If we can't be 100% sure of anything, then we should trust everything with mistrust and suspicion?

For example, the existence of other minds, the existence of the outside world, etc. is currently impossible to prove with absolute, 100% certainty.

So we should believe in the existence of other minds with a grain of disbelief?

That is, we should constantly doubt the existence of other minds?

How will it differ from agnosticism and pragmatism?

  • 5
    fwiw, which is nothing, i would urge caution, humility, etc. over disbelief.
    – user67155
    Aug 12 at 1:50
  • user66697, good point. Aug 12 at 2:15
  • 2
    How do you get form "believe with a grain of disbelief" to "constantly doubt"? We believe something when we are prepared to act on it, having a grain of salt there is fine but it makes little difference. The grain of doubt is not constant, it is set aside and forgotten about, except for rare reflective moments. It is pragmatic, but the sort of pragmatic that almost everyone subscribes to - humans are fallible, as they say. And agnosticism professes a lot more than a grain of salt.
    – Conifold
    Aug 12 at 2:54
  • if we rely on common sense then we can believe in the existence of other minds with absolute certainty, but if we use IBE or other arguments in favor of the existence of other minds then we have to believe in the existence of other minds with a lot of doubt or skepticism?
    – Arnold
    Aug 12 at 5:52
  • 1
    If other minds isn't a thing that you could ever resolve a bet about by recourse to measurement, I don't think it can have a probability in the first place. And even if it could have a probability, but nothing you can do can ever resolve the bet anyway, why care?
    – g s
    Aug 12 at 10:32

11 Answers 11


“Being 100% certain” of something implies that that mental state is meaningful in the first place. What does it mean to be 100% certain of something? Is it just a thought? A feeling? Either way, why would a conscious sensation occurring in your mind or feeling affect the world out there?

People say that you can’t prove for certain that the world is real. If one is not 100% certain that the world is real, then this means that there is a non zero % chance that the world is not real.

But how does one prove there’s a non zero % chance? In other contexts, there is proof. Landing heads 737393 times in a row has an absurdly low chance. But it’s the kind of non zero chance that one can prove and point to knowing that that sequence of coin tosses doesn’t violate any natural laws. But the world not existing isn’t provable in this sense. By definition, it being a simulation would violate natural laws and is in every sense impossible given what we know. So in what sense can we even claim there’s a non zero percent chance?

The gist of my point is that attaching credences to these kinds of beliefs is meaningless. Probability just doesn’t make sense here. The universe and the world isn’t a game of chance. As such, the entire premise of your question is lacking.


You might think you can be 100% sure of your existence following the line of Descartes' cogito ergo sum, "I think therefore I am". However in Being & Time (116) Heidegger dismantles the I by observing that any functional mind is a social development. At least we can be 100% sure that there is thought: that thought exists.

The word 'I' is to be understood only in the sense of a non-committal formal indicator, indicating something which may perhaps reveal itself as its 'opposite' in some parti­cular phenomenal context of Being. In that case, the 'not-I' is by no means tantamount to an entity which essentially lacks 'I-hood' ["Ichheit"], but is rather a definite kind of Being which the 'I' itself possesses, such as having lost itself [Selbstverlorenheit].

Yet even the positive Interpretation of Dasein which we have so far given, already forbids us to start with the formal givenness of the "I", if our purpose is to answer the question of the "who" in a way which is pheno­menally adequate. In clarifying Being-in-the-world we have shown that a bare subject without a world never 'is' proximally, nor is it ever given. And so in the end an isolated "I" without Others is just as far from being proximally given. If, however, 'the Others' already are there with us [mit da sind] in Being-in-the-world, and if this is ascertained phenomenally, even this should not mislead us into supposing that the ontological structure of what is thus 'given' is obvious, requiring no investigation. Our task is to make visible phenomenally the species to which this Dasein-with in closest everydayness belongs, and to Interpret it in a way which is ontologically appropriate.

As for everything that appears for the mind to think on . . .

In the immediate circle of beings we believe ourselves to be at home. The being is familiar, reliable, ordinary. Nonetheless, the clearing is pervaded by a constant concealment in the twofold form of refusal and obstructing. Fundamentally, the ordinary is not ordinary; it is extra-ordinary, uncanny [un-geheuer]. The essence of truth, i.e., unconcealment, is ruled throughout by a denial. This denial is, however, neither a defect nor a fault – as if truth were a pure unconcealment that has rid itself of everything concealed. If truth could accomplish this it would no longer be itself. Denial, by way of the twofold concealing, belongs to the essence of truth as unconcealment. Truth, in its essence, is un-truth. We put it this way emphatically to indicate, with a perhaps off-putting directness, that refusal in the mode of concealing is intrinsic to unconcealment as clearing. On the other hand, the sentence "the essence of truth is un-truth" should not be taken to claim that truth, fundamentally, is falsehood. Equally little does it mean that truth is never itself but, dialectically represented, is always its opposite as well. (The Origin of the Work of Art (GA5) 1935/36, p.31)

So all very saṃsāra and Maya. The final recommendation is to let go of ascertainment, in a disavowal. Again from Being & Time (386):

Arising, as it does, from a resolute pro­jection of oneself, repetition does not let itself be persuaded of something by what is 'past', just in order that this, as something which was formerly actual, may recur. Rather, the repetition makes a reciprocative rejoinder to the possibility of that existence which has-been-there. But when such a rejoinder is made to this possibility in a resolution, it is made in a moment of vision; and as such it is at the same time a disavowal of that which in the "today", is working itself out as the 'past'.1 Repetition does not abandon itself to that which is past, nor does it aim at progress. In the moment of vision authentic existence is indifferent to both these alternatives.

(footnote 1. The idea seems to be that in resolute repetition one is having, as it were, a conversation with the past, in which the past proposes certain possibilities for adoption, but in which one makes a rejoinder to this proposal by 'reciprocating' with the proposal of other possibilities as a sort of rebuke to the past, which one now disavows.)

Again, all very buddhist. "That brahmana who does not grasp at a view, with what could he be identified in the world?" Sn 4.5

So we should believe in the existence of other minds with a grain of disbelief?

Pragmatically you have to go with what you know and roll the dice. The problem comes with identification. I'll finish with this quirky quote from Alfred Korzybski, Science & Sanity, 4th ed, page 187.

Identification, or the confusion of orders of abstractions, in an aristotelian or infantile system, plays a much more pernicious role than the present official psychiatry recognizes. Any identification, at any level, or of any orders, represents a non-survival s.r [semantic reaction] which leads invariably to the reversal of the natural survival order, and becomes the foundation for general improper evaluation, and, therefore, general lack of adjustment, no matter whether the maladjustment is subtle as in daily life, or whether it is aggravated as in cases of schizophrenia. A non-aristotelian system, by a complete elimination of 'identity' and identification, supplies simple yet effective means for the elimination by preventive education of this general source of maladjustment.

So basically, don't worry about it. Don't get hung up on labels. Other minds, maybe. It's sure going to get confusing with AI.

  • @xerx593 sounds a bit Westworld, which is a mixture of human and AI. The traditional solipsism quandary is whether other minds are real or everything is a dream or a simulation, in which case best proceed nobly and not worry about it. Aug 13 at 17:43

If we can't be 100% sure of anything, then we should trust everything with mistrust and suspicion?

That feels like an all-or-nothing stance. There is a spectrum of belief values in between 0 and 1:

  • It is CERTAINLY FALSE ( P = 0 )
  • It is almost certainly false ( P = 0.000... )
  • I distrust it / I am suspicious / it is highly unlikely
  • It is unlikely / doubtful
  • It is 50/50 ( P = 0.5 )
  • It is likely
  • I trust it / it is probable
  • It is almost certainly true ( P = 0.999... )
  • It is CERTAINLY TRUE ( P = 1 )

I would argue that P = 0 and P = 1 must both be faith-based.

How will it differ from agnosticism and pragmatism?

To list the different approaches for a variable X:

  1. X-theism: P=1, or 100% belief in X
  2. X-atheism: P=0, or 100% disbelief in X
  3. X-agnosticism: degrees of confidence as listed above, this may include SOME trust or uncertainty, even if I choose "it is almost certainly true"
  4. X-pragmatism: "I don't know what P or X is, or I think they are meaningless, I'm going to suspend my disbelief and carry on about my day"

Disclaimer: These statements are not peer-reviewed and they are not necessarily widely-accepted, they are just my opinion.


Unsure whether it is philosophically correct, but I like to apply the scientific approach to questions like that.

Some Physicians think that is does not matter whether a theory is exact. What matters is whether it can help us to better use the world where we live. A good example for that is the good old Newtonian physics where two objects exchange a gravitational force proportional to their masses and to the inverse of the square of their distance. We now do know that it is wrong at small scales where we have to use the quantic mechanics, and it is also incorrect at large scales, where the relativity is more accurate. But it still offers a simple model that is largely enough to make a plane flight.

We could use the same approach here. As we cannot make sure of anything, because we would first have to believe our senses (Matrix is a nice example where the true reality is not what the hero live...), what matters is whether we can build a consistent and efficient model. So the question is how will I live if I believe that no other mind can exist? Can I imagine a consistent model based on that in which I can live?

What I mean here is that a rational approach could be to know that nothing can be 100% sure, because it is impossible to find 100% sure evidence of it. Yet it is rational to act as if we could trust our senses, because it gives a consistent model. In fact, and until we can find better, that model is both the simplest and the most efficient (in terms of helping us to manage our life).

BTW the until we can find better is here to mitigate the risk of living in a Matrix like world, whether our senses will lie. It means that if we can find more inconsistencies in trusting the model than in rejecting it, then we must with no doubt reject it and try to find a better fitted one. But I must admit that until now, I have always been fine with it...


The problem you raise is similar to that raised by Descartes in his Meditations on First Philosophy.

I shall then suppose, not that God who is supremely good and the fountain of truth, but some evil genius not less powerful than deceitful, has employed his whole energies in deceiving me; I shall consider that the heavens, the earth, colours, figures, sound, and all other external things are nought but the illusions and dreams of which this genius has availed himself in order to lay traps for my credulity; I shall consider myself as having no hands, no eyes, no flesh, no blood, nor any senses, yet falsely believing myself to possess all these things; I shall remain obstinately attached to this idea, and if by this means it is not in my power to arrive at the knowledge of any truth, I may at least do what is in my power [i.e. suspend my judgment], and with firm purpose avoid giving credence to any false thing, or being imposed upon by this arch deceiver, however powerful and deceptive he may be. Source, page 8

The attitude of suspicion elicited by Descates' thought experiment of the Evil Genius is known as "methodological skepticism" or "methodological doubt". Of course, in modern times we know this idea in pop-culture through movies like The Matrix or, perhaps even more tellingly, The Truman Show.

Of course, if we regard absolutely all of our thoughts and perceptions with this type of methodological doubt then we have no starting point for knowledge, and therefore we fall into a kind of paranoid skepticism in which we can't trust anything. Even the belief that my own hand is in front of my face becomes subject to corrosive doubt. The problem is not just that we have some level of uncertainty, it is that we have an unmeasurable level of uncertainty. There is no upper or lower limit we can place on this doubt. Since it is based on the mere possibility of being systematically deceived, we seemingly cannot limit the epistemic damage in any way.

Philosophers have of course tried very hard to find ways to stop the slide from methodological skepticism to paranoid skepticism. Descartes himself famously argued that we can in fact have certainty about some of our beliefs, such as that "I am". Based on such foundational certainties we can construct a solid edifice of knowledge. This view in modern terms would be called infallibilist foundationism. The infallibilism means there are certain beliefs that we literally cannot be wrong about, they are self-evident and undoubtable. The foundationism part means that we can build all of our knowledge on the basis of these self-evident beliefs.

However, most people aren't convinced by Descartes solution to his own problem. In my opinion, the most promising reply to the specter of paranoid skepticism is known in contemporary philosophy as the default and challenge model of knowledge. The argument for this model says, look, even in considering the possibility of Descartes' Evil Genius deceiving us, we start out thinking we know basic facts about the external world (e.g., there are tables and chairs). What the Evil Genius argument shows is that there could be a scenario which would undermine or defeat the justification of our ordinary beliefs. However, the crucial thing is that this only defeats our prima facie justification for believing in tables and chairs, if we had a positive reason to believe that we lived in the world with and Evil Genius. In other words, we'd need to undergo some kind of experience that convinces us that we really do live in the Evil Genius world. Hollywood has given us great examples to illustrate this. Neo takes the red pill and then he wakes up in the "real world" which outside of the Matrix. In The Truman Show, Truman literally bumps into the edge of the sky which turns out to be a concrete dome. The argument says that it isn't enough for a scenario which undermines our present state of knowledge to be merely possibly; it has to be actually true (and we must have reason to believe this) in order for it to undermine our present state of knowledge.

Thus, we get the default and challenge model of knowledge. This is a version of fallibilism, in that any part of our belief system is open to challenge and revision in light of new evidence. However, in this model there are no unmotivated global epistemic defeaters allowed. In a sense this model challenges the unstated assumption in Descartes that we should begin philosophy with unrestricted methodological skepticism. Instead, it argues that we always begin with some pre-existing knowledge which we are justified in trusting until it is overturned by positive evidence to the contrary. In one way then we still believe everything with a grain of salt, but this doesn't mean we have to constantly be suspicious or paranoid. We can be very confident in those of our beliefs that have stood the test of time since we know that it would take a truly radical new argument or new set of experiential evidence for us to have good epistemic reason to revise those beliefs.

For a good intro to these arguments see Problems of Knowledge by Michael Williams. Wittgentstein articulates a similar position in On Certainty as does Merleau-Ponty in Phenomenology of Perception. Many philosophers seem to have independently converged on the default-challenge model as the best response to Evil Genius type skeptical scenarios.


To an extent, degrees of belief that are only theoretically (rather than practically) non-negligible are both meaningless and interesting.

Notwithstanding the serially misled/gullible (why do you keep returning to the same used car salesman who misled you last time and the time before), if you can't test something then you cannot really find out you've been tricked (into believing e.g. in other minds).

So while there is something epiphanic about realising we don't know for absolute certain: doubt that the sun will rise tomorrow is more interesting for what it says about us.

  • what might the consistent fallibility of all induction say about our mysterious mathematical sciences?
  • if we can give up on certainty, then what other epistemic virtues are there on offer?


  • 1
    Woo! Nice answer. Both interesting and meaningless is both interesting and meaningful 😉
    – Rushi
    Aug 13 at 6:21
  • haha thanks @Rushi
    – user67155
    Aug 13 at 6:25
  • 1
    Occurs to me : Zen koans are by definition hopelessly meaningless and incorrigibly interesting. Thats what gives them at a meta level an abiding meaningfulness. ie viva Gödel
    – Rushi
    Aug 13 at 6:46

If we can't be 100% sure of anything, then we have to believe everything with a grain of salt?

Yes. This is a psychological challenge, not a logical one. Human beings have evolved to be afraid or confident in the face of uncertainty. The classic expression of this is thinking about how would a bushman in the African bush respond to the sound of a potential lion in the bush. On the one hand, a rustling in the bush can and should induce fear and lead to a bias in favor of being confident a lion is there even if it's not. On the other hand, if a lion does appear, fearlessness to engage the lion should be the default response because a person cannot outrun a lion. (Certain tribes in Kenya, for instance, consider the slaying of a lion a passage into adulthood.) Overconfidence is so regular in bad reasoning, it has its own bias.

Therefore, it can be a bit disorienting to find out that the confidence some people show in their parents, gods, loved ones, belief system, or political units is misplaced. This is particularly pronounced in individuals who begin to suffer from paralysis by analysis. Such individuals might respond by placing excessive confidence in formal logic as a tool to deal with the world to cope with the fact that reason is defeasible. Thus, one might group philosophers into three buckets based on confidence in reason and knowledge: skeptics, foundationalists, and fallibilists (IEP). From IEP:

Fallibilism is the epistemological thesis that no belief (theory, view, thesis, and so on) can ever be rationally supported or justified in a conclusive way. Always, there remains a possible doubt as to the truth of the belief. Fallibilism applies that assessment even to science’s best-entrenched claims and to people’s best-loved commonsense views.

A strong or radical skeptic will discard knowledge as an enterprise. No confidence is warranted in reason because there is no way to adequately justify belief. And a foundationalist will constantly strive to show that their premises or metaphysical presuppositions are inerrant, and therefore there logic is superior and well-grounded ensuring their conclusion. A fallibilist on the other hand simply insists that instead of confidence or doubt, some confidence is warranted, and that the devil lies in the details. In this way, a fallibilist is like a situational ethicist.

The fallibilist is likely to receive pushback from those who say fallibilism is overconfident, and also from those who insist it is not confident enough. Thus, the fallibilist must answer both responses. Among professional philosophers, such responses are likely to be very technical, but two broad outlines can be sketched. To the skeptic who undervalues certainty, a falliblist will merely point out the pragmatic benefit (IEP) of maintaining and using confidence intelligently. Certainly, in the face of a lion running at us, we should be confident harm is imminent. But for the foundationalist who insists on their argument, a fallibilist merely can undermine metaphysical presuppositions as any quietist is wont to do. After all, how can one ever truly defend presumptions except as themselves conclusions. And then the Agrippan trope of infinite regress roars loudly.

For example, the existence of other minds, the existence of the outside world, etc. is currently impossible to prove with absolute, 100% certainty.

And what of it? The beautiful thing about the brain is it allows us to adapt! What is relatively certain today becomes doubtful tomorrow. What we had no idea was possible suddenly becomes obvious. Ultimately, human learning is exactly what makes us anti-fragile, and that is a virtue.


If we can't be 100% sure of anything, then we have to believe everything with a grain of salt?

No. Even if consciousness is everywhere, you need not believe this until you are sure. Even though this is the truth, you may take even this with a grain of salt.


But without any doubt you can believe in your existence, because after your death, you must be somewhere (You may believe as it is in some form also). But if you are going to take all the rest with a grain of salt, you cannot enjoy anything in your life. How would you learn something that you are taught if you are a doubting Thomas?

If you say ‘yes’ to this question, that implies your existence also is doubtful and so there is no need to answer to your question. So, even if the logical aspect of philosophy is considered, the answer to your question is ‘No’.

How will it differ from agnosticism and pragmatism?

Regarding this type of believe (or Regarding this question), an agnostic will not achieve any good that some great people achieve/realize while a pragmatist might please others by helping them with the things to increase sensual pleasures without any self-satisfaction to oneself. Both of them cannot understand the actual value of this human life. Both these philosophies are concerned only with the maximum thing attainable in animal life. So the end result for both philosophy is disappointment.

Now, again to your first question:

You may doubt even the salt you add :) It is good to keep you always alert. However you need not doubt your existence. Some religions say “I am everything. I pervade everything.” etc. The dispute is because of the ignorance about the “I”. All these ideas point to one truth – to culminate logic (mentioned above) and religion the answer must be ‘No’. In other words, when you realize that logic and religion culminates, you will understand the answer to your question must certainly be ‘No’. Otherwise there must be more than one ultimate truth – one is what religion says (if what religions say is the supreme) and the other, what logic says. This is again illogical. Or, this might lead to the fact that no religion leads to the ultimate truth.


According to David Hume,

... In our reasonings concerning fact, there are all imaginable degrees of assurance, from the highest certainty to the lowest species of moral evidence. A wise man therefore proportions his belief to the evidence... He weighs the opposite experiments; he considers which side is supported by the greater number of experiments - to that side he inclines with doubt and hesitation; and when at last he fixes his judgment, the evidence exceeds not what we properly call "probability".

  • An Enquiry Concerning Human Understanding. Section 10.

Hume's point is that the wise or rational person does not attach too high a degree of credence to a proposition, considering it to be beyond all possibility of doubt. Equally, they do not attach too low a degree of credence, considering everything to be a mere guess or supposition. Rationality is concerned with weighing up the evidence and finding the appropriate degree of credence considering all available information.

One might add that it is also a principle of rational thought to be aware always that one might be mistaken and to be willing to change one's mind in the light of fresh evidence. So it is reaonable never to be absolutely certain that a proposition is true, nor that it is false.

Hume even uses the word 'probability' to describe this rational degree of belief. This perhaps anticipates the later development of Bayesian methods in using probabilities to represent rational degrees of credence and formulating models of belief revision based on probability theory. Within Bayesian theory, the principle that one should never assign a probability of exactly zero or one to any proposition is referred to as Cromwell's rule.

So, it is inappropriate to treat everything with mistrust and suspicion. But it is appropriate to allow for remote possibilities, even very strange ones. Maybe the external world does not exist, or it is a simulation, or other people are projections. Such possibilities are so weird, I am not willing to grant them more than a tiny degree of credence, but tiny is not zero.

Agnosticism is different in that it is more a case of weighing up the evidence relating to some proposition and coming to the conclusion that there just is no way to assess a rational of degree of credence. It is a meta-level judgment that no value for the probability is available.

Pragmatism is more a case of saying: never mind the probabilities, I am just going to form beliefs based on what seems intuitively reasonable and see how things turn out. If they turn out badly, I'll try something else.

Philosophers are highly experienced and proficient at doubting things, but even they do not live in a constant state of doubt. It is too mentally exhausting to doubt everything all the time. So, in practice we reserve most of our doubting and weighing up of evidence for more reflective moments. Most of the time we live by what we judge to be highly probable and that's good enough for practical purposes.


"So we should believe in the existence of other minds with a grain of disbelief?"

What does this achieve? If we meet, and I behave like I behave because of my mind, or I behave like I behave because you are in some simulation, or because I'm a robot that behaves exactly as if it had a mind, what difference does that make to you?

You can't express your disbelief, because if you express your disbelief that I have a mind to me, a person with a mind, there will be painful consequences for you. And if you express your disbelief to that robot, there will be exactly the same painful consequences. Although in that case, the painful consequences might be just in your mind - which doesn't make them one bit less painful.


If we can't be 100% sure of anything, then we have to believe everything with a grain of salt?

No because beliefs might not exist in order to "believe everything with a grain of salt."

See: Eliminative materialism in relation to the existence of beliefs.

You must log in to answer this question.

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