Saturday, 25 March 2017

That Kalam argument

Sooner or later as an atheist on the web, you're going to run into the Cosmological Argument.  This often comes in the form of the "Kalam argument" or its modern form promoted by William Craig.

It's typically presented as a syllogism. This a logical argument that used two related premises to reach a conclusion.

  • P1: Whatever begins to exist has a cause;
  • P2: The universe began to exist;
  • C: The universe has a cause
A Rapidly Inflating Singularity Explains Many of the Attributes of this Universe
(Figure via Shutterstock as Stock illustration ID: 489781135)

As the argument does not include any deities, it is imputed that this cause, must itself be uncaused.  Conveniently theists knew what this cause was all along. This is their favourite deity, which by fiat is eternal, existing outside time and space and, uncaused.

The popularity of the syllogism doesn't disguise its flaws.  It is these flaws that have prevented a stampede of atheists toward Christianity.  Let me elaborate.

A syllogism depends on the its premises being correct.  It is not good enough for them to be possibly correct.  And there are reasons to suspect the premises are not.

Let's look at the first.

How do we know that everything that begins to exist has a cause?

  • The reason we have the qualification "begins to exist" is to exclude deities with eternal properties.  If it was just 'exists' then the syllogism would net in gods too.  In short, it is a special pleading fallacy introduced right at the start.  How do we know deities are able to exist without a beginning?  We don't.  No evidence is attached to prove this.  It's just one more thing we have to believe about gods on faith alone.
  • It's an uncertain premise.  From what we understand about quantum mechanics, the quantum world behaves stochastically.  It's a random world at that level. Negative and positive sub-atomic particles wink in and out of existence. 
    • The above means that P1 is not self-evident
    • Induction is not strong enough to prove that P1 is correct. We've sampled a tiny fraction of the universe and our observed sample size of universes is still stuck at 1. 
  • It makes the term 'cause' do a lot of work. Here the syllogism tries to conflate causality (in the sense of purposely caused by an agent) with other causality in other contexts (e.g. occurs naturally without intercession because of environment and natural laws). As a term, causes are not really part of the conceptual toolbox of Fundamental Physics.  I'm not convinced at this level, cause is an appropriate term.
  • It is also a category fallacy. Universes don't fall into the same category as phenomenon within the universe.  
  • Similarly, it's not clear what we mean by 'begins to exist'. If time is an emergent property of the universe, then it becomes difficult to talk about a point where the universe 'begins to exist' (see the Hawking-Hartle no-boundary condition).  
In short, P1 is a barely coherent and fallacious premise that has significant scientific objections to overcome.

How do we know the universe began to exist?

We don't. The standard model of the universe converges to a singularity, where the regular laws of physics as we understand them don't apply.  This could be a genuine beginning, or it could be just the observable part of other ways the universe evolved. 

Alternatives include a bouncing universe (our universe originated on another one that collapsed, before bouncing back from a singularity), cyclical (the various ekpyrotic models), reproducing from others etc.  

P2 has yet to be proven to be true.

Does the Universe Have Cause?  


In the sense that some event occurred that initiated an inflationary period for the universe? That seems plausible. For instance, some quantum event is potentially able to cause this.  Does that mean we know that this is how it happened?  No.  But there are natural explanations.  We don't have to invoke gods, and I'm not sure why we would.  What falsifiable hypotheses could we test? What predictions does this make?  The scientific model predicts a flat, homogeneous universe with Cosmic Background Microwave radiation. The divine model predicts a bloodgod who is outwitted by a talking snake, so that he has to mate with a virgin to kill his own offspring because two people ate some fruit.  It's not a good way to build credence.

Gods have a problem as they're not mentioned in the syllogism, and as the syllogism is attempting to prove gods exist, it's not an independent proof of said gods. We do know say, that quantum perturbations do happen. The syllogism for gods starts becoming a little circular at this point.

This lack of evidence for gods makes the whole syllogism unconvincing.  

Summary: Syllogisms aren't Evidence

Syllogisms are interesting logical arguments.  We can say if the premises are true or not, and we can say whether the conclusion follows from these premises.   But they're not evidence, they're either logically true or false.  In this case, the Kalam or first cause argument suffers a plethora of philosophical and scientific flaws. 

Let me conclude with my own syllogism to illustrate this point:

P1: Everything that exists has a natural cause (by induction)*
P2: Gods are supernatural 
C: Gods cannot exist or cause anything that does exist.

* E.g. We've replaced many supernatural explanations with natural, and never a natural explanation with a supernatural.


  1. Good analysis. The KCA is dishonest in another respect. It begs the question by implying that there are two sets of things - those that begin to exist and those that don't. The god being promoted is then designated at=s the only allowable occupant of the second set so that it is conveniently there to be inserted as the only allowable cause. Try the same logic but substitute a peanut butter sandwich for the locally popular god to prove the awesome creativity of peanut butter.

    Besides, by what logic was there ever 'nothing' in the first place? How can 'nothing' exist when 'nothing' is defined as not existing? There is no logical basis for assuming the default state of existence is non-existence. This is a logical absurdity.

    As with any religious apologetic which conclude with 'therefore 'God' exists', you can legitimately ask 'Which god?" and not get a logical answer. This invalidates the argument and shows it to have been circular and to have assumed its conclusion a priori. In other words, its a dishonest fraud.

  2. P1: Everything that exists has a natural cause (by induction)*
    P2: Gods are supernatural
    C: Gods cannot exist or cause anything that does exist.

    Yes - every "thing" has a natural cause. God is not a "thing," which is an "inanimate material object as distinct from a living sentient being."

    1. Everything material has a cause
    2. The universe is material
    3. The universe has a cause

    The universe began to exist ~ 13.8 billion years ago. What powered inflation and the start of space, time, and matter? The answer is an eternal,immaterial uncaused cause: GOD.

    1. You miss the point completely. I'm not proving gods don't exist. I'm illustrating why syllogisms don't work as evidence.

      And all you did was redefine 'everything' to exclude your god. What a desperate and dishonest gambit to try!

      Kalam does *not* include deities in any of its premises. Your syllogism does *not* inlcude deities. Dieties are added as an AD HOC device by FAITH alone.

      Please show:
      1) How a quantum fluctuation could not generate the inflationary expansion of a singularity.
      2) your deity is eternal and uncaused.

  3. Please show proof that inflation was powered by a quantum fluctuation. That's an unproven hypothesis, as you very well know.

    Isn't this fun? Maybe you should unblock me now. ๐Ÿ˜‰

    1. You claimed the answer was your deity. That excludes quantum fluctuations. I merely asked to see your proof it wasn't. How odd you can't produce the relevant proof.

      Nor can you offer any evidence your deity is eternal and uncaused. That's kind of important. Especially when the slave-owners that invented your god, didn't do so until the Bronze Age.

  4. One more problem with this post by Another Atheist is that it ignores the philosophical reasoning that the universe cannot be past eternal. This is a particular disappointment since this is Craig's main reason for arguing that the universe had an absolute beginning, with scientific evidence being a backup to that philosophical reasoning. To see how this is actually argued, you can see it from the transcript of a Defenders class by Craig at The most relevant part is when he speaks about Hilberts (Grand) Hotel and the impossibility of forming an actually infinite number of events (there are some other bits in that transcript as well).

    1. Given the plethora of problems outlined in the post, you'll forgive me if I wasn't concerned about providing a refutation to Craig's philosophical 'infinity' argument. The theist's task is to show the premises are true. I'm merely outlining the Herculean task of getting Kalam to work.

      There's a reason Craig's arguments appear on blogs and debates, and not high-ranking, reputable physics journals. His talent lies largely in sifting through a large scientific literature, looking for a few nuggets he can distort and package for debates. It's a not a particularly honest, or respected approach to cosmology.

      Hilbert's Grand Hotel suffers the usual problem of wanting a deterministic and progressive process (say measured against an entropy arrow of time) to apply at the quantum level and to states where regular GR physics breaks down. Craig wants the quantum world to conform to reality above this level. It doesn't. It has no particular credibility. That's why cosmologists are still happily generating bouncing, or reproducing models of the universe in complete disregard of Craig.

      P2 remains in an unproven state.