Against TJump's Very Crazy Epistemology

A popular and very belligerent YouTuber who I recently debated has the worst epistemological view of all time

Tom Jump (better known by the acronym TJump) is an “Atheist, Philosopher, aspiring POTUS.” These words come from the mouth of none other than “Atheist, Philosopher, aspiring POTUS,” Tom Jump—a man who is a philosopher only in the sense that he produces YouTube videos where he says very confused things about philosophy. That the things he says are very confused is a judgment not just had by me, but by Alex Malpass, Michael Huemer, James Anderson, Robin Collins, and almost every other philosopher with whom he speaks, many of whose eyebrows, by the end of the conversation, have been raised so frequently and with such fervor that they’re likely to be sore for weeks.

I recently debated Monsieur Jump about the existence of God. The debate included lots of quite hilarious and bizarre claims—and quite a profound dose of belligerence, wherein Tom would claim that he had debunked me on a point so thoroughly that he’d mute me until I moved on from the point (the debate took place in his discord server where he could mute me—he muted me 8 times, mostly when he was backed against the wall and had nothing to say). The list of bizarre and hilarious claims he made is too vast to fully enumerate, so I’ll have to give just a sample of them including:

1. He claimed that the way you come to verify the reliability of your memory is by making guesses about what will happen if your memory is reliable and those being confirmed. When I pointed out the very obvious problem that this requires relying on your memory to verify that you in fact made the prediction, he claimed that you can have justified false belief (this is true of course) and that this somehow debunked my point. When I asked how the mere existence of justified false beliefs provides any sort of justification here, he claimed he had already refuted me and demanded that we move on. Okay.

2. He claimed that if Einstein had, for no reason, built into his theory of general relativity that there’s a colorless invisible square circle then it would be reasonable to believe that there is, in fact, a colorless invisible square circle. This is so even if the colorless square circle has nothing to do with the rest of his theory but is instead a random extra posit that Einstein happened to build into his theory for no reason.

3. He denied the existence of dark matter without realizing it. He claimed that you should never believe in a new kind of thing based on it explaining past data—it needs to make new predictions. When I pointed out that the main reason to believe in dark matter is because it explains anomalies, rather than making novel testable predictions, he claimed that dark matter isn’t a thing, but a label for an unknown cause of anomalies. In response I pointed out that this is wrong—dark matter is the notion that there’s invisible matter that pervades the universe, which is why there are alternative theories to the dark matter theory like MOND as even a cursory Wikipedia search would reveal. He claimed, ludicrously, that MOND is just a version of the dark matter theory, which is false, and then that he’d already refuted me on the point, proceeding to mute me and request that we move on. This means that he denies what physicists mean by dark matter, as his epistemology rules it out.

4. He claimed that we should reject the fine-tuning argument because historians don’t appeal to miracles in explaining historical events. Now, aside from misreading what the historical article he was citing was saying, this is a very ridiculous claim.

TJump is to philosophy what a crank physicist is to physicist. He thinks he knows what he’s talking about but he does not. His views on epistemology are proprietary and taken seriously by no one other than him. He combines extreme overconfidence, dishonesty (in the debate he fabricated a quote from Luke Barnes and he made up a Stanford Encyclopedia of Philosophy page in a previous debate), and titanic conceptual confusion. A philosopher friend of mine once remarked that when one listens to what he says, they feel as though they have to give him an intro to philosophy lecture before addressing any of his points given how deep the conceptual confusion is.

Still, seeing as he has many fans, I think it would be useful once and for all to explain why his views on epistemology are so confused. Hopefully this will convince some people not to adopt his backwards views.

Tom claims that it only makes sense to believe in something if it makes novel testable predictions. It makes sense to believe in the theory of general relativity because it made advanced predictions that turned out right. Just explaining the data is easy—you can always make up a theory that does that. Tom says that for this reason, you need to make predictions in advance—that’s what differentiates a real scientific theory from pseudoscience like belief in the mythical, magical, or paranormal.

Now, this simple formulation has an obvious problem that I pointed out; it would completely rule out any historical knowledge. When figuring out that Julius Caesar exists, historians didn’t make a prediction that there would be documents and later discover them. Rather, they knew there were documents, and then inferred that Caesar exists.

Tom has a response: rather than his principle being about new things, it’s about new kinds of things. Julius Caesar is a kind of human, and we have empirical evidence for humans, so we don’t need to make advanced testable predictions based on his existence. But believing in a new kind of thing—like a leprechaun—based on it explaining data would not be reasonable. It would need to make correct predictions.

What does it mean to be a new kind of thing? Sure there’s a category that Julius Caesar is part of that we have evidence for—humans. But there’s also a category that God and leprechauns are part of that we have evidence for—things that exist. Tom clarified that by a new category of things, he means something not composed of existing things. Julius Caesar is composed of things we have evidence for—atoms—but God is not, so it doesn’t make sense to believe in God based on explaining existing features of our world, but it does make sense to believe in Caesar based on his explaining those. More precisely TJump’s view is:

It’s only reasonable to believe in something if either:

1. Novel testable predictions have been made on the assumption that the thing exists.

2. The thing is composed of other things that have been used to make novel testable predictions.

Okay, now we have the theory, why is it wrong?

First, it’s totally unmotivated. It’s very complicated and gerrymandered, and there’s simply no reason to accept it. Why think epistemology works completely differently when it comes to new kinds of things vs new arrangements of existing kinds of things (e.g. atoms we know about existing arranging to form a new thing—Julius Caesar). Tom claimed in defense of it that it’s widely accepted in the sciences, but this isn’t true—science is methodologically naturalist in that it tends to only consider natural explanations of events, but scientists don’t have some well-worked out epistemology for examining supernatural explanations. It simply isn’t their field of expertise. No one prior to Jump endorsed this epistemological theory, so it’s quite odd to claim that it’s the consensus view.

Second, as explained before, it would rule out believing in dark matter. Dark matter is a new kind of thing, not composed of existing stuff. It’s believed in because it explains existing data, not because it made advanced experimental predictions. So on Tom’s epistemology, we’d have to throw it out.

Third, imagine a voice from the heavens appeared declaring “Christianity is the true religion.” It was heard by everyone in the world. Then, it performed many miracles including parting oceans, causing everyone to experience an incalculably pleasant hedonic state without changing their brain, and exploding several stars. On Tom’s view, because this wasn’t predicted in advance, it would be no evidence for Christianity. This is very silly! In fact, if none of the religions had predicted postmortem heaven, on Tom’s view, finding oneself in heaven wouldn’t be evidence for theism. This is especially weird when conjoined with Tom’s view that if you prayed to God and got a gold brick repeatedly, you would have strong evidence for theism. So God performing predicted parlor tricks is evidence for God’s existence but appearing in a cloud of glory and declaring Christianity to be the right religion is not (assuming that this was pre second-coming, so not predicted by Christians).

Fourth, it’s unclear how such a view would explain our knowledge of other people. I’ve never taken the assumption that there are other people as a hypothesis and made predictions based on that. So on Tom’s epistemology, it seems that I would have no evidence for the existence of other people. It isn’t enough for it to just be a thing I believe as I make testable predictions, as that would mean that I also have evidence that, say, my political views are right from the fact that I had them while making testable predictions.

Fifth, his view implies something rather silly. Imagine that there was a law that before one formulates any theory about physics, first they have to read about every experimental result. Imagine that every conceivable experiment had actually been performed and people could read very quickly about all of them to learn them all. By the time one had formulated theories, they knew every experimental result.

In such a case, one could never make novel testable predictions because they already know the result of every experiment. Therefore, on Tom’s view, no one would ever be justified in believing in any new kind of thing, no matter how well it explained the data. In such a world, one should deny atoms, electrons, protons, and neutrons—even if there was a theory of everything that perfectly explained all the facts and had no suitable competitors one should still deny it. This is quite silly! Reading about all the conceivable experiments as an undergraduate shouldn’t make it unreasonable to believe anything scientific!

Sixth, it can’t even be used to rule out most pseudoscientific hypotheses. It won’t rule out using leprechauns or aliens to explain phenomena, because they’re not composed of new stuff, but instead of rearrangements of existing stuff. Now, you could modify the principle to the following to avoid the problem:

It’s only reasonable to believe in something if either:

1. Novel testable predictions have been made on the assumption that the thing exists.

2. The thing only has properties that have been used to make novel testable predictions.

But 2. would rule out believing in Caesar on the basis of testimony. Julius Caesar has the property of being Julius Caesar, and yet it’s reasonable to believe in his existence on the basis of testimony. It would also rule out believing in Pangea—that all the continents used to be one—which is primarily believed in on the basis that the life forms on various continents millions of years ago were very similar to the life form on adjacent continents (e.g. Africa and South America had similar plants).

Seventh, it implies ludicrously that if little green men appeared in the sky in space ships—and we investigated them under a microscope and found they weren’t composed of any of the particles that we know exists—that would be no evidence for aliens (after all, no one predicted their existence in advance and they’re made of a new kind of thing!)

Eighth, the view is self-defeating. If the only way you can get evidence for something is based on making novel testable predictions on the assumption that it exists, you can never verify that you actually made the predictions—that requires relying on your memory which, by the very standard laid out, you have no evidence for the reliability of.

Ninth, the view still has the arbitrary reference class problem. The second condition under which it can be reasonable to believe in a thing is that “the thing is composed of other things that have been used to make novel testable predictions.” But this requires how you define the thing. For instance, it’s true that atoms have been used to make novel testable predictions, but atoms composing Julius Caesar have not. There are multiple descriptions of what it is that composes any new thing, and there’s no principled way of picking out a privileged one—but the view requires finding a privileged description of the things composing an object, for it relies on the notion that those things make future advanced predictions.

Tenth, prediction alone is clearly not enough. If a person predicts that they have a square circle in their pocket that will cause them to win the lottery, if they win it, you still shouldn’t think they have a square circle in their pocket. If Einstein had included in his papers on general relativity “oh and by the way, this theory is of course based on the assumption that I have a square circle in my pocket,” it wouldn’t be reasonable to think he actually had a square circle in his pocket.

Eleventh, his view implies not only that evidence for God doesn’t exist but that it’s impossible. If we discovered that the beginning of the universe was particles arranged to spell out “created by God,” and that shortly after the creation of the universe, an infinite number of different things appeared in violation of the laws of physics, this would be no evidence for the existence of God. Nuts!

Tom often flummoxed the philosophers he has on. They’re so confused by his strange, shifting, hard to pin down epistemology, that he often comes off as dominating them in conversation. Yet when one takes the time to carefully consider his views, they make absolutely no sense. His view of epistemology is about as implausible as any such view could be.