The order of events matters. If the coin is flipped first (with a definite outcome unknown to the players), then subjective and objective consequentialism result in different conclusions.

The issue with probabilistic concepts is that it puts objective consequentialism into a very strange place. Let's suppose that determinism is true. In this case, all future events are facts about the universe (there is only one possible outcome). In your coinflip example, playing the game either results in killing nobody or killing more than two people nonprobabilistically. And you can't really know which (unless there's information in the universe that allows you to precisely predict some future outcomes, namely the result of this coinflip.) This makes it kind of an impotent ethical approach.

If determinism is false, one can legitimately ask the question whether such a thing as an "objective probability" even exists. That is, the falsity of determinism does not imply that the universe is definitely probabilistic (meaning that future events have well-defined probabilities).

For example, we may model coinflips in an objective manner by claiming that it's a 50-50 proposition, but it's far from clear that the reality of the situation is actually that it's a 50-50 proposition. The probability of the outcome may depend on future events (such as how the coin is actually flipped and how the coin is caught), and that these future events may not have definite probabilities associated with them. Even if we allow for the idea that the coin has an inherent 50-50 probability associated with its "natural" behaviors, the coin flipping process can still be seen as not having an objective probability.

Your continued insistence that meta-ethical positions have no bearing on ethical reasoning is bizarre. You can't pretend that two positions can always reduce to identical logic under distinct meta-ethical viewpoints with enough translations of words and meaning.

For example, consider the meta-ethical position of a Divine command theorist together with strict consequentialism. Actions are either good or evil if God declares they are good or evil. But then we have an ethical understanding of good or evil dependent solely upon consequences of actions. The result is an extremely shallow ethical perspective that has some very serious implications that are pretty ridiculous.

I don't claim that subjective consequentialism implies moral relativism (or vice-versa) in some formal manner. I'm claiming that they are close to each other. They work well with each other and concepts flow well between the two.

You also need to distinguish your moral concepts carefully. Whether an action is morally good/wrong is distinct from the idea that an action is morally responsible/irresponsible. Pushing 72o is good if it wins, but that doesn't make it a responsible play.

You like to say that I'm confused by this, but you have yet to actually demonstrate any actual confusion. This results from your continued insistence of the complete absence of any consequences of holding a meta-ethical theory. If you believe that the two are independent of each other in every possible manner, and that one can always translate one to another and accept identical logic, then I would say that your understanding of ethics is lacking. Wait... you've said exactly that. So I think your understanding of ethics is lacking.

I think your understanding of ethics is lacking.

Your definition of objective consequentialism seemed to strong to me (very strange that flipping a concealed coin before or after the choice made a difference, for instance), so I googled some defs and this resolves at least some of the issues:
So yes, if we focus on "actual consequences" where you demand we need to know absolutely everything about our universe (or toy universes, as it happens), including things like the outcome of all coinflips, that flipping coins don't butteryfly effect into tornados, etc, then some of these determinism vs nondeterminism problems come up. But if we weaken to a sort of "objectively probable consequences" then we can still say that murdering a kid is bad even though we don't know for certainty that the kid doesn't grow up to become hitler.

True, you can still raise some of your problems (what does probability mean?) and questions like that are interesting philosophical problems, but I don't know if we are required to solve them before we can have a meaningful conception of objective consequentialism as compared to other moral theories.


I don't know what "flow well" is suppose to mean here. But I suspect that you are confusing a generic concept of being relative to individual and capital R Moral Relativism. I can say "it is moral to kill people taller than you" and this is a lower case r relative theory because whether it is or is not moral to kill someone is relative to who you are and your height. Likewise whenever we talk about subjectivity which refers to mental content in a person. Whether something is moral in theories like subjective consequentialism is likewise relative to who you are and your mental content. But this is not Moral Relativism, the meta-ethical theory. Indeed, my "it is moral to kill people taller than you" theory, or subjective consequentialism, can be made as universally applicable edicts, that this is good for all people, and for all cultures, regardless of whether they do or do not accept my ethical proclamations. So maybe you are right, that it "flows well" in some completely unspecified sense. But it would seem to "flow well" with meta-ethical views that are NOT moral relativism as well, and the ideas are fundamentally distinct eeven if you want to use the word "relative" somewhere in your ethical theory.



Well it depends, are the actions that God declares to be good those that have certain types of consequences? As in, DCT is a meta-ethical theory, but presumably is accompanied by an ethical theory too, such as a list of commandments in a religious book. If one of those commandments was "Thou shalt act to cause good consequences" or whatever, then there would be no tension, and if not, the tension is between two different ethical theories (consequentialism and the list of commandments) not between an ethical theory and a meta-ethical theory.

Let me be clear, my position is not that there can be no connection at all between meta-ehtical theories and ethical theories, merely that your willy-nilly interchanging between the two without even mentioning the distinction isn't valid. For instance, you stated that an ethical theory was "not that far" from a meta ethical theory without any form of note that you are comparing apples and oranges here! That said, I'd be curious if you can answer this question. Person A, a deontologist, claims "it is wrong to murder person C". Person B, a consequentialist, claims the same. Do you know what their meta-ethics are?

Here are two presentations:

http://onlinelibrary.wiley.com/doi/1...187.x/abstract

http://philpapers.org/browse/objecti...nsequentialism

The distinction between the presentations is that the second one is merely trying to parse between objective/subjective consequentialism, whereas the first is also needing to distinguish between subjunctive and possibilist forms of objective consequentialism.

----

I don't know how to break it to you, but determinsm/indeterminism comes up. It's part of the nature of the beast.

http://onlinelibrary.wiley.com/doi/1...187.x/abstract

---

Ummmmm... not really. You have no meaningfully objective measure on the probability of that event.

I never said that it wasn't meaningful. I'm just trying to draw the clear delineation between subjective and objective consequentialism. But it really does matter in a very real way whether you're talking about what the individual perceives or whether you're talking about the reality of the situation. This has analogies all over the place in mathematical modeling, including finance (fat tails) or more broadly "black swan events" or other outliers.

We've been going around in circles for a while on this, so I'm going to quote IEP and try to clear it up:

http://www.iep.utm.edu/moral-re/#H2

Let's agree to this language, taking this as a neutral arbiter of meaning. You are denying (g-2) in your presentation of "meta-ethical moral relativism." I've adopted (e) in my usage of "meta-ethical moral relativism."

I think what your objective / subjective distinction captures is the different ways we evaluate the act v evaluate the actor. It relies on allowing only that a bad decision can have a good outcome. Thus the act that was poorly motivated has a result that benefits the actor. In your example above the person who makes a bad decision is a dollar better off than the person who doesn't. It seems that the person who refused to bet does worse than the person who did.

But again we can differentiate evaluating the actual outcome from the expected or intended outcome and judge the actor irrespective of results. When evaluating the actual outcome we make no claim on how the person should have acted when placed in the decision without all the information. This is not normative this does not refer to what the actor should have done.

Yes and?

I'll borrow the example Original Position used to explain the distinction to me. Imagine that a mugger steals a persons purse and the person is sent to hospital for checks. Imagine that during those checks a diagnosis is made which had it been left would have resulted in the person dying. Would you allow that it was good for that person in that instance to be mugged? The consequentialist says yes, I say yes, you say?

Right, so I quoted an alternate definition from the SEP previously. It would seem that there are many different variants here, but that people have considered at least the following: actual consequences (your version of obj. cons.), likely, intended, forseen, and forseeable consequences. Loosely it seems that actual and likely are lumped into "objective consequentialism", intended and forseen into "subjective consequentialism" and foreseeable somewhere in the middle. Ie the subjective ones depend on the cognitive states of the individuals.

Do you also suggest I don't have a meaningful objective measure of the probability of a coinflip? Let me quote an example from the SEP that shows the kind of level claims about likelihood are being made: I feel if we are happy to say one is objectively likely to become ill on eating rotten meat, I can say a child is objectively unlikely to become the next hitler.

This is a much weaker standard. We are not asking what the actual probability is of something in a (debatably) indeterministic universe. We are not asking the deeper philosophical questions about, say, what probability even means. We are doing a more modest day-to-day type of analysis, I think, where we consider the available information (not every fact in the universe) and make predictions about the future. It is likely to rain today, but I am unlikely to be struck by lightening, and so forth. It isn't that deeper philosophical questions aren't interesting or can't be discussed, as some have, but it doesn't seem particularly necessary or constructive for our purposes here (although I'm not sure what our purposes actually are, so perhaps this is the main thing you are interested in, i don't know).

Personally, I don't like the strong version of objective consequentialism you gave (ie actual consequences) because a) we can't possibly know everything and b) can't possibly compute all the consequences even if we did. And even then it has weird artifacts like the morality of an action changing based on whether a covered coinflip happens slightly before or slightly after the choice. At best, it is a sort of aspirational standard. A weaker version where we say that based on the information that can be reasonable known, rational people would conclude that a consequence is likely, this is a pragmatic way to go about making moral decisions. So for instance, we can't know if a drunk person will or will not get into a car crash on their drive home, but reasonable people can conclude that it is sufficiently likely that this will occur to condemn the act as immoral. Perhaps you don't disagree, but just want to call this "subjective consequentialism", which might be fine (and might even be canonical....it does seem there is some differences in the literature), but then there is meaningful distinctions still to be made from this likely consequence stuff vs the intended or forseen consequences stuff.

Sure, I'm happy with this. I've also typically been operating on the weaker version in e as well. For instance, when discussing the distinction between ethics and meta-ethics, I've said versions of "one could consistently affirm [meta ethical relativism] and still insist that one particular standpoint was demonstrably superior to all others." On this view, one could accept, say, objective consequentialism as a normative framework while having the meta ethical views of a relativist.

Accepting e going forward, the question is why you think that the specific normative framework of subjective consequentialism "flows well" with meta-ethical moral relativism in a way that other normative frameworks do not. You have yet to shed light on it, so I can only guess. The guess I made earlier was that you were confusing the subjectivitity of the normative framework (ie statements are made relative to the mental content of people) and meta-ethical relativity. They both use the word "relative" but are used in fundamentally different ways on two fundamentally different levels. For instance, in my "it is wrong to kill people taller than you" ethical theory, if A is taller than B who is taller than C, it is wrong for person C to kill B, but not wrong for person A to kill B. The answer, in this ethical system, is relative to the actor, but this is nothing at all like meta ethical relativism. One could also subscribe to this ethical theory as a meta ethical moral absolutist, for instance.

I can say that it's "good for that person" without saying that it was a "morally good" act. Perhaps a better word for it would be "fortunate."

Deregs, note that one sort of resolution that might be relevant in your example is that of proximate causality. So one can say that while the act of stealing is clearly wrong according to likely, intended or forseen consequences, it is also wrong according to actual consequences because it was not proximately causal to the discovery of the fatal diagnosis.

It was an act with a good outcome or consequence.

Assume there's a clear link between the mugging and the person being sent to hospital for tests and the diagnosis.

We can still punish the mugger it would still be appropriate to charge the mugger irrespective of the actual outcome. Hence we have an act which in this instance has a good outcome but one which the actor is penalised for.

But was the outcome a good *moral* outcome? If the person dies of the particular condition that was discovered, is that a *moral* harm?

I think so, if, as the consequentialist is apt, one is a utilitarian and the discovery of the condition minimises the suffering and / or maximises the pleasure the victim experiences then that is a good moral outcome. If the person dies of the condition then the calculation needs to be remade, if their death entails less suffering then it still counts as a benefit if it doesn't then the benefit of the discovery reduces when weighed against the harm committed by the mugging.

Penalizing isn't really the question. Legally, the person violates a deontological law (that you can't mug people) and that is why they are penalized. The question is whether that action is moral. On the view of someone who only cares about the actual consequences of something, if the actual consequences are good the action is considered good. On the view of someone who only cares about hte likely consequences of something, if the actual consequence is bad, the action may well still be considered good.

This is one of the reasons why I find blind happiness-driven consequentialism to be a somewhat confusing way to see the world. It's like it doesn't want to accept death as a natural part of life.

And how would such a calculation be performed?

Coinflip, yes. I can grant a meaningful objective measure to that. "The next Hitler?" No, I don't grant a meaningful objective measure to that. It's not a well-defined condition.

This is a bad analogy. Jumping from rotten meat to Hitler loses all cause-effect relationships and any sense of foreseeable consequences. It feels you're taking the statement out of context. (See below.)

The skirting of the line of subjective/objective here should be obvious. I don't read the paragraph in context as "setting a standard" but rather exploring a variety of viewpoints. The paragraph is making distinctions between intended, foreseen, foreseeable, and likely. Here's how the paragraph starts and ends:

I wouldn't take this paragraph as an elaboration on the definition or concept of "objectively probable." At any rate, you've clearly taken it too far by jumping from the cause-effect nature of eating rotten food and people growing up to be "the next Hitler." The issue is that there's nothing in particular to "foresee" in the kid's future to declare some sort of probability of what will happen to him.

Such is the nature of the beast.

Sure, but you have to be very careful about the actual nature of the relationship you're establishing.

How likely is it?

I can accept these distinctions, but you first need to be more precise in what you mean by them. I don't know what you "foresee" in a child about whether he will become "the next Hitler."

I think we should be careful about the terms we use, laws can be based on deontological or consequentialist principles. Mugging people causing harm is a reason to have laws proscribing mugging if those laws reduce mugging.

The consequentialist can distingiuish the likely outcome, by which she judges the actor as prasieworthy or blameworthy, and the and the actual outcome of the act which she judges good or bad.

The calculations are difficult and this is why consequentialism has challenges as a decision making theory, however post hoc we can ask people and there does seem to be some understanding of how to evaluate and quantify suffering.

It doesn't require that people not accept death only that as being dead is morally neutral, no happiness or suffering, it should be preferred to suffering and not to happiness.

Not in the long-term perspective of a human life. For example, I don't think that utilitarians have successfully demonstrated that the net utility of an average human life is positive.

The bolded isn't actually true.

1) Why must we take death as morally neutral? Why can it not be a morally good thing? Indeed, if you take species overpopulation into consideration, it can be a very good moral thing for the population as a whole for some people to die.

2) On a more technical note, it's not clear to me that death must be treated as morally neutral even if it's the separation between good and bad. I can imagine re-scaling the utility so that death is 1,000,000 happiness units and the things that you've called happiness are things that are greater than 1,000,000 happiness units, and suffering are things that are less than 1,000,000 happiness units.

The larger point here is that it seems canonical to accept a weaker condition then "actual consequences" in the definition of "objective consequentialism", we can also accept "likely consequences". For those, we agree can talk about situations depending on coinflips and eating rotten meat in a "likely consequence" model. I"ll talk about distinguishing these cases from the growing up to be the next hitler case in a bit, but the basic point - that we have relaxed from requiring all possible information - has been established.

Is a lack of well definedness really the trump here? If you wish we could say something more precise like "will lead a regime that murders millions". Surely we should be able to agree that it is unlikely that a child will do this? Indeed, we can observe that very few of the billions of adults born do this, and thus it is unlikely any individual child will do this in the same way we might say it is unlikely they are to live to 115. Granted, we can't numerically compute something like this, just give a sort of vague "it is very unlikely", but then we can't for "it is likely you will get sick from eating rotten meat" either, which has somewhat similar levels of vagueness.

Similarly for the drunk driver example, the lack of an ability to quantify it precisely is something that is shared with this example and the rotten meat example. If you believe only one of these to be admissible for our purposes, how are you distinguishing between them?

It appears as if these paragraphs are meant to elaborate on the distinction between the two examples, but frankly I am not following your point. Can you state precisley what it is that distinguishes your contentment with claiming it is likely a child will get sick from eating rotten meat, but reject claiming it is unlikely a child will lead a regime that murders millions?

Oh I agree that is the function of the paragraph, but in so doing it has let us see the kind of examples the author thinks are admissible for "objectively likely" (which otherwise you and I might debate what it means).

The more interesting discussion, in my mind, is the moral relativity bit. The above is more about what the canonical definitions of words are. If you recall you came into this discussion arguing from objective to subjective consequentialism because of some argument (that seems really bad but I don't really care about, especially since you demand such strong conditions for your terms) and then concluded subjective consequentialism is rather like moral relativism, a claim that seems on its face and has had no further elaboration from you other than that it "flows well". That is the part I would prefer you focus on.

We've had a similar disagreement before, as I recall you consider there is sufficient evidence that the world contains more evil than good and so it seems reasonable to assume from your perspective that there is more harm than happiness.

I disagree I consider that there is evidence that the bar for a life worth living is rather low and as such is met often. This doesn't mean we shouldn't try to raise the standards of the worst off, we should but the repugnant conclusion carries weight because of that standard being low. That people living lives we may consider wretched try to survive in the numbers they do supports this. Even if we argue that this is a response to evolutionary pressures it seems persons are privileged when it comes to quantifying whether their lives are worth living and the actions they take support that they do.

My language may have been clumsy so I'll attempt to clarify. At the point of death a person ceases to be either a moral agent or moral subject, there may be reasons to honour the requests of the living post mortem but this is because a failure to do so would cause harm to the living who would no longer be confident those desires in death would be honoured. My point was that at the point of death there is neither harm nor happiness for that moral subject as there is no moral subject.

As a point of clarification, I claim that there's sufficient evidence in the world to claim that human nature is inclined toward evil. That's not quite the same as saying there is more harm than happiness, but one could potentially infer this from the observation.

I don't really buy this argument. People do all sorts of things out of fear that end up hurting them. For example, people make really awful financial decisions because their immediate fear is larger than their sense of delayed value. I do not find that the human fight for survival is persuasive from a consequentialist perspective.

Okay. But I still find the vague sense of thinking that a person's life might be better based on potential future happiness to be difficult to maintain. In particular, it seems to say that at a certain age (so that future happiness is smaller) random violence against persons can become a moral good because the amount of happy life they have left is smaller than the amount of suffering they will have as they age. I think that cuts against a lot of moral intuitions.

Out of curiosity, do you think that human nature is inclined towards good?

(which I suppose could be either rejected because you think this statement is incompatible with yours, or that it might be compatible, but you just don't htink it is true)

Okay fair enough.

All I am trying to do is establish a prima facie case that people consider their lives worth living. It may seem that this is a low bar but if it is hard to maintain, as you state below, that lives may be better due to some sense of future happiness then it seems to follow that most people want to live be based on a current sense of their lives worth.

I consider the first person perspective privileged in most cases when attempting to determine whether a life is worth living. There are exceptions but these are both a small minority and must be subject to rigorous judicial and medical constraints. Principles of agency and individual autonomy I'd hold are central to a conception of a life worth living and so to kill someone against their will the greatest harm that can be done. On this account random killings aren't a moral good nor can they be, further random killings cause distress beyond those killed and so I don't find this line persuasive.

In privileging the first person perspective I allow that people are best placed to choose when their lives are not worth living or when they believe that the suffering will outweigh the happiness. I'll admit that there are certain parts of that debate I am unable to answer to my own satisfaction but it's where I stand in principle.

I assume this is directed at Aaron, in any case it's an interesting question, with caveats I'd say yes.

I believe human nature is naturally inclined towards evil. This does not mean that it is incapable of doing good.

I'll respond to your other post when I have more time.

That isn't quite an answer to the question though, whether human nature is ALSO naturally inclined towards good? Like for instance, there are lot of terms here to define and this is a big oversimpliciation, but broadly I would probably suggest that humans are inclined to both.

I'll respond to your other post when I have more time.[/QUOTE]Please take your time.