If hard determinism is true, and our thoughts are merely the results of a causal chain of atomic interactions, are reason and logic illusory?

This matter has likely been discussed in another thread, but I couldn't find it. I was led to ask the question after John Lennox raised the issue in his debate with Richard Dawkins here, (approx 40:20). He claims rationality in a deterministic universe is logically incoherent.

EDIT: The answers so far have been very useful. As has been pointed out however, I misrepresented Lennox. He did not refer to deterministic processes. His actual words were:

"If in the end my beliefs, my theories, my scientific theories, are the results ultimately of the motions of atoms in my brain, produced by an unguided, random, mindless process why should I believe them?".

The answers to date acknowledge this and describe the implications both of random and deterministic processes, but this edit is important should anyone else wish to weigh in.

  • 1
    Please specify the time of the relevant passage in the video. Sep 26 at 10:31
  • 2
    No, the conclusion is too strong. One can argue that if our reasoning is predetermined then there is no causal nexus between how the world is and how we reason. But, depending on how the deterministic laws are set up, they can equally well make our reasoning inevitably misleading, inevitably sound, or anything in between. It is true that determinism undercuts the intuitive justification for soundness (bad reasoning leads to bad consequences and is corrected as a result), but we may be compelled to reason soundly anyway.
    – Conifold
    Sep 26 at 10:56
  • 3
    Btw, Lennox reprises C. S. Lewis's argument from reason, but he is not doing it very well. Here is Lewis:"those who ask me to believe this world picture also ask me to believe that Reason is simply the unforeseen and unintended by-product of mindless matter at one stage of its endless and aimless becoming. Here is flat contradiction. They ask me at the same moment to accept a conclusion and to discredit the only testimony on which that conclusion can be based."
    – Conifold
    Sep 26 at 11:05
  • 2
    @Conifold Can you reference any actual determinists who think that determinism is incompatible with causal chains? "Bad reasoning leads to bad consequences and is corrected as a result" seems like a statement any determinist would agree with, as a summary of the mechanism by which evolution resulted in human rationality. Your criticisms are coming from outside the deterministic perspective, not from within it.
    – causative
    Sep 26 at 11:38
  • 4
    Reason and logic are deterministic processes anyway: once we have accepted premises and a sound argument we have no choice but to accept the conclusion. Nobody can decide that the angles of a flat triangle sum up to something else than 180 degrees. Hard determinism has nothing to do with it, and this argument sounds more like a lazy cope out than a serious objection.
    – armand
    Sep 26 at 12:27

John Lennox's idea is that we would have no reason to believe our own logical conclusions if nature, and therefore the brain itself, was deterministic.

He says (40:15 - 40:48) that there would be no reason to trust our own logic if was based on the "unguided" and "random" processes of a materialistic world.

First, John Lennox, during the debate with Dawkins, does not justify his conclusion, but he clearly assumes that belief is somehow not a function of the brain, even though this contradicts his hypothesis of a deterministic universe, for, in a deterministic universe, belief can only be a function of the brain, and in that case, we do not have the choice but to believe whatever our brain believes, so to speak. And, presumably, what our brain believes would be consistent with whatever its own logic would be. So, Lennox is being illogical in his reasoning--unfortunately, Dawkins does not point this out.

At another level, logic is a property of the brain and as such the result of natural selection, and thus it is a cognitive capacity adapted to our environment, and therefore adapted to at least a part of our universe. In other words, we better trust it, somewhat like water better trust that the bed of the river will lead to the sea.

So, yes, it is true that we cannot trust that our logic is adapted to the whole of reality, but we have no better alternative, and therefore no real choice. We could opt for irrationality, and on occasions it will work, but mostly it won't. If our logic suddenly consistently failed to help us, for whatever reason, we would have to try something irrational, and essentially, work out a new logic through a renewed process of natural selection. We can only hope that we never need to try that, because it would hurt a lot before it could get better.


"If hard determinism is true, and our thoughts are merely the results of a causal chain of atomic interactions, are reason and logic illusory?"

In the sense Lennox is talking about, no. Reason and logic can arise deterministically if you set up a system that deterministically evolves by pruning out anything irrational and illogical.

Lennox relies on the strawman argument that because evolution by natural selection involves a "random" mechanism, not guided by any intelligence, that the result is necessarily random and unintelligent. If the rules and principles by which my reasoning worked were selected randomly, from all possible rules, it would be highly unlikely to generate reliable conclusions. It's a variant on the Argument from Design.

But evolution by natural selection is not random. It consists of two parts: random variation, and highly non-random selection. It is the selection that contributes the 'design'. A good analogy for it is topiary, the shaping of plants and bushes into artificial sculpture-like shapes. The plant grows randomly. But the gardener cuts off any branch that crosses a chosen line in space. The result is a plant that grows in a highly non-random form that precisely avoids crossing the lines. How, you might ask, could such a shape so perfectly suited for survival arise by the random growth of twigs and branches? How did the plant know so precisely when and where to stop growing? That's the wrong question! The source of the 'design' in natural evolution is not random mutation, but in the highly non-random process of survival of the fittest.

We can trust our evolved logic and reason to give us reasonably reliable answers about the world because those alternatives that gave unreliable answers all died out long ago. Only accurate logic and reason enables us to survive. Our minds are shaped by the dangers in our environment. And it is only by such a process that the genesis of reason and logic can be explained. Intelligent design simply shifts the question back a step, you have the same problem to explain the designer.

Besides the question Lennox raised, which was more about physics being unguided than it being deterministic (there is no such thing as 'randomness' in a deterministic universe!), there is a separate question whether hard determinism poses any obstacle to human reason and logic. Here we run into questions of 'free will' and the meaning of 'counterfactuals', on which human reason in aid of decisionmaking is often supposed to rely, but which are commonly held to be incompatible with determinism and materialism. That's a far more complicated debate.

Human reasoning builds simplified predictive models of the world. It divides it up into 'objects' which have their own models. A model takes a set of initial facts about the world as input, some known, some unknown, and some of which are under our control, and generates a predicted outcome as output. The problem-solving mind can then run through all the possible inputs to find which ones are most likely to give a desirable output, and do that. We model counterfactuals - things that could happen but don't - as an inherent and essential part of our reasoning process. But in a deterministic universe, the idea of "something that could happen but doesn't" is a nonsense! There is only one thing that could and does happen.

We model other humans and animals recursively, as "agents". This treats minds as black boxes that make decisions autonomously, and we use our own mind, with a modified set of inputs to account for their different goals and knowledge, to model theirs. Again, the model involves a range of hypotheses about things they could do but eventually choose not to. Our model of the world includes 'free will'.

The process of taking a model of the world, exploring all the possible alternative inputs (things we don't know, future actions), and picking the best of the outcomes can be done as a deterministic algorithm. We loop through the variables, evaluate each against our desires and goals, and deterministically select the one that works best. But this deterministic algorithm contains representations of counterfactual alternatives and agents with free will, which (depending on how you define them) in a deterministic universe are impossible.

So in this sense, in a deterministic universe there are certain features of our reason and logic that are illusory. But the reason and logic themselves are not. They still exist, and function. They still work very successfully, in the evolutionary sense of giving answers accurate and reliable enough for survival. But they are only a crude approximation to reality, ones that only work most of the time, but can give false answers if pushed beyond their domain of validity.

The debate goes far beyond that simple picture. There are alternative ways to define terms to make determinism compatible with free will. You can find further information on the debate on Compatibilism starting here.


"If hard determinism is true, and our thoughts are merely the results of a causal chain of atomic interactions, are reason and logic illusory?"

Lennox is not speaking about "hard determinism" in the section you've referenced. He speaks of a process that is "unguided" and "random", calling attention to the blindness of material processes. I am not aware of any supposed conflict between determinism and reason/logic, though determinism is thought by some to render human experience illusory as it relates to freedom of the will (see section 6 of the Stanford Encyclopedia's entry on Causal Determinism).

Lennox is offering an evolutionary argument against materialism, along the lines of Alvin Plantinga's evolutionary argument against naturalism. He is not arguing that reason and logic would be illusory under materialism, just that materialism would narrow the bounds of their usefulness to the point that materialism itself would be rendered incoherent.

The difficulty for materialism is this: assuming that evolutionary theory is correct, we can be confident that we perceive what we need to perceive in order to survive. But what justification do we have for supposing that the faculties our species has developed to protect us from the dangers of the material world should reliably inform us of anything other than the dangers of the material world? If materialism were true, how would Lennox or Dawkins be equipped to say anything credible about whether or not there is a God? How could their disagreement be significant if their brains were merely make-me-capable-of-perpetuating-the-species thingies? If survival capability translates to knowledge of reality, why place two humans behind the podiums rather than a dandelion and a jellyfish?

"[Lennox] claims rationality in a deterministic universe is logically incoherent."

Not quite. What he said is that "an argument that purports to derive rationality from irrationality ... is logically incoherent." Theism and determinism are not logically incompatible. It may seem fanciful or silly, but the notion of an all-powerful rational being having the ability to create a deterministic universe containing rational creatures is logically straightforward.

  • 2
    It could be an effect of my materialist brain, but I am under the impression that humans are better problem solvers than dandelions and jellyfishes. If the brain that evolved to make fire with two sticks is good enough to design spaceships that reach far off planets with dazzling accuracy, then it must be able to reason quite good. A lucky side effect of evolution. As for if it can reach conclusions about god, it's exclusively a problem for theists. Agnostic atheists like Dawkins are perfectly fine with the idea that we can't know god, and therefore can't know if it exists.
    – armand
    Sep 27 at 6:35
  • @armand fire and space travel are ways of interacting with the dangers of the material world, right? Under materialism, we should expect to be good at those things, and only good at interacting with non-material realities insofar as it aids our understanding/avoidance of material dangers. If that is what our problem solving abilities amount to, then it's worth pointing out that there are other creatures that have solved these problems as well, and much more efficiently than humans. It's not only theists who recognize the need to reason beyond those constraints. Sep 30 at 1:57
  • I don't think there are such animals, and if there are they do so by means of their biological bodies, not by reasoning. I.e. A chameleon is good at camouflage, but it's not by inventing a cloaking device like we did. So yes, our brain is good at interacting with the world and every aspect of our daily life proves so. Lennox then argues "but why should we trust it to solve metaphysical problems ?", and far from being an objection, this plays straight into the hand of atheists: yes, we shouldn't. Let's stop hair splitting about God, the meaning of omniscience or the trinity, and start living.
    – armand
    Sep 30 at 3:49
  • @armand materialism holds that reasoning is "by means of [our] biological bodies," so I don't think you fully appreciate the implications of that point, but it was an off-hand illustration anyway. What's more significant is that you seem to be under the impression that God is the only subject of concern here. These debates command public interest because both sides understand that justice, morality, the value of human life relative to other aspects of nature, etc. are equally at issue. Even most atheists are not content to "start living" as though none of those things matter. Oct 3 at 21:48
  • I am not referring to god only, but very clearly to all of metaphysics. All that can't be proved to "work". Justice, morality, the value of human life are problem that can be addressed in a down-to-earth, non metaphysical way. Not discussing endlessly about what is "the Right" or "the Good", but just try and see what types of behavior make for a desirable society. The value of life is a no brainer: I value my life, I want people to value it, how about we all value each other's life? Simple as that. We don't need metaphysics, so if our brains can't handle it it's a non problem for materialism.
    – armand
    Oct 3 at 22:58

I have since learned that Dawkins addressed a similar question in another conversation he had with Lennox, this time at the Oxford Museum of Natural History: Has Science Buried God? (14:10). It supports some of what has been posted in other answers.

Lennox: "It seems to me that your atheism undermines the very rationality that I assume and you assume when we go and study the universe".

Dawkins: "It seems to me a quite absurd thing to say that because we are saying that our minds are produced by brains and brains evolved by evolution by natural selection, therefore that somehow undermines our ability to understand anything. Why on earth should that be? Natural selection builds brains which are good at surviving and brains that are good at surviving are good at surviving in the world".

Lennox: "But where does the concept of truth... how do they come to recognise things like truth if those things are simply reducible to things like chemistry and neurophysiology, how do they serve truth?".

Dawkins: "Truth is what happens.... An animal that is attempting to survive and that didn't recognise truth or falsehood in some sense, at whatever level is appropriate for the kind of survival that it has... it wouldn't survive. I mean truth just means that you're living in the real world and you behave in the real world in such a way as things make sense in the real world. When you see a rock in your way you don't go charging into it. You'd die if you did that. If you jumped over a cliff you'd die. That's truth. It's perfectly obvious that natural selection would favour, in any animal, a brain that behaves in a way that recognises truth and acts upon it".

  • Dawkins' response really shows how Lennox' objection is only a problem for theists. Truth is what works. Dawkins speaks only of very mundane things, but even very elaborate concepts handled by the human brain, like quantum mechanics have at least enough validity that microchips work. Computers prove it billions of times per second. Once truth that can be verified to work is put aside, what Lennox is really arguing is "why should I trust my metaphysical beliefs, including my belief in God?" To which we can reply "Exactly. Why should you?"
    – armand
    Sep 28 at 3:51

Computers are the embodiment of deterministic rationality.

If computers are programmed properly for rational analysis, determinism is no handicap.

However, if the computer was running an app consisting of random lines of code, it would be foolish to trust its logic. Even the discipline of evolutionary programming requires careful construction of evaluation functions and of the rules of how programs are modified in successive generations. And even then, if the domain is sufficiently complex, there will be some skepticism about whether the final product is 'both complete and correct', or whether some untested execution paths would fail to work properly (or 'rationally').

Likewise with people - the question behind trustworthy rational thinking is whether the thinking infrastructure was reliably designed and whether it is now functioning well.


Determinism is only a theoretical idea, a simplified model of reality. It has no truth value. It is logically impossible to claim or believe that determinism is true, as determinism excludes concepts like claim or belief.

Determinism has no effect on reason or logic. But reason and logic categorically deny determinism.

  • 1
    as determinism excludes concepts like claim or belief. But reason and logic categorically deny determinism. You'd need to support these statements with something more...
    – AnoE
    Sep 27 at 12:23
  • @PerttiRuismäki "reason and logic categorically deny determinism" If logic somehow included any a priori truth about the universe, then humans would all agree to it, and in the case of determinism, different humans believe different things. Ergo, logic does not tell us in any way whether determinism is true or false. Logic is not the sort of things that could "deny". If anything, it is agnostic about the universe. Sep 27 at 16:04
  • A deterministic universe is a logical impossibility. Determinism ironically excludes both methods to determine anything, like the contents of a universe. A deterministic universe could not evolve or pop up randomly and it could not be deliberately created either. Determinism is not a theory or a belief. Determinism does not describe reality, it does not explain anything. It is neither true or false. It is just an imaginary construct with some practical use in classical physics. Sep 28 at 4:52
  • As always, just repeating your point won't make it more true. Can you support it with arguments?
    – armand
    Sep 28 at 5:42
  • @armand My argument is the definition of determinism. In determinism every event is uniquely determined by previous events. There simply is no room for probabilistic randomness or any kind of volition. Under determinism the laws of physics are completely different from reality. There is no concept of alternative, everything happens with 100% certainty. Therefore determinism cannot be considered as an alternative. It has no role in philosophy. Sep 28 at 6:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.