Recognizing "Design"

Loren Haarsma (
Mon, 20 Apr 1998 15:02:16 -0400 (EDT)

Is it, in principle, possible to recognize "design" in biology -- design
in the sense that the term is used by advocates of ID theory? I believe
that it is. The following isn't a rigorous definition of the term;
however, I think that it matches our intuitive sense of the term and has
the potential to be scientifically useful. (I don't know if ID
advocates will agree with these criteria. They'll have to answer that
for themselves.)

We recognize "design" in objects, systems, or events when:

1) they are otherwise improbable;
2) they happen in a recognizable pattern;
3) there is or was, or could conceivably be, an intelligent agent capable
of producing it for whom it could serve some function.

Stonehenge is an example. First, there is a small probability of any
one rock of those approximate shapes naturally occurring where it is.
Second, the rocks form a recognizable pattern. Third, we know that
humans lived there who could have made it, and for whom it could have
functioned as a primitive observatory. That clinches it. Another
example like this is prehistoric stone tools.

SETI is another. SETI looks for a radio signal with a recognizable
pattern. (If a radio signal consists of densely coded information, it
would probably appear random unless you knew the code. But if the
information coding is not dense, the signal could form a recognizable
pattern.) SETI would then try to establish that there is a very low
probability of the signal being natural in origin. As for the third
point, it's not too difficult to conceive of such an agent.

The first criteria -- improbability -- could be made empirically
rigorous, although there is always some room for debate on how the
probabilities should be calculated and where the "cutoff" for
recognizing design should be placed. The third criteria can sometimes
be answered with an unambiguous "yes" or "no," but not always. The
second criteria -- recognizable pattern -- may be the trickiest. A
single improbably shaped stone tool, no matter how useful to prehistoric
humans, might have been formed by a natural process. It is a collection
of such tools, forming a pattern, which makes us conclude they were
designed. A single improbable event, even a very useful one, makes only
a weak argument for design. The appearance of a pattern makes the
argument strong.

Discovering a pattern *might* indicate that some unknown natural process
is at work. In fact, that possibility should always be investigated.
If a new natural process is discovered, that would cause us to reassess
the first criteria, the probability of the object or event. Barring
such a discovery, however, the existence of the pattern adds weight to
the design hypothesis.

In other posts, I talked about two classes of designed objects: those
assembled by intelligent agents, and those which self-organize out of
simpler component pieces which were themselves designed. Before we talk
about design in biology, it's worth noting that the criteria listed
above must be evaluated differently for different methods of assembly.

The first criteria starts by asking the probability that an object or
system could self-organize naturally. If the probability of self-
organization is very low, it might have been intelligently assembled.
If that probability for self-organization is high, then the question
shifts. What is the probability that the component pieces would have
such convenient characteristics as to make self-organization possible?
(Another way to ask this question is, how "fine-tuned" are the component
pieces for self-organization?)

For intelligently assembled objects, the third criteria asks whether
there is an agent capable of assembling it out of existing components.
For self-organized objects, the third criteria asks whether there is an
agent capable of manufacturing such components.

What about biology? Does it meet these criteria for design? (For now,
I will set aside the idea of self-organization and use "design" as ID-
advocates do, to indicate intelligently-aided assembly.)

For theists, the third criteria is easily met. God could have
miraculously intervened in biological history to make new life forms.
It is also worth at least mentioning, to meet the third criteria, the
possibility of alien intervention in earth's biological history.

The first criteria poses the greatest challenge to ID theory. How do we
calculate the probability of a particular new lifeform, or a particular
complex organ or biochemical process, developing "naturally"? Given our
current knowledge of mutational mechanisms, cellular physiology,
developmental biology, and population genetics, estimates range from
"virtually impossible" to "virtually certain." A lot more work needs to
be done. Simply pointing out the complexity of biological systems, or
the information content of the genome, will not convince those who are
not already convinced. To make the probability argument for a
particular organ or process, the genes which govern its development and
regulation will have to be mapped; corresponding genes in related
organisms (both organisms which share that organ or system, and those
which lack it) will have to be mapped. Empirical arguments based upon
these details are possible in principle, and are increasingly becoming
possible in practice.

The intuition of ID advocates is that the biological world has many
examples of such "improbable" structures and systems, and that empirical
studies will eventually make this clear. Let us suppose that they are
correct in that intuition. Suppose that empirical studies map out the
genomes, understand the structure, function, and physiology of the gene
products, and do eventually show not just one example, but a pattern
throughout the biological world of novel forms of complexity arising
"improbably." In that case, I believe, a very strong case would be made
for recognizing design.

Several objections to recognizing biological design have been raised.
First, it has been pointed out that we don't have a rigorous, empirical
definition of design. I don't think this is necessary in order for
"design" to be recognized, nor is it necessary for design to be a
scientifically useful concept. Not all scientifically useful concepts
are empirical concepts. If there emerged a recognizable pattern of
"improbable" elements of the biological world (which would make a strong
case for extra-naturally assembly), that pattern could be used to guide
further research, to generate new hypotheses and suggest new tests.

Second, it has been pointed out that highly organized and complex
information strings (such as the genome might be if it were designed)
cannot be distinguished mathematically from strings generated by natural
processes. That may be, but I don't believe that information content,
organization, or complexity are the best concepts for evaluating whether
or not a genome is "designed" in the ID-sense of the term. A more
important question is the probability of a genome being produced by
natural mechanisms. A specific question might be whether the
genomes of two species which are believed to share a particular common
ancestor could -- with reasonable probability -- have been produced by
mutation and natural selection. The question is especially
interesting if one of the species has a particular complex feature which
the other species and the putative common ancestor lacks.

Third, the argument has been made that design implies knowledge of
intent, and we cannot fully know the creator's intentions. I do not
think recognizing design requires knowledge of intent. Recognizing a
pattern, along with a reasonable hypothesis of intent, is sufficient for
design to be recognizable, and sufficient for it to be useful in guiding
further research.

My own intuition is that empirical studies will show that self-
organization of novelty and complexity in biology is, in fact, probable.
This brings us back to recognizing "design" in self-organized objects or
systems. God is certainly capable of creating and governing in this
way. This pattern is suggested by what we know from cosmology and
geology about the formation of creation's physical structures;
therefore, I wouldn't be surprised to see this pattern continued in
biological history. Present physical theory points to an extreme fine-
tuning of the properties of physical laws ("component pieces") in order to
make all of this possible. Although there is reason to be cautious in
arguing from fine-tuning, it certainly is suggestive. To me, it
provides ample reason to see design in all of nature, including the
biological world --- provided the third criteria is not a stumbling
block, provided you think the idea of a God who creates and governs in
this way is credible. I should close by saying that I don't believe in
God because of this argument for design. Rather, because I believe in
God, I can recognize design in all these things.

Loren Haarsma