Some Thoughts on Evolutionary Theory

Scientific theories are usually not radically contested for more than about
thirty years or so after they are first proposed. A small number will be
debated for longer than that, but usually thirty years is enough for evidence
to be collected and a decision made by society as a whole. Copernicus’
theory, that the Sun is, in fact, the centre of our Solar System, was one of
the few that was not immediately accepted or discredited. Evolutionary theory
is in the same position today, and for mostly the same reasons.

In this essay, I wish to analyse the evolutionary theories of Natural
Selection, Strict Creationism, and Intelligent Design, and evaluate each as a
scientific theory. While issues of evidence are often taken to be sufficient
to demonstrate Natural Selection’s superiority to the other two, an adequate
evaluation of such evidence would require an in-depth analysis and
investigation of contemporary and historical scientific literature. Instead
of taking this painstaking approach, we can evaluate these theories in other
ways, and thus can push some of the issues related to evidence to one side.
In this way, we need not give over our personal judgement on evolution to
expert proponents of one or another theory.

It is often assumed that the principle of `verification’ is sufficient to
evaluate the scientific value of a given theory, ie, if a theory has a lot of
evidence supporting it, then it must be better than a theory which doesn’t.
However, since the early part of the twentieth century, critics have pointed
out that what qualifies as `evidence’ is intimately connected with the
attitudes of those investigating the theory. For example, suppose a
scientist, in making calculations, arrives at a value which flatly contradicts
the theory they are operating under They may take this as grounds to reject
the theory, or believe that they made some error in their calculations, or
chalk the problem up to `experimental error’. One position may be somewhat
stronger than others (for example, if this is the first such contradiction to
appear in the community of scientists working with this theory, an error in
calculation or measurement may be claimed) but that does not mean that it is
automatically `the right answer’. This is called the Duhem-Quine problem,
after two well-known philosophers of science.

More generally, the evidence used to compare theories only makes sense when
those theories can agree on what counts as evidence and how to interpret it;
this remains quite controversial in comparing our three theories here. Thus,
I offer instead of a purely empirical test the principle of `fecundity’, which
says that a given theory is better than an alternative if the given theory
allows us to understand reality better, in terms of making predictions, making
models, and interpreting sense-data. In a loose sense, a theory is more
fecund if it gives scientists lots of interesting things to think about.
Using this we can say, for instance, that because they are useful suppositions
it is rational to at least pretend middle-sized durable physical objects
(chairs) and electrons exist: we don’t have to constantly worry about whether
the thing we think we’re about to sit on is really there, and we get
transistors and other microelectronic devices.

Some definitions: By Natural Selection I refer to the theory of Darwin,
and its intellectual descendants, including Punctuated Equilibrium. All life
evolved from a common ancestor, and the features of organisms are the result
of changes over time. The late Stephen Jay Gould exemplifies the advocate of
Natural Selection, as does Richard Dawkins; we neglected the differences
between these theorists here, because they’re basically identical when
compared with the other two theories. By Intelligent Design I refer to the
theory which modifies Natural Selection by saying certain features of
organisms (examples usually given are eyes, lungs, and blood clotting) are
`irreducibly complex’, could not have evolved in a gradual way, and thus were
literally designed and implemented by some intelligent creator. William
Dembski is an excellent prototype of an `Intelligent Design-er’. And by
Strict Creationism I refer to the theory which says some demiurge brought
forth all forms of life as they are today a few thousand years ago, in the
spirit of a literal reading of Genesis, chapter one, and all sorts of
Creationists whose names I don’t know. Using these definitions and the
principle of fecundity, we may now evaluate each theory by asking of each the
question `How does this help us understand reality?’

Strict Creationism does this marginally at best. It allows us some sense
of why organisms are the way they are, but it cannot explain this to any depth
or offer us much beyond this. Strict Creationism gives us `Organisms are the
way they are because God said so’, but when pressed as to why God made
organisms this way, Strict Creationism has no response, and indeed it is
inappropriate within this theory to press further: God is ineffable, and it
is inappropriate for mere humans to ponder God’s motivations. Why do we have
appendices and wisdom teeth? Because God said so. Why did God say so? No
more questions, please. If we are to agree with Strict Creationism, we must
adopt an epistemological wall immediately in front of our noses, and biology
is reduced to descriptive physiology. Strict Creationism also offers no way
to model or predict organic change: if every aspect of our physiology was
established in a permanent form at creation, there is no reason to expect
organic change (if it is even possible), unless God ineffably decides to
introduce it. Any sort of bio-engineering, whether genetic engineering,
pharmaceuticals, or even combating disease, as thus epistemologically
impossible.

Intelligent Design does do a bit better: by accepting some degree of
evolution it allows us to say, for instance, that humans have appendices
because our ancestors had some use for them, as certain organisms that appear
closely related to us by genetic evidence still do. But, ultimately, the
theory posits another `black box’ around the workings of biological systems,
into which we cannot penetrate. We do enjoy more room to move about here than
under Strict Creationism, but trying to bridge those irreducibly complex gaps
is an impossibility: perhaps God introduced the irreducible features between
generations, or established the original community of new species by fiat, but
either way the process is not fully comprehensible by mortal minds. Hence,
any attempt to accurately model biological history is, in this theory,
necessarily incomplete. We also run into the same problem of ineffable divine
intervention: God had to step in and implant these irreducibly complex
components into our ancestry, and it is impossible for us to predict what may
happen in the future so long as we posit divine intervention always, as it
were, waiting in the wings for its next, unexpected walk-on. So while
Intelligent Design theory is a great deal more fecund than Strict Creationism,
is leaves us wanting.

Natural Selection does not have this problem. By not assuming any sort of
irreducible complexity or other epistemological barrier, Natural Selection
permits us to investigate and understand the evolution of the human eye, blood
clotting, and whatever else we might be interested in. We also never have to
postulate a necessary God: Natural Selection works whether you think it’s a
completely mechanical system, happening according to chance and quasi-economic
principles or you think some deity is working things towards a conclusion by
influencing what random variations occur to be selected for or against by the
process itself. This freedom in choice of models gives Natural Selection
considerable power for impressive feats of bio-engineering: we can, for
example, model the evolution of HIV in a patient under certain programmes of
medication with a good amount of accuracy.

We can get a better hold of the difference by considering a general
scientific theory: a set of rules that models, if only probablistically and
approximately, the workings of some part of the universe. For this theory to
be fecund at all, it must be at least probablistically and approximately
determinate; that is, the theory must allow us to make statements like `If the
system is similar to x near time t, then the probability is close to p that it
will be similar to y near time t+1′. But this means that while we can allow
some leeway into our theory, for human error or other issues of uncertainty or
randomness, we cannot permit our theory to include aspects that might occur at
any time and have completely unpredictable effects.

Consider Quantum Mechanics, as it is normally understood. Many events that
we consider impossible are merely extraordinarily unlikely, according to
Quantum Mechanics: for example, there is a non-zero probability (so it is
possible, even if extraordinarily unlikely) that I could appear in your living
room as you’re reading this sentence. There’s even a non-zero probability
that this could happen for every individual who reads these words. But
Quantum Mechanics, for all its probabilistic weirdness, does not allow
completely inexplicable causes and events, which would make prediction of even
the roughest, most probabilistic and approximate sort completely impossible.

The God in Strict Creationism and Intelligent Design is precisely this sort
of completely unaccountable cause, and divine intervention a correspondingly
completely inexplicable event. Their theistic assumptions make these theories
less fecund than the `weakly atheistic’ (in the sense that it does not involve
any divine intervention in the selection process itself) Natural Selection.

More broadly, Natural Selection has the same modelling strength and
flexbility as Quantum Theory. Both of these theories are flexible enough to
develop a wide variety of models, so we would expect both to be extremely
fecund — as they are. Quantum Theory is notable for its usefulness in all
areas of modern technology; Natural Selection has been extremely useful in
many areas of bio-engineering and medicine. Strict Creationism and
Intelligent Design are not as flexible in their ability to create models, so
we can conclude that they are less fecund than Natural Selection.

Hence, our analysis comes down in favour of Natural Selection: it gives us
a more comprehensive, more interesting, and more useful way to model and
interpret reality. It is therefore reasonable for us to conclude that Natural
Selection is a better scientific theory than either Intelligent Design or
Strict Creationism. The fact that the evidence gathered generally supports
Natural Selection over the other two happily bolsters this conclusion.

Human history becomes more and more a race between education and catastrophe. –H.G. Wells, The Outline of History

Every child ought to be brought up more intelligent than [his or her] parents. –Clarence Darrow, Scope’s “Monkey” Trial