Although rarely used today by most philosophers, certain conservative theologians and, of course, creationists continue to utilize the "argument from design" as a cornerstone of their personal world views (Geisler, 1983; Gordon, 1984; Jnana Dasa, n.d.; Sadaputa Dasa, 1984; Watchtower Bible and Tract Society of New York, 1967). Strictly from the logical perspective, such arguments are often unsatisfying to those who have not already committed themselves, consciously or otherwise, to an a priori belief in the intelligent design of the universe. But anti-creationist contentions that such arguments are made from an incomplete, and therefore invalid, analogy (Edwords, 1983; Thwaites, 1983) have rarely been developed in any real detail.
Stephen Jay Gould has contributed major steps in this direction by pointing out that many adaptations in living organisms have apparently been developed by the opportunistic utilization of previously existing partsthat is, tinkering, as opposed to real "design" (1980). The fact that living organisms would be limited in their adaptational options by prior phylogenetic history is a critical point in this analysis. Living organisms must evolve by modification of existing organs. Evolutionary innovation cannot come about by the sudden appearance of new, functioning organ systems. Instead, they result from the "exaptation" of preexisting structures for new functions. Thus, a wrist bone can become the Panda's thumb and ordinary flower parts can become the elaborate insect-luring mechanisms of orchids.
But Gould's analysis only goes so far. It is not enough to demonstrate that bricolage has apparently occurred; for once a design is demonstrated as functionally adequate in its new role, the determination as to whether or not it constitutes "good" design is largely subjective. It comes as no surprise that some creationists have already responded to this argument with the contention that, in their opinion, the Panda's thumb is not a clumsy, "jury-rigged" affair but, in fact, is an interesting, effective, and original design (Gordon, 1984).
The purpose of this article, then, is to take a look at the same issue from a slightly different perspective: instead of looking at organisms to evaluate the "goodness" of their design (by whatever subjective criteria we may choose to impose), we shall examine how designed machines as a group differ objectively from living organisms, regardless of the subjective quality of those designs. It is the contention of this article that, from an engineering perspective, living organisms fail to demonstrate the clear signs of artifice that characterize "created kinds."
The Process of Intelligent Design
The "created kinds" of modern technology have indeed evolved but in a manner quite different from living organisms. Where living organisms are constrained in their development by their evolutionary histories, machines are constrained only by the level of technology available to their designers. Design selection is based upon the unique design criteria of each individual creative act. Previous designs can be used, modified, or completely abandoned. Changes over time reflect technological advances, and such advances often manifest themselves in distinct ways.
1. The Quantum Leap. The achievement of a new technology will show up in machines with a previous design history as completely new systems or subsystems that only minimally affect the other independent systems in the machine. For example, when the technology of turbine engines became available, aeronautic engineers began designing aircraft using jet engines, while most of the other aircraft systems were unaffected. Control systems, building materials, even most of the aerodynamic design of our first jets were almost unchanged from their cylinder-propeller-driven predecessors. These other systems did change later but only with the development of other new technologies.
The point is not that this type of mosaic evolution is unlike that of living organisms. One need only look at Archaeopteryx to see that this is actually quite like a living organism. The point is that the design of a turbine engine is so radically different from that of a cylinder engine that one cannot possibly construct a "Darwinian history" that could evolve one from the other. Unlike the fusing of two clavicles to form a wishbone or the fraying of a keeled scale to form a feather, the turbine engine is a completely original design with complex parts and subsystems that have no homologs and often no analogs in the other design.
Innovations in design are often unrestrained by whatever designs existed prior to them. They are therefore often revolutionary changes. Innovations in living organisms, however, do not show this type of wholesale replacement of systems. The very fact that it is possible to construct plausible Darwinian histories for living organisms, while at the same time it is impossible to do the same for our own "created kinds," is a crucial point of comparison.
2. Contagious Technology. Living systems evolving from different directions to fill the same ecological niche often develop strikingly similar adaptations. But again, constrained by their evolutionary histories, these adaptations are formed out of the parts available, and different parts are used for similar purposes. Dolphins do look remarkably like large fish and even more like icthyosaurs. But no competent zoologist would ever confuse the three. Past the most superficial level, the differences are dramatic.
Machines, constrained only by the level of technology of their creators, need not be so dissimilar when designed for similar functions. Grumman, Northrop, and McDonnel Douglas may be designing three individual air superiority fighters, but, if the specific design criteria so dictate, they can use the identical Pratt and Whitney engines for all three aircraft. This is not convergent evolution, but it is a fact of contemporary design.
And once turbine engines became available for aircraft, they need not be limited to the aircraft "clade." Engineers have placed turbines into boats, automobiles, motorcyles, and the M-1 main battle tank. The organic equivalent to this would be for feathers, once evolved in birds, also to appear suddenly on bats and fl, ying squirrels or for whales, dolphins, and icthyosaurs to have gills. Needless to say, we do not see this.
An innovation in a "created kind" is contagious between "created kinds." Physical traits (systems, technologies) are not confined within specific clades by the limitations of genetic transfer through phylogenetic descent.
3. Extrafunctional Homology. Design engineers do tend to specialize and develop their own unique design styles. A bridge designer will often use similar designs for different bridges. A common designer should be expected to use similar designs for similar functional purposes. But what about different functional purposes?
What engineer in his or her right mind would use the same design only slightly modified to build an aircraft and a submarine? The different functions and design criteria mandate drastically different designs for these purposes. An aircraft requires a specially designed hull to hold air pressure in; a submarine requires an equally special hull to hold water pressure out. An aircraft will normally utilize internal combustion engines, even at high altitude; a submerged submarine requires electric or nuclear power. An aircraft requires airfoils for lift; a submarine requires ballast tanks for buoyancy. The list of profound differences could go on for pages.
But whales and bats are both air-breathing, warm-blooded, milk-giving mammals. Their design differences are quantitative, not qualitative. This cross clade similarity of design is completely unheard of in "created kinds," whether the design team consists of fifty engineers or only one.
If living organisms did in fact display the characteristics of "created kinds" that we witness around us, then the premiere prediction of evolutionary biology would be easily falsified. That prediction is the existence of a hierarchical, nested pattern of similarities in living organisms with which to construct a taxonomic system. If traits (technologies) were distributed among living things in the same way that they are distributed among machines, there would simply be no perceivable basis for a pattern that is so tangible that it was first perceived and formalized by creationists.
How would one go about establishing a Linnean taxonomy for machines? One could choose to sort machines by functionfor example, "flying machines." But at the organic level, that would group bats with the birds (a mistake which can, incidentally, be found in Leviticus). One could attempt to sort by the organic criteria of homology versus analogy, but then where would one put Evel Kneival's rocket motorcycle? In the Harley Davidson clade or the Saturn V clade? It rapidly becomes evident that the selection of taxonomic criteria for machines would be entirely arbitrary and that drastically different "cladograms" would result depending upon which traits were being used to determine taxonomic affinity.
Even with these problems, at least one creationist, David Johannsen, has made a token effort at suggesting such a taxonomy for machines (1984). It was largely the spectacular failure of his system to result in the nested, hierarchical pattern he was explicitly attempting to construct that inspired this article.
Furthermore, even the limited evolutionary pattern given by advancing technology changing through time would be missing from the creationist scenario. The genuine evolution of humankind's technology has at least resulted in machines that increase in complexity over time. From a strictly directional (as opposed to anatomical) perspective, aircraft have evolved since Orville and Wilbur Wright took their first flights at Kittyhawkbut not in a manner that would allow construction of a phylogenetic tree.
But creationists don't even have this much of an explanation going for them. Creationists maintain that the creative act took place essentially instantaneously by a creator with no technological limitations. What possible reason could there be for God to impose an evolutionary pattern of change over time on living organisms when he allegedly had the unlimited technology available to make the appropriate design decisions at the first moment of design?
As already pointed out, if living things were created, there would simply be no engineering rationale to not create feathered, milk-giving fish and thus completely frustrate any attempts to establish taxa higher than the species. But instead we find penguins (firmly members of the class Aves, yet, a marine niche) and seals (unambiguously mammalian, yet also in a marine niche).
A common design stretched so far across different functional requirements speaks unambiguously of a severe lack of options, not an omnipotent designer. It speaks of a near total lack of originality, not omniscience. It speaks of opportunism, not intelligent design.
When observed from the point of view of engineering, and particularly from the perspective of the design process, it becomes quite clear that, if living organisms are the result of intelligent design, it is a type of design completely alien to engineering as we know it. The design decisions creationists attribute to God are not the same type of design decisions that are made by human engineersthe only intelligent designers with whom we are reasonably familiar.
And therein lies a fundamental flaw in the argument for design. The entire analogy rests upon the comparison of living organisms to the products of human design. And as we can clearly see, such a comparison presents us with glaring inconsistencies that argue against the validity of that analogy. If the creationist chooses to take the position that God is not limited to design in the same manner as humans, fine. But doing so forfeits the entire argument. If God has not designed the natural world in the same manner that humans design, then the analogy fails even before we get to the level of discussing the similarities between the universe and a watch found on a desert beach.
One arrives at this conclusion without even bothering to ask why the mammalian retina is inside-out or why I had to get those extra teeth extracted. At this point it may be interesting to discuss "good" versus "bad" design, but it is no longer necessary. Even if I were to grant (and I will not so grant) that all living designs are good, they still appear to be the best that nature could do with what it had on handnot forward thinking, intelligent design.
Several of this nation's finest engineering schools have annual competitions for which students are given small packages of machine parts and construction materials. Using only these parts the students are to design, build, and demonstrate machines that can accomplish specific tasks, such as: traverse a three-foot table, capture a ping-pong ball, and return it to home base while simultaneously defending from someone else's machine trying to capture the same ping-pong ball from the other end of the table. Although these are talented young engineers, they are "historically constrained" in their design options by the parts with which they are provided. The results invariably are that almost all of the competitors come up with variations on a very few basic designs, none of them as effective as they could have been in an unconstrained design environment.
The competition is invariably won not by some novel, original design but by the best use of a basic design that simply did not work as well for several of the other participants. The competition and the results are an analog of Darwinian natural selection at its purest.