Mar 21

The Sherlock Holmes Defense

When you have eliminated the impossible, whatever remains, however improbable, must be true. – Sherlock Holmes (Arthur Conan Doyle, The Sign of the Four)

What must the cause of the universe be like? According to William Lane Craig, it must be spaceless, timeless, and personal. In effect, it must be a non-physical, non-temporal mind with causal efficacy and unimaginable power. Notice the word, “must.” Craig isn’t just saying he thinks it’s probable – he thinks it can be shown necessarily through argument. This point generally comes as an adjoinder to the Kalam Cosmological Argument, which Craig explains here.

But does this make sense? This is the question I’ve been debating in the comment thread of On Absurdity: William Lane Craig and Actual Infinites. How can we realistically assert some kind of mind as the cause of the universe?

The argument goes essentially as follows:

As the cause of space and time, this cause must be outside of space and time. If the cause is non-spatial and non-temporal, it must be changeless and immaterial. Now, Craig argues, the only entities which can be timeless or immaterial are either minds or abstract objects. However, abstract objects do not possess causal efficacy. Therefore, it must have been a mind. Next, only a free agent can account for the origin of a temporal effect from a timeless cause. So, it must be personal. So, there you have it – the cause of the universe must have been an immaterial mind.

We arrive at this position from a dilemma. This allows Craig and his defenders to invoke what I’m calling the Sherlock Holmes defense. They boil it down to two options and dismiss one of the options as intuitively absurd. So, however much you may be suspicious of the immaterial mind, it’s all that is left. It must have been a mind no matter how improbable you find that conclusion to be! Let’s consider this dilemma further. Either we have a false dilemma, which we can dismiss, or we have a true dilemma, which I will argue still provides no grounds for a conclusion.


What if it is a false dilemma?

In my opinion, the likely conclusion here is that we have arrived at a false dilemma. There are several suspect assumptions that bring us to Craig’s conclusion.

First, we have to decide whether we consider our own universe to be “all of space and time” or whether it is just a universe within some form of external space and time. Craig, to his credit, does attempt to engage in modern science and cosmology to provide support for his premises. He invokes the Big Bang as evidence that all of space and time began at a single point and moment in the finite past. Very few scientists question the Big Bang, though it is far from certain that this is all that exists. For example, M-Theory provides a look into what possibly existed prior to the Big Bang. Proponents of M-Theory might assert that we are a bubble in a sea of universes. This would provide a natural solution to the problem that remains within space and time.

But Craig could then move back and question the cause of all of those universes. He would assert we still have to have a beginning since there could not be an actual infinite. This argument depends on a series of paradoxes – the kind that philosophers love and most people on the street say, “Who cares?” Craig is best known for invoking Hilbert’s Paradox of the Grand Hotel, although, there are a few others he uses. The idea is that actual infinites lead to absurdities, which we can deal with in math, but they simply could not exist in reality. It raises several questions, including, “If the past was an actually infinite series of events, then how would we ever arrive at this moment?” It’s certainly an interesting question, but it is far from decisive in showing this could not have been the case. For example, see a few of Wes Morriston’s responses on actual infinites here and here. These are simply provided to show that the impossibility of actual infinites is questioned among philosophers.

Additionally, Craig’s arguments depend on a certain theory of time. Craig defends an A-Theory of time, but the argument falls apart under a B-Theory. If you don’t know those terms, don’t worry; you’re in good company. This article by the Stanford Encyclopedia of Philosophy will get you started with an overview of the issues. B-Theory strikes people as a bit odd – it’s not intuitive – but many feel that special relativity leads us to this view of time. If time is like space, then it is tenseless. It can be measured in relation to other points, but there is no true cardinality.

So, if we could have a beginningless past containing a series of physical, temporal causes, then the problem is gone. The dilemma is not needed, and we can dismiss the notion of an abstract immaterial mind as the creator of everything. This is the route I would take, and I’ve provided some routes to explore in your discussions which I think are fruitful. But let’s delve deeper and consider what would happen if we could not successfully argue any of these points.


What if it is a true dilemma?

What if we really do arrive at the decision to make with either a mind or abstract objects, like numbers, acting in a causal role?

I think we are only justified in reaching a standoff with these options. There is no evidence that either thing acts in a causal role. In fact, the very existence of both things is questionable.

But the point of the Sherlock Holmes defense is that it doesn’t matter whether you think a causal, non-spatial, non-temporal mind cannot exist. What matters is that we have a dilemma and we have eliminated one option from consideration since it is impossible. It must be the other option.

In my discussion mentioned earlier, I suggested that minds do not exist apart from brains. I was challenged to show why this was impossible. In the world of logical possibility, this is pretty difficult. That is essentially the point of Russell’s Teapot and the Flying Spaghetti Monster. Now, let’s turn the tables a bit. If I am challenged to show that minds cannot cause, then it certainly seems fair to challenge Craig and his defenders to show that numbers cannot cause.

The only real defense I’ve heard of this is that “of course numbers don’t cause anything!” A lot of people probably find this persuasive, but we need an argument that cannot be also leveled at minds. I could say, “of course minds only exist when there are brains!” Even with Artificial Intelligence, there would be some physical “stuff” grounding the mind.

So, this is the challenge for the Craig defenders: Provide an argument against numbers that does not apply to immaterial minds. Without such an argument, this dilemma goes nowhere.



To simply assert this dilemma begs the question by assuming minds can play a causal role. We must have some reason to think this. The reason given by defenders is the initial argument I described. Notice this is not empirical evidence; rather, it is a rationalist approach to deduce it purely by logic.

To combat this, I have provided some ways of attacking the premises of the argument to show we never even reach this dilemma and are justified in concluding we have no reason to think minds can cause (or even exist) without brains, at least not in virtue of this argument. If we do not have to reach the dilemma, then the argument does not concern us. This seems like a very reasonable conclusion to me. If we do reach the dilemma, then we are in the very odd stalemate of wondering whether a mind or some abstract object, like pi, caused all of space and time. But, as I’ve said, we really have no reason to prefer one over the other. Intuition about such matters is irrelevant.

Now, there is nothing inherently wrong with the Sherlock Holmes defense. It’s perfectly valid. However, how often do we find ourselves in a position to eliminate all possibilities save one? This type of argument must have such limited scope. We live in a world of uncertainty, of induction, of inference – to deal with such things requires tools equipped to handle uncertainty. This is precisely why Bayes Theorem has been so successful in gaining support. About such ideas as timeless, spaceless, infinity, and origins of everything, I find it very hard indeed to seriously entertain an argument about certainty. I will be suspicious every time of claims that such things can realistically be deduced. The defender may think he or she has made a Sherlock Holmes breakthrough, but what is the messy trail of assumptions, fraught with uncertainty, that led to that final point?

Similar Posts:


1 ping

Skip to comment form

  1. M Gehlke

    There literally is no argument which can be leveled at mathematics which would not translate into an argument against free minds, mathematics being a general enough symbology that you could embed (in theory) the mechanics of human thought, and thus any “mind” we could articulate is equivalent to the actions of some abstract mathematical object. Additionally, things not human articulable are likely also mathematical in some superset of human mathematics, but it’s rather hard to say anything about such a class of objects, if it exists.

    I have a long standing challenge: show me a concept that isn’t inherently mathematical in nature. I suspect that much of the debate over free-agency and the like is a macro-level confusion about how the universe operates.

  2. Mike

    I think I would agree with your conclusions. Although, would Godel’s incompleteness theorems be an example of concepts that cannot be covered by mathematics? I’m not really sure.

  3. M Gehlke


    The Incompleteness theorems are more of the flavor that we can’t accurately determine all of the behavior of a mathematical system. One of two things happens when we try: we had a contradiction in our axioms and so everything is true and our categorization of true things is the class of everything; or else there are true statements we cannot verify are true, by proof or mechanical action (ie, there’s no function that returns 0 for only true propositions).

    Ironically, it was only the rigorous study of what was and wasn’t possible to embed in mathematics that led to the Incompleteness theorems: Godel devised a system for converting propositions about mathematics into unique mathematical statements – an example of embedding human thought into a mathematical structure.

    The incompleteness theorems basically condense down into translating the idea “This statement can’t be proved” into mathematics, then showing that for any axiomatization of mathematics, such an idea must be embedded somewhere, and hence at least part of the structure must be indeterminate or contradictory.

  4. Hugo Juggernaut

    God describes himself as WORD, SPIRIT, LOGOS and it’s through this WORD he created the universe. He says Jesus is the Word in the briefest expatiation on existence in John 1. Understand this and you are on your way.

    This is not something without analog in human experience. We create universes too, with words: We call those who do so Authors.

    As God said, “Let there be Light” so we too think of some creation and bring it into being.

    God, in this respect, is no different. We are, after all, created in His image.


  5. Francois Demers

    Mmm. Most post-newtonian theories of Physics assign a fairly low probability to the reality of space and time, preferring to consider them perceptual side-effects of observing motion. For convenience’s sake, motion is easier to deal with if it is assumed to happen somewhere in some places. From the point of view of a photon, admittedly not easy to adopt, there is no time between emission and reception. Thereby no space as it would allow the photon to travel at infinite velocities. what the emission and reception of a photon does is make its emitter and receptor contiguous in the manifold.

    This could change inside the event horizon of a black hole: a photon has no mass at rest. If it is eventually “stopped ” by the singularity, then gravity has no effect on it and its degrees of freedom become infinite. However, if it escapes the singularity, the it acquires momentum, and thereby mass, and thereby becomes subject to gravity again. It is possible to imagine that inside the event horizon, the photon exists in two states: massless and free / massive and captured simultaneously. Or that it oscillates between the two states. This would create “real” time and space but would be forever outside the domain of observation.

    If the cosmological constant is indeed higher that 1, the implication would be that time and space had and will have “real” reality only at the instants of the bing bang and collapse and those two events would be simultaneous (better: one and the same) . everything in between would be just variations in the distribution of energy in a conventional space that has no dimensions.

  6. Mike

    Francois, I’m not sure I understand. I would agree for time, as it seems to be merely the measurement of some change, but space too? How can Einsteinian space be curved if there is no real dimensionality? Perhaps you could clarify.

  7. Francois Demers

    Mike, I apologise for not clarifying.

    All I wrote above is highly speculative and holds (shoddily) only if the observer is one photon. I doubt there is a sane physicist who would extend it to allow for any other observer or a plurality of observers.

    Space and time are useful conventions, certainly. However, it is difficult to demonstrate that they are real. Intuitively, we know space and time exist but they bravely resist mathematical proof.

    In extreme cases, space and time become meaningless: if the expansion of the universe is endless (cosmological constant < 1) it tends to a state of maximum entropy. At 0 degrees Kelvin, it is dimensionless and its state virtually identical to that of the instant prior to the Big Bang. "Virtually" is important here: there can be no outside observer unless you want to postulate "something that exists and is not part of the set of everything that exists". That would be dangerously close to a metaphysical creator.

    You can also hypothesise the same conditions at the singularity of a black hole without walking too far away from actual science. Again, there can be no observer.

    All conceivable conditions of the universe that are observable need motion to have dimensions, space and time. So, do space, time, and dimensions have an intrinsic reality?

  1. The Clueless Argument for God's Existence | Foxhole Atheism

    […] Edit: I've added some further thoughts on this type of argument in my post The Sherlock Holmes Defense] Bookmark on Delicious Digg this post Recommend on Facebook share via Reddit Share with Stumblers […]

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Notify me of followup comments via e-mail. You can also subscribe without commenting.