How Can We Be Sure About Anything at All?

The problem of the external world

Like many students before me, my first faltering steps into philosophy were guided by some of the ruminations of the 17th-century thinker René Descartes. Here is what I took from Descartes: there is no way you can be certain that even your most ordinary experiences do not radically misrepresent the real shape of the world. No matter how vivid your experiences, they might, for instance, simply be the product of a vivid dream or of the deceptive influences of some malign force.1

However, this problem of the external world, as it is called, calls into question more than just whether our experiences accurately represent the external world—it calls into question the very existence of any external world at all. For all you can tell, your mind is the only thing that exists, and the surprising and seemingly out-of-your-control aspects of “the world” are just produced by your own unconscious mind. What possible evidence could disprove such a thesis, or even render it improbable? Certainly Descartes’ own solution—deduce the existence of a benevolent God from the very idea of God, and then realize that a benevolent God would not allow his creations to be so systematically deceived—does little to convince.

Problems like these usually provide an acid test for whether or not one has the appropriate temperament for philosophy: if such problems occasionally keep you awake at night, then you have the appropriate temperament; if they do not, then you do not. I confess, the problem of the external world keeps me awake on occasion, but it compensates me for lost nightmares by being a nightmare in itself. Nevertheless, this particular problem is not the main topic I want to talk about in this article. Rather, I want to call attention to an even more severe problem that Descartes briefly mentioned, but which is not as often discussed.

Pathology and thought

Even if all of your apparent experiences of the external world are illusions (not that Descartes claimed this—he just said that, barring his proof of a benevolent God, we can’t be certain that they are not illusions), surely there still is much you can know beyond your mere existence. Surely, for instance, there can be no doubt about the truth of elementary arithmetic propositions, such as the proposition that 1 + 1 = 2. Surely! But, Descartes is not convinced:

What is more, since I sometimes believe that others go astray in cases where they think they have the most perfect knowledge, may I not similarly go wrong every time I add two and three or count the sides of a square, or in some even simpler matter, if that is imaginable? (Cottingham, et al., 1984:14)

My automatic reaction when I first read this passage was to assume that when others go astray despite claiming perfect knowledge, they must be claiming perfect knowledge about something far more complex than basic arithmetic truths; either that, or their claims of perfect knowledge must, deep down inside, be insincere. It seemed to me unfathomable that anyone reflective could sincerely come to believe something on par with 1 + 1 = 3. And so, I effectively ignored or glossed over the passage for more than a decade. Then, I learned about a few unusual neurological disorders.

There is a condition called Capgras syndrome, wherein the afflicted person, while perfectly normal in all other respects, becomes firmly convinced that a loved one has been replaced by an imposter. According to the neurologist V. S. Ramachandran (2004:7-9), the condition probably is caused by the failure of the brain to transmit information properly to the amygdala, the part of the brain responsible for producing the feeling of emotional connection that normally accompanies recognition; since the frontal cortex notices that the feeling of familiarity is not present, it infers that what so clearly looks like the one you know and love, must actually be a very clever replica. The attending neurologist may take note of the condition, but, apparently, no evidence can sway the afflicted: the conviction produced by the brain is too strong, and overwhelms all counterarguments.

Another bizarre condition described by Ramachandran (32-35) is mirror agnosia, wherein a person with damage to the right hemisphere of her brain loses all awareness of everything to her left. If asked to draw a picture, she will draw only the right half. When putting on makeup, she will attend only to the right side of her face. Most striking is what happens when one places a mirror in the right side of her field of vision, reflecting an image of an object in the left side, and asks the patient to reach out and grasp the object. Ramachandran describes:

This is a problem that can be solved without difficulty by a three-year-old child. Even a chimpanzee doesn’t confuse a mirror image for the real object. But the wise Ms. D—in spite of seventy years of experience with reflections—reaches straight into the mirror…I think what happens is the patient knows she’s looking at a reflection, therefore the object is on her left. But because left doesn’t exist in her universe, the only possible explanation, however improbable, is that the object is inside the mirror. Remarkably, all her abstract knowledge about the laws of optics and mirrors is distorted to accommodate this strange new sensory world in which the patient finds herself trapped. (35)

A third condition described by Ramachandran, called Cotard’s syndrome, is the most disturbing of all: patients afflicted with Cotard’s syndrome believes they are dead. What—dead? Surely, one would think, such a belief could easily be corrected by an appeal to the patient’s reason. Not so, as Ramachandran explains:

The delusion of Cotard’s is notoriously resistant to intellectual correction. For example, a man will agree that dead people cannot bleed; then, if pricked with a needle, he will express amazement and conclude that the dead do bleed after all, instead of giving up his delusion and inferring that he is alive. Once a delusional fixation develops, all contrary evidence is warped to accommodate it. Emotion seems to override reason rather than the other way around. (91)

What makes conditions like these so much more disturbing than Matrix-type thought experiments, is that they involve not merely deep errors of perception, but deep errors of basic understanding. In light of such conditions, even the most hard-headed, philosophy-disdaining, common-sense realist must admit that one’s brain can produce an overwhelming conviction (and not just arrogant swagger, or self-deception which one admits deep down inside, but a real sense of knowledge) of the truth of a proposition which is in fact patently absurd.

The problem is that once you admit that the brain can produce an unshakeable conviction of the truth of that which is in fact obviously false, how can you be sure that you do not suffer from an affliction of that type right now? On what basis are you so cocksure that your certainty about even the most seemingly elementary, seemingly obvious truths, is not the product of a delusion?

But, you might say, we can’t all be afflicted by a delusion—it is fine and well to take note of a few scattered people with brain damage, but a completely different matter to speculate about the possibility that we all suffer from comparable problems.

There are two rejoinders here. In the first place, it may, in fact, not be that we all suffer from comparable problems. It may just be you. It may be that no one actually agrees with you about the things you take to be elementary. Why would you think they do agree? Because they have told you so? But how do you know that your belief that they have told you so is not part of the delusion from which you may be suffering?

In the second place, why couldn’t we be wired, as a species, to suffer from certain kinds of delusions, in the same way that we are wired to experience certain kinds of optical illusions? Why couldn’t deep cognitive delusions result from the normal function of a healthy human brain? Every other species has cognitive limits. Aren’t we all familiar with cases where one can trick other animals into doing things that a rational creature would never do, simply because other animals are more constrained by instincts that natural selection has tailored to a particular environment? Yet  we, too, have evolved in a particular environment, so we should allow at least the possibility that our rock-bottom cognitive processes are constrained by adaptation in deep history. This may, for instance, account for why quantum mechanics and relativity are impossible for us to grasp intuitively. Fortunately, our power of abstract thought appears to help us to overcome these obstacles somewhat and at least generate mathematical models of nature, but isn’t it at least possible that there are biological constraints on our thought which even our capacity for abstract reasoning cannot overcome? Could we not be like the mirror agnosia patient, helpless despite her knowledge of optics?

All of this reference to brain damage is a lever to help motivate Descartes’ insight. None of it is strictly necessary to make the point, but it helps us to see that his worry wasn’t so absurd after all. How, then can we be sure we know anything at all? How can we be sure that our reasoning is not pathological through and through?

Can we at least save logic?

One suggestion follows the model that Descartes used to save at least the certainty of the proposition that he himself exists. Descartes noted that he could not coherently doubt that he himself existed, because, if he did not exist, then what was doing the doubting? Likewise, not even God could deceive Descartes into thinking he existed, because if he did not exist, then there would be no Descartes for God to deceive. In short, Descartes was able to conclude, famously, that “I think, therefore I am,” because the denial of his own existence would undercut the denial itself.2

Of course, if we are worried about the legitimacy of our reasoning, then perhaps we should be worried about Descartes’ reasoning, no matter how transparent it seems. But, so the argument goes, we do not need to be worried about our reasoning at least at this level: we must, at the very least, assent to the laws of logic, because without logic, no skeptical argument can get off the ground. Consider:

  1. If, for all we know, our conviction of the legitimacy of logic may be a delusion, then we cannot be certain that our conviction of the legitimacy of logic is correct.
  2. For all we know, our conviction of the legitimacy of logic may be a delusion.
  3. Therefore, we cannot be certain that our conviction of the legitimacy of logic is correct.

This seems like a sound argument; however, the argument is based on the logical form modus ponens (If p, then q. p. Therefore, q.). If the conclusion is correct, then we cannot be sure modus ponens works, and this uncertainty, in turn, would prevent us from drawing the conclusion from our premises. Hence, trying to doubt logic turns out to be self-defeating. We are still left with many questions, but perhaps at least there is this bare core of certainty upon which we might be able to build. And, when you consider the vast metaphysical systems rationalist philosophers (including Descartes) have attempted to spin from pure logic and the bare ideas in their minds, perhaps this is not a conclusion that must consign one to despair.

I will say that I am not entirely at peace with this last move. Half of the time, I cannot see any problem with it. The other half of the time, it feels like a parlor trick to me, because I have reminded myself that the relevance of my inability to see any problem with something is precisely what is at issue. I can do no more at this point than commend the issue to my readers.

Well, if they exist.


1 Science fiction has had an especially fun time exploring this theme. Both The Matrix and Inception, for instance, are great Cartesian movies.

2 Although, as Hume later pointed out, if Descartes means he has proved his existence as a being that has thoughts, even that may not strictly be justified—if “free-floating” (so to speak) ideas are possible, then, really, the only thing that we cannot coherently doubt is that at least one thought exists.


Cottingham J, Stoothoff R, and Murdoch D. (tr.) 1984. The Philosophical Writings of Descartes. Vol. II. Cambridge: Cambridge University Press.

Ramachandran VS. 2004. A Brief Tour of Human Consciousness. New York: Pi Press.