Hume, Paley, and Theology
Paley rebuts much of the philosophical argument against design coming from Hume in his Natural Theology, though he never directly indicates that he is reacting to Hume. There is, however, a critique that Paley does not seem to address. This is the theological critique that Hume lays out. Hume claims that after the design argument, the theologian is left with something theologically untenable. Hume writes, “I cannot, for my part, think, that so wild and unsettled a system of theology is, in any respect, preferable to none at all.”[1] The quality of Paley’s response to the issues found in Hume is such that there is no necessity in restating Paley in any more than an overview. This essay will therefore describe the essential elements of the design argument, give some indication of Paley’s philosophical and evidential response, and then focus on addressing Hume’s theological critique.
The classical design argument has the following basic structure: as the works of human beings exhibit a certain complexity and order that requires an intelligence to produce, so also the natural order itself in its complexity and order also requires an intelligence to produce. In other words, as human beings are to our products, so also God is to God’s creation. The experiential data for making such a claim relies on the complexity of the natural world. The way that nature fits together with adapted means and ends seems to be similar to complex machines of human manufacture. Further, as Hume presents the argument, the very principle of inferential analogy seems to be supported by the regularity of nature. The known circulation of blood in human beings is inferred analogically to other animals and to plants. The analogy weakens as we encounter plants and animals that are more dissimilar to those we know, yet it still seems to be a valid principle, even when the analogy is admittedly rather weak. We can therefore use the principle to draw conclusions about God based on what we know about human beings. This is, as Hume states in the voice of Cleanthes, “the Author of nature is somewhat similar to the mind of man.”[2]
The sort of inferential analogy presumed by Hume and Paley in their argumentation hinges on a logic in which a presumed similarity between God and the minds of human beings. God is an immaterial spirit and the minds of human beings are also immaterial, creative, and intelligent. This allows for an almost mathematic proportionality surrounding this middle term. God:nature::man:manufactured objects. This is a proportional analogy that admits that there is a proportional difference between God and human beings, such that God is higher than the human yet some similarity and valid inferential analogy yet remains.
Hume immediately follows this up in his main outline of the design argument by informing the reader that he will be taking issue with it. He argues that there is an acute dissimilarity between natural objects and man-made ones. We have a common experience that a house needs a builder but we do not have such an experience regarding nature. We do not have a direct experience of the sort of cause for which nature is the effect. We have seen houses built by men and thus experienced both cause and effect. We have not seen natural orders created by God and thus not experienced both cause and effect in this case. Hume argues that therefore any similarity here is at best a guess or presumption. This will become part of his theological critique later on in the essay.
Paley, on his part, does an excellent job refuting these basic criticisms of an inferential proportional analogy that allows us to predicate a designer. His famous example is that of find a watch along a wooded path. It is not necessary for us to have seen it manufactured or have experience of watch manufacturing in general in order for us validly to infer that it was manufactured. Therefore, by proportional analogy, the fact that we were not around when God created the natural order does not mean that we cannot infer that God is creator. Further, Paley points out that any defects in the watch or a lack of understanding regarding the purpose and function of some of the parts do not in fact take away from the validity of the inference. Any defect in the natural order, whether real or perceived, or a lack of understanding of the natural order does not mitigate against inferring a creator. Further observations by Paley include that no one would think that a watch would come together through a processes of ordering intrinsic to matter, or by chance, or was an illusion.
There is a deeper principle of analogy at work in Aquinas that may strengthen Paley’s rebuttal and further weaken Hume’s criticism. This is the idea that there is an analogy of being, or analogia entis, from which one can make claims about God. It is a metaphysical principle regarding how causes and effects relate in general and what we can know from them. Such analogy does not rely on a proportion that encompasses human beings and thus also avoids the weakness of inferential analogy. That is to say, it does not liken God with humans and the natural world with manufactured goods in order to make its point. It also avoids certain anthropomorphic interpretation that Hume claims causes further problems for theism.
The most basic knowledge causing syllogism, as outlined in Aristotle’s Posterior Analytics (PA), is the syllogistic demonstration in which cause and effect together result in a sure conclusion. There is a middle term by which the syllogism can operate and appearing in at least two of the terms. This is the sort of knowledge that Hume and Paley have in mind in using human intelligence and action as a middle term in their discussion.
This differs from the sort of analogy advocated and used in theistic proofs by Thomas Aquinas. Thomas notes, using Aristotle’s PA, that it is also possible to know a cause through an effect. This is what is often called a demonstration quia or fact-demonstration. It produces knowledge of the cause from the effect due to a certain convertibility in the terms, even though this fact knowledge is not quite as strong.[3] There is a certain amount of convertibility in the terms. It allows for a why, such as a first cause, to be derived from a fact. Thomas presumes that we do not know the nature or essence of God, which would be the middle term in a syllogistic proof for the existence of God. He therefore uses this type of reasoning to a cause through its effects, using the effect as a middle term in the syllogism (ST, I, 2, 2. See also, SCG, I, 12, [8].) This is particularly appropriate whenever the effect is more known than the cause (CPA, p 106). The effect is analogous to the cause rather than holding a univocal direct relation to the cause. It can tell us something about it without disclosing the cause in all of its fullness.
An example of this general principle at work, an object put into contact with fire heats up. The heat in the object is the effect from which one can posit heat in the fire, which is the cause. Another example would be that we do not have direct access to the interiority of human beings, but we have access to what human beings do as an effect of human character. I cannot know an individual human character directly but I can know a particular human character from the deeds of the individual human being.
The degree of knowledge in this manner is not as great because, as in the examples, only one aspect of the cause comes under consideration. Using this type of argument one can find the heat element of fire but not necessarily the light element. Alternatively, we may know that a particular person is a liar or is altruistic and be able to posit such qualities in her character, but there are usually more qualities in any individual.
The importance of the difference in analogical argument between the design argument portrayed by Hume and Paley and that of Thomas Aquinas comes in two places. First, the formal position for the incomprehensibility of God in Thomas’s argument (incomprehensibility does not mean that God is unknowable but that God cannot be completely known in God’s essence). This formal position of incomprehensibility addresses Hume’s contention about the design argument degrading the subsequence theology involved. Second, this type of analogical predication avoids the need to argue for a similarity between nature and manufactured items. Once order and regularity in nature is established, this is enough to make the argument work.
While Hume and Paley may not have the sort of argumentation that Thomas Aquinas used in mind in developing their arguments, it may be helpful to invoke this great natural theologian in this area. In developing his natural theology, Thomas presumes that we do not know the nature or essence of God, which would be the middle term in a syllogistic proof for the existence of God. He therefore uses an analogical type of reasoning to a cause through its effects, using the effect as a middle term in the syllogism.[4] This is particularly appropriate whenever the effect is more known than the cause.[5] Further, this sort of demonstration from effect to cause is a type of analogical predication.[6] The effect is analogous to the cause rather than holding a univocal direct relation to the cause. It can tell us something about it without disclosing the cause in all of its fullness.
Moving on to Hume’s attack on theology, Hume attacks God’s infinity, perfection, unity, and comprehensibility. Paley is apparently silent in addressing these issues. Hume first argues that the analogy of proportion in the design argument entails a finite God. He writes, “By this method of reasoning, you renounce all claim to infinity in any of the attributes of God.”[7] However, one can then argue for God as providential governor and order-giver such that a definition of God as governor and order-giver does not exhaust our knowledge of God. Order and governance are effects that we see in the natural world that point back to God as the cause.
In the face of Hume’s articulation of these theological implications of the design argument, the place and function of theistic proofs in general need to be explored. In Thomas Aquinas, his theistic proofs are part of the prolegomena of faith. Such proofs establish the ground from which one can engage in further systematic development. If God does not exist, then arguing further about the attributes of God and salvific value of Christ’s death and resurrection seems pointless. There is no reason to think that the basis for Christian claims in scripture hold the value of divine inspiration if there is no God. If God exists, then one can further argue for the inspiration of scripture and the further points of scripture. Another part of the function of theistic proofs is to articulate through natural reasoning those principles and characteristics of God that are already found in scripture. Scripture portrays God as creator of the universe, therefore arguing for a God who exemplifies the causality needed as ground for the existence of the universe coheres with how Christians think about God. So too, the design argument exemplifies God as providential governor and order-giver to the universe. In this sense, theistic proofs provide a way of thinking and talking about God.
The effect of order and directedness of nature points toward the God who providentially governs and orders nature as revealed in scripture. What is proved is one attribute of God that coheres with Christian scripture rather than the totality of God in God’s own essence. Further argument can be made to God’s infinity that does not conflict with either Christian theology or the method of argument, such as Thomas’s argument from the first cause argument to divine simplicity to the infinity of God.[8]
Hume further argues that there is no way to think that from the design argument that God is perfect. The presumption, perhaps with Liebniz’s best possible world theology in mind, is that the degree of inexplicable difficulties in the world seem at best to make it impossible to tell if the world is the product of a perfect creator. Hume argues that we would have to understand all of nature rather thoroughly and find it perfect in order to prove a perfect creator from an analogy with the natural world. Plantinga’s articulation of Liebniz’s lapse aside, it seems that again Hume is assuming that the argument stands alone and depicts God in an unqualified manner. Theologically, the argument is not necessarily meant to stand alone and portray a perfect creator without further argument.
This same point against Hume’s attack on God’s infinity and perfection also applies to his attack on the unity of God. Hume makes the comment that we cannot know from the design argument if God is one or many. He claims that there could be an unknown multitude of designers. While it could be pointed out that there are some uniformities in nature that seem to indicate that it is the product of one creator rather than many, it is not necessary to do so. The argument for the unity of God does not need to be contained in the design argument itself for the argument to be valid. Such arguments can be made later.
Hume further argues that the sort of God needed to satisfy the design argument would be beyond analogy and comprehension. Of course, being beyond comprehension does not mean that God is beyond knowing or contact with humanity. The term ‘comprehension’ classically means beyond complete understanding rather than beyond understanding at all. The God of Christianity is beyond comprehension in precisely this sort of way, and Hume does not seem to understand this. This is understood from such diverse theologians as Thomas Aquinas and John Calvin.[9] The concept is also depicted in scripture as Job learns when God addresses him from the whirlwind, and as written in Isaiah, “For as the heavens are higher than the earth, So are My ways higher than your ways and My thoughts higher than your thoughts” (Isa. 55:9, NASB). There is therefore no theological problem with Hume’s contention here. In fact, the incomprehensibility of God is the reason why God is self-revealing. God reveals Himself in his Word through scripture and the person of Jesus Christ in order to reveal to us what we cannot derive on our own.
The theological principles and conceptual tools available in Christian theology therefore are able to address Hume’s theological concerns. Further, the type of analogical reasoning found in Aquinas’s natural philosophy may be helpful in addressing Hume’s basic critique of the design argument. This is merely a beginning that holds many places for further development. Still, it seems that far from posing an insurmountable challenge to Christian theology, Hume merely betrays his ignorance of the Christian theological and philosophical tradition.
-Bruce Paolozzi
(Cross Posted at principia saceris)
[1] David Hume, Dialogues Concerning Natural Religion, part 5 para 12
[2] Hume, Dialogues.
[3] Thomas Aquinas, Commentary on Aristotle’s Posterior Analytics (CPA), (Dumb Ox Books, 2007), pp. 105-107.
[4] Thomas Aquinas, Summa Theologica (ST), I, 2, 2. See also, Summa Contra Gentiles (SCG), I, 12, [8].
[5] Thomas Aquinas, Commentary on Aristotle’s Posterior Analytics (PA), tr. Richard H. Berquist, Aristotelian Commentary Series (Notre Dame, IN: Dumb Ox Books, 2007), p 106.
[6] ST, I.13.5.ad 1.
[7] Hume.
[8] ST, I.3 & I.7.
[9] Thomas Aquinas, ST I.12. John Calvin, Institutes of the Christian Religion, book I, chapter 5.




Bruce has been married to Cathy for over nine years; they have two young sons. Bruce and Cathy are both occasional contributors to the radio show. Bruce was an early staff member of Apologetics.com, and was a regular radio show host and panelist from 2000 to 2002. His interests revolve around the problem of evil, the possibility of theological language, medieval theology, arguments for the existence of God, classical metaphysics, divine attributes, and post-modernity and religion. Bruce is currently working on a Ph.D. in Philosophy of Religion and Theology in the School of Religion at Claremont Graduate University. His area of academic specialization is in the thought of Thomas Aquinas.

