This is a question that has been bugging me that came up on a debate about the Kalam cosmological argument between Graham Oppy and Andrew Loke, both philosophers as far as I am aware: https://www.youtube.com/live/a8NrTv-Durc?si=22cjbeVgIpmP1gNz

Oppy argues that everything that occurs has an explanation but not necessarily a cause. He says there is an initial thing and an event that exists which is the universe (in this case “universe” means everything that exists) at t = 0. He also says that this initial thing is necessary and thus does not require a cause. He says it does still require an explanation but in this case, the explanation for the universe’s existence is that it is necessary. The universe is said to begin at t = 0 in his view.

Loke also argues for a first cause but in this case, the “first cause” is god and he somehow exists timelessly and “without” a beginning, unlike the universe. He argues that for things that do not have a beginning, there is no cause needed, and also no explanation needed.

When, then, does something require an explanation? If something does not have a beginning, does it not require an explanation? What would an explanation for something without a beginning look like? Can a timeless (or eternal, but within time) god more easily or plausibly be asserted to exist without explanation compared to a universe with a beginning?

Also, what is the difference between an explanation and a cause here? Can something truly have an explanation for its existence but not have a cause?

I understand that these are a lot of questions but since they’re all related to the ultimate question of what needs explanation, I thought I might as well have listed all of the chief concerns that were bugging me. This allows me to portray the concerns without the reader having to watch the entire video, although feel free to do so if you wish, since watching the video can help.

3 Answers 3


You ask:

what is the difference between an explanation and a cause here? Can something truly have an explanation for its existence but not have a cause?

A cause is imputed to be real, and an explanation is imputed to be a linguistic description of the real. That means, historically and intuitionally, causes are external to people and explanations are internal. This is in line with beliefs involved with naive realism. From WP:

In philosophy of perception and epistemology, naïve realism (also known as direct realism, perceptual realism, or common sense realism) is the idea that the senses provide us with direct awareness of objects as they really are.1

Let's take a look the example in the WP article, a person viewing a candle's wick burning.

EXPLANATION INVOKING CAUSES IN ANCIENT GREECE: For physicalists, the idea that there is a cause to the burning of a candle goes back to Aristotle's Four Causes. The final cause of the candle is to provide light, as the form the candle takes is the result of the labor of a human which is the efficient cause. The shape the candle makes is the formal cause, and the material cause of the candle is the choice of wax and wick.

EXPLANATION INVOKING CAUSES IN THE MODERN WORLD: But none of those causes explains why the candle burns in the modern world. Of course, since Priestly, oxygen takes a key role in explaining why the flame burns, and today, an explanation of the flame revolves around oxidation-reduction. Chemistry and the atom lie at the base of our understanding.

Intuitively, causes are thought to be real, physical, and external to the mind, but the explanation, which has a physical topic, can be viewed as words describing the world. Thus causes are taken to be physical activity (traditionally) and explanation is taken to be mental activity. Today in psychology, we can suggest that explanations help to satisfy curiosity and reduce cognitive dissonance and thus are psychological rather than philosophical in nature.

If one argues, however, that causes themselves are not real, but are uses of normative uses of language just like descriptions, then it is possible to have explanations and accept there are no "real causes" that do not at least weakly reduce to language normativity. After all, one cannot weigh, touch, see, or measure a cause. Some believe that at best one can measure correlations. See Judea Pearl's Causality (GB) work for a list of techniques of establishing causality through mathematical and logical reasons. This latter view moves towards scientific instrumentalism.

  • I don't see how causes don't explain why the candle burns when the explanations you give are causes in themselves. Oxidation-reduction describes the efficient cause of burning, and further chemistry explains the material causes of the candle's flammability. It seems you mistook the difference between cause and explanation as being simply one of resolution.
    – Mutoh
    Nov 6, 2023 at 15:02
  • @Mutoh I reread and see why you might think that I might think that explanations aren't narratives laying out causality. The effort was to try to show that causes change when explanations change. I'll fix.
    – J D
    Nov 6, 2023 at 15:11
  • 1
    I've updated both sections to reflect: EXPLANATION INVOKING CAUSES IN <TIME PERIOD> to show how causes are normative in explanation. Thanks for you feedback.
    – J D
    Nov 6, 2023 at 15:12
  • 1
    That's definitely clearer. Though I still would say the difference between "modern" causal explanations and Aristotelian ones is one of resolution. That a candle is made out of wax and wick does explain why it burns - were it made of ice, it wouldn't burn as well if at all. Furthermore, wax is not merely a decoration, it delays the wick burning, and so on. Modern chemistry, while still dealing with material causes, simply provides further explanations why, say, wax and wick are flammable but ice is not, or why wax doesn't burn as quickly as bare wick, etc.
    – Mutoh
    Nov 6, 2023 at 15:52
  • @Mutoh Agreeeeeed 100%. Cummins has an explanation of scientific explanatory models to show that ultimately resolution is potential and not actual. What's important for the OP to understand is that all causes are tentative epistemic constructions that have a normative basis. If causality is seen as a potential much in the spirit infinity is seen as potential, then causes aren't a real thing, and don't exist, if existence is taken to be having a specific physical essence. Causes are simple conditional logic embedded into narrative about the physical domain and change as the narrator sees fit.
    – J D
    Nov 6, 2023 at 16:58

I think it is difficult to attribute any real meaning to the kinds of ideas being discussed in that debate. Both positions seem to be synonymous with 'we don't know what caused the Universe'. If someone claimed that the Universe was created by the Giant Bong Rabbit, which was an indefinable entity existing outside of space and time, I would struggle to see any real reason why that theory shouldn't be treated in the same way as any other related idea about some omnipotent entity. They are just empty words. They have no predictive power and offer no means of falsification, so they are essentially arbitrary. You can be inclined to believe one type of story or the other, and there's no hard reason to choose between them, which probably explains why the debates have continued endlessly for more than two millennia already.

As to the difference between an explanation and a cause, usually you have to know, or imagine, a cause to give an explanation, the latter being a story about the former, unless your explanation is that there is no cause, which seems to me to be no explanation at all.


Answering the question directly, what determines the need for an explanation is the Principle of Sufficient Reason, under which everything has a reason or, synonymously here, an explanation. Everything is either self-explanatory (in which case it's necessary) or its sufficient reason is ultimately external to itself (in which case it's contingent).

As for differences, an explanation is what makes something intelligible; a cause is what imparts being to something. Causes have to be different from their effects, but explanations don't have to be different from what they explain. Not everything has a cause, but everything has an explanation. Everything which has a cause can only be sufficiently explained by something distinct from it, but whatever does not have a cause is sufficiently explained by itself. And all causes are explanations insofar as they make their effects intelligible, but not all explanations are causes.

Loke is right to say that a first cause would have no cause (otherwise, it wouldn't be the first cause), but he's wrong to say that the first cause has no explanation (unless he means specifically a contigent explanation). The first cause's sufficient explanation is itself. (But being intelligible in itself does not mean it's thoroughly intelligible to us. Cf. this article.)

Both God and the universe require explanations. Even an universe without a beginning would need an explanation. The real question is whether lacking a beginning automatically makes something self-explanatory. Aquinas, who famously believed that the universe can't be demonstrated to have had a beginning through natural reason alone, would disagree that it automatically does. Because regardless if the universe had a beginning or not, either way it has enough characteristics that make it something caused and therefore in need of an explanation that goes beyond itself. Namely, the fact that it undergoes change would suffice to show that it can't be self-explanatory, and demands an unchanged changer as an explanation. But this is the beginning of another question altogether.

  • How can an eternal thing be caused?
    – user62907
    Nov 6, 2023 at 21:32
  • @thinkingman an eternal thing can't be caused. If not having any beginning altogether means something is eternal, which I'll grant, then a qualification is in order: even if the world didn't have a beginning of time, it would still have one of creation (which can be shown independently), so it can't be eternal.
    – Mutoh
    Nov 7, 2023 at 14:45

You must log in to answer this question.