A Humean Dillema for Analogies and Artifacts

William Paley

"I knew Paley’s argument from design, knew about the watch and the watchmaker, and I knew now that these people—these Jesus freaks—were trundling out the same old argument dressed in new clothes. Intricacy requires design, that was what they said. And design requires a designer. That was as far as they could see, that was it, case closed: God exists. And the earth is ten thousand years old, just like the Bible says" (Dave, a character in a T.C Boyle story).[1]

William Paley’s analogical argument from design is simultaneously the best known and most derided argument for the existence of God. Ironically, the derision started before Paley wrote it. David Hume had been busy dismantling analogical arguments some years before Paley put quill to parchment. In this paper I will argue that though most of Hume’s objections to the argument fail, there is one that succeeds in sufficiently weakening the probability of the conclusion. I will argue that the objection forces the proponent of the analogical argument from design (AAD) into a dilemma. Either the AAD proponent must appeal to a priori knowledge or show that she has had sufficiently relevant experiences of divine designers. Unless the AAD proponent can claim to have experienced a divine designer she is forced to appeal to a priori knowledge. If she appeals to a priori knowledge the argument is no longer an argument from experience. At best this leaves us with a weak analogy that is limited to a hypothesis about a divine designer whom we can know very little about.[2] In honor of Paley’s valiance in the face of ridicule I quote him in full:

In crossing a heath, suppose I pitched my foot against a stone, and were asked how the stone came to be there, I might possibly answer, that, for anything I knew to the contrary, it had laid there for ever; nor would it, perhaps, be very easy to show the absurdity of this answer. But suppose I had found a watch upon the ground, and it should be enquired how the watch happened to be in that place, I should hardly think of the answer which I had before given, that, for anything I knew, the watch might have always been there yet why should not this answer serve for the watch as well as four the stone? Why is it not as admissible in the second case as in the first? For this reason, and for no other, viz., that, when we come to inspect the watch, we perceive (What we could not discover in the stone ) that it’s several parts are framed and put together for a purpose, e.g., that they are so formed and adjusted as to produce motion, and that motion so regulated as to point out the hour of the day; that, if the different parts had been differently shaped from what they are, of a different size from what they are, or placed after any other manner, or in any other order then that in which they are placed, either no emotion at all would have been carried on in the machine, or none which would have answered the use that is now served by it.… this mechanism being observed, the inference, we think, is inevitable, that the watch had a maker: that there must have existed at some time, and some place or other, and artificer or artificers who formed it for the purpose which we find it actually to answer: Who comprehended its construction and designed its use… every indication of contrivance, every manifestation of design, which existed in the watch, exists in the works of nature; With the difference, on the side of nature, of being greater and more, and that in a degree which exceeds all computation.[3]

It is important to to note that Paley’s argument is not merely an inductive design argument. Rather, Paley’s conclusion depends upon a sufficient analogy between the relationship between the human designer of an artifact—in this case, a watch—and a divine designer of the world. To argue by analogy is to compare two entities and their attributes. An argument by analogy catalogues the attributes that are the same for two entities and then moves to an additional attribute known to be an attribute of one of the entities and suggests that the other entity has that attribute as well. An argument by analogy is strengthened by appealing to the relevancy, number and diversity of similarities between the analogues. It is weakened by appealing to a disanalogy, pointing out dissimilar attributes, and specificity in the conclusion.[4] A rough version of Paley’s argument runs like this:

(1) We know that a watch is designed
(2) The world is analogous to a watch
(3) For all we know, the world is designed.[5]

The first premise rests upon a disanalogy. Paley asks us to consider our reaction to finding a watch as we are crossing a heath as opposed to our reaction to kicking a rock on the heath. While we would not be surprised to find a rock, our attention to the watch would produce a different reaction. The watch, a complex mechanism endowed with purpose, would force the walker to presume that an intentional action carried out by an agent brought the watch into existence.

Paley’s argument is an inductive a posteriori argument. Paley’s appeal is to the common ground of experience. Both theist and atheist can imagine discovering a watch on a heath and concluding its design. While this creates common ground between Paley and his opponent it simultaneously restricts his grounds. He cannot, for example, having appealed to experience, smuggle in any a priori premises. This will be crucial to Hume’s criticism. Hume regards experience as foundational to any claim.[6]

Paley’s first premise does not rest solely upon the appearance of design in an artifact, as if one could know what designedness is prior to any experience of artifacts, but the knowledge that artifacts have designers. While he might tell the story as if one could spot designedness merely from experiencing the artifact he thinks that knowledge of its design is possible in virtue of background knowledge including knowledge of designers. As such, his argument appeals to an analogous relationship between designers of artifacts and the designer of the world. If the world exhibits qualities that are similar to artifacts, then the process by which the world came into being is similar to how the artifact came into being. Paley knows that artifacts are designed (from prior experience of artifact design). Therefore, he concludes, the world is probably designed.

The second premise relies on building an analogy between the watch in the world. Strictly speaking, Paley argues that within the world one can see particular evidences of design. He generalizes from specific instances of design to the conclusion that the world had a designer. Paley compares an eye with a telescope. Both appear to be complex mechanisms endowed with purpose or proper function. Clearly, Paley tells us, both complex mechanisms indicate design. If design is indicated not only by the telescope, which had a human designer, but by the eye that uses the telescope, then the conclusion, that the whole world was designed, follows by generalization. The step from the analogy between the artifact and the world passes through, as it were, an analogy between an instance of design found within the world—the human eye—to a generalization about the whole world. The reason for this step is that there are no other experiences of worlds to suggest any law like relation that obtains between them.[7] On the other hand, specific instances of entities that one is able to have multiple experiences of, such as eyes and other complex functioning organisms, can be grouped into a kind according to a set of law like common features. Since there is no other world-like item to compare our world with, the argument proceeds from the observation that several organic entities possess the relevant feature of design and therefore we can infer that the whole world posses such a feature.[8]

Enough has been said about the argument itself to prepare it for Humean criticism. It is important to note that Hume’s critique of the argument from design precedes Paley’s version of it in terms of publication. The argument, however, was in common use my other apologists impressed by the Newtonian worldview.[9] The Newtonian worldview had inspired mechanism, a view that “all of the properties of physical objects…can be accounted for in terms of mass, motion, charge, and so on, and therefore the laws governing these properties can give a complete explanation of all physical occurrences that can be explained.”[10] Paley’s appeal, then, is an appeal to a common conception of nature held supposedly by Christian and opponent alike.

Humean critiques of arguments for design from analogy are generally grouped into four. First, Hume attacks the analogous relationship between artifacts and the world. This criticism focuses on the generalization from specific instances of apparent design to the whole world having the same feature. Second, Hume suggests there are other competing designers such as demons or multiple gods that could explain the order of the world. Third, Hume suggests that laws of nature would work equally as an explanation for design. Fourth, Hume argued that since ideas and the life of the mind seems to be self-organizing there is no reason to suppose that matter cannot be similarly self-organizing.[11]

Hume’s central criticism of AAD is that the analogy is, at best, weak. His criticisms focus on the relevant experiences sufficient to conclude that the world was designed. Hume suggests that the analogy between the watch and the world is not as close as the AAD proponent suggests. Hume suggests that human artefacts do not possess attributes which are similar enough to warrant the conclusion. While one can plainly see the similarity between a watch and other humanly built artifacts, there is precious little reason to suppose that these human artifacts possess the same attributes as the world. The relevant feature lacking from the analogy is found in the experience of the AAD proponent. He argues that while we know from experience that a house has a builder we cannot know any such thing when it comes to the world. Recall that, in part, Paley’s argument rests on not the intrinsic properties of two objects but on an inference rooted in experience. Hume argues that we know from experience that a house is built by human builders but because we do not know from experience that the world was built by anyone we cannot form an analogous relationship between them. What we can know from experience is that the world exhibits order but we cannot know that there is an intentional agent behind such order.

Hume argues that if the theist wants to argue from experience he can only provide instances of observed conjunctions of objects. Hume’s Philo argues that valid inferences are drawn from experiences and produce facts which are testable through observation. He concludes: “When two species of objects have always been observed to become join together, I can infer, by custom, the existence of one where ever I see the existence of the other.”[12] Since no one has observed an intelligent designer, such as the one in question, causing anything to happen to an object we cannot conclude merely why the observation of defects that the causing question is intelligent let alone divine. If, however, the apologist was to appeal instead to some idea and not experience then the argument no longer can appeal to observation. In other words, if the idea is supposed to be found in experience but experience cannot provide it, then it is not open to the theist to appeal to any a priori reasoning to support his conclusion. In essence, Hume responds to the third premise, the claim that for all we know the world is designed, by suggesting that either we do know but we do not know through experience (and thus refusing the second premise) or we do not know that the world is designed.

Paul Russell, in his book, The Riddle of Hume’s Treatise, suggests that Hume’s engagement with Joseph Butler’s analogy formed the basis for his views on inductive reasoning. Butler’s analogy was that supposed to show that our experience of God’s world serves to provide strong evidence for a religion consistent with revealed religion. Our reasoning, Butler contended, is inductive, relying on the uniformity of nature - our experiences in the past will be repeated in the future. Hume’s response to Butler demonstrates Hume’s treatment of providing a foundation for probable reasoning based on analogy. Hume distinguishes between what in practice one relies on but what one cannot know in principle. While it is true that in practice we can infer from past experience what is highly likely in the future we cannot know in principle that the future will resemble the past. This observation sheds light on Hume’s reluctance to accept any argument from analogy supported by experience that seeks to demonstrate something that cannot only be known practically but can be known in principle. Hume’s “skeptical argument leads to the conclusion that none of our probable reasoning can be justified period insofar as religious inductive reasoning is considered at all, it is just one instance of this general skeptical problem. That is to say, that granted that the uniformity of nature lacks any justification, there is no basis for any reasoning from the past to the future.”[13]

The AAD defender could suggest that the limitations we have in our experience of the world’s designer is analogous to the experience of a Martian who, upon visiting earth is confronted both with the rock and with the watch. It seems plausible to think that the Martian would concur with Paley. The watch would appear to be designed even to a person who has never met a designer. In response, it might be said that a Martian who has never experienced any design or designer is an unintelligible thing. How would the Martian be able to agree with Paley if the Martian has never experienced anything like design? Knowing about design is not obtained through a priori means and if a person has had no experience of design or someone designing something, then the Martian would not have the conceptual equipment to answer the question as to whether or not the watch or the world appears designed.

Hume assumes that designedness cannot be found in matter. It is only upon experience of an artificer designing an artifact from matter that the concept of design is possible. Once one has experienced design one can observe similar features. This line of thought clearly expresses Hume’s anti-Aristotelian worldview. Matter, on Hume’s view is inert and can have no intrinsic capacities.[14] Paley’s view supposes that since we can observe purposive design there must be a designer who gives purpose to matter. If the AAP proponent thinks matter has intrinsic proper function, then there is no need to bring God into the equation. If, however, matter requires the intentional activity of a mind to give it purpose, then it must be shown from matter that there is a purpose to it. However, Hume argues, matter itself can show us no such thing. If it is possible to think that some object is designed one must have experienced a designer designing an artifact in the past. The Martian has had no experience. Therefore, the Martian cannot know that the watch is designed.

Perhaps the Martian has experienced artifacts and their designers, but on his planet there is no such thing as telling time. Perhaps he has so many suns that it is always light and he is of a species of Martian that never sleeps. He couldn’t mark time because he is not really aware of it as we earthlings with our seasons and nights are. Would the Martian recognize the watch as an artifact or would he see it like sees the rock? It seems plausible that the Martian would see a difference. He might even infer that the watch is an artifact. Crucially, he could know this without having a clue what the watch is for. As with the watch so with the world. If one is sufficiently able to spot an artifact, then one does not need to know the particular artifact is for. Knowing what design is does not entail that one has to know what a designed thing is for.[15]

Analogously, the world, even if it is conceded that it displays features that appear to be the result of an intentional mind does not tell us what it is designed to do. What does it mean for the world to be designed? The most obvious response is to suggest that just as the watch is designed for the purpose of telling the time, the world is designed for some purpose. The trouble is that coming up with the purpose for the world is slightly more difficult that suggesting the purpose of a watch. Perhaps a child comes across a tachistoscope. Would the child know that the tachistoscope is an artifact? It seems plausible to think so. Like our Martian, the child recognizes design without knowing the purpose of the artifact. If this is true, then it is equally plausible to confer such an observation on the world. While we can’t tell what the world is for just by looking at it, it might be possible to suggest that there is a purpose for it even if we don’t know what it is.

Though this view might be plausible to some, a Humean critic may find it weak. A Humean ontology includes inert entities that are related to each other spatially and temporally. Classes of entities are defined by similarities between entities. However, in order to posit a similarity one needs a class of entities each of whose members instantiate properties that resemble other members. However, while one can infer that there are a class of artifacts that exhibit features of design (i.e. they are made for some purpose or other), there is nothing that exists that resembles the world. There is a difference between experiences of artifacts and creations. I have had frequent experiences of artifacts and the designers who made them. In contrast, I have not had any experiences of any worlds apart from this one. If, in order to determine a class of designed worlds, I need more than one, then there is no way to determine the probability of God having created this universe given the order of this universe verses some other universe. Since the argument is supposed to show the probability a posteriori, the probability is indeterminate.

According to Hume, an analogical argument of this stripe can only appeal to experience. That is, after all, the common ground the AAD proponent seeks to build on. Since no one has either observed any designer making a world nor experienced being in an undersigned world, there is no reason to think the world we do live in is designed. Hume’s central criticism then is aimed at showing the irrelevancy of the experiences in question. In this paper I have argued that the latter argument forces the proponent of the AAD into a dilemma. Either the AAD proponent must appeal to a priori knowledge or he must show that he has had sufficiently relevant experiences of divine designers. Since the AAD proponent cannot claim to have experienced a divine designer designing anything he is forced to appeal to a priori knowledge. If he appeals to a priori knowledge the argument is no longer an argument from experience.

[1] The sentiments of Dave, a character in “Bulletproof,” a story about an atheist going up against a Christian teenager in a debate over creationist curriculum in American schools. T.C. Boyle, “Bulletproof,” in Wild Child and Other Stories (New York: Penguin, 2010), 96.
[2] There is a way to strengthen the conclusion by appealing to religious experience. One might suggest that AAD coupled with an argument from religious experience might strengthen the conclusion. In this paper, I focus exclusively on Paley’s (and Cleanthes’) form of the argument and I take the argument on its own merit without the context of a more culminative case argument. An AAD may indeed offer evidence as part of a culminative case for theism but taken alone I am inclined to agree with Hume’s central criticism. Furthermore, for the sake of retaining a narrow scope, I will ignore the contemporary cousin of Paley’s argument – the fine-tuning argument.
[3] William Paley, “The Analogical Teleological Argument,” in Philosophy of Religion: Selected Readings, eds. Peterson, Hasker, Reichenbach & Basinger (New York: Oxford University Press, 2010), 212–213.
[4] Patrick Hurley, A Concise Introduction to Logic (Stamford, CT: Cengage Learning, 2008), 527.
[5] My version stresses the epistemological content of the argument. The criticism I am going to focus on depends on highlighting this aspect of the argument. If the reader prefers it can be stated more simply as one might find it in an introduction to the philosophy of religion: (1) A watch displays features of design (2) the world, like the watch, displays features of design (3) designed things have designers (4) ergo, the world has a designer.
[6] “That all inferences, Cleanthes, concerning fact are founded on experience, and that all experimental reasonings are founded on the supposition, that similar causes prove similar effects, and similar effects similar causes.” David Hume, Dialogues Concerning Natural Religion (New York: Cambridge University Press, 2007), 23.
[7] Keith Yandell, Hume’s “Inexplicable Mystery”: His Views On Religion (Philadelphia: Temple University Press, 1990), 177.
[8] Whether or not this is a fallacious move in the argument rests on whether or not one can mount a case that a sufficient quantity of experiences of entities within the world would give one warrant to generalize to the whole. But even this won’t do it. As Yandell remarks, it is possible to suggest that even though all the parts are machines, this still does not necessitate that the whole is a machine. Yandell, 179.
[9] Paul Russell, The Riddle of Hume’s Treatise (New York: Oxford University Press, 2008), 147–166.
[10] James Madden, Mind, Matter and Nature (Washington DC: Catholic University of America Press, 2013), 219–220.
[11] I am roughly following Tim Mawson’s characterization of Hume’s criticisms. T. J. Mason, Belief in God: An Introduction to the Philosophy of Religion (New York: Oxford University Press, 2005), 134–135.
[12] David Hume, Dialogues Concerning Natural Religion (New York: Cambridge University Press, 2007), 26.
[13] Russell, 139.
[14] Strictly speaking, Hume’s point was epistemological: one cannot know from experience of matter whether or not it is endowed with capacities or powers.
[15] Resting in the background of this criticism is Hume’s view of causation. Hume thought that causation was not apparent through observation even while we practically assume it. In human work? Hume argued that if Adam were to confront two billiard balls having experienced no causes or affects before and if he was asked to predict what would happen when one billiard ball was rolled into the next, Adam would not know. After seeing many such instances of one billiard ball apparently causing another to move Adam would begin to conclude that the first billiard ball was causing the movement of the second. This, Hume argued, would be a mistake for there has been no additional knowledge added to Adam’s experience of billiard balls in comparison to the first experience. Therefore, Hume concluded that though in practice we rely on cause and effect it cannot be shown by observation.