If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Main content

Video transcript

(intro music) I'm Jennifer Wang. I'm a professor of philosophy at the University of Georgia. Today, I'm going to talk about the Ship of Theseus puzzle. This puzzle was recorded by Plutarch, an Ancient Greek historian, though it's come up in many different forms over the ages. It goes like this. Theseus was this great mythical hero of Athens, who sailed off to Crete and slew the Minotaur, a creature with the head of a bull and the body of a man. After Theseus came back, his ship was left in the Athenian harbor as a memorial. Over centuries, the planks of the ship decayed and were gradually replaced. Now, it doesn't really matter that the planks decayed, or that the ship still had masts and sails and other ship stuff too. We can simplify the story. Let's pretend that the Ship of Theseus is a very simple ship, made of one thousand planks and nothing more. Let's also say that the planks are made of invincible wood, super wood, so that they never decay. In what I'll call "scenario one", the Ship of Theseus has its one thousand planks replaced very slowly, over the course of one thousand years. That's one plank a year. So here's the puzzle. Surely a ship can survive the replacement of one of its planks. In year one, when the first plank is replaced, it's still the Ship of Theseus. In year two, when the second plank is replaced, it's still the Ship of Theseus, and so on, through year one thousand. But the ship at year zero, the original Ship of Theseus, doesn't share any of the same parts with the ship at year one thousand, which we can call "A." So how can A be the real Ship of Theseus? Thomas Hobbes, a seventeenth- century English philosopher, added a twist to the story. In scenario two, a ship repairman keeps all of the old planks of the Ship of Theseus and uses them to build an exact replica of the original ship, with all of the planks in the same arrangement. So in this scenario, at year one thousand, there are two exactly similar ships: the one whose planks were gradually replaced, which we called "A" in scenario one, and the one built from the old planks, which we can call "B." Now, A has the same claim to being the real Ship of Theseus as it did in scenario one. But B also has a good claim to being the real Ship of Theseus. After all, it's made of the same parts as the original Ship of Theseus, in the same arrangement. But they can't both be the Ship of Theseus. Let's look more carefully at the underlying assumptions that generate the puzzle. One assumption is that ordinary objects survive gradual change. This is very plausible. You can't destroy a coat just by removing one of its buttons. Maybe you then ruin the aesthetic of the coat, but that's not what's at issue here. It's still the same coat. It's just changed a bit. The principle that ordinary objects survive gradual change motivates the conclusion that A is the real Ship of Theseus. Another assumption is that an object goes where its parts go, so to speak, at least in cases where the parts are in the same arrangements. Let's modify our scenario so that the planks of the ship are gradually removed, but aren't replaced with new planks. Again, the old planks are used to build an exact replica of the ship so that, at the end of the new scenario, there's only one ship, the ship we called "B." Call this modified scenario "scenario three." The principle that an object goes where its parts go motivates the conclusion that B is the real Ship of Theseus in scenario three. But it motivates this conclusion in scenario two as well, where there are two ships at the end. It doesn't look like both principles can stay. Which should go? Let's go through some possible solutions to the puzzle of the Ship of Theseus, some of which involve rejecting one principle or the other. They all come with disadvantages. Solution one is to deny the parts principle. This solution involves saying that in scenario three, the ship at the end is not the Ship of Theseus, even though it has all the same parts arranged in all the same ways. Solution two involves denying the change principle: ordinary objects survive some gradual change but not all. That is, sometime between the year zero and the year one thousand, removing a plank destroys the Ship of Theseus. The problem is that this solution seems arbitrary. Why would removing, say, plank number 543 destroy the Ship of Theseus, but not number 542? And at that moment, does the ship being built out of the old planks in scenario two suddenly become the Ship of Theseus? On solution three, the plank which destroys the Ship of Theseus is not some middling plank. Rather, as soon as plank number one is removed, the ship is destroyed. This solution involves denying the change principle as well, but it offers a stronger thesis in its place: ordinary objects never survive any change. This view was advocated by Roderick Chisholm, a twentieth-century American philosopher, who was inspired by Bishop Joseph Butler, an eighteenth-century English theologian and philosopher. Butler's thesis was that ordinary objects like ships persist only in a loose and popular sense. Whether A or B is regarded as the Ship of Theseus ends up being something of a practical matter. According to Butler's thesis, no ship really ever survives any change. However, not only is this view implausible, it implies that there are one thousand ships where we thought there was only one, as the destruction of each ship is followed by the creation of a new one. On solution four, neither the change principle nor the parts principle needs to be rejected. Rather the solution here is to say that A and B are each the Ship of Theseus. This involves rejecting the following principle, called the "transitivity of identity": if X is identical to Y, and Y is identical to Z, then X is identical to Z. On solution four, A is identical to the Ship of Theseus, and the Ship of Theseus is identical to B, but A is not identical to B. According to solution five, the Worm Theory solution, we need to change the way we're thinking about ordinary objects. Here's the idea. I introduce scenario two like this: there is a ship at year zero and two ships at year one thousand, and the challenge is to figure out which of the two ships at year one thousand is identical to the ship at year zero. The implicit assumption that worm theory rejects is that ordinary objects like ships are three dimensional objects, where the three dimensions are spatial dimensions. According to worm theory, ordinary objects really have four dimensions: three spatial and one temporal. So there are no ships wholly present at year zero or at year one thousand. Rather, there is one worm-like ship which has a part at year zero at one end, and has A as a part at the other end. And there is another worm-like ship which has a part at year zero at one end, and B as a part at the other. The two worm-like entities have overlapping parts at year zero. This solution doesn't require rejecting transitivity, the parts principle, or the change principle. After all, it's no longer clear what claim we're making when we assert "A is identical to the Ship of Theseus" or "the Ship of Theseus is identical to B." A and B are not identical to each other, but nor are either of them identical to the Ship of Theseus. They both have the object at year zero as a part. That is all. As you can see, accepting any of these five solutions comes with disadvantages, but to resolve the puzzle, it looks like we have to accept some disadvantage or other. Subtitles by the Amara.org community