Evolution by Design

By Jonathan Wells

     By shifting the evolutionary paradigm from one that rejects design to one that accepts it,
     scientists could explain various observations that Darwinian theory has difficulty
     accounting for.

     Jonathan Wells holds doctorates in both biology (Berkeley) and theology (Yale).
     He is currently a postdoctoral research biologist in the Department of Molecular
     and Cell Biology, University of California, Berkeley, and a fellow of the Discovery
     Institute in Seattle.

     Adapted with permission from the International Conference on the Unity of the
     Sciences. The original of this paper was presented at the Twenty-first International
     Conference on the Unity of the Sciences, which met in Washington, D.C., in
     November 1997.

Before the twentieth century, most Western scientists believed that God created living things by design.
Belief in God was part of the very fabric of Western civilization; and by viewing the world through the
spectacles of faith, people saw it as God's handiwork. In the words of John Henry Newman, "I believe
in design because I believe in God; not in a God because I see design."

In the eighteenth and nineteenth centuries, however, some thinkers reversed the traditional logic to
argue from design to God's existence. William Paley wrote in Natural Theology (1802) that someone
crossing a heath and finding a watch would see that "its several parts are framed and put together for a
purpose" and would conclude that it had been designed by a watchmaker. Analogously, Paley argued,
one could conclude that living things are designed by God.

Darwin's exclusion of design

Charles Darwin was born into this intellectual environment in 1809. By the time his Origin of Species
was published in 1859, Darwin had become convinced that the design that Paley claimed to see in
living things was an illusion. According to Darwin, what appears to be design in living things can be
explained naturalistically as the result of random variations and natural selection. [In this paper,
"naturalism" and "naturalistic" refer to the philosophical doctrine that the physical universe is the whole
of reality and that ideas and the supernatural are human projections.]

Darwin argued that just as domestic livestock can be modified by selecting certain variants for
breeding, so wild species are modified by a "natural selection" due to competition for survival.
According to Darwin, the continuation of such "descent with modification" over millions of years
produced all living things from one or a few original organisms. He saw no room for design in this
process. When Harvard botanist Asa Gray proposed that God had designed the variations on which
natural selection operated, Darwin rejected the idea and concluded his 1868 Variation of Animals and
Plants Under Domestication with a refutation of design. According to Darwin, the products of
random variation and natural selection cannot be regarded as designed; and human beings, as the latest
in a long series of undesigned results, are the least designed of all.

Darwin's modern followers concur. In 1967, paleontologist George Gaylord Simpson wrote: "Man is the
result of a purposeless and natural process that did not have him in mind" (The Meaning of Evolution,
revised edition). In 1970, molecular biologist and Nobel laureate Jacques Monod announced that "the
mechanism of Darwinism is at last securely founded," and thus "man has to understand that he is a
mere accident" (quoted in H.F. Judson's The Eighth Day of Creation, 1979). And in 1986, zoologist
Richard Dawkins wrote a best-selling book titled The Blind Watchmaker: Why the Evidence of
Evolution Reveals a Universe Without Design.

But the "evidence" that Dawkins cites in The Blind Watchmaker consists almost entirely of computer
simulations. He argues that Darwinism would have to be true even if there were no evidence for it,
because short of postulating the existence of a deity (which Dawkins rejects), Darwin's theory of
"cumulative selection, by slow and gradual degrees, is ... the only workable explanation that has ever
been proposed, for the existence of life's complex design." In other words, what persuades Dawkins
that Darwinian evolution is true is not the evidence, but the fact that it is the only tenable naturalistic
explanation for the history of life. As he writes in the book's opening chapter, "Darwin made it possible
to be an intellectually fulfilled atheist."

Evolutionary biologists are virtually unanimous in their rejection of design, though some (such as
paleontologist Stephen Jay Gould) sharply disagree with Dawkins over the sufficiency of Darwin's
mechanism of gradual selection. Yet if one wishes to exclude design on scientific grounds, one must do
so on the basis of a demonstrated mechanism; mere descent with modification is not enough. This point
is unintentionally illustrated by biologist Tim Berra in Evolution and the Myth of Creationism (1990):

If you look at a 1953 Corvette and compare it to the latest model, only the most general resemblances
are evident, but if you compare a 1953 and a 1954 Corvette, side by side, then a 1954 and a 1955
model, and so on, the descent with modification is overwhelmingly obvious. This is what paleontologists
do with fossils, and the evidence is so solid and comprehensive that it cannot be denied by
reasonable people.

But the historical development of the Corvette, which Berra calls "descent with modification," is
undeniably due to construction according to preexisting plans--that is, design. Ironically, therefore, his
analogy shows that descent with modification is compatible with design.

Evidence has been accumulating for decades, however, that Darwin's mechanism fails to account for
major features of evolution. The fossil record (especially where it is most complete) lacks the
innumerable transitional forms that Darwin's theory predicts; artificial breeding (no matter how intense
or protracted) fails to produce the major modifications that his theory requires; and embryonic
development (as revealed by modern comparative embryology) is radically different from Darwinian
expectations. According to molecular biologist Michael Denton (Evolution: A Theory in Crisis, 1986),
not "one single empirical discovery or scientific advance since 1859" has validated Darwin's theory that
large-scale evolution is caused by natural selection acting on random variations.

Given the empirical anomalies, and the sharp disagreement over mechanism between Dawkins and
Gould, it is clear that the modern Darwinian denial of design rests on nonempirical grounds. It is no
longer an inference from evidence but an a priori assumption based on a commitment to naturalistic

Reintroducing design

One good metaphysical a priori deserves another. Since Darwinists have shifted their ground from
science to philosophy, it is legitimate to ask whether their axiomatic exclusion of design is the only
logical possibility. The answer, obviously, is no. Before Darwin, design was taken for granted by most
Western scientists, and even today, a significant number of scientists view the world as designed.

For the remainder of this paper, I will assume that living things are designed--not necessarily in every
detail, but in at least certain aspects. Specifically, I will assume that the human species was planned
before life began and that the history of life is the record of how this plan was implemented.

The Darwinian account of the history of life begins with the most primitive organisms and works its
way forward to the appearance of human beings. Although this is how events actually unfolded, from a
design perspective the idea of human beings came first, followed by a plan to achieve the goal. In a
sense, then, the plan took shape by working backward from the goal.1

What would the plan have to include? Any plan that places humans as the intended outcome would
have to provide for such basic needs as food, water, and a suitable environment. It can be argued that
humans have other needs as well, including social interactions, intellectual stimulation, and aesthetic
enjoyment. Here I will focus entirely on physical needs.

When human beings first appeared, the environment must have been congenial to unprotected human
life. From a design perspective, this human-friendly environment was planned. Advocates of the
Anthropic Principle have pointed out that such an environment was possible only because the
fundamental physical constants of the universe had the precise values they have. But these constants
are consistent with a wide range of environments, whereas life requires a relatively narrow range of
temperature, pressure, and other physical parameters. Therefore, in addition to the universal constants,
suitable local conditions would have needed to be part of the design as well.

Humans use oxygen in their metabolism and release carbon dioxide as a waste product. Therefore,
suitable local conditions must include an atmosphere containing these gases and a mechanism that
regenerates oxygen from carbon dioxide. This mechanism is photosynthesis, which is carried out by
green plants. It uses energy from the sun and also produces carbohydrates--another raw material in
human metabolism. Photosynthesis is a remarkably efficient system for maintaining an environment
congenial to human life. Unless some other mechanism is shown to be capable of fulfilling the same
role, a design perspective implies that organisms very much like green plants were a necessary part of
the original plan.

In addition to carbohydrates, the human body needs various other nutrients, including specific amino
acids, minerals, and vitamins. Our nutritional needs are quite complex and must be met on a regular
basis, so we are absolutely dependent on a variety of food sources. These are found in the plants and
animals around us. Since our needs include complex organic molecules found only in other living things,
those organisms are necessary for our existence.

Whatever organisms may have been necessary for human nutrition, their existence required a balanced
ecosystem that accommodated their needs. The original plan must have included a self-sustaining
biosphere in which reproduction and growth were balanced by death and decay. The balance among
organisms in an ecosystem is normally quite complex, and ecologists frequently discover that organisms
previously thought to be unessential are necessary elements in that balance. It is thus clear that
planning for human beings requires planning for many other organisms as well.

Getting from there to here

The need for large numbers of organisms becomes even more evident when we try to imagine how
human beings appeared on what was originally a lifeless planet. Although there is no consensus among
paleogeologists about atmospheric conditions on the primitive earth, those conditions were almost
certainly different from today's. The first organisms must have been capable of surviving in those
conditions and transforming them into an environment more favorable to human life.

In other words, primitive organisms had to pave the way for the stable ecosystems we see today. A
barren planet had to become a garden; soils containing organic nutrients for land plants had to be
produced. To use current biological terminology, ecological niches were filled by organisms adapted to
survive under local conditions. Those organisms then transformed their conditions, and other living
things took over.

Producing a congenial environment with nutritious foods, while necessary, would not have been
sufficient. Some people believe that the first human beings were created fully grown. But even if we
ignore psychological considerations and restrict ourselves to physical ones, birth and growth are
essential aspects of human beings as we know them. A creature that begins life without passing
through birth and childhood would be so unlike us that we could not regard it as truly human, regardless
of how great the superficial resemblance. And because human babies are totally dependent on other
creatures for their survival during early development, animals capable of raising the first human babies
must have been a necessary part of the original plan.

Human babies need milk to survive and grow, so mammals had to exist before humans appeared. And
not just any mammal. The first human baby presumably had to be nurtured by a creature very much
like itself--a humanlike primate. This creature, in turn, could only have been nurtured by a creature
intermediate in some respects between it and a more primitive mammal. In other words, a plan for the
emergence of human beings must have included something like the succession of prehistoric forms we
find in the fossil record.

Similar reasoning could be applied to earlier episodes in the history of life. For example, just as
mammals were necessary predecessors of the first humans, mammallike reptiles were presumably
needed to precede the first mammals, and so on. The emergence of humans thus depended on a
progression of creatures that increasingly resembled us.

Although this process is superficially similar to the Darwinian notion of common descent, design theory
differs from the latter in maintaining that predecessors need not be biological ancestors but only
providers of essential nourishment and protection. Successive organisms are "related" in the sense that
they represent planned stages in the history of life, but they are not genetically related as ancestors and
descendants. A planned succession would not require the innumerable transitional forms that Darwin
predicted. Design theory is thus more compatible than Darwinism with the discontinuities found in the
fossil record.

Design theory also does a better job than Darwin's theory in accounting for homology. According to
Darwin, features in diverse organisms are structurally similar ("homologous") because they are
inherited from a common ancestor. Biological inheritance implies that such features are more similar
because they are produced by similar genes or developmental pathways, but this implication is
contradicted by the genetic and embryological evidence.2 In a design view, however, homologies exist
(at least in part) because new organisms need to be protected and nourished by organisms somewhat
like them. But homologies need not be produced by similar genes or developmental pathways, since
there is no insistence on the sort of mechanistic continuity required by Darwinian common descent.

In conclusion, a design perspective on the history of life might turn out to account for the biological
evidence better than Darwinian evolution can. For example, Darwinism fails to specify why any given
organism exists, beyond insisting that it be able to survive. But for design theory, a variety of
creatures--including green plants and humanlike primates--are necessary prerequisites for human life.
A design perspective requires progressive stages in the history of life, as seen in the fossil record, but
unlike Darwin's theory it does not predict innumerable transitional forms that do not exist. Design
theory also suggests that homologies exist, at least in part, so that certain organisms can prepare the
way for others intended to follow them. Unlike Darwinism, it does not imply that homologous features
are produced by similar genes or developmental pathways, and so does not run afoul of the evidence.

This analysis, although preliminary and subject to revision, demonstrates that a design perspective has
major implications for our understanding of the biological evidence. As our knowledge of ecology and
human physiology increases, and as the analysis is refined and expanded, more detailed implications
will follow. In this way, a design perspective may eventually provide a detailed account of the history
of life more faithful to the evidence than Darwin's theory and thus offer a framework for more fruitful
research programs in biology.

1. See Unification Thought Institute, From Evolution Theory to a New Creation Theory: Errors in
Darwinism and a Proposal From Unification Thought (Tokyo: Kogensha, 1996).

2. See Jonathan Wells, "Homology in Biology: A Problem for Naturalistic Science," presented at the
Conference on Naturalism, Theism, and the Scientific Enterprise, Department of Philosophy, University
of Texas, Austin, February 1997 (posted on the World Wide Web at