Brief Primer on Intelligent Design
Having read a fair amount of material on intelligent design and having been involved in various discussions on the topic, I decided to prepare this brief primer that I trust will be useful in clarifying the central issues and in helping those less familiar with intelligent design understand its basic propositions.
This is not intended to be a comprehensive analysis of intelligent design, nor is it intended to respond to criticisms. Rather, this represents my modest attempt to avoid the side roads and the irrelevancies, and outline the fundamental central tenet of intelligent design, which is that some things exhibit characteristics of design that can be objectively and reliably detected. It is my view that criticisms of intelligent design must focus on this central tenet, or risk missing the mark. It is also with this central tenet that intelligent design stands or falls as a scientific enterprise.
Setting the Stage
As with so many issues, it is important to first define our terms. In public debates, the term “intelligent design” is often incorrectly associated with anyone who believes that the Earth and all life upon the Earth were actively created by an intelligent Creator, and when used pejoratively, the term generates much more heat than light and adds no substantive insight to the discussion.
In a broader sense, the term might be applied to individuals who hold to a basic teleological view of the universe or the diversity of life on earth. In this sense, many individuals believe in some form of intelligent design, including those who hold to an initial act of life’s creation, followed by naturalistic evolutionary mechanisms.
In yet a more concrete sense, the term is often used with respect to those involved in the modern intelligent design movement, including vocal proponents such as Philip Johnson and Jonathan Wells. Although Johnson and Wells are certainly involved in the broader intelligent design movement, they largely use intelligent design as a tool for promoting change in current educational and philosophical frameworks. This use of intelligent design as a tool for change has received by far the most press coverage and is at the heart of the often-heated debates over school curricula. However, as intelligent design’s primary spokesperson, William Dembski, has pointed out, intelligent design’s use as a tool for change is secondary to intelligent design’s undertaking as an independent scientific enterprise.
Finally, therefore, intelligent design refers to the science of detecting design. In this latter sense, intelligent design is not limited to debates over evolutionary theory or discussions of design in nature, but covers the study of signs of intelligence wherever they may occur: whether in archeology, forensic science, the search for extraterrestrial intelligence, or otherwise. (Though not strictly limited to historical events, intelligent design argues that design can be detected in some things even in the absence of any reliable historical record or independent knowledge of a designing intelligence. It is in this context that we wish to discuss intelligent design.) Defined more tightly, intelligent design can thus be viewed as the science of studying the criteria, parameters and procedures for reliably detecting the activity of an intelligent agent.
Associated with this latter more limited definition are scientists involved in such a scientific enterprise. These individuals include, probably most notably, Dembski and Michael Behe, and a number of other scientists who have begun to take notice of intelligent design as a legitimate scientific inquiry.
It is in this latter sense that I wish to examine the concept of intelligent design.
What then is the basic foundation and what are the basic propositions of intelligent design?
Intelligent design begins with a very basic proposition: some things are designed. This is slightly more complicated than it sounds, but not much, if we keep a couple of points in mind.
First, one might object that many things appear to be partly designed and partly not. This, however, is simply a matter of drilling down deeply enough to identify the discrete “thing” being examined. For example, if we look at a stone wall we can see that it is made up of stones of various sizes and shapes. Even if we assume that the stones themselves were not the product of intelligent design, we would conclude that they have been used by an intelligent agent in designing and building the wall. Thus, in situations where something looks partly designed and partly not designed, we need simply drill down further and determine which aspect, portion, or piece of the “thing” we are evaluating. In this example, are we examining the individual stones, or are we examining their overall arrangement, pattern, and resulting function?
Even if we are unable to break down a particular object or system into its component parts, and we end up with a “thing” that is partially designed and partially not designed, the initial proposition of intelligent design would remain essentially the same: some parts, or portions, or components of some things are designed.
Second, when we talk about the fact that some things are designed, we are not referring only to physical objects, but are referring to anything that is the subject of design, whether it be a physical object, a system, or a message or other representation able to convey information. Thus if I took the same naturally-occurring stones, and instead of building a wall, I laid them out on the beach to spell a message, we would also have a clear indication of the actions of an intelligent agent, once again not in the stones themselves, but in the representation created by the stones and the information conveyed by that representation.
Given this basic proposition that some things are designed, intelligent design then asks the next logical question: is it possible to detect design? As others have pointed out, if the unlikely answer is “no,” then we can only say that everything may or may not be designed, and we have no way of determining whether any particular item is or is not designed. However, if the likely answer is “yes,” then this leads to a final and more challenging question that lies at the heart of intelligent design theory and intelligent design as a scientific enterprise: how does one reliably detect design?
Characteristics of Design and Limitations of Intelligent Design
What kinds of characteristics do things that are designed exhibit? When we contemplate things that are designed – a car, a computer, a carefully coordinated bouquet of flowers – a number of characteristics might spring to mind, such as regularity, order, and beauty. However, if we think for a moment, we can come up with many examples of naturally occurring phenomena that might fit these descriptions: the rotation of the Earth that brings each new day and the well-timed phases of the moon exhibit regularity; naturally-occurring crystals are examples of nearly flawless order; the rainbow or the sunset, resulting from the sun’s rays playing in the atmosphere, are paradigms of beauty.
To be sure, characteristics such as regularity and order might be strongly indicative of an intelligent agent in those instances where natural phenomena would not normally account for them, such as a handful of evenly spaced flowers growing beside the highway, or a pile of carefully stacked rocks along the hiking trail. Nevertheless, because there are many instances of naturally occurring phenomena that exhibit regularity, order, and beauty, the mere existence of these characteristics is not necessarily indicative of design. In other words, these are not necessary defining characteristics of design.
On the flip side, there are many things that are designed that do not exhibit any particular regularity or order, at least not in a mathematical sense, such as a painting or a sculpture. There are also many objects of design that do not evoke any particular sense of beauty. And this brings up an important limitation of intelligent design: we are not able to identify everything that is designed.
A related limitation arises in that we cannot say with certainty that a particular thing is not designed. This is particularly true, given that many things are purposely designed to resemble naturally occurring phenomena. For example, in my yard I have many rocks that have been purposely designed and strategically placed to resemble the random placement of rocks in a stream. In addition, when I recently remodeled a room in my home, I used a faux painting technique – carefully designed and coordinated over the course of several hours – to resemble a naturally occurring pattern.
As a result, intelligent design is limited in two important aspects: it can neither identify all things that are designed, nor can it tell us with certainty that a particular thing is not designed.
But that leaves one remaining possibility: is it possible to identify with certainty some things that are designed? Dembski and Behe would argue that the answer is “yes.”
Possibility versus Probability
In order to identify with certainty that something is designed, we must be able to define characteristics that, while not necessarily present in all things designed, are never present in things not designed. It is in defining these characteristics and setting the parameters for identifying and studying these characteristics, that intelligent design seeks to make its scientific contribution.
We have already reviewed some potential characteristics of things that might be designed, and have noted, for example, that regularity and order do not necessarily define design. I have posited, however, that regularity and order might provide an inference of design, in those instances where natural phenomena would not normally account for them, such as the handful of evenly spaced flowers or the pile of stacked rocks. Let’s examine these two examples in a bit more detail.
Is it possible that this pattern of flowers or the stack of rocks occurred naturally? Yes, it is possible. It is also possible, at least as a pure logical matter, that the sun will cease to shine tomorrow morning at 9:00 a.m. To give a stronger example, is it possible that the laws of physics will fail tonight at midnight? Sure, as a pure logical matter. But is it likely? Absolutely not. In fact, based on past observations and experience, we deem such an event so unlikely as to be a practical impossibility.
Note that in the examples of the sun ceasing to shine or the laws of physics failing we are not talking simply about unusual or rare events; rather we are talking about something so improbable that we, our precious scientific theories, and the very community in which we live are more likely to pass into oblivion before the event in question occurs. Thus for all practical purposes, within the frame of reference of the universe as we understand it and the world in which we live and operate, it can be deemed an impossibility. Dembski has already skillfully addressed this issue of logical possibility, so I will not review the matter further, except to summarize that in science we are not so interested in pure logical possibility as in realistic probability. It is within this realm of probability that all science operates, and it is in this sense that we must view the probabilities relevant to intelligent design.
However, while we need not be concerned with wildly speculative logical possibilities, we might nevertheless conclude that the pattern of flowers or the stack of rocks is possible, not only as a matter of logical possibility, but also as a matter of reasonable probability, within the realm of our experience. After all, there are lots of flowers on the Earth and surely a handful of them must eventually turn up evenly spaced as though carefully planted. In addition, we have all seen precariously balanced rocks, formed as a result of erosion acting on rocks of disparate hardness, so perhaps our pile of rocks also occurred naturally. We might admit that our flowers and our stack of rocks are rare and unusual natural phenomena, but we would argue that they are not outside of the realm of probability or our past experience.
Thus, the inference of design needs to get much stronger before we are satisfied that our pattern of flowers or our stack of rocks have been designed.
The Design Inference Continuum
Now let’s suppose that we tweak the examples a bit. Let’s suppose that instead of a handful of flowers, we have several dozen flowers, each evenly spaced one foot apart along the highway. Can we safely conclude that this is the product of design? What about a dozen identical stacks of rocks along the hiking trail? One might still mount an argument that these phenomena do not yet reliably indicate design because they could have been created naturally. Nevertheless, in making such an argument we would be relying less on realistic probabilities and what we know about the world around us, and slipping closer to the argument by logical possibility. This precisely the mistake for which Dembski takes Allen Orr to task.
Now allow me to tweak yet a bit more. Let’s suppose that the dozens of flowers are now hundreds, each in a carefully and evenly spaced pattern along the highway. At this point, the probability of natural occurrence becomes so low as to completely escape our previous experience; it becomes so low as to suggest practical impossibility. Is it the sheer number of flowers that puts us over the hump? No, it is not the number of flowers itself that provides evidence for design, but the number of spacings between the flowers, the complexity of the overall pattern, and the fact that these spacings and the resulting complexity are not required by any natural law, but are only one of any number of possible variations. In other words, it is the discretionary placement of all of these flowers, selected from among the nearly infinite number of placements possible under natural laws, which allows us to infer design. It is this placement of all the flowers, which gives the characteristics of specificity and complexity, and which Dembski terms “specified complexity.” And it is in this realm of specified complexity that the probability of non-design nears impossibility, and our confidence in inferring design nears certainty.
Yet, our examples can become even more compelling. As a last modification, let’s suppose that the flowers are now arranged by the side of the road in the outline of the state of Texas, complete with Bluebonnets in the shape of the Lone Star. Let’s suppose that our stacks of rocks are arranged so that there is one stack exactly each mile along the trail, or one stack at each fork in the trail. Now we have not only specified complex patterns, but patterns high in secondary information content. In the one case we have a shape that identifies Texas, a particular type of flower that signifies the state, and a star that is not just a pattern, but a pattern with strong symbolic meaning. Along our hiking trail we have markers that carry out a function by providing specific information regarding changes in the trail or indicating the distance traveled.
Intelligent design, as a scientific enterprise is geared toward this end of the probability continuum where the probability of non-design nears zero and the probability of design nears one. In some ways, focusing only on the area of most certainty is a rather modest and limiting approach. Yet design theorists willingly give up the possibility of identifying design in many cases where it in fact exists, in exchange for the accuracy and the certainty that a more stringent set of criteria bestow. In this way, the design inference is lifted from the level of broad intuition to a focused scientific instrument with definitive testable criteria.
As a scientific undertaking, intelligent design is not in the business of identifying all things designed, nor is it in the business of confirming with certainty that a particular thing is not designed. Indeed, intelligent design, and it is fair to say current human knowledge, is incapable of performing these tasks. What intelligent design does seek to do, however, is identify some things that are designed.
We have seen that the argument to design is essentially an inference based on probabilities. As a result, there is a continuum ranging from the likelihood of non-design to the likelihood of design. At a certain point the probability of non-design nears zero and the probability of design nears one. At that point we can say, the design theorist argues, with as much certainty as any other scientific fact or proposition, that the thing in question was designed. It is in this area of specified complexity (of which high secondary information content and Behe’s “irreducible complexity” are examples) that the theory of intelligent design operates.
Criticisms of intelligent design based on social, religious, philosophical, or cultural grounds, including complaints about the identity, motives, or capabilities of the putative designer, miss the mark. Design theorists argue that specified complexity can be objectively and reliably defined and detected so that the probability of non-design nears impossibility and the probability of design nears certainty. This is intelligent design’s central tenet. It is on this point, and only on this point, that intelligent design as a scientific undertaking can be appropriately challenged and criticized. And it is on this point that Dembski, Behe, and others are confident that intelligent design will make its greatest contribution.
September 9, 2003