Suppose there are two persons A and B.

A attests to having witnessed some extraordinary event, e.g. A claims to have had an extraordinary religious experience with an other-worldly entity. Let's say for the sake of argument that A is sincere.

B, on the other hand, is skeptical of A's story. B thinks that A is probably exaggerating or misinterpreting their own experience, and that the better, simpler, more parsimonious explanation is that A just had a hallucination ("Listen A, there is no need to make our ontology more complicated by positing the existence of other-worldly entities when there is a simpler explanation: you just had a hallucination").

However, B's objection can be easily generalized to any subjective experience. Suppose C comes by and claims to have had a mundane experience with the physical world. One could easily apply B's objection to C's claim: "Listen C, there is no need to make our ontology more complicated by positing the existence of a physical world when there is a simpler explanation: the physical world is just a hallucination".

Is appealing to our subjective experience to justify our ontology irrational, since the hallucination hypothesis is always the best explanation?

Another question: Is there a symmetry breaker between A's claim and C's claim that would render C justified in trusting their subjective experience and A unjustified? In other words, are C's subjective experiences more reliable than A's subjective experiences? Is it possible to assess the reliability of subjective experiences?

  • 2
    i mentioned something like this to a shrink. they do, occasionally, ask if you think that your diagnosis explains your experiences (my reply) "a diagnosis of psychosis can explain anything". they seemed almost upset by this, but you always find room for delulu ha
    – user67675
    Commented Nov 18, 2023 at 16:57
  • 1
    The current day idealist Bernardo Karstrup says exactly this — materialism is a much much less parsimonious explanation for phenomena in general than idealism. Don't have a ref at hand — pls excuse. But you should find this on youtube and his books
    – Rushi
    Commented Nov 18, 2023 at 17:10
  • 2
    First, there is an asymmetry between extraordinary and mundane events. The latter agree with the ordinary course of events and do not require revision of established models of them, the former do. This puts extra justificatory pressure on them ("extraordinary claims require extraordinary evidence"). And second, surmises from perception are not necessarily subjective even when drawn, in part, from personal experience because their output can be cross-validated by others. Rationality of ontology heavily depends on what kinds of experience are used in its justification.
    – Conifold
    Commented Nov 18, 2023 at 20:53
  • 1
    The simplest explanation is that everyone hallucinates everything. Actually, solipsism is even simpler.
    – Scott Rowe
    Commented Nov 21, 2023 at 11:23

5 Answers 5


When is a hallucination the best explanation?

We have a fairly good idea of when and how experiences are unreliable. Just off the top of my head:

  • When you experience something while in bed in the middle of the night: Dreams would be the obvious example, and the line between dream and reality isn't always that definitive, and sleeping involves multiple brain processes working together, which don't always align perfectly, e.g. sleep walking or sleep paralysis, which frequently involves hallucinations.
  • When you have a relevant illness. Mental illnesses, absolutely, but illnesses generally could also apply. For example, fever can cause vivid "fever dreams" that someone could conceivably mistake with reality or something more than just a dream, especially if they don't commonly experience vivid dreams.
  • When you've suffered a brain injury.
  • When you generally have a loose grasp of reality. This one's a bit hard to directly argue: if someone sees things all the time that others don't consider to exist, they could merely say all of that actually exists in some other realm. Although one could argue whether they, or anyone else, are justified in believing that to be the case.
  • When you're on drugs (recreational or prescribed) with potential hallucinogenic properties.
  • When you see something in the corner of your eye, when you see some vague shape in the dark, when you hear some voice in your head, etc.

In addition, memory is notoriously unreliable (especially during particularly emotional experiences, which commonly applies to claimed spiritual experiences). People especially frequently remember exaggerated versions of what actually happened (also especially during particularly emotional experiences, e.g. remembering one's abuser as larger than life). Never mind when they just consciously exaggerate, to try to convince someone of their preferred conclusion, or to make it sound more significant than it actually was. It may also be difficult to give someone a good idea of what you experienced (even if we're talking about e.g. seeing physical objects, but especially if we're talking about emotions and internal senses, which are common parts of spiritual experiences, although we can try to make sense of someone's described experience by relating it to our own experiences and the described experiences of others). So if someone says "I saw a cloud that spoke to me", one might reasonably interpret that as them having seen some dark spot in their vision and having had some thoughts* (but of course it depends on the details they offer).

* As a side note, not everyone thinks in words, so if someone has an experience of thinking in words when they haven't experienced that before, they may attribute some extraordinary explanation to that. Although, as a former religious person, who does think in words, I know how easily people (my former self included) attribute their thoughts to the divine.

So any or all of the above could make it being a hallucination more likely.

If you have no independent evidence, what someone claims could also just be a lie. Although I wouldn't jump straight to that, as it's more productive and respectful to give people the benefit of the doubt and accept that they believe what they're saying.

When isn't a hallucination the best explanation?

To make the case for when a hallucination wouldn't be the best explanation, one could start by turning all of the above on its head: a non-drug-user sees something clearly in broad daylight, and otherwise has a firm grasp of reality and no relevant illness nor brain injury. That alone probably doesn't really get us to the finish line though.

Beyond the above, the best way to make the case for some other-worldly entity above a hallucination would be to extend beyond a singular experience:

  • Shared experiences (although shared delusions may also be a thing)
  • Repeated experiences
  • Divine healing: this has been reported plenty, but those reports suffer from a lack of well-documented medical verification of the ailment and the healing and/or the ailment could go away by itself. As atheists commonly point out, amputation is something we've never seen go away by itself, and we conveniently also have zero verifiable examples of the divine healing of an amputee.
  • Future predictions. One could say a lot about this. But just note that there are some criteria for what makes a good prediction, e.g. being specific about exactly what will happen and when, avoiding any sort of metaphors that could be predicted in multiple ways, and predicting rare events. Of course if someone says "God told me that will happen and it happened", that wouldn't be all that compelling, compared to someone telling you ahead of time what will happen.
  • Giving knowledge: If one gains some knowledge that one didn't know before (e.g. scientific discoveries), that may make a hallucination less plausible. Although it might be the case that you merely figured it out during a hallucination, or that you remembered something you've merely forgotten.

Other explanations?

For any competing hypothesis, one also has to consider how well that explains the experience, whether other things could explain it more simply, and how well it aligns with reality generally. Some would consider the problem of evil and divine hiddenness to altogether refute the existence of a deity with a specific combination of traits (all-loving and all-powerful). That pushes the bar much higher for accepting the existence of such a deity. Although one might still accept the existence of some powerful being (deity or not) without those traits.

  • Now I would like to know what you think about this question.
    – Mark
    Commented Nov 21, 2023 at 1:12

Not if there is verifiable evidence for another hypothesis

First we can have a debate of what measure we use for "best".

But assuming there is any such measure, then the answer is: no, a null hypothesis — i.e. the assumption that there is no correlation between the reported observation and reality — is not the "best" if there is independently verifiable evidence for another hypothesis.

And if the observation is independently repeatable then a halluciantion hypothesis simply does not make sense.

Granted, one can choose to mistrust evidence — always come up with objections to the evidence and stubbornly reject them — as there is no solution to the Hard Problem of Solipsism without starting to make some assumptions.

But one does this at the hazard of being branded obtuse and unreasonable by one's peers, cf. anti-vaccers, flat earthers, conspiracy fantasists, et cetera.

So, is the hallucination hypothesis always "best"?

It depends on your definition of "best". It is always sufficient to reject any reported observation, but — as evidence mounts for another hypothesis — you cling to the hallucination hypothesis at the hazard of wrecking your credibility.

  • What do you mean by verifiable evidence? Also, what do you mean by independently verifiable evidence?
    – Mark
    Commented Nov 18, 2023 at 18:36
  • @Mark "Independently" means I do not have to rely on the initial claimant to verify the evidence. Instead I can check up on the evidence on my own, thus avoiding the problem of having to trust the initial claimant. "Verifiable evidence" means just that: evidence that I can put under scrutiny and verify its veracity. Witness testimony for instance is not that.
    – MichaelK
    Commented Nov 18, 2023 at 18:55
  • There is apparently no solution for the Hard Problem of Headedness. But one person wrote a nice book called, "On Having No Head". Highly recommended for people with heads, hard or otherwise.
    – Scott Rowe
    Commented Nov 21, 2023 at 11:44

There is a very powerful symmetry breaker- reality is experienced in broadly the same way by billions of other people. If you have an everyday experience, the idea that it might be an hallucination is unnecessary to explain it. If you have an out-of-this-world experience, the possibility of it being an hallucination is a plausible explanation. The idea of the physical world being an hallucination in the ordinary sense of the word (eg the result of imagination) is ludicrous. How would you account for the consistency of your own experience of the world over time if it was an act of imagination? And how would you account for the fact that billions of other people are imagining the world in a synchronised way?

Of course, there is another sense of the word hallucination, which means that you are not experiencing what is really there, and in that sense of the word your life is one long hallucination, since colours, smells etc are responses of the human mind to stimuli that have no inherent colour or smell.

  • How would you account for the consistency of your own experience of the world over time if it was an act of imagination? And how would you account for the fact that billions of other people are imagining the world in a synchronised way? - See en.wikipedia.org/wiki/Brain_in_a_vat, en.wikipedia.org/wiki/Solipsism, en.wikipedia.org/wiki/Evil_demon
    – Mark
    Commented Nov 19, 2023 at 10:40
  • @Mark many thanks. I am familiar with the articles you have kindly referenced. I would apply my comments with equal force to solipsism etc. Commented Nov 19, 2023 at 12:20

The difference is simple. Many people in this world share the same subjective experience. In the case of A, many people do not share that person’s subjective experience.

If everyone claims to share the same experience, let’s compare two candidate theories: hallucination or a physical world. What explains the fact that there are multiple people claiming to share the same experience better?

Now, one can imagine an infinite number of hallucinations, only one of which would be a hallucination that just so happens to include others claiming to share the same experience as you. In other words, if all we knew is that a person had a hallucination, we would have no reason to expect this particular kind. On the other hand, given a physical world with multiple people, it is completely expected for everyone to share the same experience. In that sense, the latter explains it better.

Note, however, that this process is irredeemably subjective and dependent on your intuitions. It’s why an argument against solipsism is technically undefeated despite almost all of us finding the notion ridiculous.

Note that one can build into the hallucination hypothesis the expected data. One can say, “well, if I did have a hallucination where others claim to share a subjective experience, the data is what I would expect.” But this is completely ad hoc and atleast intuitively, should be given less credence.

  • You may find this follow-up question of interest.
    – Mark
    Commented Nov 20, 2023 at 23:37

In an interconnected world , where consciousness of individuals are connected, it is possible to verify the subjective experience. Science has some understanding of the concept of interconnectedness. For example - if two particles are entangled then if we measure the entangled property of one particle then it is possible to predict the opposite outcome of entangled property of another particle , even if they are separated by large distance. Until one of them is measured we can not say what will be outcome of the other. Science offers a very rudimentary definition of interconnectedness. Religion on the other hand says God is omniscient. God knows each individual, everything about individuals. Through the medium of God it is possible to verify the subjective claims. But the condition is that the follower of God must trust God completely. Trust is not required in physical entanglement process.

Everything is not an illusion. Suppose I thrash you , will you call it illusion or hallucination ? Suppose I ask you to drink poison ? Will you call it illusion or a hallucination?

On the other hand if we meditate properly we will find that all phenomena is lacking self. Self is like an illusion. Understanding of absolute self is like an illusion. We think self is real but if we look closely we find that self and related stories are like hallucination or illusion. There is a feeling but the feeling is not self. There is a body but the body is not self . There are perceptions but the perceptions are not self. There are choices to be made but the choices are not self. There is consciousness but the consciousness is not self.

Why is it so ? Because body is impermanent. Perception are impermanent. Feelings are impermanent. Choices are impermanent. Consciousness is impermanent.

  • 1
    I wish more people would find this convincing, but without experience of meditation it is not available or not persistent enough to affect people. It is not much better than explaining to someone how their pancreas works and saying, "Now you can control your blood sugar." I don't know a way to improve on this. The pancreas thing works when knowledge and experience come together over a long time. Same with knowledge of mind. One must choose to have the right kind of experiences, because they desire to. 99.9% of people do not. ("Of 1000 people, one seeks Me. Of 1000 who seek, one finds Me.")
    – Scott Rowe
    Commented Nov 21, 2023 at 11:35

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .