The Trolley Problem

For those of you who aren’t familiar with it, this, the trolley problem is a rather famous thought experiment that goes as follows:

Suppose there is an out-of-control trolley barreling down the tracks.  Ahead of it are five people who, for whatever reason, are unable to move out of its way, and will be killed when struck by the trolley.  You are standing next to a switch that can shunt the trolley onto a different track, where one person is will be killed.

Do you throw the switch?

There are multiple variations on this, of course, because academic psychologists and ethicists seem to have plenty of time to sit around and come up with devious twists on these sorts of things.  The most famous variation, though, changes the premise slightly.

Instead of standing next to a switch, you’re standing on a bridge over the track.  Next to you is a person who, if pushed off the bridge in front of the trolley, would stop it and save the other five people, but would be killed in the process.

Do you push him?

The answers to this problem have spurred all kinds of research and attempts at explanation, but they generally fall into two camps of ethics: deontology and utilitarianism.

Deontology is based around following moral rules.  Only actions which follow the moral rule can be good, and if you’re following the moral rule, than by definition your actions are good as long as your intent is also good.  The means are also an end, basically.

Utilitarianism is based around maximizing utility, which essentially boils down to maximizing well-being across the affected group of people.  This can be summed up by the Vulcan proverb, “The needs of the many outweigh the needs of the few.”

To help clarify, let’s apply this to the trolley problem.  For deontology, we might postulate a moral rule along the lines of “Take no action causing harm to another person.”  In following that rule, we wouldn’t throw the switch, and we wouldn’t push the person, because both would actively be causing harm.  By doing nothing, the trolley will kill the other five people, but we haven’t directly done any harm ourselves.

For utilitarianism, the reverse is true.  In both cases, we should take the suggested action: throw the switch and push the person.  After all, sacrificing one person to save five would definitely maximize the overall utility across the people involved.

Or, if it would resonate better with you, the 10 Commandments are definitely a deontological system, as are many of the rules laid out in the Epistles, while Jesus’ second commandment, “Love your neighbor as yourself” sure seems utilitarian to me.  (It’s still a rule, though, so hang onto that thought.)

(Despite the prior paragraph, for what I hope are obvious reasons, we’re going to focus on well-being here on Earth, rather than any given possible afterlife.)

Now, naturally, it’s not quite that clear cut in practice.  For example, in general, many more people will elect to throw the switch than elect to push the person, probably because of the degree of separation between their action and the death.  Putting a person’s child in one of the roles can also affect people’s choices.  In short, emotion can overrule reasoning.  Go figure.

And both approaches have their problems, of course. 

The most fundamental problem with deontology is defining and agreeing on what the moral rules are.  In our application to the trolley problem, we could just as easily have postulated a moral rule like “Always save the greatest number of lives”, and come up with a completely different result.  And, from a utilitarian perspective, simply following rules regardless of the consequences can seem simplistic, cruel, inflexible, or unreactive to changing circumstances.

The most fundamental problem with utilitarianism is how you define, quantify and measure well-being across the population of the problem you’re addressing.  “Kill one to save five” doesn’t even remotely cover all the aspects to consider.  What if the five are all nonagenarians with no living relatives, while the one is the sole breadwinner for a large family, for example?  When you take secondarily-affected people, like those in the family, into account, what will create the greatest utility then?  Don’t forget to take yourself into account, as well; you’re part of this as well.  Is there a precedent being set that could affect future people?  From a deontological perspective, utilitarianism is ill-defined, relative, convoluted, or prone to slippery slopes.

So why bring this up?

At the risk of generalizing a bit, I think this split describes some of the current political divisions we’re currently seeing.  To be clear, this is nowhere near a complete description of the problems, and it only addresses people actually acting according to principles, but it's an aspect.

First of all, let’s just look at the definitions of conservative and liberal (NB: these will not necessarily match to Republican/Democrat; we’ll get to that in a bit).

Conservative: Holding to traditional attitudes and values and cautious about change or innovation, typically in relation to politics or religion

Liberal: Open to new behavior or opinions and willing to discard traditional values; favorable to or respectful of individual rights and freedoms; (in a political context) favoring maximum individual liberty in political and social reform

I think the split is pretty clear here, right? 

Conservatives are clearly more deontological; their moral rule is essentially “Keep doing things the way they’ve been done.”  Do that, and you’re good. 

Meanwhile, equate “freedom” or “liberty” with “well-being” (and I think most people would do that to at least some extent; I’m equivocating here for a reason), and those definitions of liberal essentially become the definition of utilitarianism.  And that makes sense; if you want to reform systems and you’re not out to impose an autocracy or something, then your guiding star should be trying to find the systems that benefit the most people the most.  People may disagree about how different systems grade out, but that’s par for the course.

Now, as mentioned above, the Republican/Democrat split isn’t necessarily the same as a conservative/liberal split in terms of strict definitions, and I think most people recognize that.  Really, Republicans are “social conservatives”, while Democrats are “social liberals”, and I think that’s obvious enough, along with the ethical equivalents, that I’m not going to spell it out further here.  (If it’s not, and I’ve pulled the dreaded “this should be intuitively obvious” math prof shortcut on someone here, let me know, and I’ll be happy to elaborate.)

(Also, I’m going to talk here about generic Democrats and Republicans, rather than anyone in particular who may or may not be operating with good intentions, holding to their principles, etc.)

The classical construction of definitions on the economic side is that Republicans are “economic liberals” while Democrats are “economic conservatives”, but I think that’s not quite right in a couple important ways.

Just to make sure we’re all starting from the same point (if anyone’s going to argue with a premise, I want the premises to be clear): Republicans, in general, favor free markets with few regulations, while Democrats favor more heavily regulated free markets (with a couple special exceptions).  While “free markets with few regulations” does generally fit the “maximum individual liberty” definition of liberal, “more heavily regulated free markets” doesn’t really fit the definition of conservative; if anything, it’s just not as liberal.  (I’m avoiding the term “illiberal” due to negative connotations I don’t want to introduce here.)

I would also posit that a laissez-faire capitalistic system is, in fact, the traditional system here in the US, and has been since the beginning, which actually makes maintaining it as such the “conservative” approach.

I’m not arguing here that Republicans are “economic conservatives” as well as “social conservatives”, nor that Democrats are “economic liberals”; that’s clearly not the case.  They’re just both their own different and conflicting mixes of liberal and conservative/less liberal.

Now, I’m going to take advantage of all of the caveats I’ve built into this post so far to define a couple other terms to avoid looking like I’m about to shift the goalposts wildly.

Utilitarianism comes in two main flavors: rule utilitarianism, and act utilitarianism. 

Rule utilitarianism argues that by following rules that, on average, result in the most good, you are doing good even if the end result of a given action is bad.  (The classic example is telling a murderer where to find his targets because lying is bad.)  The key distinction between rule utilitarianism and deontology is why you’re following the rule.  Remember that in deontology, following the rule is good in and of itself, without any care for the net effect on the well-being of the affected people (remember we’ve ruled out rewards in heaven for those divinely-inspired rules).

The problem with rule utilitarianism is that it’s not very stable.  You either tend to rigidly follow the rules, which winds up looking a lot like deontology, or you build in so many exceptions that you wind up with act utilitarianism.

Act utilitarianism simply judges whether an act was good or not by its results.  If the net result was an increase in well-being for everyone, then it was a good act.  If not, it was a bad act.  Pretty straightforward (as a definition, anyways.  Not necessarily as a calculation.)

The problem with act utilitarianism should be obvious: it’s basically the definition of “the ends justify the means”.  For example, should a surgeon kill one healthy person to harvest organs to save five other people?  The basic structure of the problem is the same as the trolley problem, but I imagine that there are far fewer people who would say yes to this question than even those who would push the person off the bridge.

So why am I bringing these things up now?  To me, rather than thinking of economic approaches as liberal/conservative, I think it’s more useful to think of them in these ethical terms.

Republicans, I would argue, are “economic rule utilitarians”.  They follow rules like “Government should be as small as possible” and “Taxes should be as low as possible” in the belief that these rules maximize well-being overall.  And recall that I did point out before that, to some extent, most people will equate “individual liberty” with “well-being”, so there’s some justification for this approach.

Democrats, on the other hand, would seem to be “economic act utilitarians”.  They believe that, in some cases, increased regulation can actually do better at maximizing well-being/individual liberty.  Note that this doesn’t equate to “Government should be as big as possible” or that “Taxes should be as high as possible”, it equates to “Government and taxes should be set so as to implement the desired policies.”  Given that sometimes one person’s liberty can impact another’s in an adverse manner, there’s some justification for this approach as well.

Do I have a point to all of this?  A few, actually.

First, there’s something here I’ve been hinting at and dancing around, but which I’ve so far mostly avoided addressing: systems thinking.  I haven’t really wanted to touch on it here because I think it deserves its own post (next week, hopefully!), but the takeaway here is just that the amount of systems thinking these approaches take forms something of a spectrum.  The more you have to judge the effects of an act and quantify changes to well-being, the more systems thinking you have to do.  (And it quite possibly runs the opposite direction as well, where the more systems thinking you can/are able to do, the more it affects which ethical approach you take.)

Second, I’ve been making an effort to point out the flaws in each approach because it’s important to recognize where errors and biases can creep into our own thinking.  And that’s on top of the fact that very few people rigidly fall into only one category; most people change depending on circumstances, emotional effects, etc.  We’re all very good at doing what they want and then creating a justification afterwards.

And third, I’m not going to levy a judgment here on which approach is best.  (In fact, my goal is that, if you didn’t already know where I fall, you’d be unable to tell from how I described things.)  To do so would actually undermine the real point, which is this:

When it comes to ethics and morality, because we’re talking about “good” and “bad”, “right” and “wrong”, it’s very easy to demonize the other “side” just because they don’t think about things the same way you do.  That’s part of why our politics are so fractious right now: instead of trying to understand the other side, even if you don’t agree with them, we just write them off as obtuse, clueless, or even evil, just because they don’t approach life the way we do.

But that’s just the problem.  Someone whose entire framework of thinking and judging is different from yours is inherently difficult to understand, because they’re making decisions based on different priorities and philosophies than you.  We want to look at things through our own frameworks, and sometimes positions and decisions just won’t make sense to us.  And things like the fundamental attribution error make us more likely to think that this is due to something wrong with the other person.

I can’t tell you whether or not to throw the switch; all I can do is trust you to make the decision you feel is right, even if it’s not the choice I would make.