1. Are there any other things we can be certain of?
  2. Are there any 'hooks' that can be hooked into 'I think therefore I am', that one could be 100% certain are truthful as 'I think therefore I am' itself?
  • 3
  • 2
    I challenge that the quote 'makes certain' of anything: a. truth is subjective; b. the quote's [flawed] logic.
    – user4500
    Commented Jun 23, 2014 at 20:10
  • @user4500, the quote 'makes certain' of existence. It means (inner) experience entails existence.
    – nir
    Commented Aug 3, 2014 at 21:36
  • The important point about 'cogito' is that it is self-knowledge or 'knowledge by identity'. This is the only form of certain knowledge, it being unmediated. As for what we can learn from it reports vary. Some say very little, some say everything. Note that Descartes can be read as making a statement not an argument, for he could just as well have said 'I am'. That he can state it proves it.
    – user20253
    Commented Jul 25, 2018 at 11:34

8 Answers 8


We can be pretty certain of tautologies, which are axioms of logic based on the principle of non-contradiction. After all, 'cogito ergo sum' already presupposes that something must exist in order to think, and presupposes a notion of 'ergo', both of which are (to Descartes at least) at least as certain as the full phrase itself.

Necessary truths, which are defined either as being true in all possible worlds, or by their falsification leading to contradictions (both of those definitions are essentially the same) are also certain truths, because no matter how we may imagine a universe to look, it cannot sustain contradictions. Therefore, the classic example 'all bachelors are unmarried', is a necessary truth, because the definition of a bachelor is someone unmarried. For this statement to be false, a man would have to be both married and unmarried - an impossibility. Hence, this statement cannot be false and therefore must be true. (I think. See here and here)

In addition, like logical proofs, the results of mathematical proofs are similarly trustworthy, so to speak. If a rectangle is defined as a four sided polygon whose sides interact at right angles, and we prove that a polygon has these properties, it must necessarily be a square. The same idea (I think) applies to numbers.

Some may argue these points, but some argue on Descartes' argument as well. It seems like these are some things that we can be at least as sure of as we may be of 'cogito ergo sum'.

  • But there can be no bachelors, nor geometry, nor math. You know these things through your experiences that you got trough your senses and all your life as you remember could have been a awesome hallucination on uber-LSD, and logic that you believed in is by-product of your malfunctioning mind and all shapes are only round squares but you saw and felt them as something different than they actually were. Commented Jul 4, 2014 at 8:29
  • 1
    @LIUFA true, but then you don't know that you exist either
    – That Guy
    Commented Jul 4, 2014 at 13:34
  • There has to be something that raises the question. Commented Jul 4, 2014 at 13:42
  • 1
    @are you sure? Maybe that's just logic based on a twisted (malfunctioning, as you put it) mind? Now you'll tell me, 'ha, see, you admitted to it being from the mind, so it must exist', but that's really because am limited by the same construct, whatever it may be, and have no other way to express this notion
    – That Guy
    Commented Jul 4, 2014 at 13:49
  • You are right, I am forced to base think -> exist being based my experience (logic). This is a very good point indeed. Commented Jul 4, 2014 at 13:58

I'd say 'I think, therefore I am' isn't something we are 100% certain of. Or rather, what it entails is nebulous.

The fact that I am a thinking 'being' at this instance says nothing about 1 second ago. I "remember" it, but personal recollection is hardly a solid source of evidence. I could refer to external sources for validation of myself, but there is nothing about being a thinking being that confirms that my understanding of reality actually correlates to what reality is.

'Cogito ergo sum' is just the acceptance that we have to start somewhere if we want to attempt getting an 'objective' understanding of reality. It's an agreed upon baseline, that exists mainly because to not accept it as fact means we can't really know anything.

  • You might not be certain, but that doesn't mean the parent isn't :)
    – yters
    Commented Jun 16, 2014 at 21:21

Law of non-contradiction is part of Cogito ergo sum. Other similar logic principles, where it is self defeating to doubt them. Such as saying "there is no absolute truth" which is itself an absolute truth if true. Finally, you can't prove that uncertainty means lack of knowledge.

  • Why is doubting the law of non-contradiction self defeating? After all it amounts to accepting that for some truth bearer both it and its negation are true. What is self defeating about that?
    – sequitur
    Commented Jun 16, 2014 at 13:29
  • If the parent is certain of cogito ergo sum, the law of non-contradiction is a part of that.
    – yters
    Commented Jun 16, 2014 at 14:06
  • Yeah, but why is it self defeating to deny the LNC?
    – sequitur
    Commented Jun 16, 2014 at 16:59
  • 2
    If it's false then it's true? Does that work?
    – yters
    Commented Jun 16, 2014 at 21:21

Often the things we can be pretty certain of are ontological commitments.

Exploiting ontological commitments

A typical example is a Henkin style completeness proof for first order predicate logic. We are talking about formulas and deductions that we "can" write down, so we can be pretty certain that we can write down things. We use this certainty to construct a syntactical model of the axioms. (The consistency of the axioms enters by the "non-collapse" of the constructed model.)

I don't know whether the earliest (Gödel/Tarski style) completeness proofs also relied on the same sort of ontological commitment. The last section below indicates why it is highly likely that some sort of ontological commitment is required for any completeness proof, as long as no explicit notion of "existence" relative to which we talk about "completeness" is specified.

What ontological commitments are really there?

One point of contention is how much ontological commitment is really there. Just because I can write down some things doesn't mean that I can write down an arbitrarily huge amount of things. Or maybe I can write them down, but I thereby might destroy things I wrote down earlier.

What ontological commitments are really needed?

For a completeness proof, we must show that for each unprovable formula, there exists a structure were the unprovable formula is false and all axioms are true. This structure must "exist" in a suitable sense, because what else could be meant by "completeness"? If only the structures that can be represented in a computer with 4GB memory would be said to "exist", then first order predicate logic would not be "complete" relative to this notion of "existence".

  • Yes but according to Descartes it could be demon writing that stuff for you, and the axioms do not require proof are based on obvious and 'obvious' is based on our previous experience of same thing happening again and again (otherwise it would be 'extraordinary' and could not be used as base to derive). Commented Jul 4, 2014 at 9:17
  • @LIUFA I'm not sure whether you understood the intent of the original answer. We can't prove that the Peano axioms are consistent, but we can prove that first order predicate logic is sound and complete. Isn't that surprising? How can that be? The original answer stated that one way to prove completeness is to exploit implicit ontological commitments. I think I understood now why such ontological commitments are really required, and added the corresponding explanation to the answer. Maybe that's not the type of answer you were looking for, but I'm quite pleased now with what I understood. Commented Jul 5, 2014 at 1:28

I'd say that we can't have certain knowledge regarding the real world, and that certain knowledge is restricted to logic and mathematics. This includes statements such as "cogito ergo sum", see the answers to "Could `cogito ergo sum' possibly be false?". As soon as we consider statements about reality (rather than some artificial system constructed from rules and axioms) we have to put up with the idea that we can only have uncertain knowledge, with different levels of support/corroboration from the evidence/observations. IMHO the search of certain knowledge regarding reality is misguided.

  • Downvoter, some feedback on the reasons for the downvite would be appreciated. AFAICS what I have written here is fairly standard. Commented Dec 22, 2015 at 12:45

Dubito ergo Cogito, ergo Sum


I doubt, therefore I think, therefore I am.

Ergo ... ergo or therefore ... therefore are the signs and significations of reason. Can one doubt 'doubt' by doubting 'reason'?

Al-Ghazali thought so (Deliverance from Error, page 22):

Then sense-data spoke up: "what assurance have you that your reliance on rational data is not like your reliance on sense-data? Indeed, you used to have confidence in me. Then the reason-judge came along and have me the lie. But were it not for the reason-judge, you would still accept me as true. So there may be, beyond the perception of reason, another judge. And if the latter revealed itself, it would give the lie to the judgments of reason, just as the reason-judge revealed itself and gave the lie to the judgments of sense. The mere fact of the nonappearance of that further perception does not prove the impossibility of its existence."

So perhaps one has to learn to judge and not merely reason.


Al-Ghazālī, A. H. (2006). Al-Ghazālī’s Path to Sufism: his Deliverance from Error [al-Munqidh min al-Dalal], R.J. McCarthy translator.


Another aspect of "ego cogito ergo sum" ignored by the current answers is that it is a statement about the subject itself. But it is more than that, it is also a sort of instruction for anybody to convince himself that he exists, so it is more than just a subjective analytical statement, but also the predicted outcome of an experiment anybody could try to verify for himself.

You might have trouble to imagine anybody who would not be convinced of his own existence after performing this experiment. But assume somebody with a severe psychiatrical condition. For him, maybe the assertion "I am mad" would work better:

mad man: "I think I finally understood. I am mad!"
partner: "But you are crazy, why do you think that?"
mad man: "See, now you also tell me that I am insane!"

But Descartes was a philosopher, not a mad man. So it seems more appropriate to compare his instructions with teachings of other great philosophers like Socrate. He also made statements about himself, like: "As for me, all I know is that I know nothing, for when I don't know what justice is, I'll hardly know whether it is a kind of virtue or not, or whether a person who has it is happy or unhappy." We may assume that Socrate himself knew whether this assertion was true or false. The more interesting question is whether his discussion partners could know whether this statement was true or false. In fact, I believe they could not know this. All they could know was that Socrate was criticizing people who believed that they knew things which felt squarely outside their domain of expertise.


I've heard this one from Matt Dillahunty of the Atheist Experience of Austin-

There is a reality / Something exists

You must log in to answer this question.

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