If the alternative that's being offered to what physicists now talk about - a big bang, a spontaneous singularity which gave rise to the origin of the universe - if the alternative to that is a divine intelligence, a creator, which would have to have been complicated,[1] statistically improbable, the very kind of thing which scientific theories such as Darwin's exists to explain, then immediately we see that however difficult and apparently inadequate the theory of the physicists is, the theory of the theologians - that the first course was a complicated intelligence - is even more difficult to accept. They're both difficult but the theory of the cosmic intelligence is even worse. What Darwinism does is to raise our consciousness to the power of science to explain the existence of complex things and intelligences, and creative intelligences are above all complex things, they're statistically improbable. Darwinism raises our consciousness to the power of science to explain how such entities - and the human brain is one - can come into existence from simple beginnings. However difficult those simple beginnings may be to accept, [2]they are a whole lot easier to accept than complicated beginnings. Complicated things come into the universe late, as a consequence of slow, gradual, incremental steps.[3] God, if he exists, would have to be a very, very, very complicated thing indeed. So to postulate a God as the beginning of the universe, as the answer to the riddle of the first cause, is to shoot yourself in the conceptual foot because you are immediately postulating something far far more complicated than that which you are trying to explain. Now, physicists cope with this problem in various ways, which may seem somewhat unconvincing. For example, they suggest that our universe is but one bubble in foam of universes, the multiverse, and each bubble in the foam has a different set of laws and constants. And by the anthropic principle we have to be - since we're here talking about it - in the kind of bubble, with the kind of laws and constants, which are capable of giving rise to the evolutionary process and therefore to creatures like us. That is one current physicists' explanation for how we exist in the kind of universe that we do. It doesn't sound so shatteringly convincing as say Darwin's own theory, [4] which is self-evidently very convincing. Nevertheless, however unconvincing that may sound, it is many, many, many orders of magnitude more convincing than any theory that says complex intelligence was there right from the outset. If you have problems seeing how matter could just come into existence - try thinking about how complex intelligent matter, or complex intelligent entities of any kind, could suddenly spring into existence, it's many many orders of magnitude harder to understand. Lynchburg, Virginia, 23/10/2006

[1]I'm really curious as to what exactly the area of statistics can say about the existence of God, improbable according to who and what exactly? Even if you are to believe his assertion as true, does improbability exclude things from the existence or can improbable things also happen?

[2] How does this exactly work that the complexity of a thing has any bearing on whether it exists, I don't see it. And how is evolution a less complicated explanation anyway?

[3] I'm sure being hard or impossible to understand does not have all that much to say about whether things exist.

[4] You will excuse me if I don't take your word on it, not so self-evident to me.

He sure has some oddball assertions and I wonder where he gets some of his ideas from, I don't know what kind of logic is at work here but it does not sound to me like the good kind.

  • 2
    1) agreed: the idea of God Dawkins is speaking about seems to be a sort of "scientific" God-like thing, to be evaluated according to scientific procedure. The idea of God of e.g. rationalist "classical" philosophers (Descartes, Spinoza, Leibniz) is not "statistically improbable". Jan 8, 2020 at 11:51
  • 2) in the same vain, if we look at evolutionary biology or thermodynamics, "simple" facts/creature are less complex than later ones. This is true for the phisico-biological world; why must it apply also to an omnipotent/omniscient/eternal being ? Jan 8, 2020 at 11:52
  • In conclusion, the author is an atheist (it's ok) scientific-minded (it's ok) that is trying to "prove scientifically" that the idea of God is "contradictory/absurd" etc. (which is impossible). Jan 8, 2020 at 11:55
  • @MauroALLEGRANZA How do you define biological complexity?
    – Cell
    Jan 8, 2020 at 14:18
  • 1
    @Cell Biological complexity is defined in terms of measurement. Humans have more cells than molluscs; humans have more differentiated systems than molluscs; humans have organs that are larger and more articulated than molluscs; the behavior of humans is combinatorially larger than molluscs. In no sense is the A&P of a human equivalent or simpler to that of a mollusic.
    – J D
    Jan 8, 2020 at 17:12

3 Answers 3


After having read a lot of Dawkins, I would put the following gloss on the big overarching argument he has developed over the entirety of his oeuvre:

  • P1 - The only argument for God worth taking seriously is as the intelligent designer of the complexities of life.

  • P2 - The complexities of life are better explained by evolution than by God.

  • C - Therefore there is no argument for God worth taking seriously.

He spends a lot of time on P2, but tends to take P1 as self-evident and therefore not requiring support, which is, to say the least, controversial.

The piece you quoted is a more fine-grained argument parallel to the coarser one:

  • P1 - In general, things can only be created by things that are more complex than themselves.
  • P2 - The one exception to P1 is through the process of evolution.
  • P3 - The more complex something is, the less probable it is that it could come into existence without an adequate explanatory process.
  • P4 - God must be maximally complex in order to have created the universe
  • C - therefore God must be maximally improbable.

P1/P2 and P4 strike me as the controversial premises here.

It's worth noting here that Dawkins' reputation is higher outside the philosophical and scientific communities than within them. It's generally understood, even among philosophers sympathetic to his conclusions, that his arguments are not particularly rigorous. His role in the world of ideas is as an influential popularizer of religious, philosophical and scientific concepts.


You're missing his point: which is that scientific explanations of intelligent life are better explanations than 'God'. So when you say

I'm really curious as to what exactly the area of statistics can say about the existence of God

You've glossed the statistical probability of 'evolution' -- and his claim that "complicated things" need to be explained via gradual changes -- with atheism.

Of course he is also an atheist, but it does you no favors.

  • There is no scientific explanation of intelligent life. There is an explanation of evolving complexity, but this is not the same thing. ,
    – user20253
    Jan 9, 2020 at 13:08
  • @PeterJ [Evoutionary psychology] deals thoroughly with the origins and nature of intelligent life, and cognitive science even more broadly. You may reject their explanation of intelligent life, but that doesn't make it not an explanation.
    – J D
    Jan 13, 2020 at 1:21
  • @JD - Do you have a reference? I wonder how they explain intelligence without explaining consciousness, and what exactly is meant here by 'explain'. I wasn't aware they'd even explained why anyone bothers to survive.
    – user20253
    Jan 13, 2020 at 11:55
  • @PeterJ Blackwell's Companion. Chapter entitled "Explanation", p. 127. For intelligence, a good start would be Gardner.
    – J D
    Jan 13, 2020 at 17:59
  • Intelligence is not only a phenomenon, but actionable; see any introductory work in organizational psychology or educational psychology.
    – J D
    Jan 13, 2020 at 18:02

I'm not particularly a fan of Dawkins, but it's worth the time to consider his position fairly. His argument boils down to this:

  1. Complex things are developed from (and thus come after) simple things
  2. God (should God exist) would seem to be an exceptionally complex thing
  3. Therefore, God cannot come before the less complicated things that compose the universe.

There's a confusion of physicalism and metaphysicalism here, as though we can blithely measure the complexity of a putative god on the same scales and dimensions that we measure the complexity of material substances, but it's not a horrible argument on the face of it. I take him as using the word 'probability' in the loose colloquial sense of 'likely' or 'unlikely,' not in the analytical sense of statistical measurements. Given his presumptions, then yes, it would seem odd that a highly complex entity would exist before even the simplest particles are formed. It's not a question of God being hard to understand; it's merely that God would have to be capable of function on a scale of complexity far beyond anything we imagine, at t-minus-nothing.

And again, given his presumptions, Darwin's work does appear self-evident. Darwin posits an adaptive world, in which creatures complexify themselves over time to face and overcome new environmental challenges.

To a certain extent, Dawkins is guilty of pseudoscience. He makes these broad, grand assertions without operationalizing them, defining his terms, or bothering to provide evidence. For instance, it is not at all clear how he is measuring 'complexity.' Genetically speaking, trees are far more complex than human beings, yet trees existed long before we did. Is that a problem? There is — oddly enough — a kind of quasi-religious concept of mankind as the pinnacle of creation lurking in Dawkin's work, which I think really gets at the root of Dawkin's ideology. He wants to assert that mankind if the pinnacle without God, not because mankind is a reflection of God.

  • he could be applying the principle of evolution to God, but it would be a very ugly argument to apply genetic trends -- rather than universal features of 'explanation' -- to the existence of a creator God. dawkins isn't a very clear writer, it seems
    – user38026
    Jan 9, 2020 at 11:44
  • The major issue seems to be Dawkins' assumption that God was of the same 'kind' as our physical universe, following the same laws. This assumption hardly works as an argument against theists, especially considering traditions like Spinozism (God is substance, the universe only antecedent) and Tsimtsum of the Kabbalah (God created the universe out of Himself as something that is different from Him) being based on the opposite. It's like saying "God cannot be like you say He is and if He's not, He probably does not exist" without defending the first, central premise at all.
    – Philip Klöcking
    Jan 9, 2020 at 12:34
  • +1 A much better answer than mine. But where does P2 come from, and why is it necessary? P2 creates a straw.man that is easy to knock-down, and the qualification 'would seem to be' is mealy-mouthed. Okay, so God cannot be complex. This is the start of an investigation, not the end of one.
    – user20253
    Jan 9, 2020 at 13:04
  • 2
    @another_name — Well, Dawkins isn't really a philosopher: more of an intelligent pundit with an axe to grind. And I don't think he's saying that God evolved in the biological sense. He's abstracting the principle of evolution to a universal tendency (in the way we might say that heavy elements 'evolve' from light elements in the furnaces of stars). Jan 9, 2020 at 16:55
  • @PhilipKlöcking — But you have to contextualize Dawkins properly. He isn't really thinking about 'God' in that philosophical way. He's reacting to the anthropomorphic, personalized, authoritative conception of 'God' presented in conservative branches of the Abrahamic faiths, particularly in the aggressive fundamentalisms of Islamism and Rightist Evangelicalism. His work is part of a political conflict over whether civic moral authority should be religious or secular; philosophy is at best irrelevant to him, and at worst mere collateral damage. Jan 9, 2020 at 17:06

You must log in to answer this question.

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