When it comes to the cosmological argument, proponents point out that a first cause is needed to stop an infinite regress of causes.

In epistemology, foundationalism is used to break the infinite regress of justification. If every belief needs another to be justified, the process would go on forever, and hence some beliefs don’t need to be justified by others “stopping” the regress.

My question is about how these things are supposed to “stop” the regress and what it actually means. For example, how does a first cause “stop” the regress? If this first cause was always there, isn’t this still creating a regress since there is no “first” moment in time where it existed?

So in essence, what does it mean to “stop” this chain and how does one know that these propositions actually stop them?

  • There is a sort of epistemic "optical" illusion, here. Say we have reached a cause that we don't want to go further back from. If we are yet tempted to explain why this cause is the first, rather than some other, we will start regressing at a diagonal to principles of comparison and explanation. Eventually we'll reach another off-kilter edge of ideas, and we'll spiral about; either some angle of the spiral loops back on another, or the regress continues apace; or else we just stop moving, exhausted, wherever we find ourselves standing dizzily for all our spinning. Sep 13, 2023 at 20:06
  • In the cosmological argument, causation is not restricted to causation in time. In its sense of "causing", the parallel postulate "causes" (is the reason for) the sum of the angles of a triangle to be two right angles, and time is not involved there at all. “Stopping the regress” is just an idiom too. It simply means that a causal chain must end somewhere (if one subscribes to foundationalism), and where it ends is the first cause that "stops" or "breaks" the regress.
    – Conifold
    Sep 13, 2023 at 22:44
  • 1
    @Conifold How does one know that “timeless” causes are possible? I can see justification not depending on time of course since it’s epistemic but don’t seem to understand how one can just assert the concept of a timeless cause without any evidence for it. If one can assert things with no evidence, why not just assert the universe started out of nothing?
    – user62907
    Sep 14, 2023 at 0:59
  • 1
    To pre-understand what it even means to break infinite regress for this contingent Saha world if it's possible at all perhaps you should contemplate why propositions in classic logic or its weaker forms such as intuitionistic logic are all timeless, and if you add time for your rumination like in physics, it seems time is constantly flowing in sync with the clock on the wall such that we have idioms like 'time flies so fast', but how fast it really flows? Can you really talk about space emerges inside another space or emergent (illusory) time inside another time dimension ad infinitum?... Sep 14, 2023 at 5:40
  • 1
    @Conifold I'd separate physical(ish) causes from logical conclusions and structural implications and such. The latter may potentially be a matter of knowledge more than anything else, and doesn't seem capable of giving rise to physical effects. If you have a triangle, you can conclude that the angles sum to 180 degrees, but this doesn't cause that sum or that equality to be created as a physical object (except maybe in some view of Platonism, and existing as a thought). On the other hand, if I were to drop something, there's a physical change in terms of that object falling to the ground.
    – NotThatGuy
    Sep 14, 2023 at 10:10

2 Answers 2


The claim of a first cause wouldn't be that it was always there, it would be that at some point there was nothing, and then the first cause made it so that there suddenly started being something. Or perhaps the universe was in some perfectly static state, and then the first cause caused stuff to start happening (although then one would also need to posit that the universe was in that initial state).

But if there is nothing (or if things were perfectly static), there would be nothing to be a first cause, so that seems to be a not-insignificant problem.

This leads to some odd ideas, like something causing itself, or something causing an event in the past (that ultimately causes itself). I've heard hypotheses of such things in quantum mechanics, although I'm not too well-versed in that and I don't know how popular such hypotheses are.

Also, with the spacetime physics model, it's at least theoretically conceivable (if not most likely) that time and how we think of causation (something always having a cause, always being caused by things that happened before it, and only being able to cause things after it) breaks down at the earliest points of the universe. At that point, it may be rather difficult (if not nonsensical) to try to discuss what was "first".

The much more common way that a first cause is posited is by calling it "timeless".

In theory, this sidesteps the problem by just asserting that ideas such as "before" and "always" doesn't apply to the cause.

In my view, this assertion is unjustified and contrived in apologetics (it's asserted as a property of God, so God can serve as the only solution to a presented problem, rather than being a conclusion reached based on any sort of evidence or rational justification). It also introduces a far bigger problem of what it even means to be timeless, how something could exist timelessly, and how a timeless entity could have the properties attributed to it (e.g. consciousness).

Yes, I did just allude to how we think of causation breaking down in the section above, which could conceivably allow for the possibility of timelessness. But the breakdown of causation is a counter-intuitive conclusion (not assumption) that would apply to elementary physics particles. If you want the first cause to have consciousness, this introduces additional problem of how it's possible to think without time (given that we tend to consciously think fairly sequentially) and space (given that the only consciousnesses we know of are tied to physical bodies*). Also, if timelessness, and causation from it, is possible in physics, God would not be necessary to terminate the regress, which rather undermines e.g. people trying to prove God's necessity using the cosmological argument.

* Idealism posits non-physical consciousness, and that reality is a result of consciousness. This renders the entire discussion of causes within reality somewhat moot. It might solve the first cause of observed reality (that cause then trivially being consciousness), but it mostly just pushes back the problem, and creates a few more, because then the question would become what caused that consciousness, or what was the "first cause" within that consciousness, and where and how does that consciousness exist, and how is that consciousness linked to observed reality, and what explanatory power or justification does the claim of such a consciousness have.


how does a first cause “stop” the regress? If this first cause was always there, isn’t this still creating a regress since there is no “first” moment in time where it existed?

I think of it as short circuiting the claim of an infinite regress. So, you still go backwards, but you do not go back forever.

there is no “first” moment in time where it existed

A first cause exists eternally, so you cannot go before him, even conceptually. When you get to god, you "stop", and are no longer thinking of a regress any more, because has doesn't have a cause and/or because you cannot think of anything before him,

You have to take add up all terms CAUSAL INFINITE REGRESS, not just 'causal', 'infinite' and 'regress'.

  • 1
    The point is that if god is eternal and he exists in time, technically there is an actual infinite regress happening: each moment in time has another moment that precedes it…infinitely. So if an infinite regress exists here, why can’t a regress exist with causality?
    – user62907
    Sep 14, 2023 at 13:21
  • @thinkingman i guess that does kinda make sense... but have you "stopped" the infinite series of causes?
    – user67675
    Sep 14, 2023 at 13:50

You must log in to answer this question.