The "Laws" of Thought

I'm writing in response to this video I happened upon. 
https://www.youtube.com/watch?v=Ulmvy4QczV0

The second speaker says that logic is "a description of existence" and uses the analogy of a map on the one hand and the thing that the map refers to, on the other. This is bad. Part of the problem (perhaps) is that 'logic' has more than one use. There's the ordinary usage as evidenced by statements like, "I don't agree with your logic" in response to an argument. Here, 'logic' seems to refer to something like one's line of reasoning. Of course, this is very different from 'logic' as a philosopher's and logician's term of art. 'Logic' in this latter sense refers simply to formal languages like first-order propositional logic, predicate logic, modal logics, meta-logic and perhaps multi-valued systems. The problem is, in neither of these uses of the term does it make sense to say that logic is a "description of existence." And I really don't get the map analogy. If the second speaker wants to get away from reifying rules of valid inference (qua abstract entities), then this analogy is not helpful because by his own words, the map represents things that exist. So if the rules of a formal system are to reality what a map is to the things it represents, then we would actually expect the rules of a given logic to represent things (and what else could they be, but either concrete objects or abstract entities?).
Moreover, 'the laws of thought' need not be coextensive with the rules given in various logical systems, but this exchange seems to presuppose this very thing. There is much debate in the philosophy of logic (this is the subfield of philosophy that studies logic) about just what formal systems like first-order propositional logic are meant to represent (though nobody I know of says it "describes existence"). We know that logics are formal, artificial languages. We also know that natural languages (e.g., English) suggest to us certain patterns of reasoning---that is to say, there seem to be certain rules beyond mere syntax which we presume are in some fashion, truth-preserving which is to say that we can get from one statement or set of statements that are deemed 'true' and then, in some suitable sense, "get to" another statement that is also true. But this is really complicated terrain. I just think whoever talks about the "laws of thought" should not conflate that with systems of logic. I have similar complaints about the first speaker as well, but it'll needlessly lengthen this post. The point is, neither of them seem to be well-versed in philosophy or logic, but it doesn't stop them from using technical jaron, and the result is confusion.

Which brings me to the Gangadeanian at the beginning of the video who raises the question, "do you accept the laws of thought?" Gangadean and his people talk a lot about the laws of thought. They enumerate them as the law of non contradiction, the law of excluded middle, and the law of identity. Gangadean draws the analogy between the laws of nature and these laws. Insofar as the physical laws of nature cease, life ceases and so it is with the laws of thought--if they cease, thinking ceases, so he claims. Notice this isn't a deductive argument. You've got something resembling an argument from analogy (at best an inductive argument). Lest I be accused of misrepresenting Gangadean and his crew, just keep reading and I'll get to ways that they try to motivate this claim. Anyway, with this analogy, you're suppose to just hear this and somehow "recognize" that it's true and that the analogy is apt. It's a picture he paints for you to elicit a certain kind of response. Again, none of this is proof of anything.

Now importantly, it's not a settled matter what a law of nature is to begin with. So this analogy is not going to prove that helpful anyway. There's considerable and ongoing debate concerning laws of nature. A Humean for instance thinks that laws of nature are purely descriptive regularities of the way that nature tends to behave. Or at the very least the Humean thinks we have no epistemological grounds to say more than this about laws of nature. Humeans question whether laws are anything more than a generalized trend based on a bunch of observations. In a terribly simplified example: a lot of things drop to the ground and so we reason that there is this tendency for things to fall to the center of the earth and call it the law of gravity. But there isn't something like a law over and above the regularity denoted. It's certainly not like you can bump into a law of nature. So they don't seem to be concrete entities by anyone's lights. Of course laws of nature could be abstract entities, but it's hard to say much more if we go down that route. The very nature of abstracta is really murky and mysterious. Abstract entities are supposed to be things that exist without being "in" spacetime. They are acausal. They don't interact with anything else that exists. But they are thought to exist nonetheless or be real in some mysterious sense but are usually posited for purely theoretical reasons (for instance, the existence of numbers is one straightforward way to account for how it is that mathematical formulas can be true). And there are other views. Some tend to reify laws of nature as if they are robust, exceptionless rules that all physical stuff "follows." But at best we use words like 'law' and 'follow' and 'rule' loosely. We have some idea of what it takes to be a man made rule like the rules of grammar, or the legislative laws or even social mores---they are prescriptions and proscriptions that cover a particular domain of behavior. But it's not like these laws actually control what people do. Instead they tell us what people should do given certain interests after all, people violate the norms of grammar all the time (and still manage to communicate). But presumably laws of nature are supposed to exert more "control" on their objects. But it's hard to say precisely what this amounts to.

Anyway, the point is not to get into a substantive debate about the nature of laws in general. It's a fascinating issue, but highly complex. The point I want to drive home is that contrary to how the Gangadeanian's present things, the notion of a law of nature is not something we have a satisfying account of. It's not a settled matter that for instance, Humean's must necessarily be wrong in their deflationary account. Laws of nature might simply be generalizations or regularities that we observe or they might not be, but the jury is still out. If the Gangadeanian disagrees, she had better be in a position to disprove any alternatives to her view and explain just what she takes laws to be ontologically speaking.

What this means for present purposes is that the analogy is not very elucidating. If laws of thought are to thinking as the laws of nature are to life, then we need to first settle the matter of the nature of the laws of thought. And part of that will involve determining whether a Humean analogue to the laws of thought is correct. That is, minimally, we need to figure out whether the laws of thought are merely descriptive generalizations about the way that humans characteristically think. If both kinds of laws end up being merely regularities that are determined inductively by observation, then the normative import is inclined to be limited. We have to make room for exceptions (generalizations after all admit to these and are formed by instances and counter-instances). More precisely, it would seem that we would have to put less credence in the notion that every rational being must necessarily think in accordance with them insofar as they are actually thinking.

Now faced with this issue the Gangadeanian is likely to pull the transcendental move. They will ask you to give them one instance of a thought which does not instantiate the laws of thought. They will say, "look, take any thought, aren't you assuming 'a is a' in having that thought? Can you give an example where you don't?" And they will think that they have proven something substantial. The correct response to this is to point out that they've done nothing more than appeal to intuition or common sense. For anybody to determine that a particular thought instantiates the law of identity (or not) for instance, requires that they just immediately "see" it as such. There's no argument. It's an immediate apprehension of sorts. So the Gangadeanian should recognize that they are resting ultimately on intuitions which won't satisfy the skeptic they are always intent on answering.

More crucially, my inability to provide an example of a thought that doesn't (intuitively) instantiate the laws of thought is not any proof that the laws of thought are not merely descriptive generalizations. That simply doesn't follow. Here's an analogy. Suppose you claim that every raven is black. That is you make an absolute claim about all ravens just as the Gangadeanians make a claim about all thought. Suppose I question you. I point out that it isn't obvious to me that all raven's are black. It does nothing to prove your point, if you merely ask me to produce a non-black raven even if I am unsuccessful. Sure, provided that I have access to a large sample of ravens, my inability to produce a counter-instance of your generalization does provide some inductive evidence for it. But that's not the same as proving your categorical rule! 

In response, the Gangadeanian will attempt to pull a reductio. They will say. "Ok let's assume that the laws of thought are merely descriptive generalizations of the way that people tend to reason. If so, then there's no reason why people should adhere to them. But all arguments presuppose that people should adhere to certain principles of reasoning otherwise there's no point. So, you see, you actually accept that they are more than purely descriptive generalizations." Unfortunately, this is just another pull on your intuition strings. The claim that all arguments presuppose that people should adhere to certain principles is far from trivially true. It's stated as a platitude, but that doesn't make it one. The problem is that you've got to just intuit that what was said is true. Once again I don't even know what would count as a proof here. What would possibly prove that all arguments presuppose that certain principles be adhered to? At best what we can say is that all the things we so far have determined to count as arguments appear to presuppose certain principles. But that's not nearly strong enough to support the line we're considering. So here too, the Gangadeanian would have to rest their case on intuition.

Another criticism is that the Gangadeanians seem to be generalizing from a limited sample to the whole which not only definitionally falls short of proof, it's bad induction in the form of a hasty generalization. In other words, you might challenge them on grounds that they have at best weak inductive evidence for their claim that all thinking requires the laws of thought. This is because they have experience with only those thoughts they've encountered in their lifetimes. And suppose we ignore the fact that they are merely using their intuitions to make generalizations about this set of thoughts--still this is a tiny subset of the set of all thoughts in the universe (present, past and future). So it's no good to conclude from this limited sample, anything general about all thoughts.

In response Gangadeanians are inclined to say something about "grasping a concept." Gangadean claims, but has no way of proving, that we grasp the essences of things whenever we "have" a concept. He just says this is so. Hence he and his followers sometimes claim that they grasp the concept of thought---and just by sort of thinking about thinking, they uncover the categorical truth that all thought presupposes or requires or exemplifies the laws of thought. But here again, this is little more than banking on their own raw intuitions. There's no argument given for the claim that "in a concept, we grasp the essence of things." More importantly, how does one argue or prove that one's concept is correct over another's? Suppose that you think it's essential to the nature of water that it is composed of hydrogen and oxygen molecules and I disagree because I think that water in a fundamentally different universe might have been composed of different elements. How can you possibly prove me wrong? There's no basis in argument about whether water is necessarily H20. 

Another move they might try in response to the charge of generalizing hastily is to say that thoughts come in types. That is, they concede the point that they couldn't possibly have encountered a sufficient number of thought tokens to make a good case for their claim about all thoughts. So they might try to claim that they have at least experience of a sufficiently wide range of the kinds of thoughts that there are and they generalize from there. Of course, this still falls short of deduction--it's induction. But it's also very dubious for other reasons. First of all, they would need infallible access to the properties of every single kind of thought there is--which would beg for a comprehensive taxonomy (for proof of concept). What counts as an argument type? Moreover, there's still just no way to know that we've nailed down every type of thought that there is, possible. And again, if there are any debates about what counts as a thought type (which there are bound to be), there would be no way of demonstrably settling it.

So don't be impressed by Gangadean's claim that the laws of thought are transcendental, or are preconditions to thought and that all thinking would cease without them. Or at least don't take his word for it that he knows these things with certainty (or that these things are absolutely clear to reason). He hasn't shown us that much.  True many philosophers accept the law of identity, and the law of non-contradiction (while fewer except the law of excluded middle). But they don't claim absolute certainty regarding them and freely admit that they accept them on the basis of intuition not via some deductively sound argument. My point here is not to say that the laws of thought are bad in some way to accept. My claim is that Gangadean should admit that he's got no privileged epistemic position regarding them. In the end, if he's consistent, he's just an intuitionist like the rest of us.