Why Paul Graham is (sort of) Wrong About Philosophy


A couple years ago, Paul Graham wrote an essay about philosophy. The essay, on my reading, makes three central claims:

  • Philosophy has largely been driven by our failure to realize that our words are imprecise.

  • This failure is the result of failing to search for useful knowledge.

  • We can do philosophy better if we filter out thoughts that don’t meet a usefulness criterion.

While I agree with a lot of what Graham says about philosophy in his essay, I think Graham is wrong on some of the more important points he makes. Before I say why I disagree with him, let me say that I think that Graham is a smart guy with a lot of interesting things to say and that I respect the dude a lot. We just happen to disagree on philosophy, and I, as a philosophical hacker, can’t resist expressing that disagreement.

Are we really just confused about our words?

Here’s Graham:

I would say that this has been, unfortunately for philosophy, the central fact of philosophy. Most philosophical debates are not merely afflicted by but driven by confusions over words. Do we have free will? Depends what you mean by "free." Do abstract ideas exist? Depends what you mean by "exist."

I sympathize the most with Graham’s first central claim, his claim that philosophy is largely driven by confusion over words. I have often wondered whether this sort of confusion was driving some of the discussion in my philosophy classes, but I have never worried that most of the questions I’ve discussed in my philosophical career are the result of tripping up over words.

What bothered me most about Graham’s point is that he seems to be ignoring large branches of philosophy that are less plausibly charged with addressing questions that are “driven by confusions over words.” Graham mostly seems to discuss metaphysics in his essay, but there are (at least) two other branches of philosophy, ethics and epistemology, that he seems to ignore almost entirely. Its less plausible, moreover, that these two fields are driven by confusion over words.

Let’s start with ethics. Suppose I’m wondering whether it is it right for me demand extra equity because I’m the guy who came up with the idea for our startup. This is not a question about the meaning of my words. If someone responded to my wonderings by saying, “It depends what you mean by the word, ‘right’,” we’d probably think they have missed something.¹ I don’t even think Graham would disagree with this. But why shouldn’t this count as evidence that there are a lot of philosophical questions that aren’t plagued by a confusion over words, especially given the fact that ethics has been a part of philosophy since before Aristotle’s Metaphysics was written.

Then there’s epistemology, the study of knowledge. How do we know that we’re not in the matrix? How can we know anything if all our beliefs rest on beliefs that are unsupported by evidence? (We’ve got to have some “foundational beliefs.”)

Again, if someone responded to these questions by saying that we are confused about the meaning of the word “know,” we’d feel that they have missed the point.² When I ask, “How do I know I’m not in the matrix?,” I’m asking, “Why should I believe that I’m not in the matrix?” That question makes no reference to the word “know,” and yet it is still, in my opinion, an interesting question. Epistemology is another big branch in philosophy, so if we think that there are epistemological questions that aren’t driven by confusion over words, then we have to admit that there’s another large section of philosophical questions that aren’t misguided in the way that Graham suggests.

Now, lets talk about metaphysics. Like I said, I’m more sympathetic to Graham’s point with respect to some kinds of metaphysics.³ Some metaphysical questions may be motivated by confusions over words and some metaphysical questions may be moral questions in disguise, but they are still important questions. Whether someone was “free to act” is something that we consider often in legal settings, and I think it’s right for us to ask this question before we lock someone up. If turns out that “I” don’t exist, to take another one of Graham’s examples, this might change the way we think about, for example, death.4

Even if Graham and I are together in thinking that some metaphysical questions are confused, I don’t think the way that he argues for his point is very compelling. In fact, Graham doesn’t really seem to argue for this point at all. I think he comes closest to arguing for his point in the following passages:

Most philosophical debates are not merely afflicted by but driven by confusions over words. Do we have free will? Depends what you mean by "free." Do abstract ideas exist? Depends what you mean by "exist."

To say that a question can be answered by saying, “Depends on what you mean by X” is far from proving that the question is motivated by a confusion over words. Case in point: Did evolution happen? Depends on what you mean by “evolution.”

Here’s another passage:

There's no way to prove a text is meaningless. The closest you can get is to show that the official judges of some class of texts can't distinguish them from placebos.

This point seems more promising if I understand what’s being proposed. Graham seems to be saying that if a philosophy professor can’t distinguish a “real” philosophy paper from a fake, placebo paper that is purposely filled with nonsense, then the real paper is meaningless. If this is Graham’s proposal, I think that many philosophy professors could pass this test for many philosophical texts. There are definitely some charlatans posing as philosophers out there, but that doesn’t mean that all or most of philosophy is B.S.

Maybe Paul wasn’t really trying to argue that philosophy is based on confusions over words. Maybe he was addressing his essay to people who already bought in to the idea that philosophy is a waste of time and who wanted to know how philosophers have managed to be so wrong for so long. If this is the intent of his essay, we should cut him some slack, but I’d be curious to know why he’s so sure that philosophers are just a sad, unintelligent bunch of people who got confused about their words.

I’ll address Graham’s other two points in subsequent posts. Maybe I’ll have that next post up by Thursday, but hopefully, Graham will pop in, make some comments, and I’ll have to delay that post until after him and I have talked a little.

Here’s the discussion on HackerNews.

Here’s my response to some of the comments.


  1. I take this to be one of the central points in G.E. Moore’s Principia Ethica.

  2. This is probably a controversial point. There are plenty of philosophers who are trying to analyze away skeptical worries by focusing on the meaning of the word “know.” Obviously, I think this is misguided, but I’m just a kid with a Master’s degree in philosophy, so what do I know.

  3. Even though I’m suspicious of metaphysics, no one should take my suspicion very seriously.  I’ve only taken 1 metaphysics class, and I know people who could probably make me doubt whether I should harbor any suspicion towards metaphysics.

  4. In Reasons and Persons, Derek Parfit argues that there are practical implications of ceasing to believe that the “I” exists.

On Philosophypaul grahamphilosophy

A response to Your Comments on "Why Paul Graham is (sort of) wrong about philosophy"

How to Break into a Mobile Dev Career in 10 months or less

comments powered by Disqus