The Scholarly Focus of Intelligent Design

Intelligent design proponents posit that the universe, or at least components of it, have been designed by an "intelligence." They also claim that they can empirically distinguish intelligent design from design produced through natural processes (e.g., natural selection). This is done through the application of two complementary ideas, one promoted by a biochemist and the other by a philosopher-mathematician.

Irreducible Complexity. The biochemist Michael Behe contends that intelligence is required to produce irreducibly complex cellular structures (ones that couldn't function if a single part were removed) because such structures could not have been produced by the incremental additions of natural selection (Behe 1996).

Critics of Behe have pointed out that it is not clear that irreducibly complex structures actually exist—except perhaps by definition. Critics have also argued that the examples Behe gives of irreducibly complex structures can often be reduced and still be functional. Behe commonly uses a mousetrap as his example of an irreducibly complex structure, claiming that if any one of the five basic parts of a mousetrap (platform, hammer, spring, catch, and hold-down bar) is removed, it can no longer catch mice. Scientists gleefully set about producing four-part, three-part, two-part, and even one-part mousetraps to demonstrate the reducibility of Behe's prime example of an irreducibly complex structure.

Similarly, supposedly irreducibly complex biochemical structures such as a bacteria's flagellum can function with fewer parts than Behe originally claimed in Darwin's Black Box. Ultimately, of course, it is possible to reduce a structure to so few parts that, indeed, removal of any one part will make the structure cease functioning. More important than whether irreducibly complex structures actually occur other than by

Figure 6.1

An Irreducibly Complex Mousetrap. A mousetrap has five parts. Behe contends that all five pieces need to be assembled or it is impossible to catch mice, hence the mousetrap is irreducibly complex. Courtesy of Sarina Bromberg.

Figure 6.1

An Irreducibly Complex Mousetrap. A mousetrap has five parts. Behe contends that all five pieces need to be assembled or it is impossible to catch mice, hence the mousetrap is irreducibly complex. Courtesy of Sarina Bromberg.

definition, however, is the critical question of whether they can be produced by natural mechanisms.

Behe answers no, claiming that natural selection, the main mechanism of evolutionary change, is inadequate to the task. He views natural selection as assembling a complex structure by stringing together components one at a time, with each addition requiring a selective advantage. Behe's view is that for a structure like the bacterial flagellum, consisting of more than forty proteins and enzymes, it is extraordinarily unlikely that so many elements could be assembled—by chance—one by one, and even more unlikely that there would be selective advantage to each addition. This piece-by-piece assemblage of the flagellum, one enzyme at a time, one after another, is also envisioned by William Dembski (2001), who claims that the probability against this occurring is astronomical; a bacterial flagellum, Dembski and Behe agree, cannot be produced through natural causes.

Critics have noted that Behe presents an incomplete picture of how natural selection operates: it is not the case that components of a complex structure must be added one after another, piece by piece, like stringing beads. It is clear from the study of components of a cell that a great deal of borrowing and swapping of bits and pieces takes place: each cellular structure is not composed of unique proteins and enzymes, or even of wholly unique combinations of proteins, and this is also true of the proteins composing the bacterial flagellum. In a recent analysis of the structure and origin of the flagellum, the authors note, "Three modular molecular devices are at the heart of the bacterial flagellum: the rotorstator that powers flagellar rotation, the chemotaxis apparatus that mediates changes in the direction of motion and the T3SS that mediates export of the axial components of the flagellum. In each module, the apparatus is fashioned from recycled parts that occur elsewhere in nature" (Pallen and Matzke 2006; emphasis added).

The cross-linking proteins of flagella, for example, have other functions elsewhere in the cell. Somewhat fewer than half of the proteins found in a bacteria's flagellum are the same as or very similar to those found among other bacteria in a structure called the type-III secretory apparatus, which performs some of the functions of a flagellum. An adaptive advantage for a structural element may exist and cause it to be selected for—but for a different purpose and perhaps in a different cell component than that of the final, supposedly irreducibly complex structure under discussion. Natural selection thus can produce complex structures without having to separately string protein after protein together, in which each addition requires a separate action of natural selection. Behe's critics thus have argued that some of the components of an irreducibly complex structure could be assembled separately for some purposes and then combined for other functions. This is not through a random or chance process: natural selection is intimately involved during all stages.

There is another way in which natural selection can be more flexible than Behe appears to allow. So-called irreducibly complex structures may indeed have been assembled piece by piece, but the pathway of assembly may not be obvious because of a process evolutionary biologists have called scaffolding.

Consider an arch made of stone. The keystone, the stone at the top of the arch, must be in place or the arch will fall down; an arch is an irreducibly complex structure. To build it, stonemasons will often build a scaffolding to support the sides of the arch as they build toward the center. When the keystone is laid and the arch is thereafter stable, the scaffolding is removed, leaving the irreducibly complex arch.

An irreducibly complex biochemical or molecular structure may be built in a similar way, in the sense that at an earlier time in the history of the structure, there might have been components that supported the function of the structure, much as a scaffolding supports the sides of an arch until the keystone is in place. These supporting biochemical components may be made redundant by the addition of more efficient components, much as the arch's scaffolding is made redundant for holding up the sides of the arch once the keystone is laid. The now-superfluous components can be removed by natural selection. Without knowing the entire history of the structure, it might seem that all the parts of the structure appeared all at once, fully formed and functional in their final configuration, with no history of earlier, simpler structural predecessors. But just as an arch attains irreducible complexity only at the end of construction, so too do these supposedly irreducibly complex biochemical structures: they actually had a history, though the exact events may not yet have been traced. Evolutionary biologists have proposed a number of similar ways in which a complex structure could be formed (Pond 2006).

But even if natural selection were unable to explain the construction of irreducibly complex structures, does this mean that we must now infer that intelligence is required to produce such structures? Only if there are no other natural causes—known or unknown—that could produce such a structure. Given our current knowledge of the mechanisms of evolution, there is no reason natural selection cannot explain the assembly of an irreducibly complex structure—but it is also the case that a future researcher might come up with an additional mechanism or mechanisms that can explain irreducibly complex structures by some other natural process.

Some scientists have described Behe's approach as an "argument from ignorance" (Blackstone 1997: 446) because the intelligent creator is used as an explanation when a natural explanation is lacking. This is reminiscent of the God-of-the-gaps argument, in which God's direct action is called on to explain something that science has not yet explained. Both theologians and scientists reject God-of-the-gaps arguments. To scientists, using God to explain natural phenomena of any kind violates the practice ofmethodological naturalism, in which scientific explanations are limited only to natural causes. To theologians, the God-of-the-gaps approach creates theological problems of the irrelevance or diminution of God when natural explanations for natural events ultimately replace the direct hand of God. Intelligent design proponents, however, claim that the issue is not current ignorance of a discoverable natural cause but the impossibility of a natural cause.

Behe's idea of irreducible complexity was anticipated in creation science; much as in Paley's conception, creation science proponents hold that structures too complex to have occurred "by chance" require special creation (Scott and Matzke 2007). Behe, following ID convention, doesn't mention God directly, but the logical consequence of the irreducible complexity argument is that irreducibly complex structures—unable to be produced by natural causes—are evidence for God's direct action. As such, ID verges on being a variety of progressive creationism in which God intervenes at intervals to create irreducibly complex structures like DNA, bacterial flagellum, the blood-clotting cascade, and so on. Although many ID proponents find the progressive creationist position attractive, ID is not necessarily wedded to the progressive creation position. Some of its proponents suggest that design could have been prearranged (or "front-loaded"). In Darwin's Black Box, Behe suggests that perhaps all the irreducibly complex structures of all living things were somehow present in the first living cell, and then appeared through time as various organisms evolved. "Suppose that nearly four billion years ago the designer made the first cell, already containing all of the irreducibly complex biochemical systems discussed here and many others. (One can postulate that the designs for systems that were to be used later, such as blood clotting, were present but not 'turned on.' In present-day organisms plenty of genes are turned off for a while, sometimes for generations, to be turned on at a later time.)" (Behe 1996: 228).

As noted by the cell biologist Kenneth Miller, such an ubercell would somehow have to avoid the mutational drift of genes controlling such structures for billions of years until it was "time" for such structures to appear (Miller 1996: 40). The probability of genes for, say, the bacterial flagellum remaining intact for so long violates much of what we know about the behavior of genes in the absence of natural selection. Such genes tend to accumulate mutations that make the gene nonfunctional.

William Dembski has expanded this concept and proposes that God might have front-loaded everything in the universe in the Big Bang—all the irreducibly complex structures are merely unfolding like so many homunculi as time passes (Dembski 2001). In this view, God would not be progressively creating but would have acted only once.

Complex Specified Information: The Design Inference. Dembski's design inference takes a probability theory approach to distinguish those phenomena in nature that are designed by intelligence from those that are the result of natural causes or chance. Although arguments against evolution based on probability have long been a mainstay in creation science (Gish 1976; Morris 1974; Perloff 1999), Dembski's design inference is at least superficially more impressive, couched as it is in a mathematical idiom. In proposing an explanatory filter decision tree, Dembski contends that there are three ways to explain phenomena on the basis of their frequency of occurrence (see Figure 6.2).

Things that occur commonly or with high predictability can be attributed to the unfolding of natural laws. That the moon goes through phases every month can be explained by the passage of the moon around Earth and the changing angle between the moon and sun, as we see here on Earth; it is not necessary to attribute design to this phenomenon. Phenomena that occur at intermediate probability can be attributed to chance—even very low-probability events will occur some of the time, just by chance alone. But some kinds of low-probability phenomena—Dembski refers to them as specified low-probability events—that are not due to law or to chance compose the class of phenomena that must be attributed to intelligent design. Dembski proposes that complex specified information distinguishes intelligently designed phenomena.

Specification is a sort of side information that we add about a phenomenon or event. Consider the explanation for finding an arrow in a bull's-eye. If we see an arrow in a bull's-eye, we might consider that the archer got lucky, but if we see ten arrows in ten bull's-eyes, we attribute this to an archer with a high level of skill. On the other hand, if we knew that the archer shot the arrows first, and then drew the targets around them, we would not attribute the perfect shots to skill. Knowing that the targets were present before the arrows is a specification or additional information that allows us to attribute the arrows in the bull's-eyes to design rather than chance (or cheating).

Dembski's filter (Figure 6.2) allows the assignment of the causes of some phenomena to natural law, chance, or design, using the combination of probability and specification. The natural law filter stops high-probability events; medium or low, unspecified-probability events are attributed to chance, and only low, specified-probability events are attributed to intelligent design. Dembski's filter is therefore an elimination algorithm: something is explained by design when it is not explained by law or chance. But this approach allows false positives where something is attributed to design because of missing or unknown information at the first, natural law level.

For example, let's say that while walking through the forest, we come upon a circle of toadstools that has sprung up overnight.1 The ring wasn't there yesterday, and in a few days it will largely be in tatters: fungi are fragile things. If this walk were taking place in the ninth century in Europe, as peasants we would recognize the circle of toadstools as a fairy ring, a location indicating the presence of fairies the night before. Applying Dembski's filter, we would conclude that the sudden and random appearance of the circle, and of course its symmetrical shape, certainly were not the result of natural processes: rings of toadstools crop up with no warning, unlike the phases of the moon. So in the year 800, a fairy ring would pass through the first (regularity) filter. It would also not be attributed to chance, as the likelihood of a fairy ring occurring at a given place is very improbable. However, this low-probability event has a specification, its circular shape. Therefore, following Dembski's filter, we would attribute the appearance of a fairy ring to ID; European peasants of the year 800 knew that fairy rings were the remains of midnight revels held by tiny fairies in the woods.

Perhaps because no one ever found tiny beer cans next to the toadstools, eventually a natural explanation was found for fairy rings: they are the result of one of the ways toadstools reproduce themselves. These fungi send out underground, threadlike mycelia from a central point, and when circumstances of moisture and temperature

Figure 6.2

Dembski's Explanatory Filter. William Demb-ski proposes that design can be detected by eliminating regularity and chance. Small probability events that are specified are the result of design, according to Dembski (1998: 37). Courtesy of Alan Gishlick.

start

chance

are suitable—and when two mycelia mate—toadstools form aboveground. These toadstools produce spores that are carried by the wind and eventually land and start growing new fungi. Because the mycelia radiate from a center, circles of toadstools are formed. With additional information, it can be seen that fairy rings actually are not improbable, though many variables are involved in their appearance, and they actually do not appear randomly but in specific environments. In the twenty-first century, we recognize that fairy rings have a natural explanation; in the ninth century, the circles were explained by design. Because Dembski's filter depends on the extent of scientific knowledge of the time, it thus fails to be a reliable predictor of design by intelligence (Wilkins and Elsberry 2001).

The Problem of Natural Intelligent Cause. As shown in Table 6.1, ID proponents contend that design can be produced both by natural causes (e.g., natural selection has some limited ability to shape organisms to meet some environmental pressures) and by intelligent causes. An intelligently designed phenomenon could be the product of transcendent intelligence such as a creator God, or it could be the product of material

Table 6.1

Natural and Intelligent Causes: The Intelligent Design View

Table 6.1

Natural and Intelligent Causes: The Intelligent Design View

Natural causes

Intelligent causes

Natural selection

Transcendent agent (God)

Natural (material) agents: Humans,

Higher primates, Extraterrestrials

agents such as extremely intelligent extraterrestrials—an argument first made, in fact, in the original ID book The Mystery of Life's Origin.

Unfortunately, the dichotomy between natural and intelligent is artificial, because some of the agents on the intelligent side are actually natural. To a scientist, anything that is the result of matter, energy, and their interactions is a natural phenomenon, whether nonliving phenomena such as stars and rocks or living phenomena such as plants and animals. Material agents such as humans, higher primates, or extraterrestrials (if such beings exist) are therefore natural, as are their behaviors. No one disputes that the behavior of humans and animals can be studied and understood through the application of scientific principles; such behavior is the subject matter of physiology and psychology. We study bird or whale communication, for example, and attempt to explain it by using general theories about regularities of these behaviors. Behavior is the product of natural entities and is thus itself natural.

To answer whether the intelligent behavior of material beings can similarly be studied and explained requires a definition of intelligent behavior—which is not as easy as it seems. Psychologists define intelligence broadly, as having elements of problem solving and some degree of abstraction. But problem solving is also a broad category: bees solve the problem of communicating the location of nectar through their waggle dance, which surely is a complex behavior having elements of abstraction (the dance indicates the direction and distance from the hive to the food source), but is this behavior intelligent? Intelligent behavior is usually conceived of as having some element of choice involved rather than as the result of largely uncontrolled or genetically wired causes—yet clearly choice is a continuum. A bee may be largely hardwired to return to the hive when a source of pollen is discovered and to perform the waggle dance: her genes make it extremely likely that she will respond to a food source by returning directly to the hive. But what if, on the way home, the bee finds a larger source of nectar? Can she make a choice not to report on the original source but to bring back a message about the second one?

In other instances of behavior, choices clearly are made. A chimpanzee attains a high or low social status through complex interactions with many individuals over a long period of time. Such actions are not genetically hardwired like the waggle dance of the bee, and in fact even involve examples of conscious manipulation of other group members, including efforts to deceive one other (especially over food sources). When primatologists and psychologists study such intelligent behaviors, they attempt to explain them through theoretical principles—in other words, they study them scientifically. Certainly economists, psychologists, and political scientists also study the intelligent (and sometimes unintelligent!) behavior of human beings and attribute

Table 6.2

An Alternative View of Natural and Intelligent Causes

Natural causes Transcendent causes

Natural selection Transcendent agent (God)

Natural (material) agents: Humans, Higher primates, Extraterrestrials all or part of it to patterns. The intelligent behavior of material creatures is therefore natural and an appropriate subject for scientific investigation.

Therefore, all of the natural intelligent agents on the right side of Table 6.1 should be moved to the left side. If intelligence produced by material beings is moved to the natural side of the equation, as in Table 6.2, the ID dichotomy of natural and intelligent must be restructured. What ID proponents wish to label "intelligent" reduces to one agent: God. The natural-intelligent dichotomy is in reality a natural-transcendent dichotomy. This is not only more empirically accurate but also a more logically satisfying relationship: if there is a transcendent, omniscient, and omnipotent agent such as God, then such an agent by definition could not be explained by natural causes and more properly would form a dichotomy with natural cause.

Intelligent design supporters cannot accept this, however, because appealing to transcendent causes is, of course, a form of religion. They are well aware that the First Amendment does not allow the advocacy of creationism in public schools. Hence it better suits the ID movement to try to combine all forms of intelligent cause into one heterogeneous list—whether or not such a division is empirically or logically defensible.

Intelligent design supporters are hostile to methodological materialism and propose a new kind of science: theistic science. This is an alleged subclass of science concerned with those scientific problems that deal with origins (i.e., origins science), which are unrepeatable. Such phenomena as the origin of life and the evolution of living things (unspecified) constitute origins science. Although the majority of science may be performed in a methodologically materialistic fashion, explaining only with reference to natural causes, origins science allows (indeed, requires) the occasional intervention of intelligence—by which is meant the direct hand of God. Theistic science, then, is a proposal to radically change how we do science by abandoning methodological materialism in favor of allowing explanation by supernatural causes—and still call the process science. It is not a position that either philosophers of science or scientists have embraced (Pennock 1999). Theistic science leads to the second focus of the ID movement, an effort to promote a sectarian religious view.

Was this article helpful?

0 0

Post a comment