Michael Czapkay Sudduth
Reformed theologians since John
Calvin have emphasised the immediacy of at least some of our natural knowledge
of God, as well as the necessity of spiritual regeneration and the epistemic
efficacy of the Holy Spirit for much of the rest of our knowledge of God.[1]
In Warranted Christian Belief (hereafter WCB), Alvin Plantinga
provides a rigorous philosophical development and defence of these
long-standing Reformed ideas over against various epistemic objections to
theistic and Christian belief. In particular one must be impressed with
Plantinga’s careful analysis of important epistemological details, something
typically absent from earlier Reformed thinkers.
These merits notwithstanding, advocates of natural theology and more
traditionally minded religious philosophers may be disappointed with WCB
on two counts. First, unlike other works in philosophy of religion and
religious epistemology, no attempt is made to provide arguments for God’s
existence, though considerable and laudable effort is put toward answering
various objections to Christian belief. Of course, Plantinga has elsewhere
argued for the importance of theistic arguments. In fact, he has outlined and
developed a variety of such arguments himself, some of which he thinks are good
arguments.[2]
In WCB Plantinga is presenting a model of how theistic and Christian
belief can be warranted, and his main argument is that there are no plausible
objections to this model that do not presuppose the falsity of theism. So we
must conclude, I think, that the presentation of theistic arguments simply
falls outside the scope of the project Plantinga is carrying out in WCB.
Secondly, though, there might be a related concern. The model Plantinga
presents involves theistic and Christian belief having warrant in a basic way.
Consequently little is said about the nature, extent, or significance of inferential
knowledge of God, or, more precisely, the epistemic function of theistic
arguments. Although Plantinga concedes in a few places in WCB that
theistic arguments can confer warrant on theistic belief (WCB, pp. 179,
258), from the rather sparse and undeveloped comments at this juncture it
appears that it is not essential, or at least not central, to the
epistemological model he presents. But it might be worth considering the
avenues open to Plantinga here, and which might retrospectively call for
modifications in the model he presents.
In this paper I hope to provide some suggestions as to how Plantinga
could have developed his model of properly basic theistic belief in a way that
would assign a more explicit and prominent role to inferential reasoning
vis-à-vis our putative natural knowledge of God. These suggestions will arise
by way of a critique of Plantinga’s use of the Reformed tradition in the
development of his epistemological model. In both WCB and in his earlier
publications, Plantinga has drawn considerably on the Reformed tradition in his
thinking about the knowledge of God, specifically utilising theologians such as
John Calvin, Abraham Kuyper, and Herman Bavinck. However, Plantinga has revised
the Reformed view on our natural knowledge of God in some crucial ways, and as
a result of these revisions an important function for natural theology in the
epistemology of theistic belief has been marginalised, if not eliminated
altogether.
In Chapter 6 of WCB Plantinga
proposes the A/C (Aquinas/Calvin) model of warranted theistic belief. According
to this model, humans are created with a disposition to form various theistic
beliefs in wide variety of experiential circumstances, from the sight of the
starry night sky to the sound of a Mozart Symphony. In much the same way that
sensory perceptual experiences provide the circumstances that trigger our
disposition to form various sensory perceptual beliefs, there are widely
realised experiential circumstances that spontaneously trigger an innate
disposition to form various theistic beliefs. Such beliefs are formed in a
basic way, not by logical inference from other beliefs taken as evidence for
theistic propositions. In this way, theistic belief shares an important
similarity to sensory perceptual, memorial, testimonial, and a priori beliefs.
Plantinga claims that this account is a philosophical development of insights
found in the Reformed theological tradition, originating with John Calvin’s
discussion of the sensus divinitatis (sense of divinity) in the early
chapters of Book I of the Institutes of the Christian Religion. Although
Plantinga’s model of properly basic theistic belief has several philosophical
merits, it deviates in some important ways from a common Reformed view of
natural knowledge of God.
When discussing the natural knowledge of God, it has been commonplace in
the Reformed tradition to distinguish between the implanted knowledge of God (cognitio
Dei insita) and the acquired knowledge of God (cognitio Dei acquisita).[3]
The distinction arose among Reformed theologians shortly after Calvin in their
attempt to elucidate and develop Calvin’s account of the natural knowledge of
God. In Chapter 3 of book I of the Institutes, Calvin articulates his
doctrine of the sensus divinitatis, an immediate or non-inferential
knowledge of God that arises spontaneously in the human mind.[4]
But in Chapter 5 of book I of the Institutes Calvin affirms a knowledge
of God grounded in the manifestation or evidence of God’s perfections in
creation. “Not only has God sowed in men’s minds that seed of religion of which
we have spoken but revealed himself and daily discloses himself in the whole
workmanship of the universe” (Institutes 1.5.1).[5]
Although there has been some dispute in Reformed circles as to how exactly we
should understand the distinction, if any, Calvin is drawing here, Reformed
theologians have traditionally understood Calvin to be at least implicitly
distinguishing between two distinct modes of natural knowledge of God, the sensus
divinitatis and a knowledge of God associated with or derived from the
general revelation of God in the created order.[6]
This distinction became deeply entrenched in the period of Protestant scholasticism
under the categories of implanted and acquired natural knowledge of God, the
former being non-inferential and the latter inferential. Equally important,
though, there has also been a tendency to regard these two modes of knowledge
as differing in their content or scope. For instance, according to Charles
Hodge, the implanted or immediate knowledge of God is simply “the general sense
of a Being on whom we are dependent, and to whom we are responsible”, whereas
analysis and inference aims “to show what that Being is”.[7]
Again, Calvin may be regarded as the genesis of the basic insight here. The
sense of divinity appears to be a somewhat vague religious sense or awareness,
a sense that there is some God and that he ought to be worshipped. By contrast,
the knowledge of God acquired from reflection on the created order provides, at
least in principle, a variety of details about the nature of God.[8]
I will consider this in greater detail below.
At first glance, Plantinga’s account revises this traditional view in
several ways. According to Plantinga (WCB, p. 174), the sensus
divinitatis produces a broad range of beliefs about God: there is a God,
God is great, God is creator, God should be thanked and praised, God
disapproves of what I have done, God forgives me for what I have done, God
hears my prayer, God can deliver me out of this trial, God is speaking to me,
etc. Many of Plantinga’s basic theistic beliefs involve the recognition of
various divine attributes (power, knowledge, goodness, for example) and a range
of beliefs with personal indexicals (such as God is speaking to me now).
But in Calvin’s treatment of the sensus divnitatis little if anything is
said about these sorts of highly specified theistic beliefs. Secondly,
Plantinga takes Calvin’s Institutes 1.5 “evidences” to be circumstances
in which the sensus divinitatis is triggered and various theistic
beliefs are formed immediately, without any inferential reasoning, in much the
same way that our sensory perceptual or memory beliefs are formed (WCB,
pp. 175–176). Furthermore, rather than treating the sense of divinity as a
knowledge of God itself (as Calvin does), for Plantinga it is simply a faculty
or mechanism utilised to produce theistic belief and knowledge when stimulated
by the Institutes 1.5 evidences. The innate or implanted knowledge of
God is replaced by an innate “capacity” to acquire knowledge of God in a basic
way, in circumstances that correspond to an experience of the Institutes
1.5 “evidences”. The grounds for acquired knowledge of God become the mere
occasion for triggering the innate capacity. Plantinga thus loses the Reformed
distinction between the implanted and acquired knowledge of God. Two distinct
sources of theistic knowledge become two different elements involved in a single
process of belief formation and knowledge.
In losing the Reformed distinction
between implanted and acquired natural knowledge of God, Plantinga’s revision
seems to lose a common Reformed basis for natural, inferential knowledge
of God. Limits on the scope of properly basic theistic beliefs enlarge
the cognitive space for inferential knowledge of God. The latter function as a
supplement to the former by filling out and extending some minimal truth about
God grasped immediately by the human mind. Conversely, by maximising or
enlarging the scope of what is given by way of the sensus divinitatis,
as Plantinga does, one reduces the importance of the inferential character of
our natural knowledge of God. Such a move does not undermine the project of
natural theology per se. Plantinga can certainly maintain, and in fact
does, that there are good theistic arguments and they are important to
Christian apologetics and philosophical theology. The criticism here is that an
important epistemic function of theistic arguments or evidences has been
weakened, the necessity of such arguments or evidences for the warranted status
of a range of theistic beliefs that form an essential part of the natural
knowledge of God. Of course Plantinga could still argue, and in fact does, that
natural theology can contribute and even be necessary to the warrant of
theistic beliefs in various circumstances for some people, especially given the
acquisition of defeaters.[9]
So the argument is not even that, on Plantinga’s model, theistic arguments have
no epistemic function. The present argument suggests a stronger
epistemic function, a more radical proposal. On the model I propose here, the
sense of divinity does not provide the range of natural knowledge of God
Plantinga suggests. So the warrant of these beliefs will likely depend on
inference.
Let’s consider how this would work
in greater detail.
Clearly, a crucial question is the scope of the propositional
content associated with the two modes of knowledge under discussion. Reformed
theologians do not speak with one voice concerning the exact content of the
implanted knowledge of God, and space does not permit any detailed analysis of
this. As already stated, for Calvin the sensus divinitatis is simply the
awareness that there is some God and that he ought to be worshipped. Since
Calvin takes idolatry to be evidence of the existence of a sense of divinity,
this awareness of God is logically consistent with a great deal of
fundamentally false beliefs about God. Hence, the sense of divinity must be
sufficiently general. But in both The Institutes and in his commentaries
on the Psalms, Acts, and Romans, Calvin argues that various divine attributes
can in principle be known by inference from features of the created order.
Among such attributes are God’s goodness, wisdom, justice, power, mercy, and
eternity.[10] These more
detailed descriptions of God are never associated with the sense of divinity
itself, but left rather to the development of the primordial awareness of a
supreme being through the process of reasoning and reflection.
What Calvin says is more explicitly articulated in later Reformed
theologians. As already noted, according to Charles Hodge, the implanted
knowledge of God is simply “the general sense of a being on whom we are
dependent, and to whom we are responsible”.[11]
The knowledge of God’s personality, moral attributes, and intelligence arise
only upon observation of the created world and reflection. The function of the
theistic arguments, as Hodge sees it, is not simply to prove that God exists,
but to inform us about the nature of God.[12]
Samuel Harris maintained a similar distinction: “By rational intuition man
knows that absolute Being exists; his knowledge of what it is, is progressive
with his progressive knowledge of man and nature”.[13]
Augustus Strong also argued, “a consideration of these [theistic] arguments may
also serve to explicate the contents of an intuition which has remained obscure
and only half conscious for lack of reflection”.[14]
According to William Shedd, theistic arguments “assist the development of the
idea of God, and contain a scientific analysis of man’s natural consciousness
of the deity”.[15] By limiting
the scope of what can be immediately known about God, these Reformed
theologians left important cognitive space for an inferential, natural
knowledge of God.
We may compare the current proposal to our epistemic situation in
relation to beliefs we have about other minds. Plantinga and other Reformed
thinkers have frequently argued that belief in other minds is relevantly
similar to belief in God.[16]
It is basic and properly so. However, in the case of belief in the existence of
other minds presumably what is immediate is my belief that there are other
minds or perhaps my belief that ‘here’ or ‘there’ is a mind (connected with the
body I see before me). In neither case is anything specified about the characteristics
of other minds. One might argue, therefore, that it is only by inference that I
acquire knowledge about the characteristics of other minds. So perhaps even
knowing that one’s friend or wife exists requires inference of
some sort, for identifying minds with these properties seems to require
inference. I know that Mr. Gush is my friend on the basis of my knowledge of
what is necessary and sufficient for friendship and that my relationship with
Mr. Gush satisfies these conditions. The knowledge is inferential. Perhaps I
directly know that there are other minds, but on what basis do I know that
‘this bloke’ is my friend, ‘that lady’ my wife, and ‘this man’ a colleague at
Saint Michael’s College? Similarly, inference would seem to play an important
role in the knowledge of propositions involving more specific descriptions of
the characters, physical properties, and mental states of people. True, in all
these cases I must make the assumption that there are other minds and that they
are somehow connected to the bodies that appear to and interact with me. But
unless I did this and made certain inferences, what could I know about the
other minds whose existence I grasp in some immediate way? Precious little I
should think.
If we restrict our attention solely to the similarity between belief in
the existence of other minds and belief in the existence of God,
we may lose sight of the other side of the analogy. Just as our knowledge about
the nature of other minds depends largely on inference, so also our knowledge
of the nature of God depends largely on inference. An epistemology of religious
belief must be concerned not only with how we know that there is a God
but also with how we know various other propositions about the nature of God,
for instance that he is omniscient, omnipotent, omnibenevolent, eternal,
immutable, necessary, and so on. Moreover, when such crucial epistemological
details are introduced it is clear that for some people even the belief there
is a God may depend considerably on inference.[17]
The proposition expressed by the sentence there is a God depends on what
is meant by ‘God’, and the epistemic devil is in the conceptual details.
There are two important qualifications to my argument here though.
First, many beliefs about other minds and God can be accepted in a psychologically
basic way. One could accept a variety of beliefs about other minds in a
psychologically basic way via testimony, and presumably this would also be true
with respect to many theistic beliefs. For example, I might accept on the
testimony of a pastor or theologian that God is timelessly eternal, in much the
same way that I accept on the testimony of a close friend that Mr. Gush is a
virtuous person. But my basic belief in each case would be warranted via
testimony only if these beliefs were warranted for the testifier. But why are
they warranted for the testifier? This will set off a regress of warrant
transfer for the relevant belief that will eventually require some sort of
inferentially warranted belief in the target proposition. So there must be
inferential warrant somewhere down the line in the ancestry of the belief.
Hence, even a warranted basic belief in this situation will have propositional
evidence available for it, and it would not otherwise be warranted. Hence,
theistic beliefs of this sort, as well as various beliefs about other minds, would
be similar to what Stephen Wykstra calls “evidence-essential” beliefs, beliefs
for which there must be evidence available, even if a given person does not
accept it on the basis of such evidence.[18]
Secondly, if Plantinga’s extended
model of warranted Christian belief is roughly correct, then many
Christian beliefs about God, including those that can also be known by our
natural cognitive powers, will have warrant in a basic way for the Christian.
The Scriptures declare God’s goodness, mercy, knowledge, providential control
of the world, etc. On Plantinga’s extended model, belief in these truths is by
way of the internal instigation of the Holy Spirit. Although inference plays an
important role in the development of the Christian’s knowledge of God, it looks
like it will not be needed to know many of the theistic propositions that qua
natural knowledge of God depend on inference. But it is only the latter that
interests me in the present paper. What the present account implies is an
important difference between the unregenerate person’s knowledge of God and the
Christian’s knowledge of God even when it is the same theistic proposition
under consideration. The Christian can know these in a basic way on the basis
of Scripture and the testimony of the Holy Spirit. The unregenerate is in a
significantly different epistemic position. Furthermore, given the noetic
effects of sin, the present account may also imply that the unregenerate person
often does not in fact know what can in principle be known about God by reason
and which is known by faith by the Christian.
So I take it that the present account of the necessity of inferential
knowledge of God must be qualified in the ways outlined above. Thus, even if we
suppose that there is some immediate, natural knowledge of God’s existence, the
Reformed tradition seems to regard such knowledge as quite minimal. The
development and extension of this knowledge is left to inference, subject to
the above qualifications. In this context, the philosophical project of natural
theology will be the attempt to codify and formally articulate what can, in
principle, be known about God by inference from reason and observation.[19]
I think that the above proposal also addresses several, related concerns
that Plantinga expresses about inferential, natural knowledge of God, from
which he derives additional support for the advantages of his own model.
Plantinga’s main concern with taking Calvin’s Institutes 1.5 evidences
as suggestive of inferential knowledge of God is that as arguments they look
quite weak and inadequate to ground any warranted theistic belief. Plantinga writes: “It isn’t that
one beholds the night sky, notes that it is grand, and concludes that there
must be such a person as God: an argument like that would be ridiculously weak”
(WCB, p. 175). Plantinga’s basic argument seems to be that Calvin’s
account of our natural knowledge of God is more philosophically plausible if we
adopt the sort of model Plantinga is proposing. Put more technically,
Plantinga’s argument is that the epistemic conditional probability of the
existence of God on the evidence Calvin cites and relevant background knowledge
would be low. Hence, it looks like theistic belief will not get much by way of
warrant this way.
Even if we suppose that Plantinga is correct here, there is an important
assumption being made in the analysis. When Plantinga introduces the
inferential interpretation of Calvin, he takes the target proposition,
supported by the relevant evidence, to be ‘God exists’. Reasonably enough, the
background knowledge will exclude theistic propositions. But on the
interpretation of Calvin I have argued, the target proposition is not ‘God
exists,’ but some other proposition about the nature of God, which will in turn
serve to extend the content of an original, immediate knowledge of God. The
conditional epistemic probability that ‘God exists’ given evidence E and
background knowledge K (which excludes theistic beliefs) will be very different
from the conditional epistemic probability that ‘God is provident’ or ‘God is a
personal, intelligent being’ given E and K, where K includes knowledge that
there is some God. Again we can draw on the analogy from other minds. The
conditional probability of ‘this here is another mind’ given my
observations of the behaviour of local physical bodies (and relevant background
knowledge) may be very low. But suppose that what is at issue is the
conditional epistemic probability that Mr. Gush is my friend or that my
wife is a virtuous woman. The conditional probability of these propositions
may be very high given the same evidence in conjunction with background
knowledge now revised to include my knowledge that that there are other minds
and that they are connected to certain local bodies and behaviour.
Given the discussion on the sensus divinitatis antecedent to the
elaboration of the manifestation of God in creation, it is reasonably clear
that Calvin does not intend the Institutes 1.5 evidences to be the
grounds for de novo belief in or knowledge of God’s existence. Theistic
arguments always presuppose some prior sense of divinity. So theistic arguments
or inferences aim to extend and develop a rudimentary, antecedent knowledge of
God, in much the same way that inference develops and extends our minimal
knowledge of ‘this’ or ‘that’ other mind. Put otherwise, argument or inference
is not the sole ground for our natural knowledge of God as a whole, and
theistic arguments carry the logical force they do for their respective
conclusions only because there is an antecedent sense of divinity in each
person.
There is, of course, a potential objection to my critical comments on
Plantinga lurking in the pages of WCB. In Chapter 6 of WCB, Plantinga
does leave open the possibility of construing our natural knowledge of God as
inferential, and he presents an alternate inferential model as a reasonable
interpretation of Calvin. Plantinga argues that the role of the sensus
divinitatis might be to enable a person to see the truth of a crucial
premise in a simple sort of theistic argument. For example, perhaps the sensus
divinitatis enables one to see the truth of the premise that ‘the heavens
can be gloriously beautiful only if God has created them’ (WCB, p. 176). In
this case, the belief that God created the heavens would be inferred from this
explanatory conditional in conjunction with the belief that the heavens are
gloriously beautiful. Hence, it would seem that Plantinga does not eliminate
the traditional Reformed notion of acquired, inferential knowledge of God after
all.
There are two important observations in this suggestion by Plantinga.
First, we need not think of inferential knowledge as requiring the actual giving
of an argument or even constructing one explicitly in one’s mind. As I
have argued elsewhere, much of our inferential knowledge probably proceeds in a
temporally immediate fashion without a conscious process of inferring one
proposition from another.[20]
Secondly, Calvin’s notion of inferential knowledge of God is largely a matter
of God’s existence explaining features of the world. Hence, the
inferences are largely explanatory in nature.
However, Plantinga’s alternate model still differs significantly from
the proposal in this paper. First, the function of the sensus divinitatis
is different. Rather than constituting a basic awareness of God, it is still
being conceived of as a faculty that produces beliefs of a certain sort, and in
this case producing beliefs of a very different sort than I think Calvin and
the Reformed tradition have associated with the sensus divinitatis.
Secondly, Plantinga still manages to lose the distinction between the implanted
and acquired knowledge of God, though this time it is the immediate knowledge
of God that is edged out. In Plantinga’s alternate model, the sensus
divinitatis is not nor by itself yields any knowledge of God. It simply
yields knowledge of certain conditional propositions that operate in
explanatory inferences that lead to theistic knowledge. Lastly, and I think
this is crucial, the alternate model at best obscures the central claim of this
paper, namely that the acquired knowledge of God is an extension or development
of a minimal immediate knowledge of God. The function of inferential knowledge
of God is primarily to reach a fuller notion of the being whose existence is
vaguely intuited or grasped immediately. Of course, due to sin few will
actually make much progress in such knowledge. This is the other side of the
epistemic coin. But Calvin holds such people accountable for lacking such
knowledge since they nonetheless possess a sense of divinity and ought to know
God in a deeper way given the evidences they have all around them.
So we seem left with three possible models for our natural knowledge of
God. In Plantinga’s main model, the scope of immediate natural knowledge of God
undermines the importance of inferential natural knowledge of God. In his
alternate model, inferential natural knowledge of God becomes significant, but
this knowledge is not a development of a pre-existing immediate natural
knowledge of God. In fact, the immediate knowledge of God seems to be
eliminated on this account. Moreover, since the sensus divinitatis
produces, presumably in a basic way, the premise that links observational
phenomena with God as the explanans, it isn’t clear how the philosophical
project of natural theology can be developed given this model. Lastly, there is
the model I have presented in this paper. In contrast to the Plantingian models,
this model represents a traditional Reformed view of the natural knowledge of
God. By placing limits on the scope of immediate knowledge of God, inferential
knowledge of God takes on a specific role of extending and developing a very
rudimentary immediate knowledge of God. The philosophical project of natural
theology will then be concerned with formally articulating the logic of this
inferential knowledge of God.
Unlike many of his Reformed
predecessors, Plantinga’s account of our natural knowledge of God does without
the long-standing Reformed distinction between the implanted and acquired
knowledge of God, as well as the related difference in content between them.
Although Plantinga’s revision has strengthened the apologetic defence of the
positive epistemic status of theistic belief against various evidentialist
objections, it has weakened an interesting, important, and strong epistemic
function of theistic arguments or evidences that has been available in the
Reformed tradition. Some may judge this an acceptable trade off. But an
interesting future challenge to Reformed philosophers would be to maintain the
sound apologetic against the undesirable elements of evidentialism, while also
developing a stronger role for inference with respect to our knowledge of God,
natural and otherwise. In this paper, I have drawn on the Reformed tradition to
present one model of how this can be done, specifically with respect to the
natural knowledge of God. I suspect that Plantinga will welcome some of these
suggestions, whether as a critical engagement of his current way of thinking
about theistic knowledge or as suggestive of ways of modifying and extending
the A/C model presented in WCB. In short, the general recommendation
would be to set limits on proper basicality (with respect to warrant) to make
room for inferential knowledge of God. I haven’t proposed any detailed account
of just where we should draw this line. I’ll leave this for future
reflection and dialogue. Lastly, I have no doubt that the project of natural
theology may exist harmoniously with the claims of Plantinga’s Reformed
epistemology as they are developed in WCB. The present recommendation
represents one way of strengthening the connection between them. Perhaps it
will generate further dialogue at a crossroad somewhere between Jerusalem and
Athens.[21]
[1]. William Shedd, Dogmatic Theology, Vol. 1
(reprint, Thomas Nelson Publishers, 1980), pp. 198–199; Augustus Strong, Systematic
Theology (1907; reprint, Judson Press, 1979), pp. 52–62; Herman Bavinck, Doctrine
of God, tr. and ed. by William Hendriksen (1951; reprint, Eerdmans, 1979),
pp. 57–62; Auguste Lecerf, Introduction to Reformed Dogmatics
(Lutterworth Press, 1949), pp. 37, 40, 111, 120; Abraham Kuyper, Principles
of Sacred Theology, tr. by J. Hedrik De Vries (1898; reprint, Baker Book
House, 1980), pp. 263–275, 372–389, 553–563.
[2]. See Alvin Plantinga, ‘The Prospects for Natural Theology’’ in Philosophical Perspectives 5: Philosophy of Religion, ed. James Tomberlin (Ridgeview, 1991), pp. 311-312; ‘Christian Philosophy at the End of the Twentieth Century’ in The Analytic Theist: An Alvin Plantinga Reader, ed. James Sennett (Eerdmans, 1998), pp. 339-340; ‘Two Dozen (or So) Theistic Arguments’ (1986 unpublished lecture notes), http://www.homestead.com/philofreligion/files/Theisticarguments.html; ‘Belief in God’ in Introduction to Philosophy, ed. R. Boylan (Harcourt Brace Jovanovich, 1992), pp. 390-396.
[3]. See Heinrich Hepp, Reformed Dogmatics: Set Out and
Illustrated from the Sources, ed. Ernest Bizer, tr. G.T. Thomson (Baker
Book House, 1978), pp. 1–3; Louis Berkhof, Systematic Theology, 4th
edition (1941; reprint, Eerdmans, 1984), pp. 35–36; Herman Bavinck, The
Doctrine of God, pp. 52–64. On the historical development and influence of
this distinction, see Richard Muller, Post-Reformation Dogmatics, Vol.
1: Prolegomena to Theology (Baker Books, 1987), especially Ch. 5; and John
Platt, Reformed Thought and Scholasticism: The Arguments for the Existence
of God in Dutch Theology, 1575–1650 (E.J. Brill, 1982), especially Chs.
3–6, and 8.
[4]. Reformed theologians in the nineteenth and twentieth
centuries frequently refer to this immediate knowledge of God as the intuitive
knowledge of God’s existence. See Shedd, Dogmatic Theology, Vol. 1, pp.
198–199; Charles Hodge, Systematic Theology, Vol. 1 (reprint, Eerdmans,
1982), pp. 191–199; Augustus Strong, Systematic Theology, pp. 53–55.
[5]. John Calvin, Institutes of the Christian Religion,
tr. Ford Lewis Battles, in the Library of Christian Classics, Vols.
XX–XXI (The Westminster Press, 1960).
[6]. Edward Dowey explains Calvin’s view as follows: “Man
empirically observes the order of nature and from this draws conclusions about
its Maker and Governor. Rather than an immediate sensus divinitatis, we
have immediate experience of the world as the raw material or sense data on the
basis of which the mind of man says ‘therefore’ with respect to God” (Dowey, The
Knowledge of God in Calvin’s Theology, 3rd. Edition (reprint; 1952,
Eerdmans, 1994), p. 74). For a defence of this traditional interpretation of
Calvin, see my ‘The Prospects for Mediate Natural Theology in John Calvin’ Religious
Studies vol. 31 (1995), pp. 53–68.
[7]. Hodge, Systematic Theology, Vol. 1, pp.
195, 202.
[8]. See Dowey, The Knowledge of God in Calvin’s
Theology, pp. 50–52, 72–81.
[9]. For a detailed development of this, see my ‘The
Internalist Character and Evidentialist Implications of Plantingian Defeaters’,
International Journal for Philosophy of Religion, vol. 45 (1999), pp.
167–187.
[10]. Calvin, Institutes of the Christian Religion, 1.5.1–3; Commentary
on the Book of Psalms, in Calvin’s Commentaries, vol. 4 of 22, tr.
and ed. James Anderson (Baker Book House, 1979), Psalm 19, pp. 308–311; Commentary
Upon the Acts of the Apostles in Calvin’s Commentaries, tr.
Christopher Fetherstone, ed. Henry Beveridge, vol. 19 of 22, Acts 14:17; 17:22,
pp. 19–20, 154, 158; Commentary on the Epistle to the Romans in Calvin’s
Commentaries, tr. and ed. John Owen, Romans 1:19–21, pp. 69–72.
[11]. Hodge, Systematic Theology, Vol. 1, p. 195.
[12]. Ibid., pp. 194–195, 199–200, 202.
[13]. Samuel Harris, The Philosophical Basis of Theism (Charles
Scribner’s Sons, 1883), p. 288.
[14]. Strong, Systematic Theology, p. 72.
[15]. Shedd, Dogmatic Theology, Vol. 1, p. 221.
[16]. Prior to Plantinga’s analogy between God and other minds, John Baillie
made considerable use of the analogy in his critique of inferential knowledge
of God in Our Knowledge of God (1939; reprint, Oxford University Press,
1963).
[17]. Stewart Goetz has developed a similar sort of objection, though with
more radical implications, namely that no theistic belief can be
properly basic since all such beliefs require attributing properties to God,
and no such property attribution can proceed without inference. See Stewart C.
Goetz, ‘Belief in God is not Properly Basic’, Religious Studies, vol. 19
(1983), pp. 475–484.
[18]. See Stephen Wykstra, ‘Toward a Sensible Evidentialism: On the Notion of
Needing Evidence’, in W.L. Rowe and W.J. Wainwright (ed.), Philosophy of
Religion: Selected Readings, 2nd ed. (Harcourt Brace Jovanovich, 1989).
[19]. Of course, according to Calvin and the Reformed tradition the noetic
effects of sin compromise this natural knowledge of God in various ways. Hence,
the only successful formal project of natural theology may be a distinctly
Christian one.
[20]. See my application of Robert Audi’s distinction between structurally and
episodically inferential belief in ‘The Prospects for ‘Mediate’ Natural
Theology in John Calvin’, Religious Studies, vol. 31 (1995), pp. 57–63.
[21] I would like to thank Alvin Plantinga for his generous and challenging comments on the penultimate draft of this paper.