René Descartes famously said in his Second Meditation on First Philosophy “Cogito, ergo sum” which translates to “I think, therefore
I am” (Descartes 1641). For Descartes, this was the first step to attaining certainty within knowledge, by which he means knowledge that is immune from a method of radical skepticism proposed by Descartes, later called cartesian skepticism. In addition to arguably being Descartes’ most famous dictum, it is also his most misunderstood, even by Descartes himself. This dictum is often put forward as an argument that answers the question, “Does my mind exist?” This is a pervasive philosophical folly as this question, I will argue, cannot, in principle, be answered in argument form and thus inferential methodology is insufficient to account for our knowledge.
In his later work Principles of Philosophy, Descartes says his cogito is the conclusion of a syllogism so his intent is to draw an inferential method to know his own existence (Descartes 1644). This is problematic as Bertrand Russell points out that this phrase, if intended to be an argument, would be circular regardless of what form we attempt to put it in. All arguments' premises assert the existence of its content. If the 'I' is in the premises, its existence is assumed. If an argument is circular, it hasn’t provided a reason to believe its conclusion, it merely assumes it.
So what alternative option may Descartes offer? In his response to his critics in Meditations, Descartes argues he is discussing an immediate intuition, an apprehension of a fact absent inferential relations and thus the cogito is in fact not the conclusion of a syllogism (Descartes 1642). Essentially, he means he is not combining one piece of information with another piece of information and then drawing the conclusion “I exist.” Instead, he is establishing the limits of methodological doubt, the breaking point at which his method of skeptical scenarios is no longer viable, by pointing to a piece of knowledge that is immediately apparent and not propositionally componential.
This might lead to a totally different inferential relation, as some have argued, the reason for believing in the existence of your mind is in virtue of its immunity to cartesian scenarios. That is “if the belief x is immune to radical skeptical scenarios then the believer is prima facie justified in believing x.” This has also been called into question in recent discourse and some interesting cartesian scenarios are offered as a counter to this position. For the sake of brevity I won’t discuss those here but for an overview of the most potent one the Youtube channel carnaedes.org offers a great summary you can find here: (Doubting "I Think").
The issue with answering the initial question, “does my mind exist,” is then at least three fold. First, the knowledge of our own mind’s existence is either intuitive and not attained via inference or it is not intuitive and thus attained via some degree of inference. Second, which I argue is the explanation and resolution for the first issue, is that the problem arises from a category error where we assume we can argue for the existence of something so immediate. The resolution that is the second issue leads to our third issue, some consider the notion of non-inferential knowledge problematic as a solution because this is not reason giving for a belief.
According to Plato, knowledge is best defined as “justified, true, belief" where the belief has some set of reasons (a justification) that count in favor of or are reason giving for the belief (Plato 369 BC). The contention then is that non-inferential methods fail to satisfy the conditions of justification and so do not seem to count as knowledge.
My goal here is not to contend with the third point’s objections but only point to it being the only tenable solution to the second issue. The second issue, the category error, is - in my view - the reason the first issue arises to begin with. A category error is an error in predication where some quality is applied to an object that cannot, in principle, contain that quality. Examples are things like “the book sleeps fast.” The book is the object and the predicates being misapplied are sleep and fastness. Books do not sleep and sleep does not have a speed. Argumentation is a linguistic method for combining objects and predicates in a way that produces inferential conclusions. The issue is the phrase “I exist” or “my mind exists” lacks a predicate. Let’s explore why. Example: P1) Socrates is a man P2) All men are mortal C) Socrates is a mortal
Notice the argument’s conclusion follows a result of the predicates (all men, and mortal) being combined in such a way with the object (Socrates) so that there is some shared relation producing the conclusion. Semantically the phrase “Socrates is a man” seems similar to “Socrates exists” where manness and existence are the predicates to the object (Socrates). However, as noted by Immanuel Kant in The Critique of Pure Reason, “being is evidently not a real predicate” (Kant 1781). A predicate is something that in some way changes the conception of an object in our perception. For instance, imagine a purple unicorn. Now imagine a white unicorn. The two colors of the unicorn have augmented your conception of the unicorn thus white and purple are predicates. Now attempt the same with existence. Imagine a purple unicorn. Now imagine a purple unicorn that exists. Your conception shouldn’t have been augmented.
Argumentation is only useful in a system where predicates are augmenting objects - that is that objects and predicates must have relations for there to be an inference. However, absent a predicate, we are only left with an object, the “I.” So what is the category error I am referencing if there is no predicate and a category error is an error in predication? Because existence is not a predicate and argumentation requires predicative relations with objects, the proposition “my mind exists” cannot be argued for. Otherwise stated, the proposition “The proposition 'my mind exists' can be argued for” contains a category error where argumentation cannot be applied to existence because existence is not predicative. This entails that the inferential methods of knowing we call argumentation (the predicate) is insufficient by its nature to argue for the object (my mind). This leads to an argument founded on the dichotomies this line of reasoning produces.
P1) Either I know I exist or I do not know I exist. (Dichotomy 1)
P2) If I do not know I exist then I do not have any knowledge
P3) I do have knowledge
P4) I do know I exist (modus tollens from 2 to 3)
P5) If I know I exist then the knowledge that I exist is either inferential or non-inferential. (Dichotomy 2)
P6) If the knowledge that I exist is possibly inferential then category errors are possible
P7) Category errors are not possible
P8) The knowledge that I exist is not possibly inferential (modus tollens from 6 to 7)
P9) The knowledge that I exist is non-inferential (from 5 to 8)
C) Some knowledge is non-inferential (from 1 to 9)
The inferential method (i.e. argumentation) results in a category error and thus the premise “If I know I exist then some knowledge is non-inferential” is the only option to retain knowledge. The radical skeptic will likely attack premise 3 and if their attack is successful its entailments are philosophically catastrophic. This objection will be covered in an upcoming article “Does Radical Skepticism Pose a Serious Threat to Knowledge” Conclusions:
In short, my answer to the question is yes, my mind does exist and yes I do know it. Both options (non-inferential knowledge or epistemic nihilism) seem unjustified and therefore untenable at face value. However, I believe the tenability of the two is not symmetric. If I do not know I exist then I do not know anything at all so then how is any of the argument I have provided above been reason giving for any belief at all? How would any counter argument be reason giving to believe the opposing position? This seems self defeating. If my argument against inferential methodology being adequate to attain knowledge of my mind’s existence is not reason-giving, then why would any of this persuade me or you to believe that I do not know I exist? If my argumentation here is successful, this is the first step to epistemic intuitionism which would most ambitiously call into question our classical conceptions of knowledge as justified, true, belief - or at minimum alter the requirements of what satisfies a justification condition for a belief beyond strictly inferential methods. I will continue this line of reasoning in my next few posts “Does Radical Skepticism Pose a Serious Threat to Knowledge” and “Doubting The Principle of Sufficient Reason: Does everything require a reason for rational inquiry to be valid.” I will also discuss this article alongside notions of justifications alternative to the strictly inferential methodology (namely Justificatory pluralism) required of argumentation on the upcoming CallIn Show on the Pangburn Channel. I hope to see you there.
Citations and Resources: https://www.britannica.com/topic/cogito-ergo-sum https://plato.stanford.edu/entries/knowledge-analysis/ Russell, Bertrand, 1945. A History of Western Philosophy, New York: Simon and Schuster.