I am seeking a comprehensive and rigorous analysis of the concept of Intelligent Design (ID) in order to determine whether it meets the essential requirements to be considered a scientific theory. I am particularly interested in gathering insights from the philosophy of science perspective.

Intelligent Design proposes that certain features of the natural world are best explained by an intelligent cause rather than an undirected process such as natural selection. While proponents argue that ID should be recognized as a scientific theory, critics often assert that it lacks scientific legitimacy.

To address this question, I would appreciate detailed and well-supported explanations regarding the essential characteristics and criteria that a scientific theory must fulfill. This would include but is not limited to:

  1. Falsifiability: Does Intelligent Design make testable predictions that, if proven false, would contradict its central claims?

  2. Empirical evidence: What empirical evidence supports or refutes the claims made by Intelligent Design?

  3. Methodology: Does Intelligent Design follow established scientific methodologies, such as proposing hypotheses, conducting experiments, and engaging in peer review?

  4. Consilience: Does Intelligent Design integrate with other scientific fields and theories, fostering a cohesive and interconnected body of knowledge?

  5. Naturalistic explanations: How does Intelligent Design address the requirement of offering naturalistic explanations, which is commonly expected within scientific inquiry?

  6. Consensus within the scientific community: What is the general opinion among experts in relevant scientific disciplines regarding the recognition of Intelligent Design as a scientific theory?

I kindly request that answers be framed within the context of the philosophy of science and be as rigorous and objective as possible.

EDIT: I disagree that this question is a duplicate of this one. You can learn more about my reasons in this meta question: Can question X be on-topic, question Y be off-topic, and X be closed as a duplicate of Y?

  • Comments have been moved to chat; please do not continue the discussion here. Before posting a comment below this one, please review the purposes of comments. Comments that do not request clarification or suggest improvements usually belong as an answer, on Philosophy Meta, or in Philosophy Chat. Comments continuing discussion may be removed.
    – Philip Klöcking
    Commented Jun 4, 2023 at 21:17
  • 1
    Are the 14 Answers satisfactory?
    – Scott Rowe
    Commented Jun 9, 2023 at 19:41

13 Answers 13


These two answers to prior questions spell out almost everything you have asked here:


  1. Yes, but ID advocates don't highlight them. They have mostly been proven false by evolutionary advocates who specified them on TalkOrigins.

  2. The imperfect optimization of life refutes omni design, and the lack of common structures and DNA refutes common design.

  3. The ID community maintains peer review journals and publishes papers. There are hypotheses too, such as specified complexity as a measure of step size that evolution can achieve, and irreducible complexity as the steps needed to be achieved by evolution (these two theoretical points are among the best products of the movement). The community is light on experiments, and particularly light on looking for refuting data.

  4. ID attempts to embrace the established conclusions of science in other fields, the forms of science, and its commonality with other design-focused science programs like anthropology and SETI. So this is attempted.

  5. The whole point of ID is that applying methodological naturalism IS a naturalistic explanation. Naturalism does not pre-specify ontology, so agency IS a possible outcome of a naturalistic explanation.

  6. The science community treats ID as a politically driven pseudo-science program, attempting to subvert science by miming its outward forms for the purely propagandistic of objective of smuggling religion into school systems. The philosophers of science such as myself who try to look beyond the ad hominem reasoning that drives most science response, agree that ID COULD be a scientific programme. BUT -- lots of potentially "science" subjects become non-science when their advocates hold by them despite blatant refutations. The KEY to being science is not the outward forms, but the openness to and willingness to search for refutations. And the ID community is NOT open to reconsidering its basic assumptions, nor looking for refutations. By miming science forms, and claiming to be science, but rejecting this key attitudinal approach, in practice ID is explicitly a pseudoscience movement.


Commonly "Intelligent Design" does not denote anything theoretic, but a political anti-scientific movement aiming to replace science teaching in schools by teaching of religion.

The theoretic parts of ID do not form a single theory, but instead are a dynamic accumulation of ideas, based on the presumption that in the history of mankind at some point a non-earthly power intervened to shape mankind or life or both, and that if we keep looking hard enough, we will find evidence of that, and refute the naturalistic theory of evolution that claims that abiogenesis and evolution species happened as a result of only natural selection (and human breeding efforts maybe).

ID is not a single theory, as demonstrated like this: One part of ID claims that some bacterial flagellum exhibits irreducible complexity not explainable by natural selection. If this was refuted, would ID be refuted?

No, because ID can then add a part that some blood clotting cascade exhibits irreducible complexity not explainable by natural selection. If this was refuted, would ID be refuted?

No, because ID can then add a part that the cambrian explosion exhibits diversity not explainable by natural selection. If this was refuted, would ID be refuted?

No, because ID can then add a part that the universe was fine-tuned to support life which requires an intelligent designer. If this was refuted, would ID be refuted?

No, as ID can keep adding new claims about any part of history not being explainable by natural selection.

There is no central claim or prediction that if refuted would refute Intelligent Design.

Publications about ID are not integrated well with actual science fields. While ID authors will cite scientific publications to add weight to their arguments, ID authors are not being cited. That means ID is not adding value to other fields of science. There are no experiments conducted by ID authors that have advanced our understanding of nature or history.

Since ID proponents reject the findings of mainstream science, their writing is divisive and incongruent with the rest of science, while being congruent with Creationist religion.

This is all well recognized by experts in other scientific disciplines.


Allow me to get properly philosophical here and point out that the phrase 'scientific theory' is at best a red herring (and at worst a holdover from mid-20th century scientism). Theory is proposed explanation (a belief assertion); science is a system for evaluating proposed explanations. Trying to talk about 'scientific theories' is like trying to talk about 'un-dug holes'. People can argue till the cows come home about what they're going to find when they dig a hole (that's theory), but no one is going to know unless they crack out the shovels (that's science).

Intelligent design is a perfectly valid theory, if that is the belief assertion one wants to make. But when we start digging for evidence we don't find a lot of intelligent designers, or a lot of designs that appear to be the products of active intelligence. Instead, we find what appears to be a slow drift of living things changing to optimize within ever-changing environments. Does that preclude the existence of an intelligent designer? No, of course not. But — if we are being properly scientific — it would necessarily change the nature of that intelligence, because that intelligence designed a system of eons-long experimentation in which humanity was by no means a necessary outcome, much less an apex or culmination of the process.

If Intelligent Design were taught in science classes as a science, only two outcomes would be possible:

  1. It would be treated as a failed theory, because of the extremely poor evidence in its favor and the vast amount of evidence that contradicts it.
  2. It would have to be so modified to work within the available evidence that it would lose all of the theological characteristics that make it attractive to the people who propose it.

We shouldn't worry about whether Intelligent Design is a proper 'scientific theory'. We should just be good scientists, and let theories prove themselves or hoist themselves on their own petards.


The elephant in the room is that ID apparently originated as pseudoscience:

Intelligent design (ID) is a pseudoscientific argument for the existence of God, presented by its proponents as "an evidence-based scientific theory about life's origins". Proponents claim that "certain features of the universe and of living things are best explained by an intelligent cause, not an undirected process such as natural selection." ID is a form of creationism that lacks empirical support and offers no testable or tenable hypotheses, and is therefore not science. The leading proponents of ID are associated with the Discovery Institute, a Christian, politically conservative think tank based in the United States. (from Wiki)

Not to mention the conceptual problem that intelligent design presupposes an "intelligent designer". And both words are important here, A) that it's a designer and B) that he/she/it is intelligent. Because if you take that for granted, then the whole quest of understanding "what is", "how it works" and "what will be next", is either futile (if the designer is still at it and can change natural laws on a whim) or at least some tedious version of reverse engineering compared to question of "WHY did the designer intended it like that". So as a surprise to no one this would rather hint at a more fruitful approach by a "psychology of the creator", who as he/she/it is so much more intelligent than us (who can't create on that level), and is thus more in the domain of theology and religion.

In terms of falsifiability, well if indeed faith comes first, you'd be out of luck. The thing is, the value of a scientific theory would stem from, at least partial, internal consistency (required to make it any useful in practical application), but cardinally from it's correlation with nature. So scientific experiments are rather looking for the edge cases, where no data exists, to find whether the correlation holds and whether the predictions turn out to be useful (within margins of error).

While in terms of faith the value of a conclusion is not contingent on it's internal consistency or external correlation, so even if you dispel an argument they could come up with another because their actual argument (usually something psychological) remains unstated and thus almost impossible to guess and consequently to disprove.

So that already would either mark a dead end in terms of the questions as to whether it's a science, because the entire sciency look&feel would be nothing more than a mask for scam OR you'd need to narrow it down significantly, need to get down to the nitty gritty of one or a few particular theories in that domain.

Like it's not ideal, but probably rather human that scientists are married to their hypothesis and it's not directly problematic as long as the arguments that they propose are based in science. So you can believe your hypothesis stronger than what you hold as evidence but the strength of your argument still relies on your evidence not your believes, charisma or whatnot. So it's technically possible to have proponents of ID that genuinely believe it but also operate on scientific principle. The thing is if you are correct, and even if it is by accident, nature would consequently behave as expected.

So there could theoretically be proponents of theories by design who operate genuinely according to the scientific method, despite the original ID proponents being pseudoscience.

Not to mention that this stripped down argument:

Intelligent Design proposes that certain features of the natural world are best explained by an intelligent cause rather than an undirected process such as natural selection.

is not universally false. Think of artifacts from previous cultures. Like archeologists probably regularly have to weight between hypotheses of design (indicating intelligent cultures) and those by natural processes (indicating interesting phenomena).

However these would be localized hypotheses, with testable predictions, while the broad claim of "it's complex therefor it needs to have originated from an intelligent design" is unfalsifiable because you can always hide the god in the gap. "Oh yeah what we called 'too complex' last week is actually rather simple, but this other thing that we currently don't understand is obviously far too complex and thus evidence of an intelligent design".


I'm surprised that nobody has mentioned the Discovery Institute's Wedge Document, which is interesting reading for anyone who wants to know whether to take Intelligent Design seriously.

The social consequences of materialism have been devastating. As symptoms, those consequences are certainly worth treating. However, we are convinced that in order to defeat materialism, we must cut it off at its source. That source is scientific materialism. This is precisely our strategy. If we view the predominant materialistic science as a giant tree, our strategy is intended to function as a "wedge" that, while relatively small, can split the trunk when applied at its weakest points....

This does not read as if ID has been created as a bona fide explanation of something in biology. Phillip Johnson, "the father of intelligent design", may have a valid point about the social consequences of materialism; social consequences do indeed form an part of sociology, but they are not part of biology, chemistry, or physics.

To see intelligent design theory as an accepted alternative in the sciences and scientific research being done from the perspective of design theory. ... To see design theory permeate our religious, cultural, moral and political life.

To summarize, ID is an exercise in social engineering, not a scientific theory, if you believe the people who created ID.

  • 1
    This was a really illuminating comment to read.
    – TKoL
    Commented May 13 at 13:14

Before we can answer the question, we first need to understand what it means for theory to be scientific. In my view, your framing of this question in six questions is not helpful, and a distraction from addressing the problem.

Science is a process. It is the process of understanding the world around us through observation and experiment. Scientific theories are generated by finding areas in which our current understanding fails to explain, predict, or describe some feature of the world we are interested in and then figuring out a theory which does. This theory, if it is to be scientific, needs to be based in our current knowledge and prior observation of the world. Obviously, we move forward by rejecting or refining early ideas, but we do not do so willy-nilly, a new theory should be at least as explanatory as ones that have gone before. New theories are then tested by developing experiments which will produce different results depending on whether the new theory or the old is true; and seeing what happens. For any substantial theory, a great many experiments and observations are required to generate confidence in a theory.

Intelligent Design, in total contrast, was scientifically known to be completely 100% false when it was first invented in the 1980s¹. There was no scientific problem or observation that inspired it, instead it was invented for political reasons in order to push religious ideology into classrooms by members of the Creation "Science" Movement. Clever rhetorical tricks like "irreducible complexity" were designed to fool layman into thinking that the scientific evidence was unsound, rather than to generate new knowledge. Those pushing the theory engaged in fake experimentation and sham publishing of papers in order to try and produce a false impression of scientific vigour whilst not actually doing anything of the sort.

As such Intelligent Design is not a scientific theory.

¹ Don't confuse the late 20th century idea called Intelligent Design with the more general concept that could have used that name.


(This answer was invited in chat by the question-asker.)

Your fourth point about consilience should be viewed in light of Quine (WP, SEP) and his view of science. Quine wrote a book The Web of Belief about what we now call confirmation holism. In this approach, theories and empirical experiments are not definable/executable in isolation, but within a context which includes the bulk of scientific knowledge.

For example, modern laboratory glassblowing requires understanding of (roughly from small to large) quantum mechanics, thermodynamics, chemistry, materials science, organic chemistry, and biology. Note that the glass problem is still an open question; scientific progress relies on glass, regardless of whether our model of glass is incomplete.

ID does not fulfill Quine's requirements. In particular, through the development of genetic algorithms and advances in statistical mechanics, we understand that goal-directed evolution need not be agentive. The differences between intelligent and non-intelligent design in biology will be exponentially small, and after millions of Earth years, any evidence of ID will have faded quite a bit. ID proponents therefore have a burden of proving not just that ID has substantiating evidence, but that that evidence points specifically to an agent performing husbandry or genetic engineering.

  • 2
    Glad to see you are contributing! I'd reconsider on the objection-orientation. I'd be curious to see your views.
    – J D
    Commented Jun 7, 2023 at 16:01
  • 3
    Nah, it's not like ID is wrong for only one reason. Like most wrongness, it's wrong for multiple reasons. I'm just putting this in a place that's searchable, instead of chat.
    – Corbin
    Commented Jun 7, 2023 at 16:08

The Problem with ID is Metaphysical

Does Intelligent Design fulfill the necessary criteria to be recognized as a scientific theory?

It depends on who you ask. The fact is there are philosophically two sides to this issue: creationists and neo-creationists who maintain yes, and everyone else, including theologians like the Pope of the Catholic church (who is a trained scientist), who say no. It is the position of the latter, much larger group that a case should be made why creationists calling their belief system 'science' are committing a category mistake. Therefore, the problem with ID is primarily a problem of metaphysics since the implication of supernatural causation cannot be falsified, has no empirical evidence, does not rely on methodological naturalism, is meant to be consilient with theological creationism, suggests a supernatural explanation, and therefore is largely in violation of major consensus within the scientific community.

Naturalism, Supernaturalism, and Equivocation on Science

One of the fundamental philosophical issues going back to the Ancient Greeks is the question of supernatural versus natural causation. Among the Pre-Socratics, the first and greatest philosophers in the West and the fathers of our tradition, there were theologi and physiki, those who admitted, and those who denied supernaturalism. So, it is the primary and fundamental presumption of natural science that there is no supernatural causation. That's built into what science is after thousands of years moving from naturalism introduced by the Pre-Socratics into natural philosophy and natural theology into modern science. According to natural science, there CANNOT be supernatural causes to physical events.

From a metaphysical standpoint, there CAN be supernatural causation. To many (I'd argue everyone before rational powers come to fruition), it seems obvious there can be. And this position is acceptable philosophically because it is generally take that metaphysical necessitation and explanation (SEP) is privileged and foundational. But if one says there is supernatural causation, then one is not doing natural science, which is what most people mean when the use the term science by itself.

The Rejection of Supernaturalism

It is the viewpoint of the majority in this debate, that creation science and theistic science are not natural science:

Natural science is one of the branches of science concerned with the description, understanding and prediction of natural phenomena, based on empirical evidence from observation and experimentation.5 Mechanisms such as peer review and repeatability of findings are used to try to ensure the validity of scientific advances.

The key to this is the term 'natural phenomena', and it is key because it excludes supernatural causation. Theistic science appeals to natural science to achieve proof of a supernatural agency, attacks the conclusions of natural science (evolution is the obvious target), and uses a term "Designer" which is an obvious synonym for Creator. And yet, the opposite of science isn't religion, because Catholicism is a religion that preaches science, and I'm sure there are atheists out there who believe in supernatural forces. Since conventional Christianity embraces science, this debate is not about religion versus science, but about naturalism versus supernaturalism, which is a core tenet of creationism and religious fundamentalism.

I have no science textbook that starts off, "Now, because we exclude supernatural causation....". It's taken for granted that the methodology of natural science excludes supernaturalism. In Ernst Mayr's This Is Biology, he devotes none of his book to refuting supernaturalism in biology. So, for a creationist to use the term science to describe their views is an attempt to poach the effectiveness of the method of natural science while maintaining a fundamental belief in supernaturalism. It is a category mistake to consider theistic science a form of naturalism. Creation "science" and theistic "science" are attempts (somewhat disingenuous by some) to move have all of the fideist certainty of fundamentalism and literal interpretations of Scripture without having to have the pesky skepticism and fallibilism (IEP) that comes with natural science. It is the viewpoint of many atheists, philosophers, and theologians alike that claiming that theistic science is a natural science is not only bad metaphysics, but is an obvious attempt to craft a pro-creationist doctrine to slip pass by the separation of church and state which is a philosophical prescription in political philosophy arising out of the Enlightenment. The US does not allow the teaching of theology in public science classes.

Methodological Naturalism

According to WP:

Methodological naturalism, this second sense of the term "naturalism", seeks to provide a framework of acquiring knowledge that requires scientists to seek explanations of how the world around us functions based on what we can observe, test, replicate and verify. It is a distinct system of thought concerned with a cognitive approach to reality, and is thus a philosophy of knowledge. It is a self-imposed convention of science that attempts to explain and test scientific endeavors, hypotheses, and events with reference to natural causes and events.

It should read "world around us functions based on what we can observe, test, replicate and verify by presuming there are only natural causes". It is a metaphysical necessity of natural science and natural theology that causes, as we understand them, are free of meddling as would be the case with sorcerers, ghosts, goblins, and Greek gods. Christians largely hold that the Christian God doesn't cavort about the earth like Zeus did, using magic, and disguising himself to interact with people. And it is not the claim of natural theology that God cannot do these things, it is the claim that he (or she or it) doesn't do these things.

The philosophical metaphysics of natural science are well articulated. For instance, natural science presupposes a regularity to the universe, affords a special place for induction in reason, and accepts that sometimes the evidence doesn't provide an answer on account of underdetermination. Also, highly influential on natural science is David Hume's views particularly on causality. Natural science embraces the principle of sufficient reason, and and even among natural theologians, who profess faith in God, one just cannot decide a supernatural cause exists even in if the lack of natural explanation is fact. Theistic science specifically attempts to undermine the theory of evolution which is a very durable and more than adequate theory in biology whose scientific controversy ended in the 19th century. It does so by trying to disprove evolution is a naturalistic explanation by engaging in certain details of biology.

The Details of Biology Are Irrelevant

Theistic science muddies the water philosophically by casting doubt on the conclusions of natural science. In increasingly sophisticated attacks using the language biology, mathematics, and chemistry, it marshals details to cast doubt on evolution, which is non-controversial within natural science. But metaphysically (and legally in Kitzmiller), the primary motivation of theistic science is to convince people it is a (superior) form of science leading people back to the supernatural claim of a Designer or Creator. In this way, it not only equivocates on the term 'science' taking people's confusion or ignorance for granted, and not only rejects the supreme presumption that there are no supernatural causes, but goes on to cloak itself by adopting the actual practices of natural scientists (poorly as many critics maintain).

And yet, even if one attempts to use naturalism to discredit certain naturalistic findings, one cannot then arrive at a supernatural conclusion and continue to call one's theory natural science since it natural science excludes supernaturalism as a presumption. This is the metaphysical inconsistency in theistic science which makes it theological and not scientific in the primary sense of the word. It claims that the findings of natural scientists support a supernatural conclusion. Obviously, the logical contradiction is apparent.

This is not happenstance. A number of prominent philosophers of science maintain that theories are laden with metaphysical presumption. Therefore, theistic scientists arrive at different biological conclusions than natural scientists precisely because they philosophically start with presumptions about the viability of supernatural causation. In modern parlance, their theory arrives at different conclusions because scientistic theories (be they creationist or natural with scientistic indicating the superficial practices as distinct from the fundamental metaphysical presumptions) are theory-laden:

In the philosophy of science, observations are said to be "theory-laden" when they are affected by the theoretical presuppositions held by the investigator. The thesis of theory-ladenness is most strongly associated with the late 1950s and early 1960s work of Norwood Russell Hanson, Thomas Kuhn, and Paul Feyerabend, and was probably first put forth (at least implicitly) by Pierre Duhem about 50 years earlier.

Modern philosophers of science have argued that rather successfully (and contributing to the defeat of the programme of the logical positivists) that scientific theories are intimately bound to the nature of psychological observation. In other words, observation sentences fail to be objective, because observers are by definition subjective, and that attempts at objectivity are at best idealized (philosophical) intersubjectivity. I suspect this is less of a problem for Continental philosophers of science, than analytic philosophers who are used to carving the observer out of the theory. But language is normative, conventional, and teleological, if you believe latter Wittgenstein. (Philosophical Investigations is of course open for interpretation, but see language-games and the private language argument for a start.)

Summary of Why ID Isn't Natural Science

So, now let's put ID in the proper philosophical context. ID is a both a philosophical theory and a political agenda (addressed elsewhere, but in the words of the ID folks themselves the Wedge Strategy) that is promulgated almost entirely by theistic scientists (there are outliers, of course). Setting aside the political agenda for replacing natural science with theistic science, stand the metaphysics of the theory itself.

Theistic science and the ID it advocates is theology and NOT natural science because while it pays lip service to methodological naturalism, it is inherently hostile to naturalism by way of the belief of the reality, truth, or adequacy of supernatural causation. It conflates two important theological statements: 'God CAN intervene in the universe because He is all powerful' with 'God DOES intervene in the universe because He is all powerful'. Catholicism, as a prominent example, rejects this (as I understand Catholic doctrine with little effort to master the nuance of the encyclicals or history being non-Catholic).

Intelligent Design is a research methodology, but it is not one of natural science. In fact, it is has been crafted to replace natural science. It may call itself science, wear the lab coat, get degrees in natural science, talk like natural science, and apply to the local public school to "teach the (non-existent) controversy". But it is not natural science, because natural science rejects supernaturalism in causation and has done so in increasingly better language from Thales to Bacon to Carnap. It may efface the Christian God, and even spawn an Islamic and ET version, but the concepts, under the lamp of philosophical analysis, ID can't conceal the clear fingerprints of fundamentalist theology that begs for supernatural forces to run amok through the union of naturalism, materialism, and physicalism that undergird natural science today.


The Dream of Reason by Anthony Gottlieb
Early Greek Sciences: Thales to Aristotle by G.E.R. Lloyd
Greek Science after Aristotle by G.E.R. Lloyd
The Origins of Modern Science by Herbert Butterfield
Husserl and the Sciences by Richard Feist
This is Biology by Ernst Mayr
Causality by Judea Pearl
A Companion to the Philosophy of Science edited by W.H. Newton-Smith
Abusing Science: The Case Against Creationism by Philip Kitcher
Why Intelligent Design Fails: A Scientific Critique of the New Creationism by Young and Edis (with the caveat I only poked around in Google Books, and Amazon still has it on the way).
Kitzmiller v. Dover, Robert T. Pennock, and John F. Haught

  • 1
    Wow, what a juggernaut! I don't understand why religion even tries to meet science on the same battlefield. It is its own thing. You don't see artists, musicians, dancers and chefs claiming to follow scientific methods... But those things can be taught in schools. What did religion do to get booted out?
    – Scott Rowe
    Commented Jun 6, 2023 at 22:49
  • 2
    @ScottRowe "What did religion do to get booted out?" Burned too many people at the stake, perhaps? Started too many religious wars? Commented Jun 6, 2023 at 23:32
  • @SimonCrase "You catch more flies with honey than with vinegar." It's just bad marketing!
    – Scott Rowe
    Commented Jun 7, 2023 at 2:01
  • JD, what is the wall between church and religion?
    – Scott Rowe
    Commented Jun 7, 2023 at 10:28

This is a very thoughtful question. Since you’ve specified the criteria you’re asking about, I’ll forego any comments on what the criteria should be and discuss the first two of yours.


Evolutionary Algorithms

Different people seem to mean different things by “Intelligent Design”—which is not necessarily a problem for any one of them, but makes this impossible to answer without clarifying which ideas you are asking about. In this context, I’ll discuss the ideas of William Dembski.

William Dembski has strongly asserted, “evolutionary algorithms cannot generate specified complexity.” He concedes that, if that is wrong, so is his whole theory. That seems highly falsifiable these days. Does this prove that nothing, absolutely nothing, an AI model created by reinforcement learning can produce can possibly be “complex specified information?” If the answer is that it cannot be, we will have a clear falsification when any such model creates even a single bit of complex information that matches a specification. Which we more often call, a prompt.

But if on the other hand we bite the bullet and say: nothing created by the product of any evolutionary algorithm can ever be complex specified information? In that case, there must be a whole lot less of it than Dembski, Michael Behe or anyone else believed until very recently. If our intuition was that wrong about what types of information evolutionary algorithms can create, how can we be sure that anything is completely out of its reach? Will no reinforcement-learning model ever be able to create any virus or nanobot capable of killing all humans? Are we absolutely certain from first principles that that will forever be beyond that type of AI, because no product of an evolutionary algorithm, no matter how advanced, could ever potentially create anything as complex and specified as even one part of the very simplest biological organism? Such as, to give one of his examples, a flagellum. Does it follow that taking any precautions would be foolish, because William Dembski proved it impossible? Would you bet your life on that?

Or, if someone asks ChatGPT 1701-D, “Write a genome for a humanoid organism with humanlike intelligence and personality, and also green skin, but no genetic similarity to homo sapiens, and save it to my projects folder,” and it does, what exactly would that prove?

Or perhaps ID proponents are in fact acknowledging that certain evolutionary algorithms can produce complex specified information, but only through some kind of loophole that means it doesn’t really count? Dembski, for example, responded to a well-known counterexample (METHINKS IT IS LIKE A WEASEL) by claiming that the fact an evolutionary algorithm was able to generate this string proved that it was only “apparent complex specified information,” and was possible because “the search function” was as complex as the information it produced.

Well, all right then, if evolutionary processes are, by Dembski’s own express admission, able to create specified information as complex as “the search function,” hoe many bits of complexity does the search function of natural and sexual selection on planet Earth have? Compared to the human genome?

We can easily determine an upper bound on the number of bits of information in the human genome: each base-pair is one of four possible types, so the uncompressed size of one human’s genetic code is two bits per base pair, or less than 13 gigabits. However, Dembski defines the actual information content as the size after the maximum possible compression, which would be strictly less. The fact that every human with XX chromosomes has two copies of each gene on a pair of chromosomes alone cuts this nearly in half. (Observe that, if there’s more than half because two different variants of a gene on a pair of alleles have more information than two identical copies, mutations to a copy of a gene always add bits of information to the genome, by William Dembski’s definition.) Additionally, much of the genome is completely non-functional “junk DNA,” so the information actually needed to grow a human brain is a fraction of that, although whatever led to the human genome must have been complex enough to create all that junk, too. So we can be reasonably confident that (at eight bits to the byte), the human genome contains around a gigabyte of information, and not ten times more or less.

You run some apps with more bits than that on the supercomputer you carry in your pocket. Operating systems, for example, crossed the threshold of needing to be able to run executables larger than four gigabytes (which is 32 gigabits) a decade ago, and that’s not even counting the size of all the data files. And the download for it was compressed, so the amount of information it contains is probably not much less than the download size.

Now, how many bits of complexity would a completely realistic computer simulation of natural selection, running an evolutionary algorithm, have? Here are a few of the necessary components:

  • An algorithm that accurately predicts when any individual human being will die. You could also sell this to insurance companies.
  • An algorithm that accurately predicts which two humans will mate and how many children they will have. You could also sell this to dating sites.
  • A realistic simulation of the ancestral environment, which you might release as a spiritual sequel to Ancestors: The Humankind Odyssey.

If all of this put together, code and all the data files they would need, and packaged for download with the best compression algorithm, would be as large as Microsoft Office. William Dembski has conceded that natural selection has enough bits of complexity to produce the human genome through evolution. Whether he realizes it or not.

Empirical Evidence

Examples from Nature

The previous section got long for an answer on this site, so I’ll keep this section short and then stop. You already know what most scientists think about Creationism and whether it posits “then a miracle occurs.”

If you find a detailed log of precipitation and temperature going back hundreds of years, is that complex specified information that could not be that accurate by chance?

Many museums of natural history have slices of tree trunks whose rings encode exactly the same information. This information is not in their genetic code. The organism added and incorporated information from the surrounding environment into its body. This is also what happens when evolution adds information to the genetic code.

Or. let’s say you find a particle detector with a record of every time a neutrino or gamma ray collided with it, and someone discovers something in that data that they publish as a physics paper and wins a Nobel prize. Did that data contain any complex specified information? If so, so does the record of mutations in our genetic code, recording every mutation caused by a particle hitting one of its molecules.

  • 2
    I strongly disagree that "evolutionary algorithms cannot generate specified complexity" is falsifiable, at least until "specified complexity" is clearly defined. ID proponents commonly say the human eye is "irreducibly complex" and cannot come about through evolution. We've then pointed out the various evolutionary steps (with reference to organisms currently alive) that could result in the formation of an eye. Many of them have just kept saying that despite this, while others may have moved the goalpost to some other form of complexity. That's stock-standard unfalsifiability.
    – NotThatGuy
    Commented Jun 5, 2023 at 11:30
  • 2
    @NotThatGuy William “CSI” Dembski and Michael “Irreducible Complexity” Behe are two different people, but both have given specific examples of things that they say cannot be the result of evolution. Behe was more lucid about what he meant, and so has already been falsified (not that that ever stops a crackpot). As for Dembski, his motivation is obviously to claim that life cannot have evolved, so an algorithm designing a virus would disprove his original claims. He might later move the goalposts to some other claim, like “Okay, a bacterium yes, but not human brains.” But so far, he hasn’t.
    – Davislor
    Commented Jun 5, 2023 at 17:53

Intelligent Design seems to fail at all points.

The main problem is that the identity of the Intelligent Designer is not established. Trying to explain a natural phenomenon by attributing it to an unidentified entity is downright absurd.


Does Intelligent Design fulfill the necessary criteria to be recognized as a scientific theory?

It doesn't.

There's no experiment can be designed such that its results can only and exclusively be interpreted as an evidence of Intelligent Design.

Suppose you managed to summon God. He opened a portal to New Jerusalem from which angels emerged, started raising the dead and doing all sorts of supernatural things, magnificent and terrifying.

Provided that you survived the apocalypse, suppose you figured out how to reproduce it indefinitely. From now on, everything can be perfectly studied, measured, analyzed, and converted into a rational knowledge.

But you still cannot determine if God is really THE creator.

  • Maybe he is a chthonic monster or a munchkin power user who emerged into existence along with the universe and seized the control of it.
  • Maybe he is a NPC in the simulated reality, where life just appeared by a random chance.
  • Maybe he is AI/Golem that created this universe by blindly optimizing some utility function, and, again by just a chance, it happened to generate the existence that we observe.
  • But maybe he really IS the creator. Which, if true, there's no way you can be sure of anyway, since you still cannot exclude all the other possibilities.

Domain of scientific knowledge is limited to statements the ground truth of which can be challenged by logical rigor and practical experiment.

The ground truth behind the idea of Intelligent Design cannot be challenged. It is fundamentally unreachable. Thus it is not scientific.

Basically any idea that makes claims about unreachable things (e.g. the universe is a result of a vacuum fluctuation, we are living in a matrix, Elvis was abducted by the aliens, Ctuhlu will rise from Sagittarius A* and devour us all, etc.) belongs to the same camp of ideas which are non-excludable but aren't necessary either. There always can be found such an interpretation of the universe's origin that replaces subjective intention with an inanimate process, and some other interpretation which contends the opposite. These are doomed to fight each other indefinitely to no final outcome. (Which is by the way the reason why people learn to live in peace with never-ending discordance in human beliefs.)

So the simple heuristic is:
Whenever you want to address some idea, but the only means available to you are convictions and interpretations, you're standing outside the walls of science. Which isn't good and isn't bad in itself; just the fact of being beyond the scientific scope.

  • "There's no experiment can be designed such that its results can only and exclusively be interpreted as an evidence of Intelligent Design." I fail to see how the theory of evolution is not vulnerable to the same criticism. This standard would render evolution unscientific as well. There is no known experiment that can be run such that its result can only and exclusively be interpreted as an evidence of (macro) evolution. All known, feasible experiments are open to alternative interpretations.
    – Mark
    Commented Jun 8, 2023 at 21:19
  • 1
    @Mark "how the theory of evolution is not vulnerable to the same criticism" – Because unlike products of sheer belief and imagination, the material is always here for us for studying. Take a genetically uniform population of some species, divide it into groups and put them into a separate environments with critically different requirements for phenotypic fitness. Check what happens after a few dozen generations.
    – blakkwater
    Commented Jun 8, 2023 at 21:47
  • @Mark This will reproduce one of the main scenarios thought to give a push to evolutionary processes. Despite that it theoretically can appear to be not working as expected, it can be arbitrarily tested to a large extent, which is why it belongs to scientific scope.
    – blakkwater
    Commented Jun 8, 2023 at 21:49
  • But again, let's go back to the standard you proposed: "There's no experiment that can be designed such that its results can only and exclusively be interpreted as an evidence of [fill in the blank]." Even if you manage to run an intergenerational experiment for 1 million years to see macroevolution in action, you still can't rule out alternative explanations such as "there was a hidden intelligence guiding evolution", or "the universe was created 5 minutes ago with the appearance of age", or "this experiment is an illusion inside the Matrix", etc.
    – Mark
    Commented Jun 8, 2023 at 22:07
  • @Mark The point of science is to gain reliable, useful knowledge. When you're conducting an experiment but then decide to add into the mix random what-ifs, it doesn't change (in the either way) the predictive power of the knowledge gained from the experiment. Scientific knowledge is very strict in its interpretation exactly because it is intended and ensured to be reliable; it's what it exists for. When you trust your SSD to not mess up, say, more than 1 bit per 1 TBW, you can exclusively attribute it to the god of statistics which prevents quantum tunneling from messing up more bits.
    – blakkwater
    Commented Jun 8, 2023 at 22:53

The core of ID really comes down to one fundamental claim:

Physical systems cannot generate more complex, specified structure than is put into them. This is known as "conservation of information".

This claim is entirely empirical and testable. We can also prove mathematical theorems about this claim. Hence, at least this aspect of ID is scientific.

Now, responding to each of the criteria:

  1. Falsifiability: Yes, the above claim is eminently falsifiable. Many scientific and mathematical books have been written to try and falsify it.
  2. Empirical evidence: Conservation theories are everywhere in the empirical sciences, so there is a large amount of empirical evidence.
  3. Methodology: Yes, for example here is Dembski's IEEE article proving conservation of information.
  4. Consilience: As mentioned in #2 conservation of information is very compatible with other fields.
  5. Naturalistic explanations: Rather, ID is proposing the complete absence of naturalistic explanations for the origin of information, which is very testable. Only a single contrary example is necessary to disprove ID.
  6. Consensus within the scientific community: Surprisingly more than is commonly realized. For example, a corollary conservation of information theorem known as the data processing inequality is central to the field of machine learning.

Furthermore, since conservation of information is central to ID, then any piece of contrary evidence falsifies not only CoI but also ID. Hence, the theory as a whole is empirically falsifiable due to the CoI, making the theory as a whole scientific.

  • 4
    "Conservation of information theorems indicate that any search algorithm performs, on average, as well as random search without replacement unless it takes advantage of problem-specific information about the search target or the search-space structure" - the entire principle of evolution is that it "takes advantage of problem-specific information...". So that simply doesn't apply to evolution at all. Or the evidence for evolution (even just modern evolution, which most ID proponents accept) has already falsified that - your choice.
    – NotThatGuy
    Commented Jun 6, 2023 at 6:21
  • 3
    Your entire argument seems to be "we can define ID as little more than conservation of information, and all of this is true for conservation of information". That's not how anything works. Intelligent design is claiming that there is ... a designer ... that is intelligent. This is a lot more than simply claiming that information is conserved. You're basically arguing from ignorance and special pleading: if the known alternative is false, then there must be a designer, who is also not subject to the same principles. Never mind that many/most ID proponents make much more specific claims.
    – NotThatGuy
    Commented Jun 6, 2023 at 6:29
  • 4
    From en.wikipedia.org/wiki/Specified_complexity: "Specified complexity is a creationist argument introduced by William Dembski, used by advocates to promote the pseudoscience of intelligent design. [...] The concept of specified complexity is widely regarded as mathematically unsound and has not been the basis for further independent work in information theory, in the theory of complex systems, or in biology." Really, please spare us creationist pseudo-science. There is a forum here for that, christianity.stackexchange.com , where such answers will get you all the applause you want.
    – tkruse
    Commented Jun 6, 2023 at 8:37
  • 2
    @NotThatGuy conservation of information is the foundation of ID. If natural processes can create information, then there is no need for a designer. If they cannot, then we have to look elsewhere for the source of information. This piece is clearly testable, and since it is a crucial component of ID, then we can definitively say this aspect of ID is scientific.
    – yters
    Commented Jun 6, 2023 at 11:26
  • 2
    "Physical systems cannot generate more complex, specified structure than is put into them" - it just means that some people have no imagination. The universe has done a lot with 3 particles and 4 forces. Raspberry to this entire stubborn group of folks.
    – Scott Rowe
    Commented Jun 7, 2023 at 10:36

Just so other viewpoints are considered, I would like to quote this answer (exact link: https://christianity.stackexchange.com/a/95690/61679) (NOTE: I didn't write this, I'm literally quoting verbatim an answer written by an ID proponent over on Christianity Stack Exchange, everything below is inside a quotation block, and I'm referencing said answer, so this is not plagiarism.) (You may also feel free to downvote/upvote this answer as if you were downvoting/upvoting the original answer posted on the other site. I won't take it personally.)

1. Does Intelligent Design make testable predictions that, if proven false, would contradict its central claims?

Yes. I'm not going to try to summarize, but ID's predictions tend to revolve around science failing to determine viable mechanisms for producing various observed features via any purely naturalistic mechanism. For the most part, these predictions have so far proven correct.

(Cue scoffers whining that "a God-of-the-gaps argument is not much of a 'prediction'". While there is a certain amount of truth to this, the purpose of such predictions is that they are falsifiable; that is, if it could be shown that design is not necessary for life to exist, that would refute ID. While it's true that the positive case for ID is partly based on absense of evidence, the idea that we cannot accurately infer the action of intelligence is ludicrous. Archaeologists and forensic scientists do exactly that all the time. SETI endeavors to do so; thus far without success. There is no non-philosophical reason why such inferences should be excluded from biology.)

Of course, ID makes or made many other predictions or retrodictions which are confirmed by our current knowledge. Several of these can be grouped as "supposed evidence against ID will turn out to be incorrect" and cover topics such as "vestigial" organs (that aren't) and "junk" DNA (that isn't), and "poor" design (that actually serves a purpose). In general, life simply looks like design was involved.

See also:

Although you asked about testable predictions, a related criticism is whether ID (or, more accurately, the detection of apparent design) is falsifiable. The answer to that is clearly "yes". ID prefers to eschew design where non-design can be shown to be a plausible alternative, much as methodological naturalism prefers to eschew miracles where a non-supernatural explanation is feasible. Thus, all that is necessary to falsify a design hypothesis is to show that a purely natural causal chain can produce the observed effect. This is true not only for ID, where design detection is applied to biology, but for other fields that attempt to detect design such as archaeology or forensics.

For more on falsifiability, see:

2. What empirical evidence supports or refutes the claims made by Intelligent Design?

Sorry, I don't feel like writing a book. Please read one of the excellent books available. 🙂 (Suggestions at end of Answer.)

Very briefly, we simply don't have any evidence that an organism can change sufficiently to become a totally novel organism. Indeed, we have no evidence that natural selection can produce anything truly novel outside of some very strict bounds. What we do have is any number of features that are consistent with design.

3. Does Intelligent Design follow established scientific methodologies, such as proposing hypotheses, conducting experiments, and engaging in peer review?

It certainly tries to, and the ID space does publish. ID is hampered in this respect, however, by extreme philosophical hostility. "Mainstream" journals automatically reject anything that even hints of ID, and funding is a serious problem, particularly as materialists tend to do everything possible to deny funding to ID research.

That said, it should be noted that ID is mostly a historical science. We don't presently have the ability (and perhaps never will) to design organisms for ourselves, so much of ID research is more akin to paleontology; that is, not so many experiments in the sense you'd expect from, say, material science.

4. Does Intelligent Design integrate with other scientific fields and theories, fostering a cohesive and interconnected body of knowledge?

This is a difficult question. While ID certainly does not ignore other fields, it's unclear how it contributes to, say, materials science. (Of course, it's unclear what materialism has to offer, as well.) This, however, should not be seen as a difficulty, any more than paleontology is invalidated by having minimal application to particle physics.

Perhaps a better response is to not that ID is not inconsistent with other branches of science, aside from those with which it directly competes. In particular, it is not inconsistent with any observational science.

5. How does Intelligent Design address the requirement of offering naturalistic explanations, which is commonly expected within scientific inquiry?

First off, it needs to be clearly stated that requiring only naturalistic explanations leads to incorrect science. Consider a bicycle. If one excludes intelligent causes, it is going to be very difficult to arrive at a correct explanation of how the bicycle came to exist. Similarly, if an intelligence is responsible for life, ruling out that possibility a priori is obviously going to lead to incorrect conclusions being drawn.

The question ID attempts to answer is "was intelligence involved?" This is a philosophically neutral question, as it does not presume to speculate on the nature of the intelligence. As far as ID goes, it does not (directly) rule out the possibility that terrestrial life was designed by an intelligence which originated via purely naturalistic processes. (Of course, if ID is correct, the origin of this potential designer is likely to represent a serious difficulty.)

6. What is the general opinion among experts in relevant scientific disciplines regarding the recognition of Intelligent Design as a scientific theory?

The most meaningful thing I can say here is that it is highly polarized. It may be worth observing, however, that there is substantial pressure brought to bear against ID; individuals expressing sympathy to ID are subject to any manner of social and economic reprisals.

Some comments with respect to the responses on Philosophy.SE:

Lots of potentially "science" subjects become non-science when their advocates hold by them despite blatant refutations.

...and if you think this doesn't apply to materialism, well...

There are many lines of evidence that present serious challenges, not only to Abiogenesis and Common Descent, but also to the idea that Earth is billions of years old. Yet adherents still insist on calling their beliefs "science".

The KEY to being science is not the outward forms, but the openness to and willingness to search for refutations. And the ID community is NOT open to reconsidering its basic assumptions, nor looking for refutations.

When is the last time you encountered a materialist that was open to reconsidering their basic assumptions, or looking for refutations? Given how trivially easy it is to find such refutations...

When we start digging for evidence we don't find a lot of intelligent designers, or a lot of designs that appear to be the products of active intelligence. Instead, we find what appears to be a slow drift of living things changing to optimize within ever-changing environments.

While it's true that those who are out to avoid God at all costs have a hard time "finding" Him, the claim that we don't find apparently-intelligent designs is utterly duplicitous. Indeed, Richard Dawkins, one of the best known defenders of Common Descent, defines biology as "the study of complicated things that give the appearance of having been designed for a purpose". People that claim something "doesn't look designed" are almost always desperately trying to deny Design. Claims of "slow drift" are largely unsubstantiated (what we actually see is sudden¹ "appearance" followed by relative stasis) and involve a healthy dose of wishful thinking, particular through the assumption (which is itself questionable) that life has been around for millions of years.

(¹ Materialists would argue that "sudden" is a relative term, to which ID says a) "produce the intermediates" and b) demonstrate how life can get from A to B within the relatively short time period that is available. Particularly around the Cambrian Explosion, the complexity of known organisms takes a giant leap forward in what seems to be a period of time that does not seem adequate to account for the degree and diversity of change.)

[Intelligent Design] would be treated as a failed theory, because of the extremely poor evidence in its favor and the vast amount of evidence that contradicts it.

Evidence is subject to interpretation, and many people feel that the evidence very much supports ID, and that there is little evidence against it (and that most such can be explained as a result of entropy acting on a design that occurred in the distant past).

Media Bias rates [Evolution News(?)] as a "Strong conspiracy website based on promoting biblical verses as science. We also rate them low for factual reporting for the same reason."

This is just hilarious. Ignoring the obvious pro-materialist bias (a group which has a significant history of known fraud), the Discovery Institute isn't even a Christian organization (some of its members are Christian) and many of EN's articles make no mention of God or Scripture. Ironically, such attacks are worthwhile for revealing materialism's true agenda, which is not seeking the truth but attacking Christianity. Therefore, they view anyone who does not completely embrace materialism is a traitor to their anti-Christian crusade, and consequently believe such people must support what they are trying to oppose.

Is ID "science"? It's proponents certainly feel that it is, although they're clearly biased. It's opponents think it isn't, but are no less biased. A trite article on the internet is hardly the best way of answering the question. Instead, I cannot recommend strongly enough to read some books by ID proponents, especially those published in the last few years. (Noteworthy authors include Douglas Axe, Michael Behe, William A. Dembski, Michael Denton and Stephen C. Meyer.)


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .