Is a literal violation of a current law of nature the only way we can point to a potential divine explanation? This is not to say that a violation of a current law of nature necessarily implies a divine explanation. There could be a natural explanation that we are not aware of yet. But it is easy to see how it may atleast introduce the possibility of a divine explanation, or may justifiably increase our credence in god after observing the miracle.

Without this violation though, what possible evidence could we possibly have for a divine explanation that isn’t a miracle? What remains is what is possible under nature. Even if this is extremely improbable, it would only be evidence for God if God already exists and created that extremely improbable thing. But God’s existence is the very thing we have to prove.

For example, suppose I had a dream in my sleep that detailed events of the next day and God is displayed in that dream and He says “Here are 10,000 events that will happen tomorrow.” Suppose they all happen. Despite intuition, even though this seems very improbable, this still doesn’t seem like evidence for God. This is because these events are still possible under naturalism and a very improbable event doesn’t increase the probability of God doing this unless you already assume God exists. But that is the very thing we are trying to prove.

One may argue, using Bayesian terms, that the dreams happening may atleast increase your confidence in god. But I fail to see how this can be the case, since you would first have to find some independent reason to justify god having a non zero prior probability. But we have no prior evidence of god’s existence, much less operating in the world.

Does this mean we can de facto dismiss or not even consider any possible evidence for God unless it is a violation of a law of nature?

  • 3
    I am puzzled how we would ever prove that an event was a violation of a law of nature (as opposed to an apparent violation of something that is thought to be a law of nature at the time). There is always the possibility of a natural explanation that we are not aware of yet, and, unless you already believe in supernatural intervention, that will always be the more probable explanation.
    – Ludwig V
    Commented Jul 9, 2023 at 10:09
  • The point is that a violation of a law of nature can arguably atleast introduce the possibility of a supernatural cause. Or maybe, increase our confidence in a supernatural cause even if that confidence ends up being lower than the natural one. The question is whether an event that does NOT violate a law of nature can ever justifiably increase your confidence in god.
    – user62907
    Commented Jul 9, 2023 at 10:37
  • 1
    Not necessarily. God not only manifests by means of miracles. Perhaps you discover it on a mathematical formula which allows you to create planets, life or spacetime.
    – RodolfoAP
    Commented Jul 9, 2023 at 11:19
  • That wouldn’t be the traditional conception of a god: all knowing, all powerful, and most importantly an actual being. Instructions/formulas aren’t conscious beings
    – user62907
    Commented Jul 9, 2023 at 11:47
  • Yes. I'm apologize for misinterpreting your question.
    – Ludwig V
    Commented Jul 9, 2023 at 13:46

6 Answers 6


This question has several major thinking flaws embedded in it.

The first is in the title, where you use "supernatural" as an ontological category. (others include the mistaken idea there are "Laws of nature" rather than hypotheses and theories, plus regularities, and the presumption that physicalism is the default view of the world and should be presumed until overturned). Back to Supernatural. What is "supernatural" ontologically? What is "natural" ontologically? These are terms with no useful meaning.

If you try to define natural ontologically in terms of physicalism -- you quickly run into Hempel's dilemma -- nothing can be excluded from future advances in physics, including Gods. You also run into the problem of philosophy which destroyed Logical Positivism -- physics needs philosophy to precede it, to establish the scientific method, and the metrics/criteria it uses -- and none of THAT is physics but must exist for physics to exist.

The only useful meaning for "supernatural" in philosophy is its epistemological usage -- the category of things for which one cannot use methodological naturalism to investigate. Some theists assert that God is in such a category, hence if one uses this epistemological definition, AND assert God to be epistemologically supernatural, THEN by simple derivation, one cannot provide "evidence" for God.

However, as Sam Harris argued extensively in The End of Faith -- in practice basically no theists treat God as epistemologically supernatural. Theists hold by their religious views for reasons, and these are most clearly on display in the debates BETWEEN different factions of theists.

Outside theists, the most common claim of epistemelogic supernaturalism I have seen is the "New Mysterians" among philosophers of mind. They hold that consciousness is intrinsically beyond our ability to understand, and it will always remain "mysterious". Note these are mostly physicalists, or physicalist leaning philosophers, who are explicitly arguing for part of our world being supernatural.

If you want to understand the reasons for theistic faiths, then you would be best off actually reading the evidence cases that theists put together, rather than asking other atheists to regurgitate straw man arguments against theism. Have you actually read any such arguments?

Many of the evidence cases put together by theists -- tend to establish the plausibility of their worldview, rather than its definitive truth. Among the best authors I have encountered in this set of writers are JP Moreland, Richard Swinburne, John Pokinghorne, and Paul Davies. The first three are Christians, Davies is a deist.

You also misunderstand the nature of evidence in methodological naturalism. NO evidences can EVER be definitive. Quine pointed out that theory is ALWAYS underdetermined by evidence! Therefore any hypothesized worldview or assumptions set is infinitely klugable to account for even the most embarrassing observations. There is also no "default view" that a competing one has to overcome before it can be accepted -- all worldviews or claims must make their cases against all evidence.

Because of the infinite klugeability of hypotheses/worldviews, the metric for goodness of aq theory or claim in the face of challenging evidence is not "can this evidence possibly be interpreted differently" but "how useful is the assumption set/worldview after it has been kluged to fit this evidence". The best theory I have found in Philosophy of Science for how this is done is Lakatos' Research Programme methodology. https://bertie.ccsu.edu/naturesci/PhilSci/Lakatos.html Another very useful framing is to describe the accumulation of a usefulness case for a worldview or hypothesis to be that of consilience https://science.jrank.org/pages/7591/Consilience.html

What one discovers, when one pokes, prods, and challenges a philosophic worldviews aggressively enough, it that every one of them provides incomplete explanations and justifications, and has incoherencies embedded in them. This is not sufficient to dissuade philosophers from holding by those worldviews. For instance the problems I noted for physicalism have not prevented physicalism from being the predominant worldview of Western philosophers for the last century. This is because philosophical worldviews ARE Lakatosisn Research Programmes -- and they accept the incompleteness of their current solution.

For me, a theist of a sort, my evidence base for accepting theism has been

  1. the superiority of spiritual dualism as an explanation for consciousness over all physicalist, idealist, and neutral monist models.

  2. The large supporting evidence set of "supernatural" experiences (I am using the term colloquially, not either ontologically or epistemologically) of direct apprehension of ghosts, synchronicity, and warnings.

  3. the supporting evidence set of successful parapsychology experiments, which presume either dualism or idealism. https://parapsych.org/articles/36/55/what_is_the_stateoftheevidence.aspx

  4. the extensive history of mystic experiences, where humans communicate with and interact with a spiritual universe.

  5. My own success in doing so, following a process developed by a Gnostic: http://www.bswett.com/1990-03TwoWayPrayer.html

  6. the direct experiences of a God in two way prayer.

This is a set of evidence compiled as part of methodological naturalism. It is a highly useful and effective worldview in navigating the world, and holds up far better to philosophic challenges than any other I know of. This is a consilience and progressive research programme case for a form of theism (di-theism).


Essentially, what you are asking to be presented is natural evidence of a non-natural thing. This is an impossibility.

Even if a supernatural thing existed, there could, by definition, never be any natural evidence that it existed. This is because natural evidence is only concerned with what is natural, and seeks to explain everything as either a result of known natural processes or unknown natural processes, and excludes all other possibilities, irrespective of their probability.

At best, you could potentially show that something non-natural must exist. For instance, if all natural causes have a cause, and if there is not a circular set of causes, then there must exist some cause which is not natural. That itself isn't really a natural evidence, but it is an evidence from naturalism.

It's the same way that math cannot define how beautiful a picture is. It can give you numbers about how symmetrical it is, how well the colors contrast, and so forth, but beauty, in the traditional sense, is not a mathematical concept. Whether God created and caused natural objects and natural laws or not is not a question that can be answered with appeal to natural evidence.

Miracles don't necessarily have to be supernatural. While most people expect that miracles defy current understanding of natural law to some extent, depending upon the definition used, it may just be an occurrence which is highly improbably according to currently understood natural principals. Whether rational or not, most people accept the power of inductive reasoning, and when the probability of one set of explanations becomes sufficiently low, people would naturally begin to accept other explanations which they might have otherwise considered improbable.

The evidence for the existence of God is not often concerned with what is or is not natural. Many people who believe God exists also do not hold to the belief that all important or relevant things are purely natural or obey currently understood natural laws. Instead, they might attempt to disprove Naturalism itself. Examples of miracles is potentially one way to attempt that, by showing that in some scenarios the likelihood that Naturalism is the best explanation is low enough to be counterintuitive. However, more formal arguments against Naturalism do not include the discussion of Miracles.


Is a miracle the only possible evidence for the supernatural?

Since you haven't provided definitions for "miracle" and "supernatural", I will provide the online dictionary definitions

  • miracle :A surprising and welcome event that is not explicable by natural or scientific laws and is therefore considered to be the work of a divine agency.
  • supernatural: (of a manifestation or event) attributed to some force beyond scientific understanding or the laws of nature. "a supernatural being"

The only difference I see is that a miracle is God's work and the supernatural isn't.

Based on these definitions, the answer is no.


That's one viewpoint on miracles, and a common and widely accepted one. It's arguably traceable back to Hume's On Miracles ("A miracle is a violation of the laws of nature; and as a firm and unalterable experience has established these laws, the proof against a miracle, from the very nature of the fact, is as entire as any argument from experience can possibly be imagined.")

That's not the only way of looking at miracles, however, and it's arguably NOT how most theists conceive of miracles. What makes something a miracle from a theistic point of view is not whether it has a naturalistic explanation or not, but whether we can perceive God through it. A (real or seeming) suspension of the natural order is one place someone might perceive a miracle, but not the only one. For instance, Augustine experienced it as a "miracle" when the child next door said "Take up and read" at the exact moment where he was looking for spiritual guidance. It wasn't supernatural, but he perceived the hand of God at work.

CS Lewis dramatized this view of miracles in his book The Silver Chair, which is part of his Narnia series of Christian allegories. There is an incident where the protagonists are in need of divine guidance. They open their window, and see giant letters that say "UNDER ME," which they take as a miracle. In the morning, however, they learn that the letters are just the last remnants of an ancient monument--the naturalistic explanation. So which explanation is correct? Lewis' point is that both are true, within the framework of the story. The letters are simultaneously an ancient inscription, unrelated to anything current, and a timely divine message to the protagonists. Of course, it only works this way because Lewis set it up this way. But if we view Lewis as standing in the position of creator of this particular fictional world, then it still serves his point.


You ask "Without this violation (sc. of natural law) though, what possible evidence could we possibly have for a divine explanation that isn’t a miracle?"

There is/was a doctrine that took natural laws themselves as evidence for the existence of God. It is called "Occasionalism". See SEP - Occasionalism. The high point of this doctrine was in the 17th century. I'm uncertain how much philosophical support there is for this doctrine nowadays.

The two best-known names for this are Nicolas Malebranche and George Berkeley. A fundamental idea for both is their understanding that God is the only possible real cause of anything that happens or exists. The key text for Malebranche is his De la recherche de la vérité. (Of the Search after Truth). Berkeley's "proof" of the existence of God is the high point of his Treatise Concerning the Principles of Human Knowledge; there's another version in Three Dialogues between Hylas and Philonous.


The only possible evidence for "divine explanation" is the divinity itself.

As long as you have not established the existence of a divinity, you cannot use it as an explanation.

  • What if one argues (and I’m merely playing devil’s advocate here) that my dream example increases the probability that god exists. For otherwise, it is extremely improbable that it happened through blind, natural means.
    – user62907
    Commented Jul 9, 2023 at 13:12
  • @thinkingman You cannot talk about probability of an event that has already/never happened. Commented Jul 9, 2023 at 15:08
  • 1
    I have downvoted this answer because I think its claim that evidence is ever identical to the thing it is evidence for is clearly false.
    – g s
    Commented Jul 9, 2023 at 15:22
  • I think what he means is direct observation @gs
    – user62907
    Commented Jul 9, 2023 at 16:15
  • @gs Divinity is not evidence of itself, naturally. The existence of a divinity is evidence supporting the design theory. Commented Jul 9, 2023 at 16:59

You must log in to answer this question.