A Defense of Knowing Nothing
by Robert Gressis
In this essay, I argue against the idea that in order to count as knowing some proposition P, P should be true, you have to believe P, and you should have some justification for believing P.
I have several arguments for this view, but honestly, I don’t need any because I know things and I have no justification for many of these things I know. So, that should end the discussion right there. Also, since my thesis is right, I don’t have to give arguments anyway, because it’s part of my thesis that people don’t need to have any justification to count as knowing things.
Now, let’s say, somehow, you make a compelling case that my view is wrong. You may want me to recant and get frustrated when I don’t. But remember, it’s not part of my view that you need to believe something to know it – and luckily for me, I don’t believe my own view, so there’s nothing for me to recant.
In addition, even if my view is false, so what? I can know things even if they’re false. So, in disproving my view, you proved it.
That was the introduction. You got a sense of what I’m going to say from that first paragraph. Note how clear the introduction was. And note how insane my view is! Don’t you just want to attack it?
Also, look at the first sentence: no narrative hook at all. I just, like, started. This is how you should teach your students to write!
Situating My View in the Literature
Lots of philosophers think that to count as knowing some proposition P at least the following have to be true: you have to believe P, you have to have justification for P, and P has to be true. Now, many philosophers think that, in addition, you need something else to know P: “warrant” or “non-luckiness” or something like that. And some philosophers go the other way: they’ll say you can know P without believing it , or they’ll say that you can know P without its being true, or they’ll say you can know P without having any justification. But I’m the only philosopher who thinks that you can know P without any of these.
Call my view the “Knowledge Needs Nothing” View (KNNV). My acronymizing this view — and also my using the letter P to stand for “proposition” — makes my view seem scientific, and therefore, daunting. Also, the fact that I’m the only one who accepts KNNV makes my view a significant contribution to the literature. 
Arguing for KNNV
I have a battery of arguments for why KNNV is right.
The Linguistic Practice Argument
First, look at how people talk. Imagine something unexpected happens. Not only did this unexpected thing happen, but you wanted it to happen. You may say, “I knew it!” This is a perfectly acceptable thing to say even though you didn’t have any justification for knowing it. If you say it, you probably won’t get push-back, though people may roll their eyes. And if someone does push back, people will probably roll their eyes at him. So, our linguistic practices show that we accept the idea that you can know something without justification.
Second, imagine that something unexpected happens, and you say you knew it, but then later something else happens nullifying the unexpected thing. For example, imagine that many epidemiologists say that the coronavirus will kill between 200,000 and 1.7 million people. You don’t want this to be true, so you disagree. And then, after a few months, it looks like the coronavirus will kill only about 60,000 people. You say, “I knew it!” At this point, only Aspergery jerks will try to correct you. But then, states open up and 300,000 people die by the end of the year. So, the epidemiologists were right, and you were wrong. But it was still OK to say “I knew it!” back when you said it, even though, technically, you didn’t have good reason for saying it, and even though, technically, what you said turned out to be false.
Third, after 300,000 people die, someone dredges up your Twitter feed where you said “I knew it!” They point it out to you. What a jerk! But it’s OK. You can just say, “obviously, when I said ‘I knew it!’ I didn’t believe that. I was just venting. And if you’re familiar with conversational implicature, you’d know that’s what I meant.” That would be a good response, and it fits into a single tweet.
In summary, our linguistic practices say you can know something even if you have no justification, what you say is false, and you don’t believe it. The thing is, philosophers figure out their intuitions about concepts like knowledge simply by looking at linguistic practices. So philosophers themselves have to accept that our linguistic practices are what show how we may use our concepts.
So, here’s my argument:
 Our linguistic practices show that we can count as knowing things that we don’t have good justification for, that are false, and that we don’t even believe.
 The way we’re permitted to use concepts comes from our linguistic practices.
 Don’t even get me started on warrant.
 Therefore, we can count as knowing things that we don’t have good justification for, that are false, and that we don’t even believe.The Argument from Disagreement
There’s just so much controversy in philosophy. I mean, you can say that I’m wrong about some philosophical proposition P, but lots of philosophers would disagree with you. And if lots of philosophers disagree with you, then you should lose confidence in what you’re saying. Now, I admit that this argument applies to my position as well. But that’s OK. I’m not saying what you must say. I’m saying what you’re permitted to say. And it’s perfectly OK to say, “I’m not confident about P, but you may say P.” But it’s not OK to say, “I’m not confident about P, but you must say P.” So, my argument in this section goes like this:
[1′] There is lots of disagreement about what the necessary conditions for knowledge are.
[2′] If there is lots of disagreement about X, then you shouldn’t be confident about X.
[3′] Therefore, you shouldn’t be confident about the necessary conditions for knowledge.
[4′] If you shouldn’t be confident about something S, then you shouldn’t say, “here’s how you must talk about S.”
[5′] However, even if you shouldn’t be confident about S, it’s still OK to say, “here’s how you may talk about S.”
[6′] Therefore, you may say that you can know something even if you have no good reason to believe it, it’s false, and you don’t even believe it.
The Modal Argument
Philosophers love modal stuff. For example, here’s a good modal argument:
[1”] In some possible world W, there is a decisive argument for showing that it’s necessarily true that God exists.
[2”] If it’s possible that an argument decisively shows that necessarily that God exists, then necessarily God exists.
[3”] Therefore, it’s necessarily true that God exists.
The beautiful thing about this argument is that I don’t even have to come up with the argument that decisively shows that God exists! But premise 1 has got to be true. Think about it: there are infinitely many possible worlds. Does it really seem likely to you that in none of them there’s a decisive argument for God’s existence? Me neither. But this argument iterates. So, here’s another argument for my conclusion:
[1”’] In some possible world W, there is a decisive argument showing that necessarily, you can know KNNV even if it’s false that KNNV and you don’t have good reason to believe KNNV and you don’t even believe KNNV.
[2”’] If it’s possible that an argument decisively shows that necessarily KNNV, then necessarily KNNV.
[3”’] Therefore, it’s necessarily true that KNNV.
Of course, someone may point out that there’s also an argument in possible space that shows, necessarily, ~KNNV. Against this, I offer the following rejoinder.
Obviously enough, there is a decisive argument for ~KNNV. But remember, if KNNV is right, then I know it even if it’s false, I have no good reason for it, and I don’t believe. So, what do you expect this decisive argument for ~KNNV to show, really?
In this paper, I have shown, using three arguments, that you can know something even if you don’t believe it, even if it’s false, and even if you don’t have any good reason for it. This is a very significant result, because no one has ever said this before.
So far, what I’ve said is pretty similar to the introduction. But it’s a conclusion, so I need to add something, so here’s a promissory note: in future work, I will investigate whether KNNV is not only sometimes true, but always, and necessarily true. But doing that would take me beyond the scope of this paper.
 See Jonathan Jenkins Ichikawa and Matthias Steup, “The Analysis of Knowledge”, Stanford Encyclopedia of Philosophy, 2017.
 See Alvin Plantinga, Warrant and Proper Function (Oxford: Oxford University Press 1993).
 See Linda Zagzebski, “The Inescapability of Gettier Problems”, The Philosophical Quarterly, vol. 44, no. 174 (1994: 65-73). Note that Linda Zagzebski is a woman, so my citing to her is inclusive.
 See Colin Radford, “Knowledge—By Examples”, Analysis, vol. 27, no. 1 (1966: 1-11) and Blake Myers-Schulz and Eric Schwitzgebel, “Knowing that P without Believing that P”, Noûs, vol. 47, no. 2 (2013: 371-384).
 See Allan Hazlett, “The Myth of Factive Verbs”, Philosophy and Phenomenological Research, vol. 80, no. 3 (2010: 497-522).
 See Hilary Kornblith, “Knowledge Needs No Justification”, in Quentin Smith (ed.), Epistemology: New Essays (Oxford: Oxford University Press, 2008), pp. 5-23. Eagle-eyed readers will realize that pretty much everything I’m citing comes from a super-cursory reading of the first couple of paragraphs of my first citation.
 Or do I? See section I.
 You ever heard the adage, “work expands to fill time”? Well, here’s another one: “philosophy expands to fill logical space.” I’m doing my part here.