# Is my argument against Descartes's “I think, therefore I am”, logically sound?

Disclaimer: I have answered each and every answer here on the comments where I think they are wrong. So far, I have not been able to find my mistake or anyone clearly admitting Descartes's.

I am not disputing that doubt is thought or not. In fact, I would agree that doubt is thought under another part of Philosophy, but here I am arguing under the ambit of Descartes's LOGIC. Let's take a deeper look into the ORDER of the arguments AND the assumptions involved.

Descartes first says that "I can doubt everything". What is established here, before we can make this statement? A doubt exists, a thought exists to doubt everything, and everything(Universe) exists, which contains both thought and doubt. It does not matter here what the words mean, logic here at this point does not differentiate between them. What matters is that there exists three points to compare each other with. You can say one equals another, but not at this stage. This is before logic has been applied. Think of it as starting tools you got. It only matters that you knew that these existed, you need not even define them.

Does he mean here that doubt is thought? NO, he establishes that later, not at this point. Let me explain why.

Let's change the order of arguments for a moment.

Compare this with.
Argument 1 ( We need to establish that there is thought, doubt and everything to go ahead)
I can doubt everything. ( Rule 1)
Doubt is thought ( Rule 2)
But, I cannot doubt my thought, therefore there is definitely thought. ( Logic for argument 2)

What's the piece of logic here? " But, I cannot doubt my thought".
It is established under prior two rules.

In this the logic has a paradoxical rule. Why? Because Rule 1 says I can doubt everything. That everything is a superset which includes observation or "doubting that doubt is thought", because doubt is thought comes from observation. So under Rule 1 which is established FIRST, Rule 2 is paradoxical, and the logic which is established now has a flaw.

Argument 2 ( We need to establish that there is thought, doubt and everything to go ahead)
Doubt is thought. (Rule 1)
I can doubt everything. (Rule 2)
But, I cannot doubt my thought, therefore there is definitely thought. (Logic for argument 1)
Here there is again a paradoxical set of rules. Rule 1 clashes with Rule 2. Why? Because we first said that Doubt is thought is definite, then we said we can doubt everything which was a superset including all the observations we can make. So this is not absolute as well.

Argument 3:( We need to establish that there is thought, doubt and everything to go ahead)
Doubt is thought. (Rule 1)
I can doubt everything, but my observation or that "Doubt is thought" (Rule 2)
Therefore there is definitely thought. (NO Logic for argument 1)
No paradoxical set of rules here, but this is true by definition. There is NO logic involved at all. Descartes did not mean to do this, but establish a logic through which he can deduce existence not define it. This is absolutely true, but redundant. Everything, doubt and thought needed to be established BEFORE the argument began. Did it mean here that doubt was thought or doubt was not thought? NO. It does not matter BEFORE the argument. The argument begins with an assumption or rule. Before that there are simply three quantities or things we know we are comparing each other with.

Argument 4:( We need to establish that there is thought, doubt and everything to go ahead)
I can doubt everything(Rule 1)
Doubt may or may not be thought ( No Rule here since this is a generic statement which exhausts the Universe of possibilities)

That's it. Only 1 Rule here or only 1 assumption here. So we keep doubting everything till we come to doubt and thought. And say that doubt may or may not be thought. What can we establish from this? " I think, therefore I must be". Why must? Because it reflects that small amount of doubt leftover, indicating that under Rule 1, I can still doubt my thought, but mostly there is no doubt left, so I must be. It's because any other assumption would be paradoxical. This statement is "absolutely true", under 1 assumption, because there are no paradoxical set of statements here.

These are all the permutations and combinations possible of logic(There is one more trivial one, but let's not waste time on the obvious) and the set of rules here. There are none left. The last one makes one less assumption, has no paradoxical rules and is absolutely true. First two have paradoxical rules, therefore are not absolutely true(under established rules). Third one is redundant.

Written word takes so long to communicate. I will throw another bounty if no one still gets it. Current answers are mostly wrong or not getting the point. I am not arguing over semantics, but over his logic. The logic has a flaw I think. Mine is argument 4. Descartes's is Argument 1.

Do I say in my argument if doubt is not thought? No. It might very well be. It is, under everything we know. But for us to say this " I think, therefore I AM", we need to go under argument number 3, which is redundant.

I think there is a flaw, which has simply gone unnoticed, because people think " It is too obvious that doubt is thought". The flaw is in the logic which has been applied. Nothing is obvious.

I apologize if my words seem a little harsh, but this has gone on unnoticed and misunderstood for far too long.

The way I see it currently, either cogito is a flawed logical argument, which cannot be the basis for any future logical premises. Or it is simply true by definition. Little disappointed as well. I thought in Philosophy we questioned everything. Why does it matter who said it.

Descartes wants to establish something. The only means given to man in order to establish something to be true is logic. First thing we check is if the logic is absolutely correct or not.

• I have just had a minor eye surgery, so kindly bear with me for the moment, if I do not respond fast enough. – user12196 Aug 31 '17 at 11:11
• The "argument" is not a logical one, meaning that it can be assessed "by logic alone" (no logical argument can prove the existence of something...). It is a sort of "intuition"; the key point is not about "reliability" of the content of the act of doubting but about the impossibility (?) of conceiving a "mental act" whatever without simultaneously conceiving the subjcet (the "I") performing that act. – Mauro ALLEGRANZA Aug 31 '17 at 11:36
• If so, I can doubt about my thinking to a doubt about ..., but still there is an "I" performing this act. This, IMO, is in a nutshell D's cogito. – Mauro ALLEGRANZA Aug 31 '17 at 12:04
• As you know, Cogito Ergo Sum "raises numerous philosophical questions and has generated an enormous literature." "As Descartes writes: When someone says “I am thinking, therefore I am, or I exist,” he does not deduce existence from thought by means of a syllogism, but recognizes it as something self-evident by a simple intuition of the mind. (Replies 2, AT 7:140)" – Mauro ALLEGRANZA Aug 31 '17 at 12:11
• In what way is a doubt not a species of a thought. The original paraphrases Augustine's 'Dubito ergo sum'. So the distinction between think and doubt is not part of the original argument, it is an attempt by Descartes to clarify, which may fail. If I doubt that I doubt, it is still me doubting. – user9166 Aug 31 '17 at 19:07

Disclaimer, some of this post may not make sense to you, as the OP has rewritten his argument numerous times, and I am not deleting any of this so, skip to the end for newest most relevant information.

Descartes has made a mistake in logic which has not been caught for the past 350 years.

No, he hasn't. You are falling into a fallacy of false premise, the error being believing further doubt invalidates the logic of Descartes's argument. Lets quickly analyze cogito Ergo Sum. Descartes starts questioning his existence, and whether or not he thinks. However the fact that he is questioning necessitates his thought and existence as someone has to be asking the question. (The thought cannot exist without the thinker thinking.) There for since Descartes is thinking he must exist.

Now what you did, you add another doubt (question) to this argument. The issue is that does not invalidate the logic of the initial argument. No matter how much you doubt this it remains logical.

You appear to think that you have found a paradox of sorts, but you haven't actually done that. The thing about a paradox is that it is an argument that can be neither true or false. For example the statement "This statement is false." is illogical because if the statement is true it must by false, and if it is false that would make it true so it can repeat indefinitely.

However with your modification cogito ergo sum is not rendered false. At every step it is rendered true. You doubt (A thought) and there for must be real and thinking, or you could not have had that doubt (or thought). If you again doubt you there for must be real and thinking, or you could not have had that doubt. This is not a contradiction it is just an infinite repetition of the proof.

It appears this has still not gotten my point across clearly so I will now analyze this argument from the current question.

Let A be the object: Doubt Let B be the object: Thought

Descartes's Idea: I can apply A to all objects except B, because even if I am able to apply it to B, A is also B, and hence B for sure is, therefore " I am"

My idea: I can write this now: I apply A to B first. Then B might be ( Let's not make the leap from might to is here so quickly, and add a might instead of definitely, because doubting is the act applied to thought, so there is a fine distinction) But this can be re written as: then B might be, given A applied to B. Now I can write: A can be applied to { B might be, given A applied to B}, because it still makes logical sense. I can add A to B before the sentence and B to A before it infinitely. Hence it is not possible to remove doubt from assertion or belief using Descartes's idea.

Hence, a better statement would be " I think, therefore I must be", indulging both doubt and belief.

This appears to be not false equivalence, but instead false non-equivalence. You draw this distinction between doubt and thought, but the doubt is a type of thought.

Let doubt = A

Let thought = B

Let the thinker = C

Now all A is a type of B, and all B requires C. (Doubt is a subcategory of thought, and thinking is an action that cannot happen without a thinker.) Therefor when A is given then B is given and C is given.

You cannot have A without also having B, so attempting to have A without the necessity of B is illogical.

Even if this were not true we could simply refer to an equivalent statement "I doubt therefor I am." With this slight tweak the act of doubt can now act as proof, as I must be in order for me to be able to doubt. (Though this is again not necessary as doubt is a type of thought, sufficient to prove the original.)

I'm going to try to make this clear one more time, and that is it. Quoting from chat.

Every definition is an assumption. Agree or not?

No. Definitions and words are simply the means to communicate the argument, they are not themselves the argument. The logical side works, arguing wording is just semantics. If you don't agree with the words, that does not change the meaning Descartes refers to with them. Its like if I were to call your argument invalid because I don't think you should use the word must. The argument is not about the meaning of words, so that is irrelevant.

The thought happened in his mind, as per his observation.

But Descartes has begun by doubting everything. And I am now saying let us doubt this observation of senses as well.

But this isn't an observation of the senses. This is a thought exercise, that can be completed without the use of sight, sound, or any other sense. This thought exercise cannot be accomplished by something that doesn't exist. (Obviously if something doesn't exist it can't do this.) Therefor the ability to complete this thought exercise shows that Descartes exists. In essence the ability to have ANY thought proves your existence, as you must exist to think. (or doubt.)

In argument one and two you make an error. The argument is not paradoxical because "I can doubt everything" is simply where he starts, not a universal rule that is supposed to govern everything in the universe. He can doubt anything until he has a logical reason not to. However where paradoxes actually do come in is when you consider doubting doubt. You can't doubt doubt unless you can doubt, so your arguments about doubting doubt are paradoxical if anything is. This means there is no logical reason to doubt your ability to doubt.

• @Novice how is it an infinite regression? The fact that he can have a single thought proves his existence in some form. Doubting this further does not invalidate it. He can have further doubt about the nature of his existence, but he has proven that he exists in some form, as in order to ask the question, "do I exist" he must exist, or there would be no one to ask the question in the first place. – Braydon Aug 31 '17 at 14:24
• @novice But you have no logical basis for establishing doubt. Just because you claim to doubt logic does not invalidate it. – Braydon Aug 31 '17 at 14:26
• @Novice Not logically. You seem to be mistaking emotional uncertainty with having logical reason to doubt. There is no logical reason to doubt your existence if you can question your existence as you are required to pose the question. There is no logical reason to question this again, as it is redundant. The thing is your loop does not disprove anything even if you do ask another question. Repeating the question again will again lead to the same answer that you must again exist in order to ask the question. This so called regression only proves Descartes infinite times. – Braydon Aug 31 '17 at 14:51
• @novice it is a proof of both existence and thought. Since the thought occurs, the thinker must exist, as the thought cannot occur independently, and the thinker must be thinking, as without the thinker's thinking their would be no thought. I've edited my post with more information to hopefully explain why you have not successfully challenged cogito ergo sum. – Braydon Aug 31 '17 at 23:55
• Do you not understand anything I say? Try reading it again before criticizing. I never actually related it to physical phenomenon I related it to the laws of nature if anything, and again, missing the point. The point is that this rule applies only when you do not have a logical reason to ignored it. It was never claimed to be a universal rule that applies to all logic, it was merely the starting point where you do not assume. – Braydon Sep 10 '17 at 4:13

Can we doubt that “doubt is a thought”? Yes, we can. But let's see what it does for cogito.

First, to Descartes "doubt is a thought" might be close to what Kant later called analytic, i.e. it simply reflects the meanings of "doubt" and "thought". The idea that doubt is more than thought (or ought to be to count) appears much later (in Peirce and other anti-Cartesians). Presumably, Descartes's doubting was for substantive issues, not verbiage. But more importantly, in the crucial passage we can replace every use of "think" by "doubt" and still get the intended meaning:

But immediately upon this I observed that, whilst I thus wished to doubt all, it was absolutely necessary that I, who thus doubted, should be something; And as I observed that this truth, I doubt, therefore I am, was so certain and of such evidence that no ground of doubt, however extravagant, could be alleged by the Skeptics capable of shaking it.

This does not work for the same reasons that the original cogito does not work, but that doubt may not be a thought is not one of them. One of commonly pointed out reasons is the inserting of the "I". Descartes starts with doubting, finds an obstacle, and concludes... "I, who thus doubted, should be something". Well, either the "I" was there from the beginning, in addition to doubting, and the doubting did not do its job, or it wasn't, and he is "inferring" the "I" as "something" out of the doubting alone, and that is a big leap.

Second, "can" is ambiguous. Descartes does not assume that he can (as in, is able to) doubt everything upon consideration, only that he can (as in, allows himself to) doubt everything at the outset. In the end, he finds himself unable to doubt cogito, "no ground of doubt... is capable of shaking it". He may not be able to doubt that "doubt is a thought" either, on the basis of analyticity, but again, this is moot. What he finally says is not true by definition (i.e. in virtue of meanings). He allowed himself to doubt everything, he then found out that there was something he was unable to doubt, namely his doubt. That that would happen was not clear from the outset in virtue of meanings alone, it needed to happen. This is why in defending cogito against criticisms Descartes disavowed it as an inference, and described it as a non-inferential surmise, where "I think" (replaceable with "I doubt") simply serves as a reminder of the experience that motivates "I am", not as a premise of an inference:

"When someone says 'I am thinking, therefore I am, or I exist' he does not deduce existence from thought by means of a syllogism, but recognizes it as something self-evident by a simple intuition of the mind."

We might call this a "fact of reason" (as Kant called the moral law), or like Peirce, "compulsion of thought". If cogito is taken as an inference then it does make a mistake of presuming its conclusion, and much more besides: the "I", the "think", the "am", and a good chunk of conceptual language required to understand what those mean, including truth and inference. If all of that is made into a background then cogito can be made into a valid inference (but that defeats its purpose). Indeed, if we happen to have a database about individual X containing "X thinks" but not "X is", due to oversight, we are justified to infer the latter from the former, and with more background assumptions even that "X is human".

The failing behind the cogito is common to all attempts to derive something out of nothing. They overlook that when this is taken at face value the lack of conceptual background in nothing turns everything into gibberish. Conversely, it is always possible to infer background assumptions from non-gibberish (at least under some allowance for presuppositional inference, as in Kant's transcendental arguments), but that is pointless if the point is not to presuppose them. Here is Peirce:

"Descartes thought this "très-clair"; but it is a fundamental mistake to suppose that an idea which stands isolated can be otherwise than perfectly blind. He professes to doubt the testimony of his memory; and in that case all that is left is a vague indescribable idea. There is no warrant for putting it into the first person singular. "I think" begs the question. "There is an idea: therefore, I am," it may be contended represents a compulsion of thought; but it is not a rational compulsion. There is nothing clear in it. Here is a man who utterly disbelieves and almost denies the dicta of memory. He notices an idea, and then he thinks he exists. The ego of which he thinks is nothing but a holder together of ideas. But if memory lies there may be only one idea. If that one idea suggests a holder-together of ideas, how it can do so is a mystery. [CP 4.71]

• Hi, you still have it slightly wrong. I am not saying that doubt is not thought or doubt is thought. I am saying if you say either statement then you are assuming something. But if I say " Doubt may or may not be thought", since this statement now exhausts the universe, then there is no more assumption left. That doubt is a thought comes from observing thought. I am simply saying that using Descartes's method I am now allowed to doubt my observation. Hopefully things are more clear and you edit your answer to reflect this as well! – user12196 Sep 8 '17 at 17:46
• I only meant to point out one paradoxical assumption in Descartes's argument. One first assumption or rule is "I can doubt everything", the second rule is " I cannot doubt my observation", or doubt that " doubt is thought", both statements cannot be simultaneously absolutely true. How do you catch a paradox? A statement and it's converse if both true, constitute a paradox: Example: Liar's paradox. It is the same here. Hence Descartes has failed to establish an existence for certain. I am adding the words "must be", to reflect that small doubt which is left over, and removing one assumption. – user12196 Sep 8 '17 at 17:49
• And this is not relying on semantics at all!, but an argument from informal logic challenging the basic assumptions in Descartes's argument. I hope things are more clear now, but please let me know if any clarifications are needed. – user12196 Sep 8 '17 at 17:50
• I am not saying if doubt is thought or not! Why should I need say either statements? It is Descartes who says doubt is thought. He says that this is for certain. But how does he arrive at it? My observing his thought. But before all of this he has said that he can doubt everything. So let's doubt his observation as well. Now after doing this, he cannot establish existence for certain, because his first assumption does not allow the second assumption which he has made, because that reasoning can only be applied by NOT doubting his observation. – user12196 Sep 8 '17 at 17:56
• Just wrote my edit 2. You are getting it slightly wrong. – user12196 Sep 8 '17 at 19:20

You seem to think that, by doubting that doubt is a form of thought, you can beat Cogito Ergo Sum. This is incorrect, as you're not applying logic to beat Descarte's assertion, but you're relying on semantics more than anything else.

We can rewrite Descarte's conclusion like this:

Something 'I' is doing something doubting or thinking, therefore something 'I' exists

(for something cannot do something without something existing)

Whether you call 'doubt' a form of thought or not, is wholly irrelevant to the conclusion that something exists, and Descartes chooses to call that something 'I'.

Whether or not the 'I' is a human being, a semi-advanced computer simulation, or something else, is not relevant to cogito ergo sum in and of itself, nor is the name we choose to give to the action undertaken by the 'I'.

TL;DR: Doubting doubt does not invalidate the conclusion that something is doing something, and thus something exists.

• Changed my question to make it simpler. You have it wrong. – user12196 Sep 8 '17 at 23:45

That's an intelligent question. Whether the argument is sound or not depends on how you read it. I my view, Descartes's argument even though maybe imperfectly articulated is a useful mental exercise if only for yielding a better understanding of our mind and our existence. I view the Cogito to be just an attempt at logically establishing what is evident to us through intuition but the argument doesn't at least explicitly address many questions that may emerge in subseqeunce which are however not really to its detriment if we note that no intuitive knowledge can be expressed in a logically sound expression maybe because human intuition doesn't work discretely as does logical thinking.

Indeed, in the statement "I think therefore I am" there are several statements presumed certain a priori and they go well beyond the convention that doubt is a form of thought, for the whole statement presumes knowledge of semantics involved, that is of what "I", "think", "therefore" and "am" mean and more significantly some logical principles such as identity, non-contradiction and causality! Descartes might have had a point if he said that our intuitive, non-discursive, non-deduced self-knowledge doesn't depend on recognition of prior principles of logic but the Cogito is meant as an argument not a pointing to our intuition.

Nonetheless the Kartesian doubt can be applied to each of the presumed semantics and prove right: I may doubt what all these concepts mean including "doubt" and "think", yet again I can't doubt that I'm doubting them!

The point of this observation then being that regardless of how logically you argue, there are already a lot of things presumed with certainty such as a set of definitions, some basic logical and philosophical principles (e.g. identity, non-contradiction, causality), and that in our most radical acts of doubt, we are never detached from them. In fact it is because of them that we are able to think and doubt in the first place. Therefore, even though Descartes in his notion of methodic doubt claims that he applies radical doubt to any dubitable thought, he is applying his doubt on a foundation of very certain but implicit principles, and it is these certain principles that enable him to move beyond doubt in the first place. Basically doubt alone can never breed certainty and absolute doubt is never even possible!

Furthermore, I find it noteworthy that, among all the prior premises and definitions presumed by our mind, existence can be argued to be the highermost assumption in each act of thinking. In the Cogito argument the existence of I and each of the concepts are presumed because even though I can doubt for example that the external world exists, but I can't doubt that the concept of "external world" exists in my mind as well as all concepts in the Cogito statement, and since all of these are subordinate to my mind I can then deduce my own existence from those perceptions. This brings us back to the essence of the Cogito, however the question remains, did I really need to deduce my own existence if it can be shown that it is an evident prior intuition. I doubt if Descartes disagreed as he seems to have been primarily concerned with refuting the radical dialectical skeptics who went out of their way to even deny the existence of self, rather than implying that intuitive recognition of self really required any argument. Therefore given the weakness of prior assumptions, the Cogito fails if is considered a logical argument based on sound premises. But nevertheless it would be a useful experiment if presented as only an intellectual pinch on radical skeptics to have them admit their own existence by starting from their own premise that absolute doubt is possible.

We maybe then recognize the genius of Muslim philosophers such as the 12th century philosopher, Avicenna, who had already cited the essence of Cogito argument (centuries before Descartes) only to dismiss it as invalid based on the claim that we can never experience our thoughts separate from our existence, hence in all acts of thinking the existence of self is presumed. For Avicenna therefore existence of self was self-evident and needless of demonstration and any attempt at demonstration would be imperfect (imperfections of the Cogito being a testimony). This is also in keeping with the Muslim philosopher's concept of "knowledge by presence", their term for unmediated intuitive knowledge that is distinct from and the ground of all discursive knowledge (that is thoughts).

The obvious but often mysteriously missed reason for evidence of self-existence have to be the fact that self is ontologicaly prior to thoughts as thoughts can never exist without self existing first hence no thought can be experienced prior to it. This may render the cogito argument as an argument from effect to cause, whereas the cause is already evident, even though this self-evidence is usually and mysteriously missed by the average man. Much later, the ontological precedence and yet co-existence of existence with all thoughts became the focus of Martin Heidegger. He articulated that no knowledge is prior to the sense of existence (or being) and even yet, no sense of being itself is equatable to Being (with capital B) per se as Being itself always stands above all categories. This is all too consistent with the idea of Muslim philosophers including Avicenna that self as a being is not thoughts (whereas Descartes believed that self is a substance whose whole nature consist in thoughts). Until Mulla Sadra a 17th century Muslim philosopher who brought about an entire revolution to peripatetic philosophy by arguing from logical and ontological precedence of Being as well as its indefinition and irreducibility that only being captures the true essence of God as God and Being seem to be identical in these properties! They are both omnipresent yet ineffable, undefinable and inescapable!

Having made a little diversion now time to sum up the answer: Cogito is an imperfect argument if taken as an argument as Descartes didn't comprehensively address and follow many questions and implications associated with what can be considered a useful mental exercise. The greatest fruit of the exercise I believe is that it shows that all roads lead to (and at the same time come from) being!

• So you agree that Descartes argument is flawed? And my criticism of it is valid? (Just making things simpler here). Are there any of my points that you disagree with as well? Thanks for the answer! I will read it a few times again, just that I am recovering from an eye surgery right now. I disagree with what you sum up though. All roads might lead to being, from the point that Descartes starts. He cannot remove all doubt, by the act of doubting everything, when he starts that as the initial point of his argument. – user12196 Sep 2 '17 at 12:50
• You are right that "I cannot doubt that I am doubting them", but I can still doubt if doubt is thought, still reducing Descartes's argument to null and void when it comes to establishing existence of an "I". – user12196 Sep 2 '17 at 12:54
• @infatuated. Great answer. "This may render the cogito argument as an argument from effect to cause," - Yes! His 'I am' was enough and 'cogito ergo' is redundant. But Western philosophers rarely see past their thoughts to examine the 'I am' on which they depend. Kant, meanwhile, saw that the intellect depends on something prior. The poet Paul Valery writes "Sometimes I think, sometimes I am". This may be a much more revealing formulation. . – user20253 Sep 2 '17 at 13:11
• @infatuated That is exactly what I am disputing. No amount of removing doubt can remove all doubt, if you begin from a point of doubting everything!, and therefore cannot establish anything for certain. – user12196 Sep 2 '17 at 13:37
• And you do get credit for recognizing the flaw in that assumption and the weakness in the argument. That's why I commended you in opening of my answer. – infatuated Sep 2 '17 at 13:56

You take as Descartes' "first assumption" the idea that one can doubt everything - but I would prefer to say that the cogito ergo sum is simply the first principle he arrives at in his process of steady inquiry, as I believe this more carefully captures the rationale for Descartes' process and his representation of that process. I would not see Descartes' formulation of his argument as a strict representation of a process of logic, but rather as an act of persuasion - similar to a process of logic, in that he wants us to agree with the logical intuitiveness of his steps in that process of steady inquiry. But I think that Descartes would regard his own process as inadequate, which evidently he did not, if he saw himself as taking as his first principle/assumption the idea that he could doubt everything. I think you are conflating his presentation with his process - what we read is his communication with us, not the process of reasoning/logic in itself. In fact, The process Descartes is hoping that we follow and agree with his intuitions about, is supposed to occur "prior" to any application of logic or science, as the cogito ergo sum is supposed to operate as the first principle upon which any subsequent exercise of logic can assuredly stand, without further questioning, provided that we agree intuitively with Descartes' process of establishing that first principle, as he presents it.

• I think I have just applied a logic, prior to which Descartes's logic can stand upon. Please read my edited question. – user12196 Sep 8 '17 at 23:48
• I do not agree with his first principle at all. His logic has paradoxical assumptions. – user12196 Sep 8 '17 at 23:49

'Cogito ergo sum', 'I am thinking, therefore I am' or 'I think therefore I must be' is an existence conditioned on thought. Once thought stops, you don't exist. Other than demonstrating that experience is dependent, conditional, subject to a frame of reference, the statement says no thing interesting.

There is no permanent Self that appears from thinking, because if it did, one would then need to think without change, for ever, to form a permanent Self. In fact - what you? Can I ask your 5 year old self of Descartes' conundrum? Does your retired self have the same opinion as you now?

If one chooses to not rely on observation because of a speculated deceiver, one must give reasonable grounds for supporting such a deceiver. Drop a ball, any ball, a million times from a certain height. Measure the time it takes to land as accurately as it needs. If the hypothesis 'there is no deceiver' is not rejected, good good. No deceiver has ever been found within experience using the scientific method.

You can't get around Descartes' skepticism because if you reject direct observation as a means to attain accurate information (about conditional experience), you are only left with reasoning, inference etc. Inference is only a valid mode of gaining information subject to accurate observations of experience.

If I chose to never observe apples falling down onto the earth (or were too skeptical to care), I could state - without a sound basis (don't ask the path, it's a-scientific) - that apples in fact fall upwards, and given this information, in 50 years time Earth will be Apple free. The inference is perfectly reasonable, it's the initial observation (or lack thereof) that is at fault.

Direct observation offers a clue - all observed things arise dependent on conditions (mother and father for a human), subsist dependent on conditions (food), and cease dependent on conditions (old age). No thing, even a proton or a black hole has been deemed to last for ever. All things are observed to be impermanent.

As such, any notion of a permanent 'thing' or Self - an object that exists, with defined characteristics, independent of observation ('I am thinking' is an observation) - is entirely alien to what is seen, heard and sensed.

Now Descartes went wrong because positing a permanent deceiver goes against the observational evidence of impermanence. You have less reason to doubt observation in a world showing and acting impermanently and empty of Self, because the deceiver, a 'thing' posited outside of observable experience - a being hypothesized as permanent, a consistent net force in some direction across All (whether making left seem as right or peacefulness seem as violence) - is definitively unobservable in a relational world (the act of observation is by itself a condition of observed properties). Such a deceiver offers more ground for doubt than does relying on direct observation.

(If the deceiver is picky and does not affect All unconditionally, then his choices are conditioned, and so not substantially different (not a true deceiver) from the impermanence and non-Self (anatta) that observation of experience offers)

(https://aeon.co/essays/the-logic-of-buddhist-philosophy-goes-beyond-simple-truth for a more interesting take on the ineffable!)