Is It Epistemically Certain that Whatever Begins to Exist is Caused?

I wrote that 

1) Whatever begins to exist is caused

is not epistemically certain. I don't deny that (1) is true; I deny that it can be known with certainty.  (As I explained earlier, truth and certainty are different properties.) And then I wrote that 

If an argument is presented for (1), then I will show that the premises of that argument are not, all of them, certain.

That is to say: if you try to show that (1) is certain by producing a valid deductive argument all of the premises of which are certain, an argument that transmits the certainty of its premises to its conclusion, then I will show that the premises of that argument are not, all of them, certain. I am using 'certain' as short for 'epistemically certain.' 

Lukas Novak responded:

Let us play that game. I believe I have an argument to prove (1) that can be reduced exclusively to obvious conceptual truths. Let's go step by step; you say which premise you doubt and I will produce an argument for it.

My kick-off:

(1.1) Whatever does not have a cause and yet exists, exists necessarily.
(1.2) Whatever begins to exist never exists necessarily.
Ergo etc.

Which one do you doubt?

I have no problem with (1.2).  I would say, however, that (1.1) is not certain.  The negation of (1.1) is: Something exists contingently without cause.  This is not a formally self-contradictory proposition. So we cannot rule it out on formal-logical grounds alone the way we can rule out Something exists that does not exist. It is therefore logically possible (narrowly logically possible) that (1.1) be false.

Is (1.1) a conceptual truth as Lukas appears to be maintaining?  Well, can we know it to be true by sheer analysis of the concept uncaused existent?  Not as far as I can see. Analyzing that concept, all I get is: existent that is not the effect of any cause or causes. That every EFFECT has a cause is a conceptual truth, but not that every EVENT has a cause, or that every EXISTENT has a cause.

If Lukas is right, then it is epistemically certain that the physical universe, which is modally contingent (i.e., not necessary and not impossible) cannot be a brute fact.  So if Lukas is right, then it is epistemically certain that the physical universe cannot exist both contingently and without a cause.

Here is where I disagree. I believe that the physical universe (together with finite minds) exists, exists contingently, and is caused. But I don't believe that we can know this to be the case with certainty.

It may be that Lukas is thinking along the lines of Reginald Garrigou-Lagrange.

Garrigou-Lagrange thinks that one violates the Law of Non-Contradiction if one says of a contingent thing that it is both contingent and uncaused.  He thinks this is equivalent to saying:

A thing may exist of itself and simultaneously not exist of itself. Existence of itself would belong to it, both necessarily and impossibly. Existence would be an inseparable predicate of a being which can be separated from existence. All this is absurd, unintelligible. (Reality: A Synthesis of Thomistic Thought, tr. Patrick Cummins, O. S. B., Ex Fontibus 2012, p. 65) 

Suppose that a contingent existent is one that is caused to exist by a self-existent existent.  If one then went on to say that such an existent is both contingent and uncaused, then one would embrace a logical contradiction.  But this presupposes that contingency implies causal dependency.

And therein lies the rub.  That the universe is contingent I grant.  But how does one get from modal contingency to the universe's causal dependence on a causa prima?  If one simply packs dependency into contingency then one begs the question.  What is contingent needn't be contingent upon anything. 

Related articles

Disagreement in Philosophy: Notes on Jiří Fuchs

On the Logical Independence of Person and Proposition

Comments

[Jim Soriano]

That “something exists contingently without cause” cannot be ruled out seems consistent with what little I know of Aquinas (ST, I, Q. 46). If God had wanted the universe to have always existed He could have willed it to be so. He didn’t.

Think of a matrix with four boxes. Across the top axis, label the two boxes “ex nihilo” and “ab aeternum.” Down the left side of the matrix label the two boxes “can be proved” and “can be disproved.” Then write “no” in all four boxes.

Thus the proposition that the universe was created from nothing can neither be proved (epistemically certain?) nor disproved, and the the proposition that the universe always existed can neither be proved nor disproved.

The “correct” answer, of course, is that the universe exists contingently and is caused, but natural reason cannot be certain of that. We know it true because God told Moses.

[BV]

Jim,

You appear to be conflating two different questions:

Q1: Did the universe always exist, or did it have a temporal beginning?

Q2: Could the universe exist without a cause or must it have a cause?

As I understand Aquinas, he holds that we can know, apart from divine revelation, by natural reason alone, that the universe must have a transcendent cause. We know this by the Five Ways. What we can't know by unaided natural reason is whether the universe always existed or had a beginning in time. We know that by divine revelation. Using natural reason alone we can answer (Q1) but not (Q2).

You may also be conflating the temporal with the modal. If something exists at every time, or always exists, it does not follow that it is necessary. It could still be contingent. And if contingent, then in need of a cause of its existence.

My question is this:

Q3: Is it certain that the contingent must have a cause, or not?

I say it is not certain; Novak says that it is.

[Lukáš Novák]

Bill,

allow me, for the moment, to pass over your argumentation and just continue the "game" (I believe that it will help us to clarify things more effectively). You've doubted premise (1.1), "Whatever does not have a cause and yet exists, exists necessarily." Let us see whether I can eliminate that doubt. I offer the following argument for (1.1):

(1.1.1) Every uncausable (or "incausable"?) existent is a necessary existent.
(1.1.2) Every uncaused existent is an uncausable existent.
Ergo, etc.

What do you think of these premises?

Two more points obiter dicendo (I don't want them to distract us at the moment):

(1) I do not say that it is epistemically certain that the physical universe, which is modally contingent cannot be a brute fact. I say that it can be made epistemically certain, to someone. This is a much weaker claim. I don't think that conceptual truths are automatically epistemically certain (to anyone or to everyone).

(2) Concerning conceptual analysis: if you mean just definitional analysis, I claim that is not sufficient to reveal all the implicit conceptual content. Analyse the concept of an Euclidean triangle and you will get something like "planar shape with three edges"; but you will never obtain properties like "having the sum of its angles equal to 180°"; "whose centroid, orthocenter and cicrcumcenter are located on a straight line" etc. Analyse the concept "red" as you will, but you never find the notion "not blue" as part of its definition. Etc. Ergo, there is (always!) more to concepts than has been defined into them (the medievals called that "virtual containment", Kant made up his synthetic a priori judgements to account for that phenomenon). Certain conceptual truths are not "analytic" in the sense "reducible to identities by definitional analysis"; but they are "analytic" in the sense "true in virtue of conceptual content", or "in virtue of the meaning", if "meaning" is taken sufficiently broadly.

[BV]

Lukas,

Thanks for the response.

I grant your first premise. If something has its necessity from itself, then it cannot have a cause. And conversely.

But I do not consider your second premise to be certainly true. Suppose the universe U exists, exists contingently, but has no cause. Your second premise implies that U cannot have a cause, i.e., is necessarily uncaused. Not clear. Why can't U be contingently uncaused? Why can't it be both contingently existent and contingently uncaused?

[Lukáš Novák]

Dear Bill,

you have doubted that (1.1.2) every uncaused existent is an uncausable existent. I argue for it thus (compressing several steps that seem trivial to me by assuming, hopefully unproblematically, that what is essential is necessary, that what is not accidental is essential, and that causality involves dependency):

(1.1.2.1) Causal independency cannot be an accidental property.
(1.1.2.2) Every uncaused existent is causally independent.
Ergo, every uncaused existent is essentially, and consequently necessarily, causally independent, i.e. uncausable.

And if you should doubt (1.1.2.1), then I say:

(1.1.2.1.1) An accidental property is possessed in dependency on something beyond its subject's essence.
(1.1.2.1.2) Causal independency cannot be possessed in dependency on something beyond its subject's essence.
Ergo, etc.

There is, I believe, another possible way to vindicate (1.1.2) that I will only sketch informally:

Note that the notion of causality has a modal dimension - it extends across the possible worlds, so to speak. For it involves dependecy (of the effect on a cause), which in turn involves impossibility for the effect to exist without a cause. Thus, if something is caused in the actual world, there is no possible world in which it exists without a cause. Thus, conversely, if something actually existing does exist without a cause in a possible world, it exists without a cause in the actual world; and since we are talking conceptual truths, all the propositions are necessary (i.e. the inference is valid for any possible world considered as actual), and so if anything exists as uncaused in a possible world, it exists as uncaused wherever it exists.

[Owen Anderson]

I'm comparing this to what Aristotle said about arguing for the law of non-contradiction.
Owen Anderson

[Lukáš Novák]

Owen,

just curious: what do you think Aristotle said? (Interpretations vary...)

[BV]

Lukas assumes that >>what is essential is necessary, that what is not accidental is essential, and that causality involves dependency<<

Not clear!

1. If a thing has a property essentially, it does not follow that it has it necessarily. To say that Socrates is essentially F is to say that Socrates is F in every metaphysically possible world in which he exists. But he is a contingent being: he exists in some but not all possible worlds. Therefore, Socrates is human essentially but not necessarily. By contrast, God, a necessary being is omnipotent both essentially and necessarily.

2. It is impossible that anything be both round and not round. Nothing has this property essentially and nothing has it accidentally. Therefore, what is not essential need not be accidental.

3. Is it entirely clear that the effects of secondary causes depend for their existence on their causes? No. The primary cause, God, produces his effects, bring them into existence. Not clear in the case of secondary causes. Occasionalism remains a livee option for theists. See the work of Hugh McCann.

[BV]

Lukas argues:

(1.1.2.1) Causal independency cannot be an accidental property.
(1.1.2.2) Every uncaused existent is causally independent.
Ergo, every uncaused existent is essentially, and consequently necessarily, causally independent, i.e. uncausable.

This is a non sequitur given what I said in the immediately preceding comment.

And while I accept the second premise, I reject the first. I mean of course that I reject that it is certainly knowable, knowable with certainty.

Lukas gives the following argument in support of the first premise:

(1.1.2.1.1) An accidental property is possessed in dependency on something beyond its subject's essence.
(1.1.2.1.2) Causal independency cannot be possessed in dependency on something beyond its subject's essence.
Ergo, etc.

The first premise may well be true for non-existential properties. But it is not clear why the universe cannot just exist and exist accidentally without being dependent on anything beyond its essence.

[BV]

Owen,

What Lukas and I are discussing is whether there are any substantive philosophical theses that are knowable with epistemic (not psychological) certainty. Examples include *God exists* and *There are substances.* L. says that these two and others can be known with certainty. I say they can't.

As for LNC, is it certain that it is a law of Being and not just a law of thought? I say No. L. says Yes.

[Lukáš Novák]

Dear Bill,

I see I should have resisted the temptation to abbreviate :-)

Ad 1): By "necessary property" in "what is essential is necessary" I meant "in every possible world where the subject exists". This is sufficient for my argument, as I hope is clear. (The true necessity of existence of the uncaused is safeguarded by premise (1.1.1) which has already been conceded; in this branch of the argument we are merely exploring the connection between uncausedness in the actual world, and uncausedness in whichever world the rewspective thing exists).

Ad 2): I would say that "not to be (possibly) both round and not round" is essential in the broad sense employed here: it flows from the essence of everything, because it flows from the ratio entis which is included in the essence of everything. At any rate, this is not a counterexample to the argument: you would need a non-accidental, non-essential contingent property.

Ad 3): I define causality as bringing into existence, or "giving being". If secondary "causes" do not impart being to their "effects", they are not their true causes. At any rate, this factual question has no bearing on the conceptual relations we are discussing - if you don't like my definition of cause, replace "cause" with "giver of being" everywhere in my argument.

So I believe the argument is not harmed by these three objections at all.

[Lukáš Novák]

Bill,

concerning your criticism of (1.1.2.1.1): please note the property in question is not existence but causal independency. Existence would be trivially "essential" in my usage, since it (trivially) accompanies a thing wherever it exists (by saying that, I am not taking any position in the "real distinction between essence and existence" quarrel - if you considered the real distinction certain (which I don't), my task would be much easier).

Does causal independency count as an "existential property" for you, I wonder? If you say yes, does it mean that you concede (hurray, hurray!) that causal independency implies existence (in the non-trivial way, meaning it is impossible for a causally independent thing not to exist)?

We are exploring here, whether a thing is causally independent due to its essence or due to something else. The difference between causal dependency and causal independency is a real one, and so causal independency must make some ontological difference, whose root is located either in the essence of the thing or elsewhere.

Now I say that it must be located in the essence, and thus causal independency must accompany a thing in every world where it happens to exist. Because if it were located elsewhere, then we would have the impossible situation that a thing was rendered independent through dependency on that "something else" (no matter what).

You can look at it also this way: Is it possible that a thing were causally independent in one world where it existed and causally dependent in another world where it existed? (We are completely disregarding the worlds where it does not exist - that's why objection 1 is irrelevant.)

I say no: causal (in)dependency status cannot vary across possible worlds, since essence being the same across the worlds, such a variation would have to be due to various accidental forms, and that is absurd.

But to argue formally for (1.1.2.1.1):

(1.1.2.1.1.1) Every property (possibly except existence) is possessed either due to the essence of the respective thing, or due to something else.
(1.1.2.1.1.2) An accidental property is not possessed due to the essence of the respective thing.
Ergo, an accidental property (which is not existence) is possessed due to something else than the essence of the respective thing.

This is a sufficient formulation of (1.1.2.1.1).

[BV]

Lukas,

I am losing the thread of this discussion. Your claim, I take it, is as follows:

One can know, a priori, and with epistemic certainty, that necessarily, every modally contingent existent has a cause of its existence.

Is that your thesis? If so, can you present a valid deductive argument for it all of the premises of which are epistemically certain?

[saracen]

here's the argument that establishes the thesis:

The contingent does not have necessity from itself
What does not have necessity from itself is not uncaused
Therefore, the contingent is not uncaused

[saracen]

apologies, the major should say:

what doesn't have necessity from itself is caused

and the conclusion altered accordingly.