Shoving a fat man under the trolley: abstraction or fiasco?

Konishiki: Less easy to push than the Trolley Problem assumes

A commenter objects to my having fun with a moral conundrum much mulled over by psychologists, philosophers, and behavioral economists: should you instantly decide to push a fat man under a runaway trolley to save the lives of five other people? One standard formulation is:

As before, a trolley is hurtling down a track towards five people. You are on a bridge under which it will pass, and you can stop it by dropping a heavy weight in front of it. As it happens, there is a very fat man next to you – your only way to stop the trolley is to push him over the bridge and onto the track, killing him to save five. Should you proceed?

I noted that very fat men tend to be hard to push.

My commenter scoffs:

Normally I object to the learned autism of most contemporary moral philosophy, but in this case the problem is clearly your and many of your commentators' learned retardation. Inability to abstract shouldn't be a point of pride. 

Here's the beginning of another thought experiment. To explain Newton's Laws, a physicist begins "imagine a frictionless world composed of a single billiard ball sitting motionless on an infinite plane. The ball is set into motion at a constant velocity of 10 m/s . . ." Would you have your nit-picking reaction then (why 10 m/s? Is the ball stripes or solids? I've never seen an infinite plane!)? Of course not, because you and every other 3-digit IQer would understand that it is a simplification intended to show clearly some essential features (in this case, why an object set into motion stays in motion along its vector). This, despite the fact that the frictionless plane is more alien to your experience than any version of the trolley problem. 

I chose Newton because his example is very influential. Like Newton, many scientists begin with simplified, unrealistic foundations, then try to build atop that more and more complexity until we have a reasonably accurate predictive model. It doesn't always work, but that is the fault of the scientist, not some flaw in reductionism. The Trolley Problem needs to be understood in this light. 

The Trolley Problem is obviously a simplified situation. You aren't the first person to realize that. Everyone knows it, philosophers and psychologists included. The simplifications are necessary, because even small details can change people's responses dramatically. Wright and Greene talk about this in their bloggingheads talk -- changing the scenario from one where you push the fat man to one where you drop him onto the tracks through a trap door by pulling a lever drives a lot of people into the sacrifice-fatty camp. But, the simplifications are ok because were left with a scenario that has, for most, a strong intuitive tug and that can be easily tweaked get different reactions. That last part is important, because by tweaking we can begin to figure out the (sometimes surprising) things that can affect moral cognition. 

Moral psychologists are interested in the responses for their own sake, and moral philosophers are interested for exactly the reasons Greene talks about -- if our moral intuition can be tossed hither and tither by something so silly as whether we touch the fat man or use a trap door, should we trust our gut reaction for any moral situation? Maybe we still should, but it's deep question, and one that gives only limited support to utilitarianism.

Before the trolley problem was invented, science fiction writers devoted much effort to crafting more airtight scenarios, such as Tom Godwin's 1954 classic The Cold Equations. Space travel scenarios involving shortages of fuel to boost everybody on board home have a couple of advantages over the trolley problem:

- Plenty of time to think things through logically.

- No outside resources available to offer easy win-win solutions. 

But I like the unintentionally comic Trolley Problem precisely because, when you stop to think about it, it raises what ought to be an important issue in moral philosophy: the "Law of" (or, better, tendency toward) Unintended Consequences. Even within the narrow moral imagination of the utilitarian philosophy, there are good reasons why it's usually a bad idea to decide to try to kill somebody with your bare hands even if it seems like the logical thing to do after one or two seconds of reflection. Having a general emotional prejudice against starting a lethal brawl because it seemed like a good idea at the time may not be as illogical as utilitarians assume.

If you try to shove a fat man to his death, will everything ensue exactly as the problem assumes? Very likely not. Instead, you are more likely to wind up with a lot of grappling, grunting, and sweating. Maybe in your struggle, both of you will eventually fall to your deaths, after the trolley has shot past, thus winding up with seven deaths, not five.

This helps explain why people tend to be less averse to less messy, more mechanistic -- indeed, more deterministic -- version of the Trolley Problem, like throwing a switch or opening a trapdoor that sacrifices one to save five. Among much else, being mechanical, these methods seem more likely to work than getting into a shoving match with somebody larger than yourself.

The messiness of trying to throw a fat man off a bridge shouldn't be conceived of as merely a unique, isolated issue in moral philosophy. It has broader implications for morality in the real world. 

Say there is an unlovable dictator who, after a long nasty career of attacking other countries, has settled down to more or less minding his own business. Should we invade his country to bring his oppressed people democracy? 

What could possibly go wrong?