Reply to Bentham's Substack

This is the second article I will be posting in an Exchange with Bentham's Substack. I will be replying to his opening article, seen here.  Let's get started right away.

Argument 1

Almost all of the considerations Matthew adduces in his blog article against deontology are paradox's or strange/unintuitive results in general. Many are parasitic on intuitions that are not independent of each-other. Such things were only one part of my case against utilitarianism on my blog alongside the epistemic, theoretical, explanatory and practical arguments I raised. So, I suppose the first thing to point out, is that a case which rests nigh solely on the same intuitions, in regards to matters where, perhaps, we have reason to distrust our intuitions especially about bizarre hypotheticals. Is, if not an inherently weak case, one we have immediate antecedent reason to be skeptical of. One which I doubt can have the "overwhelming force" Matthew claims.

Anyways, first intuition pump.

Suppose a person puts a bomb in a park for malevolent reasons. Then, they realize that they’ve done something immoral, and they decide to take the bombs out of the park.  However, while they’re in the process, they realize that there are two other bombs planted by other people. They can either diffuse their own bomb, or the other two bombs. Each bomb will kill one person.

It seems very obvious that they shouldn’t diffuse their own bomb—they should instead diffuse the two others. But this is troubling—on the deontologist’s account, this is hard to make sense of. When choosing between diffusing their own bomb or two others, they are directly making a choice between making things such that they violate two people’s rights or violate another person’s rights.

This is one where I'm confused what the force of this is supposed to be. You have a duty to diffuse any of the bomb's you are able to diffuse, and I don't even think that's for consequentialist reasons. Suppose there is a bomb in your vicinity, say one of the one's that was planted, then that fact confers obligation on you to diffuse it, to ignore it would be immoral, as it violates your greatest obligation, respect for humanity as such. If you are not able to diffuse all the bombs, and yours goes off, then that's just to say you aren't responsible, since you did your moral duty by diffusing the other bombs. Obviously, you are responsible insofar as you planted the bomb, but that's just to say the separate act of you planting it was immoral, but in respect of the act of diffusing the other bombs you are not responsible for your failure to diffuse yours given that you couldn't diffuse all three.

Argument 2

Suppose you’re deciding whether or not to kill one person to prevent two killings. The deontologists hold that you shouldn’t. However, it can be shown that a third party should hope that you do. To illustrate this, suppose that a third party is deciding between you killing one person to prevent the two killings, or you simply joining the killing and killing one indiscriminately. Surely, they should prefer you kill one to prevent two killings to you killing one indiscriminately.

Thus, if you killed one indiscriminately, that would be no worse than killing one to prevent two killings, from the standpoint of a third party. But a third party should prefer you killing one indiscriminately to two other people each killing one indiscriminately. Therefore, by transitivity, they should prefer you kill one to prevent two killings to the two killings happening—thus they should prefer you kill one to prevent two. To see this let’s call you killing one indiscriminately YKOI, you killing one to prevent two killings YKOTPTK, and the two killings happening TKH.

YKOTPTK < YKOI < TKH. < represents being preferable. Thus, the deontologist should want you to do the wrong thing sometimes—a perfectly moral third party should hope you do the wrong thing.

(Worth noting that it is absolutely permissible, even on deontology, to kill 1 to save 2 if it is self-defense, or protecting them from a killer, but we'll suppose henceforth that the 1 is innocent).

There is a couple ways to interpret what is being said here, so it'll perhaps be helpful to disambiguate them. Insofar as this argument is attempting to pick out an incoherence between how an agent should act vs what they should prefer to happen given the truth of deontology. Well, as a deontologist, I'm not committed to a position about what people "should hope" or "should prefer". Deontology is, exclusively, a theory about what makes actions right or wrong, my position, qua deontologist, is silent on the matter of what events we should prefer occur from the third person. Perhaps one way in which it might be relevant, is as follows; having a preference for the "killing 1 to prevent 2 killings" scenario, is more conducive for agents to cultivate respect for humanity(rational nature) as such, and therefore more virtuous dispositions. But if that's the case, it's unclear what the incoherence or tension in the position is supposed to be. Both not killing to save 2 and preferring 1 being killed to 2 being killed, are both unified under the maxim of respecting for humanity as such. At worst, it is a slightly weird entailment, but slightly weird entailments are hardly unique to deontology.

Another way of understanding it, is that when we calculate the total value of each state of affairs from the third person. The state of affairs involving one person being killed to prevent two deaths is all-things-considered better than the state of affairs wherein one killing happens. So, for this reason you ought to prefer the killing to save 5 to happen. Yet, the deontologist thinks you shouldn't kill one to prevent two killings. If this is all that's being said, then what's being said doesn't seem to be much more than a restatement of deontological commitments. There is no "paradox", just different questions being asked, "is the action right?" and "is the state of affairs more valuable?". Deontologists do not think an actions rightness depends on the total value of the states of affairs, but on other considerations, so it isn't the least bit surprising that the right-making features of actions and the total value of states of affairs are judged differently, and are not reducible to each-other. 

Additionally, I think what little intuitive force this has going for it, is counterbalanced, by my lights at least, when we then look at intuitive right-making features of actions that seem to be over and above the total goodness of the value of states of affairs. Such as; the intentions and character-traits the agent has, the maxims the agent acted on, whether the act involves violating autonomy/self-determination, humanity or other things we take to be important. etc.

Argument 3

Imagine one thinks that it’s wrong to flip the switch in the trolley problem. While we’ll first apply this scenario to the trolley problem, it generalizes to wider deontological commitments. The question is, suppose one accidentally flips the switch. Should they flip it back?

It seems that the answer is obviously not. After all, now there’s a trolley barreling toward one person. If you flip the switch it will kill five people. In this case, you should obviously not flip the switch.

However, there’s a very plausible principle that says that if an action is wrong to do, then if you do it, you should undo it. Deontologists thus have to reject this principle. They have to think that actions are wrong, but you shouldn’t undo them. More precisely, the principle says the following.

Reversal Principle: if an action is wrong to do, then if you do it, and you have the option to undo it before it has had any effect on anyone, you should undo it.

This problem can also apply to the footbridge case. Suppose you push the guy. Should you pull him back up? No—if you come across a person who is going to stop a train from killing five, you obviously shouldn’t preemptively save him by lifting him up, costing five lives.

A few things can be said here. I feel myself under no initial pressure to accept the reversal principle. There seems to be no reason to think that such a principle is a necessary truth that governs all morally salient actions. The only necessary universal moral truth I'm committed to, is that an action is wrong iff it violates the categorical imperatives, and there seems to be no reason to think reversal actions cannot fall under that category. At best what looks plausible to me is the following principle;

Weak Reversal Principle: if an action is wrong to do, then if you do it, and you have the option to undo it before it has had any effect on anyone, that fact is some defeasible reason to think you should undo it.

I think this principle is clearly far more intuitive than the strong one, and that any intuitive plausibility the strong principle has going for it is parasitic on the intuitive plausibility of the weak one. Suppose it were the case that undoing some wrong action would involve violating someone's autonomy, murder, cultivating really terrible character-traits and dispositions, it seems like considerations like these, can clearly outweigh the fact that the act you are undoing is a wrong act when deliberating whether the reversal act is right to do. Ironically, the only reason I can think of to deny this, would be if one is a consequentialist, which I am not.

To answer the cases then; I think you should probably flip the trolley switch so there isn't an immoral act to undo. No, you shouldn't save the fatman. No, you shouldn't remove the 5 people's organs to save the one. The fact that you are reversing a wrong action is some defeasible reason to do it, and that in many cases all-else equal you should undo wrong actions, but I reject that this fact is overriding in these cases.

Another point worth making is, depending on how reversal is construed. There either seems to be a tu quoque objection here, or we should be skeptical of our intuition in these cases. Does reversal mean the action is already done, so you go back and undo the effects of your action? Then it looks like a consequentialist should be committed to thinking there are cases where you shouldn't reverse wrong actions too. Suppose you harvested 1 person's organ's to save 5, then you realize one of the 5 you saved is a depraved serial killer, and will kill, rape, and torture many more victims thanks to your action. So, all-things-considered, the action was wrong on consequentialism. Suppose you can undo it, but the act of doing so, has massive causal ramifications, inscrutable to you, which would eventually lead to the birth of Super Hitler, who will initiate a global nuclear holocaust. Strange scenario, but there you have it, a case where undoing a wrong action would be wrong on consequentialism. So, by the reversal principle consequentialism is absurd! Another way of interpreting reversal is undoing the action so that it doesn't even happen, perhaps by way of a time machine. But I simply don't put much stock in our intuitions about abstract scenario's like what we should do given a time machine, which may not be even possible.

Argument 4

This one is Huemer's paradox of deontology, and is the first argument that I feel the force of. Read Huemer's paper for more context. It starts by putting forth two principles.

Individuation Independence: Whether some behavior is morally permissible cannot depend upon whether that behavior constitutes a single action or more than one action.

And;

Two Wrongs: If it is wrong to do A, and it is wrong to do B given that one does A, then it is wrong to do both A and B.

Now, here is the case which generates the paradox;

Torture Transfer: Mary works at a prison where prisoners are being unjustly tortured. She finds two prisoners, A and B, each strapped to a device that inflicts pain on them by passing an electric current through their bodies. Mary cannot stop the torture completely; however, there are two dials, each connected to both of the machines, used to control the electric current and hence the level of pain inflicted on the prisoners. Oddly enough, the first dial functions like this: if it is turned up, prisoner A’s electric current will be increased, but this will cause prisoner B’s current to be reduced by twice as much. The second dial has the opposite effect: if turned up, it will increase B’s torture level while lowering A’s torture level by twice as much. Knowing all this, Mary turns the first dial, immediately followed by the second, bringing about a net reduction in both prisoners’ suffering. Did Mary act wrongly?

There is a general issue that I think is shared with some of these 'paradoxes' and it is a problem shared with the previous 'paradox' as well. It only arises if we assume principles of judging the moral features of actions, that seem plausible at first blush, but are ones I'm simply not committed to, and are not part of my normative theory. In general, if you propose a principle that is supposed to be a necessary truth regarding the moral features of all actions, and you have no independent argument for it, apart from brute intuition. My response is, naturally, to practice a high degree of skepticism that these principles are, at least universally, true.

With that noted, I'm skeptical of the 'Individuation Independence' principle. There seems to be no strong reason to think, that if two or more actions are individually wrong, it thereby follows that a conjunction which contains both actions will thereby also be wrong. Just as a set of bricks can have various properties that each brick lacks (different weights, size etc.), it doesn't seem completely absurd that a conjunction of the same actions could have right-making properties that those actions taken alone lack (and that this couldn't be one such example). There'd need to be a further argument given to think ethical action is a case where the properties of one, necessarily transfers across a conjunction.

But let's say you really, really want to keep these principles. Then, given those principles, deontology commits you to thinking Mary did act wrongly. This is out of accord with folk intuition. But perhaps we can put pressure on the intuitions by;

1) Pointing out that the scenario is one that is extremely bizarre; it involves having two dials that can increase and decrease suffering levels on a whim, but the dials also work in weird ways that prevent you from reducing one's torture without increasing another's. This is not a scenario anything like what anyone has, and most likely ever will encounter. Furthermore, accurate reasoning about such off the wall hypothetical's that have no connection to our ordinary lives, is highly unlikely to have survival or reproductive benefits so we wouldn't expect natural selection to select for true beliefs about these hypotheticals.
2) Focusing on the kinds of beliefs that lead us to judge Mary's action as right and putting pressure on those. For example, the action Mary commits' are one's that involve violating consent, but the result is good for both parties. But plausibly, we shouldn't force someone to stop smoking, even if it's good for them. We shouldn't promote paternalism for the good of everyone. So we shouldn't, generally, violate consent for a rationally informed agent to bring about that which is good for them. This seems especially true in cases where violating consent involves causing significant pain.

TL;DR. This argument is strong but not insuperable. If you are a deontologist, and your intuition of the truth of the principles is stronger than the intuition that Mary acted wrongly, reject that she acted wrongly, if your intuition that she acted wrongly is stronger, reject at least one of the principles.

Argument 5

The argument is roughly as follows—every time a person drives their car or moves, they affect whether a vast number of people exist. If you talk to someone, that delays when they next have sex by a bit, which changes the identity of the future person. Thus, each of us causes millions of future people’s identities to change.

This means that each of us causes lots of extra murderers to be born, and prevents many from being born. While the consequences balance out in expectation, every time we drive, we are both causing and preventing many murders. On deontology, an action that has a 50% chance of causing an extra death, a 50% chance of preventing an extra death, and gives one trivial benefit is wrong—but this is what happens every time one drives.

One way of pressing the argument is to imagine the following. Each time you flip a coin, there’s a 50% chance it will save someone’s life, a 50% chance it will kill someone, and it will certainly give you five dollars. This seems analogous to going to the store, for trivial benefits—you might cause a death, you might save someone, and you definitely get a trivial reward.

I just don't get this one. Not only do I think the entire notion that this is an objection to deontology is confused.  I think it's, if anything, an argument against consequentialism.  

I think actions that involve directly violating certain rights is wrong,  and intentionally producing outcomes that leads to people being harmed and having their rights violated is also wrong, ceteris paribus. What I completely reject, is that it is wrong to cause rights violations and people to be harmed as a long-term, unintended, and unforeseen outcome. I don't think you're responsible in those cases. I reject the consequentialist assumption that long-term unforeseen, and unintended, identity-affecting outcomes are equally morally relevant in regard to the rightness or wrongness of actions as directly intended, or foreseen outcomes, as should any deontologist worth their salt. In fact, I go so far as to reject those kinds of consequences as even having any bearing, whatsoever, on what makes an action right or wrong. Only other things do, such as known consequences, prima facie duties, the maxims incorporated into the action, the character-based dispositions leading to the act etc. The argument thus, fails to get off the ground.  

The paper Matthew cites does respond to something close enough to this objection, by way of responding to Lenman, who also draws a distinction between foreseen vs unforeseen consequences in his paper. I think the response leaves much to be desired. 

What does it mean for some outcome to be foreseeable (to an ideally conscientious agent)? We can't say that some outcome is foreseeable just in case you were (in principle) in a position to know that the outcome would follow from your choice of that act. There is certainly no plausibility in the suggestion that the potential for some action to bring about some consequence can provide a reason against performance of that action only if the agent is (in principle) in a position to know that the consequence will surely obtain if the action is performed. Any such view would permit reckless endangerment. 

I take 'foreseeable' to mean the agent can reasonably expect a given outcome to occur given their intentions, background evidence, and beliefs (and other mental states) which are directly accessible to them. If there is nothing accessible to the agent which would permit the agent to know that the outcome is something that will obtain as a result from the act, then the agent cannot be responsible for the outcome which obtains. It's just begging the question to call this reckless endangerment. It is only reckless endangerment if the agent is blameworthy for what occurs, but nothing like this has been shown. It seems to me utterly implausible that the mere potential for some act to generate bad consequences confers reasons on the agent, if such reasons are not accessible to the agent.

So, for long-term consequences of ordinary acts. On my view the agent isn't blameworthy for 3 reasons. 

1) It wasn't intended
2) No one was directly caused to be harmed or used as mere means in the process of the act. 
3) The agent performing the act did not foresee any particular bad outcome that was to occur as a result of the act.

This also refutes Mathew's example. Your doing some course of action is not the direct nor proximal cause of an innocent family dying in a car accident, nor was it intended or known that such an outcome would occur. Further, importantly, your act could have still been done, and if some asshole didn't drink and drive, or if the driver wasn't distracted ,or if they weren't speeding etc. the death wouldn't have happened. On the other hand, your action of flipping the coin and it's landing on the wrong side does directly cause someone to be killed, and you knew it would. The act, by stipulation, couldn't have been done without leading to the death. Finally, it is false that in ordinary actions, the agent's epistemic situation is that there is a 50% chance some particular bad outcome obtains.  Instead, particular long-term identity-affecting outcomes of ordinary actions are completely inscrutable to the agent. No given outcome is expected. 

The paralysis argument against deontology is a failure, it looks to me. But I think a decent argument can be salvaged when it is properly understood as an argument against consequentialism. Consequentialists should factor in these unforeseeable possibilities when deliberating what action they should do, since on consequentialism the distinction between foreseeable vs unforeseeable consequences is not morally relevant. So consequentialists should not risk the unforeseen consequences caused by ordinary actions. Matthew hints at a response by saying that 'the consequences balance out in expectation'. However, this fails for the same reason Shelly Kagan's "canceling out" response to the epistemic objection fails, see my previous article for that. 

Argument 6

1. If deontology is true, then you shouldn’t kill one person to prevent two killings.
2. If you shouldn’t kill one person to prevent two killings, then, all else equal, you should prevent another person from killing one person to prevent two killings.
3. All else equal, you should not prevent another person from killing one person to prevent two killings
Therefore, deontology is false

There is a general problem with this argument. Any non-question begging reason (that is, reason that assumes the falsity of deontology) you can give to think 3 is true, is a reason to think 2 is false by the deontologists lights. Indeed, all the argument's that Matthew gives in favor of 3, are all, as a deontologist, I think pretty good reasons to think 2 is false. So it's hard to see how this argument is particularly effective against deontology, just reject 2 for the same reasons Matthew gives to accept 3.

My response is therefore to reject 2, because of the reasons Matthew gives, and roughly the same reasons I reject the strong reversal principle, I think there is a defeasible reason to prevent wrong actions it just isn't universally overriding. This should be sufficient to preserve our intuitions and avoid this argument. But let's quickly look at Matthew's justification for 2.

The idea here is pretty simple. It seems really obvious that you should prevent people from doing wrong things if you can at no personal cost. In fact, as this paper. which started the entire idea of this worry for deontology in my mind notes, this produces a strange result when it comes to deterrence. Presumably, if we think that killing one to save five is wrong, we’ll think that it’s a good thing that laws against murder prevent that. But if we think that third parties have no reason to prevent killing one to save five, then deterrence is not a reason to ban deontic rights violations with good outcomes.

If you have no reason to prevent organ harvesting, then it isn’t wrong. One should prevent wrongdoing, if all else is held equal.

The reasons he gives are;

1. "It's obvious", but this is ineffective, I don't think it's obvious. You have, at most, a defeasible reason, but I think we have good reasons to believe it is overturned in this case. 
2. It creates problems for deterrence. But this is also ineffective. I reject that deterrence is the reason we ban killing 1 to save 5, and I reject that legality is morality applied. Laws in the minimal sense should exist to guarantee people's freedoms and rights. So, it's obvious that we should ban murder, even in cases where 5 is saved. Also, Consequentialism holds that the act of killing 1 to save 5 is actually good, but presumably we shouldn't deter good actions by making them illegal. If that's a bad argument against consequentialism, which I think it is, then it's also a bad argument against deontology.

Argument 7

Take the following example of theft. Suppose that there are 1,000 aliens, each of which has a stone. They can all steal the stone of their neighbor to decrease their suffering very slightly any number of times. The stone produces very minimal benefits—the primary benefit comes from stealing the stone.

The aliens are in unimaginable agony—experiencing, each second, more suffering than has existed in human history. Each time they steal a stone, their suffering decreases only slightly, so they have to steal it 100^100 times in order to drop their suffering to zero. It’s very obvious, in this case, that all of the aliens should steal the stones 100^100 times. If they all do that, rather than being in unimaginably agony, they won’t be badly off at all.

The following seem true.

1. If deontology is true, it is wrong to steal for the sake of minimal personal benefits.
2. If it is wrong to steal for the sake of minimal personal benefits, it is wrong to steal repeatedly where each theft considered individually is for the sake of minimal personal benefits.
3. In the alien case, it is not wrong to steal repeatedly where each theft considered individually is for the sake of minimal personal benefits.

Therefore, deontology is false.

This argument rests on, roughly, the same intuitions as Huemer's paradox. So as an argument which forms part of a cumulative case in addition to Huemer's argument, it doesn't really add more weight it seems to me. I'll just briefly make 3 points here;

1. I'm skeptical that, if an action is wrong, it thereby follows that a conjunction which contains the same action in repetition will thereby also be wrong. For roughly  the same reasons I am skeptical of the Individuation Independence principle.
2. This hypothetical scenario is the most abstract and bizarre yet, so our intuitions here probably don't count for much.
3. There may be a tu quoque concern here. Suppose you randomly select someone from the global population and give them a medical needle jab, the action is wrong, you caused them minor pain and didn't benefit them in any way. But you keep doing this a billion times, and doing so, basically statistically guarantee's that at-least 1 person will benefit from the jab, and have horrifically painful diseases many orders of magnitude worse than the pain caused by all jabs combined prevented. The individual act is wrong, and you are not in a position to know any of the acts, on an individual basis, wouldn't be wrong (indeed statistically they probably are wrong), yet the conjunctive act is right on consequentialism. So, if Matthew accepts the principle that if an action is wrong, the same action in repetition must also be wrong, which he needs for 2 to be motivated, it looks like his own position is under some heat. 

Argument 8

We have lots of scientific evidence that judgments favoring rights are caused by emotion, while careful reasoning makes people more utilitarian. Paxton et al 2014 show that more careful reflection leads to being more utilitarian.

People with damaged vmpc’s (a brain region responsible for generating emotions) were more utilitarian (Koenigs et al 2007), proving that emotion is responsible for non-utilitarian judgements. The largest study on the topic by (Patil et al 2021) finds that better and more careful reasoning results in more utilitarian judgements across a wide range of studies.

Ok, so this is one I have alot less to say about, I'm not all that familiar with the research on the topic.  It is case by case, I'd imagine. A general hurdle for alot of these studies is that  there are relevant differences between the cases presented, and alot of it might be accounted for by people reacting emotionally to the relevant differences. It could be that the same people use different parts of the brain when making judgements about different scenario's, utilitarian's and deontologists alike. Rather than it's just being the case that utilitarian's are cold rational thinkers and deontologists emotional thinkers. 

I also have my doubts that the fact that people with damage to the pre-frontal cortex have a positive correlation with utilitarianism is something that says much in utilitarianisms favor. It's not implausible that emotional ability allows agents to better empathize, and is an important part of moral reasoning. 

But at the end of the day, if the studies do overwhelmingly show that utilitarian's are generally more rational than deontologists in their judgements, so be it, I'll concede that this is a point in utilitarianisms favor albeit a non-decisive one. 

Argument 9 

(1) Deontic constraint (for reductio): Protagonist acts wrongly in One Killing to Prevent Five, and ought instead to bring about the world of Five Killings.

(2) If an agent can bring about W1 or W2, and it would be wrong for them to bring about W1 (but not W2), then W2 ≻ W1. (key premise)

(3) Five Killings ≻ One Killing to Prevent Five. (from 1, 2)

(4) One Killing to Prevent Five ≻≻ Failed Prevention. (premise)

(5) Failed Prevention ⪰ Six Killings. (premise)

(6) Five Killings ≻ One Killing to Prevent Five ≻≻ Failed Prevention ⪰ Six Killings. (3 - 5, transitivity)

(7) It is not the case that Five Killings ≻≻ Six Killings. (definition of ‘≻≻’)# Contradiction (6, 7, transitivity).

 I reject (2). I reject that the rightness/wrongness of an act is a function of the world-states that the action brought about/could have brought about. Rather,  it is a function of the reasons/principles which governed the act. (2) seems, pretty much, tantamount to a denial of deontology.

(I) Denying (2)—e.g. by claiming that we should prefer wrong actions to be performed— would rob deontic verdicts of their normative authority.

 But I don't think denying (2) robs them of their normative authority. At the very least, I'd need to see an argument for that claim, for this argument to move me (or any non-consequentialist, really). I also argued, in the previous article, wherefrom deontic verdicts derive their normative authority. Never did I appeal to the total goodness of world-states, but only, the unconditional value of humanity. Until the charge of loss of normative authority is motivated, I have nothing further to say here. 

 Argument 10

Deontology holds that there are constraints on what you should do. But this produces a very strange result—it ends up forcing the conclusion that sometimes putting perfect people in charge of decisions makes things worse. Suppose that a person while unconscious sleepwalks and is about to kill someone and harvest their organs to save five people. Then they wake up and have a choice of whether to actually harvest their organs or not.

Given that harvesting organs makes things better—it causes one death instead of five—it would be bad that they woke up. But this is strange—it seems like putting people who always do the right things in charge of a situation can’t make things worse.

This is another one I just don't feel much of the force of. I already believed that there were cases where an agent doing the right thing can make things overall worse off, (Killing 1 to save 5 being an example). So, given this, I'm not sure what pointing out that giving a morally perfect agent control can make things worse off adds. It's just entailed from both my normative theory, and my pre-theoretic moral judgements. Not a surprising result.

Perhaps you might think it is a strange result, but I also think there is a similar result for utilitarianism which seems equally, if not more, strange. There seem to be situations where utilitarianism is self-effacing. Human's are very bad at judging even some of the immediate consequences of their actions, let alone the long term consequences. So, it seems very plausible that human's acting like, or being deontologists and following a set of deontological rules would be overall better, at the very least in some cases than leaving human's to their own devices and attempt to figure out what course of action will maximize well-being/minimize suffering. In which case, making utilitarianism publicly accepted would be bad on utilitarianism. But that's an odd result, surely it can't be wrong for everyone to hold to, and attempt to act in accord with the correct moral theory. 

Argument 11 

1. If deontology is true, it’s impermissible to kill one person to harvest their organs and save five.
2. If it’s impermissible to kill one person to harvest their organs and save five, it’s impermissible flip the switch in the trolley problem.
3. It’s not impermissible to flip the switch in the trollye problem.

Therefore, deontology is false.

I reject 2,  but let's examine Matthew's argument for it.

Suppose that flipping the switch was less wrong than pushing the guy off the bridge. In this case, we should expect that, if one is given the choice between the two actions, they ought to flip the switch. After all, it’s better to do less wrong things rather than more wrong things.

Thus, the argument is as follows.

1. If flipping the switch is significantly less wrong than pushing the man, then if given the choice between the two options, even if flipping the switch is made less choiceworthy, one ought to flip the switch.
2. If given the choice between the two options, even if flipping the switch is made less choiceworthy, one ought not flip the switch.

         Therefore, flipping the switch is not significantly less wrong than pushing the man

Not sure what 1 means. What factors are we considering when we say an action is made "less choice-worthy"? This is vague to me. Matthew claims this premise is obvious in the article, but, as we've seen, he says a lot of things are obvious that I think are false. So I'm unsure whether to accept 1. 

I probably reject 2 depending on what is meant, but let's let Matthew disambiguate; 

Imagine the following scenario. There’s a train that will hit five people unless you do something. There’s a man standing above, on a bridge. There’s a track that you can, by flipping a switch, redirect the train onto, which leads up to the man on the bridge above the track. However, the train moves very slowly, so if you do that, the man will be very slowly and painfully crushed.

However, you have another option. You can, by pushing the man, cause him to die painlessly as he hits the tracks, and he’ll stop the five people. Which should you do?

Now, the deontologist is in a bit of a pickle. On the one hand, they think that you should flip the switch in general to bring about one death while saving five, but you shouldn’t push a man to save five. But in this case, it seems obvious that, given that the man would be far, far better off, it’s much better to push him than it is to flip the switch.

So, I lean towards thinking pushing the fat man is immoral and pulling the trolley lever is not. The main reason is because in the footbridge case you are using the fat man as a means, in the trolley case you're not using the person on the track as a means. What happens in each situation if the man isn't there? In the trolley case, you can still pull the lever and save those on the trolley. With the fat man, you can't. So you're using the fat man as a means to an end, but not the guy standing on the other track. Another important reason has to do with the Doctrine of Double effect. In the trolley case, your intent isn't (necessarily at least) to kill someone, the goal is to switch the track to save the five. This is what Anscombe highlights as a problem. 

With this in mind, what to think of the scenario? It highly depends. If the agent is deliberating between two ways to kill the fat-man, then it's immoral to do either, since in both cases he'd be intending to kill the fat man and use him as a mere means. If on the other hand the agent see's the switch, and pulls it with the goal to save the five, that's good. Now, in fairness, I can see how this, is intuitively a weird result so I do see this as having some intuitive pull. However, I don't think it's all that strong when we consider the morally salient differences between the two actions qua actions, as opposed to the two states of affairs qua states of affairs. All in all, I reject 2 (both arguments).

Argument 12

It doesn’t really matter if you or someone else violates someone’s rights. From, to quote Sidgwick, the “point of view of the universe,” it seems unimportant whether it’s you violating rights.

Indeed, it seems very clear that in the organ harvesting case, it’s more important that five don’t die than it is that you don’t get your hands dirty. This seems like an unassailable moral datum. But morality should pick up on the things that really matter.

This intuition is completely at home with my view. I grant that whether you, or someone else acts wrongly and violates someone's rights is not better or worse for the world. Where I disagree is that morality "should pick up on the things that really matter" in the sense Matthew means. Morality is not axiology.  It tells you what actions we should and shouldn't do,  not what is truly valuable from the point of view of the universe, if there even is such a thing, which I doubt. But even if there is, why should we expect moral judgements about the rightness or wrongness of human actions to track that? It seems we should only expect this if, antecedently, we are consequentialists, which I'm not. 

Argument 13 & 14

Moderate deontology holds that rights are valuable, but not infinitely so. While a radical deontologist would be against killing one person to save the world, a moderate deontologist would favor killing one to save the world. Extreme deontology is extremely crazy. An extreme deontologist friend of mine has even held that one shouldn’t steal a penny from Jeff Bezos to prevent infinite Auschwitz’s. This is deeply counterintuitive.

Depending on what is meant, I'm probably an 'extreme deontologist', if what that amounts to is denying threshold deontology and accepting some acts, such as murder, torture, rape etc. are categorically forbidden.  Though, I'm not as extreme as most others who we'd call 'extreme deontologists'. Yes, you should steal from Bezos. Yes, you should lie to the Nazi.  In any case, I'm extreme enough so that the explosion problem doesn't apply to me. Since I'm not a 'moderate deontologist', argument 14 fails to apply to me as well.  I'll leave this as an avenue for Matthew to attack my view on the grounds of 'craziness'. Moving on. 

Argument 15

There are various explanations of our deontological intuitions being produced by various biases. There are lots of biases that we should expect make us more willing to cause us to believe something like deontology even if it is false.

1. Status quo bias. This describes people’s tendency to prefer things as they are, in fact, going. For example, if you ask people whether they think a hypothetical person should invest in A or B, their answer will depend on whether you tell them that they’re actually investing in A or B. But this explains pretty much all of our deontological intuitions. All of them are intuitions about non-interference—about keeping things as they are. These are thus prime candidates for things that can be chalked up to status quo bias.
2. Loss Aversion. This describes a tendency for people to regard a loss of some harm as more significant than just losing out on a gain. Losing five dollars is seen as worse than not gaining an extra five dollars that one would have otherwise. But people being averse to causing extra losses, combined with the idea that various losses in, for example, Bridge are incorporated into their deliberation, means that they will be likely to oppose pushing the person.
3. Existence Bias. People treat the mere fact that something exists as evidence that it is good. But this intersects with status quo bias and explains why we don’t want to change things.

I'm willing to concede this is some evidence for utilitarianism over deontology. Nonetheless a few points are worth making here.

For one, I have my doubts that these bias's (at least completely) explains all, or even most deontological intuitions. Intuitions about the overriding wrongness of violating autonomy/self-determination, torture etc. even in cases where the outcomes are overall worse. As well as, the intuition that the wrongness of, say, promise-breaking being over and above the bad outcomes don't seem to be explained by any of these bias's.  

For two, this seems to cut both ways to some extent.  Utilitarian's seem to ignore many seemingly salient factors in ethical decision-making, they only judge actions based on the state of affairs brought about by it. In general, all human's are prone to bias, and it is doubtful that utilitarianism truly solves it, rather than merely being a framework for cloaking many of our pre-existing moral bias's under the guise of mathematical calculation and certainty. 

For three,  most people who we would expect to be most reflective about the subject, and thus least prone to bias, as well as most knowledgeable e.g professional normative ethicists are non-consequentialist with more being deontologists than any particular other moral theory (although I think the divide between consequentialism and non-consequentialism is much more significant than the divide between deontology and virtue ethics, which I take to collapse in many ways) and consequentialism being the least popular of the main three only a sub-set of which will be hedonic act utilitarians (Matthew's view).  Of course they will be prone to bias too, as all humans are, but I think this consideration is enough to greatly off-set the force of Matthew's point.

Conclusion

Most of Matthew's argument's seemingly have little to no force, or just fails in the case of the paralysis argument, or seemingly applies equally to consequentialism, and rely on the same sorts of intuitions about extremely bizarre hypotheticals which we shouldn't expect to be reliable. The best arguments and considerations raised here are, at best, some evidence against non-consequentialism but are not nearly as devastating as Matthew claims.  By contrast, my article provided direct argument's for Kantian constructivism's parsimony and explanatory power, as well as serious epistemic and practical concerns for consequentialism, and, the part of my case that did rest on intuition focused on ordinary cases rather than abstract reasoning about strange hypotheticals. I conclude that we should reject consequentialism and embrace deontology based on the arguments provided so far. 

Deontology vs. Consequentialism - My Opening for an Exchange with Bentham's Substack

 Introduction

This is my opener for a written cross-blog dialogue between myself and Bentham's Substack, also known as Matthew Adelstein, and by his Youtube handle 'Deliberation Under Ideal Conditions'.  I will be presenting my case for deontology, and against consequentialism, and he, on his blog, will present his case for consequentialism and against deontology. Here I define deontology as the thesis that the right-making and wrong-making features of actions are rights-based concerns, prima facie duties, or maxims which move a will. Contrasted with consequentialism, which is the thesis that the right-making and wrong-making features of actions are the total value of states of affairs.  The flavor of deontology I prefer, and will attempt to motivate, is Kantian constructivism. My methodology here for the majority of this article will be, broadly, abductive. I think we should prefer, and I do not believe my opponent will disagree, the theory which scores highest on theoretical virtues, such as explanatory scope, fit with background beliefs & intuitions,  theoretical simplicity, ontological parsimony etc. In particular, we should prefer the view which better explains various features of the moral landscape, such as our moral intuitions, practice, moral responsibility, moral knowledge etc. while managing the trade-off between explanatory power and the other virtues I listed. I will argue that my view scores higher, overall, on these explanatory/theoretical virtues then my opponents view. The article will be divided into two main sections, one covering direct considerations which are motivations for my view, and another covering indirect considerations which are, what I take to be, shortcomings of my opponents view. Since I believe some of the best motivations for deontology are the unsavory implications of rival views.  Section 1 will have 3 sub-sections, one introducing Kantian Constructivism and the other two arguing for it. Section 2 will have 4 subsections, each covering a separate objection for consequentialism. With that said, let us begin.  

Section 1 - Direct Considerations

1.1 Kantian Constructivism

As said, I will start by motivating the view I currently am sympathetic to, which is Kantian Constructivism. First, I will need to explain the view. It is divided into two views, one metaethical, the other normative ethical. Those being metaethical constructivism and Kantian ethics. There are many ways of understanding metaethical constructivism, most notably, there is Rawls' view, and Korsgaards' view. I take Korsgaards' view. I think Sharon Street offers the best characterization of the constructivism I favor (Street 2010). 

When a creature values something – or, as I will also put it, when he or she takes or judges this, that or the other thing to be valuable – he or she occupies what we may call for convenience the practical point of view. More broadly, we may say, the practical point of view is the point of view occupied by any creature who takes at least some things in the world to be good or bad, better or worse, required or optional, worthy or worthless, and so on – the standpoint of a being who judges, whether at a reflective or unreflective level, that some things call for, demand, or provide reasons for others. The claim is that we have an understanding of this attitude even if we do not yet understand what value itself is. In addition to having an understanding of the attitude of valuing and the associated idea of the practical point of view, we also have an understanding of the idea of entailment from within the standpoint of any given set of values. Quite apart from whether we think a given set of values is correct, in other words – indeed, even if we aren’t clear yet on what it is for a set of values to be correct – we can nevertheless think about and discuss what follows, as a purely logical and instrumental matter, from a given set of values in combination with the non-normative facts.

Street calls this the "Practical Standpoint characterization of constructivism". But, this alone is not  enough to capture my view, since such a characterization is equally consistent with Humean constructivism and Kantian constructivism. The difference between these two views, is their views on practical rationality. As Hume famously stated, "It is not contrary to reason to prefer the destruction of the world to the scratching of my finger". A Humean constructivist will hold that there are no irrational ends, no common ends entailed by the practical standpoint of all agents or valuers. Whereas a Kantian constructivist will demur, and believes there are irrational ends, and certain ends entailed from the practical standpoint of all rational agents. The Kantian will say that it is constitutive of agency that our ends our worth-bestowing in virtue of our rational nature, thus making rational nature the source of all normativity. We will next get into the motivation for this. 

1.2 The Normative Question

The first motivation for Kantian Constructivism, is that the Kantian Constructivist has the most satisfying answer to what Korsgaard calls "The Normative Question" (Korsgaard 1999). The question of why moral considerations have the authority to guide our conduct, and give laws that we all have reason to follow. The question of, what justifies the claims morality makes upon us. Following Korsgaard, there seem to be 3 conditions a satisfying answer to the Normative question must meet. 

1. It must succeed in addressing an agent in the first-person position who demands justification for the claims morality makes on them.
2. When the agent comes to know what justifies their action as required, they come to the belief that their action is justified and makes sense. 
3. It must appeal, to a deep sense of who we are, our identity.

The second follows from the first, the third condition is perhaps not a necessary condition for a successful answer, but is plausible in light of the fact that there are plausibly cases where morality demands us to sacrifice our life. What can be worse than death, other then something equivalent to it, such as losing our own sense of who we are?  It seems to be our practical conception of ourselves that matters most, that gives rise to unconditional obligations. They are those obligations which, if violated, we lose our identity, and are no longer able to think of ourselves as being 'under the description under which we see ourselves as valuable and our goals as worth pursuing'.  Which is no better, perhaps worse, than being dead. If a theory can satisfy condition 3, and others can't, then that is, at the very least an impressive explanatory virtue of the theory indeed.

The main competitor to constructivism in answering the normative question is Moral Realism, which Matthew himself is committed to. There is also voluntarism, and eliminativism about normativity, but I'll put these aside for our purposes. The realist believes, rather than normativity being imposed on us by an authority, or contract, that  normativity, obligation etc. is an irreducible  notion. Some actions just intrinsically instantiate the property of being right, and there is no further explanation or reason to appeal to. The problem here is simple, this doesn't answer the normative question, in fact, it doesn't meet any of the conditions. What the realist has done, is merely relocate the problem.  Since there cannot be obligations unless there are actions which we necessarily have reason to do,  the realist simply put the necessity where he desired to find it - those moral duties which we already thought were obligatory. But the normative question is asking if there really is anything we necessarily have reason to do, and if so, which things and why them? To put the problem another way, belief in the existence of irreducible intrinsically normative properties must be sustained by our confidence that we do, in fact, have binding obligations. So, appealing to the existence of these properties cannot be used to support our confidence. Yet, the normative question is precisely about what justifies our confidence, and it arises when said confidence is rattled. Thus, it's extremely hard to see how realism can answer it.

By contrast, we have the constructivist answer to the normative question, by taking the view that morality is grounded in human nature, or more precisely, our identity as rational agents. Moral properties are projections of human dispositions put to a test of reflective endorsement. We as humans are subject to practical constraints such as those from our own interests, and the interests of others.  These claims arise from one's practical identity which, as stated above, is a description under which you value yourself and see your ends as worthy of pursuit. Some examples; your identity as a man or a woman, a lover, or friend, or family member, and each of these identities give rise to a set of reasons and commitments, those reasons and commitments being constitutive of those identities . The normativity is 'built into' these roles.  In this case, we are interested in your practical identity qua rational agent, qua  'citizen of the Kingdom of ends'. The three conditions for answering the normative question, therefore, seem to be satisfied, it addresses any agent who asks it from the first person, because these constraints, ex hypothesi, are constitutive of the practical point of view of all rational agents, to deny such reasons would be for a rational agent to deny oneself as a rational agent which would be incoherent.  It answers the third by appealing to our practical identity, our deep sense of who we are. The remaining threat implicit in the normative question then, is that the various constraints our nature imposes on us may be incongruent. The normative question would then be fully answered by showing that they are not, in fact, incongruent. That meeting the constraints made from one point of view will not entail violating the constraints imposed by another. Perhaps it may have occurred to you that there may be cases wherein the reasons conferred  on you by your other practical identities. Such as parent, worker, romantic lover etc. may conflict with reasons conferred on you by virtue of your identity as a rational agent. However, your identity as a rational agent is unique in that it makes your other identities possible,  it is a more fundamental aspect of your identity making it, also more normatively fundamental. Other practical identities derive their importance and normativity from our rational nature. If humanity is not treated as normative then neither can our other identities be normative. Because of this, our rational nature governs and confers value on our other identities, so that conceptions of our identity that do not accord with the value of humanity/rational nature cannot be rationally acceptable. Perhaps then, there might be a tension between self-interested reasons as a rational agent, and other-regarding reasons as a rational agent. I don't think this is right. Suppose you take yourself as having self-interested  reasons to violate rational nature. You take those reasons as worth-pursuing in virtue of selecting them as your ends with your rational capacities. But if those reasons entail violating rational nature, then you are violating the very thing which made those reasons worth-pursuing in the first place. So, by reductio, you cannot intelligibly have reasons which entail violating rational nature. 

Another way of seeing that a Kantian constructivist view like mine is in a better position to answer the normative question would be a consideration regarding the regress of one's values. Suppose you wanted to eat ice-cream. The fact that you desire to eat ice-cream isn't a reason (by this I mean a reason sufficient to justify action) to eat the ice-cream. You can ask the normative question "why ought I act according to this desire?". You can weigh it against other desires, such as your desire to lose weight and ask "which of these desires is a better reason to act?". Notice, for any given impression or inclination with the content of some desirable outcome, you again, can ask the normative question. The same point applies to facts external to your will, such as natural properties e.g "do I have a reason to act in accord with this natural property?". For any heteronomous fact, it seems the normative question will keep coming back.  Since it will always be possible for the normative question to rear it's ugly head for heteronomous facts, the idea is, for it to be answered, value must be prior to heteronomy. In this sense, reasons are wholly contingent upon one's autonomous rational will, which, I take to be one's reflective capacity to pick out which of one's inclinations, impressions, and aims is a better reason to act. One's rational autonomous will is therefore the source of all reasons, values and normativity. This is the Kantian view. 

One worry the realist may have is, this only gets us extrinsic normativity, when what we  really  want is intrinsic normativity. On the contrary, the test of reflective endorsement shows that human nature is intrinsically normative. To quote Korsgaard;

Within human nature, morality can coherently be challenged from the point of view of self-interest, and self-interest from the point of view of morality. Outside of human nature, there is no normative point of view from which morality can be challenged. But morality can meet the internal challenge that is made from the point of view of self-interest, and it also approves of itself. It is human nature to be governed by morality, and from every point of view, including its own, morality earns its right to govern us. We have therefore no reason to reject our nature, and can allow it to be a law to us. Human nature, moral government included, is therefore normative, and has authority for us.

There is much, much more that can be said here, much more objections that can be made on both sides, I haven't nearly done justice to the case. Note that none of the arguments made in this section are at all original to me, for more, read Korsgaard's Sources of Normativity, and some of her other works. 

1.3 The Kantian Argument 

Next I'll present the a priori argument for Kantianism and briefly address a couple objections. Here is a good formulation of the argument from Julia Markovits' Moral Reason (Markovits 2014).

1) I value the ends I rationally set myself, and take myself to have

reason to pursue them.

(2) But I recognize that their value is only conditional: if I did not set

them as my ends, I would have no reason to pursue them.

(3) So I must see myself as having a worth-bestowing status.

(4) So I must see myself as having an unconditional value—as being

an end in myself and the condition of the value of my chosen

ends—in virtue of my capacity to bestow worth on my ends by

rationally choosing them.

(5) I must similarly accord any other rational being the same unconditional value I accord myself.

(6) So I should act in a manner that respects this unconditional

value: I should use humanity (that is, rational nature), whether

in my own person or in the person of any other, always at the same

time as an end, never merely as means.

(1) is true almost definitionally, we might say it's just constitutive of being a rational agent. (2) is where moral realists will demur (at-lest non-Kantian moral realists), I've already argued, borrowing arguments from Korsgaard, that realism fails to answer the normative question, so we should accept (2). I will provide more reasons in the next section to favor Kantian Constructivism to realism. (3) and (4), when properly understood, are just entailments of the previous premises. (5) is where the Humean and the anti-realist will demur. My goal here is mostly just to show my view is more palatable than Matthew's, not battle the Humean or the anti-realist. So I'll make this quick, a Humean will respond to the argument by saying there is no irrationality in valuing their own rational nature, and their own ends, and not valuing other beings with rational nature. They value it simply because it's theirs. The problem is this is, it seems to me, at worst inconsistent, and at best, intolerably arbitrary. For, it is not your rational nature that is the precondition for you valuing anything, or setting your ends and viewing them as worth-pursuing it is rational nature as such. Any principled reason you can provide, it seems, for valuing your rational capacities would be a reason that would apply, mutatis mutandis, to other beings with rational capacities. There is nothing special about your rational nature, it is rational nature in general that is the source of normativity. We can continue going back and forth in this dialectic, but as said, that isn't my purpose here so I'll just move on. 

1.4 Kantian Constructivism is theoretically virtuous

I will finish off the case for Kantian Constructivism by arguing that Kantian constructivism is theoretically virtuous. I have already, in part, made the case by arguing for it's explanatory power in answering the normative question. I'll list off some further ways in which Kantian constructivism or Kantianism broadly scores highly on theoretical virtues. 

1. Through the Kantian argument and the categorical imperative test, we have an a priori way of testing the correctness or incorrectness of some maxim, thus a firmer ground for moral epistemology and discovering moral truths, as well as a guide for ethical decision-making, couldn't be asked for. Something which, as we'll get to later, is not afforded to the consequentialist. All you need to do is ask whether some maxim or action X involves using rational agents as a mere means, or cannot be rationally endorsed, if universalized. If it fails the first test or both tests X is impermissible, if it fails the second test it is a bad maxim so avoid doing X and you are not permitted to make a principle of doing X.

2. Kantian constructivism is reason internalist, connecting moral reasons to our motivational set, thus explaining why they are something we deeply care about, whereas it is a bit more mysterious why we would or should care about external reason-giving properties. Also, making the theory fit better with our background knowledge since we already posit internal reasons such as desires, beliefs, goals etc. Despite this, it also has a boost in explanatory power relative to other anti-realist views since it accounts for many peoples, and most philosophers intuition that there are categorical reasons binding all rational agents including those imperatives that command us not to murder, rape etc.

3. There are also potential further problems with reason externalism. One immediate concern might be about the intelligibility of stance-independent external reasons. Lance Bush has made this point in a number of places, reasons seem to be relational properties, conceptually tied to goals, desires or other such internal motivations, as well as consistency relations therein. The concept of non-relational external reasons would then be, no less confused then non-relational tallness, or non-indexical northness. While I believe this is much too strong, and I do at-least have the concept of non-relational external reasons, I'll agree at least that there is something odd about it that makes it hard to wrap ones head around. This is something my view avoids entirely, as reasons are internal, and relational to the practical standpoint of an agent.  A similar point would be, following Jonas Olson (Olson 2014), external reasons are very queer to the point that we are better off rejecting them.  To briefly present the case, the realist posits that there is a counts-in-favor of relation, that is a relation which gives us reason to perform one action rather than another, I'll call it the C-relation (X counts in favor of Y).  Realism is committed to the C-relation obtaining independent of the agent's psychology. The problem is, it's difficult to understand what the truth-maker for the C-relation obtaining would be, if it's not the fact that, given X, the agent has some motivation to Y.  The conjunction of there being a C-relation, and it's obtaining not in virtue of any motivational factor seems to be extremely queer, difficult to parse, does not accord with our ordinary experience, and is unlike anything else we know. Maybe you don't buy  these arguments, but these are potential issues which internalism side-steps.

4. Even if you, for whatever reason, find reasons externalism plausible, there are plenty of realist accounts of Kantian ethics to accommodate such a view, most notably in the works of Allen Wood, Paul Guyer, and Peter Railton. 
5. Kantian constructivism is less ontologically profligate. It does not involve positing additional ontological commitments such as non-natural, irreducibly normative facts. The same can be said about moral naturalisms, but naturalist realisms seem to have hard problems of normativity. There is quite alot I can say about moral naturalism, but for now I'll just cite these sources which argue against it (Fitzpatrick 2014) (Parfit 2011) (Dancy 2006) (Enoch 2011).
6. To expand on (5), positing these non-natural irreducibly normative properties, or even natural normative properties does not seem to yield sufficient explanatory benefits so as to off-set it's lack of ontological parsimony. Indeed, Gilbert Harman (Harman 1977) has argued that moral facts are explanatorily impotent. But even if they aren't totally impotent, there's still  a question of whether theories invoking moral facts exhibit the requisite explanatory virtues. However, explanations which posit moral facts  lack predictive novelty, they do not unify  disparate pieces of data under one theoretical framework, and as (Wright 1992) points out, while perhaps being able to explain cognitive effects, fail to explain sensuous effects, interactive effects, and brute physical effects. Making them extremely unimpressive explanations. 
7. Kantian ethics has good explanatory power for accounting for our intuitions about human rights, human dignity, what makes us valuable e.g our rational natures, and accurately diagnoses what makes promise-breaking, rape, callousness, and racial and sexual discrimination wrong. Where many other views, including my opponents, fail.
8. Finally, Kantian constructivism is better equipped to address evolutionary debunking concerns. The debunker wants to argue that the evaluative attitudes, and the content of our moral beliefs are formed through unguided evolutionary pressures, because they promote survival and reproduction not because they are truth-tracking. It would be an extraordinary coincidence, if the content of our beliefs and attitudes was formed both by evolutionary pressure's and by objective moral truths'. The Kantian Constructivist does not have to affirm the crazy coincidence, as the realist must, that the content of our beliefs and evaluative attitudes happen to align with independent moral truths. The constructivism I lean towards, holds that evaluative truth is a function of that which is entailed from an agents/valuers practical point of view which includes our evaluative attitudes and beliefs, the content of which, we can happily grant, happened to be shaped by non-truth-tracking evolutionary pressures. While natural selection caused the set of evaluative attitudes and beliefs we started with, we can accurately reason about what is entailed from these beliefs and attitudes, and our broader practical point of view as an agent. As a Kantian, I would make the case that what is entailed through reflective endorsement is that, since all our values, and ends are set via choosing them with our rational capacities, it is constitutive of our practical standpoint as a rational agent that rational nature is the source of all our values, of all our reasons for action. Of course the debunker can then levy the charge that our ability to do such practical  and theoretical reasoning to arrive at evaluative truth, is also marred by evolutionary, non-truth-tracking process's but I think it is at the very least, a much harder case to make that accurate practical and theoretical reasoning ability does not correlate with survival benefits. 

Section 2 - Indirect Considerations

2.1 The Epistemic Objection

The first objection I will levy against consequentialism is one I think sits among the most powerful. For this sub-section, I will be taking my cues from the excellent paper 'Consequentialism and Cluelessness' (Lenman 2000). The basic idea being that Consequentialism implies we are hopeless judges of right and wrong. If consequentialism is right, then the right and wrong-making features of actions is the total goodness of their consequences. Yet, it is not possible to really know what the future holds, and in particular the many unforeseen outcomes which will determine whether the given act was right or wrong to perform. You're betting on total outcomes, over periods of time you couldn't possibly know enough about to make informed decisions. Thus, making consequentialism unusable as a practical guide for action, and ultimately lead to moral skepticism. James Lenman gives the following example;

Imagine we are in what is now southern Germany a hundred years before the birth of Jesus. A certain bandit, Richard, quite lost to history, has raided a village and killed all its inhabitants bar one. This final survivor, a pregnant woman named Angie, he finds hiding in a house about to be burned. On a whim of compassion, he orders that her life be spared. But perhaps, by consequentialist standards, he should not have done so. For let us suppose Angie was a great great great great great great great great great great great great great great great great great great great great(..) mother of Adolf Hitler. The millions of Hitler's victims were thus also victims of Richard's sparing of Angie. It must be stressed that there is nothing unlikely about this story. Let us bear in mind that in Hitler's family tree there are 2^100 slots for grandparents of this order. Because that is a number astronomically larger than the then available population of the world, there must be many people occupying more than one such slot. Nonetheless, it is very likely that there were back then large areas, of Europe at least, where a high percentage of the population were ancestors of Hitler (very possibly all those whose bloodlines endured into the twentieth century). Anyone who saved the lives of any of these people or any of their intermediate descendants or who missed some opportunity to kill them before they fathered or mothered the relevant child shares in Richard's wrongdoing.

Now, there are of course,  responses consequentialists have to this line of argument. None of them, I believe, work but I will only address a few and leave Matthew to provide his own way of resisting the argument. 

The first objection I will look at goes something like; Perhaps there is a small chance of a unforeseen disaster that will result from your act, but equally, there might be unforeseen great benefits that result from the act. Since we have no antecedent reason to favor one over the other, these possibilities will 'cancel each-other out' when deliberating which action you should do, leaving the known consequences to break the tie. This is Shelly Kagan's response to the worry, and as compelling as it seems at first blush, unfortunately, I'm going to have to agree with Lenman that it doesn't work. First off, this response doesn't seem to take seriously enough how likely it is for a given action to have disastrous consequences, this is made more clear when we look at the example I quoted above. It appears stronger than some bare epistemic possibility that for-all-we-know might be true.  Second, it doesn't take nearly seriously enough the extreme limitations of our knowledge when it comes to ultimate consequences and their moral value. Suppose we did have complete knowledge of the future, what we'd have to do is look at, at least, two alternatives and possible future histories, (almost certainly much more than 2, but I'm making this assumption to be generous to the consequentialist)  where you perform (or fail to perform) some action or other. You'd then calculate the total value of both complete futures and pick the one that is more on-balance optimific. Were this possible, it would require incomprehensibly vast knowledge that isn't available to us, complete propositional knowledge about a endless, or at least astronomically long future which is many orders of magnitude greater than the futures we will live or that our human brain could even begin to comprehend, super complex reasoning from historical counterfactuals and their massive and inscrutable-to-us causal ramifications, as well as knowledge of goods and evils, and entailments and causal relations therein, as well as what weights they should be assigned relative to other goods and evils which could have been instead instantiated had your choice been different. As such, when considering the epistemic distance between what we would do given complete knowledge, and their inscrutability relative to our epistemic situation and how much we do not know, it looks like the knowledge we get from the foreseen consequences of our actions should, by the consequentialists own lights be nigh weightless in comparison. 

The next objection, is a possible expansion of the last. Perhaps the consequentialist can use the principle of indifference to avoid running afoul of the epistemic objection and to lend weight to Kagan's response, The principle of indifference goes as follows; 

"If we have no evidence favoring any of n set of mutually exclusive jointly exhaustive possibilities, assign each possibility a probability of 1/n"

Using this principle we can justify the canceling-out response stated above. Since we have no evidence favoring any of the n mutually exclusive jointly exhaustive future outcomes, assign each outcome a probability of 1/n and let the foreseen outcomes break the tie. There are problems with this. For one, the unrestricted principle of indifference is false and is known to have paradoxical implications, see Bertrand's Paradox. For two, while perhaps there are restricted principle's of indifference which avoid the paradox, by e.g limiting it to cases where we already know the full range of mutually exclusive, jointly exhaustive possibilities in question. This does not seem to be the case for our knowledge of possible future outcomes. In fact, it's unclear how at all we would partition the possibilities that our actions might lead to, it seems nigh inscrutable given our epistemic situation what the possibility space even could be.  Finally,  even if we could appeal to the principle of indifference here, you should still be in doubt about what actions you should do,  suppose we assume, (Again, extremely generously, as if there are more than two then we should assign a lower probability a-priori to each possibility, according to the principle of indifference) that you are only picking from two alternatives, A and ~A, and you assign the a-priori probability of each complete future being more optimific on the whole a 1/2 chance. Then you factor in the foreseen consequences, which we'll stipulate are optimific if you do A, which we'll call F.   So, Pr(A&F > ~A&F) > Pr(~A&F > A&F), in other words the probability that A&F is as a whole optimific is very slightly higher then the probability that ~A&F is  as a whole optimific. Perhaps the consequentialist will consider this a victory, but I cannot overstate how slight the probability advantage here is. Since the complete futures and their value properties that stem from A & ~A are, for reasons explicated before, astronomically large, it is therefore astronomically improbable, that F makes any difference whatsoever as to whether A or ~A are as-a-whole optimific. So, while F would give some reason to do A rather than ~A, it would be an extremely weak reason, and again, this is even granting the super generous assumptions the consequentialist can make use of, which we shouldn't. 

The last objection I'll look at is the tu quoque objection. The objection, also from Kagan, is that if the Epistemic argument is sound, it attacks not only consequentialism but other normative views which have any plausibility. This is so, because every sane normative view holds that consequences are relevant to the moral status of actions, so we too should be in the dark about what actions we should do. Whether this claim is true, depends on what is meant. It is certainly true that every sane view, mine included, holds that at-least some consequences are minimally morally relevant to which actions are better to do. What is not shared however, is that all consequences, including those which are long-term, indirect and unintended, even in part, determine  the right or wrongness of actions. But it is the latter claim that is needed in order to mount the tu quoque objection.  I may well hold that at least some consequences are morally relevant, but this does not entail that I think all consequences, especially ones which could not be anticipated,  factor into the moral equation when deliberating what one should do, and it certainly does not entail that I view consequences as determining what makes an action right and wrong. That would be other things, such as rights-based concerns, a good will, the maxims the agent acted on, or the prima facie duties which are violated. Insofar as consequences factor into our moral decisions, it is the visible consequences which matter, a difference which is morally irrelevant by the consequentialists lights. 

Of course, it may be pointed out that the epistemic objection is not a knockdown defeater for consequentialism. Consequentialism may be true, and we truly are hopeless judges of right or wrong. That's true, that is indeed possible, but for most of us who seek practical guidance for actions, and find moral skepticism unattractive, the epistemic objection shows consequentialism simply isn't a live option. At the very least, knowledge of right and wrong, is a feature of the moral landscape we take for granted, but is one consequentialism is powerless to account for. 

2.2 The Moral Demandingness Objection

This next objection against consequentialism is a common one, which I'll dub the "Moral Demandingness" objection. It is as it sounds, the objection being that consequentialism (for simplicities sake I'll just look at hedonic act utilitarianism which is Adelstein's view) is too demanding. There are two versions of this claim. The first is the stronger claim that utilitarianism is so demanding that, similar to the epistemic objection, it is virtually useless as a guide for action. The second, is the weaker claim that utilitarianism is so demanding that it just fails to track our ordinary moral practice and beliefs. I think both claims are right, but I only need the second for the objection to form part of my abductive case against consequentialism. Consider how much more you could be doing right now to increase well-being, instead of reading this blogpost, you could be working and studying to become a doctor and move to a country where there is a shortage of doctors, or an epidemic of illness or injury, you could be constantly working to promote broader political change either by aiming towards becoming a UN member or a domestic politician, or doing activist work and starting fundraising orgs, among a plethora of other things. Relevant to this, is Peter Singer's famous argument (Singer 1997), if we see a child drowning in a shallow pond, we would immediately feel ourselves to have a moral obligation to save them.  The excuses that someone else could do it, and that it is a minor inconvenience are unacceptable. So, since it is not morally relevant how far away the child is, equally we should feel ourselves obligated, to donate items or support charity. Indeed, every-time you purchase luxuries, such as going to a night-out at a restaurant, going to the movies, buying a game. That decision can be weighed against what you could have done, the lives you could have possibly saved, and the well-being of the world you could increase using that money.  Since you could be doing these things, and these things would increase well-being much more than your current course of action, utilitarianism finds you blameworthy for not doing them. Yet, in our ordinary moral practice and our beliefs we do not find them blameworthy. We take it that charity, while good and praiseworthy, is not obligatory, and we are permitted to spend our resources on personal projects, luxuries, and loved ones. Once again, we might even endorse the stronger claim that, since all of our individual actions are such that we could have done more to maximize well-being, and utilitarianism demands this of us, utilitarianism cannot be a practical guide for human decision-making.  It demands too much to be a useful guide. 

I will just consider two responses for now. The first is the tu quoque response. Every plausible ethic runs afoul of Singer's argument, because every plausible ethic would agree that you should save the child in the shallow pond.  This is true, as far as it goes. What is not true however, is that on every plausible ethic, there is no relevant difference between saving the drowning child and constantly weighing everyday decisions you make against what you could be doing to save lives and increase well-being. Kantian ethics is, or can be, more lax and flexible when it comes to positive duties. You have a perfect duty to treat rational agents as ends in themselves, never as mere means, that is non-negotiable. You also have imperfect duties to help people when in need, and cultivate your capacities so that you have a virtuous disposition. But the thing with imperfect duties, is there is no specific pre-defined set of rules on how to follow them, such that if you fail, you would be violating your moral duties. Constantly donating to charity would be one way you can fulfill your imperfect duties, but it is not the only way. While, it would be impermissible to make a principle of not donating to charity, there is no particular time T such that you are obligated to donate to charity. I believe a Rossian intuitionist can help themselves to a similar response, as can other non-consequentialist views e.g virtue ethics. What about the drowning child case? In that case, you are directly observing a rational agent, (or potential rational agent) drown. This difference is indeed relevant on Kantian ethics, and most other non-consequentialist views, since one of the most important parts of these views is the intentional component. In not donating to charity, it does not follow that you are intending to permit children to squalor, starve, and languish. In the drowning case, it's plausible that, by failing to save them, you must judge them as not worthy of saving, thus not respecting their autonomy as an end in itself which is impermissible.  At least, you would have to cultivate very bad dispositions and character traits to actively not care enough about saving the child that is within your vision, which is plausibly impermissible. 

The next objection, is to accommodate human psychological limitations. Since ought implies can, and perhaps we can't always maximize well-being, utilitarianism will not demand this. Maybe if we do too much and sacrifice too much of our resources, this will make ourselves less happy and healthy and thus lessens our ability to do good in the long-term, leading to worse outcomes. I worry this doesn't seem to really engage with the objection. The question we are interested in here, is which individual actions when assessing a set of options, you are obligated to do. If you have two possible actions, and action1 will maximize well-being & minimize suffering, and action2 will not (at-least relative to action1), act utilitarianism implies we are obligated to do action1, always. Yes, you can also weigh concerns such as those regarding your own health, and happiness, and that of your families, but those concerns are surely miniscule when compared to the quality of life you can (indeed, it is the case that you can) increase for those living in squalor, any time you deliberate between whether you should help yourself to luxuries such as the newest video game, or donate resources or said money to starving children in Africa. It is surely the case, if any fact about the consequences of our actions is obvious, that the well-being gained both short-term and long-term is greater if you donate. As for the concern that it will weaken your ability to do good, insofar as that is talking about your own personal motivation, it's plausible that act utilitarianism should find you blameworthy for having such a weak motivation that is not sufficiently responsive to moral concerns. It doesn't seem plausible that we shouldn't find weak motivations blameworthy, our judgement of whether an action is right or wrong should be independent of that. Suppose we cede ground to this response and admit that always maximizing utility is not psychologically possible, and spending some resources on ourselves is necessary for us to be of benefit to others. It is still undeniable, that virtually everyone, including utilitarian's themselves, could be doing far more than they are, and are thus failing to meet their obligations, thus utilitarianism remains more demanding then our ordinary practice and beliefs. 

2.3 The Unintuitiveness Objection

 
Another common objection to utilitarianism is that it entails unintuitive ethical beliefs, and poorly accounts for our intuitions in a number of ways. There are famous thought experiments which tease-out unintuitive implications of utilitarianism, the most well-known ones are from Nozick, and those are the utility monster and the Experience machine. While, I think these have some force, and I do certainly share the anti-utilitarian intuitions in these cases, as do most, these cases are also very abstract and strange and perhaps we have good reason to distrust our intuitions in these strange hypothetical cases. There are also, strange, abstract hypotheticals one could come up with where deontology appears to get it wrong according to the intuitions of most people, so were we to play this game, it would be, roughly, a wash.

Instead of very bizarre hypotheticals, I'll focus on more ordinary cases, cases which utilitarianism gets wrong according to ordinary intuitions. 

Suppose a woman is in a comatose state, you break into the hospital in the middle of the night, and you can rape her while in this state without the trauma and suffering which rape towards the victim regularly implies, and you can get away with everyone being none-the-wiser. Suppose also that she probably won't wake up from the coma, but if she does she has a boyfriend so if she gets  pregnant she'll attribute it to her boyfriend. You increased your pleasure/well-being from the sexual activity and did not cause any pain; so even if you aren't obligated to rape her (which there's a non-zero chance you might be on utilitarianism) it would still at-least be a morally good act as far as you know, so utilitarianism would not judge the action as wrong or evil. This is very counter-intuitive.

 The utilitarian might respond that it can still lead to very bad consequences in the long-term so you shouldn't do it, there's a chance someone will see you do it, or that the woman will awaken from the state and later figure out that she was raped leading to a disaster. Surely, the utilitarian insists, you'd have to be talking about an abstract vacuum scenario if you don't take into account these possibilities. Two problems with this: First, it seems we can run what I take to be the best response to the epistemic objection here which is the canceling-out response, (see above) if the utilitarian takes that response then it should also apply here. Granted, it's possible, though not probable as far as you can tell, that the rape will lead to disastrous long-term outcomes, but it's also possible though not probable that it leads to even greater outcomes than anticipated, perhaps she awakens, becomes pregnant and lives an enjoyable life as a mother, or even better the son becomes a medical doctor and saves hundreds of lives. So you should let the foreseen outcomes break the tie, which is the pleasure you get from the rape and the lack of pain it causes anyone else. The utilitarian can only reject that response on pain of giving up their best response to the epistemic objection thus making the epistemic objection worse, so it is a 'pick your poison" situation. Secondly, we can just stipulate of the hypothetical that in-fact the overall outcomes are optimific. This is not an absurd vacuum scenario, it is still nonetheless true that raping her does not guarantee that no-one will be harmed and could lead to really bad outcomes, or to put it another way, that in a sufficiently close possible world it really does lead to really bad outcomes. Nonetheless in the actual world (or more aptly, the world where the scenario in question occurs)  it doesn't. Surely, it is the actual consequences which are morally relevant on the consequentialists' own view not counter-possible consequences.

Another response might also be to appeal to rule consequentialism. Rather than doing actions which maximize utility, you should follow rules which are justified in virtue of maximizing utility. Raping coma patients violates a rule. It occurs to me that the consequentialist might make use of this to respond to the epistemic and demandingness objection to some extent as well.  Problem; rule utilitarianism collapses into act utilitarianism or is non-consequentialist. Suppose you have a rule, "Do not lie", this rule is justified because it generally maximizes utility, but then you have a case where there is a Nazi at the door looking for Jews. Why would I not break the rule in this case? If the motivation for the rule in the first place, was that following it generally maximizes utility, why follow it in the instances where it doesn't? Perhaps the rule utilitarian could argue, that if breaking the rule in this case maximizes utility, then rule utilitarianism wouldn't have in the first place, picked this rule but a different one. This won't work.  If on rule consequentialism, the reason you pick one rule rather than another is because it maximizes utility/is overall optimific, then, for any rule you have, if there is a case where violating it maximizes utility, you amend it and have a better rule such that, the action which maximizes utility does not violate it. If you have a rule where following it maximizes utility 80% of the time, surely you can have a better rule, one where following it maximizes utility 100% of the time. So, you keep amending the rule, until it covers every circumstance  where the act utilitarian would lie, steal, kill, and rape thus making the views extensionally equivalent. Otherwise, you do not adjust the rule to accommodate cases where violating it maximizes utility, but in that case the view looks pretty non-consequentialist. You'd just be following a set of rules even in cases where doing so is not optimific, and the motivation for not breaking the rules in those cases is undercut by the very same motivation for having those rules in the first place. There is more that can be said here, but I will not say it yet, unless Matthew pushes back.. 

Here's another one. The act of gang-raping and painlessly killing a homeless woman is an action that plausibly can, or at the very least we can reasonably stipulate will, maximize total well-being. The gang rape will cause pain for one person, and generate pleasure for multiple people and the woman will not experience immense trauma after-words since she will be painlessly killed. Painlessly killing her will also prevent the miserable life she can reasonably be expected to have even had you not done the gang-rape. We can also stipulate, and as well it can be reasonably inferred given her predicament, that she does not have family or loved ones who will greatly suffer for her having been killed. While it is plausible that the disutility of the woman being raped is higher than the pleasure each rapist gets, it's far from obvious that the total known utility gained from the act is such that it does not outweigh the disutility. The utilitarian should at-least be in doubt that the act as currently stated is immoral, and in the case where I stipulate that the good outcomes of the act outweigh the bad outcomes the utilitarian should judge it as good. The utilitarian can make the same responses to this as the previous example, and I would levy the same responses to those responses.  

Here's another. You aren't permitted to sacrifice yourself to save a friend with poorer life conditions than you on utilitarianism. Normally we would think it is supererogatory to sacrifice your life to save a friend who will live a slightly less happy life than yours. On utilitarianism, it's forbidden since you are actually minimizing utility. I'll just point out that this one also isn't an abstract vacuum scenario, it's surely not someone no-one ever has or will encounter and totally alien to human life. 

Here's another. Suppose a thief tries to steal grandma's purse but in the process of doing so, pulls her away and saves her from being hit by a car. On consequentialism, since it's only the consequences of an  action that matter, the action of the thief was good, intuitively, it was not a good action since the thief's intentions were wrong. 

Here's another; You have a rich friend who is on his death bed and his dying wish is for you to make sure his fortune goes to his son. It is overall more optimific to lie and say his wish was for his fortune to go to charity. On utilitarianism, you therefore should lie.  This is unintuitive.

Finally, a classic. On utilitarianism, it's not impermissible, in fact it's probably obligatory, for a doctor to kidnap someone off the street to harvest their organs and save 5 others. 

I could go on, but instead of continuing with utilitarianisms' unintuitive answers to specific ethical questions, I'll end this sub-section by listing 4 general ways in which utilitarianism seems unintuitive.

1. Utilitarianism fails to sufficiently account for human dignity and reduces ethics to a numbers game. Either by doing arithmetic to save the most lives, or calculating "units of utility" where if some threshold is crossed, it's ok to murder someone. By most people's intuitions, mine included, this seems to miss the point of ethics. 
2. Utilitarianism is not egalitarian, nor does it entail justice or fairness. The utility monster thought experiment reveals this to some extent, showing we should privilege the experiences of a being who can feel much higher pleasures, even over all of humanity assuming the difference is so great. A less abstract scenario would be, suppose we pass a legislation which directly and intentionally is racist and greatly reduces the living conditions for black people, but at the same time greatly increases the living conditions for both rich and white people, so that more utility is gained on net. On utilitarianism, such a legislation would be good. 
3. Utilitarianism fails to accurately diagnose what makes actions wrong. There seems to be something that makes breaking promises, discriminating on the basis of race and sex, violating bodily autonomy, cruelty, and lack of sensitivity to the plights of others wrong, that is not reducible to consequences.
4. Utilitarianism does not sufficiently account for intentions when it comes to judging moral actions. 

2.4 Weak Motivations

The last objection I will present will be that I think the motivations for Utilitarianism fail. It would be one think if utilitarianism had really strong motivations that are sufficient to off-set the problems I mentioned.  Since this post is getting much longer than anticipated, I'll just look at a few arguments and briefly explain why I find them uncompelling, or at-least non-decisive. 

2.4.1 Theoretical Simplicity

The first considerations supporting utilitarianism is theoretical simplicity. The idea being that we should prefer simpler theories, because simpler theories are more likely all other things equal, and utilitarianism is simpler in terms of how it's theoretical content is cashed out. This is because it only has one rule "maximize utility", It only posits one thing as being identical to or coextensive with, the good, that is pleasure (at least on hedonic utilitarianism).

The utilitarian's claim to theoretical simplicity is not indubitable and can be questioned. Perhaps we might say, while the utilitarian can help themselves to some initial description under which their theory appears simple, it might be highly complex when we investigate exactly what such a description actually implies. Utilitarianism states that we ought to maximize utility, but for this to have content, we need to specify what utility it is we are maximizing so we need to posit further concepts, most notably the phenomenal state of pleasure and the claim that it has moral value, and we also need a way of understanding what it is to maximize it, so we need to specify a standard by which we measure it. Then we specify what it is for there to be disutility which requires invoking yet more concepts, the phenomenal state pain and it's having moral disvalue, and a standard by which we can weigh the utility of pleasures against the disutility of pains so that we can figure out which action is right and wrong. This gets yet more complex when we reason about how we  apply these concepts and calculations to calculate the total value of long term outcomes. Additionally while it seems simple to claim only one thing, pleasure, is coextensive with goodness, there is a sense in which that claim is unmodest since it involves being committed to the falsity of a massive set of propositions that state any property P is the good rather than, or in addition to pleasure. I'm not going to make the claim that utilitarianism is less simple. The upshot here is just that assessing theoretical content is very complicated, and it is therefore extremely non-obvious that utilitarianism comes out as the most theoretically simple theory overall. 

I can even grant that utilitarianism is more simple in terms of it's theoretical content, even so this consideration is far from a game-changer. Even if utilitarianism wins on theoretical simplicity, it probably loses relative to my view, on ontological parsimony, since presumably we'd need to posit some strange irreducible external property of stance-independent badness that is conferred on pains, and stance-independent goodness conferred on pleasure. This is both ontologically uneconomical, and very metaphysically strange such that many people, myself included, have a hard time understanding what is even being said or picked out. Kantian constructivism, on the other hand, does not require these strange ontological commitments. The utilitarian can perhaps weaken the force of this point by being a moral naturalist, but as stated in section 1, moral naturalism has problems of normativity. Furthermore, the most plausible form of moral naturalism, which I would say is Neo-Aristotelianism, is hostile to utilitarian commitments. Finally, even granting utilitarianism theoretical simplicity, this does not off-set it's poor explanatory power when it comes to our intuitions and the other objections I laid out. 

2.4.2 Phenomenal Introspection

Another motivation for utilitarianism/moral realism that Matthew likes is the argument from phenomenal introspection. Here is Matthew's summary from this article. 

1. Phenomenal introspection is the only reliable way of forming moral beliefs.

2. Phenomenal introspection informs us of only hedonism

Therefore, hedonism is true — pleasure is the only good. Phenomenal introspection involves reflecting on experiences and forming beliefs about what they’re like (e.g. I conclude that my yellow wall is bright and that itching is uncomfortable). Premise 2 is true — when we reflect on pleasure we conclude that it’s good and that pain is bad. Premise 1 is also true — other moral beliefs are arrived at through intuitions, but those are unreliable given that many people’s intuitions throughout history have supported slavery and genocide — and people’s moral intuitions lead to totally different disagreement. If some method of forming beliefs constantly contradicts itself and concludes genocide is good, then it’s unreliable, especially when we have no evolutionary reason to think that it’s reliable. However, phenomenal introspection is reliable; beings who can form reliable beliefs about their mental states are fitter than ones who can’t. 

 I'll make this brief. Premise 1 is extremely dubious for two reasons. For one, a friend of mine on his blog, has argued, succinctly I believe, that phenomenal introspection does not get you to moral beliefs about the stance-independent badness of pain and goodness of pleasure, the  very notion that it does, is confused. The belief is formed on the basis of a second-order judgement about one's phenomenology which requires referencing a certain representational, most likely inferential process of forming the attitude and the folk psychological framework the attitude is embedded in as well as presumably, a set of background beliefs. Insofar as these further assumptions are invoked, contra Matthew, there is no immunity to debunking considerations. Check out my friends article, and his second response to Matthew for more detail and Lance Bush makes a similar point here. 

For two, even if phenomenal introspection is a reliable process for forming true moral beliefs, it's implausible that it's the only reliable process. It is strange that Matthew, a phenomenal conservative would argue that moral intuition is unreliable for forming moral beliefs. Regardless, my case for Kantian constructivism does not rest on intuition, and Kantian constructivism implies certain moral beliefs. I've already provided an a priori argument for Kantian ethics, as well as abductive considerations, and while there is disagreement on the arguments' respective soundness and cogency, surely a priori arguments and abductive arguments are a reliable guide to truth. 

2.4.3 Irrational Desires

 The last argument is that there are irrational desires, and we have powerful intuitions that desires for well-being/pleasure are inherently worth pursuing, and failing to pursue well-being is irrational. 

This is typically motivated by an appeal to certain examples or thought experiments. I'll once again just quote Matthew.

 A person doesn’t care what happens to them on a future Tuesday. When Tuesday rolls around, they care a great deal about what happens to them; they’re just indifferent to happenings on a future Tuesday. This person is given the following gamble — they can either get a pinprick on Monday or endure the fires of hell on Tuesday. If they endure the fires of hell on Tuesday, this will not merely affect what happens this Tuesday — every Tuesday until the sun burns out shall be accompanied by unfathomable misery — the likes of which can’t be imagined, next to which the collective misery of history’s worst atrocities is but a paltry, vanishing scintilla.

Another example.

Suppose a person hates picking grass -- they derive no enjoyment from it and it causes them a good deal of suffering. Suppose on top of this that they are terribly allergic to grass -- picking it causes them to develop painful ulcers that itch and hurt. However, despite this, and despite never enjoying it, they spend hours a day picking grass.

Last one.

 A person is indifferent to suffering that’s on the left side of their body. They still feel suffering on the left side of their body just as vividly and intensely as it would be on the right side of their body. Indeed, we can even imagine that they feel it a hundred times more vividly and intensely — it wouldn’t matter. However, they do not care about the left side suffering.

I'll note that it's not argued for, or explained why these desires are irrational. We are just supposed to share the intuition that they are irrational. The idea then is since on hedonic utilitarianism pleasure is all that is good, and all that is bad is pain, utilitarianism has tremendous explanatory power for explaining why these cases are irrational, it is because maximizing well-being is the only goal worthy of rational pursuit. 

I'll make a couple points in response. Firstly, I simply do not share the intuition that these cases are irrational.  It is certainly true that they all involve very bizarre desires for a rational agent to possess, but this does not entail that they are making a mistake. I do not know what Matthew means by irrationality, but I take it that irrationality is a means-end incoherence from one's practical point of view, or inconsistency relations between one's beliefs, desire's, goals etc. There is no inconsistency here, just strangeness.  Even if they are being irrational, it's quite unclear to me what the truth-maker for that claim would be if there is no incoherence. 

Secondly, it is precisely because these scenario's are so strange that we should not trust our intuition that they are irrational.  These cases are so strange that it is impossible to imagine what it is like to be in the brain states of such weird agents. Who's to say they cannot be so constituted that, were you in their perspective, you would not judge them of making any error. 

As a general point, it's not clear to me how hendonism/utilitarianism actually gets a boost in explanatory power here relative to my view. It only explains these cases in the sense that, given the truth of hedonism, they would be irrational. But it's not actually explaining what makes them irrational, if the answer is just "because well-being is necessarily worth-pursuing" that seems to me to just reassert the content of hedonism. Whereas my view explains why certain actions are irrational by drawing out an inconsistency from the practical standpoint of the agent.

Conclusion

To summarize the case, Kantian constructivism offers the most satisfying answer to the normative question, there is an a priori argument for Kantian ethics, and Kantian constructivism is theoretically virtuous as it explains our intuitions, it accounts for moral knowledge, it is in a better position to avoid debunking considerations, while being less ontologically committing, and not taking a strange view like externalism about reasons. Utilitarianism runs afoul of the epistemic objection, which is not only disastrous for moral epistemology but implies we are hopeless judges of right and wrong, it is too demanding, it fails to properly account for our intuitions in many ways, and it's motivations are manifestly insufficient to off-set the costs. I conclude that we should be Kantian constructivists and reject utilitarianism. I look forward to Matthew's rejoinder.