Oct 16th, 2008 at 3:11 pm
The Moral Sense Test
If you’ve ever taken a philosophy class, you know that a lot of reference gets made to our moral intuitions. This, of course, raises some empirical issues about what “our” intuitions are. The Moral Sense Test” is part of a project to address those issues. Check it out.
40 Responses to “The Moral Sense Test”
Register
RSSLogin
Advertisement
Visit Our Affiliated Sites
Books By Matthew Yglesias
Topic CloudView topics sorted by popularity
Abortion
Afghanistan
Basketball
Blanche Lincoln
Budget
Cars
China
climate
Congress
Crime
DC
Demographics
Economics
Economy
Energy
EU
Finance
Football
Gender
Health Care
History
Inequality
Infrastructure
International Relations
Iran
Israel
Media
Monetary Policy
National Security
NBA
Pakistan
Palestine
planning
Political Reform
Public Health
Public Opinion
Race
Regulation
Sarah Palin
Sports
Stimulus
taxes
Technology
Trade
transportation
Featured
Subscribe to the Progress Report
Contact Matthew Yglesias
Use this form to contact blog author Matthew Yglesias.
ArchivesBlog RollAbout Matt YglesiasContact Me
Donate
October 16th, 2008 at 3:20 pm
Oops, forgot the link:
http://vegasblog.latimes.com/vegas/2008/10/politics-vegas.html
October 16th, 2008 at 3:20 pm
In the spirit of “moral sense,” is a “Topless Sarah Palin Lookalike Contest” sexist?
October 16th, 2008 at 3:25 pm
Here is a background article on “experimental philosophy”:
http://chronicle.com/free/v55/i06/06b00901.htm
October 16th, 2008 at 3:32 pm
I answered the question that it was Extreme Moral to smother designers of moral sense tests.
But I forgot to add that I’d put lavender on the pillow.
That way my response would pass the sniff test.
October 16th, 2008 at 3:35 pm
Total, unmitigated nonsense. That was neither experimental nor philosophy.
October 16th, 2008 at 3:38 pm
I think the “of course” in Matt’s post is anything but the kind. I use intuitions, and I couldn’t care less about empirical research into intuitions. Intuitions are only valuable insofar as they are carefully considered and due to the role they play in rational deliberation aimed at reflective equilibrium.
October 16th, 2008 at 3:43 pm
I hate these kinds of tests. They’re always “if you knew for sure X would happen, would you do Y?”
It’s hard NOT to be a consequentialist when you take these tests, because they guarantee the consequences of your actions! But as outcomes become more uncertain, one should be LESS of a consequentialist.
I mean, we all think it’s wonderfully moral to cut open a living person…to remove a diseased appendix and save their life. Or to remove a healthy kidney and save someone else’s life. So, when it comes to medical surgery we’re all consequentialists, because it’s all very clear.
But if the question is whether I should push a guy off a boat and definitely drown him in order to get the boat fast enough to save 5 drowning people…eh, there’s too many unknown variables to justify doing something awful. Maybe those people are better swimmers than you think, or your passenger is a better swimmer, or if he lies down in the back he’ll be light enough or whatever. I’m not willing to be a consequentialist when the outcome is dependent on so many unknowable variables. Better to do your best and hope to create a solution that doesn’t involve killing someone.
Isn’t there a moral integrity in recognizing that someone has presented you with a scenario that is impossible? I mean, of course we’d all kill Hitler when he was 6 years old if we knew he was Hitler. But hopefully we’re all smart enough to realize you can’t reliably predict Hitlerism in a 6 year old, so certain solutions should be off the table no matter how hard someone tries to convince you that no, really, this 6 year old is Hitler.
I hate these quizzes.
October 16th, 2008 at 3:46 pm
There is still way too little context in the ‘moral sense’ questions. If I’m in a situation where I get to decide if I’m going to throw a passenger over board to drown so I can save five others, it is going to depend on who the passenger is and how he got there. A total stranger who just happens to be there because somebody else put him there is going to be easier to toss overboard than somebody I like and who I invited.
I’d evaluate others by the same or at least similar standards I’d use to decide myself.
October 16th, 2008 at 3:47 pm
“GENEVA (AP) — Iran and Iraq have agreed to work together in tracing thousands of people still missing after the war between the countries in the 1980s, the International Committee of the Red Cross said Thursday.
The neighboring Middle Eastern countries have established how they will gather and share information about the missing and hand over any remains uncovered, the Red Cross said.
The agreement — signed Thursday by representatives from Iran, Iraq and the Red Cross at the neutral organization’s headquarters in Geneva — represents “an important step toward easing the heavy burden of thousands of bereaved Iraqi and Iranian families,” said Beatrice Megevand-Roggo, the ICRC’s head of operations for the Middle East and North Africa.
It was the first time Iran and Iraq signed a direct agreement to tackle the problem together. Previously they each dealt separately with the Red Cross, ICRC spokeswoman Dorothea Krimitsas said. Iranian and Iraqi officials in Geneva did not immediately return calls.”
No doubt this is a new phase in the Neocons plan to take over the planet. Too bad Saddam Hussein isn’t still in charge.
October 16th, 2008 at 3:49 pm
I agree with anonymiss @ 3:43pm.
The “only punishment is a fine” questions annoy me too.
And you don’t even get a payoff from the survey of some description of your “results.”
October 16th, 2008 at 3:50 pm
If you need to push somebody off your boat to get it to go faster, it’s probably not going to go very fast even after pushing that person off the boat.
October 16th, 2008 at 3:51 pm
The question about the scientist and test subjects and vials of poison and vaccine doesn’t make sense – the question asks about the scientist taking a course of action that will certainly result in one fatality, but there remains an unmentioned alternative that only makes a death 50% likely. So, fail.
October 16th, 2008 at 4:05 pm
Do we get to see what the results are?
October 16th, 2008 at 4:28 pm
It’s hard NOT to be a consequentialist when you take these tests, because they guarantee the consequences of your actions! But as outcomes become more uncertain, one should be LESS of a consequentialist
Okay, let’s try this out.
There’s a train heading down a track, approaching a mechanism switching between two tracks. 100 people are tied to one track and on the other is tied a single person. Before you is a lever which controls the switching mechanism…
1) If you leave the lever as it is, the train will run over 100 people. If you pull the lever, then the train will switch tracks and run over the single person. What should you do?
2) Unfortunately, the mechanism is run down and doesn’t work quite correctly. If you leave the lever as it is, there’s a 90% chance the train will continue down its current track and run over 100 people and a 10% chance it will switch tracks and only run over the single person. If you pull the lever, then there’s a 90% chance it will switch tracks and a 10% chance it will remain on its current track. What should you do?
3) Unfortunately, the mechanism is run down and doesn’t work quite correctly. If you leave the lever as it is, there’s a 51% chance the train will continue down its current track and run over 100 people and a 49% chance it will switch tracks and only run over the single person. If you pull the lever, then there’s a 51% chance it will switch tracks and a 49% chance it will remain on its current track. What should you do?
4) Unfortunately, the mechanism is run down and doesn’t work quite correctly. Not being a train track switching mechanism expert, you have no idea what the chances are of the train switching tracks and how they depend on the lever being pulled. But you can tell that if you pull the lever, it’s more likely to switch tracks than if you don’t pull the lever. What should you do?
October 16th, 2008 at 4:52 pm
I agree with anonymiss. I was utterly consequentialist because all the questions are posed as certainties. It is specifically because we can’t have certainties that I would not choose those options in a real situation.
When you have god-like omniscience, it isn’t necessarily immoral to usurp other powers traditionally allotted to God.
The probative value of the test is virtually nil.
October 16th, 2008 at 4:56 pm
So you’re all in agreement that as you go from 1-4 in my scenarios you become less consequentialist? Personally, I think you should pull the lever in scenario 4 for consequentialist reasons, even though the outcome is highly uncertain and, in fact, the amount of uncertainty is uncertain.
October 16th, 2008 at 4:59 pm
Nick,
That situation isn’t what we’re complaining about. In that case, I would do what I thought would result in the fewest deaths.
Here is the real world version of your problem.
There is a train headed for a three way switch. One track has 100 workers on it who are ignoring the train, one has one worker on it ignoring the train and the other has none.
It looks like the switch is set so the train will hit the 100 workers. You think you can switch it to hit the one guy. There is some unknown chance that the switch is automated and will route the train onto the empty track provided no one messes with it. What do you do?
October 16th, 2008 at 5:10 pm
Nick,
That’s not really a good scenario. If you do nothing, 100 people die for sure. Therefore any action that might have any chance at all at changing the track is best (51/49 is really irrelevant).
Njorl is off too, for the same reason, but close. The relevant scenario is something like: if you do nothing, it’s going to hit one person. You’re pretty sure you can redirect it to an empty track, but there’s a small (<10%) chance it might go onto a third track with ten people (if it could possibly hit 100 people vs 1, I’m pretty sure you’d never do it in any circumstance).
October 16th, 2008 at 5:24 pm
It’s true that I kind of gamed my scenarios but 51/49 really is more uncertain than 90/10 so the claim that uncertainty (full-stop) is the problem strikes me as off.
In Njorl’s example, if p is the chance that the switch is automated then you can expect to save lives as long as p is less than 99%. In real life, one might have some sense of what p is likely to be and I feel that the correct course of action would again be to “do what I thought would result in the fewest deaths.” Given no information about p, I’m not quite sure…
October 16th, 2008 at 5:34 pm
The relevant scenario is something like: if you do nothing, it’s going to hit one person. You’re pretty sure you can redirect it to an empty track, but there’s a small (<10%) chance it might go onto a third track with ten people
As written, I think I would definitely pull the lever. Heck, even if you changed it to “<=10%”, I’d still do it. Even with no other information about this “small chance”, you can always expect to save lives by pulling the lever.
(if it could possibly hit 100 people vs 1, I’m pretty sure you’d never do it in any circumstance).
Well, you expect to save lives if the “small chance” is less than 1%. Though maybe we need to have an increasing marginal utility of human lives?
October 16th, 2008 at 5:48 pm
But if the question is whether I should push a guy off a boat and definitely drown him in order to get the boat fast enough to save 5 drowning people…eh, there’s too many unknown variables to justify doing something awful. Maybe those people are better swimmers than you think, or your passenger is a better swimmer, or if he lies down in the back he’ll be light enough or whatever. I’m not willing to be a consequentialist when the outcome is dependent on so many unknowable variables.
How many is “so many?” You list three. It would appear there are at least as many “unknown variables” in other situations in which most people would probably support taking an action that causes serious, demonstrable harm for some uncertain greater good (e.g., punishing people convicted of a crime, bombing enemy targets during a war). Are we to understand you oppose all such acts?
October 16th, 2008 at 7:14 pm
Moral intuition huh.
My intuition tells me that it is immoral to ask people to contemplate under what circumstances they are willing to murder someone.
October 16th, 2008 at 8:55 pm
I was getting into the whole quiz thing until the end. Then they lost me. There is a huge difference between triage and knowing killing someone who you know will die soon to prevent the loss of life, the survey did not give you the option of ‘it depends’ in the end. The implied equivocation of those two things at the end invalidates the whole thing. If you agree with the general concepts of OSHA, law enforcement, and fire protection, then there are no reasonable answers to the last several questions. In that general sense, I also agree with Guy P.
October 16th, 2008 at 9:28 pm
I had trouble because I kept thinking that people don’t always make their decisions based on what they consider to be the morally correct answer.
October 16th, 2008 at 10:27 pm
Big typo alert… I meant to say ‘knowingly killing someone who you know will not die…’
Big difference…
October 16th, 2008 at 10:52 pm
This test was totally bogus. I gave it up after 4-5 almost identical questions. Am I willing to sacrifice the one-or the few to save the many ?
In real life you are never sure that the one will die,neither have you any guarantee that the many will be saved.
I hope I would be willing to risk my own life to save people I love.spec my children,but I’m not willing to die for some patriotic BS, and certainly not for Halliburton or ExxonMobile.
October 16th, 2008 at 11:53 pm
See, crap like that is why I always carry an extra life jacket and oxygen tank with me wherever I go.
October 17th, 2008 at 4:58 am
anonymiss @7 is largely right I think. Sometimes hypotheticals are useful, sometimes they are not. In this situation, it is not very helpful or even telling of how we ought to live our lives. These scenarios are assuming that certainty about the results of our actions that doesn’t exist.
One thing I object to though:it’s not true that as certainty becomes less, consequentialism should become less. Perhaps if you’re an act utilitarian this is true, but if you follow a more indirect system of utilitarianism this would encourage us to NOT push the baby out of the window, to NOT kill the patient to get at their organs: following these rules/dispositions increases overall utility.
October 17th, 2008 at 8:22 am
That’s the point. In reality, you never know “p”. You don’t have godlike omniscience so you are not entitled to godlike rights of life and death.
October 17th, 2008 at 9:50 am
Yes, but the point of my final example was that sometimes you don’t need to know p. In Adam’s final example, again, you don’t know p. You just know one fact about p, namely that it is less than 10%. This is pretty weak stuff. It’s not even a distribution on p or an idea of what a distribution on p might look like. And yet that’s enough (in my opinion) to still make a decision based on consequentialist reasoning.
Yes, in reality, you never know p. But I think sometimes you might have a good enough idea about p to be justified in making some of these types of decisions. Where to draw the line, though, I’m not quite sure.
February 23rd, 2009 at 4:06 pm
You are the greatest!
==
http://street-wear.org/jenniferfarriswordpresscom/
March 1st, 2009 at 5:58 am
viagra
I want to say – thank you for this!
March 12th, 2009 at 11:44 pm
Great site. Good info
March 14th, 2009 at 5:34 am
Very interesting site. Hope it will always be alive!
xanax
March 22nd, 2009 at 6:23 am
tramadol
Very interesting site. Hope it will always be alive!
March 22nd, 2009 at 10:33 am
buy viagra online
It is the coolest site,keep so!
April 3rd, 2009 at 4:19 am
Incredible site!
cheap brand pfizer viagra
April 9th, 2009 at 6:48 am
It is the coolest site,keep so! viagra