Why Intelligent Design is a scientific theory

This article is from the book “Signature in the cell” by dr. Stephen Meyer

How to define science ?
Stephen Meyer write that he as a philosopher of science, always have thought there was something odd and even disingenuous about the objection that intelligent design is not scientific. The argument shifts the focus from an interesting question of truth to a trivial question of definition. To say that an idea or theory does or does not qualify as science implies an accepted definition of the term by which to make that judgment. But to say that a claim about reality “is not science” according to some definition says nothing about whether the claim is true—unless it can be assumed that only scientific theories are true. A definition of science does not, by itself, tell us anything about the truth of competing statements, but only how to classify them (whether as scientific or something else, such as philosophical, historical, or religious statements). 

So, at one level, Meyer regarded the debate about whether intelligent design qualifies as science as essentially a semantic dispute, one that distracts attention from significant questions about what actually happened in the past to cause life to arise. Does life exhibit evidence of intelligent design or just apparent design? Did life arise by undirected processes, or did a designing intelligence play a role? Surely such questions are not settled by defining one of the competing hypotheses as “unscientific” and then refusing to consider it. 

At another level the debate is tacitly a debate about the basis of the theory itself. Since the term “science” connotes a rigorous experimental or empirical method for studying nature, denying that an idea is scientific implies that rigorous empirical methods played no role in its formulation. To emphasize this impression, many critics of intelligent design insist that the theory is not testable and, for this reason, is neither rigorous nor scientific.8 Because many people assume that only “the” scientific method produces justified conclusions, the charge that the theory isn’t science seems to justify dismissing it as merely a subjectively based opinion or belief. The objection “ID isn’t science” is code for “It isn’t true,” “It’s disreputable,” and “There is no evidence for it.” 

That is why the claim that intelligent design is not science—repeated often and with great presumed authority—has led many to reject it before considering the evidence and arguments for it. Meyer realized that in order to make his case—and open minds to the evidence in favor of it—he needed to defend the theory of intelligent design against this charge. To do so, indeed to defend any theory against this charge and to do so with intellectual integrity, requires one to navigate some treacherous philosophical waters. To claim that intelligent design is science implicitly invokes a definition of science—some understanding of what science is. But which definition? 

Because of Meyer´s background, he knew that historians and philosophers of science—the scholars who study such questions—do not agree about how to define science. Many doubt there is even a single definition that can characterize all the different kinds of science. In the philosophy of science this is known as the “demarcation problem,” the problem of defining science and distinguishing (or “demarcating”) it from “pseudoscience,” metaphysics, history, religion, or other forms of thought or inquiry. 

Typically, philosophers of science have tried to define science and distinguish it from other types of inquiry (or systems of belief) by studying the methods that scientists use to study nature. But that’s where the trouble started. As historians and philosophers of science studied the methods that scientists use, they realized that scientists in different fields use different methods. 

This, incidentally, is why historians and philosophers of science are generally better qualified to adjudicate the demarcation question than scientific specialists, such as inorganic chemists, for example. As they say of the catcher in baseball, the philosopher and historian of science has a view of the whole field of play, meaning he or she is less likely to fall into the error of defining all of science by the practices used in one corner of the scientific world. I already had some inkling of this from my work as a geophysicist. I was aware that historical and structural geology use distinct (if partially overlapping) methods. But as I delved into the demarcation question, I discovered that different sciences use a wide variety of methods. 

Some sciences perform laboratory experiments. Some do not. Some sciences name, classify, and organize natural objects; some sciences seek to discover natural laws; others seek to reconstruct past events. Some sciences seek to formulate causal explanations of natural phenomena. Some provide mathematical descriptions of natural phenomena. Some sciences construct models. Some explain general or repeatable phenomena by reference to natural laws or general theories. Some study unique or particular events and seek to explain them by reference to past (causal) events. 

Some sciences test their theories by making predictions; some test their theories by assessing their explanatory power; some test their theories by assessing both explanatory power and predictive success. Some methods of scientific investigation involve direct verification, some employ more indirect methods of testing. Some test theories in isolation from competing hypotheses. Some test theories by comparing the predictive or explanatory success of competing hypotheses. Some branches of science formulate conjectures that cannot yet be tested at all. Some sciences study only what can be observed. Some sciences make inferences about entities that cannot be observed. Some sciences reason deductively; some inductively; some abductively. Some use all three modes of inference. Some sciences use the hypothetico-deductive method of testing. Some use the method of multiple competing hypotheses. 

This diversity of methods has doomed attempts to find a single definition (or set of criteria) that accurately characterizes all types of science by reference to their methodological practices. Thus, philosophers of science now talk openly about the “demise” of attempts to demarcate or define science by reference to a single set of methods. 

To say that an idea, theory, concept, inference, or explanation is or isn’t scientific requires a particular definition of science. Yet if different scientists and philosophers of science could not agree about what the scientific method is, how could they decide what did and did not qualify as science? And how could Meyer argue that the theory of intelligent design is scientific, if he could not say what I meant by “science”? Conversely, how could critics of intelligent design assert that intelligent design is not science without articulating the standard by which they made this judgment? How could any headway in this debate be made without an agreed-upon definition? 

Meyer discovered that though it was difficult to define science by reference to a single definition or set of methodological criteria, it was not difficult to define science in such a way that either acknowledged the diversity of methodological practices or refused to specify which method made a discipline scientific. Such an approach allows science to be defined more broadly as, for instance, “a systematic way of studying nature involving observation, experimentation, and/or reasoning about physical phenomena.” So far, so good. The difficulty has come when scholars tried to equate science with a particular systematic method of studying nature to the exclusion of other such methods. 

The situation was not hopeless, however. I discovered that although it was impossible to describe the rich variety of scientific methods with a single definition, it was possible to characterize the methodological practices of specific disciplines or types of science. This made sense. It was precisely the diversity of scientific methods that made defining science as a whole difficult in the first place. Focusing on a single established scientific method as the relevant standard of judgment eliminated the practical problem of deciding how to assess the scientific status of a theory without an established definition of science. Furthermore, from my own studies, I knew the methodological practices of the sciences directly relevant to the questions I was pursuing—the sciences that investigate the causes of particular events in the remote past. Stephen Jay Gould called these sciences the historical sciences.11 I knew that the inference to design followed from a rigorous application of the logical and methodological guidelines of these disciplines. As shown in his book Signature of the cell, Chapter 15, Meyer carefully followed these guidelines in constructing his own case for design, so he concluded that there was a good (if definition-dependent) reason to regard intelligent design as a scientific—and, specifically, historically scientific—theory. In fact, there are several such reasons.

Reason 1: The Case for ID Is Based on Empirical Evidence The case for intelligent design, like other scientific theories, is based upon empirical evidence, not religious dogma. Contrary to the claims of Robert Pennock, one of the expert witnesses in the Dover trial, design theorists have developed specific empirical arguments to support their theory. Stephen Meyer have developed an argument for intelligent design based on the discovery of digital information in the cell presented in his book Signature in the cell. In addition, other scientists now see evidence of intelligent design in the “irreducible complexity” of molecular machines and circuits in the cell, the pattern of appearance of the major groups of organisms in the fossil record, the fine-tuning of the laws and constants of physics, the fine-tuning of our terrestrial environment, the information-processing system of the cell, and even in the phenomenon known as “homology” (evidence previously thought to provide unequivocal support for neo-Darwinism). Critics may disagree with the conclusions of these design arguments, but they cannot reasonably deny that they are based upon commonly accepted observations of the natural world. Since the term “science” commonly connotes an activity in which theories are developed to explain observations of the natural world, the empirical, observational basis of the theory of intelligent design provides a good reason for regarding intelligent design as a scientific theory. 

Reason 2: Advocates of ID Use Established Scientific Methods

The case for intelligent design follows from the application of not one, but two separate systematic methods of scientific reasoning—methods that establish criteria for determining when observed evidence supports a hypothesis. In Chapter 15 in the book Signature of the cell, Meyer discussed the primary method, the method of multiple competing hypotheses, by which he inferred and justified the inference to intelligent design as the best explanation for the origin of biological information. As noted there, this method is a standard method of scientific reasoning in several well-established scientific disciplines. Below he discuss in more detail how this method makes it possible to test intelligent design. Additionally, advocates of intelligent design have developed another method that complements the method of multiple competing hypotheses. 

In The Design Inference (and in subsequent works), William Dembski established criteria by which intelligently designed systems can be identified by the kinds of patterns and probabilistic signatures they exhibit. Based on these criteria, Dembski developed a comparative evaluation procedure—his explanatory filter—to guide our analysis and reasoning about natural objects and artifacts and to help investigators decide among three different types of explanations: chance, necessity, and design. As such, it constitutes a rigorous, systematic, evidence-based method for detecting the effects of intelligence, again suggesting a good reason to regard intelligent design as scientific in accord with common definitions of the term. 

Reason 3: 

ID Is a Testable
 Theory Most scientists and philosophers of science think that the ability to subject theories to empirical tests constitutes an important aspect of any scientific method of study. But for a theory to be testable, there must be some evidential grounds by which it could be shown to be incorrect or inadequate. And, contrary to the repeated claims of its detractors, the theory of intelligent design is testable. In fact, it is testable in several interrelated ways. 

First, like other scientific theories concerned with explaining events in the remote past, intelligent design is testable by comparing its explanatory power to that of competing theories. Darwin used this method of testing in On the Origin of Species. That the theory of intelligent design can explain the origin of biological information (and the origin of the cell’s interdependent information-processing system) better than its materialistic competitors shows that it has passed an important scientific test. 

This comparative process is not a hall of mirrors, a competition without an external standard of judgment. The theory of intelligent design, like the other historical scientific theories it competes against, is tested against our knowledge of the evidence in need of explanation and our knowledge of the cause-and-effect structure of the world. Evaluations of “causal adequacy” guide historical scientific reasoning and help to determine which hypothesis among a competing group of hypotheses has the best explanatory power. Considerations of causal adequacy provide an experience-based criterion by which to test—accept, reject, or prefer—competing historical scientific theories. When such theories cite causes that are known to produce the effect in question, they meet the test of causal adequacy; when they fail to cite such causes, they fail to meet this test. As an example, the earthquake hypothesis fails the test of causal adequacy because we do not have evidence that earthquakes cause layers of volcanic ash to accumulate, whereas the volcanic eruption hypothesis passes the test of causal adequacy because experience has shown that eruptions do cause this phenomenon. 

Since empirical considerations provide grounds for rejecting historical scientific theories or preferring one theory over another, such theories are clearly testable. Like other historical scientific theories, intelligent design makes claims about the cause of past events, thus making it testable against our knowledge of cause and effect. Moreover, because experience shows that an intelligent agent is not only a known, but also the only known cause of specified, digitally encoded information, the theory of intelligent design developed in this book has passed two critical tests: the tests of causal adequacy and causal existence. Precisely because intelligent design uniquely passed these tests, Meyer argued that it stands as the best explanation of the DNA enigma. 

Finally, though historical scientific theories typically do not make predictions that can be tested under controlled laboratory conditions, they do sometimes generate discriminating predictions about what we should find in the natural world—predictions that enable scientists to compare them to other historical scientific theories. The theory of intelligent design has generated a number of such discriminating empirical predictions. These predictions not only distinguish the theory of intelligent design from competing evolutionary theories; they also serve to confirm the design hypothesis rather than its competitors. 

Consider the case of so-called junk DNA—the DNA that does not code for proteins found in the genomes of both one-celled organisms and multicellular plants and animals. The theory of intelligent design and materialistic evolutionary theories (both chemical and biological) differ in their interpretation of so-called junk DNA. Since neo-Darwinism holds that new biological information arises as the result of a process of mutational trial and error, it predicts that nonfunctional DNA would tend to accumulate in the genomes of eukaryotic organisms (organisms whose cells contain nuclei). Since most chemical evolutionary theories also envision some role for chance interactions in the origin of biological information, they imply that nonfunctional DNA would have similarly accumulated in the first simple (prokaryotic) organisms—as a kind of remnant of whatever undirected process first produced functional information in the cell. For this reason, most evolutionary biologists concluded upon the discovery of nonprotein-coding DNA that such DNA was “junk.” In their view, discovery of the nonprotein-coding regions confirmed the prediction or expectation of naturalistic evolutionary theories and disconfirmed an implicit prediction of intelligent design. 

As Michael Shermer argues, “Rather than being intelligently designed, the human genome looks more and more like a mosaic of mutations, fragmented copies, borrowed sequences, and discarded strings of DNA that were jerry-built over millions of years of evolution.” Or as Ken Miller argues: “The critics of evolution like to say that the complexity of the genome makes it clear that it was designed…. But there’s a problem with that analysis, and it’s a serious one. The problem is the genome itself: it’s not perfect. In fact, it’s riddled with useless information, mistakes, and broken genes…. Molecular biologists actually call some of these regions ‘gene deserts,’ reflecting their barren nature.” Or as philosopher of science Philip Kitcher puts it, “If you were designing the genomes of organisms, you would not fill them up with junk.” 

ID advocates advance a different view of nonprotein-coding DNA. The theory of intelligent design predicts that most of the nonprotein-coding sequences in the genome should perform some biological function, even if they do not direct protein synthesis. ID theorists do not deny that mutational processes might have degraded or “broken” some previously functional DNA, but we predict that the functional DNA (the signal) should dwarf the nonfunctional DNA (the noise), and not the reverse. As William Dembski explained and predicted in 1998: “On an evolutionary view we expect a lot of useless DNA. If, on the other hand, organisms are designed, we expect DNA, as much as possible, to exhibit function.” The discovery in recent years that nonprotein-coding DNA performs a diversity of important biological functions has confirmed this prediction. It also decisively refutes prominent critics of intelligent design—including Shermer, Miller, and Kitcher—who have continued to argue (each as recently as 2008) that the genome is composed of mostly useless DNA. 

Contrary to their claims, recent scientific discoveries have shown that the nonprotein-coding regions of the genome direct the production of RNA molecules that regulate the use of the protein-coding regions of DNA. Cell and genome biologists have also discovered that these supposedly “useless” nonprotein-coding regions of the genome: 
(1) regulate DNA replication, 
(2) regulate transcription, 
(3) mark sites for programmed rearrangements of genetic material, 
(4) influence the proper folding and maintenance of chromosomes, 
(5) control the interactions of chromosomes with the nuclear membrane (and matrix),
(6) control RNA processing, editing, and splicing, 
(7) modulate translation, 
(8) regulate embryological development, 
(9) repair DNA, and 
(10) aid in immunodefense or fighting disease among other functions. 

In some cases, “junk” DNA has even been found to code functional genes. Overall, the nonprotein-coding regions of the genome function much like an operating system in a computer that can direct multiple operations simultaneously. Indeed, far from being “junk,” as materialistic theories of evolution assumed, the nonprotein-coding DNA directs the use of other information in the genome, just as an operating system directs the use of the information contained in various application programs stored in a computer. In any case, contrary to the often heard criticism that the theory makes no predictions, intelligent design not only makes a discriminating prediction about the nature of “junk DNA”; recent discoveries about nonprotein-coding DNA confirm the prediction that it makes. 

Reason 4: The Case for ID Exemplifies Historical Scientific Reasoning

Not only do scientists use systematic methods to infer intelligent design; the specific methods they use conform closely to established patterns of inquiry in the historical sciences. Indeed, the theory of intelligent design and the patterns of reasoning used to infer and defend it exemplify each of the key features of a historical science. During Meyer´s doctoral studies he discovered several distinctive characteristics of historical scientific disciplines—disciplines that try to reconstruct the past and explain present evidence by reference to past causes rather than trying to classify or explain unchanging laws and properties of nature. He found that historical sciences generally can be distinguished from nonhistorical scientific disciplines by reference to four criteria. And the theory of intelligent design (and the modes of inference used to establish and test it) provides a good example of each of the key features of a historical science. 

A Distinctive Historical Objective

Historical sciences focus on questions of the form, “What happened?” or “What caused this event or that natural feature to arise?” rather than questions of the form, “How does nature normally operate or function?” or “What causes this general phenomenon to occur?” Those who postulate the past activity of an intelligent designer do so as an answer, or a partial answer, to distinctively historical questions. The theory of intelligent design attempts to answer a question about what caused certain features in the natural world to come into existence—such as the digitally encoded, specified information present in the cell. It attempts to answer questions of the form “How did this natural feature arise?” as opposed to questions of the form “How does nature normally operate or function?” 

A Distinctive Form of Inference

The historical sciences use inferences with a distinctive logical form. Unlike many nonhistorical disciplines, which typically infer generalizations or laws from particular facts (induction), historical sciences employ abductive logic to infer a past event from a present fact or clue. Such inferences are also called “retrodictive.” As Gould put it, the historical scientist infers “history from its results.” Inferences to intelligent design exemplify this abductive and retrodictive logical structure. They infer a past unobservable cause (in this case, an instance of creative mental action or agency) from present facts or clues in the natural world, such as the specified information in DNA, the irreducible complexity of certain biological systems, and the fine-tuning of the laws and constants of physics. 

A Distinctive Type of Explanations 

Historical sciences usually offer causal explanations of particular events, not lawlike descriptions or theories describing how certain kinds of phenomena—such as condensation or nuclear fission—generally occur. In historical explanations, past causal events, not laws or general physical properties, do the main explanatory work. To explain a dramatic erosional feature in eastern Washington called the Channeled Scablands, a historical geologist posited an event: the collapse of an ice dam and subsequent massive flooding. This and other historical scientific explanations emphasize past events as causes for subsequent events and/or present features of the world. 

The theory of intelligent design offers such a distinctively historical form of explanation. Theories of design invoke the act or acts of an agent and conceptualize those acts as causal events, albeit ones involving mental rather than purely physical entities. Advocates of design postulate past causal events (or a sequence of events) to explain the origin of present evidence or clues, just as proponents of chemical evolutionary theories do. 

Use of the Method of Multiple Competing Hypotheses

Historical scientists do not mainly test hypotheses by assessing the accuracy of the predictions they make under controlled laboratory conditions. Using the method of multiple competing hypotheses, historical scientists test hypotheses by comparing their explanatory power against that of their competitors. And advocates of the theory of intelligent design use this method. In sum, the theory of intelligent design seeks to answer characteristically historical questions, it relies upon abductive/retrodictive inferences, it postulates past causal events as explanations of present evidence, and it is tested indirectly by comparing its explanatory power against that of competing theories. Thus, the theory of intelligent design exhibits each of the main features of a historical science, suggesting another reason to regard it as scientific. 

Reason 5: ID Addresses a Specific Question in Evolutionary Biology

There is another closely related reason to regard intelligent design as a scientific theory. It addresses a key question that has long been part of historical and evolutionary biology: How did the appearance of design in living systems arise? Both Darwin and contemporary evolutionary biologists such as Francisco Ayala, Richard Dawkins, and Richard Lewontin acknowledge that biological organisms appear to have been designed. Nevertheless, for most evolutionary theorists, the appearance of design is considered illusory, because they are convinced that the mechanism of natural selection acting on random variations (and/or other similarly unguided mechanisms) can fully account for the appearance of design in living organisms. 

In On the Origin of Species, Darwin sought to show that natural selection has creative powers comparable to those of intelligent human breeders. In doing so, he sought to refute the design hypothesis by providing a materialistic explanation for the origin of the appearance of design in living organisms. Following Aleksandr Oparin, chemical evolutionary theorists have sought to provide similarly materialistic accounts for the appearance of design in the simplest living cells. 

Is the appearance of design in biology real or illusory? Clearly, there are two possible answers to this question. Neo-Darwinism and chemical evolutionary theory provide one answer, and competing theories of intelligent design provide an opposite answer. By almost all accounts the classical Darwinian answer to this question—“The appearance of design in biology does not result from actual design”—has long been considered a scientific proposition. But what is the status of the opposite answer? If the proposition “Jupiter is made primarily of methane gas” is a scientific proposition, then the proposition “Jupiter is not made primarily of methane gas” would seem to be a scientific proposition as well. The negation of a proposition does not make it a different type of claim. Similarly, the claim “The appearance of design in biology does not result from actual design” and the claim “The appearance of design in biology does result from actual design” are not two different kinds of propositions; they are two different answers to the same question, a question that has long been part of evolutionary biology and historical science. If one of these propositions is scientific, then the other is scientific as well. 

Reason 6: ID Is Supported by Peer-Reviewed Scientific Literature

Critics of the theory of intelligent design often claim that its advocates have failed to publish their work in peer-reviewed scientific publications. For this reason, they say the theory of intelligent design does not qualify as a scientific theory. According to these critics, science is what scientists do. Since ID scientists don’t do what other scientists do—namely, publish in peer-reviewed journals—they are not real scientists and their theory isn’t scientific either. 

Critics of the theory of intelligent design made this argument before and during the Dover trial in support of the ACLU’s case against the Dover school-board policy. For example, Barbara Forrest, a philosophy professor from Southeastern Louisiana State University and one of the expert witnesses for the ACLU, asserted in a USA Today article before the trial that design theorists “aren’t published because they don’t have any scientific data.” In her expert witness report in support of the ACLU, Forrest also claimed that “there are no peer-reviewed ID articles in which ID is used as a biological theory in mainstream scientific databases such as MEDLine.” Judge Jones apparently accepted such assertions at face value. In his decision, he stated not once, but five separate times, that there were no peer-reviewed scientific publications supporting intelligent design. 

But Dr. Forrest’s carefully qualified statement gave an entirely misleading impression. In 2004, a year in advance of the trial, I published a peer-reviewed scientific article advancing the theory of intelligent design in a mainstream scientific journal. As I mentioned in the Prologue, the publication of the article evoked a huge backlash at the Smithsonian Institution, where the journal, The Proceedings of the Biological Society of Washington, was published. Moreover, controversy about the editor’s decision and his subsequent treatment spilled over into both the scientific and the mainstream press, with articles about it appearing in Science, Nature, the Wall Street Journal, and the Washington Post among other places. Both Dr. Forrest and Judge Jones had every opportunity to inform themselves about the existence of at least one peer-reviewed scientific article in support of intelligent design. 

In any case, as Discovery institute informed the court in an amicus curiae (friend of the court) brief, Meyer article was by no means the only peer-reviewed or peer-edited scientific publication in support of the theory of intelligent design. By 2005, scientists and philosophers advocating the theory of intelligent design had already developed their theory and the empirical case for it in peer-reviewed scientific books published both by trade presses and by university presses. Michael Behe’s groundbreaking Darwin’s Black Box was published by the Free Press in New York. William Dembski’s The Design Inference was published by Cambridge University Press. Both were peer-reviewed. In addition, design proponents have also published scientific articles advancing the case for intelligent design in peer-reviewed scientific books and anthologies published by university presses and in scientific conference proceedings published by university presses and trade presses. Advocates of intelligent design have also published work advancing their theory in peer-reviewed philosophy of science journals and other relevant interdisciplinary journals. Moreover, since the publication of Meyer article in 2004, several other scientific articles supporting intelligent design (or describing research guided by an ID perspective) have been published in mainstream peer-reviewed scientific journals. 
Of course, critics of intelligent design may still judge that the number of published books and articles supporting the theory does not yet make it sufficiently mainstream to warrant teaching students about it. Perhaps. But that is a judgment about educational policy distinct from deciding the scientific status, or still less, the merits of the theory of intelligent design itself. Clearly, there is no magic number of supporting peer-reviewed publications that suddenly confers the adjective “scientific” on a theory; nor is there a tribunal vested with the authority to make this determination. If there were a hard-and-fast numerical standard as low as even one, no new theory could ever achieve scientific status. Each new theory would face an impossible catch-22: for a new theory to be considered “scientific” it must have appeared in the peer-reviewed scientific literature, but anytime a scientist submitted an article to a peer-reviewed science journal advocating a new theory, it would have to be rejected as “unscientific” on the grounds that no other peer-reviewed scientific publications existed supporting the new theory. 

Critics of intelligent design have actually used a similarly circular kind of argument to claim that ID is not science. Before 2004, critics argued that the theory of intelligent design was unscientific, because there were no published articles supporting it in peer-reviewed scientific journals (ignoring the various peer-reviewed books that existed in support of ID). Then once a peer-reviewed scientific journal article was published supporting intelligent design, critics claimed that the article should not have been published, because the theory of intelligent design is inherently unscientific. Indeed, critics accused the editor who published my article of editorial malfeasance, because they thought he should never have considered sending the article out for peer review in the first place. Why? Because, according to these critics, the perspective of the article should have immediately disqualified it from consideration. In short, critics argued that “intelligent design is not scientific because peer-reviewed articles supporting the theory have not been published” and that “peer-reviewed articles supporting intelligent design should not be published because the theory is not scientific,” apparently never recognizing the patent circularity of this self-serving, exclusionary logic. 

Logically, the issue of peer review is a red herring—a distracting procedural side issue. The truth of a theory is not determined or guaranteed by the place of, or procedures followed, in its publication. Many great scientific theories were first advanced and published without undergoing formal peer review. Though modern peer-review procedures often do a good job of catching and correcting factual mistakes, they also can enforce ideological conformity, stifle innovation, and resist novel theoretical insights. Scientific experts can make mistakes in judgment and, being human, they sometimes reject good new ideas because of prejudicial attachments to older, more familiar ones. The history of science is replete with examples of established scientists summarily dismissing new theories that later proved able to explain the evidence better than previously established theories. In such situations, proponents of new theories have often found traditional organs of publication closed to them. Thus, it is neither surprising nor damning to intelligent design that currently many scientific journals are implacably opposed to publishing articles supporting the theory. 

Yet if science is what scientists do, and if publishing peer-reviewed scientific books and articles is part of what scientists do that makes their theories scientific (as critics of ID assert), then there is another good, convention-dependent reason to regard intelligent design as scientific. The scientists who have developed the case for intelligent design have begun to overcome the prejudice against their ideas and have published their work in peer-reviewed scientific journals, books, conference volumes, and anthologies. 


As I examined the question of whether intelligent design qualified as a scientific theory, it was clear to me that the answer to this question depended upon the definition of science chosen to decide the question. But as I considered both common definitions of science and what I had learned about the specialized methodological practice of the historical sciences, it seemed equally clear that there were many good—if definition-dependent—reasons for considering intelligent design as a scientific theory. 

But maybe there was some other, better definition of science that I should have considered. Perhaps there was some specific feature of a scientific theory that intelligent design did not possess, or some specific criterion of scientific practice that its advocates did not follow. I knew that the theory of intelligent design met the criterion of testability, despite what many critics of the theory asserted, but perhaps there were other criteria that it could not meet. If so, then perhaps these definitional criteria would establish a good reason for disqualifying intelligent design from consideration as science after all. Certainly, many critics of intelligent design have argued that the theory lacks many key features of a bona fide scientific theory—that it fails to meet criteria by which science could be defined and distinguished from nonscience, metaphysics, or religion. In the next chapter, I examine why critics of the theory—including the judge in the Dover case—have insisted that, despite the arguments developed in this chapter, intelligent design does not qualify as a scientific theory.