The Problem of the Criterion: How Do We Know What We Know?

Here’s a strange thing philosophers noticed: Suppose you want to figure out whether you actually know something—whether your friend’s name is Alex, whether the sun will rise tomorrow, whether you’re dreaming right now. You seem to need two things. First, you need a list of things you already know, so you can study them and figure out what makes them different from things you don’t know. But second, you need a criterion—a rule or test—for distinguishing knowledge from non-knowledge, so you can tell which things belong on that list in the first place.

Do you see the problem? You can’t get the list without the criterion, and you can’t get the criterion without the list. It’s like needing a map to find the map shop. This is called the problem of the criterion, and a philosopher named Roderick Chisholm spent most of his career trying to solve it. But instead of trying to prove that the problem doesn’t exist, he did something different: he said, “Let’s just assume we already know a bunch of stuff, and work backward from there to figure out how.”

How to Build Knowledge from the Ground Up

Chisholm started with a simple idea: some things are just obviously true to you, right now, in a way that’s hard to doubt. For example, it seems to you that you’re reading these words. You might be wrong—you could be dreaming, or hallucinating, or an evil demon could be tricking you—but right now, it seems that way. That’s what Chisholm called a self-presenting state. It’s a state you’re in that presents itself to you directly.

Chisholm thought that if you’re in a self-presenting state—like seeming to see black text on a white background—then it’s evident to you that you’re in that state. Not certain beyond all possible doubt, but evident enough that you can build on it. You don’t need to prove that you’re not dreaming first; you can just start from the fact that it seems this way to you.

But how do you get from “it seems like I’m reading about philosophy” to “I actually know something about philosophy”? That’s the hard part. Chisholm proposed a series of principles—rules that connect what seems obvious to you to what you can reasonably believe about the world outside your mind.

Here’s one of his principles: If you believe you’re perceiving something to have a certain property—say, you think you’re seeing something red—then it’s reasonable for you to believe that you’re actually seeing something red. That’s not yet knowledge. But Chisholm added a further principle: if the property is something you can perceive with your senses (like redness, or roundness, or loudness), and you believe you’re perceiving it, and there’s no special reason to doubt (like you just took hallucinogenic drugs), then it’s evident to you that there’s really something out there with that property.

This gets complicated, but here’s what it accomplishes: Chisholm was trying to show how we can have knowledge of the external world—tables, chairs, other people, the past—starting only from what’s going on inside our own minds. He wasn’t trying to refute the skeptic who says “you can’t prove you’re not dreaming.” Instead, he was trying to show that if you accept a few basic principles about what’s reasonable to believe, you can explain how ordinary knowledge is possible.

The Problem of the Criterion: Methodists vs. Particularists

Remember the problem of the criterion? Chisholm had a name for the two sides of this puzzle. Methodists are people who think you should first figure out a method or criterion for knowledge, and then use it to decide what you know. Particularists are people who think you should start with particular things you’re pretty sure you know, and work backward to figure out what makes them knowledge.

Chisholm was a particularist. He admitted that he had no argument against methodism—he couldn’t prove that starting with the criterion was the wrong way to go. He just said: look, I’m pretty sure I know that I’m reading this page, that I have two hands, that 2+2=4. Let’s start there and see what kind of theory of knowledge we can build.

This might sound like cheating. But think about it: if the problem of the criterion is real, then any answer is going to seem question-begging. You have to start somewhere. Chisholm chose to start with common sense, following a philosopher named G. E. Moore, who once famously argued for the existence of the external world by holding up his hands and saying “Here is one hand, and here is another.”

The Strange Case of the Ship of Theseus

Chisholm also thought deeply about what it means for something—or someone—to persist through time. Here’s a puzzle that’s been around for thousands of years:

Imagine a ship. Over many years, every single plank and nail is replaced, one by one. Eventually, the ship contains none of its original parts. Is it still the same ship? Most people would say yes, because the change was gradual and the ship was always the same ship. But now imagine that someone collects all the old planks and nails and reassembles them into a ship that looks exactly like the original. Now we have two ships claiming to be the original. Which one is really the Ship of Theseus?

Chisholm thought there was something deeply weird about this. He said that ordinary objects like ships and tables can be said to persist through time only in a “loose and popular” sense—not in a “strict and philosophical” sense. In the strict sense, if you change any part, you’ve got a different thing. But we don’t talk that way in everyday life, and that’s fine.

Where this gets personal is when we apply it to ourselves. Chisholm noted that your body replaces its cells over time. After a few years, you might not have a single atom in common with the baby you once were. If you are your body, then in the strict sense, you’re not really the same person you were as a child. You’re just a loose-and-popular continuation.

But Chisholm thought this was obviously wrong. He gave an example: Suppose you need surgery, and the surgeon offers you a cheaper option—no anesthesia, but you’ll be given drugs afterward to make you forget the pain. Now suppose your friends tell you, “Don’t worry, we’ll just make it a convention that the person on the operating table isn’t you—it’s someone else named Smith.”

Would that reassure you? Of course not. No matter what convention people adopt, the person feeling that pain will be you. The question “Will that person be me?” has an answer that isn’t up for negotiation. This suggests that you are not a loose-and-popular construction like a ship. You are something more fundamental—something that persists strictly, not just conventionally.

Chisholm concluded that persons must be simple entities—not made of parts that can be swapped in and out. This put him in a tricky position, because your body clearly is made of parts. So if you’re not your body, what are you? Chisholm suggested, cautiously, that perhaps you are a simple, immaterial substance—something like a soul.

Can You Have Done Otherwise?

Another big question Chisholm tackled was free will. When we say someone “could have done otherwise,” what do we mean? This matters because moral responsibility seems to require that you could have chosen differently. If you were forced to do something, it’s not your fault.

Chisholm looked at several attempts to explain “could have done otherwise” and found them all inadequate. For example, some people say that “he could have gone to Boston” just means “if he had chosen to go to Boston, he would have gone.” But Chisholm pointed out problems: What if he didn’t know the way, so choosing to go would have caused him to get lost? Then he could have gone (he had the ability), but the conditional is false. And what if he was psychologically incapable of choosing to go? Then the conditional is true (if he had chosen, he’d have gone), but he couldn’t have gone because he couldn’t choose it.

Chisholm’s own solution involved a controversial idea: agent causation. Usually, when we think about causes, we think about events causing other events—the bowling ball’s motion causing the pins to fall. But Chisholm thought that sometimes, people cause things directly, without being caused to do so by prior events. When you freely raise your arm, you’re not just the place where a chain of causes and effects passes through. You are a genuine source of action.

This is a form of libertarianism (the philosophical kind, not the political kind). It says that free actions are not determined by prior causes. Chisholm admitted this means free actions are in a sense random—there’s no sufficient reason why you chose one way rather than another. But he thought that was the price of genuine freedom and moral responsibility.

The Puzzle of Conflicting Obligations

Chisholm also found a paradox in ethics that bears his name. Consider this situation: You have an obligation to help your neighbor. You also have an obligation to tell him you’re coming, if you’re going to help him. But if you’re not going to help him, you have an obligation not to tell him you’re coming. And in fact, you’re not going to help him.

Now, from the first two obligations, it seems you should tell him you’re coming. But from the last two, it seems you should not tell him you’re coming. You can’t do both. This is called Chisholm’s Paradox (or the contrary-to-duty imperative problem). It shows that our ordinary moral thinking contains logical tensions that aren’t easy to resolve.

Philosophers are still working on this. Some have developed special logics of “conditional obligation” to handle cases like these. But Chisholm’s simple example revealed a deep problem in how we think about morality.

The Big Picture

Chisholm’s work touches on some of the biggest questions in philosophy: What can we know? What kind of thing am I? Do I have free will? What makes something good or bad? His answers were often unfashionable—he believed in souls, in agent causation, in starting from common sense rather than from doubt. But his arguments were always careful, precise, and honest about their difficulties.

He thought philosophy should start with puzzles, not with certainty. He would present a puzzle, propose a solution, invite objections, and then revise. His definitions would get longer and more complicated as objections piled up. (A joke among philosophers: “to chisholm” means to make repeated small changes to a definition.) But the core ideas stayed surprisingly stable over fifty years of work.

And the problem of the criterion? Chisholm never claimed to have solved it. He just chose his starting point and built from there. Maybe that’s the best any philosopher can do.

Key Terms

Term	What it does in the debate
Problem of the criterion	A puzzle about whether you need a rule for knowledge before you can list what you know, or a list before you can figure out the rule
Self-presenting state	A mental state that seems directly obvious to you (like “it seems to me I’m seeing red”)
Methodist	Someone who thinks you should find a criterion for knowledge first, then apply it
Particularist	Someone who thinks you should start with things you’re pretty sure you know, then figure out the criterion
Loose and popular identity	The kind of “sameness” an ordinary object has when it changes parts over time
Strict and philosophical identity	The kind of “sameness” that requires having exactly the same parts
Agent causation	The idea that people (not just events) can directly cause things to happen
Chisholm’s Paradox	A puzzle showing that moral obligations can conflict in tricky ways when you’ve already failed to do something

Key People

Roderick Chisholm (1916–1999): An American philosopher who spent most of his career at Brown University. He thought philosophy should start with puzzles and build careful systems from a few simple principles. He was famous for constantly revising his definitions in response to criticism.
G. E. Moore (1873–1958): A British philosopher who argued that we know many things for certain (like that we have hands) even if we can’t prove them. Chisholm followed Moore’s approach of starting from common sense.
David Hume (1711–1776): A Scottish philosopher who argued that when he looked inside himself, he found only perceptions, not a single “self.” Chisholm disagreed, arguing that Hume’s own description revealed there must be someone doing the looking.
Franz Brentano (1838–1917): An Austrian philosopher who argued that all mental states are “about” something. Chisholm spent decades trying to formulate Brentano’s idea precisely.

Things to Think About

Suppose you’re completely wrong about something—say, you’re dreaming right now. Does it still make sense to say you know you’re reading this? Or does knowledge require that you’re not mistaken, even if you can’t prove it?
If you learned that every atom in your body will be replaced within seven years, would you worry about whether you’ll still be “you” seven years from now? Or does it feel obvious that you’ll be the same person?
Chisholm thought that for free will to be real, some of our choices must have no sufficient cause. But if a choice has no cause at all, in what sense is it your choice? Could a random event really be something you’re responsible for?
Think of a time you failed to do something you should have done. Did you then have new obligations (like apologizing) that you wouldn’t have had if you’d just done the right thing in the first place? Does that seem fair?

Where This Shows Up

The problem of the criterion shows up whenever you’re trying to figure out what counts as reliable information. Is Wikipedia reliable? How would you decide? You need a rule, but you also need examples. This is the same puzzle.
The Ship of Theseus is relevant whenever someone gets a new phone or computer. If you replace the screen, then the battery, then the case—is it still “your” phone? What if someone assembles a phone from your old parts?
Agent causation and free will debates appear in courtrooms, where judges and juries decide whether someone is responsible for their actions. If everything we do is caused by prior events (our genes, our upbringing, our brain chemistry), can anyone ever be blamed?
Chisholm’s Paradox about obligations appears whenever someone has made a mistake. If you break a rule, you now have new obligations (to apologize, to make up for it) that you wouldn’t have had otherwise. But those obligations seem to come from the fact that you already failed—which is a weird thing to build morality on.