Causation Survey (A Real One!)

Christopher Hitchcock and Joshua Knobe are running a study on intuitions about causation that they’d like you to take. ‘You’ here including any readers of this blog, and friends you have, any random stranger you meet on the street, etc. Unlike the little test I did here, this one features some careful controls, and has even been approved by an Institutional Review Board. So go at it!


One of my weaknesses as a writer is that I’m not the best at keeping track of what other people have said about stuff I’m writing about. I’m sure in the past this has led to me citing less papers than I should have, simply because I didn’t know about the paper to be cited. Frequently, it leads to me finding out when well into a project about something that I should have read at the start.

This is all a roundabout way of saying that I just read, and quite liked, Samir Okasha’s paper What Did Hume Really Show About Induction?. Samir takes Hume to be making what I call the exhaustive argument for inductive scepticism. More importantly, he takes the problem with the argument to be that the argument for the ‘induction can’t be justified empirically’ premise assumes that we decide in advance how we’ll respond to new evidence. That’s, er, very close to what I think is wrong with the argument. So it’s a good paper, and I encourage you all to read it. (There is also a back and forth with Marc Lange in later issues of the Philosophical Quarterly on this paper.)

Yet More Good News for Cornell

Cornell is very pleased that Tad Brennan has accepted an offer to join the Sage School. Cornell obviously has a great tradition in ancient philosophy, and Tad continues the excellent tradition. He recently published a book on the Stoics, but has written on historical figures from the preSocratics to the so-called “late Platonists” through the 6th Century.

This just about completes a very active hiring season for Cornell. Depending on just when everyone starts, we’ll soon be welcoming Derk Pereboom, Karen Bennett and Tad Brennan as senior appointments, Erin Taylor as an assistant professor, and Wylie Breckenridge as a post-doc. We may have more news to announce yet too. So it’s an exciting time up here.

More Compass Articles

As always, clicking on the link will take you to an abstract of the paper. For the full article, you need to get your library to subscribe.

Around the Web

A few links and stuff for the morning, while we grade and/or wait for the world cup to start.

  • The Order of the Science Scouts of Exemplary Repute and Above Average Physique has a very large selection of scout badges for successful (and not so successful) scientists. I especially like the I left the respectable sciences to pursue humanistic studies of the sciences badge, but you may have your own favourite. (HT: Elizabeth Lee)
  • A new romantic comedy starring a middle-aged Jack Nicholson, with a certain resemblance to a Stanley Kubrik movie, has a “trailer on YouTube. (HT: Andrew McGonigal)
  • If you’d rather win the world cup from your own computer, and have several hours between grading to spare playing online cricket games, you’ll love Stick Cricket. If you are susceptible to timesinks, you’ll hate me for posting this.
  • The Michigan grad students have a new blog, Go Grue. I’m always pleased to see new grad student group blogs; it’s a good way to get ideas talked about.
  • There is a nice article about Terence Tao in the NY Times this morning. As many of you will know, I was one year away from being Prof Tao’s teammate on the Australian Maths Olympiad team. There are a few interesting comments in the article about the move from doing maths competitions to becoming a professional mathematician.

Good News for Cornell

Two more pieces of good news for Cornell.

Erin Taylor (UCLA) has accepted a tenure-track position at Cornell. Erin works on moral conflicts and associative obligations, and has developed really interesting positions on both topics. She argues, for instance, that it is possible to have situations where one person is obliged to do X, and another is obliged to prevent X happening. And she has a new take on why conventions of promise-keeping play an important role in generating the moral force of the obligation to keep promises. Erin is currently at the UCLA Center for Society and Genetics.

Wylie Breckenridge (Oxford) has accepted a Mellon Postdoctoral Fellowship to Cornell. Wylie’s thesis is on the meaning of the verb ‘looks’. He has papers on, well, approximately everything. (At this stage if we were doing PECOTA comparisons, Wylie’s main comparable philosopher would be, I’d guess, me.) As well as having a number of striking philosophical views, his forthcoming Philosophical Perspectives paper for instance argues against the view that visual experience has representational content, he is a great philosophical interlocutor, and will be a real asset to Cornell.

Obviously we’re very pleased each of them is coming to Cornell. (Hopefully they’ll each be active on the northeast conference circuit, because they are great people to talk philosophy with.) Cornell next semester will be almost unrecognisable compared to this semester. There will be at least six people teaching classes who are not teaching this semester, and maybe several more than that. Students considering Cornell should bear this in mind when thinking about what the program will look like over the next few years.

Causation Survey

UPDATE: The survey is now closed. See below for results and discussion, and thanks to everyone who took part!

This is a bit of a gratuitous request. I’m very interested in cases like the one below, but my intuitions about the case are exceedingly unclear. So I’m hoping that if I ask all my friends for their reaction I’ll understand the case (and one interesting class of cases of which it is a member) much better. So if you could take the poll below I’ll be very grateful. (I’m not sure why there is so much space above the survey, by the way, but I couldn’t figure out how to get rid of it.)

Consider the following story, and say which of the sentences below are true in it.

Host is hosting a dinner party at which Guest is a guest. Host lives on the eighth floor of a building with a single elevator. The party is going well until Host mentions the war. This upsets Guest who storms out and calls the elevator, by pressing the down button on the elevator console. Unfortunately, the elevator has just left and won’t return until it has gone to the ground floor. While Guest is waiting, and before the elevator has reached the ground, another person on this floor, Neighbour, comes to the elevator. She would have called the elevator, which had not yet reached the ground, but saw that Guest had already called it. If Neighbour had called the elevator, it would have arrived at the time it actually arrived, since the elevator would not have started back up until it reached the ground in any case. Some time after that the elevator arrives and Guest and Neighbour ride it to the ground floor.

UPDATE: I’ve now closed the survey after getting 200 responses. Here are the responses.

The number in brackets at the end is the number of the respondents to have said the sentence is true in the story.

  • Guest’s calling the elevator caused the elevator to arrive (197)
  • If Guest had not called the elevator, it would not have arrived (10)
  • Neighbour caused the elevator to arrive (0)
  • Host’s mentioning the war caused the elevator to arrive (33)
  • Neighbour was a cause of the elevator arriving (4)
  • Host’s mentioning the war was a cause of the elevator’s arrival (95)

I’ll write this up in more detail soon, but there are two big things I wanted to draw attention to.

First, the difference between ‘caused’ and ‘a cause of’. My impression is that a lot of the literature on causation runs these two together. Now it is possible that there is no truth-conditional difference between the expressions, but I would say that’s rather unlikely. So people who want to analyse these expressions should be aware of the possibility, perhaps probability, that the same analysis will not work for each.

Second, the difference between causal relations that ground causatives and those that don’t in what we’d ordinarily call pre-emptive causation networks. It’s clearly true in the story Host’s mentioning the war set in chain a series of events that led to the elevator being called, and almost everyone agreed that caused the elevator to appear. I’d bet that in any case where that chain sustained a causative, say if Host had asked Guest to call the elevator, then people would say that Host did cause the elevator to appear. Or at least a lot more than 50% of people would say that Host is a cause of the elevator appearing. Normal analysis of pre-emptive causation is not sensitive to this point.

I’m primarily interested in the second point. Say that c merely causes e if c causes e, but we cannot use a causative (opened, called, killed etc) to describe the relation between c and e. Then my hunch is that there is no mere pre-emptive causation. That is, if c merely causes e, then e is counterfactually dependent on c. We need a lot more than this case to show that is true, but that’s my current working hypothesis.

Thanks again to everyone who took part!