How Intelligent Is Intelligent Design?
Letters in Response to Steve Meyer's "DNA and Other Designs" First ThingsPhillip E. Johnson writes:
Stephen C. Meyers article ”DNA and Other Designs“ (April) captures the heart of the scientific case against the materialist ideology that rules biology. Neither physical laws nor chance can write meaningful text (complex specified information). Chance produces only meaningless disorder, and law produces only simple repetition. That is why no one has ever observed natural selection or any other natural process creating new genetic information by a combination of law and chance; it is every bit as impossible as a perpetual motion machine. Professor Meyers article will produce angry and baffled responses not because there is any real objection to the logic, but because the aim of biology in the era of Darwin has been to support a materialist worldview rather than to investigate the data impartially. Thanks to Stephen Meyer and to First Things for helping to bring that era to a close.
Phillip E. Johnson
Boalt Hall (Law School)
University of California
Berkeley, California
Kenneth R. Miller writes:
Stephen C. Meyers article DNA and Other Designs presents, no doubt unintentionally, a compelling case for the intellectual desperation of the Intelligent Design movement.
Professor Meyer seems to think that if he can show that evolutionary biologists have not yet solved the many problems associated with the origin of life, then he can claim that intelligent design is a fair alternative. Curiously, however, he then ignores nearly two decades of research on this very subject, failing to tell his readers of experiments showing that very simple RNA sequences can serve as biological catalysts and even self-replicate. He follows with elaborate probability calculations showing that sequence-specific large protein molecules would be unlikely to self-assemble on the primitive earth, overlooking the fact that no scientist has ever proposed that today’s highly evolved proteins were assembled in such fashion.
A reasonable person might well argue, despite strong evidence to the contrary, that the first living forms on planet Earth might have been the product of some act of intelligent design. Prof. Meyer, however, insists on going much further, and here his thesis stumbles from unlikely speculation into demonstrable error. Waving Michael Behes banner of irreducible complexity, he flatly states that Darwinian processes cannot produce complex biochemical systems. This claim is simply wrong, as even a cursory search of the biochemical literature will establish.
Lastly, Prof. Meyer allies himself with the curious informational arguments of William A. Dembski, who asserts that the informational complexity of DNA can be explained only by the direct, intelligent design of DNA sequences. Dembskis theories may be attractive, but only to those who are unfamiliar with the wealth of direct, experimental evidence on the mechanisms of molecular evolution. When one can watch new genes, with novel biochemical capabilities, evolve under controlled laboratory conditions, the claim that DNA sequence complexity requires intelligent design vanishes into thin air.
Why does Prof. Meyer find it necessary, despite overwhelming evidence for evolution, to take arms against Darwin? Quite clearly, as he puts it, what is at stake is nothing less than free will, meaning, purpose, and God. Stirring words, to be sure, but absolute logical nonsense. Free will has never been threatened by evolution, and issues of meaning, purpose, and Divinity clearly lie beyond the reach of science, even the Darwinian science he fears so much. If Stephen Meyer wishes to argue for the importance of these values he will have many allies, including John Polkinghorne, Paul Davies, Ian Barbour, and me. However, when he and his small group of friends claim scientific status for their attacks on evolution, the readers of FIRST THINGS would be well advised to inquire exactly how many papers have appeared in the scientific literature to document intelligent design. The answer will surprise no one who has followed the development of Prof. Meyers ideas. None.
Kenneth R. Miller
Professor of Biology
Brown University
Providence, Rhode Island
Bob Puharic writes:
Stephen C. Meyers article on design in nature merely confirms an observation that we scientists have about design theory: theres no there there. Professor Meyer spends lots of time telling us what didn’t happen via evolution. In fact he spends so much time that he can even calculate the odds against evolution happening. What he doesnt tell us is what did happen. To say that design produces order because computer programmers create information ignores the fact that we can hire programmers. If Prof. Meyer is suggesting we can hire God, that’s a pretty neat trick. Design theory isn’t science for one simple reason: there is no mechanism for design. Prof. Meyer spends all of his article saying what didn’t happen, as if that’s proof of design. What it proves is that Prof. Meyer, like William A Dembski before him, is parading ignorance as science.
Bob Puharic
Whitehall, Pennsylvania
Donald J. Wink writes:
Stephen C. Meyer outlines well one of the great voids in materialist science: an explanation of how DNA arose as the genetic code. Professor Meyer does a good job of documenting that an inference of design is reasonable when we talk about the origins of the DNA-based biota we know now. Certainly chance is not acceptable if we feel we need to explain how DNA or related RNA life forms could have risen, full-blown, as a coding system. But, as a chemist, I think there are flaws in his argument, both at the level of basic chemistry and in philosophical terms.
Prof. Meyer is implying more than he should about the way we understand the operation of the coding system in the evolution of species, including our own. Few scientists feel there is a problem in understanding how chance mutations and natural selection explain changes in the genetic code. Chance, not design, is also a reasonable inference when we notice that the genetic code of many organisms, including our own, includes astounding waste and uneconomical variation. The full genome does have high information content, but it also has high nonsense content. Is nonsense (including repeats and stretches that do not code for anything used in that organism) a sign of intelligence, or of chance?
To the extent that Prof. Meyers work focuses on origin of biotic systems from prebiotic ones, design could be an explanation. But I sense that he is building on the same error that is behind arguments based on origin-by-chance: that we can consider the current coding system as a clear indication of the system that was present during the original change from prebiotic to biotic systems. Our current genetic system is based on some robust chemistry, far more robust and successful than any cruder original hereditary system may have been. In other words, we may be looking at only the most successful system, not the original system. And the genetic code leaves no fossils. This is a very frustrating part of a lot of good science: we cannot always access the basic mechanism of a particular event. The answer, in those cases, is to look for more creative probes of the question, not to overemphasize one mechanism that, while consistent with the data, possesses little of the predictive power of a good scientific theory.
I am always wary of arguments that infer anything by induction, and that is what I think Prof. Meyer does when he writes, Because mind or intelligent design is a necessary cause of an informative system, one can detect the past action of an intelli¬gent cause. All we can say, as scientists, is that mind or intelligent design is the only origin of complex coding that we know of.
Indeed, an inductive counterargument could be advanced for chance in the original code. It is the argument I think most materialists would cling to, and I do not think Prof. Meyer addresses it. Specifically, since chance is generally accepted as adequate to explain shifts in that coding, there is a strong inductive reason to think chance may, in ways we cannot fathom right now, have played a role in origin. In my opinion this induction of chance, and his induction of design, are equally problematic.
In the end I think the greatest danger of Prof. Meyers view is that it invites scientists to conduct experiments that falsify the design theory by coming up with even one counterexample. These scientists would then claim that they had proven that there was no need for design. Well, then what? If a chance-based example of an original code were to appear, would our faith in Gods design be rendered false? I for one think not, because I do not see that we need to have science on our side to justify our faith in design.
If faith is indeed hope in things–including God’s designs–that are unseen, then we should not worry if science cannot see what we know in faith. Faith in things invisible to modern science is important, and we must not anchor our faith on the dis¬coveries of science, nor abandon it to the possibility of scientific refutation.
Donald J. Wink
Associate Professor of Chemistry
University of Illinois at Chicago
Edmurid Weinmann writes:
In DNA and Other Designs, Stephen C. Meyer writes that the probability of constructing a rather short, functional protein at random becomes so small (1 chance in 10,125) as to approach the point at which appeals to chance become absurd even given the probabilistic resources of our multibillion-year-old universe.
However, no one yet has attempted to estimate what the size of those resources might be, and this is the major flaw, as I see it, in the argument from improbability, which has replaced the old argument from design. The larger those resources might be, the less improbable the occurrence of the protein becomes. So the improbability is dependent on ones reckoning of the “probabilistic resources.”
Hence, while the chance for a protein being constructed by chance may be very, very small, the difference between those who accept such an argument from improbability and those who do not will lie not in their recognition of how small the chance is, but in their personal reckoning of the probabilistic resources of our universe over its apparent lifespan, which is currently some twelve to fourteen billion years.
It is the appreciation of the size and scale of the universe and of our place in it–brought about by the development of modern science in just the last four centuries of human history–that is really responsible for the decline in the popular plausibility of theistic arguments from design or improbability. We simply did not have the means of reckoning–of getting anything like an accurate personal sense–of our place within the scope of things. We on our planet could scarcely have helped considering ourselves the crown jewel in the divine creation, given how little we knew of our world compared to what we know now.
Edmurid Weinmann
Huntington, New York
Stephen C. Meyer replies:
Kenneth Miller’s letter makes only one point relevant to my central argument. In his second paragraph he faults me for failing to tell [my] readers of experiments showing that very simple RNA sequences can serve as biological catalysts and even self-replicate. But Professor Miller fails to tell readers the whole story. In my essay I argued that intelligent design constitutes the best (most causally adequate) explanation for the origin of the genetic information necessary to produce a living cell in the first place. To support this claim I argued that all major classes of naturalistic explanation, whether based upon chance, physical-chemical necessity, or some combination of the two, fail to explain how information-rich biomacromolecules could have arisen from pre-biotic (nonliving) chemistry.
The RNA world scenario that Prof. Miller cites does not solve this problem. Indeed, the RNA world was never proposed as an explanation for the sequencing or information problem (thus my silence about it). Rather it was proposed as an explanation for the origin of the interdependence of nucleic acids and proteins in the cell’s information-processing system. In extant cells, building proteins requires genetic information from DNA, but information on DNA cannot be processed without many specific proteins and protein complexes. This poses a chicken-or-egg problem. The discovery that RNA (a nucleic acid) possesses some limited catalytic properties (similar to those of proteins) suggested a way to solve this problem. RNA first advocates proposed an early state in which RNA performed both the enzymatic functions of modern proteins and the information storage function of modern DNA, thus allegedly making the interdependence of DNA and proteins unnecessary in the earliest living system.
Nevertheless, recent scientific publications detail many fundamental difficulties with the RNA world scenario. First, naturally occurring RNA possesses very few of the specific enzymatic properties of the proteins that are necessary to extant cells. Second, RNA world advocates offer no plausible explanation for how primitive RNA replicators might have evolved into modern cells that do rely (almost exclusively) on proteins to process genetic information and regulate metabolism. Third, attempts to enhance the limited catalytic properties that RNA molecules do have inevitably have involved extensive investigator manipulation in so-called ribozyme engineering experiments, thus simulating, if anything, the need for intelligent design, not the adequacy of an undirected chemical evolutionary process. Fourth, synthesizing (and/or maintaining) many essential building blocks of RNA molecules under realistic conditions has proven either difficult or impossible. Further, the chemical conditions required for the synthesis of ribose sugars are decidedly incompatible with the conditions required for synthesizing nucleoside bases. Yet both are necessary constituents of RNA
In any case, the RNA world presupposes, but does not provide, a solution to the sequence specificity or information problem addressed in my article. For strands of RNA to perform enzymatic functions (including enzymatically mediated self-replication) they must, like proteins, have very specific arrangements of constituent building blocks (in this case, their nucleoside bases). Further, they must be long enough to fold into complex three-dimensional shapes (to form so-called tertiary structure). Thus, any RNA molecule capable of enzymatic function must have the same property of sequence specificity that DNA and proteins do. Indeed, such molecules must possess considerable (specified) information content.
Nevertheless, explaining how the building blocks of RNA might have arranged themselves into functionally specified sequences has proven no easier than explaining how the constituent parts of DNA might have done so, especially given the high probability of destructive cross reactions between desirable and undesirable molecules in any realistic pre-biotic soup. Further, for a single-stranded RNA-catalyst to self-replicate it must find an identical RNA molecule in close vicinity to function as a template, since a single RNA molecule cannot function as both enzyme and template. It has been shown, however, that to have a reasonable chance of finding two identical RNA molecules of a length sufficient to perform enzymatic functions would require an RNA library of some 1054 RNA molecules. The mass of such a library vastly exceeds the mass of the earth, suggesting the extreme implausibility of any chance origin of a primitive replicator system. Yet one cannot invoke natural selection to explain the origin of primitive replicators, since natural selection ensues only after self-replication has arisen. Likewise, RNA bases, like DNA bases, do not manifest self-organizational bonding affinities that can explain their specific sequencing. In short, the same kind of evidentiary and theoretical problems emerge whether one proposes that genetic information arose first in RNA or DNA molecules.
Prof. Miller dismisses my elaborate probability calculations showing that sequence-specific large protein molecules would be unlikely to self-assemble on the primitive earth, complaining that I overlook the fact that no scientist has ever proposed that today’s highly evolved proteins were assembled in such fashion. Here he makes a concession, but states it as an objection. Prof. Miller recognizes, as almost all origin-of-life researchers do, that proteins, RNA, and DNA are far too complex and specified to have arisen by chance given the time available from the cooling of the early earth (or even from the big bang) to the first appearance of life on earth. I included the calculations that I did, not to critique a straw man, but to illustrate why this consensus against chance emerged during the 1960s. (Before that time some scientists, notably George Wald, did, contra Prof. Millers claim, advocate chance alone.)
With the one exception noted above, therefore, Prof. Miller does not dispute my empirical argument. Instead, he responds mainly by critiquing Michael Behe’s argument from irreducible complexity and by defending Darwin’s theory of biological (not chemical) evolution. He claims that I flatly state that Darwinian processes cannot produce complex biochemical systems. But I did not. Behe claims that, or at least something close to it. While I happen to agree with Behe, Prof. Miller cannot refute my argument by citing arguments that allegedly undermine Behe’s. (I will leave rebutting Prof. Millers critique of Behe to Behe, but interested readers may want to know that Behe has responded specifically to the papers Prof. Miller depends upon at several recent academic conferences. Discovery Institute has also posted Behes responses to Prof. Miller, to the papers he depends upon, and to his other critics on its website at www.crsc.org [also accessible at www.discovery.org].)
Prof. Miller similarly fails to address my argument when he claims that one can watch new genes, with novel biochemical capabilities, evolve under controlled laboratory conditions. Prof. Miller thinks recent work has established this possibility and thus states that the claim that DNA sequence complexity requires intelligent design vanishes into thin air. Yet Prof. Miller’s wording shows that he has confused the issue. My article challenged the claim that undirected natural processes sufficed to explain the original genetic information necessary to the first living cell (starting in a pre-biotic environment). To talk about new genes evolving novel capabilities implies the existence of pre-existing genes in an already living organism. Yet genes by definition contain large amounts of genetic information. Thus, citing examples that putatively document how pre-existing genes can evolve new genetic information, as Prof. Miller does, only assumes the point at issue in my article, namely, the origin of biological information in the first place.
Prof. Miller also objects to my assessment of the philosophical implications of Darwinism and other nineteenth-century materialistic theories of origin. He insists that science can say nothing about purpose. Yet curiously his own textbook asserts that Evolution works without either plan or purpose and is random and undi¬rected (emphasis in original).
Prof. Miller closes by asking exactly how many papers have appeared in the scientific literature to document intelligent design? None, he answers. But his answer assumes the dominant theoretical framework at issue. As molecular biologist Scott Minnich has noted, scientists who are sympathetic to the design hypothesis see considerable evidence of intelligent design in the scientific literature. Readers might consult, for example, the February 1998 special issue of Cell devoted to the description of biomolecular machines. The question is not whether scientific papers document evidence of intelligent design. Many scientists now think that they do. The more important question is: Given the ideological climate evidenced by Kenneth Millers letter, will scientific journals allow well-credentialed scientists who do see evidence of intelligent design to say so openly in their scientific papers? On this question, the jury is–sadly–still out.
Bob Puharics letter raises an objection that I addressed directly in my essay. He casts my argument as an argument from ignorance because I critique naturalistic explanations for the origin of information without tell[ing] us what did happen. As he puts it, Prof. Meyer spends all of his article saying what didn’t happen, as if that’s proof for design. Yet I do say what happened. Specifically, I argued that an intelligent cause operated [acted] in the past to produce the information necessary to the origin of life.
Moreover, I did not present my argument for intelligent design as a proof, but rather as an inference to the best explanation. Such arguments do not commit the fallacy of arguing from ignorance. Arguments from ignorance claim that because’ some proposition X has been disproved, some proposition Y has, therefore, been proved. Best explanation arguments do not commit this fallacy, since 1) they do not claim proof, and 2) they do not just provide negative argument against a range of possible hypotheses (say, A, B, and C), but also positive argument for some alternative hypothesis (say, D, where A-D represent the set of hypotheses currently under consideration). In this case, I not only critiqued the causal adequacy of three broad types of naturalistic explanations for the origin of information, but also discussed the known (positive) causal powers of intelligent agents.
Thus, I noted in my article that design theorists infer [intelligent] design not just because natural processes cannot explain the origin of biological systems, but also because we know that intelligent agents can produce information-rich sequences. As Henry Quastler, an early pioneer in the application of information theory to molecular biology, recognized, the creation of new information is habitually associated with conscious activity. Quastler’s observation suggests conscious activity or intelligent design as at least a possible explanation for the origin of information. If, as I argued in my article, no other adequate causal explanation exists, then intelligent design also stands as the best explanation. Moreover, we routinely make such inferences to intelligent design because of our knowledge, not our ignorance, of what agents as well as various types of naturalistic entities can and cannot produce.
Edmund Weinmann and Donald Wink challenge my critique of the chance hypothesis. Weinmann argues correctly that the larger those [probabilistic] resources might be, the less improbable the occurrence of the protein becomes. He then asserts that no one has yet attempted to estimate what the size of those resources might be. Unfortunately, this statement is incorrect. In the sixth chapter of The Design Inference (1998), William A. Dembski, following earlier probability theorists such as Emile Borel, does calculate the probabilistic/specificational resources of the universe. His calculations imply that chance can be eliminated as a plausible explanation for specified systems of small probability, where small, on the cosmological scale, is less than 1/2 x 1/10150. Dembskis work on chance elimination merely refines and develops already well-accepted statistical methods and procedures. It also simply makes explicit the implicit logic that origin-of-life biologists have employed to eliminate the chance hypothesis since the late 1960s.
In addition to concerns about my assessment of the chance hypothesis, Donald Wink objects to the inductive basis of my argument. While I would characterize the overall structure of my argument as an inference to the best explanation, Professor Wink is correct that such arguments do depend upon generalizations about causal powers derived from (often) repeated experience, that is, from induction. Nevertheless, it does not follow that identifying an inductive basis for an argument makes it inherently weak or suspect as fur as practical reasoning is concerned. When assessing empirical arguments the question is not, Is this argument based upon inductive reasoning? Most scientific arguments are. Indeed, most laws of physics have been accepted on essentially inductive grounds. We do not distrust them–practically speaking–for that. Instead, the question should be, Are the inductive grounds for this argument strong or weak? Prof. Wink effectively concedes that the inductive basis for my argument is strong when he states that all we can say, as scientists, is that mind or intelligent design is the only origin [cause] of complex coding that we know of. Indeed, I commend intelligent design as the best explanation for the encoded information in DNA for this very reason.
Nevertheless, Prof. Wink expresses concern about the provisional nature of empirically based scientific arguments and cautions against basing religious faith on them. He worries that scientists may later prove that there was no need for design and asks, Then what? Well, clearly, a decisive refutation of the DNA to Design argument would leave theists with one less reason for their belief in an intelligent designer. Conversely, if the argument stands, theists will have one more such reason. Nevertheless, those who insist that faith does not require reasons should have no objection to marshaling evidence and argument that may support some aspect of theistic belief, as long as the conclusions are warranted by the evidence. Such arguments would, at best, provide welcome but unnecessary rational support for theistic belief. At worst, the refutation of such arguments would leave theists in exactly the same position that the fideists say is sufficient, that is, without rational support for their belief. In short, nothing is lost by trying. In any case, consequentialist arguments about what does or does not promote faith do not bear on the status of arguments for (or against) intelligent design. Thus, design theorists are committed to following the evidence wherever it leads, even if it happens to lead to a conclusion with theistic implications.
Finally, I thank Phillip E. Johnson for his supportive comments.