you believe the existence of the external object which you perceive? This belief, sir, is none of manufacture; it came from the mint of Nature; it bears her image and superscription; and, if it is not right the fault is not mine: I even took it upon trust, and without suspicion. Reason, says the skeptic, is the only judge of truth, and you ought to throw off every opinion and every belief that is not grounded on reason. Why, sir, should I believe the faculty of reason more than that of perception? – they came both out of the same shop, and were made by the same artist; and if he puts one piece of false ware into my hands, what should hinder him from putting another?” Reid, Thomas. Inquiry Into the Human Mind. Chicago University Press. 1970. p. 207

[24] In fact Plantinga’s theory only pertains to the first two parts: 1 and 2. Questions concerning 3 will be bracketed in this thesis.

[25] Here we are reminded of the comments that Wright made towards the end of (WC) that any epistemology that relies upon proper function needs its own version of the serenity prayer. The prayer is needed because any epistemological theory that takes it as a “entitlement” that our cognitive powers give us reliable information, because it seems to be a kind of faith – if we do not have any reason so distrust our cognitive powers, then we continue to trust them.  (see p. 31 of chapter 1)

[26] See pages 199-211 in WPF for naturalistic attempts of analysis of proper function. Plantinga thinks that naturalistic epistemology flourishes best in supernatural metaphysics.

[27] This is McDowell’s term for what is available to the experience both in veridical and deceptive cases.

[28] He could perhaps be an agnostic.

[29] E.g. D. Hume, J.S. Mill, A. Flew, J.L Mackie who think that there is indeed evidence for premises from which the conclusion has the negation of (THEIST) III.

[30] E.g. Smarts deliberations in. Atheism and Theism., ed. Smart, J. J. and Haldane, J. J. Atheism and Theism. Oxford: Blackwell Publishers. 1996. For other examples: Danniel Dennet’s Darwins Dangerous Idea, Touchstone, NY. 1996.

[31] If he was thoroughly convinced, even though he responded to the input as he did, then he would be noting but a ‘half skeptic’.

[32] See: Luke 24:38-41 and John 20:24-27.

[33] As already mentioned, the present paper looks only at the Calvin/Aquinas model: sufficing it to be the ground for Judaism, Islam, and Christian beliefs, general theism; leaving out of consideration the extended Calvin/Aquinas model: specific Christian belief. Islamic and Judaic theologians will not accept this specific presentation, but the essentials of the model will be accepted by those religions as well.

[34] Summa Theologiae I, q. 2, a.1, ad.1., quoted from WCB, p. 170.

Augustus H. Strong calls the faculty responsible for “the knowledge of God’s existence […] a rational intuition. Logically, it precedes and conditions all observation and reasoning. Chronologically, only reflection upon the phenomena of nature and of mind occasions its rise in consciousness.” Systematic Theology, p. 52.

V.S. Ramachandran, a neuroscientist, says after P. Churchland has informed him about a Canadian scientist who stimulated his temporal lobe and experienced God that he “always suspected that the temporal lobes, especially the left lobe, are somehow involved in religious experience.”  The question remains is there a “God module” in our heads he asks? See. Chapter 9 in his Phantoms in the Brain. The quotation is from p. 175.

[35] Romans, 1:18­­–20

[36] Calvin, John. Institutes of the Christian Religion I, iii, 1, p.44, quoted from (WCB p.171)

[37] Included in these circumstances are the glories of nature: “the marvelous, impressive beauty of the night sky; the timeless crash and roar of the surf that resonates deep within us; the majestic grandeur of the mountains; the ancient, brooding presence of the Australian outback; the thunder of a great waterfall. But it isn’t only grandeur and majesty that counts; he (Calvin) would say the same for the subtle play of sunlight on a field in spring, or the dainty, articulate beauty of a tiny flower, or aspen leaves shimmering and dancing in the breeze.” WCB p. 174.

[38] In this fashion Patricia Churchland says that “Truth, whatever that is, definitely takes the hindmost” in discussing the purpose or function of the cognitive organs. Journal of Philosophy, 84 (October 1987), p. 548, quoted from WPF, p. 218.

[39]  See J. J. Smart and J. J. Haldane Atheism and Theism p. 168.

[40] Smart says: “I concede that theism is an emotionally attractive doctrine. Perhaps it even is true.” Ibid. p. 77.

[41] This would be something like (a sort of) complex and peculiar theological Gettier problem, Plantinga observes.

[42] The question is asked specifically to Christian belief.

[43] Fred, Dretske. “Epistemic Operators”, Journal of Philosophy 67: 1007-23. 1970. and “Conclusive Reasons” Australian Journal of Philosophy 49: 1-22. 1971. Nozick, N. Philosophical Explanations, Oxford: Oxford University Press. 1981.