I am thinking that simple statement such as "I am", "I think" are all beliefs that are also knowledge and conditional statements such as "This dog does exist in this world if the world is real" are also knowledge since they are true even if proven to be an illusion.

  • 7
    No, and it makes little sense. To a radical skeptic every statement is in doubt, specific statements cannot be counterexamples to an attitude. Especially one that applies to all statements, them included.
    – Conifold
    Commented May 22, 2021 at 23:46
  • 5
    "To a radical skeptic every statement is in doubt..." I hope you're ready to back that up! ;) Commented May 24, 2021 at 14:42
  • "are also knowledge since they are true even if" - I think "knowledge" is about more than just statements that are necessarily true. The statement also probably needs to actually tell you something about the world, which a conditional statement arguably doesn't do. The example given might also be problematic because you might be imagining the dog even if the world is real. Even if you assume the world is real, there is still very little you can say for sure if you're skeptical of your perception of it.
    – NotThatGuy
    Commented May 24, 2021 at 15:26
  • @Conifold see my answer. I am interested in your feedback.
    – Nikos M.
    Commented Jul 18, 2022 at 14:23
  • This is confusing soundness and truthiness. "This dog is real if the world is real" is sound, but true only if the world is indeed real. As long as the premise is not established with an adequate level of confidence, it has no knowledge about the world to offer.
    – armand
    Commented Jul 23, 2022 at 0:17

6 Answers 6


A good question.

Every machine is fallible, including the human brain; this means that when the human brain comes up with any conclusion, there is always the possibility it made an error and came to the wrong conclusion. Even if the brain in question is your own. Computers can make occasional errors in calculation due to manufacturing defects, cosmic ray impacts, hard disk drive failure, software bugs, or even quantum tunneling (the latter with extremely low, but nonzero probability, and the fact it is nonzero is all that matters to the point). That means that when a computer tells you the result of any calculation, you cannot rationally trust it 100%. Perhaps you can trust it very close to 100%, but never exactly 100%.

The human brain is a piece of hardware subject to many of the same kinds of faults that affect computers. Humans are known to make calculation errors at a much higher rate than computers do. Therefore, we should not say we "know" something, unless by that we mean that we believe it with high confidence. Knowledge is a kind of belief, not something fundamentally different from belief. Everything is in doubt.

"Even that?"

Yes. It's a consequence of the physical basis of your brain.

Note that, if everything is uncertain, the degree of uncertainty is also uncertain. If you calculate a 99.999999% chance that you're right about X, you can't trust that calculation perfectly. Maybe you made an arithmetic error and really the chance of being right about X is only 30%!

As a practical matter, almost everyone is more confident of their own views than is objectively warranted. People, probably including you and me, would be more rational if they were less certain. Doubting your views is the only way to change them, and changing your views is the only way to become more right over time. Of course, it is not guaranteed that change is for the better; doubt also the change. But because like most people you likely have a bias towards overconfidence in your established views, it is necessary to put more effort into doubting your own views than doubting the change.

I've sometimes seen a strange reaction to the fact that all knowledge is uncertain. Some people associate uncertainty with the notion that truth is subjective, or that because we can never certainly reach absolute truth, it doesn't matter to us. This is like saying that because maps are sometimes printed with mistakes, mountains don't exist or don't matter to us. It's a complete non sequitur. It is entirely possible to both believe that a mountain exists because it's shown on the map, and that the mountain might not exist because the map might be misprinted. Not only is it possible, it's the most reasonable perspective. One believes in objective reality at the same time as one acknowledges that all of one's beliefs may be mistaken.

  • 1
    I am quite surprised by the certainty coming out of this answer in defense of radical scepticism!
    – Nikos M.
    Commented Jun 3, 2022 at 21:50
  • 1
    @NikosM. Does it matter how certain I am? Suppose that there's only a 5% chance that I'm right. This would be a 5% chance that we can't rationally be perfectly certain of anything. Well, wouldn't that mean that indeed we can't rationally be certain of anything? Since 5% of whatever small chance of error I'm proposing is still nonzero.
    – causative
    Commented Jun 3, 2022 at 22:22
  • Even if we can be certain of one thing among millions (imagine that) this means radical scepticism is false, but I am hungry right now, I'll pass.
    – Nikos M.
    Commented Jun 3, 2022 at 22:28
  • 1
    @NikosM. So you see, to consistently say that I'm wrong, you would have to be completely, 100% certain that I'm wrong, since any small chance that I'm right admits my entire point. And how would you be so certain, when the points I'm making about human fallibility are hard to deny?
    – causative
    Commented Jun 3, 2022 at 22:28
  • 2
    @NikosM. "Eat your words but don't go hungry. Words have always nearly hung me."
    – Scott Rowe
    Commented Jul 20, 2022 at 17:52

One ancient counterexample for radical skepticism is just ask such a skeptic whether he or she accepts the skeptic's own knowledge that "one should reject any knowledge". If the skeptic accepts then he or she contradicts own POV literally, if one rejects then one rejects one's own POV directly. Either result seems logically awkward for the radical skeptic.

Some modern radical skeptics use Brain in a vat (BIV) to argue that technology can be so perfect that one can never know if one is real or just a "disembodied" brain in vat, thus knowledge is impossible:

The simplest use of brain-in-a-vat scenarios is as an argument for philosophical skepticism and solipsism. A simple version of this runs as follows: Since the brain in a vat gives and receives exactly the same impulses as it would if it were in a skull, and since these are its only way of interacting with its environment, then it is not possible to tell, from the perspective of that brain, whether it is in a skull or a vat. Yet in the first case, most of the person's beliefs may be true (if they believe, say, that they are walking down the street, or eating ice-cream); in the latter case, their beliefs are false. Since the argument says one cannot know whether one is a brain in a vat, then one cannot know whether most of one's beliefs might be completely false. Since, in principle, it is impossible to rule out oneself being a brain in a vat, there cannot be good grounds for believing any of the things one believes...

Please note that any asserted knowledge like "I have a hand" could be leveraged by a radical skeptic as an argument for radical skepticism holding the view that one cannot know she is a BIV or not since technology may be prefect enough not to be able to distinguish any more. How? First the skeptic will claim we all agree with a self-evident principle of epistemic closure:

Epistemic closure is a property of some belief systems. It is the principle that if a subject S knows a (proposition) p, and S knows that p entails q, then S can thereby come to know q.

Since the proposition you have a hand entails the proposition you are not a BIV, once you know you have a hand, then you must also know you are not a BIV. But this is a contradiction with the undisputable fact that one can never know if one is BIV or not as introduced context above. So you cannot claim you know you have a hand...

Then how to construct a counterexample for this hopeless situation? Some philosophers like Robert Nozick proposed we have to abandon above "self-evident" principle of epistemic closure and change the philosophical JTB definition of knowledge. If you adopt this approach, then by external experimental confirmation of the existence of our hands then we can somehow confidently claim we have some certain real knowledge if common probabilistic types of knowledge doesn't qualify per your standard...

  • 8
    One should doubt any knowledge, not reject it. And yes, one should doubt the idea that one should doubt all knowledge. There is no contradiction here, just a surplus of uncertainty.
    – causative
    Commented May 23, 2021 at 0:49
  • 3
    @causative For a thorough empiricist one may doubt math since everything is synthetic a posteriori by deviating from Kant... A radical skeptic not only healthily doubt any knowledge, he or she may completely reject any knowledge outright... Commented May 23, 2021 at 1:22
  • 3
    @causative in your reference I found "radical skepticism can be used as an objection for most or all beliefs" so seems to me it's conceivable to reject (object) all beliefs (JTB knowledge) as a radical skeptic. Btw, there's another more radical ancient school called Cratylism (en.wikipedia.org/wiki/Cratylism), "even if something can be known about it, knowledge about it cannot be communicated to others. And, finally, even if it can be communicated, it cannot be understood" Commented May 23, 2021 at 1:39
  • 2
    It can be used as an objection, but this objection is not about negating them and saying they are false - it is saying they are uncertain. You get a self-referential paradox from saying everything is false, but you don't get a paradox from saying everything is uncertain.
    – causative
    Commented May 23, 2021 at 2:50
  • 1
    If I'm in the Matrix, and there are no glitches in the Matrix, then it really does not matter whether or not I'm in the Matrix.
    – RonJohn
    Commented May 26, 2021 at 0:32

A realist and recovering ex-engineer weighs in.

The process of living one's life and (metaphorically) getting from A to B constitutes the solution of a never-ending series of problems. Engineer are trained to solve certain classes of problems in the physical world and with the passage of time and experience, they eventually get good at it and are able to pay bills, buy houses, support families and so on.

This process is imperfect, meaning you make mistakes and must recover from them and with luck converge on the truth through an iterative process which is built on the accumulation of facts and data.

As such there is no place for radical skepticism in this paradigm. Passenger planes, cars, houses, computers, bridges, freeways, MRI machines, sewage treatment plants and so on are not invented nor are they built by radical skeptics.

In this sense, therefore, radical skepticism is amply refuted by the ability of humans to solve problems, which has furnished us with the world we currently live in.

It is an imperfect world to be sure, but as a very recent cancer survivor I owe my current existence to MRI and CT machines and the people who know how to use them to solve problems.

  • I think the point of skepticism is that we can never be sure that we know anything. Sure, MRI and CT machines are great, but we don't really know how they work. Our current explanation is based on particles and fields which behave a certain way in space and time, but even most physicists admit they don't really understand quantum mechanics and relativity. Any day a new Einstein could come along and figure out that all our current beliefs about physics are actually wrong... Commented May 23, 2021 at 16:33
  • 2
    @jcsahnwaldtReinstateMonica, The people who manufacture them do. Note that neither relativity nor quantum mechanics are needed to explain their operation. Note also that it has been known for decades that neither relativity nor quantum mechanics can be the ultimate explanation for all physical systems, in particular those which existed (only) in the very early universe. Finally, remember that whatever comes after GR and QM will in no way invalidate them but rather replace them in those special cases where they are already known to fail. V isit the physics SE for more about this. -NN Commented May 23, 2021 at 19:14
  • 1
    But I guess this discussion hinges on the exact meaning of words like "knowledge" and "validity". If we accept that "knowledge" is never absolutely reliable and may change any moment, then we "know" quite a bit - enough to build intricate machines. But there's no "knowledge" that is 100% "valid". Commented May 23, 2021 at 22:29
  • 3
    P.S.: I just read my first comment here again and realized it's indistinguishable from some utterances by postmodernists, esotericists etc., who may say stuff like "our knowledge of science/astronomy/MRI machines isn't 100% reliable; therefore, religion/astrology/Reiki is just as valid". Of course, that's nonsense. See the last paragraph of @causative's excellent answer for details. Commented May 24, 2021 at 11:21
  • 1
    The stories here, about the mistaken but operable theories of Niels Bohr and Arnold Sommerfeld, show that it would be possible to design a machine that could operably harness some natural phenomenon even though the designer actually misconceived that phenomenon. Commented May 25, 2021 at 10:50

There's two types of knowledge: Knowledge from looking at the universe, and created knowledge.

To be skeptical of created knowledge makes much less sense than knowledge derived from observation.

An example of created knowledge is the author of a story: How does he know where the hero has hidden secret widget? Because he decided.

Another example might be where you live: How do you know you live at 123 Acme Street? Because you decided to live there. I guess this latter combines observed knowledge that a house existed there in the first place, and an agent put a contract under your nose. But still there is a decision point there.

I guess it is possible to doubt your own decisions, but that route insanity lies, and is a common theme in psychological horror, such as Shutter Island.

  • I don't see the relevance of decisions on knowledge. You know where you live because you may have seen the housing contract, the name of the street, the number of the house and saw your address on Google Maps. It has nothing to do with you deciding to live there (especially if it's the house you grew up in). Some people may have made decisions at some point, but the knowledge comes via observation. And the obvious counterexample against trusting "knowledge from looking at the universe" would be: if someone thinks they've seen Bigfoot, should they just believe that, or be skeptical of it?
    – NotThatGuy
    Commented Feb 17, 2023 at 10:21
  • @NotThatGuy You said, "knowledge comes via observation". Some knowledge comes from observation. But some knowledge is created knowledge. How does JK Rowling know that Harry Potter's name is Harry? She decided it. Her readers observed it. But she decided it. She did not write down a random name, and then look at what she wrote and go, "Aha! I observe his name is Harry Potter!"
    – Stewart
    Commented Feb 17, 2023 at 15:52
  • Technically I said "the knowledge comes via observation", meaning the specific knowledge that the example relates to (not all knowledge). In any case, I probably wouldn't see an author's knowledge of the name of a character as relating to a decision as such, as much as it's about our ability to know our own minds. And that knowledge cannot be considered perfectly reliable for the past, given that it's well-established that memory is unreliable. And as far as everyone else is concerned, an author's choice of character name is knowledge that comes through observation.
    – NotThatGuy
    Commented Feb 17, 2023 at 16:10

Is there any counterexample given against radical skepticism?

Here is my humble opinion: radical skepticism has no practical value. The level of confidence someone needs before acting rises or falls with the importance or lack of importance of the decision under consideration. If there is a life-altering decision to be made, then nearly 100% certainty might be called for. For a lot of day-to-day decisions, it is often sufficient to accept what is true randomly.

But what if you don’t know the level of confidence that is justified by the data, or even whether the data itself is accurate? The answer is that you do what you can. It could be anything: run an experiment, check the newspaper.

But the answer to the basic philosophical question, while interesting, does not affect daily life.

  • Good point on the practical value, or rather lack thereof, of radical skepticism!
    – Nikos M.
    Commented Jul 18, 2022 at 21:08
  • "The level of confidence someone needs before acting rises or falls with the importance or lack of importance of the decision" - this assumes that any of your existing beliefs are correct, assumes you'd reach the correct conclusion on first analysis and assumes your judgement of importance is accurate (also assuming you don't care if you're right about not-so-important things). All of that is what skepticism challenges, and the practical value is where we're wrong to assume some of those things (which could e.g. tell us whether we should believe a god exists or change our views on morality).
    – NotThatGuy
    Commented Feb 17, 2023 at 10:10

What is radical skepticism?

  1. If it means "I do not assert anything", it is itself an assertion thus self-refuting.
  2. If it means "I only assert that I do not assert anything else". Then it is also impossible in the sense that it cannot be maintained. Since in order to maintain it while acting leads to an infinite regress, which makes any act while holding this position impossible. Any act, even eating something to satisfy hunger, necessarily is an assertion of something else (eg I am hungry, I eat, apples fall). If the skeptic says "I do not assert that, I allow that is probably true or false", then it means that there is a rule to act under uncertainty, which is itself asserted in practice (and through this, other things asserted). The radical skeptic can either accept this something without further doubt, thus refuting the position, else go into an infinite regress, making any act impossible. Since the radical skeptic demonstrably exists and acts this means the skeptic has refuted the position even without admitting it.
  3. If it means "Sometimes I may be in doubt", it is not radical.
  4. If it means "I always doubt, but I act/assert as if certain", then it is rather meaningless, if it is not simply insincere.

According to the above analysis, B. Russell's point is demonstrated:

Skepticism, while logically impeccable, is psychologically impossible, and there is an element of frivolous insincerity in any philosophy which pretends to accept it.

In the sense of the above analysis, it is no wonder the archetypal depiction of the skeptic philosopher is sitting down, doing absolutely nothing (if "doing absolutely nothing" could really be depicted), while, in contrast, a philosopher like Aristotle was always walking.

skeptic philosopher depiction

  • 4
    One can both assert things, and doubt them, at the same time. If I am 99.9% sure of X - meaning, I think there's an 0.1% chance that I am wrong - then I will assert X. The assertion of X can be translated as a claim that X is probably true (above some vague threshold of confidence). So, we can add a fourth option for radical skepticism to the 3 you have listed: "I doubt everything, and assert whatever I am sufficiently confident of, which happens to include this statement."
    – causative
    Commented Jul 18, 2022 at 17:30
  • 1
    You don't assert with full certainty, that's the point. You assert with high but imperfect confidence. That's what it means to assert. Of course, much of the time when someone asserts something, they are wrong - you agree with that? And if someone can be wrong when asserting something, then they must not have had perfect deductive certainty, because if they had, they could never have been wrong.
    – causative
    Commented Jul 18, 2022 at 19:03
  • 2
    It's perhaps worth mentioning Russell also argued for skepticism as one of his maxim: The whole problem with the world is that fools and fanatics are always so certain of themselves, and wiser people so full of doubts... Commented Jul 18, 2022 at 21:46
  • 3
    You are trying to assimilate skeptics to your own views, and, as a result, they are lost in translation. "Any act, even eating something to satisfy hunger, necessarily is an assertion" is false on most accounts, regardless of skeptics. The difference between "sometimes I may be in doubt" and "I always doubt" is just psychology, the philosophical question is whether one ought to doubt everything (including that), not whether feeble humans follow through. "Maintaining while acting" is moot on their view, and not only theirs. This is why Russell calls it "logically impeccable".
    – Conifold
    Commented Jul 19, 2022 at 5:38
  • 2
    Actions can be used to make assertions (and vice versa), but need not be so used. One can assert things and act without any correlation to them, animals act, but assert nothing. It would be a tall order to ask of humans, especially philosophers, that they live up to their assertions, so why should skeptics be held to that standard. One can make all sorts of ethical charges and psychological surmises about it, insincerity, hypocrisy, "paper" doubt, etc., but none of that matters. Neither skepticism nor any other philosophy has to be embodied in action to be "meaningful". So they'd say.
    – Conifold
    Commented Jul 19, 2022 at 10:27

You must log in to answer this question.

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