Religion, Mind, and Society Critical Commentary

DeCruzDeSmedt_IntroChs48.pdf

Home >Psychology homework help >Religion, Mind, and Society Critical Commentary

4 Teleology, the Design Stance, and the Argument from

Design

For many scientists and philosophers, Hume ’ s philosophical critique (1779) and Darwin ’ s theory of natural selection (1859) have undermined the idea that we can draw any analogy from design in artifacts to design in nature. Yet the argument from design stands as one of the most intuitively compel- ling arguments for the existence of a divine creator. It enjoys an enduring appeal, going back as early as Socrates (recorded in Xenophon ’ s Memora- bilia ), Plato ’ s Timaeus , and Cicero ’ s De Natura Deorum . The design argument can also be found outside of Western religious traditions, for instance in Hinduism, with authors such as Ś a ṅ kara and Udayana defending it. From the seventeenth to the nineteenth century, it was the most popular natu- ral theological argument for the existence of God, with Paley as the best- known, albeit a late, example. Since the later decades of the twentieth century, the design argument has been experiencing a renaissance, espe- cially in the form of the fine-tuning argument, which holds that the physi- cal constants, the laws of nature, and the initial conditions of the universe are finely tuned to be life-permitting.

In what follows, we will examine empirical studies from developmen- tal and experimental psychology to investigate the cognitive basis of the design argument. We will focus on two aspects: the tendency of humans to discern teleology in nature and the way they intuitively assess probabili- ties. A better understanding of these features not only elucidates the lasting popularity of the design argument but can also help theists and nontheists to construct a rational basis for disagreement.

De, Cruz, Helen, and Smedt, Johan De. A Natural History of Natural Theology : The Cognitive Science of Theology and Philosophy of Religion, MIT Press, 2014. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/miami/detail.action?docID=3339922. Created from miami on 2018-01-25 05:48:13.

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

62 Chapter 4

The Argument from Design

The argument for the existence of a divine creator, based on evidence of design in nature, has recently enjoyed a revival in theology and philoso- phy (as exemplified by papers collected in Manson 2003 ). Even a hardened atheist like Dawkins (1986 , 4 – 5) praises the argument for its explanatory coherence and intuitive appeal. Paley ’ s image of the watch on the heath was certainly not the earliest formulation of the design argument, nor even of the watch analogy, but its familiarity makes it a suitable starting point. His argument can be summarized as follows: if one encounters a watch, the complexity of this artifact and the interrelations of its parts lead to the inference that it is the product of purposive design. From this, Paley ([1802] 2006 , 7 – 15) concluded that complexity in nature is also the product of a designer because proposing that it could have been brought about by chance would be absurd.

Elliott Sober (2004 , 118) formalizes Paley ’ s design argument (and other arguments like it) as follows: there is an observation O and two possible hypotheses ( H 1 and H 2 ) to explain it:

O : The watch has features, in particular, design, complexity, and a certain delicacy, that is, if its parts were a bit different, it would function less well or not at all. There are two hypotheses to explain these features: H 1 : The watch was created by an intelligent designer. H 2 : The watch was produced by a mindless chance process.

Paley relies on the intuition that O would be unsurprising if H 1 were true but highly surprising if H 2 were true. O therefore favors H 1 over H 2 . Paley surmises that organisms are similar to watches in that they exhibit com- plexity, design, and fine-tuning. Therefore, running the same argument, H 1 is a better explanation for these features than H 2 . From this we can con- clude that organisms are created by an intelligent designer.

This way of arguing is not unusual. The Hindu philosopher Ś a ṅ kara (8th c. CE ) posited an intelligent, purposive craftsman (Brahman) as the origina- tor of the universe. His design argument was intended to show that mate- rialism, the view that the world originates from a non-intelligent, material first principle is a worse explanation for the apparent design of the world than creationism:

Consider that in ordinary life no non-intelligent entity is observed to produce mod- ifications suitable for satisfying the purposes of some particular person, by itself,

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

Teleology, the Design Stance, and the Argument from Design 63

without being superintended by an intelligent agent. In ordinary life what we do see is that houses, palaces, couches, seats, pleasure-gardens, and the like, which are useful for obtaining pleasure and avoiding pain at appropriate times, are constructed ( racit ā ) by intelligent craftsmen. In like manner, observe that this entire universe, externally consisting of the earth and other elements, is suitable for experiencing the fruits of various acts. … Since even the most competent craftsmen cannot com- prehend (the world ’ s construction), how could the non-intelligent Material Nature ( pradh ā na ) devise ( racayet ) it? In the case of such things as a lump of earth or a stone, no (power of contrivance) is seen, but the design ( racan ā ) of special forms out of such things as clay is seen when they are superintended by potters and the like. In the same way, Material Nature (transforms itself) only when connected with a superintending, external intelligence. … Therefore, since the design of the world is otherwise inexplicable ( racan ā – anupapatti ), its cause is not to be inferred as non- intelligent. ( Ś a ṅ kara, cited in Brown 2008 , 108)

These arguments from design have interesting epistemic properties: they are both an analogy and an argument to the best explanation. Analogies map the structure of a well-known domain (the source domain) onto a lesser-known problem (the target domain). In distant analogies, the struc- tures of source and target domains greatly differ in their basic ontological properties. The design argument is a distant analogy in that it maps the artificial domain (source domain) onto the natural world (target domain). Artifacts exhibit goal-directedness in their design: they are intentionally cre- ated by designers who have their functions in mind. As organisms exhibit goal-directedness, they must also be products of a purposeful designer.

Although this analogical structure has often been attacked on the ground that it is inconclusive (e.g., Hume 1779 ; Frank 2004 ), it is worth noting that distant analogies are well-established in scientific practice as a way to gain insight into new problems ( De Cruz and De Smedt 2010b ). When the conceptual structure of the target domain is relatively unknown, as in the case of scientific discovery, scientists often resort to analogical reasoning. Historical examples include Johannes Kepler ’ s mapping of the properties of gravity onto the properties of light (i.e., the fact that sunlight dissipates with increasing distance between the Sun and the planets it is cast upon) to explain why planets further from the Sun move more slowly, in this case, caused by a weakening of the gravitational force with increasing dis- tance from the Sun ( Gentner et al. 1997 ). A contemporary example is the Swiss army knife analogy as a way to conceptualize the evolved structure of the human mind in evolutionary psychology ( Cosmides and Tooby 1994a , 88). The analogical structure of the design argument is thus epistemic; it

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

64 Chapter 4

is performed to gain insight into an unfamiliar domain, God ’ s creation. In many early versions, the design argument served a heuristic, rather than a strictly argumentative purpose, as in the works of seventeenth-century natural philosophers such as Nieuwentijt, van Leeuwenhoek, and Ray, who perceived design as a source of wonderment about nature.

Traditionally, design arguments had an inductive argumentative struc- ture. They began with the empirical observation that all complex, func- tional objects of known origin were products of intelligent design. Then came the inductive step, in which one infers that what is true for some members of a class is true for all members. Hume (1779 , 56 – 60) dispensed with this way of reasoning by arguing that artifacts and biological organ- isms are too dissimilar to be classed together. When we see a house, we can reasonably infer that it has an architect or builder because we know from experience that this particular effect flows from that particular cause. But we have no guarantee that the universe bears such a resemblance to a house as to invoke a cosmic designer; we do not know in how far our analogy is reliable.

Although Paley did not mention Hume explicitly, his watchmaker anal- ogy sought to escape this criticism by adopting a different strategy. It is an argument to the best explanation that relies on an inference to the best explanation (IBE), which has the following structure:

IBE Given evidence E and a pool of plausible, potential explanations H 1 , … , H n of E , if H i explains E better than any of the other hypotheses, infer that H i is closer to the truth than any of these others. ( Douven 2002 , 359)

IBE enables us to probabilistically infer that a given hypothesis is closer to the truth than other hypotheses because it explains the available evi- dence better than rival explanations. In this probabilistic aspect, the design argument differs from deductive proofs for the existence of God, such as Anselm ’ s ontological proof (see Sober 2002 ; Swinburne 2004 , for probabilis- tic discussions of the design argument). Arguments to the best explanation are meant to avoid Hume ’ s critique by not relying on induction. Rather than having to make a positive case for why artificial design is analogous to divine design, they argue that there is no better explanation for order and complexity than design, regardless of where these properties are instanti- ated ( Gliboff 2000 ). Having established apparent design and probabilistic inference as two key properties of the design argument, we will now exam- ine their cognitive basis.

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

Teleology, the Design Stance, and the Argument from Design 65

How We Infer Design

The Design Stance To Paley, the conclusion that a watch is purposefully designed was self- evident. Our perception of its interrelated parts, formed and adjusted to each other — the coiled elastic spring, the flexible chains, the cogwheels — each fashioned out of the materials that suit their intended function best, should lead us to infer that the watch must have had a maker who formed it for a specific purpose. Yet Paley acknowledged that such seemingly spon- taneous inferences require contextual knowledge about the artifact under consideration: “ It requires indeed an examination of the instrument, and perhaps some previous knowledge of the subject, to perceive and under- stand it ” (Paley [1802] 2006, 8). In the case of the watch, Paley could infer the intent of the designer, as he was familiar with the class of artifacts to which the watch belongs.

What would happen if Paley had pitched his foot against an unfamiliar object, such as a laptop? Would he have inferred design as automatically as in the case of the watch? Its sleek shape, carefully integrated buttons, and intended functions would have presented a puzzle to him. Unfamil- iar artifacts can be so outlandish that people can be led to believe that they are not the work of human designers. From the nineteenth-century wave of European colonization onward, and especially during World War II, the indigenous inhabitants of Papua New Guinea were confronted with a cornucopia of Western goods. They believed that these were gifts of the ancestors. This led to the emergence of cargo cults, whose adherents are still trying to ritually lure airplanes into dropping more cargo, more Western goods. And what to think of people who observe UFOs? Often these alleged spaceships are no more than military reconnaissance airplanes, crashing weather balloons, or even bright planets. Nevertheless, people do infer that the objects are intentionally designed by supernatural agents and by extra- terrestrials respectively. To gain a better understanding of the design argu- ment, it is therefore useful to examine how humans infer design and how creator and artifact are causally linked.

Bloom (1996) has argued that manufacturing and understanding arti- facts is governed by an intuitive design stance, a term he borrowed from Daniel Dennett (1987 , 16 – 17). As we saw in chapter 2, humans take the design stance when they consider an object as designed to fulfill some

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

66 Chapter 4

function or as belonging to some class of objects. Remarkably, neither com- plexity nor order is a necessary or sufficient condition to decide whether an object is purposefully created. Gelman and Ebeling (1998) showed two- year-olds a stain vaguely resembling a bear. They told some subjects that the spot was created accidentally, by kicking over a bottle of paint, whereas another group learned that the shape was painted intentionally. Only the children in the latter group called it a bear. Thus the perception of the stain as accidental or representational depends critically on the prior infor- mation the children received on how it was brought about. In a similar experiment ( Gelman and Bloom 2000 ), older children and adults saw an object but were given two diverging reports on how it came into being. In the unintentional version, a strip of cloth was accidentally caught in a machine, resulting in holes being punched at regular intervals. In the intentional account, a person took a pair of scissors and carefully cut evenly spaced holes. Participants were more prone to call the object a belt if they believed it was intentionally created.

The role of background information in our appraisal of design intentions is also detectable at the neural level. In an elegantly designed fMRI study ( Steinbeis and Koelsch 2009 ), participants listened to a piece of electronic music. Half of them were told that the composition was written by a human composer, whereas the others were led to believe it was purely computer- generated. The first group of participants, but not the latter, exhibited a high activation in the anterior medial frontal cortex, superior temporal sul- cus, and temporal poles, brain areas involved in the attribution of mental states and the inference of intentions of others. These experiments suggest that attributing design requires some background knowledge of the inten- tions of the maker: our judgment that something is an artifact depends on our knowing that the object was intentionally created. When natural theo- logians appeal to perceived design, they rely on background assumptions about the designer, that is, they assume theism. As John Henry Newman ([1870] 1973) revealingly put it, “ I believe in design because I believe in God, not in a God because I see design. ”

Once we are familiar with specific classes of artifacts, we can reasonably infer that a particular member of a class was created with the intention of belonging to that class. When we see a broken chair in a pile of rubbish, we conclude that the object was created to fulfill a specific function (to sit on) and to be of a specific class (chairs). This stance also provides useful

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

Teleology, the Design Stance, and the Argument from Design 67

inferences when we have to identify classes of nonutilitarian objects, such as ships in bottles: although these boats will never sail, we still call them ships because the designer intended them to belong to this class of objects. It was on this inference that Paley relied in his assertion that the watch was a product of intentional design.

Conversely, knowledge of the designer ’ s intention can help us to infer an object ’ s intended function or identity. Take as an example Bloom and Mark- son ’ s (1998) experiments in which three- and four-year-olds were shown featureless ovals that were purportedly drawn by a child with a broken arm, who, because of this, could not draw well. The young subjects were told that these were drawings of chickens (three vertical ovals) and a pig (one horizontal oval). When prompted, the preschoolers effortlessly identified the pig, because they reasonably inferred that the artist would draw objects from another category in a different way. Young children intuitively regard the creator of an artifact as having privileged knowledge about both its name and its intended function. This was illustrated by an experiment in which preschoolers saw objects that were given an anomalous label; for example, the experimenter showed the child a key-shaped object and said, “ You are not going to believe this, but this is actually a spoon ” ( Jaswal 2006 ). Children were only willing to adopt the anomalous name if the experimenter referred to the object as something he had created, not as something he had merely found. Recognizing that the creator of an artifact has the prerogative to name it marks an important step in the development of the design stance.

The human propensity of inferring design may be due to the distinct evolutionary history of our species, which relies to a unique extent on tools. Whereas other primates mostly use unmodified objects as tools, archeologi- cal evidence for stone-knapping in hominids goes back as far as 2.5 million years ago ( Semaw et al. 1997 ). By adopting the design stance, hominid chil- dren might have learned to use and fashion tools more efficiently. Indeed, comparative studies of social learning in children and chimpanzees (e.g., Horner and Whiten 2005 ) reveal stark contrasts in the way new tools are used: whereas children take the intention of the person who demonstrates these tools as guidance, chimpanzees rely more extensively on the physi- cal properties of tools to figure out how they work. The design stance pro- vides children with a useful heuristic to learn about their environment. It allows them to “ ignore the actual (possibly messy) details of the physical

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

68 Chapter 4

constitution of an object, and [relying on] the assumption that it has a certain design, predict that it will behave as it is designed to behave under various circumstances ” ( Dennett 1987 , 16 – 17). Without the design stance, we would not possibly learn to use and name hundreds of tools and other artifacts, but would perhaps be limited to the less than ten tool types a typi- cal community of wild chimpanzees entertains ( Whiten et al. 1999 ).

Intuitive Teleology Whereas the design stance might have evolved in response to the regular use of artifacts throughout human evolution with the function of rapidly categorizing and using artifacts, humans also possess a natural propensity for teleological reasoning. As we saw in chapter 2, this inclination is most pronounced when thinking about biological entities, but it can apply to almost all categories of objects. Cross-cultural studies (e.g., H. C. Barrett 2004 ) indicate that humans have the intuition that animals and plants possess adaptations that are self-beneficial, such as claws for defense or thorns for protection against being eaten. Young children, however, not only attribute purpose to artifacts and biological adaptations, but also to entire organisms ( “ What are lions for? ” “ To go in the zoo ” ) and nonliving natural kinds like clouds ( “ For raining ” ) — a tendency Kelemen (2004) terms promiscuous teleology. Moreover, when given a choice between teleologi- cal and non-teleological explanations, preschoolers and elementary school children prefer teleological accounts. When asked whether rocks are pointy because of natural processes, such as “ bits of stuff piled up for a long period of time, ” or because of teleological functions, for example, “ so that animals could scratch on them when they got itchy, ” children typically endorse the latter ( Kelemen 2003 ).

Around ten to twelve years of age, the preference for teleological expla- nations lessens, probably because adolescents acquire elaborate coherent mechanistic explanations through schooling. Although mountains can be climbed, few adults would claim that mountains are there to climb on. This is because our learned knowledge that mountains are formed by tec- tonic activity or volcanism is incompatible with teleological explanations, where the function provides a sufficient reason for why the structure exists. Remarkably, patients with Alzheimer ’ s disease show a reemerging prefer- ence for teleological explanations. For example, they think that rain is there so that plants and animals can have water to drink and grow rather

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

Teleology, the Design Stance, and the Argument from Design 69

than the acquired explanation that rain occurs by water condensing into clouds and forming droplets ( Lombrozo, Kelemen, and Zaitchik 2007 ). An increased tendency for teleology is also observed in adults with little schooling ( Casler and Kelemen 2008 ). Formal education seems to reduce a preference for teleological explanations but cannot eradicate it. Indeed, when educated adults are forced to make very quick judgments, they too show a heightened acceptance of teleological explanations: when judging at a glance whether a statement is correct, they tend to endorse teleological, incorrect explanations such as “ the Sun radiates heat because warmth nur- tures life ” ( Kelemen and Rosset 2009 ). Reasoning strategies observed in chil- dren persist into adulthood but may be masked by secondary explanatory strategies. Once these become impaired (in the case of Alzheimer ’ s patients) or are unavailable (in the case of speeded judgments or lack of education), the intuitive, evolved strategies of childhood reemerge. Although scientific education tends to lessen teleological reasoning, the tendency to apply tele- ology is not absent even among trained scientists. While their acceptance of incorrect teleological explanations is lower than in comparison groups, chemists, geoscientists, and physicists from top-ranked American universi- ties are also more prone to endorse teleological explanations under speeded conditions than when they have unlimited time (Kelemen, Rottman, and Seston 2013). These experiments measure the extent to which people endorse spurious teleological explanations (such as “ Trees produce oxygen so that animals can breathe ” ) as a way to examine teleological tendencies. However, these experiments do not by themselves warrant the conclusion that teleological reasoning is mistaken, or that teleology applied to bio- logical entities is always wrong. They only indicate that it is a pervasive reasoning strategy. We will now examine whether teleological reasoning in biology is incorrect.

Immanuel Kant ([1790] 1987, sec. 75) was still confident that there could never be a Newton of biology who could explain how a blade of grass comes about by reference to laws of nature that do not invoke intentions or goals. He thought teleology was indispensable to biology, as it brings order to our observations. However, even as Kant was writing, pre-Darwinian evolution- ary theorists attempted to do precisely this — explain apparent design and teleology by reference to purely mechanical, mindless processes (see this vol., 75 – 76). By the twentieth century, philosophers were convinced that teleology or teleological function had no place in philosophy or scientific

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

70 Chapter 4

practice. Still, explanations in teleological terms remained (and still remain) pervasive in biology. As Ruse (1998 , 16) observes, “ Design language reigns triumphant in evolutionary biology. ” Although evolutionary biologists insist this language is metaphorical and just used as easy shorthand, it is clear that they cannot do without it, if only on a pragmatic level.

By the second half of the twentieth century, teleology became rein- troduced in philosophical discourse, and there is continued debate on whether or not teleology has a place in evolutionary biology. Some philos- ophers of biology (e.g., Cummins 2002 ) have urged that teleology should be excised from biological discourse. Other philosophers and biologists (e.g., Ayala 1970 ) attempt to provide a naturalistic grounding for teleology, whereby the function of a particular trait can be regarded as a product of design through natural selection. Note that this does not require a con- scious designer: the natural design is a consequence of the interactions of ancestral organisms that possessed this trait for their descendants ’ fitness. Authors using this principle are indeed quite adamant in their claim that biological function is a mind-independent, natural property of biological organisms “ that does not involve the goals or purposes of a psychological agent ” ( Allen and Bekoff 1995 , 611).

Are Humans Intuitive Theists?

Does the tendency to infer design in nature also entail an inference to a divine designer, as Paley, Ś a ṅ kara, and others have suggested? At this point, developmental and experimental psychological data do not present a unified picture. Lombrozo et al. (2007) found that although Alzheimer ’ s patients reasoned more teleologically, they nevertheless did not invoke God more frequently as an explanation than neurotypical older people. This experiment found no difference in religiosity between Alzheimer ’ s patients and older people without this condition. In a study that probed Dutch primary school children ’ s intuitive theories on the origin of species ( Samarapungavan and Wiers 1997 ), the answers clustered together in dif- ferent categories, including spontaneous generation, Lamarckism, and pure essentialism (i.e., animals and plants have always existed in their present form). Although many children made teleological inferences, only about 10 percent made explicit reference to God or intelligent design. On the other hand, a comparable investigation by Margaret Evans (2001) in the United

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

Teleology, the Design Stance, and the Argument from Design 71

States found that a majority of ten-year-olds endorsed creationist accounts of the origin of species, regardless of their religious background. Kelemen and DiYanni (2005) obtained comparable results with British elementary school children, although the percentage of creationist accounts was sig- nificantly lower than with American subjects.

Several possible explanations might account for these findings. A strong position holds that humans are intuitive theists. In this view, creationism is a natural mode of reasoning that is only altered when children acquire explicitly nonreligious beliefs from their cultural environment. Bering (2006) defends this position, arguing that religious beliefs are biological adaptations that were directly selected for to enhance cooperation, altru- ism, and group cohesion. Kelemen (2004) argues that children are intuitive theists, as their intuitive teleology and design stance lead spontaneously to a belief in God. It is important here to note that the teleological stance that children have toward artifacts and natural items changes over time. Younger children, under the age of five, are more flexible about the telos of objects, whereas older children and adults think more rigidly about the functions of objects (functional fixedness, see chapter 2). The intuitions that drive the design argument are these mature, functional fixedness intuitions.

A weaker position (e.g., Bloom 2007 ) holds that religious belief itself is not innate but a byproduct of other cognitive adaptations such as agency detection and theory of mind. In this view, children acquire culturally transmitted religious beliefs easily because they key in on evolved propen- sities of the human mind. Here, the step from design to designer is not automatically made but needs to be made explicit, as Paley and others in fact did.

A third view holds that intuitive theism may be an innate cognitive capacity that depends on external cultural circumstances for its develop- ment. This position was proposed by Calvin, who argued that humans are endowed with an innate propensity to believe in God ( sensus divinitatis ) but that this propensity still requires cultural input for its development. Cal- vin ([1559] 1960, book 1, chapter 3) acknowledged that, depending on the environment where one is raised, the sensus divinitatis can result in a wide variety of religious beliefs. It seems to us that the empirical evidence sur- veyed here does not support the strong intuitive theism advocated by Ber- ing and Kelemen. What is still required is an explicit link between design and a divine designer, a judgment that relies on intuitive probability.

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

72 Chapter 4

Intuitive Probability: Can Chance Events Produce Order and Complexity?

Several fields of cognitive science, such as artificial intelligence, linguistics, and the cognitive psychology of reasoning, have experienced what Oaksford and Chater (2007) call a “ probabilistic turn ” : when we cognize, we do not require or assume certainty (as when we make a deductive argument), but often rely on probabilistic reasoning. As the theologian Joseph Butler (1736) already remarked, we are finite beings restricted in time and space, and we cannot claim absolute knowledge but have to make do with probabilistic assumptions. Humans, like other animals, are sensitive to probabilities, and reason probabilistically in the face of uncertainty. Decades of research on perception, for instance, have shown that our detection of shapes depends on a variety of probabilistic inferences that our visual system automatically makes, based on shading, texture, boundaries, and shape recognition (e.g., de Gardelle and Summerfield 2011 ). Human infants rely on the probabil- ity with which syllables co-occur to distinguish words within continuous streams of speech sounds ( Aslin, Saffran, and Newport 1998 ). Even animals with simple nervous systems, such as bumblebees, tune their behavior to probabilistic information, such as the likelihood that flowers of a particular type will contain nectar ( Real 1991 ).

The design argument draws on an ability to assess posterior probability, the probability that is assigned after the relevant evidence is taken into account. How likely is it that the apparent design in nature was intention- ally created or, alternatively, that it happened by chance? Early proponents of the design argument have taken the intuition that chance does not pro- duce order as a starting point:

At this point must I not marvel that there should be anyone who can persuade himself that there are certain solid and indivisible particles of matter borne along by the force of gravity, and that the fortuitous collision of those particles produces this elaborate and beautiful world? I cannot understand why he who considers it pos- sible for this to have occurred should not also think that, if a countless number of copies of the one-and-twenty letters of the alphabet, made of gold or what you will, were thrown together into some receptacle and then shaken out on to the ground, it would be possible that they should produce the Annals of Ennius, all ready for the reader. I doubt whether chance could possibly succeed in producing even a single verse! (Cicero [45 BCE ] 1967, 213)

Cicero discarded the atomists ’ idea that chance collisions of elementary building blocks (atoms) formed the material world on the basis that chance

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

Teleology, the Design Stance, and the Argument from Design 73

has a low probability of producing order. Assuming that the twenty-one letters of the Roman alphabet are equally distributed into his “ countless number, ” the chance of the first letter falling in the correct place is 1/21, the chance that the first two letters are correct is 1/21 × 1/21 = 1/441. The chance that the letters would produce the approximately 7,000 charac- ters of the 600 lines that survive of Ennius ’ s Annales (a now fragmentary epic poem on the history of Rome) is vanishingly small, being 1/21 7,000 . Cicero ’ s intuition has been reiterated many times, including astronomer Fred Hoyle ’ s image of hurling around scrap metal at random and happen- ing to assemble a Boeing 747 ( “ Hoyle on Evolution ” 1981). 1 Although all arrangements of the scrap metal are equally improbable, very few of them will fly; similarly, although all combinations of the twenty-one Roman let- ters are equally unique, very few of them will produce a legible text, let alone the Annales . William Dembski (1998, section 6.2) follows a similar line of thought in his generic chance elimination argument : one can infer that an event occurred by design if the probability of its occurrence is suffi- ciently small.

Although it is intuitively compelling, rejecting chance as an explanation for complexity and design is problematic because, as Sober (2002) notes, there is no probabilistic equivalent of modus tollens. In other words, we cannot state that

If hypothesis H were true, observation O would be highly improbable.

But O .

Therefore, H is not true.

The law of likelihood in statistics stipulates that it is not the absolute value of the probability of data under a single hypothesis that is to be considered but rather how the probability values compare under different hypothe- ses. The intuitive idea that improbability strengthens the existence of God is problematic in that it tacitly relies on an analogy between human and divine agency. When deciding whether human design or chance is respon- sible, we rely on empirical knowledge of what human agents in fact do. In an example from Kenneth Himma (2005), adapted from Dembski, sup- pose a political candidate ’ s name appears first on the lists of voting ballots forty out of forty-one times. The probability of such an event occurring by chance is very small. But when we suspect that a county clerk rigged the list, we rely on tacit knowledge: that undecided voters are more likely to choose

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

74 Chapter 4

the first on the list and that the county clerk wants a particular candidate to win. Being an intelligent agent, it is not unlikely that he rigged the list. We also know of cases in which ballots were tampered with to win an election. Hence the hypothesis that the name was placed first forty out of forty-one times by design rather than mere chance becomes very plausible indeed.

In the case of divine action, by contrast, we do not have empirical knowl- edge to draw upon, and thus no assumptions can be made about what God would or would not do. While revealed theology can appeal to scripture to make assumptions about God ’ s intentions, this route is not available for the natural theologian, whose knowledge about this does not appeal to sources outside of reason and empirical observation. Some authors, like Swinburne, appeal to inductions, made on the assumption of God ’ s perfect goodness, to make predictions about what God would or would not do. For instance, assuming that God would perform the best action, he would create free agents like humans. However, such inferences rely on Swinburne ’ s moral intuitions, which, as he admits, may not be reliable: “ The moral intuitions on these matters that I am commending to my readers must inevitably be somewhat tentative, and, even if they are correct, very precise numerical values may not capture the resulting probabilities about what God will cre- ate ” ( Swinburne 2004 , 123). Therefore, a natural theologian cannot straight- forwardly appeal to God ’ s intentions by using empirical evidence. Without the necessary background data to make the design argument an IBE, this argument relies on an analogy between human and divine agency. Again, this version of the design argument becomes an argument from analogy. As we have seen, this was successfully attacked by Hume, and it was precisely for this reason that Paley recast the argument from design into an argument to the best explanation, relying on an IBE.

The reliability of IBE depends on the amount and quality of the data and the relevance of the data to the conclusion. If insufficient evidence is available, it may well lead one to choose “ the best of a bad lot ” ( van Fraassen 1989 , 143). In fact, the bad-lot concern applies even if one has all the possible evidence, because one may simply have failed to conceive of the true theory with this evidence in hand. Cicero overlooked the prin- ciple of cumulative selective retention as an alternative to brute chance and intelligent design. According to this principle, the nonrandom reten- tion of random events can, over time, result in complex, orderly features. An illustration of this is Dawkins ’ s (1986) weasel program. The likelihood

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

Teleology, the Design Stance, and the Argument from Design 75

that a computer program that generates random combinations of letters will produce by pure chance a phrase from Shakespeare ’ s Hamlet , such as METHINKSITISLIKEAWEASEL, is vanishingly small. However, allowing the computer to retain the correct letters at each attempt will produce the sen- tence in no more than 23 × 26 = 598 runs. Similarly, in Cicero ’ s example, if each letter that falls in the correct place is selectively retained, we need at most 21 × 7,000 trials to complete what is now left of Ennius ’ s Annales . When using an IBE strategy, most modern versions of the design argument do not take natural selection and its principle of cumulative selective reten- tion into account as a viable explanation. Dembski holds that regularity, chance, and design exhaust the possibilities. He considers regularity and chance separately but fails to take the combination of chance and regularity into account, and it is precisely this combination that we know as natural selection (with random mutations and nonrandom, regular selective pres- sures). On their own they cannot generate the appearance of design, but combined, they can.

To be sure, in 1802, let alone in 45 BCE , natural selection was not in the pool of possible explanations. However, as Sander Gliboff (2000) demon- strates, Paley did have a range of alternative materialist explanations, of which we mention four. First, necessity: because everything has to have some form, it may as well be the present form; for example, the eye is the actual realization of one of the possible ways to fill an eye socket. Second, he considered infinite trial and error: given an infinite time and universe, every possible configuration of matter could be produced, some of which turned out to be viable life-forms that persisted and reproduced — an inter- esting precursor to the concept of natural selection, proposed by Epicurus, and described in detail by Lucretius ([ca. 50 BCE ] 2007). Third, he discussed the claim that parts of organisms could arise before their function was deter- mined, a forerunner of exaptation theory (Paley [1802] 2006, 38 – 41). These alternatives were being explored and hotly debated in Paley ’ s time by early evolutionists like Buffon, Diderot, and d ’ Holbach. Paley seemed to have been familiar with these authors, as he mentioned Buffon explicitly and others implicitly. Fourth, he briefly discussed the special biological forces or organizing principles proposed by the so-called G ö ttingen school of German biologists, such as Blumenbach, Kielmeyer, and Reil (Paley [1802] 2006, 218 – 225). Although now rejected, the latter ’ s Newtonian approach to biology that stipulated forces acting on biological entities, analogous

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

76 Chapter 4

to physical forces acting on physical entities, was conceivable and widely accepted at the time ( Larson 1979 ).

Paley considered these alternative explanations carefully but ultimately dismissed them all, primarily for lack of evidence and lack of explanatory potential. For instance, he criticized Buffon ’ s (1766) appeal to “ internal moulds, ” which organized the organic particles that made up any individual creature (a sort of building plan) as a naturalistic explanation for why ani- mals belonged to particular species. Paley thought this explanation was not an explanation at all — no more than the Greek essential forms explained design satisfactorily. By contrast, he thought that a designing mind was a satisfying explanation because it is “ comprehensible to our understanding, and familiar to our experience ” (Paley [1802] 2006, 224). Ultimately, Paley believed that even if these proposed naturalistic mechanisms and propensi- ties are genuine, they still require a designer for their explanation. Biologi- cal forces and internal molds might explain the origin of biological design, but where did they originate? “ I am unwilling to give to it the name of an atheistic scheme … because, so far as I am able to understand it, the original propensities and the numberless varieties of them … are, in the plan itself, attributed to the ordination and appointment of an intelligent and design- ing Creator ” (Paley [1802] 2006, 224 – 225). This may strike one as a circular argument: he rejected these naturalistic mechanisms because he thought such mechanisms still require a designer, but that was the proposition that had to be proven. A more charitable interpretation is that Paley assigned a high prior probability to the existence of God, and hence to divine design, and a low prior probability to metaphysical naturalism. As will be argued in the next section, the likelihood of data can be meaningfully assessed only in relationship to hypotheses, which are accorded a prior probability.

A Rational Basis for Disagreement

If humans are prone to discern design and teleology in nature, why do some find the design argument more compelling than others? This may be due not to intrinsic differences in the way design and teleology are discerned but to differences in the prior probability people place on the existence of a divine designer. An interesting way to approach this prob- lem is through an examination of how humans regard coincidences. For

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

Teleology, the Design Stance, and the Argument from Design 77

Griffiths and Tenenbaum (2007) , an event is a coincidence if it is judged to have a lower probability of occurring under our current theory of how the world works than under an alternative hypothesis. Coincidence plays an important epistemic role in scientific discovery: the geophysicist Alfred Wegener (1912) noted that the coastlines of West Africa and South America fit into each other like pieces of a jigsaw puzzle, that their geological strata match, and that the distribution of species on both sides of the Atlantic is highly correlated. He thought that this pattern was not a mere coinci- dence but that these continents were once joined and had drifted apart. The physician John Snow (1855) noted that cholera outbreaks in London tended to cluster at public water pumps and inferred that this was not a coincidence but provided evidence for his theory that cholera was transmit- ted through polluted water (rather than bad air, the then-favored theory). These examples suggest an intimate connection between coincidence and evidence. A coincidence occurs when the likelihood ratio in favor of an alternative theory is insufficient to overwhelm the prior odds against it. A coincidence becomes evidence when the likelihood ratio in favor of an alternative theory overcomes the prior odds against it and leads us to accept that alternative theory. Because people differ in the prior probabilities they assign to alternative hypotheses, what is a coincidence to one person can be compelling evidence to another.

In the case of the design argument, the competing hypotheses are H mat (apparently purposive and complex structures arose strictly through natu- ral, material causes) and H deo (design as the result of a designer). In the framework of H mat , the occurrence of ordered complexity and apparent design presents a coincidence. Given that chance events tend to produce disorder, the probability of this occurrence is extremely low. The theory of natural selection has successfully solved this dilemma because it relies on a combination of chance and lawlike processes. Indeed, no other naturalistic theory can explain why living things are improbably complex, why the interrelationships between their parts are highly functional, and why they exhibit features that enhance their probability of surviving and reproduc- ing in their environment. Proponents of H mat can find justification in the evolutionary explanation of design. To justify why they favor their view rather than H deo , they can cite examples of maladaptedness and appeal to ontological parsimony because their explanation is restricted to physical

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

78 Chapter 4

processes. Under H deo , the occurrence of design is not improbable because this theory explicitly proposes a designer who made the universe orderly and purposeful. Under these epistemic circumstances — but not under H mat — design in nature becomes corroborative evidence for the existence of a divine creator. Next to this, natural theologians can also appeal to onto- logical parsimony because it reduces many kinds of explanation to one under H deo ( Swinburne 1968a ).

This model of prior probabilities explains why evolutionary thinkers writing before 1859 did not accept the design argument. Even in Paley ’ s time, not everyone was led to accept H deo , although natural theological arguments were widespread and widely accepted. Early evolutionists such as Erasmus Darwin sought to describe biological forces that could assemble complexity in the same way that Isaac Newton had done for mechanics. These authors had a strong commitment to a materialist worldview, lead- ing them to adopt the view that H deo was unlikely even though they did not have a compelling causal explanation for the apparent design. The importance of prior probabilities may explain why the design argument, despite its intuitive appeal, failed to convince them. It may also explain the considerable cross-cultural variability in the acceptance of evolutionary theory. In one survey ( Miller, Scott, and Okamoto 2006 ), about one-third of Americans believed evolutionary theory was absolutely false, whereas only 7 percent of Danes thought so. These differences correlate well with levels of religiosity in these countries, with Denmark having a lower regular church attendance rate (3 percent) than the United States (44 percent) ( Ver- weij, Ester, and Nauta 1997 ).

For theists, design in nature provides compelling evidence for the exis- tence of a creator. Take as an illustration the Thomistic tradition, which emphasizes the role of understanding and knowledge in belief. In this view, a successful natural theology would start out from self-evident premises, proceed by valid arguments, and reach the conclusion that there is a per- son such as God. As we have seen, humans are prone to discern design and teleology in nature. Within the epistemic context of H deo , the perceived design in nature that is a universal feature of human cognition can be taken as a self-evident premise from which the existence of a creator can be argued. It is not a stand-alone argument that can convince those who do not believe in God, especially not since plausible naturalistic explanations have become available.

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

Teleology, the Design Stance, and the Argument from Design 79

Is There Still a Place for the Design Argument?

Undeniably, the power of the design argument has been seriously weak- ened since Darwin and Wallace came up with natural selection as a natural- istic explanation of design. Given that the combination of random events and selective retention can explain most of the apparent design around us, can natural theologians still reasonably invoke the design argument?

Misrepresenting or altogether ignoring natural selection is the strategy most commonly adopted by proponents of the recent Intelligent Design (ID) movement. While it spans many fields (e.g., geology, biochemistry), its chief aim is to challenge Darwinian evolutionary theory. ID advocates that science points to irreducible complexity in nature that cannot be explained by naturalistic mechanisms ( Behe 1996 ) and hence that such complex structures were intelligently designed. It differs from natural the- ology in that it relies exclusively on scientific, empirical research to reach this conclusion and does not equate, or at least not manifestly so, the intel- ligent designer with God. According to its proponents, ID does not require prior belief in God, as anyone who objectively looks at nature will realize that “ a designer of remarkable talents is responsible for the physical world ” ( Dembski 1999 , 111). ID nevertheless has a religious agenda: it wants to make religious belief a palatable, reasonable option for modern westerners; it aims to influence school curricula and ultimately wants to undermine the secularism and naturalism of modern science and society ( Johnson 2000 ).

ID is not a very desirable position to take. Theologically, it employs a narrow notion of faith as the opposite of reason ( Mongrain 2011 ). By insisting that religious belief should be exclusively based on reason, it goes much further than natural theology, which regards reason and faith as complementary, and natural theological arguments as faith seeking under- standing. Ever since Augustine, theology (including natural theology) has been conceptualized as a form of seeking to understand what one already believes (we come back to this in the last section of chapter 5). To put it with Anselm of Canterbury ([11th c.] 2000, 93): “ But I yearn to understand some measure of Your truth, which my heart believes and loves. For I do not seek to understand in order to believe, but I believe in order to under- stand. For I believe even this: that unless I believe, I shall not understand. ” This is the classical understanding of what Christian theology does, and ID is in diametrical opposition to this.

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

80 Chapter 4

As a scientific research program, ID has not been able to establish itself as a contender to evolutionary theory; unlike the latter, it has not pro- duced substantive results. Its key concepts, design and intelligent , are not sufficiently operationalized. For instance, the designer remains elusive, as ID has scrapped all explicit references to theism. 2 However, without a spe- cific referent (God), ID remains stuck in analogies to human design ( Sarkar 2011 ). Dembski (1999) defines design in terms of complex specified infor- mation . Although formally sophisticated, this notion only operationalizes complex as a measure of improbability, but does not operationalize specified nor information . As we have shown, recognizing design requires background assumptions about the nature and intentions of designers. Without such background assumptions, the notion of design in ID will remain vacuous. There is no serious attempt to define intelligent either, except by vague analogy to human intelligence. For instance, it is unclear why irreducible complexity, a term employed by Behe (1996) to denote the fragility and interdependence of components in biological systems, should point to intelligent design. If anything, should an intelligent designer not build in some redundancy to make a system more robust and less prone to break- down because of failure in a single component ( Miller 2007 , chapter 4)? Human designers, for example, think of fire exits and emergency lighting in buildings, spare parachutes and multiple motors in airplanes. As Sahotra Sarkar (2011 , 300) remarks, “ All we have is a metaphor that works — to the extent it does — because we all seem to share cultural ideas about what prop- erties a supposedly intelligent theistic designer should have. We have no positive specification of intelligence. ”

There is also a pragmatic reason not to align oneself with ID. Its aim to help Christianity gain intellectual respectability through reason failed, as its proponents are neither taken seriously in the scientific community nor in mainstream natural theology. Augustine already warned against taking positions that invite ridicule and decrease respectability.

Now it is quite disgraceful and disastrous, something to be on one ’ s guard against at all costs, that they [non-Christians] should ever hear Christians spouting what they claim our Christian literature has to say on these topics, and talking such nonsense that they can scarcely contain their laughter when they see them to be toto caelo , as the saying goes, wide of the mark. And what is so vexing is not that misguided people should be laughed at, as that our authors should be assumed by outsiders to have held such views and, to the great detriment of those about whose salvation we

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

Teleology, the Design Stance, and the Argument from Design 81

are so concerned, should be written off and consigned to the waste paper basket as so many ignoramuses. (Augustine [416] 2002, 186 – 187)

A more productive way for theologians and scientists to look at the argu- ment from design is to treat it not as a scientific principle (as ID does) but as a metaphysical principle. Within this perspective, there are at least two cases in which a design position may still be defensible. A first case is presented by a position that endorses evolutionary biology but argues that God inter- venes occasionally to fashion structures that could not have arisen through natural selection. One endorses intelligent design as a philosophical posi- tion but not as a scientific research program that conceptualizes evolution and design as competing scientific explanations. Theologically, it follows a distinction that is commonly made between God ’ s general actions (which pertain to the universe as a whole and can be seen in the laws that govern physical, chemical, and biological processes) and special actions (which lie beyond normal physical processes). Whereas natural selection and other evolutionary processes belong to the former category, intervention in these or occasional design belongs to the latter.

This strategy is followed by Johnson and Potter (2005) , who propose that human natural language may be the product of purposive creation. Adaptationist explanations require a plausible reason for why an adapta- tion evolved. Adaptations evolve in response to specific selective pressures and enhance the survival and reproduction of their bearers. For language, there are as yet no persuasive adaptationist explanations. We do not know what language is an adaptation for or how and when it evolved. Despite the proliferation of adaptationist accounts of the origin of language, such as social grooming, technological intelligence, cooperative hunting, and sex- ual selection, none of these hypotheses has been able to substantiate itself into a generally accepted theory. This leads Johnson and Potter (2005) to infer to the best explanation that purposive design brought language into being. Their position is distinct from ID in that they explicitly endorse evo- lutionary theory as the best explanation for complexity in the living world in general terms. One problem with this line of reasoning is that it relies on a God of the gaps argument. While God of the gaps arguments are not necessarily fallacious (as we will see in chapters 6 and 7), they have proven to be notably vulnerable in the past: once a convincing scientific account of the evolution of language is outlined (which is not unlikely, given the advances in our understanding of the genetic and developmental bases of

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

82 Chapter 4

language, its cultural evolution, and its emergence in human evolution), Johnson and Potter ’ s argument may dissolve.

A second, perhaps stronger case is found in scientists and theologians who regard design and evolution as complementary rather than mutu- ally exclusive explanatory frameworks. It avoids the vulnerability asso- ciated with God of the gaps arguments but relies on assumptions about how a divine designer would create the world. If the analogy with watches is maintained, such theistic evolutionary accounts could point out that watchmakers do not build watches from scratch but rely on the gradually accumulated innovations in timekeeping technology, which can be traced back to sundials and water clocks, to the introduction of the spring, to the modern digital watch. In fact, upon close scrutiny, very few inventions appear de novo ; most are the result of a gradual and cumulative retention of favorable variations ( Basalla 1988 ). For instance, the streamlined design of Polynesian canoes, which is close to optimum, can be traced through archeological and historical data as the gradual and unconscious retention of favorable variations, with the perilous ocean as the selecting agent ( Rog- ers and Ehrlich 2008 ).

Since evolutionary processes play a pervasive role in human design, it is not an ad hoc move for the proponent of the design argument to regard evolution as a basic design feature of the created universe. The botanist Asa Gray (1888 , 57) defended the view that natural selection is an “ a-fortiori extension to the supposed case of a watch which sometimes produces bet- ter watches, and contrivances adapted to successive conditions, and so at length turns out a chronometer, a town clock, or a series of organisms of the same type. ” Theodosius Dobzhansky (1973 , 127), one of the founding fathers of the modern synthesis, wrote, “ The organic diversity becomes, however, reasonable and understandable if the Creator has created the liv- ing world not by caprice but by evolution propelled by natural selection. … Evolution is God ’ s, or Nature ’ s, method of Creation. ” The cell biologist Kenneth Miller argues that God has initiated natural selection and other natural evolutionary processes as an indirect way to create complexity and design. The undetermined nature of evolution through natural selec- tion and other natural processes enabled the evolution of truly free, truly independent beings ( Miller 2007 , chapters 7 and 8). Theistic evolutionary accounts like these can invoke evolution, with its enormous creative poten- tial, as a plausible mechanism for divine design.

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

Teleology, the Design Stance, and the Argument from Design 83

Interestingly, both positions, materialism and theistic evolution, already existed in the earliest stages of evolutionary theory. Whereas Thomas Hux- ley did not provide room for God in his explanatory framework, Alfred Wallace and Asa Gray were theists who treated divine action as comple- mentary with a scientific worldview, not as a competitor. Wallace (1870 , 343), while continuing to endorse natural selection as the chief principle guiding the evolution of plants and animals, invoked intelligent design for the human mind: “ The brain of prehistoric and of savage man seems to me to prove the existence of some power, distinct from that which has guided the development of the lower animals through their ever-varying forms of being. ” Wallace ’ s theism, rather than a volte-face , was an integral part of his evolutionary thinking ( Fichman 2001 ). Today, both schools of thought continue to exist side by side, with Richard Dawkins and Daniel Dennett as examples of strict materialists, and Kenneth Miller and Francis Collins as proponents of theistic evolution.

While theistic evolution may be an attractive option for the theist, it requires additional assumptions about the intentions of the designer. For example, Miller (2007) argues that God used a stochastic process because he wanted to create free beings that were genuinely distinct from their creator. The problem with this method of creation is that it seems to implicate God in an unacceptable amount of animal suffering that is a direct result of the winnowing process of natural selection. Animal suffering caused by preda- tion, competition, and starvation is a byproduct of natural selection that seems difficult to reconcile with a benevolent creator (but see Southgate 2008 for a theological exploration and attempt to meet this challenge). While death and suffering were traditionally explained by appeal to a his- torical Fall of humankind, the current best scientific evidence is that these features of the world predate the evolution of humanity (see also De Smedt and De Cruz 2014 ). The atheist can reply that features of evolution, such as its generation of natural evil, undermine the design hypothesis. Some Christian theists have in turn responded to this by presenting more open, process-like theological conceptions, where God is not an aloof, external orchestrator, but became incarnate to suffer along with his creation. For example, John Polkinghorne writes:

In the lonely figure hanging in the darkness and dereliction of Calvary the Christian believes that he sees God opening his arms to embrace the bitterness of the strange world he has made. The God revealed in the vulnerability of the incarnation and the

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .

84 Chapter 4

vulnerability of creation are one. He is the crucified God, whose paradoxical power is perfected in weakness, whose self-chosen symbol is the King reigning from the gallows. ( Polkinghorne 1989 , 68)

Discussions on the plausibility of theistic evolution indicate that the- ism and metaphysical naturalism cannot be purely evaluated by empirical evidence but rely on the prior probabilities their defenders assign to H mat and H deo . Indeed, the theist can argue that natural selection is not meta- physically incompatible with divine design, even if the mutations that are generated appear to be entirely stochastic from our perspective. The non- theist can argue that metaphysical naturalism is the most parsimonious explanation for natural selection, as it can explain features like suboptimal design and natural evil. A productive way forward for dialogue on design would therefore not be to focus exclusively on scientific evidence and its compatibility with the design hypothesis or naturalism, as proponents of ID or some atheist authors do. Rather, it would be to consider whether divine design through evolution provides an overall plausible worldview, or whether metaphysical naturalism is a more compelling framework.

Summary

This chapter has examined the cognitive basis of intuitions that drive argu- ments from design. These intuitions can be traced back to evolved infer- ence mechanisms: the design stance, which leads us to treat complex and purposive structures as the product of design, and intuitive teleology, the propensity to discern purpose in nature. These inference mechanisms are universal, although they can be masked by formal education or strength- ened by religious upbringing.

The step from design to designer is perhaps more explicit and relies on an argument to the best explanation. The plausibility of this argument relies on the prior probability one places on the existence of God. By making these differences in prior probability more explicit, theists (natural theolo- gians, biologists, and philosophers) and materialist scientists and philoso- phers have a rational basis for disagreement. The reason why some find the design argument compelling and others do not lies not in any intrinsic differences in assessing design in nature but rather in the prior probability they assign to complexity being produced by chance events or by a creator.

C op

yr ig

ht ©

2 01

4. M

IT P

re ss

. A ll

rig ht

s re

se rv

ed .