As our research into the cell progresses over the decades, we are seeing greater and greater levels of complexity. According to our current understanding, the cell resembles some sort of miniaturized city as can be seen here and we are still far from complete understanding.

At what point should one conclude that an intelligence was involved here?

Is there a limit to what can be attributed to unguided natural processes? Or perhaps there is no limit and whatever level of complexity we ever confront can always be attributed to unguided natural processes?

  • Let us continue this discussion in chat.
    – MichaelK
    May 18, 2018 at 10:04
  • Comments are not for extended discussion; this conversation has been moved to chat.
    – user2953
    May 18, 2018 at 15:41
  • Am I right in saying the question and all responses do not address a key prerequisite issue, objectively quantifying complexity? Or at least as "objective" as we measure quantities like mass, length, time, energy, etc. Over the years I've tried to find an answer in this forum, google, and university library resources but it seems to be a completely subjective, context-sensitive or system-dependent property. What are the units of complexity and what is a mechanical algorithm to measure it? I don't think such a thing exists.
    – user6552
    Feb 15, 2019 at 20:40
  • "Is there a limit to what can be attributed to unguided natural processes?" They are not "unguided". They are guided by physics and other simple rules. "At what point should one conclude that an intelligence was involved here?" At the same point or standard of evidence we have for all other things in science: direct observation of the cause and the effect together at the same time; i.e. the intelligence itself changing or physically altering life as mutation, natural selection, sexual selection and other mechanisms of evolution have been directly observed.
    – user6552
    Nov 19, 2020 at 21:23

6 Answers 6


As I understand it, the problem with the intelligent-design argument isn't a matter of the complexity of the objects under study, which are indeed very complex. The logical difficulty with this argument is that it posits the existence of an intelligent designer who is (presumably) at least as complex in its own construction as the things it is purported to have designed. This naturally begs the question: who designed the designer? To put this another way, if the existence of high complexity in nature requires an intelligent designer as an explanation, and if a designer is at least as complex as the thing it designs, then it follows that we must have an infinite regress of designers; it's turtles all the way down!

In the context of consideration of the complexity of a cell, yes, it is very complex. But if a single cell is very complex, and all the intelligent beings we know of are made of cells, then we know that these intelligent beings (who are not as intelligent as the purported intelligent designer) are more complex than the cell. The intelligent designer is presumably more complex than us, and hence, by transitivity, it must be more complex than the cell, whose complexity we are trying to justify.

  • "at least as complex in its own construction as the things it is purported to have designed" - by intelligent designer, i presume you mean some sort of eternal being, i.e. God. if He is eternal, then He was not "constructed". thus He is not made of parts, bolts, springs, cells, etc. since that would need to be preceded by something else and hence would not be eternal.
    – michael
    May 18, 2018 at 5:31
  • 2
    But that is the point isn't it: if God is complex, but requires no causal explanation in terms of a designer, then complexity requires no causal explanation in terms of a designer. In other words, the problem for the argument is that you can't have it both ways -i.e., asserting that complexity of objects requires a causal explanation from a designer, but asserting that God is eternal and non-caused and so we don't have to explain Him.
    – Ben
    May 18, 2018 at 6:13
  • i am saying something constructed, with parts, properties, yes requires a previous designer/cause. but something which is Eternal is not constructed by definition
    – michael
    May 18, 2018 at 8:52
  • That just sounds like special-pleading to me. It also locks you into the view that God has no properties, which makes Him not really a God (or anything at all) in any reasonable sense of the word.
    – Ben
    May 18, 2018 at 9:15
  • not anything we can understand. ultimately "something" has to be eternal and uncaused though otherwise nothing would exist. unless you want to invoke things like infinite regress
    – michael
    May 18, 2018 at 9:39

There's really two issues here. The first is a scientific question, and not a philosophical one. The second is a matter of philosophy, specifically epistemology (the study of knowledge).

1. Scientific Question: At what point would life be too complex for evolution by natural selection to realistically account for it?

If the complexity and specific structures observed appeared too early in the history of life to be compatible with biological possibility. Biological possibility would be related to things like mutation rate, population size and geographical distribution of species, average generation time, etc.

So, to take an absurd example, evolution by natural selection would not account for an observation of Homo sapiens appearing during the time of only bacteria and archebacteria, billions of years ago. Less absurd might be if we observed a brain with mammalian complexity from an annelid nervous system, evolving in 5,000 years--still far too short a time for that to occur by evolution alone.

The vast majority of biologists think life has had sufficient time to evolve to where it is now. To give detailed explanations of it all would be beyond the scope of a philosophy post. But it might be helpful to briefly underscore the magnitude of the amount of time and the amount of biological interactions that have happened since life began.

We tend to forget that many of the structures found in life (proteins, phospholipids, etc.) evolved in bacteria, and bacteria have a generational time on the order of a half hour to a day. Let's say it's a half day for convenience. That means each day there are two generations of bacteria if you start with one cell. Each year, 730 generations. The 1.5 billion years of bacterial life, that's over a trillion generations. If it were once every half hour, it'd be 26 trillion generations.

But it's far more extreme than even that, since mutations are happening in parallel, meaning that there are unimaginable numbers of bacteria out in the world at any given time (one estimate puts it currently at "five with 30 zeroes after it") , all of which can be candidates for producing new traits.

In fact, an actual experiment in a lab saw an important new trait evolve in E. coli in "just" ~30,000 generations.

2. Epistemological Question: Should we accept explanations without evidence?

One way to look at this issue is reforming the question to be, "Is it ever warranted to default to Explanation B, for which we have no evidence, just because Explanation A fails?"

That question reduces to this question: "Is it ever warranted to accept an explanation without any evidence in favor of it?"

The answer to that is a huge area in epistemology, beyond the scope of a single SE answer. But you can start with the idea of "Foundationalist Theories of Epistemic Justification" to explore why we have warrant to accept any belief.

But to explore that issue in this context, let's assume for the sake of argument that we decide that the complexity in nature means evolution by natural selection fails as an explanation. In this case, filling that in with these particulars would give us, "Is it ever warranted to default to Intelligent Design as an explanation, for which we have no evidence, just because evolution by natural selection fails as an explanation?"

I tend to work under the rule that one does need evidence in order to hold a belief about the physical nature of the world. Therefore, I don't think it is a good idea to default to intelligent design.

Instead, what we should do at that point is (given this hypothetical where we decide evolution fails as an explanation) simply admit we don't know what explains the complexity in nature.

Let me push this envelope even further. Let's assume that next year, a geneticist noticed that on human chromosome 17 there is a sequence of base pairs that, if interpreted through the right decoding rules, spells out a 20,000 word message in English, something to the effect of, "I am the Intelligent Designer, and I designed you. Here are my instructions for life...". Assume the story is not fake and the research is done extremely rigorously and is reproduced by dozens of honest top scientists.

Would we then have warrant to accept with absolute certainty that there is an Intelligent designer behind the complexity of life? At first, it would seem like it, but then consider other similarly evidence-less explanations:

  • The human mind somehow has causal power over the genetic code and there is some collective unconscious that is programming that into us.
  • Rogue scientists created this DNA sequence in the early 2000s and disseminated it via a highly contagious virus that incorporated the sequence into everyone's DNA within a few years.
  • Time traveling scientists/aliens did this to us at some time in the past 2000 years.
  • It turns out to be proven that every sequence of any 20,000 word book can be found in the DNA if you use a supercomputer to find the right decoding rules.

These seem far-fetched, but are they any more far-fetched than the idea of what amounts to a magic person doing this merely by thinking about it?

  • the question is is there a limit to what can be explained by natural selection, not whether current life is too complex
    – michael
    May 20, 2018 at 18:12
  • i have heard of that e.coli experiment. but also that the e.coli already had an ability to work with citrate and that it was a very minor change. do you know of any other evidence regarding the random mutations aspect of evolution producing new structures?
    – michael
    May 21, 2018 at 10:02
  • @m.r. I changed the #1 section based on your correction.
    – Chelonian
    May 21, 2018 at 16:29
  • @m.r. I don't know what you mean by "work with" citrate. Here's the actual author's words: "They have since evolved in a glucose-limited medium that also contains citrate, which E. coli cannot use as a carbon source under oxic conditions. No population evolved the capacity to exploit citrate for >30,000 generations, although each population tested billions of mutations." Read it yourself:pnas.org/content/105/23/7899
    – Chelonian
    May 21, 2018 at 16:35
  • i am familiar with that. a few bits of dna to modify an existing protein's binding site. it is within what you'd expect for a random dice throwing experiment. keep in mind though that an average protein is about 1000 dna base pairs. i don't see what you want to show from this. (furthermore, there has been documented cases showing dna changes itself in a non-random way somehow)
    – michael
    May 21, 2018 at 17:09

Alvin Plantinga’s chapter “Design Discourse” in Where the Conflict Really Lies examines Michael Behe’s arguments in Darwin’s Black Box and The Edge of Evolution for “irreducible complexity” leading to intelligent design. The designer(s) need not be a theist's version of God, but only the involvement of some intelligent agent(s) rather than random processes.

Plantinga also considers Paul Draper’s “Irreducible Complexity and Darwinian Gradualism: a Reply to Michael J. Behe”, Faith and Philosophy 22 (2002) pp. 3-21.

He concludes, after introducing a distinction between an argument and a discourse which can have defeaters and deflectors of those defeaters, the following (page 264):

On balance then: Behe’s design discourses do not constitute irrefragable arguments for theism, or even for the proposition that the structures he considers have in fact been designed. Taken not as arguments but as design discourses they fare better. They present us with epistemic situations in which the rational response is design belief--design belief for which there aren’t strong defeaters.

Consider the question: At what point should one conclude that an intelligence was involved here?.

One could say there is enough evidence now to accept the design discourse at least until those supporting unguided evolution can provide defeaters for that belief.

  • I read Platinga's conclusion in the context of that question differently: That there is no argument at all that warrants Intelligent Design, but no elements of the discourse that are able to defeat a certain (he calls it rational, which has a peculiar meaning in Platinga IIRC) response to the discourse as well. Evidence can be used as an argument. In other words: NO amount of complexity is able to warrant Intelligent Design, but there is no way to outright falsify the kind of natural response to complexity, i.e. assuming Intelligent Design. Anthropomorphism is natural, see Kant's CPJ.
    – Philip Klöcking
    May 21, 2018 at 17:44
  • @PhilipKlocking Plantinga agrees with Draper that Behe's arguments are not conclusive. That is why he shifts from "argument" to "discourse". I will have to see how Plantinga views "warrant" in this context. May 21, 2018 at 18:05
  • @PhilipKlöcking On page 244, he writes "degree of warrant or positive epistemic status". This makes me think that his conclusion: "They present us with epistemic situations" implies there exists some degree of warrant for design beliefs based on Behe's design "discourse". May 21, 2018 at 18:30
  • "One could say there is enough evidence now to accept the design" - I say, one is free either to accept or reject without risk of being against the science.
    – rus9384
    May 21, 2018 at 19:08

The theory of evolution explains how more complex biological organisms can arise from simpler organisms. The theory of evolution starts with the idea of replicators - physical systems that can exist in multiple variants and are copied faithfully with some finite error rate in a particular environment. For example, the DNA in an amoeba's body is copied by the amoeba making copies of its DNA and them splitting into two independent cells. The environment allows some variants of a replicator to be copied and not others. New variants of a replicator, including more complex variants, can arise by exploiting features of the environment not exploited by other replicators. For more detail see "The Selfish Gene" by Dawkins.

There is no reason to think that the cells or replicators that exist now bear much resemblance to the first cells or replicators. The first cells and replicators may have been a lot less complex. Nor is there any reason to think that any explanation other than evolution by natural selection is required to explain existing biological complexity. Nor is it clear what sort of observation could refute the theory of evolution. You can only judge an observation as a refutation of a theory in the light of another explanation that replaces the theory. For a discussion of this issue in the context of evolution, see "The Beginning of Infinity" by David Deutsch, Chapter 4.

  • Comments are not for extended discussion; this conversation has been moved to chat.
    – user2953
    May 18, 2018 at 15:41

The problem with the argument from intelligent design, as it is often made, is the fact that it is made with reference to complexity at all.

When Paley's watch argument is made (an argument from analogy based on intuition which represents the general argument from intelligent design), the real reasons we identify a watch as intelligently designed and not a rock are as follows:

1) From past experience we know that a watch is connected to some 'maker'. Whenever we have seen a watch in the past we have correctly (or at least it would seem so) inferred the existence of some designer. Whether it be a company or some individual watch maker etc...

2) By comparison to surroundings. We don't suspect that a rock, for example, is designed in the same way as a watch (by a human being) because it fits in exactly with it's surroundings and, similarly to reason '1', we have no past experiences from which we haven infer that rocks require some kind of human designer.

Paley's analogy abuses our prior knowledge about what we expect to find in nature on Earth and our prior knowledge about watch designers. We are able to infer watch designers from past experience and so apply our inference onto a watch we see in the present. The same cannot be said for the universe. You also kind of undercut yourself by arguing that complexity = intelligent design since you could never say that something is nature is 'natural'. Since everything is intelligently designed you can no longer make strong comparisons between things in the universe like rocks and pocket watches.

Similarly, we cannot make a comparison between 'intelligently designed' universes and 'non-intelligently designed' universes. Unless we consider what the 'intelligent design' hypothesis predicts for reality and what the 'non-intelligent design predicts'. The non-intelligent design hypothesis doesn't predict much, as far as I can tell. The intelligent design hypothesis, on the other hand, certainly predicts flawless design (no illness, no deadly elements within nature etc) and there is no need for a God to create a universe so large and inaccessible to his sentient creation if the creation itself plays some central role. You can patch up these problems with ad hoc additions to a hypothesis about God but all that does, mathematically speaking, is make the truth of your hypothesis less likely since a hypothesis that requires stacked assumptions to entail evidence that exists in the universe becomes less likely as more assumptions are stacked (Occham's razor). The addition of ad hoc hypotheses to explain away problems with the idea of God is very common and it is best to hear it now, sooner than later: the more you 'explain away' problems, the bigger your problem becomes unless you have powerful evidence for every individual explanation you give.

In conclusion:

Complexity is not how you distinguish 'design'.

Saying that Complexity = Design undercuts your ability to make comparisons between objects in our universe since everything is complex and hence designed.

The hypothesis of intelligent design predicts many things of which the exact opposite is manifested in our universe and ad hoc attempts to patch up this issue don't help.

  • it is not just experience. a watch has a high degree of order and we know that order tends to disintegrate unless the hand of an intelligence was involved
    – michael
    May 21, 2018 at 20:29
  • @m.r. Sorry I haven't looked at this post for so long since you replied. I would argue that the statement "we know that order tends to disintegrate unless the hand of an intelligence was involved" is wrong/poorly stated. We could talk about it in a chat if you wanted? Jun 13, 2018 at 16:56
  • "we know that order tends to disintegrate unless the hand of an intelligence" Like it doesn't do for many things in the universe, such as planetary motion? No, whether "order" [which you have not defined or measured objectively] decreases, is not a function of intelligence [which, in humans, the only known high intelligence, is also not due to intelligence. It is due to natural laws operating on matter inside the brain.] It is a function of prior conditions, environment, and the amount of dissipative forces involved, not intelligence.
    – user6552
    Nov 25, 2019 at 16:05
  • "order and we know that order tends to disintegrate unless the hand of an intelligence was involved" This is so wrong I found the need to comment twice. Did you even attempt to quantify the cases where this happens? Did you consider adding up the total known systems in this universe along with their "order decrease" as a function of intelligent interaction? Restricting to the human world (which is really a subset of nature), I think you will find chaos happens more often than not. If you consider the natural universe, you will find many systems can maintain order for billions of years.
    – user6552
    Nov 25, 2019 at 16:16

The key to the intelligent design argument is not the absolute level of complexity, but whether or not there is an satisfactory explanation for the observed complexity. In particular, the controversy typically attached to this term is whether biological evolution, as currently understood, is an adequate explanation for the observed complexity of life. The scientific consensus is that it is, given the shaping power of biological evolution, and the vast amount of time involved --although significant gaps in understanding do still exist, chiefly around the origins of life.

It's worth noting that biological intelligent design as an argument for belief in God has never figured particularly prominently among serious theological thinkers --its current prominence is arguably largely a response to the challenge evolutionary theory poses to literalist Biblical Creationism.

A more sophisticated and arguably more interesting version of the argument perceives intelligent design less in the individual creature, and more in the fecund universe. For example, much of the complexity in biological life is fractal, or otherwise directly related to the principle of chaos theory that vast complexity can arise from the reiteration of very simple rules or processes. Yet fractals themselves do not evolve. The complexity represented by fractals, therefore, is arguably built into the fabric of the universe itself, a resource intrinsically available to evolution (similar to how the built-in resources of a programming language are available to a programmer).

  • "as currently understood, is an adequate explanation for the observed complexity of life" - but how did they conclude this. as far as i know there is very little if any evidence for it. it is accepted as part of the naturalistic, mechanistic viewpoint with which the scientific community views the world
    – michael
    May 21, 2018 at 20:24
  • @m.r. That's a separate and different question. Also, your pattern of comments makes it look less like you're legitimately seeking an answer to your question, and more like you're subtly advancing an ID argument, which is not a correct use of this site. May 21, 2018 at 20:32

You must log in to answer this question.

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